{ "name": "NANYTE photoresist process recipe library", "url": "https://nanyte.com/photoresists", "license": "Data compiled from manufacturer datasheets; each recipe cites its source. Free to use with attribution.", "disclaimer": "Manufacturer datasheet values are starting points; optimal parameters depend on your substrate, equipment and environment. Trademarks belong to their respective owners. NANYTE is not affiliated with the manufacturers listed.", "count": 53, "recipes": [ { "slug": "ar-n-4340", "name": "AR-N 4340", "manufacturer": "Allresist", "productLine": "AR-N 4300 series", "aliases": [ "ARN4340", "AR-N4340" ], "tone": "negative", "chemistry": "car", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not marketed for greyscale or 3D lithography; targeted at IC production with high-contrast (Contrast: 5.0), high-resolution (0.5 µm) binary patterning.", "status": "active", "successorSlug": null, "summary": "AR-N 4340 is Allresist's highly sensitive chemically amplified negative-tone photoresist for integrated-circuit fabrication, coating 1.4 µm at 4000 rpm, i-line/g-line/broadband sensitive, with undercut lift-off profiles obtainable by extending development time.", "thicknessRange": { "min_um": 1.4, "max_um": 2, "basis": "stated", "source": "stated — two concrete example film thicknesses are named for AR-N 4340 in the datasheet's own SEM-example captions: 1.4 µm ('Resist structures' figure, 0.7 µm L/S) and 2.0 µm ('Structure resolution' figure, 4.0 µm structure), the former matching the reference 'Film thickness/4000 rpm (µm): 1.4' value in the Properties I table. Not a stated achievable min-max range in prose, and not a curve span (see spinNotes for why the figure's fuller range is not used)." }, "spinCurves": [ { "label": "AR-N 4340", "points": [ { "rpm": 500, "um": 4.16 }, { "rpm": 1000, "um": 2.95 }, { "rpm": 2000, "um": 2.09 }, { "rpm": 3000, "um": 1.71 }, { "rpm": 4000, "um": 1.47 }, { "rpm": 5000, "um": 1.31 }, { "rpm": 6000, "um": 1.21 } ], "source": "read from figure, 'Spin curve' (D₀/µm 0.0–5.0 vs rpm 0–8000), p.50 of Allresist AR-N 4300 product-information sheet (as of January 2014); single-grade chart, one legended trace labeled 'AR-N 4340' — no curve-identification ambiguity (unlike the AR-U 4000 series page); 7 diamond markers digitized at rpm 500/1000/2000/3000/4000/5000/6000 (the fitted line continues unmarked to 8000 rpm at ~1.05 µm but that tail was not digitized as a point, since it carries no discrete marker); anchor check: the figure's 4000 rpm marker reads ≈1.47 µm, about 5% above the Properties I table's printed anchor 'Film thickness/4000 rpm (µm): 1.4' (p.50, cross-referenced by the Coating box, p.51: '4000 rpm, 60 s, 1.4 µm') — the table value is the more authoritative single number, the figure read is close but not exact; digitized 2026-07-12", "figureRead": true } ], "spinNotes": "Spin-speed-vs-thickness figure ('Spin curve', p.50, axes D₀/µm 0-5.0 vs rpm 0-8000) digitized 2026-07-12: single-grade chart, one legended AR-N 4340 trace with 7 diamond markers from 500-6000 rpm, anchored against the Properties I table's 4000 rpm -> 1.4 µm reference point (the figure's own 4000 rpm marker reads ≈1.47 µm, a ~5% deviation from the table value — both are recorded, see spinCurves[0].source). No dispense volume, acceleration ramp, or edge-bead-removal step is described anywhere in this document.", "adhesion": { "hmds": null, "notes": "Datasheet specifies Allresist's own adhesion promoter AR 300-80 (Process chemicals table, p.50); does not mention HMDS specifically, so hmds is left null rather than assumed true or false. 'Good adhesion' is listed as a general Characteristic (p.50)." }, "rehydration": null, "softbake": { "temp_c": 90, "time_s": 60, "method": "hotplate", "notes": "± 1 °C tolerance. Alternative: 85 °C, 25 min (1500 s) convection oven. Datasheet calls this step 'Softbake' in the Process conditions diagram and 'Tempering' in the Process parameters table (85 °C, 60 s, hot plate — note this table gives a slightly different temperature, 85 °C not 90 °C, for the same nominal step; both are reproduced here for transparency).", "source": "Process conditions diagram, p.51; Process parameters table, p.50" }, "exposureDose": { "doses": [ { "wavelength_nm": null, "value_mJcm2": 140, "source": "Process conditions diagram, p.51 of Allresist AR-N 4300 series product information (English); dose stated as \"Exposure dose (E0, broadband UV stepper): 140 mJ/cm², 1.4 µm\"." } ], "datasheetBasis": "'UV exposure: Broadband UV, 365 nm, 405 nm, 436 nm. Exposure dose (E0, broadband UV stepper): 140 mJ/cm², 1.4 µm' (Process conditions diagram, p.51). Separately, Characteristics (p.50) lists spectral sensitivity as 'i-line, g-line', and the Process parameters table (p.50) states 'Exposure: i-line stepper (NA: 0.65)' without giving a dose for that specific tool.", "_note": "The 140 mJ/cm² dose is explicitly a broadband figure spanning i-line/h-line/g-line (365/405/436 nm) together on a broadband UV stepper — the datasheet does not isolate a dose to any single wavelength, so it is recorded as an unattributed value_mJcm2 rather than filed under at365 or at405 (per this project's rule against conflating a broadband number with a single-line dose). No dose is given for the separately mentioned i-line-only stepper (NA 0.65) process." }, "peb": { "temp_c": 95, "time_s": 120, "notes": "Referred to as 'Crosslinking bake' — this is the post-exposure step that crosslinks the exposed novolac/PAG/amine system into the negative image. ± 1 °C tolerance. Alternative: 90 °C, 25 min (1500 s) convection oven. A TCD (time-to-clear-development)-vs-bake-temperature table (p.51) shows this crosslinking bake temperature is critical: at 70 °C the clearing dose is 480 mJ/cm² (TCD 20 s), at 100 °C it drops to 65 mJ/cm² (TCD 41 s), and above 130 °C the resist 'is not developable any more'; optimum bake temperatures are stated as 90-100 °C.", "source": "Process conditions diagram, p.51; TCD vs. bake temperature table, p.51" }, "floodExposure": null, "develop": { "developer": "AR 300-475", "dilution": null, "time_s": 60, "method": "puddle", "rinse": "DI-H2O, 30 s", "source": "Process conditions diagram, p.51; Process parameters table, p.50", "_note": "AR 300-475 used undiluted at 60 s, 21-23 °C (± 0.5 °C), puddle method, per the Process conditions diagram; the Process parameters table separately states 60 s at 22 °C via AR 300-475. Note: 'By extending the development time, an undercut (lift-off) of the resist structure can be obtained at minimum possible exposure dose.' A separate small 'Development recommendations' table (p.51) also lists AR-N 4340 against three developer columns (AR 300-26, AR 300-35, AR 300-40) with entries '1:1', 'undil.', '300-475' respectively — transcribed verbatim here because its exact intended meaning (alternate acceptable developers and their dilutions, vs. a substitution note) is ambiguous in the source table layout as extracted; a human reviewer should re-read this small table directly against the PDF, p.51. developerFamily tmah: AR 300-475 is a TMAH-based, metal-ion-free developer per Allresist's own product page (https://www.allresist.com/portfolio-item/developer-ar-300-475/, classified 2026-07-12)." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Datasheet states only an optional ceiling capability — 'Hardening of structures up to 300 °C (optional)' — without a specific temperature/time recipe or bake method (hotplate vs. oven), so left null rather than assigning a single value within that ceiling. A related 'Customer-specific technologies' step lists a 150 mJ/cm² flood exposure followed by a 115 °C, 1 min hot-plate bake for 'generation of e.g. semiconductor properties or lift-off' — distinct from the AR-U 4000 series' true image-reversal flood-exposure step, and not modeled in the floodExposure field per this project's schema convention (reserved for genuine image-reversal resists).", "source": "Process conditions diagram, p.51" }, "descum": null, "applications": [ "etch-mask", "lift-off" ], "etchResistance": "\"Plasma etching resistant, temperature-stable up to 220 °C after subsequent treatment\" (Characteristics, p.50). Plasma etching rates (5 Pa, 240-250 V Bias, Properties II table, p.50): Ar-sputtering 8 nm/min, O2 173 nm/min, CF4 33 nm/min, 80% CF4 + 16% O2 93 nm/min.", "liftoffSuitable": true, "platingSuitable": null, "stripper": "AR 300-76 or AR 300-72 (Process chemicals table, p.50); alternatively AR 300-76 or O2 plasma ashing per the Process conditions diagram, p.51.", "storage": "\"Storage 6 month (°C): 10 - 18\" (Properties I table, p.50) — i.e. a 6-month shelf life is stated when stored at 10-18 °C.", "notes": "AR-N 4340 is Allresist's highly sensitive negative-tone photoresist in the AR-N 4300 series, described as 'novolac with photochemical acid generator and amine-based crosslinking agent' — a chemically amplified negative chemistry (mapped here to the 'car' enum value) rather than the older bisazide-novolak crosslinking chemistry used in resists like ma-N 1400. Development strongly depends on the crosslinking-bake temperature: the datasheet's own TCD table shows optimum crosslinking-bake temperatures of 90-100 °C, with clearing dose rising sharply outside that window (480 mJ/cm² at 70 °C vs. 65 mJ/cm² at 100 °C) and the resist becoming completely undevelopable above 130 °C. Undercut (lift-off) profiles are obtained not by underexposing but by extending development time at the resist's minimum clearing dose, per the datasheet's explicit recommendation. The published exposure dose (140 mJ/cm² for a 1.4 µm film) is a broadband figure spanning i-line/h-line/g-line together (365/405/436 nm) rather than a single-wavelength dose, so it is recorded as an unattributed value rather than filed under either the i-line or h-line field — a separately mentioned i-line-only stepper configuration (NA 0.65) is given no dose at all. A single-grade spin-speed-vs-thickness figure exists in the datasheet (0-8000 rpm, 0-5 µm), but only the exact 4000 rpm -> 1.4 µm anchor from the accompanying numeric table is published here as structured data; additional curve points should be read directly from the source figure by a human reviewer.", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://www.allresist.com/wp-content/uploads/sites/2/2016/12/allresist_produktinfos_ar-n4300_englisch.pdf", "datasheetVersionOrDate": "p.50: 'As of January 2014'; p.51: 'As of January 2016' (the two pages of this two-page spread carry different revision dates as printed)", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "Document is Allresist's own 'Negative Photoresist AR-N 4300' product-information sheet (2-page spread, catalog pages 50-51), specifically and entirely covering the AR-N 4340 grade — a genuine manufacturer technical datasheet (process conditions, spin curve, plasma-etch rates, TCD-vs-bake-temperature table), not a flyer or third-party mirror. Retrieved directly from allresist.com.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": true }, { "slug": "ar-p-3510t", "name": "AR-P 3510T", "manufacturer": "Allresist", "productLine": "AR-P 3500 series", "aliases": [ "AR-P 3510 T" ], "tone": "positive", "chemistry": "dnq-novolak", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Document never mentions grayscale or 3D lithography; it targets standard binary IC patterning, plasma-etch masking, and lift-off.", "status": "active", "successorSlug": null, "summary": "Positive-tone, DNQ-novolak resist from Allresist's AR-P 3500 series with a wide process range (broadband UV / i-line / g-line); AR-P 3510T is the TMAH-compatible ('T') variant of the 3510 grade, coating to 2.0 µm at 4000 rpm and rated suitable for 0.26 N TMAH developer.", "thicknessRange": { "min_um": 2, "max_um": 2, "basis": "stated", "source": "stated — the 'Film thickness / 4000 rpm (µm)' row in 'Properties I' (p.1) gives 2.0 µm as a SINGLE value spanning the whole 'AR-P 3510 / 3510 T' column pair (unlike the adjacent solids-content/viscosity/resolution/contrast rows, which ARE split per grade) — i.e. the document itself states this thickness is shared between the base and T grades. Confirmed again by the worked 'Coating' row of the process-conditions table: '4000 rpm, 60 s, 2.0 µm' under its 'AR-P 3510' example column. No independent min–max achievable range is stated for this grade; recorded as a single documented anchor, not a spread." }, "spinCurves": [], "spinNotes": "The 'Spin curve' figure (p.1, bottom-left; axes: film thickness 1.0–6.0 µm vs. spin speed 0–8000 rpm) plots TWO traces, and per the manufacturer's OWN legend each trace already covers a combined pair of grades: 'AR-P 3510/3510 T' as one trace, 'AR-P 3540/3540 T' as the other — this document itself, not an extraction error, presents 3510 and 3510T as sharing one spin curve. This extraction did not read numeric points off the figure: the PDF's text layer surfaces only axis tick labels, not the underlying vector-curve geometry, and a low-confidence multi-grade read is worse than none (per the SU-8 2050 / LOR 3A precedents in this project). The only confirmed numeric anchor is 2.0 µm at 4000 rpm, 60 s spin time (from the 'Coating' row of the worked process-conditions table, p.1, 'AR-P 3510' column — unsplit for this grade pair per the Properties I table). No accel or edge-bead detail is published.", "adhesion": { "hmds": false, "notes": "Manufacturer states 'very good adhesion properties' as a series-wide (AR-P 3500(T)) characteristic. The 'Process chemicals' box (p.1) names AR 300-80 as the adhesion promoter — NOT HMDS — but that box is positioned alongside the AR-P 3540T worked example specifically, so it is not confirmed to be identical for AR-P 3510T; included here as the only adhesion-promoter product named anywhere in this document." }, "rehydration": null, "softbake": { "temp_c": 100, "time_s": 60, "method": "hotplate", "notes": "'Tempering (± 1 °C): 100 °C, 1 min, hot plate OR 95 °C, 25 min, convection oven' — this row of the worked process-conditions table (p.1) is NOT split between the table's two example columns (AR-P 3510, AR-P 3540 T), i.e. it is presented as one shared condition for the worked AR-P 3500(T) example, which is why it is attributed to AR-P 3510T here despite that exact SKU not being one of the two labeled columns. Oven alternative: 95 °C for 25 min (1500 s).", "source": "\"Tempering\" row, \"Process conditions\" table, p.1, of allresist_produktinfos_ar-p3500_3500t_englisch.pdf" }, "exposureDose": { "doses": [], "datasheetBasis": "Broadband UV, 365 nm, 405 nm, 436 nm", "_note": "The 'UV exposure' row states the exposure basis (\"Broadband UV, 365 nm, 405 nm, 436 nm\") as a shared, unsplit descriptor for the whole worked example — a genuinely broadband figure spanning g/h/i-line together, so per the wavelength-routing rule any dose under it would go to value_mJcm2, never at365/at405. However, the actual 'Exposure dose (E0, broadband UV stepper)' row IS split by grade: 55 mJ/cm² under the 'AR-P 3510' column, 120 mJ/cm² under the 'AR-P 3540 T' column. Neither column is literally 'AR-P 3510 T', and dose is clearly grade-sensitive in this table (differs 55 vs. 120 mJ/cm² between the two shown grades), so the AR-P 3510 (non-T) value is NOT assumed to carry over to AR-P 3510 T. No exposure dose specific to AR-P 3510 T is published in this document — value_mJcm2 is left null rather than guessed." }, "peb": null, "floodExposure": null, "develop": { "developer": "AR 300-26 (1:2) or AR 300-35 (undiluted) or AR 300-44", "dilution": "AR 300-26 at 1:2; AR 300-35 used undiluted; AR 300-44 listed as the third (ready-to-use) option — three alternative developer/dilution combinations named for the combined 'AR-P 3510 T, 3540 T' row of the 'Development recommendations' table (p.1). This differs from the base AR-P 3510/3540 row, which instead lists AR 300-26 at 1:5, AR 300-35 at 1:1, and 'AR 300-40 (300-47), 1:1'.", "time_s": null, "method": "puddle", "rinse": "DI-H2O, 30 s", "source": "\"Development recommendations\" table (AR-P 3510 T, 3540 T row) + \"Development\"/\"Rinse\" rows of the \"Process conditions\" table, p.1, of allresist_produktinfos_ar-p3500_3500t_englisch.pdf", "_note": "No develop TIME is given specifically for the AR-P 3510 T / 3540 T dilution row. The worked process-conditions table gives 60 s develop times, but its two labeled example columns are 'AR-P 3510' (untreated) and 'AR-P 3540 T' — neither is literally 'AR-P 3510 T' — so time_s is left null rather than assumed, even though both known examples happen to use 60 s. Method ('puddle') and rinse ('DI-H2O, 30 s') are unsplit rows spanning the whole worked example and are used here as reasonably series-wide." }, "hardbake": { "temp_c": 115, "time_s": 60, "notes": "Labeled 'Post-bake (optional)' in the source, not 'hardbake', but functionally the same post-develop bake step: 115 °C, 1 min hot plate, OR 115 °C, 25 min (1500 s) convection oven. This row is unsplit across the table's two example columns (AR-P 3510, AR-P 3540 T), so attributed here to AR-P 3510T on the same reasoning as softbake.", "source": "\"Post-bake (optional)\" row, \"Process conditions\" table, p.1, of allresist_produktinfos_ar-p3500_3500t_englisch.pdf" }, "descum": null, "applications": [ "lift-off", "etch-mask" ], "etchResistance": "Manufacturer states the AR-P 3500(T) series is 'plasma etching resistant, temperature-stable up to 120 °C' (Characterisation, series-wide, p.1). A 'Plasma etching rates (nm/min)' table (5 Pa, 240–250 V bias: Ar-sputtering 7, O2 165, CF4 37, 80 CF4 + 16 O2 88) is given but is grouped with the AR-P 3540T-labeled Cauchy coefficients, so it is not confirmed to represent AR-P 3510T specifically and is not carried into a numeric field here.", "liftoffSuitable": true, "platingSuitable": null, "stripper": "AR 300-70 or O2 plasma ashing (\"Removal\" row, \"Process conditions\" table, p.1 — unsplit, shared across the worked example). Note: a separate 'Process chemicals' box elsewhere on p.1, positioned alongside the AR-P 3540T worked example, instead lists 'Remover AR 300-76, T: AR 300-76' — that box is not confirmed to apply to AR-P 3510T and is not used as the primary value here.", "storage": "10–18 °C, approximately 6-month shelf life (\"Storage 6 month (°C)\" row, \"Properties I\" table, p.1 — a single value spanning both the 3510/3510T and 3540/3540T column groups).", "notes": "AR-P 3510T is the TMAH-compatible ('T') variant of Allresist's AR-P 3510 grade within the wider AR-P 3500(T) series, a classic DNQ/novolac positive resist for IC production with broadband UV / i-line / g-line sensitivity. The manufacturer states the 3500T sub-line is 'suitable for TMAH developer 0.26 n', distinguishing it from the base (non-T) grades. Several process parameters in this document are given only for the two SPECIFIC worked example columns the table actually labels — 'AR-P 3510' (base) and 'AR-P 3540 T' — and NOT for 'AR-P 3510 T' itself; most notably the exposure dose (55 vs. 120 mJ/cm² for the two shown grades) is clearly grade-sensitive, so no dose is recorded here rather than borrowing the base grade's number. By contrast, film thickness at 4000 rpm (2.0 µm) and the spin-curve figure are explicitly shared/combined between 3510 and 3510T by the manufacturer's own table structure and legend. A large block of dose-range, depth-of-focus, and SEM linearity/dark-field-erosion data on p.1 (bottom) and all of p.2 is explicitly labeled 'AR-P 3540 T' and belongs to that different grade, not to AR-P 3510T — excluded here despite the superficially similar SKU name.", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://www.allresist.com/wp-content/uploads/sites/2/2014/03/allresist_produktinfos_ar-p3500_3500t_englisch.pdf", "datasheetVersionOrDate": "As of January 2014", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "A manufacturer's own product-information catalog excerpt (pages 22–25 of a larger Allresist catalog, page footer 'As of January 2014'), hosted directly on allresist.com — not a third-party mirror or distributor sheet.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "ar-p-5350", "name": "AR-P 5350", "manufacturer": "Allresist", "productLine": "AR-P 5300 series", "aliases": [], "tone": "positive", "chemistry": "dnq-novolak", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Document never mentions grayscale or 3D lithography; the series is purpose-built for undercut/lift-off profiles, not grayscale relief.", "status": "active", "successorSlug": null, "summary": "Positive-tone, DNQ-novolak resist purpose-built for lift-off from Allresist's AR-P 5300 series; AR-P 5350 is the thinner, higher-resolution grade (1.0 µm at 4000 rpm, 0.5 µm resolution) whose elevated softbake temperature generates the undercut profile needed for clean metal lift-off.", "thicknessRange": { "min_um": 1, "max_um": 1, "basis": "stated", "source": "stated — the 'Film thickness/4000 rpm (µm)' row in 'Properties I' (p.1) gives 1.0 µm for AR-P 5350, confirmed independently by the worked 'Coating' row of the 'Process conditions' table (p.2): '4000 rpm, 60 s, 1.0 µm', AR-P 5350 column. No min–max achievable range is stated for this grade; recorded as a single documented anchor, not a spread." }, "spinCurves": [ { "label": "AR-P 5350", "points": [ { "rpm": 1000, "um": 1.96 }, { "rpm": 1500, "um": 1.58 }, { "rpm": 2000, "um": 1.38 }, { "rpm": 3000, "um": 1.11 }, { "rpm": 4000, "um": 1.01 }, { "rpm": 6000, "um": 0.78 }, { "rpm": 7000, "um": 0.73 } ], "source": "read from figure, 'Spin curve' (D₀/µm 0.0–12.0 vs rpm 0–8000), p.1 of AR-P5300_english_Allresist_product_information.pdf (Allresist, as of January 2018 per p.1 footer); two individually-legended traces plotted together — the lower diamond-marker trace explicitly labeled 'AR-P 5350' beside its rightmost point (the upper square-marker trace is the separately labeled 'AR-P 5320', not extracted here); 7 diamond markers digitized at rpm 1000/1500/2000/3000/4000/6000/7000 (no marker plotted at 5000 rpm — the line passes through that gap unmarked — and both traces terminate at 7000 rpm, not extending to the axis's 8000 rpm end); anchor check: the figure's 4000 rpm marker reads ≈1.01 µm, matching the Properties I table's printed anchor 'Film thickness/4000 rpm (µm): 1.0' for AR-P 5350 (p.1) and the Coating row of the process-conditions table (p.2: '4000 rpm, 60 s, 1.0 µm'); digitized 2026-07-12", "figureRead": true } ], "spinNotes": "Spin-speed-vs-thickness figure ('Spin curve', p.1, axes D₀/µm 0.0-12.0 vs rpm 0-8000) digitized 2026-07-12: two separately labeled traces, AR-P 5320 (upper) and AR-P 5350 (lower) — only the AR-P 5350 trace was digitized here, 7 diamond markers from 1000-7000 rpm (no marker at 5000 rpm; the trace does not extend to 8000 rpm). The figure's own 4000 rpm marker reads ≈1.01 µm, matching the Properties I table's 1.0 µm anchor closely (see spinCurves[0].source).", "adhesion": { "hmds": false, "notes": "Manufacturer states 'good adhesion properties' as a series-wide characteristic. The 'Process chemicals' box (p.1, general/series-wide, not tied to either grade's SEM caption) names AR 300-80 as the adhesion promoter — not HMDS." }, "rehydration": null, "softbake": { "temp_c": 105, "time_s": 240, "method": "hotplate", "notes": "105 °C, 4 min hot plate — confirmed identically in two places: the 'Process parameters' box (p.1, explicitly labeled 'AR-P 5350' worked example) and the 'Tempering' row of the process-conditions table (p.2, which also gives a convection-oven alternative: 100 °C, 40 min = 2400 s). Processing instructions (p.2) note: 'Higher tempering temperatures are required to produce the undercut' — the elevated bake (vs. a typical ~90–100 °C positive-resist softbake) is deliberate and load-bearing for the lift-off undercut profile, not incidental.", "source": "\"Process parameters\" box, p.1 (AR-P 5350) + \"Tempering\" row, \"Process conditions\" table, p.2, of AR-P5300_english_Allresist_product_information.pdf" }, "exposureDose": { "doses": [ { "wavelength_nm": null, "value_mJcm2": 55, "source": "\"UV exposure\" and \"Exposure dose\" rows, \"Process conditions\" table, p.2, AR-P 5350 column, of AR-P5300_english_Allresist_product_information.pdf" } ], "datasheetBasis": "Broadband UV, 365 nm, 405 nm, 436 nm", "_note": "'UV exposure' row states the basis as broadband UV spanning 365/405/436 nm together — a genuinely broadband figure, so per the wavelength-routing rule the dose goes to value_mJcm2, not at365/at405 (both null; no monochromatic i-line or h-line dose is published). The 'Exposure dose (E0, broadband UV stepper)' row is split cleanly by grade: 58 mJ/cm² for AR-P 5320, 55 mJ/cm² for AR-P 5350 — this document names AR-P 5350 explicitly as its own column, unlike the sibling AR-P 3500/3500T sheet where the target grade's dose could not be confirmed." }, "peb": null, "floodExposure": null, "develop": { "developer": "AR 300-35 (primary, fully worked example); AR 300-26 or AR 300-47 listed as alternatives", "dilution": "AR 300-35 at 1:2 (worked example); alternatively AR 300-26 at 1:7, or AR 300-47 at 2:3 — per the 'Development recommendations' table, AR-P 5350 row", "time_s": 60, "method": "puddle", "rinse": "DI-H2O, 30 s", "source": "\"Process parameters\" box, p.1 (AR-P 5350) + \"Development\"/\"Rinse\" rows of the \"Process conditions\" table, p.2 + \"Development recommendations\" table, p.2 (AR-P 5350 row), of AR-P5300_english_Allresist_product_information.pdf", "_note": "Two independent places in the document agree: AR 300-35 at 1:2, 60 s, 22 °C, puddle, for AR-P 5350 — used as the primary recorded value (time_s = 60). The 'Development recommendations' table additionally lists AR 300-26 at 1:7 and 'AR 300-47' at 2:3 as alternative developer/dilution options for AR-P 5350, but gives no develop time for either alternative. Developer temperature 21–23 °C ± 0.5 °C per the process-conditions table header. Processing instructions (p.2) note: 'The undercut of resist structures is generated during aqueous-alkaline development.' developerFamily buffered-alkaline: AR 300-35 (the primary, fully-worked developer) is sodium metasilicate/phosphate-based per Allresist product information (classified 2026-07-12); the alternative AR 300-47 is TMAH-based (metal-ion-free) and the alternative AR 300-26 is buffered-alkaline like AR 300-35 — the primary developer's family is recorded here, not the TMAH alternative." }, "hardbake": null, "descum": null, "applications": [ "lift-off", "etch-mask" ], "etchResistance": "Manufacturer states the series is 'plasma etching resistant, temperature stable up to 120 °C' (Characterisation, series-wide, p.1). A 'Plasma etching rates (nm/min)' table (5 Pa, 240–250 V bias: Ar-sputtering 7, O2 161, CF4 39, 80 CF4 + 16 O2 90) is given but is NOT labeled to a specific grade in this document (unlike the sibling AR-P 3500/3500T sheet, where an equivalent table WAS explicitly tagged). The glass-transition temperature (108 °C) and dielectric constant (3.1) given alongside it are numerically IDENTICAL to the same two fields in the AR-P 3500/3500T sheet, suggesting these may be generic Allresist-family boilerplate rather than AR-P 5350-specific measurements — flagged, not asserted as grade-specific.", "liftoffSuitable": true, "platingSuitable": null, "stripper": "AR 300-76 or O2 plasma ashing (\"Removal\" row, \"Process conditions\" table, p.2, shared/unsplit across both grades). The 'Process chemicals' box (p.1, series-wide) additionally lists AR 600-71 as an alternative remover.", "storage": "10–18 °C, approximately 6-month shelf life (\"Storage 6 month (°C)\" row, \"Properties I\" table, p.1 — a single value shared across the AR-P 5320 and AR-P 5350 columns).", "notes": "AR-P 5350 is the thinner (1.0 µm at 4000 rpm), higher-contrast, higher-resolution (0.5 µm) grade of Allresist's AR-P 5300 series, a DNQ/novolac positive resist purpose-built for lift-off. Unlike most positive resists, an elevated softbake (105 °C, 4 min — well above a typical ~90–100 °C positive-resist prebake) deliberately produces the negative-sloped undercut profile needed for clean metal lift-off; the manufacturer states this explicitly ('higher tempering temperatures are required to produce the undercut'). Post-bake is explicitly 'Not required' for this series (process-conditions table, p.2) — recorded here as hardbake: null rather than an unread field. Development is aqueous-alkaline (AR 300-35, 1:2, 60 s in the fully worked example, confirmed twice in the document); AR 300-26 (1:7) and 'AR 300-47' (2:3) are named as alternatives with no stated time. The two-page source disagrees with itself on revision date: page 1's footer reads 'As of January 2018', page 2's reads 'As of January 2017' — both are preserved in provenance rather than silently reconciled.", "developerFamily": "buffered-alkaline", "provenance": { "datasheetUrl": "https://www.allresist.com/wp-content/uploads/sites/2/2020/03/AR-P5300_english_Allresist_product_information.pdf", "datasheetVersionOrDate": "Inconsistent across pages: 'As of January 2018' (p.1 footer) vs. 'As of January 2017' (p.2 footer) — recorded verbatim, not reconciled.", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "A manufacturer's own product-information sheet (pages 30–31 of a larger Allresist catalog), hosted directly on allresist.com — not a third-party mirror or distributor sheet. Unlike the sibling AR-P 3500/3500T sheet, this document names AR-P 5350 as an explicit, unambiguous column/label in every table, so no SKU-existence caveat is needed for the name itself (only for a few series-wide-vs-grade-specific attribution questions, noted inline above).", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "ar-u-4030", "name": "AR-U 4030", "manufacturer": "Allresist", "productLine": "AR-U 4000 series", "aliases": [ "ARU4030", "AR-U4030" ], "tone": "image-reversal", "chemistry": "bisazide-novolak", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not marketed for greyscale or 3D lithography; targets binary IC patterning, switchable between positive-tone and negative/image-reversal-tone processing of the same coated film.", "status": "active", "successorSlug": null, "summary": "AR-U 4030 is Allresist's image-reversal photoresist in the AR-U 4000 series, a novolac/bisazide negative formulation that can also be run as an ordinary positive resist, coating roughly 1.8-2.5 µm and producing pronounced undercut lift-off profiles when processed with an image-reversal bake plus flood exposure.", "thicknessRange": { "min_um": 1.8, "max_um": 2.5, "basis": "stated", "source": "stated — two concrete example film thicknesses are named for AR-U 4030 in the datasheet: the reference 'Film thickness/4000 rpm (µm): 1.8' (Properties I table, p.32, matching the Coating boxes on both process diagrams), and the 'Structure resolution' SEM example caption 'AR-U 4030, Undercut negative structures at a film thickness of 2.5 µm' (p.32). Not a stated achievable min-max range in prose, and not a curve span (see spinNotes for why the figure's fuller range is not used)." }, "spinCurves": [ { "label": "AR-U 4030", "points": [ { "rpm": 1000, "um": 4.12 }, { "rpm": 2000, "um": 2.93 }, { "rpm": 3000, "um": 2.4 }, { "rpm": 4000, "um": 2.09 }, { "rpm": 5000, "um": 1.89 }, { "rpm": 6000, "um": 1.79 } ], "source": "read from figure, 'Spin curve' (D₀/µm 0-6.0 vs rpm 0-8000), p.32 of Allresist AR-U 4000 product-information sheet (as of January 2017); genuine multi-grade chart with three individually legended traces — AR-U 4030 (topmost, dark-navy diamonds), AR-U 4040 (middle, red squares), AR-U 4060 (bottom, olive triangles). TRACE IDENTITY, re-proven 2026-07-12: the topmost (navy diamond) trace is AR-U 4030, established two independent ways: (1) its marker color matches the dark-navy 'AR-U 4030' legend label positioned at that trace's right-hand endpoint; (2) it sorts correctly against the Properties I table's per-grade nominal 'Film thickness/4000 rpm (µm)' row (AR-U 4030 = 1.8, AR-U 4040 = 1.4, AR-U 4060 = 0.6, p.32) and the Coating-box captions ('4000 rpm, 60 s, 1.8/1.4/0.6 µm', p.33/p.34) — the topmost/thickest plotted trace must be the grade with the largest nominal thickness (4030), the middle trace 4040, the bottom (olive triangle) trace 4060; no ambiguity in trace assignment. ADJUDICATED 2026-07-12 (third pixel read, WL directive: figure over table): the prior extraction's 6 points were systematically ~15-30% too low (a mis-calibration, not a wrong-trace read — same topmost/navy trace, same 6 rpm columns). Re-extracted via PyMuPDF: axes calibrated from the 13 horizontal gridlines (rows 18-269, evenly spaced, matching the 13 y-axis labels 0.0-6.0 in 0.5 steps exactly, giving 269=D0-0µm, 41.833 px/µm) and the 9 vertical gridlines (columns 54-433, matching the 9 x-axis labels 0-8000 rpm in 1000-rpm steps exactly, giving 54=0rpm, 0.047375 px/rpm); AR-U 4030 marker centroid at each rpm column taken as the midpoint of the longest contiguous run of navy-diamond-colored pixels within +-6 px of the expected column (rejects gridline/legend-text color bleed, which a naive min/max-span search does not). New read: 4.12/2.93/2.40/2.09/1.89/1.79 µm at 1000/2000/3000/4000/5000/6000 rpm — closely matches the independent cross-check extraction (4.13/2.94/2.41/2.09/1.88/1.71) to within 1-5% at every point, confirming the cross-check was correct and this file's prior read erred. ANCHOR CHECK — CONFIRMED CONFLICT, RESOLVED PER D11: the figure's own 4000 rpm marker reads 2.09 µm, ~16% above the Properties I table's printed anchor 'Film thickness/4000 rpm (µm): 1.8' and the matching Coating-box captions. Per WL's domain rule (a figure/measured curve usually wins over a printed nominal table value), the figure value (2.09 µm) is retained as the recipe's spin-curve data; the 1.8 µm table anchor is a nominal/rounded spec value, not necessarily this specific measured curve's own reading. This is a genuine figure-vs-table disagreement in the source datasheet, not a misread — figure retained per adjudication policy.", "figureRead": true } ], "spinNotes": "Spin-speed-vs-thickness figure ('Spin curve', p.32, axes D₀/µm 0-6.0 vs rpm 0-8000): genuine multi-grade chart, three individually legended traces (AR-U 4030 topmost/navy, AR-U 4040 middle/red, AR-U 4060 bottom/olive); only the AR-U 4030 (topmost navy diamond) trace was digitized here, 6 markers from 1000-6000 rpm. ADJUDICATED 2026-07-12 (third pixel read, WL directive: figure over table): trace identity re-proven via both the legend color match AND the Properties I nominal-thickness ranking (4030=1.8 > 4040=1.4 > 4060=0.6 µm at 4000 rpm, matching top>mid>bottom trace order) — no ambiguity. The prior digitization was ~15-30% too low (mis-calibration); re-extracted with gridline-exact axis calibration (see spinCurves[0].source), now matching the independent cross-check extraction to within 1-5% at every point. The figure's own 4000 rpm marker (2.09 µm) is ~16% above the Properties I table's printed anchor (1.8 µm) — a genuine figure-vs-table disagreement in the source datasheet; per D11 the figure value is retained. No dispense volume or acceleration ramp is described anywhere in this document.", "adhesion": { "hmds": null, "notes": "Datasheet specifies Allresist's own adhesion promoter AR 300-80 (Process chemicals table, p.32); does not mention HMDS specifically, so hmds is left null rather than assumed true or false." }, "rehydration": null, "softbake": { "temp_c": 90, "time_s": 60, "method": "hotplate", "notes": "Referred to as 'Tempering' in the datasheet. ± 1 °C tolerance. Alternative: 85 °C, 25 min (1500 s) convection oven. Same schedule is used for both the positive-tone and negative/image-reversal processes. Corroborated by Processing instructions prose (p.35): for pure positive-tone use, 'a softbake at only 85 °C (oven) or 90 °C (hot plate) after coating is recommended... since this resist has the potential to be crosslinked due to its specific components.'", "source": "Process conditions diagrams, p.33 (positive) and p.34 (negative); Processing instructions, p.35" }, "exposureDose": { "doses": [ { "wavelength_nm": null, "value_mJcm2": 42, "source": "Process conditions, NEGATIVE (image-reversal) process diagram, p.34 of Allresist AR-U 4000 series product information (English); dose stated as \"Exposure dose (E0, broadband UV stepper): 42 mJ/cm²\". The positive-process diagram on p.33 gives a different dose (38 mJ/cm²) and is deliberately not used." } ], "datasheetBasis": "'Image-wise UV exposure: Broadband UV, 365 nm, 405 nm, 436 nm; 90 % layer build up. Exposure dose (E0, broadband UV stepper): 42 mJ/cm²' for AR-U 4030 (Process conditions negative process diagram, p.34).", "_note": "42 mJ/cm² is the image-wise dose for the NEGATIVE / IMAGE-REVERSAL process specifically (this resist's headline use, matching tone: 'image-reversal'). The SAME grade used instead as a plain POSITIVE-tone resist (no reversal bake, no flood exposure) uses a DIFFERENT dose, 38 mJ/cm² for AR-U 4030 (Process conditions positive-process diagram, p.33) — the two are not interchangeable and are not conflated here. Both the 42 and 38 mJ/cm² figures are broadband (365/405/436 nm) values, not attributed to a single wavelength, so recorded as an unattributed value_mJcm2 rather than at365/at405. No h-line- or i-line-isolated dose is published anywhere in the document; the separately mentioned g-line stepper (NA 0.56, Process parameters table p.32) is given no dose value of its own." }, "peb": { "temp_c": 115, "time_s": 240, "notes": "Referred to as 'Image reversal bake' in the datasheet — occurs after the image-wise exposure and before the flood exposure, and is specific to the negative/image-reversal process (it is not performed in the plain positive-tone process). Alternative: 110 °C, 25 min (1500 s) convection oven. Per Processing instructions (p.35): 'Intensifying the reversal bake supports the formation of vertical walls' for vertical-edge negative structures, while 'low temperature during reversal bake' is one of three levers (alongside low image-wise exposure and extended development time) that increase undercut for lift-off.", "source": "Process conditions negative process diagram, p.34; Processing instructions, p.35" }, "floodExposure": { "dose_mJcm2": 74, "notes": "'Flood exposure: Broadband UV stepper: approx. twice of image-wise without mask' — for AR-U 4030 specifically, 74 mJ/cm² (roughly 1.76x the 42 mJ/cm² image-wise dose). Performed without a mask, after the image reversal bake and before development, and is what converts the still-soluble (unexposed) areas into an alkali-soluble state so the final development produces a negative image. Broadband (365/405/436 nm), not attributed to a single wavelength.", "source": "Process conditions negative process diagram, p.34" }, "develop": { "developer": "AR 300-35", "dilution": "4:3", "time_s": 60, "method": "puddle", "rinse": "DI-H2O, 30 s", "source": "Process conditions negative process diagram, p.34", "_note": "This is the NEGATIVE (image-reversal) process recipe. Confirmed by the 'Development recommendations' table for the negative process (p.34): AR-U 4030 (1.8 µm) row lists AR 300-26 = 1:4, AR 300-35 = 4:3, AR 300-47 = 3:2 — the process diagram's primary recipe (AR 300-35, 4:3) matches this table's AR 300-35 column. The POSITIVE-mode process (p.33) instead develops AR-U 4030 in AR 300-35 undiluted (1:1), 60 s — a different dilution for a different process mode, not modeled as a separate array entry here. Development temperature 21-23 °C ± 0.5 °C in both modes. developerFamily buffered-alkaline: AR 300-35 is sodium metasilicate/phosphate-based per Allresist product information (classified 2026-07-12)." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Datasheet explicitly states post-bake is 'Not required' for both the positive and negative processes.", "source": "Process conditions diagrams, p.33 (positive) and p.34 (negative)" }, "descum": null, "applications": [ "image-reversal", "lift-off", "etch-mask" ], "etchResistance": "No qualitative etch-resistance claim is made in the Characteristics list (unlike Allresist's AR-N 4300 sheet); only a raw rate table is given. Plasma etching rates (5 Pa, 240-250 V Bias, Properties II table, p.32): Ar-sputtering 8 nm/min, O2 169 nm/min, CF4 40 nm/min, 80% CF4 + 16% O2 89 nm/min.", "liftoffSuitable": true, "platingSuitable": null, "stripper": "General Process chemicals table (p.32): AR 300-76, AR 300-72. Positive-process removal (p.33): 'AR 300-76 or O2 plasma ashing'. Negative-process removal (p.34): 'AR 300-70 or O2 plasma ashing' — this last figure ('AR 300-70') differs from the AR 300-76 used elsewhere in the same datasheet; reproduced verbatim as printed rather than corrected to 300-76, and flagged here for human verification against the source PDF, p.34, in case it is a distinct product or a transcription variant.", "storage": "\"Storage 6 month (°C): 8 - 12\" (Properties I table, p.32).", "notes": "AR-U 4030 is Allresist's image-reversal resist in the AR-U 4000 series — described as a 'combination of novolac and bisazide' — that can be processed either as an ordinary positive resist or, with an image-reversal bake and flood exposure, as a negative resist with pronounced undercut lift-off profiles. In image-reversal (negative) mode the sequence is: image-wise broadband exposure (42 mJ/cm² for this grade) -> an 'image reversal bake' (115 °C/4 min hotplate, or 110 °C/25 min oven) that crosslinks the exposed areas via the resist's amine component -> an unmasked flood exposure (74 mJ/cm², roughly twice the image-wise dose) that renders the remaining, still-unexposed areas developable -> puddle development in AR 300-35 (4:3). Undercut and vertical-wall profiles are controlled by the same three levers in opposite directions: low image-wise dose + low reversal-bake temperature + longer development time maximizes undercut for lift-off, while high dose + high bake temperature + shorter development time produces vertical walls suited to etch masking. Critically, the image-wise exposure dose differs by process mode for the same grade — 38 mJ/cm² if used as a plain positive resist versus 42 mJ/cm² in the image-reversal negative mode — so the two figures must not be conflated; this recipe records the negative/image-reversal dose since that matches the resist's headline classification (tone: image-reversal), with the positive-mode figure noted separately. No h-line- or i-line-isolated dose is published anywhere; both the 42 mJ/cm² image-wise and 74 mJ/cm² flood-exposure doses are broadband (365/405/436 nm) figures.", "developerFamily": "buffered-alkaline", "provenance": { "datasheetUrl": "https://www.allresist.com/wp-content/uploads/sites/2/2016/12/allresist_produktinfos_ar-u4000_englisch.pdf", "datasheetVersionOrDate": "As of January 2017", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "Document is Allresist's own 'Positive and Negative Photoresists AR-U 4000' product-information sheet (4-page spread, catalog pages 32-35), covering three grades (AR-U 4030, 4040, 4060) with separate positive-process and negative/image-reversal-process conditions plus general processing-instructions prose — a genuine manufacturer technical datasheet, not a flyer or third-party mirror. Retrieved directly from allresist.com. This recipe (ar-u-4030) extracts only the AR-U 4030 column/curve from each shared multi-grade table and figure; AR-U 4040 and AR-U 4060 numbers appearing in the same tables were read but deliberately not used here.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "az-10xt", "name": "AZ 10XT", "manufacturer": "Merck", "productLine": "AZ 10XT Series", "aliases": [ "AZ 10XT Series", "10XT" ], "tone": "positive", "chemistry": "dnq-novolak", "_chemistryNote": "This AZ 10XT technical datasheet does not itself use the words 'DNQ' or 'novolak'; it only says AZ 10XT gives 'improved sidewall profiles, aspect ratios, and photospeed vs. typical thick DNQ type materials' (p.1) and that PEB is merely 'optional' (p.12), not required. Platform classification as DNQ is corroborated by AZ's own separate P4000-series thick-film roadmap table (AZ P4620 Photoresist Data Package, p.3), which lists '10XT' with Platform = DNQ. Flagged for human QC since it rests on a cross-document read, not an explicit statement in this datasheet.", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Datasheet does not address grayscale or 3D patterning for AZ 10XT; no such use is claimed anywhere in the document.", "status": "active", "successorSlug": null, "summary": "AZ 10XT is a thick positive-tone photoresist series for plating and etch-mask applications, sold in 520/220/100 cP viscosity grades, giving single-coat films from 4 µm to over 20 µm.", "thicknessRange": { "min_um": 4, "max_um": 20, "basis": "stated", "source": "stated — p.1 APPLICATIONS bullet: 'Single coat thicknesses from 4.0 to >20µm', verbatim. The '>20µm' upper bound is inherently open-ended; max_um is recorded as 20 (the floor of the stated inequality), not a hard ceiling. A 24 µm process is separately demonstrated (p.10) but is explicitly a double coat (2 x 12 µm), not a single-coat number, so it does not contradict this single-coat spec." }, "spinCurves": [ { "label": "AZ 10XT 520cP", "points": [ { "rpm": 1000, "um": 14 }, { "rpm": 1500, "um": 11.7 }, { "rpm": 2000, "um": 10.2 }, { "rpm": 2500, "um": 9 }, { "rpm": 3000, "um": 7.8 } ], "source": "read from figure, \"SPIN CURVES (200MM SILICON)\", p.1 of AZ 10XT Series Technical datasheet (Rev. 03/21) — one of three viscosity-grade curves on the same chart, identified by its red 520cP legend entry", "figureRead": true }, { "label": "AZ 10XT 220cP", "points": [ { "rpm": 1000, "um": 9.8 }, { "rpm": 1500, "um": 7.7 }, { "rpm": 2000, "um": 6.6 }, { "rpm": 2500, "um": 5.9 }, { "rpm": 3000, "um": 5.3 } ], "source": "read from figure, \"SPIN CURVES (200MM SILICON)\", p.1 of AZ 10XT Series Technical datasheet (Rev. 03/21) — identified by its blue 220cP legend entry; consistent with the '220cps, 6µm thick film' reference process elsewhere in this document (p.3-4, p.6-7)", "figureRead": true }, { "label": "AZ 10XT 100cP", "points": [ { "rpm": 1000, "um": 6.7 }, { "rpm": 1500, "um": 5.6 }, { "rpm": 2000, "um": 5 }, { "rpm": 2500, "um": 4.4 }, { "rpm": 3000, "um": 4 } ], "source": "read from figure, \"SPIN CURVES (200MM SILICON)\", p.1 of AZ 10XT Series Technical datasheet (Rev. 03/21) — identified by its green 100cP legend entry", "figureRead": true } ], "spinNotes": "Spin curves cover 1000-3000 rpm on 200 mm silicon for three viscosity grades (520 cP, 220 cP, 100 cP), each a single, unambiguously legended line on one chart (not a multi-SKU chart, so no grade-identification ambiguity) — values above are read from the figure, not a published numeric table, and need visual QC. Coating note (p.12): the spin-curve graphs assume coating to equilibrium; thicker coats can be produced off-curve by shortening spin time and letting the film 'self level', which this document does not quantify. No dispense volume, spin ramp, or edge-bead detail is published anywhere in this datasheet.", "adhesion": { "hmds": true, "notes": "Oxide-forming substrates (e.g. Si) should be HMDS primed prior to coating AZ 10XT (PROCESS CONSIDERATIONS > SUBSTRATE PREPARATION, p.12)." }, "rehydration": "A rehydration delay of 30-60 minutes between soft bake and exposure is required for films thicker than 5.0 µm; delay time varies with film thickness and ambient humidity. The front-page process summary lists a 30-minute rehydration hold as the typical value, and the 6 µm and 12 µm reference processes each specify a 30-minute 'Post Bake Delay'; the 24 µm double-coat reference process uses a 45-minute delay. (Source: FILM REHYDRATION, p.12, and TYPICAL PROCESS (p.1) / Reference Process tables (p.3-4, p.6-7, p.9-10) of AZ 10XT Series Technical datasheet (Rev. 03/21))", "softbake": { "temp_c": 110, "time_s": 120, "method": "hotplate", "notes": "This is the reference 6 µm-film softbake, repeated identically across all four 6 µm reference processes (dense lines and holes, on Si and on Cu; p.3, p.4, p.6, p.7). The datasheet states soft bake time is 'film thickness dependent' (p.1) and that AZ 10XT soft bake temperature should generally be in the 95-110°C range (p.12, PROCESS CONSIDERATIONS > SOFT BAKE), with ramped temperatures possibly needed for very thick films to avoid solvent-outgassing bubbles. Other documented thicknesses use different bakes: 12 µm at 110°C/180 s (p.9); a 24 µm double coat at 110°C/80 s (first layer) then 115°C/180 s (second layer) (p.10).", "source": "Reference Process tables, p.3-4 and p.6-7 of AZ 10XT Series Technical datasheet (Rev. 03/21) (6 µm film on Si/Cu)" }, "exposureDose": { "doses": [ { "wavelength_nm": 365, "value_mJcm2": 380, "source": "Reference Process tables, p.3-4 of AZ 10XT Series Technical datasheet (Rev. 03/21)" } ], "datasheetBasis": "\"Expose: i-line @ 380mJ/cm2 nominal (0.48NA)\" — verbatim, Reference Process tables, dense lines and holes in 6 µm film thickness on Si", "_note": "This is the reference 6 µm-on-Si dose (both the dense-line and hole reference processes cite the identical 380 mJ/cm² i-line nominal value, p.3-4). Other i-line reference processes in this same datasheet report different doses for 6 µm-on-Cu, and the datasheet is internally inconsistent about their exact value: the Cu dense-lines process states 'i-line @ 455mJ/cm2 nominal' in its process-step table but the same page's chart headers read 'LINEARITY @ 450MJ/CM2' and 'DOF @ 450MJ/CM2' (p.6, a 450-vs-455 discrepancy within one page); the Cu holes process similarly states 'i-line @ 445mJ/cm2 nominal' in its process-step table but 'LINEARITY @ 440MJ/CM2' / 'DOF @ 440MJ/CM2' in its chart headers (p.7, a 440-vs-445 discrepancy). Those Cu-specific numbers are not reported here to avoid picking an arbitrary one of two disagreeing values; flagged for QC instead. Separately, the 24 µm double-coat process (p.10) reports 'Dose: 1875 mJ/cm2' (Ultratech 1500) and 'Dose: 1785 mJ/cm2' (Suss MA200) with no wavelength stated for either exposure — both tools are described elsewhere as 'g-h line', not confirmed i-line, so these two doses are deliberately excluded from at365_mJcm2 rather than misrouted." }, "peb": null, "floodExposure": null, "develop": { "developer": "AZ 400K", "dilution": "1:4", "time_s": 420, "method": "immersion", "rinse": null, "source": "Reference Process tables (dense lines / holes, 6 µm film, Si and Cu), p.3-4 and p.6-7 of AZ 10XT Series Technical datasheet (Rev. 03/21)", "_note": "AZ 400K 1:4 immersion 420 s is used identically in all four 6 µm reference processes and is also one of the two branches of the 24 µm process (600 s spray, Ultratech branch, p.10). The 12 µm reference process instead uses AZ 400K 1:4, 260 s spray (p.9). The other 24 µm branch (Suss MA200 aligner) uses AZ 300 MIF (TMAH, undiluted), 720 s (p.10). PROCESS CONSIDERATIONS (p.12) states AZ 10XT is compatible with either MIF (TMAH) or inorganic developers and recommends 'AZ 435MIF and AZ 400K 1:3 or AZ 400K 1:4' with no single preferred developer named; higher-normality (less dilute) developer improves photospeed but may increase CD non-uniformity and dark-film loss." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Hard bake temperature should be in the 90-100°C range (verbatim) to minimize thermal distortion of the pattern; no time is specified, so the scalar fields are left null rather than assuming one. Improves adhesion in wet-etch or plating applications and improves pattern stability in dry-etch processes.", "source": "PROCESS CONSIDERATIONS > HARD BAKE, p.12 of AZ 10XT Series Technical datasheet (Rev. 03/21)" }, "descum": null, "applications": [ "electroplating-molding", "etch-mask" ], "etchResistance": null, "liftoffSuitable": null, "platingSuitable": true, "stripper": "AZ Kwik Strip, AZ 300T, or AZ 400T (solvent-based removers), per PROCESS CONSIDERATIONS > STRIPPING, p.12 of AZ 10XT Series Technical datasheet", "storage": null, "notes": "AZ 10XT is a thick positive-tone plating resist pitched by the vendor as an upgrade over conventional thick DNQ resists in sidewall profile, aspect ratio and photospeed. Like other thick DNQ-class positive resists it needs a rehydration wait — 30-60 minutes between soft bake and exposure for any film over 5 µm — before it develops reliably; skipping it is a common source of process trouble on thick coats, and the wait grows to 45 minutes for the datasheet's 24 µm double-coat process. Post-expose bake is explicitly optional and every reference process in this datasheet runs with none, consistent with a non-chemically-amplified resist. It develops in either TMAH (AZ 300MIF/AZ 435MIF) or a buffered alkaline developer (AZ 400K, used in most of the reference processes here) — the datasheet does not name one as preferred. HMDS priming is called out explicitly for oxide-forming substrates such as silicon. Exposure dose and softbake time both scale sharply with target film thickness and substrate: for the 6 µm Cu reference processes, the datasheet's own process tables and chart headers even disagree with each other by a few mJ/cm² on the exact i-line dose (450 vs. 455, and 440 vs. 445 mJ/cm² across the two Cu processes) — close enough to be rounding noise, but a reason to verify the working dose on your own stepper rather than trust either printed figure blindly. A first-time user should follow the specific reference-process table for their target thickness and substrate rather than a single rule of thumb.", "developerFamily": "buffered-alkaline", "references": [ { "type": "paper", "title": "High-resolution projection lithography for MEMS-applications using thick photoresist AZ 10XT", "authors": "Schermer et al.", "journal": "2022 Smart Systems Integration (SSI)", "year": 2022, "doi": "10.1109/SSI56489.2022.9901437", "url": "https://doi.org/10.1109/SSI56489.2022.9901437", "accessedDate": "2026-07-12", "summary": "Thick AZ 10XT under projection (stepper) lithography for MEMS etch masks" } ], "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_10xt_photoresist.pdf", "datasheetVersionOrDate": "Rev. (03/21)", "accessedDate": "2026-07-10", "secondarySources": [] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "az-125nxt", "name": "AZ 125nXT", "manufacturer": "Merck", "productLine": "AZ 125nXT Series", "aliases": [ "AZ 125nXT Series", "AZ 125nXT-7B", "AZ 125nXT-10B" ], "tone": "negative", "chemistry": "photopolymer", "_chemistryNote": "[RESOLVED, WL decision 2026-07-10] The datasheet repeatedly calls this a \"Photopolymer Negative Tone Photoresist\" and explicitly distinguishes it from both \"conventional DNQ type materials\" and \"typical chemically amplified photoresists\". Rather than misfile it under `car` or `acrylate` — neither of which the document supports — the `photopolymer` key was ADDED to components/recipes/taxonomy.js. This follows the `bisazide-novolak` precedent (RECIPES-PLAN design decision 4): widen the vocabulary, never bend the fact.", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not discussed; the document's own applications list (TSV, plating, RIE etch) and every worked example are binary line/hole/post patterns, not grayscale profiles.", "status": "active", "successorSlug": null, "summary": "AZ 125nXT is a photopolymer negative-tone photoresist for ultra-thick single-coat films (roughly 18-120 µm across its two viscosity grades), built for Cu/Au/solder electroplating and RIE etch masks in advanced packaging, with no rehydration hold and no PEB required.", "thicknessRange": { "min_um": 18, "max_um": 120, "basis": "stated", "source": "stated - explicit THICKNESS GRADES table (p.2): AZ 125nXT-7B '~18-35µm', AZ 125nXT-10B '~35-120µm'. Combined min/max of the two stated grade ranges. (The front-page bullet separately gives a looser 'Single coat thicknesses from 20 to >100µm', p.1 - kept in notes rather than overriding the more specific grade table.)" }, "spinCurves": [ { "label": "AZ 125nXT-10B", "points": [ { "rpm": 600, "um": 120 }, { "rpm": 800, "um": 100 }, { "rpm": 1000, "um": 75 }, { "rpm": 1500, "um": 55 }, { "rpm": 1900, "um": 48 }, { "rpm": 2100, "um": 43 }, { "rpm": 2500, "um": 38 } ], "source": "read from figure, p.10 of AZ 125nXT Series Technical Datasheet (Merck, Rev. 01/24), 'COATING GUIDELINES' spin-speed chart. Two-grade chart; traces identified by the chart's own legend (filled triangle = AZ 125nXT-10B, filled circle = AZ 125nXT-7B), not by color alone.", "figureRead": true }, { "label": "AZ 125nXT-7B", "points": [ { "rpm": 600, "um": 57 }, { "rpm": 1000, "um": 35 }, { "rpm": 1300, "um": 24 }, { "rpm": 1500, "um": 21 }, { "rpm": 1900, "um": 18 }, { "rpm": 2300, "um": 14 } ], "source": "read from figure, p.10 of AZ 125nXT Series Technical Datasheet (Merck, Rev. 01/24), 'COATING GUIDELINES' spin-speed chart. Two-grade chart; traces identified by the chart's own legend (filled triangle = AZ 125nXT-10B, filled circle = AZ 125nXT-7B), not by color alone.", "figureRead": true } ], "spinNotes": "The datasheet explicitly cautions (p.10) that, unlike thin-resist spin curves, 'films will continue to thin with extended spin times' for a resist this viscous, and that the plotted curves 'may be used as general guidelines for coating films of 30µm thickness and above' - i.e. the document itself flags the low-thickness end of its own chart as unreliable. That caveat is directly visible in the data: my read of the AZ 125nXT-7B trace extends from ~57 µm (600 rpm) down to ~14 µm (2300 rpm), both outside the grade's own stated 18-35 µm range (p.2 THICKNESS GRADES table) - an internal inconsistency a QC reviewer should note, not something I resolved. A full example coating sequence with acceleration and function-per-step (dispense/spread/snap-spin/set-thickness/backside rinse-dry/edge-bead-flatten) is given on p.10 for building a real recipe, rather than a single spin-speed number.", "adhesion": { "hmds": true, "notes": "'Oxide forming substrates (Si, etc.) should be HMDS primed prior to coating AZ 125nXT. Contact your product representative for detailed information on pre-treating with HMDS.' (p.11, Substrate Preparation)." }, "rehydration": "None required - the datasheet states explicitly, in both the Typical Process summary ('Rehydration Hold: None', p.1) and in bold in the Process Considerations section ('NO POST BAKE REHYDRATION DELAYS ARE REQUIRED', p.11), that no rehydration hold is needed, unlike thick DNQ resists.", "softbake": { "temp_c": null, "time_s": null, "method": null, "notes": "Stated only as a range: '115°-140°C/5-20min' (p.1 Typical Process; footnoted 'SB time is film thickness dependent'), and again '115°-140°C range' with no time in Process Considerations (p.11). Worked examples span the whole range by thickness/grade: 115°C stepping-proximity bake up to 120s@0.002in for 20µm-7B (p.3); 130°C/13min hotplate for 50µm-10B (p.4); 120°C stepping-proximity bake up to 400s contact for 20µm-7B on a different tool (p.5); 135°C/25min for 120µm-10B (p.6). Left the scalar null per protocol rather than picking one of these.", "source": "p.1, p.11 (range); pp.3-6 (worked-example values)" }, "exposureDose": { "doses": [], "datasheetBasis": "'AZ 125nXT is sensitive in the 365-435nm wavelength range' (p.11, Exposure); Typical Process lists 'Expose: broadband sensitive' (p.1); worked-example exposures specify 'ghi mode' (combined g+h+i-line, i.e. broadband) or are captioned simply 'broadband exposure' (pp.2-3).", "_note": "365-435 nm spans i-line through g-line and is described as broadband, not a single attributable wavelength, so both at365 and at405 are left null per protocol and the value is not consolidated into value_mJcm2 either: dose is also strongly thickness/tool dependent, with different worked examples giving 1120 mJ/cm2 (Suss MA-200, ghi mode, 20µm-7B, p.3), 1800 mJ/cm2 (broadband, 15µm features in 70µm-thick film, p.2), and 2000 mJ/cm2 (Ultratech AP300 stepper, different focus, 20µm-7B, p.5) - no single general product dose is stated, so none of these was picked as 'the' value." }, "peb": null, "floodExposure": null, "develop": { "developer": "AZ 300MIF", "dilution": "0.26N (2.38%) TMAH, ready-to-use; puddle mode recommended", "time_s": null, "method": "puddle", "rinse": null, "source": "p.1, p.11 (developer/mode); pp.3-6 (worked-example times)", "_note": "Time is thickness/grade dependent: 2x25s puddles @20µm-7B (p.3), 2x30s @50µm-10B (p.4), 2x60s @20µm-7B on a different tool (p.5), 3x60s @120µm-10B (p.6). Left the scalar null per protocol; see spinNotes/notes for the pattern." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "'Hard baking (post develop bake) is generally not required with AZ 125nXT. However, hard baking may improve pattern stability in aggressive dry etch processes. Hard bake temperatures should be in the 130°-140°C range.' No time or usage threshold is given.", "source": "p.11 (Hard Bake)" }, "descum": null, "applications": [ "etch-mask", "electroplating-molding", "high-aspect-ratio" ], "etchResistance": "Described only qualitatively: 'etch resistance, chemical resistance, and thermal stability far superior to typical chemically amplified photoresists' (p.1) and 'RIE etch applications' (p.1). No etch rate or selectivity number is given anywhere.", "liftoffSuitable": null, "platingSuitable": true, "stripper": "AZ 400T, 75°C, 20-25 min with agitation (p.11, Stripping); corroborated by worked electroplating examples: 'Strip: AZ 400T @ 75C' for Cu (p.7) and 'Photoresist Strip AZ 400T @ 75°C' for Au (p.8)", "storage": null, "notes": "AZ 125nXT is a photopolymer (not conventional DNQ or the vendor's own 'typical' CAR) negative resist purpose-built for ultra-thick, high-aspect-ratio plating molds and RIE masks up to and beyond 100 µm in a single coat. Two properties make it distinct from thinner thick-film resists: no rehydration hold is needed after softbake, and a PEB is explicitly not required at all - both are called out in bold in the datasheet, in direct contrast to the classic DNQ rehydration wait this recipe library flags elsewhere. Real electroplating results are documented for Cu, Ni-compatible, and Au processes with post-plate/post-strip micrographs (pp.7-8), and the stripped resist shows no reported underplating. Coating this resist is its own discipline: the vendor explicitly warns that, unlike thin-resist spin coating, films keep thinning with extended spin time, so both spin speed and spin time must be tuned together (p.10), and the datasheet's own coating-guideline curves are only endorsed for films of 30 µm and above even though they are plotted further down. HMDS priming is required on oxide-forming substrates such as Si.", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_125nxt_serie.pdf", "datasheetVersionOrDate": "Rev. (01/24)", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "Document is Merck's own branded 'Technical datasheet' (Merck logo and copyright, '© 2024 Merck KGaA, Darmstadt, Germany and/or its affiliates', Rev. 01/24), but is hosted/mirrored on distributor MicroChemicals' site rather than merckgroup.com directly. Discovered via MicroChemicals' AZ-125nXT-10B product page (https://www.microchemicals.com/AZ-125nXT-10B-Photoresist-3.785-l/1125nXT10B), not by guessing the filename; content itself is manufacturer-authored, not a distributor summary.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "az-12xt", "name": "AZ 12XT", "manufacturer": "Merck", "productLine": "AZ 12XT-20PL Series", "aliases": [ "AZ 12XT-20PL Series", "AZ 12XT-20PL-05", "AZ 12XT-20PL-10", "AZ 12XT-20PL-15" ], "tone": "positive", "chemistry": "car", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not discussed; positioned for TSV/RDL plating and RIE etch-mask lithography with defined line/space, hole, and post patterns, not grayscale 3D profiles.", "status": "active", "successorSlug": null, "summary": "AZ 12XT is a chemically amplified positive-tone thick photoresist for single-coat films from about 3 to over 20 µm, built for fast, high-throughput TSV, RDL, and RIE etch-mask lithography with a required post-exposure bake.", "thicknessRange": { "min_um": 3, "max_um": 20, "basis": "stated", "source": "stated - 'Single coat thicknesses from 3.0 to >20µm' (p.1, Applications). The upper bound is explicitly open-ended ('>20µm') in the source text; recorded as 20 here since no higher number is stated. This closely matches the curve-span of the three spin curves below (roughly 3.0 to 22.5 µm across grades 05/10/15), which corroborates the stated range rather than conflicting with it." }, "spinCurves": [ { "label": "AZ 12XT-20PL-15", "points": [ { "rpm": 1000, "um": 22.5 }, { "rpm": 1500, "um": 17.5 }, { "rpm": 2000, "um": 14.7 }, { "rpm": 2500, "um": 13 }, { "rpm": 3000, "um": 11.7 }, { "rpm": 3500, "um": 10.7 }, { "rpm": 4000, "um": 10 } ], "source": "read from figure, p.1 of AZ 12XT-20PL Series Technical Datasheet (Merck, Rev. 03/21), 'SPIN CURVES (150MM SILICON)'. Three-grade chart; traces identified by the chart's own legend (red triangle = AZ 12XT-20PL-15, blue circle = AZ 12XT-20PL-10, green diamond = AZ 12XT-20PL-05), not by color alone.", "figureRead": true }, { "label": "AZ 12XT-20PL-10", "points": [ { "rpm": 1000, "um": 14.3 }, { "rpm": 1500, "um": 11 }, { "rpm": 2000, "um": 9.5 }, { "rpm": 2500, "um": 8.3 }, { "rpm": 3000, "um": 7.5 }, { "rpm": 3500, "um": 7 }, { "rpm": 4000, "um": 6.7 } ], "source": "read from figure, p.1 of AZ 12XT-20PL Series Technical Datasheet (Merck, Rev. 03/21), 'SPIN CURVES (150MM SILICON)'. Three-grade chart; traces identified by the chart's own legend (red triangle = AZ 12XT-20PL-15, blue circle = AZ 12XT-20PL-10, green diamond = AZ 12XT-20PL-05), not by color alone.", "figureRead": true }, { "label": "AZ 12XT-20PL-05", "points": [ { "rpm": 1000, "um": 6.2 }, { "rpm": 1500, "um": 4.8 }, { "rpm": 2000, "um": 4.2 }, { "rpm": 2500, "um": 3.8 }, { "rpm": 3000, "um": 3.5 }, { "rpm": 3500, "um": 3.2 }, { "rpm": 4000, "um": 3 } ], "source": "read from figure, p.1 of AZ 12XT-20PL Series Technical Datasheet (Merck, Rev. 03/21), 'SPIN CURVES (150MM SILICON)'. Three-grade chart; traces identified by the chart's own legend (red triangle = AZ 12XT-20PL-15, blue circle = AZ 12XT-20PL-10, green diamond = AZ 12XT-20PL-05), not by color alone.", "figureRead": true } ], "spinNotes": "QC flag: the chart's own y-axis is printed as 'Film Thickness (nm)' (p.1), which is almost certainly a labeling typo in the vendor's document - the plotted values (2-24) and the product's stated 3.0->20µm single-coat range (same page) only make sense as micrometres, and the grade names (-05/-10/-15) line up with the 5/10/15 µm worked-example film thicknesses used later in the same datasheet (pp.3-5). I read and recorded the values as µm; a human reviewer should still see the original 'nm' axis label and confirm this reading against the PDF. No numeric spin-curve table is printed elsewhere, so this figure read is the only source for spinCurves.", "adhesion": { "hmds": true, "notes": "'Oxide forming substrates (Si, etc.) should be HMDS primed prior to coating AZ 12XT.' (p.9, Substrate Preparation). Every worked process example in the document starts with an explicit 'Prime: HMDS 140°C/60s (vapor)' step (pp.3-5,8)." }, "rehydration": "None - 'Rehydration Hold: None' is stated explicitly in the Typical Process summary (p.1).", "softbake": { "temp_c": 110, "time_s": 120, "method": "hotplate", "notes": "Typical Process states a single value: '110ºC/120s' (p.1). Process Considerations separately gives a broader acceptable range, '95°-110°C' (p.9), with no time. Every worked example specifies 'direct contact hotplate' and scales time with thickness: 110°C/120s @5µm (p.3), 110°C/180s @10µm (pp.4,8), 110°C/240s @15µm (p.5). A footnoted caution applies to thicker coats: 'Thicker films may require a ramped soft bake process to avoid bubble formation due to rapid outgassing of solvents' (pp.4-5). A separate caution applies to thinner coats: films under 6µm may pick up airborne amine contamination if softbake-to-expose delay is excessive and should be exposed/developed within 30-45 minutes of softbake (p.3 footnote) - this is an amine-poisoning risk specific to CAR chemistry, not a rehydration/moisture issue, and is recorded here rather than in the rehydration field.", "source": "p.1 (Typical Process, used for temp_c/time_s above); p.9 (range); pp.3-5,8 (worked examples)" }, "exposureDose": { "doses": [], "datasheetBasis": "'AZ 12XT requires exposure energy at the 365nm wavelength.' (p.9, Exposure); Typical Process: 'Expose: 365nm sensitive' (p.1).", "_note": "Wavelength is explicitly and solely i-line (365nm), but no single general product dose is stated - every dose number in the document is tied to a specific thickness/substrate/tool combination: 100 mJ/cm2 nominal @5µm-Si, 0.48NA (p.3); 110 mJ/cm2 nominal @10µm-Si, 0.48NA (pp.4,8); 185 mJ/cm2 nominal @15µm-Si, 0.48NA (p.5); 140 mJ/cm2 @6.3µm holes on Si, 0.50NA (p.7); 250 mJ/cm2 @10µm on Cu, 0.50NA (p.7); 200 mJ/cm2 @10µm on Au, 0.50NA (p.7). Left at365 null per protocol rather than picking one of these as 'the' dose; the correct field would have been at365 had a single value existed." }, "peb": { "temp_c": 90, "time_s": 60, "notes": "PEB is REQUIRED, not optional, for this chemically amplified resist - explicitly emphasized twice: a starred footnote on the Typical Process page ('* PEB is required for proper imaging', p.1) and again in prose ('A PEB is required for proper imaging of AZ 12XT. PEB temperatures and times may be application specific. As a general rule, PEB temperatures should be in the 90° to 100°C range.', p.9). The Typical Process gives a single specific value, 90°C/60s (p.1), and every worked example in the document uses exactly that value regardless of film thickness (pp.3-5,8) - so the acceptable range (90-100°C) is recorded here in notes, but the recommended single value used throughout is what is stored in temp_c/time_s.", "source": "p.1 (Typical Process + footnote); p.9 (range); corroborated pp.3-5,8" }, "floodExposure": null, "develop": { "developer": "AZ 300MIF", "dilution": "0.26N (2.38%) TMAH, ready-to-use", "time_s": null, "method": "puddle", "rinse": null, "source": "p.1, p.9 (developer); pp.3-5,7,8 (worked-example times)", "_note": "Time is thickness/substrate dependent: 2x30s puddles @5µm-Si (p.3), 2x60s @10µm-Si (pp.4,8), 2x60s @15µm-Si (p.5), 2x45s @6.3-10µm on Si/Cu/Au substrates (p.7). Left the scalar null per protocol; see notes for the full pattern." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "'Hard baking (post develop bake) improves adhesion in wet etch or plating applications and improves pattern stability in dry etch processes. Hard bake temperatures should be in the 100° to 115°C range to ensure minimal thermal distortion of the pattern.' No time or usage threshold is stated. A comparison figure (p.9) shows sidewall stability for 10µm lines (6.5µm film) at no-bake vs 105°C/110°C/115°C hard bake, without recommending a single condition.", "source": "p.9 (Hard Bake)" }, "descum": null, "applications": [ "etch-mask", "electroplating-molding", "high-aspect-ratio" ], "etchResistance": "Described only qualitatively: 'Excellent for Through Silicon Via (TSV), plating, and RIE etch applications' (p.1); hard bake (100-115°C) is said to improve 'pattern stability in dry etch processes' (p.9). No etch rate or selectivity number is given.", "liftoffSuitable": null, "platingSuitable": true, "stripper": "AZ Kwik Strip, AZ 300T, or AZ 400T (solvent-based removers, all three explicitly recommended, p.9)", "storage": null, "notes": "AZ 12XT is a chemically amplified positive-tone counterpart to the vendor's negative nXT line, aimed at the same TSV/RDL plating-mask and RIE etch-mask space but developing and exposing faster for higher tool throughput. Unlike the thick DNQ resists this recipe library also covers, no rehydration hold is needed - but the tradeoff is a PEB that is explicitly required (not optional) for the latent image to develop at all, and a real amine-sensitivity window: coats thinner than 6 µm must be exposed and developed within 30-45 minutes of softbake or airborne base contamination can degrade the pattern, a classic chemically-amplified-resist failure mode distinct from DNQ rehydration. Exposure dose and develop time both scale strongly with film thickness across the worked examples (100-250 mJ/cm2; 2x30s to 2x60s puddles), so there is no single number to quote - see the exposureDose and develop _note fields for the full table. HMDS priming is required on Si. Hard bake is optional and helps adhesion/etch stability rather than being required for imaging.", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_12xt_photoresist.pdf", "datasheetVersionOrDate": "Rev. (03/21)", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "Document is Merck's own branded 'Technical datasheet' (Merck logo and copyright, '© 2021 Merck KGaA, Darmstadt, Germany and/or its affiliates', Rev. 03/21), but is hosted/mirrored on distributor MicroChemicals' site rather than merckgroup.com directly. Discovered via MicroChemicals' AZ-12XT-20PL-10 product page (https://www.microchemicals.com/AZ-12XT-20PL-10-Photoresist-3.785-l/1A012XT1000), not by guessing the filename; content itself is manufacturer-authored, not a distributor summary.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "az-1505", "name": "AZ 1505", "manufacturer": "Merck (AZ Electronic Materials)", "productLine": "AZ 1500 series", "aliases": [ "AZ1505", "AZ 1505 Photoresist" ], "tone": "positive", "chemistry": "dnq-novolak", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Datasheet does not address grayscale or 3D lithography anywhere; AZ 1500 is marketed only for standard binary patterning (wet-etch masking, general purpose).", "status": "active", "successorSlug": null, "summary": "AZ 1505 is the thinnest grade in Merck's AZ 1500 series of general-purpose positive-tone photoresists, spin-coating to roughly 0.65-1.6 µm and intended for wet-etch masking applications requiring strong substrate adhesion.", "thicknessRange": { "min_um": 0.6, "max_um": 1.6, "basis": "curve-span", "source": "curve-span: taken from the min/max of the AZ 1505 curve on the p.1 spin-speed chart (500-4000 rpm tested); the datasheet's stated '0.5 to 6µm' figure (p.1, APPLICATION bullets) is a whole-series span across all four grades (1505/1512/1518/1529), not a per-grade figure for AZ 1505 alone, so it was not used here. ADJUDICATED 2026-07-12 (third pixel read, WL directive: figure over table): min updated from 0.65 to 0.6 um to match the corrected curve span after the 3000/3500/4000 rpm points were fixed by a fresh pixel read (see spinCurves[0].source); max (1.6 um at 500 rpm) is unchanged." }, "spinCurves": [ { "label": "AZ 1505 as supplied", "points": [ { "rpm": 500, "um": 1.6 }, { "rpm": 1000, "um": 1.1 }, { "rpm": 1500, "um": 0.95 }, { "rpm": 2000, "um": 0.85 }, { "rpm": 2500, "um": 0.8 }, { "rpm": 3000, "um": 0.65 }, { "rpm": 3500, "um": 0.6 }, { "rpm": 4000, "um": 0.6 } ], "source": "read from figure, \"SPIN CURVES (150mm Wafers)\" chart, p.1 of AZ 1500 Series Technical Datasheet (Merck, Rev. 03/21); AZ 1505 identified by its blue diamond legend marker, distinct from AZ 1512 (magenta square), AZ 1518 (red triangle) and AZ 1529 (green circle) plotted on the same axes. ADJUDICATED 2026-07-12 (third pixel read, WL directive: figure over table): the 3000/3500 rpm points were the crawl-era eyeball read and disagreed ~17% with the independent cross-check extraction; a fresh PyMuPDF pixel read of the embedded chart raster (axis calibrated from the y-axis tick-label text centroids -- value0 row=251, value10 row=51, 20 px/unit -- and the x-axis tick-label centroids -- rpm0 x=74.3, 0.1048 px/rpm; marker centroid taken as the column of maximum vertical run-length within +-7 px of each expected rpm column, i.e. the diamond's widest point, not the connecting line) reads 3000 rpm = 0.65 um, 3500 rpm = 0.60 um, 4000 rpm = 0.60 um -- confirming the crawl-era raw read erred (too high) and the cross-check extraction was essentially correct; 4000 rpm was also nudged from 0.65 to 0.60 um (not itself flagged as a >15% mismatch) because leaving it unchanged made the curve non-monotonic against the corrected 3500 rpm point, and the fresh pixel read independently supports 0.60 um there too (same marker column height/position as 3500 rpm).", "figureRead": true } ], "spinNotes": "Datasheet does not state spin ramp/acceleration, dispense volume, or static vs dynamic dispense for the spin-curve chart. Spin coating is named as one of several compatible coating methods (spray and roller coating are also mentioned, p.6 COATING) but no method-specific parameters are published beyond that. No edge-bead-removal recipe is given, only that AZ EBR Solvent or AZ EBR 70/30 are the companion EBR products (p.2).", "adhesion": { "hmds": true, "notes": "\"Oxide forming substrates (Si, etc.) should be primed with HMDS (hexamethyl disilazane) or other suitable primer prior to coating AZ 1500. Contact your AZ products representative for detailed information on pre-treating with HMDS.\" Source: SUBSTRATE PREPARATION, p.6." }, "rehydration": null, "softbake": { "temp_c": null, "time_s": null, "method": null, "notes": "Datasheet states soft bake temperature should be in the 90-110°C range (higher end improves adhesion to metals); bakes may be performed on hotplate or in a vented oven, but no single temperature, no time, and no specific method is given for AZ 1505.", "source": "TYPICAL PROCESS, p.1; PROCESS CONSIDERATIONS / SOFT BAKE, p.6" }, "exposureDose": { "doses": [], "datasheetBasis": "AZ 1500 series stated sensitive to exposure wavelengths 310-450 nm; 365-436 nm recommended (broadband exposure, no single wavelength or dose specified for the series in general).", "_note": "No exposure dose (mJ/cm²) is published for AZ 1505 anywhere in this datasheet. The only doses printed in the document (80-100 mJ/cm² for AZ 1512, 130-150 mJ/cm² for AZ 1518) come from resolution/depth-of-focus test figures captioned for those two other grades specifically, not AZ 1505 — not applied here." }, "peb": { "temp_c": null, "time_s": null, "notes": "PEB is described as optional, used to maximize process latitude and mitigate standing-wave effects from monochromatic exposure. When used, temperature should be in the 105-115°C range; this is stated as a range, not a single value, and no time is given.", "source": "TYPICAL PROCESS, p.1; POST EXPOSURE BAKE, p.6" }, "floodExposure": null, "develop": { "developer": "AZ 300MIF, AZ 726MIF, AZ 917MIF, or AZ 400K", "dilution": "AZ 400K used 1:4 for tank immersion; dilution of the MIF developers is not stated in this datasheet.", "time_s": 60, "method": null, "rinse": null, "source": "TYPICAL PROCESS, p.1; DEVELOPING, p.6", "_note": "Datasheet gives develop as \"60s Puddle or immersion\" — both methods are explicitly offered for the series, so a single enum value would misrepresent it; left null and both stated in notes/dilution fields instead. AZ 400K 1:4 is recommended for tank immersion and AZ 917MIF for puddle developing; AZ 300MIF appears elsewhere in the document but only in test captions for AZ 1512/1518, not tied to AZ 1505 specifically. developerFamily tmah-or-buffered-alkaline: the datasheet expressly offers AZ 300/726/917MIF (TMAH-based, metal-ion-free) or AZ 400K (buffered KOH) as peer developer options for AZ 1505 (classified 2026-07-12)." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Hard bake temperature should be in the 100-110°C range to minimize thermal pattern distortion; improves adhesion in wet-etch or plating applications and pattern stability in dry etch processes. No time is given, and temperature is stated as a range, not a single value.", "source": "HARD BAKE, p.6" }, "descum": null, "applications": [ "etch-mask", "electroplating-molding", "general-prototyping" ], "etchResistance": "Marketed for \"demanding wet etch applications\" with \"excellent substrate adhesion\"; no quantitative etch rate or resistance data is published in this datasheet. Source: APPLICATION, p.1.", "liftoffSuitable": null, "platingSuitable": true, "stripper": "AZ 300T, AZ 400T, or AZ Kwik Strip — removers designed for DNQ/novolac type photoresists. Strip times vary with thermal history; patterns processed above 140°C may cross-link and become harder to strip, and charred resist will not dissolve in solvent-based removers. Source: STRIPPING, p.7.", "storage": null, "notes": "AZ 1505 is the thinnest member of Merck's AZ 1500 series, a general-purpose positive DNQ/novolak-type resist family sold for demanding wet-etch masking where strong substrate adhesion matters (the datasheet's stripper guidance — removers \"designed for DNQ/novolac type photoresists\" — is the basis for this chemistry classification, since the document never states the chemistry as a standalone claim). Soft-bake, optional-PEB, and hard-bake temperatures are all given only as ranges (90-110°C, 105-115°C, 100-110°C respectively) rather than single set points, so a specific process must be optimized on-tool rather than read off this sheet verbatim. No exposure dose, PEB time, or storage/shelf-life data is published for AZ 1505 specifically — the only doses and resolution/depth-of-focus data in this datasheet were captured for the AZ 1512 and AZ 1518 grades, not AZ 1505. Develop is 60 s by puddle or immersion in a metal-ion-free (TMAH, e.g. AZ 300/726/917MIF) or inorganic (e.g. AZ 400K 1:4) developer, both explicitly supported. HMDS (or another suitable primer) is recommended ahead of coating on oxide-forming substrates such as silicon.", "developerFamily": "tmah-or-buffered-alkaline", "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_1500_series.pdf", "datasheetVersionOrDate": "Rev. (03/21)", "accessedDate": "2026-07-10", "secondarySources": [] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "az-1512", "name": "AZ 1512", "manufacturer": "Merck", "productLine": "AZ 1500 series", "aliases": [ "AZ1512", "AZ 1512HS", "AZ1512HS" ], "tone": "positive", "chemistry": "dnq-novolak", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed anywhere in the datasheet; the series is positioned as a general-purpose wet-etch/plating resist, not for grayscale/3D lithography.", "status": "active", "successorSlug": null, "summary": "AZ 1512 is the mid-thickness grade in Merck's AZ 1500 series of general-purpose positive DNQ/novolac photoresists, offering fast throughput and compatibility with both TMAH (metal-ion-free) and inorganic developers for wet-etch and plating applications.", "thicknessRange": { "min_um": 1.45, "max_um": 3.9, "basis": "curve-span", "source": "curve-span: min/max of the AZ 1512 trace on the p.1 'SPIN CURVES (150mm Wafers)' chart (500-4000 rpm plotted span; 4000 rpm→1.45 µm, 500 rpm→3.9 µm). The datasheet's '0.5 to 6µm' APPLICATION line is a whole-series span across all four grades, not per-grade, so it was not used (same ruling as az-1505/az-1518). Adjudicated 2026-07-12." }, "spinCurves": [ { "label": "AZ 1512", "points": [ { "rpm": 500, "um": 3.9 }, { "rpm": 1000, "um": 2.7 }, { "rpm": 1500, "um": 2.3 }, { "rpm": 2000, "um": 2 }, { "rpm": 2500, "um": 1.8 }, { "rpm": 3000, "um": 1.65 }, { "rpm": 3500, "um": 1.55 }, { "rpm": 4000, "um": 1.45 } ], "source": "read from figure, \"SPIN CURVES (150mm Wafers)\", p.1 of AZ 1500 Series datasheet (Merck, Rev. (03/21)); trace identified as the magenta/pink square-marker series per the chart's own legend (AZ 1505 blue diamond / AZ 1512 magenta square / AZ 1518 red triangle / AZ 1529 green circle), read at each of the 8 plotted marker positions from 500-4000 rpm (the full plotted span; chart x-axis extends to 5000 rpm but no series has a point past 4000); anchor check: the AZ 1512-specific resolution/DOF/exposure-latitude test figures (p.3-5) all cite FT=1.30µm, modestly below this reading's highest-rpm point (~1.45µm at 4000 rpm), consistent with that test film having been coated at a speed slightly above this chart's plotted ceiling rather than a misread; digitized 2026-07-12", "figureRead": true } ], "spinNotes": "Multi-grade chart (\"SPIN CURVES (150mm Wafers)\", p.1) plots AZ 1505 / AZ 1512 / AZ 1518 / AZ 1529 together with distinct markers (diamond/square/triangle/circle). The AZ 1512 trace (magenta/pink square) has now been digitized directly from this chart at its 8 plotted marker positions, 500-4000 rpm — see spinCurves. The AZ 1505 and AZ 1512 traces run visually close together on the compressed 0-12µm y-axis at rpm ≥ 2000 despite differing by roughly 2x in absolute thickness there — exactly the kind of overlap that has produced misread spin data elsewhere in this project, so the two series were distinguished carefully by marker shape/colour against the legend, not by proximity or thickness-ordering assumptions. No spin accel/dispense parameters are published. AZ EBR Solvent / AZ EBR 70/30 are listed as companion products for thinning/edge-bead removal (COMPANION PRODUCTS, p.2) but no edge-bead procedure or parameters are given. No rehydration hold is mentioned (not applicable to this thin-film series).", "adhesion": { "hmds": true, "notes": "\"Oxide forming substrates (Si, etc.) should be primed with HMDS (hexamethyl disilazane) or other suitable primer prior to coating AZ 1500. Contact your AZ products representative for detailed information on pre-treating with HMDS.\" (SUBSTRATE PREPARATION, p.6). No specific HMDS bake temp/time is published." }, "rehydration": null, "softbake": { "temp_c": null, "time_s": null, "method": "hotplate", "notes": "Series-wide range only: \"Soft bake temperatures for AZ 1500 should be in the 90-110°C range. Temperatures towards the high end of this range will improve adhesion to metals.\" No time is published; \"optimum soft bake times and temperatures may be application specific.\" Bakes may be performed on hotplates or in vented ovens — method left as hotplate (the more common of the two, and what all downstream resolution/DOF/latitude test figures use: \"Soft Bake: 100°C/90s (hotplate)\"), but that 100°C/90s figure is a test condition for the resolution figures, not the datasheet's stated typical-process recommendation, so it is not promoted to the scalar fields here.", "source": "SOFT BAKE, p.6 of AZ 1500 Series datasheet; test-condition caption on p.3–4." }, "exposureDose": { "doses": [], "datasheetBasis": "\"AZ 1500 is sensitive to exposure wavelengths between 310 and 450nm. 365-436nm is recommended.\" (EXPOSURE, p.6). All resolution/DOF/exposure-latitude test figures for AZ 1512 explicitly state \"G-line exposure\" (436 nm).", "_note": "No single nominal dose is stated for AZ 1512. The datasheet only gives sweep points from test figures: resolution at 80/90/100 mJ/cm² (FT=1.3µm, p.3), depth-of-focus at 90/100 mJ/cm² (p.4), and exposure latitude swept 70–140 mJ/cm² (p.5) — all under G-line (436 nm) exposure. A range/sweep is not a value per the extraction rules, so at365/at405/value all stay null. G-line is neither i-line (365 nm) nor h-line (405 nm), so even if a single value existed it would route to value_mJcm2, not at365/at405." }, "peb": { "temp_c": null, "time_s": null, "notes": "Optional step: \"A PEB may be employed to maximize process latitudes and mitigate standing wave effects caused by monochromatic exposure. PEB temperatures and times may be application specific. As a general rule, PEB temperatures should be in the 105°C to 115°C range.\" No time is published.", "source": "POST EXPOSURE BAKE, p.6 of AZ 1500 Series datasheet." }, "floodExposure": null, "develop": { "developer": "AZ 400K or AZ 300MIF (tank immersion); AZ 917MIF (puddle)", "dilution": "AZ 400K 1:4; AZ 300MIF and AZ 917MIF used as supplied (both metal-ion-free/ready-to-use)", "time_s": 60, "method": null, "rinse": null, "source": "TYPICAL PROCESS, p.1 (\"Develop: 60s Puddle or immersion Developer type: MIF or IN\"); DEVELOPING, p.6; developer list, COMPANION PRODUCTS, p.2.", "_note": "The datasheet offers both puddle and immersion processing, each with a different recommended developer, and states a flat 60 s develop time for either — no single developer/method pair is designated as primary, so method is left null and both options are quoted verbatim. developerFamily tmah-or-buffered-alkaline: AZ 400K (tank immersion, buffered KOH) or AZ 300MIF (tank immersion, TMAH-based) or AZ 917MIF (puddle, TMAH-based) are all expressly offered as peer developer options (classified 2026-07-12)." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "\"Hard bake temperatures should be in the 100°C to 110°C range to ensure minimal thermal distortion of the pattern.\" No time is published. Improves adhesion in wet-etch/plating applications and pattern stability in dry-etch processes.", "source": "HARD BAKE, p.6 of AZ 1500 Series datasheet." }, "descum": null, "applications": [ "etch-mask", "electroplating-molding" ], "etchResistance": "Datasheet claims \"excellent substrate adhesion for demanding wet etch applications\" (APPLICATION, p.1) but publishes no etch-rate or selectivity data.", "liftoffSuitable": null, "platingSuitable": true, "stripper": "AZ 300T, AZ 400T, or AZ Kwik Strip™ (STRIPPING, p.7): \"Under normal process conditions, AZ 1500 strips readily in removers designed for DNQ/novolac type photoresists.\" Patterns baked above 140°C may cross-link and become harder to strip.", "storage": null, "notes": "AZ 1512 is the ~1.3 µm-thick grade in Merck's AZ 1500 series of general-purpose positive DNQ/novolac photoresists, sharing a single four-grade product line (1505/1512/1518/1529) with a common process window. The series is compatible with both TMAH (metal-ion-free) and inorganic developers, and a puddle or immersion develop is quoted at a flat 60 s regardless of method. Resolution, depth-of-focus and exposure-latitude figures are all reported under g-line (436 nm) exposure rather than i-line, and only as multi-point sweeps (e.g. a 70–140 mJ/cm² latitude scan) rather than a single nominal dose — so no dose is published here. Soft bake, PEB and hard bake are each specified only as series-wide temperature ranges (90–110 °C, 105–115 °C, 100–110 °C) with no grade-specific values or times.", "developerFamily": "tmah-or-buffered-alkaline", "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_1500_series.pdf", "datasheetVersionOrDate": "Rev. (03/21)", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "The archived PDF is Merck's own \"Technical datasheet AZ 1500 Series\" (© 2021 Merck KGaA), retrieved from a MicroChemicals-hosted mirror of that Merck document — not a MicroChemicals-authored sheet. [RENAMED during audit, WL decision 2026-07-10] This recipe was extracted under the slug `az-1512hs`, but the document's legend and body text say \"AZ 1512\" throughout and the \"HS\" suffix appears nowhere in it. The recipe is therefore named after the product the document actually names. \"AZ 1512HS\" is retained as an alias; if it is a genuinely distinct high-speed variant it needs its own datasheet and its own recipe.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": true }, { "slug": "az-1518", "name": "AZ 1518", "manufacturer": "Merck (AZ Electronic Materials)", "productLine": "AZ 1500 series", "aliases": [ "AZ1518", "AZ 1518 Photoresist" ], "tone": "positive", "chemistry": "dnq-novolak", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Datasheet does not address grayscale or 3D lithography anywhere; AZ 1500 is marketed only for standard binary patterning (wet-etch masking, general purpose).", "status": "active", "successorSlug": null, "summary": "AZ 1518 is a mid/thick grade in Merck's AZ 1500 series of general-purpose positive-tone photoresists, spin-coating to roughly 1.9-5.6 µm; this datasheet documents it specifically at a 2.4 µm reference film thickness with resolution and depth-of-focus data.", "thicknessRange": { "min_um": 1.9, "max_um": 5.6, "basis": "curve-span", "source": "curve-span: taken from the min/max of the AZ 1518 curve on the p.1 spin-speed chart (500-4000 rpm tested); the datasheet's stated '0.5 to 6µm' figure (p.1, APPLICATION bullets) is a whole-series span across all four grades (1505/1512/1518/1529), not a per-grade figure for AZ 1518 alone, so it was not used here." }, "spinCurves": [ { "label": "AZ 1518 as supplied", "points": [ { "rpm": 500, "um": 5.6 }, { "rpm": 1000, "um": 3.9 }, { "rpm": 1500, "um": 3.3 }, { "rpm": 2000, "um": 2.9 }, { "rpm": 2500, "um": 2.6 }, { "rpm": 3000, "um": 2.3 }, { "rpm": 3500, "um": 2.1 }, { "rpm": 4000, "um": 1.9 } ], "source": "read from figure, \"SPIN CURVES (150mm Wafers)\" chart, p.1 of AZ 1500 Series Technical Datasheet (Merck, Rev. 03/21); AZ 1518 identified by its red triangle legend marker, distinct from AZ 1505 (blue diamond), AZ 1512 (magenta square) and AZ 1529 (green circle) plotted on the same axes. Cross-checked for plausibility against the reference thickness (2.40 µm film) named in the p.2-4 resolution/DOF figures, which sits between the 1000 and 1500 rpm figure-read points, consistent with a typical process spin speed in that band.", "figureRead": true } ], "spinNotes": "Datasheet does not state spin ramp/acceleration, dispense volume, or static vs dynamic dispense for the spin-curve chart. Spin coating is named as one of several compatible coating methods (spray and roller coating are also mentioned, p.6 COATING) but no method-specific parameters are published beyond that. No edge-bead-removal recipe is given, only that AZ EBR Solvent or AZ EBR 70/30 are the companion EBR products (p.2).", "adhesion": { "hmds": true, "notes": "\"Oxide forming substrates (Si, etc.) should be primed with HMDS (hexamethyl disilazane) or other suitable primer prior to coating AZ 1500. Contact your AZ products representative for detailed information on pre-treating with HMDS.\" Source: SUBSTRATE PREPARATION, p.6." }, "rehydration": null, "softbake": { "temp_c": 100, "time_s": 90, "method": "hotplate", "notes": "100°C/90s hotplate soft bake is the condition used for the AZ 1518 resolution ladder (p.3, FT=2.4 µm on Si) and the AZ 1518 depth-of-focus series (p.4, FT=2.40 µm on Si). The series-wide TYPICAL PROCESS section (p.1) only states a 90-110°C range without a single value or time, so this is a specific documented test condition for AZ 1518 rather than the generic series spec.", "source": "RESOLUTION OF AZ 1518 at FT=2.4µm on Si, p.3; DEPTH OF FOCUS FOR 2.0µM LINES AZ 1518 AT FT=2.40µM ON SI, p.4" }, "exposureDose": { "doses": [ { "wavelength_nm": 436, "value_mJcm2": 150, "source": "p.2 SEM image caption; p.3 RESOLUTION OF AZ 1518 at FT=2.4µm; p.4 DEPTH OF FOCUS FOR 2.0µM LINES AZ 1518" } ], "datasheetBasis": "g-line (436 nm) exposure — verbatim from the p.2 SEM caption: \"AZ 1518 Photoresist, 1.0µm lines in 2.40µm film, 150mJ/cm2 g-line exposure, AZ 300 MIF Develop (60s)\". The series-wide EXPOSURE section (p.6) separately states AZ 1500 is sensitive 310-450 nm with 365-436 nm recommended (broadband, no single wavelength).", "_note": "150 mJ/cm² is the dose captioned for the highlighted SEM cross-section on p.2. It is one point within a wider dose sweep documented elsewhere for AZ 1518: 130/140/150 mJ/cm² in the RESOLUTION OF AZ 1518 figure (p.3) and 140/150 mJ/cm² in the DEPTH OF FOCUS figure (p.4) — those sweep values were not written into the single-value field to avoid false precision; only the one explicitly captioned headline dose is recorded. Because this dose is attributed to g-line (436 nm), not i-line (365 nm) or h-line (405 nm), it is intentionally placed in value_mJcm2 and NOT copied into at365_mJcm2 or at405_mJcm2, per the no-cross-wavelength rule." }, "peb": { "temp_c": null, "time_s": null, "notes": "PEB is described as optional (series-wide statement), used to maximize process latitude and mitigate standing-wave effects from monochromatic exposure. When used, temperature should be in the 105-115°C range; this is stated as a range, not a single value, and no time is given. No AZ 1518-specific PEB condition is documented anywhere in this datasheet.", "source": "TYPICAL PROCESS, p.1; POST EXPOSURE BAKE, p.6" }, "floodExposure": null, "develop": { "developer": "AZ 300MIF", "dilution": null, "time_s": 60, "method": "puddle", "rinse": null, "source": "p.2 SEM caption (\"AZ 300 MIF Develop (60s)\"); p.4 DEPTH OF FOCUS caption (\"Develop: AZ 300MIF (60s) puddle\"); p.3 resolution caption (\"Develop: AZ 300MIF (60s)\")", "_note": "AZ 300MIF is the specific developer named in every AZ 1518 test caption in this datasheet, and the p.4 caption specifically states \"puddle\" (the p.3 caption does not restate the method, assumed identical to p.4 since both describe the same FT=2.4µm AZ 1518 test series). Dilution of AZ 300MIF is not stated anywhere. The series-wide TYPICAL PROCESS section (p.1) separately allows \"60s Puddle or immersion\" with \"MIF or IN\" developer type in general — that broader statement is not developer-specific to AZ 1518." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Hard bake temperature should be in the 100-110°C range to minimize thermal pattern distortion; improves adhesion in wet-etch or plating applications and pattern stability in dry etch processes. No time is given, and temperature is stated as a range, not a single value. No AZ 1518-specific hard-bake condition is documented.", "source": "HARD BAKE, p.6" }, "descum": null, "applications": [ "etch-mask", "electroplating-molding", "general-prototyping" ], "etchResistance": "Marketed for \"demanding wet etch applications\" with \"excellent substrate adhesion\"; no quantitative etch rate or resistance data is published in this datasheet. Source: APPLICATION, p.1.", "liftoffSuitable": null, "platingSuitable": true, "stripper": "AZ 300T, AZ 400T, or AZ Kwik Strip — removers designed for DNQ/novolac type photoresists. Strip times vary with thermal history; patterns processed above 140°C may cross-link and become harder to strip, and charred resist will not dissolve in solvent-based removers. Source: STRIPPING, p.7.", "storage": null, "notes": "AZ 1518 is a mid-to-thick grade in Merck's AZ 1500 series, a general-purpose positive DNQ/novolak-type resist sold for wet-etch masking needing strong adhesion (the datasheet's stripper guidance — removers \"designed for DNQ/novolac type photoresists\" — is the basis for this chemistry classification, since the document never states the chemistry as a standalone claim). Unlike AZ 1505, this datasheet documents AZ 1518 with dedicated resolution and depth-of-focus figures at a 2.4 µm reference film thickness: 100°C/90s hotplate soft bake, g-line exposure on a Nikon 1755G7A (0.54 NA) stepper, and a 60 s AZ 300MIF puddle develop, resolving down to roughly 1.0-1.2 µm lines and holding pattern fidelity across a documented ±0.8-1.2 µm focus window. PEB and hard-bake conditions, by contrast, are given only as series-wide ranges (105-115°C optional PEB; 100-110°C hard bake), not single values specific to AZ 1518. On the p.1 multi-grade spin-speed chart (AZ 1505/1512/1518/1529 plotted together), the AZ 1518 curve was identified by its red triangle legend marker; the reported points are a figure read, not a printed table, and warrant a visual QC pass against the source chart.", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_1500_series.pdf", "datasheetVersionOrDate": "Rev. (03/21)", "accessedDate": "2026-07-10", "secondarySources": [] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "az-15nxt", "name": "AZ 15nXT (450 CPS)", "manufacturer": "AZ Electronic Materials", "productLine": null, "aliases": [ "AZ15nXT", "AZ 15nXT (450cps)", "AZ EXP 15nXT" ], "tone": "negative", "chemistry": "car", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not discussed anywhere in the document; the material is positioned for binary Cu RDL / TSV plating and etch masks (via fill, line/space patterns), not grayscale 3D profiles.", "status": "active", "successorSlug": null, "summary": "AZ 15nXT (450 CPS) is a chemically amplified negative-tone thick photoresist for Cu redistribution-layer and through-silicon-via plating, exposed with i-line steppers and developed in standard TMAH developer.", "thicknessRange": { "min_um": 6.4, "max_um": 14.6, "basis": "curve-span", "source": "curve-span - min/max of the 6 marked data points (1000-4000 rpm) on the document's own 'AZ 15nXT (450 CPS) Spin Speed Curve' figure, p.7. Not extrapolated to the unmarked ends of the drawn line (500 rpm/~22 µm, 4500 rpm/~5.8 µm). See spinNotes for a real internal inconsistency with two other thickness figures in the same document." }, "spinCurves": [ { "label": "AZ 15nXT (450 CPS)", "points": [ { "rpm": 1000, "um": 14.6 }, { "rpm": 1500, "um": 11.3 }, { "rpm": 2000, "um": 9.4 }, { "rpm": 2500, "um": 8.3 }, { "rpm": 3000, "um": 7.4 }, { "rpm": 4000, "um": 6.4 } ], "source": "read from figure, p.7 of AZ 15nXT (450 CPS) Photoresist - Lithographic and Plating Performance Comparison (AZ Electronic Materials, Jan 2009), titled 'AZ 15nXT (450 CPS) Spin Speed Curve'. Single-grade chart (one trace, no legend ambiguity); values estimated at the six marked data points and cross-checked against the same trace redrawn with square markers on the comparison chart p.8 ('AZ 15nXT (115 CPS) and AZ 15nXT (450 CPS) Spin Speed Curves'), which shows consistent point positions.", "figureRead": true } ], "spinNotes": "Coating: hand dispense on 150 mm silicon, Opti-Trak Coat and Bake, spin 1000-4000 rpm for 30 sec, softbake 110°C/3 min integrated with the coat track (p.7-8). EDGE BEAD is severe and explicitly studied (p.15, 200 mm wafer): at 1000 rpm the film rises from a ~17.8 µm field thickness to ~25 µm at 1 mm from the wafer edge; even at 3000 rpm the edge rises from ~7.8 µm field to ~9.6 µm at 1 mm from the edge. Field thickness stabilizes by roughly 5-6 mm in from the edge at every speed tested. REHYDRATION is not mentioned anywhere in this document (unlike AZ 125nXT and AZ 12XT, which both explicitly state 'Rehydration Hold: None') - left the schema's rehydration field null rather than assuming 'not required' just because this is a chemically amplified resist. INTERNAL INCONSISTENCY for QC: the primary 'Spin Speed Curve' (p.7, used above) reads ~14.6 µm at 1000 rpm and ~6.4 µm at 4000 rpm, but two other figures in the same deck disagree - the coating-uniformity wafer-map means (p.14) report 17.8 µm at 1000 rpm / 7.8 µm at 3000 rpm, and the edge-bead study's own field-thickness values (p.15) match those higher numbers, not the p.7 curve. That is roughly a 20% gap at 1000 rpm between two nominally equivalent 'hand dispense, Opti-Trak coat and bake, 150 mm Si' runs in the same document. I used the p.7 curve for spinCurves because it is the figure explicitly captioned as the spin-speed curve, but a human reviewer should know real coats in this deck ran measurably thicker in two other places.", "adhesion": { "hmds": null, "notes": "Not discussed. Process examples in the document run directly on Si (photospeed testing) and Cu (imaging) wafers with no HMDS priming step mentioned." }, "rehydration": null, "softbake": { "temp_c": 110, "time_s": 180, "method": null, "notes": "Stated as a single value (not a range) in Process Conditions: '110°C / 180 seconds' (=110°C/3 min), and repeated identically in every other process example in the deck (pp.7,8,16-19). Processed on an 'Opti-Trak Coat and Bake' track; the document never states whether this is a hotplate or oven step, so method is left null.", "source": "p.5 (Process Conditions); corroborated pp.7,8,11,16-19" }, "exposureDose": { "doses": [ { "wavelength_nm": 365, "value_mJcm2": 400, "source": "p.5; corroborated as the boxed/highlighted reference dose in Exposure Latitude figures pp.11,22" } ], "datasheetBasis": "Exposure tool: ASML (i-line); Dose = 400 ± 50 mJ/cm2; Focus: 1 ± 0.5 µm (p.5, Process Conditions). Later figures confirm 'ASML 0.48 NA 0.55σ i-Line' stepper (pp.16-19) and 'ASML i-Line Stepper' (pp.11-13).", "_note": "400 mJ/cm2 is the document's own nominal/reference dose (with a stated ± 50 mJ/cm2 tolerance) - it is the dose value boxed/highlighted as the reference condition in the Exposure Latitude figures (pp.11,22) and used as the anchor for all delay-study CD data (pp.16-19), so it is used here per the protocol's 'typical/recommended/nominal' exception rather than being treated as an unusable range." }, "peb": { "temp_c": 120, "time_s": 60, "notes": "Consistent 120°C/60 s across every process example in the document (pp.5,11,16-19); no alternate range is stated (unlike softbake).", "source": "p.5 (Process Conditions); corroborated pp.11,16-19" }, "floodExposure": null, "develop": { "developer": "AZ 300 MIF", "dilution": "2.38% TMAH (AZ 300 MIF, ready-to-use)", "time_s": null, "method": "puddle", "rinse": null, "source": "p.5 (Process Conditions); corroborated 'AZ 300 MIF/3x50sec Spray/Puddle @23°C' pp.11,16-19", "_note": "Datasheet specifies 3 x 50 second spray/puddle steps (150 s total contact time) at 23°C, identically across every example in the document (pp.5,11,16-19). Recorded as a repeated-step spec in this note rather than forcing it into the single develop.time_s scalar." }, "hardbake": null, "descum": null, "applications": [ "etch-mask", "electroplating-molding", "high-aspect-ratio" ], "etchResistance": "Cover page title lists 'Cu RDL, TSV, and other plating & etch applications' but the document gives no etch rate, selectivity, or other quantified etch-resistance data anywhere.", "liftoffSuitable": null, "platingSuitable": true, "stripper": "AZ Kwik Strip, 70°C, 3 min (p.21 Ni/Cu Plating Compatibility; corroborated p.23 Summary: 'stripped completely in AZ Kwik Strip at 70°C for 3 min')", "storage": null, "notes": "AZ 15nXT (450 CPS) is a chemically amplified negative-tone resist built specifically for Cu RDL and TSV plating/etch masks, not a general-purpose thick film. Because it is CAR chemistry, the post-exposure bake (120°C/60 s here) is the step that actually defines the pattern - unlike a DNQ resist, exposure alone does not develop the image. The document reports both exposure-to-PEB and coat-to-exposure delay studies on copper out to 23 hours with only minor CD drift (roughly 4.85-4.95 µm at a 5 µm nominal line across most of the delay window), suggesting this grade is comparatively forgiving of queue-time variation once softbaked, though no rehydration statement is made either way. Edge bead is pronounced at low spin speed (see spinNotes) and should be planned for on production coaters. Stripping is a single documented step (AZ Kwik Strip, 70°C/3 min) with no underplating reported after Cu, Ni, or Au electroplating.", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_15nxt_450cps_photoresist.pdf", "datasheetVersionOrDate": "January 2009", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "This is not a conventional one-page TDS: it is an AZ Electronic Materials internal-style technical presentation deck ('Lithographic and Plating Performance Comparison at 10 µm FT on Cu wafers', dated January 2009), hosted by distributor MicroChemicals under its 'tds' URL path. Page 2's own product roadmap places AZ 15nXT under 'Materials in sampling' (not yet 'Commercialized materials') as of that January-2009 date. MicroChemicals still lists and links this exact document as the current datasheet for the product it sells today, so status is recorded as active, but a QC reviewer should know the source document itself is 17 years old and was, at the time it was authored, describing a pre-commercial sampling-stage material.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "az-40xt", "name": "AZ 40XT-11D", "manufacturer": "Merck KGaA, Darmstadt, Germany", "productLine": "AZ 40XT", "aliases": [ "AZ 40XT", "AZ 40XT-11D Photoresist", "40XT-11D" ], "tone": "positive", "chemistry": "car", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Datasheet does not address grayscale or 3D patterning for AZ 40XT-11D; no such use is claimed anywhere in the document.", "status": "active", "successorSlug": null, "summary": "AZ 40XT-11D is a chemically amplified, thick positive-tone photoresist for TSV, plating and RIE-etch applications, giving single-coat films from 20 µm to over 60 µm with no post-softbake rehydration wait but a required post-exposure bake.", "thicknessRange": { "min_um": 20, "max_um": 60, "basis": "stated", "source": "stated — p.1 APPLICATION bullet: 'Single coat thicknesses from 20 to >60µm', verbatim. The '>60µm' upper bound is inherently open-ended; max_um is recorded as 60 (the floor of the stated inequality), not a hard ceiling. Demonstrated reference processes in the document (40 µm and 45 µm single coats, p.2-6) sit comfortably inside this stated range." }, "spinCurves": [ { "label": "AZ 40XT-11D", "points": [ { "rpm": 1000, "um": 65 }, { "rpm": 1500, "um": 46 }, { "rpm": 2000, "um": 35 }, { "rpm": 2500, "um": 29 }, { "rpm": 3000, "um": 25 }, { "rpm": 3500, "um": 21 }, { "rpm": 4000, "um": 18 } ], "source": "read from figure, \"SPIN CURVE (200mm Silicon)\", p.1 of AZ 40XT-11D Photoresist technical datasheet (Rev. 7/2016) — a single curve for one product, no multi-grade ambiguity", "figureRead": true } ], "spinNotes": "Chart conditions (p.1, in-figure annotation): coat by hand dispense at 30 rpm, spin 1000-3000 rpm for 20 seconds, then bake 125°C/7 min. As an 'ultra-high viscosity' material (p.7, PROCESS CONSIDERATIONS > COATING), careful control of nozzle height, dispense rate, dispense volume, and spin parameters is needed to avoid bubbles/voids; hand-coating should use a beaker with an integrated pour spout (a pipette or dropper is explicitly NOT recommended), pouring close to the wafer surface after letting any bubbles dissipate. The datasheet itself notes final film thickness depends on the combination of spin speed AND spin time, not spin speed alone, so the plotted curve (spun to equilibrium) is only one data point in that space.", "adhesion": { "hmds": true, "notes": "Oxide-forming substrates (e.g. Si) should be HMDS primed prior to coating AZ 40XT (PROCESS CONSIDERATIONS > SUBSTRATE PREPARATION, p.7)." }, "rehydration": "None required — this is a key differentiator from the DNQ thick resists elsewhere in this recipe set. p.1 (APPLICATION) states 'No post bake rehydration delays required', and every reference process in this datasheet lists 'Post Bake Delay: None' (p.2, p.4, p.6). (Source: TYPICAL PROCESS, p.1, and REFERENCE PROCESS tables, p.2, p.4, p.6 of AZ 40XT-11D Photoresist technical datasheet (Rev. 7/2016))", "softbake": { "temp_c": 125, "time_s": null, "method": "hotplate", "notes": "Consistently identical across all three reference processes (40 µm on Si, p.2; 40 µm on Cu, p.4; 45 µm on Cu, p.6): a 3-stage ramped/proximity bake at 125°C — 120 s at 1.27 mm gap, then 120 s at 0.63 mm gap, then 180 s in full contact (420 s total). time_s is left null because this is a multi-stage ramped-proximity sequence, not a single flat-contact soak; collapsing it to one number would misrepresent the process. PROCESS CONSIDERATIONS (p.7) states soft bake temperature for AZ 40XT should generally be in the 115-125°C range, and that ramped temperature or plate proximity is required specifically to prevent film bubbling from rapid solvent evaporation in this very thick coat.", "source": "Reference Process tables, p.2, p.4, p.6 of AZ 40XT-11D Photoresist technical datasheet (Rev. 7/2016)" }, "exposureDose": { "doses": [ { "wavelength_nm": 365, "value_mJcm2": 400, "source": "REFERENCE PROCESS (40µm Film Thickness on 200mm Si), p.2-3 of AZ 40XT-11D Photoresist technical datasheet (Rev. 7/2016)" } ], "datasheetBasis": "\"Expose: 365nm sensitive\" (p.1 TYPICAL PROCESS) and \"AZ 40XT requires exposure energy at the 365nm wavelength\" (p.7, PROCESS CONSIDERATIONS > EXPOSURE) — explicitly i-line, verbatim.", "_note": "400 mJ/cm² is the reference dose for the 40 µm-on-Si process (REFERENCE PROCESS, p.2-3; also the dose used for the front-page '20µm holes in 40µm thick AZ 40XT' demo image, p.1). The datasheet reports substantially higher doses for the same nominal film on copper: 900 mJ/cm² for 40 µm on Cu (p.4-5), and 1000 mJ/cm² for a 45 µm film in the high-speed Cu-plating process (p.6). Dose scales with both substrate (Cu vs. Si) and thickness; reporting one number for all cases would be false precision, so only the Si reference dose is recorded in the scalar field and the Cu variants are documented here instead." }, "peb": { "temp_c": 105, "time_s": null, "notes": "PEB is REQUIRED for proper imaging — the critical step that distinguishes this chemically amplified resist from the DNQ resists elsewhere in this recipe set (p.1: 'PEB is required for proper imaging'; p.7: 'A PEB is required for proper imaging of AZ 40XT'). General guidance (p.7) is 100-110°C. The reference-process tables (p.2, p.4, p.6) consistently specify a 3-stage ramped/proximity PEB at 105°C: 10 s at 1.3 mm gap, then 10 s at 0.6 mm gap, then 80 s in full contact (100 s total) — time_s is left null here for the same multi-stage reason as softbake. Internal inconsistency worth flagging for QC: the p.1 TYPICAL PROCESS summary states this bake as a single figure, '105ºC/120s', which is 20 s more than the 100 s the detailed reference-process tables actually sum to.", "source": "TYPICAL PROCESS, p.1, and REFERENCE PROCESS tables, p.2, p.4, p.6 of AZ 40XT-11D Photoresist technical datasheet (Rev. 7/2016)" }, "floodExposure": null, "develop": { "developer": "AZ 300MIF", "dilution": "undiluted (ready-to-use 0.26N / 2.38% TMAH developer; no dilution ratio stated)", "time_s": 240, "method": "puddle", "rinse": null, "source": "REFERENCE PROCESS (40µm Film Thickness on 200mm Si), p.2 of AZ 40XT-11D Photoresist technical datasheet (Rev. 7/2016): '4 x 60 second puddles', matching the p.1 front-page demo caption 'AZ 300 MIF Develop (240s)'", "_note": "240 s is the total of four 60-second puddle cycles for the 40 µm-on-Si reference process, not one continuous immersion. The other reference processes use different puddle counts/durations on the same developer: 3 x 40 second puddles (120 s total) for 40 µm on Cu (p.4); 3 x 45 second puddles (135 s total) for the 45 µm high-speed Cu-plating process (p.6). PROCESS CONSIDERATIONS (p.7) recommends AZ 300MIF generically without specifying puddle counts." }, "hardbake": { "temp_c": 80, "time_s": 300, "notes": "General guidance (p.7, PROCESS CONSIDERATIONS > HARD BAKE) is 80-85°C to ensure minimal thermal distortion; hard baking is typically NOT required for plating applications and, where used, may improve adhesion for wet-etch or pattern stability for dry-etch. The one reference process that does hard bake (20 µm Cu-plated studs after strip, p.4) used 80°C/5 minutes specifically, which is the value recorded here.", "source": "20µm Studs Post Cu Plate and Strip, p.4, and PROCESS CONSIDERATIONS > HARD BAKE, p.7, of AZ 40XT-11D Photoresist technical datasheet (Rev. 7/2016)" }, "descum": null, "applications": [ "electroplating-molding", "high-aspect-ratio", "etch-mask" ], "etchResistance": null, "liftoffSuitable": null, "platingSuitable": true, "stripper": "AZ 400T (solvent-based remover), recommended generically per PROCESS CONSIDERATIONS > STRIPPING (p.7). Demo strip conditions vary between reference processes: AZ 400T @ 55°C/10 minutes (p.4, 40 µm Cu-plating process) vs. AZ 400T @ 70°C/5 minutes (p.6, 45 µm high-speed Cu-plating process).", "storage": "Combustible liquid; store in sealed original containers in a well-ventilated, dry area away from heat, light, oxidizers, reducers, and sources of ignition. Recommended storage temperature 30-55°F (as printed, p.8, STORAGE).", "notes": "AZ 40XT-11D is a chemically amplified thick positive resist, the odd one out among AZ's thick-film positive resists in this recipe set: unlike the DNQ resists (P4620, 10XT), it requires NO post-softbake rehydration wait (every reference process here lists 'Post Bake Delay: None') but DOES require a post-exposure bake, and skipping that PEB is the classic failure mode for this material — without it the pattern will not image properly. Both softbake and PEB are run as ramped/proximity bakes (three discrete stages at decreasing hotplate proximity) rather than a single flat contact bake, specifically to avoid solvent-outgassing bubbles in these very thick (20 to over 60 µm) single coats. It develops in AZ 300MIF, a 0.26N (2.38%) TMAH developer, applied as puddle cycles rather than a single immersion step. Exposure dose scales sharply with substrate: roughly 400 mJ/cm² at 365 nm for a 40 µm coat on silicon versus 900-1000 mJ/cm² for a comparable coat on copper. Hard bake is generally unnecessary for plating use and, when applied, should stay in the 80-85°C range to avoid thermally distorting the pattern.", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_40xt_11d_photoresist.pdf", "datasheetVersionOrDate": "Rev. 7/2016", "accessedDate": "2026-07-10", "secondarySources": [] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": true }, { "slug": "az-4533", "name": "AZ 4533", "manufacturer": "Merck (AZ Electronic Materials)", "productLine": "AZ 4500 series", "aliases": [ "AZ4533", "AZ 4533 Photoresist" ], "tone": "positive", "chemistry": "dnq-novolak", "_chemistryNote": "Datasheet describes a 'photoactive compound (PAC)' with 'reduced nitrogen content' in a positive-tone thick-film resist, wording consistent with DNQ-based chemistry, but this document never uses the words 'DNQ' or 'novolak' anywhere, and gives no resin-base information. Left null to avoid overclaiming a specific formulation this particular document doesn't state directly (contrast the sibling AZ 1500 series datasheet, which explicitly ties its resist type to 'DNQ/novolac' via a stripper-compatibility statement).", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Datasheet does not mention grayscale or 3D lithography anywhere. Its entire focus is achieving high single-coat film thickness (reduced spin time, multi-coat with bake cycles, adjusted exposure dose to fully clear thick films) and general processing guidelines — no grayscale-specific guidance (e.g. partial-exposure dose modulation, grayscale mask compatibility) is given.", "status": "active", "successorSlug": null, "summary": "AZ 4533 is the lower-viscosity of the AZ 4500 series' two thick-film grades — the everyday pick for single-coat films in the low-few-micron range, where the higher-viscosity AZ 4562 is reserved for much thicker coats and its special multi-coat techniques. AZ 4533 is a thick-film member of Merck's AZ 4500 series of positive photoresists, formulated with a low-absorption photoactive compound so it can be coated in a single spin step at thicknesses of roughly 2.7-4.7 µm.", "thicknessRange": { "min_um": 2.69, "max_um": 4.67, "basis": "curve-span", "source": "curve-span: taken directly from the printed \"FILM THICKNESS [µm] as FUNCTION of SPIN SPEED\" table (p.2), which tabulates AZ 4533 at 2000-6000 rpm (4.67 µm at 2000 rpm down to 2.69 µm at 6000 rpm). The series-wide prose claim of usable thicknesses 'up to 50 µm' (p.1, GENERAL INFORMATION) is a whole-series statement covering special multi-coat/reduced-spin-time techniques and the higher-viscosity AZ 4562 grade, not a single-coat figure for AZ 4533 specifically, so it was not used here. [RECLASSIFIED during audit: the extraction agent labelled this \"stated\", but a printed thickness-vs-spin-speed table IS the spin curve, so its min/max is a curve span, not a manufacturer-stated achievable range. The reasoning above is unchanged and correct.]" }, "spinCurves": [ { "label": "AZ 4533 as supplied", "points": [ { "rpm": 2000, "um": 4.67 }, { "rpm": 3000, "um": 3.81 }, { "rpm": 4000, "um": 3.3 }, { "rpm": 5000, "um": 2.95 }, { "rpm": 6000, "um": 2.69 } ], "source": "numeric table \"FILM THICKNESS [µm] as FUNCTION of SPIN SPEED (characteristically)\", p.2 of AZ 4500 Series Technical Datasheet (Merck, Rev. 03/21)." } ], "spinNotes": "Datasheet states the common spin time is about 30-40 s for standard coating. No dispense volume, ramp/acceleration, or static-vs-dynamic dispense is given. The spin-speed table above is captioned 'characteristically', i.e. representative rather than guaranteed values. Dilution/edge-bead removal uses AZ EBR Solvent or AZ EBR 70/30 (PROCESSING GUIDELINES, p.2). AZ 4533-specific reduced-spin-time or multi-coat techniques for extra-thick films are not discussed in this datasheet — those special techniques (3 s spin, multi-coat with bake cycle) are described only for AZ 4562 (the highest-viscosity grade).", "adhesion": { "hmds": null, "notes": "Datasheet does not address substrate priming or HMDS for the AZ 4500 series; there is no SUBSTRATE PREPARATION section (unlike the AZ 1500 series datasheet from the same manufacturer)." }, "rehydration": null, "softbake": { "temp_c": 100, "time_s": 50, "method": "hotplate", "notes": "Series-wide PROCESSING GUIDELINES value, not stated as AZ 4533-specific vs AZ 4562-specific — the table applies to both grades. For very thick coatings the datasheet separately advises (p.1, GENERAL INFORMATION) leaving the resist at room temperature at least 15 minutes before prebake to let solvent evaporate, and preferring a hotplate with a ramped temperature over an oven, to avoid trapped-solvent bubbling and adhesion failure.", "source": "PROCESSING GUIDELINES, p.2 (\"Prebake: 100°C, 50\\\", hotplate\")" }, "exposureDose": { "doses": [], "datasheetBasis": "PROCESSING GUIDELINES states exposure is \"broadband and monochromatic\"; PHYSICAL and CHEMICAL PROPERTIES gives spectral sensitivity as 310-440 nm for AZ 4533. No single dose value is published for this grade.", "_note": "GENERAL INFORMATION (p.1) warns that in extreme thick-film cases 'exposure doses above 1000 mJ/cm² have to be applied,' but this is a cautionary statement about very thick coatings in general (and the side effects of overdosing, such as surface crosslinking), not a specific recommended dose for AZ 4533 at its standard tested thicknesses (2.7-4.7 µm) — not used as a value here to avoid false precision or misattribution." }, "peb": { "temp_c": null, "time_s": null, "notes": "\"PEB: not required, optional with monochromatic exposure.\" No temperature or time is given for the optional case.", "source": "PROCESSING GUIDELINES, p.2" }, "floodExposure": null, "develop": { "developer": "AZ 340 (AZ 400K also usable)", "dilution": "AZ 340 diluted 1:5 with water; no dilution stated for AZ 400K.", "time_s": null, "method": null, "rinse": null, "source": "PROCESSING GUIDELINES, p.2 (\"Development: AZ 340, 1:5, 30\\\"/µm film thickness\"); GENERAL INFORMATION, p.2", "_note": "Develop time is published as a rate — 30 s per µm of film thickness — not a fixed duration. Since AZ 4533's own thickness varies by spin speed (2.69-4.67 µm across the tabulated range), collapsing this to one seconds value would require assuming a specific film thickness not otherwise specified; left null and the verbatim rate recorded here instead. GENERAL INFORMATION separately recommends targeting a development rate of about 2 µm/min. (≈30 s/µm), consistent with this rate. Develop method (immersion/puddle/spray) is not stated." }, "hardbake": { "temp_c": 115, "time_s": 50, "notes": "Hotplate condition (115°C, 50 s). Datasheet gives an alternative of 115°C for 60 minutes (3600 s) in an oven.", "source": "PROCESSING GUIDELINES, p.2 (\"Postbake: 115°C, 50s hotplate or 60 min. oven\")" }, "descum": null, "applications": [], "etchResistance": null, "liftoffSuitable": null, "platingSuitable": null, "stripper": "AZ 100 Remover, concentrated. Source: PROCESSING GUIDELINES, p.2.", "storage": "Store in sealed original containers between 0°C and 25°C; brief excursions do not adversely affect properties (+27°C for 10 hours, +32°C for 6 hours, or +35°C for 5 hours). Shelf life is 1 year at recommended storage conditions; the expiration date is printed on each bottle's label. Source: HANDLING ADVISES, p.3.", "notes": "AZ 4533 is a thick-film member of Merck's AZ 4500 series, formulated (per p.1) with a photoactive compound of \"low absorption and reduced nitrogen content\" specifically so film thicknesses above the ~3 µm ceiling of standard positive resists can still be fully exposed and cleared without the crosslinking or nitrogen-trapping side effects that appear when a standard resist is pushed to extreme thickness. Reach for AZ 4533 when a moderate thick coat in the few-micron range is enough and a straightforward single-spin process is wanted; step up to AZ 4562 only when the target film is thicker than AZ 4533's standard spin range or needs the reduced-spin-time or multi-coat methods. Unlike the datasheet's introduction, which discusses AZ 4562's special coating techniques (reduced spin time, multi-coat) for very thick films, no such special technique is documented specifically for AZ 4533. This datasheet has no dedicated APPLICATION section and does not name specific end-uses (etch mask, lift-off, plating, etc.) — its content is limited to achieving and processing the thick coating itself, so the applications field is left empty rather than inferred. Develop time is a rate (30 s per µm of film thickness) rather than a fixed duration, and PEB is optional and only relevant to monochromatic exposure. No HMDS/adhesion-promotion guidance is given, unlike the AZ 1500 series datasheet from the same manufacturer. Chemistry classified as dnq-novolak from MicroChemicals' statement that AZ/TI resist resin is novolak and the photoactive compound belongs to the diazonaphthoquinone (DNQ) group (chemistry classified 2026-07-12 from manufacturer SDS/TDS).", "developerFamily": "buffered-alkaline", "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_4500_series.pdf", "datasheetVersionOrDate": "Rev. (03/21)", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://www.microchemicals.com/AZ-4533-Photoresist-3.785-l/1A004533", "what": "MicroChemicals states: 'The resin of AZ and TI resists is Novolak... photo active compound... belongs to the group of diazonaphthoquinones (DNQ)'; the basis for classifying AZ 4533 as dnq-novolak." } ] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "az-4562", "name": "AZ 4562", "manufacturer": "Merck (AZ Electronic Materials)", "productLine": "AZ 4500 series", "aliases": [ "AZ4562", "AZ 4562 Photoresist" ], "tone": "positive", "chemistry": null, "_chemistryNote": "Datasheet describes a 'photoactive compound (PAC)' with 'reduced nitrogen content' in a positive-tone thick-film resist, wording consistent with DNQ-based chemistry, but this document never uses the words 'DNQ' or 'novolak' anywhere, and gives no resin-base information. Left null to avoid overclaiming a specific formulation this particular document doesn't state directly (contrast the sibling AZ 1500 series datasheet, which explicitly ties its resist type to 'DNQ/novolac' via a stripper-compatibility statement).", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Datasheet does not mention grayscale or 3D lithography anywhere. Its entire focus is achieving high single-coat and multi-coat film thickness (up to 50 µm with special techniques) and the exposure/development adjustments that thick-film coating requires — no grayscale-specific guidance (e.g. partial-exposure dose modulation, grayscale mask compatibility, tone-ramp behavior) is given for AZ 4562, despite it being the series' highest-viscosity, thickest-film grade. Grayscale suitability should not be assumed from thick-film capability alone; it is not addressed here.", "status": "active", "successorSlug": null, "summary": "AZ 4562 is the AZ 4500 series' thick-film flagship — the grade to choose when a single AZ 4533 coat cannot reach the required thickness, or when a process needs the extreme films only its special coating techniques provide. AZ 4562 is the highest-viscosity, thickest-film member of Merck's AZ 4500 series of positive photoresists, capable of single-coat films of roughly 5-9 µm at standard spin speeds, and — using reduced spin time or multi-coat bake cycles — films up to 50 µm.", "thicknessRange": { "min_um": 5.06, "max_um": 8.77, "basis": "curve-span", "source": "curve-span: taken directly from the printed \"FILM THICKNESS [µm] as FUNCTION of SPIN SPEED\" table (p.2), which tabulates AZ 4562 at 2000-6000 rpm (8.77 µm at 2000 rpm down to 5.06 µm at 6000 rpm). The series-wide prose claim of usable thicknesses 'up to 50 µm' (p.1, GENERAL INFORMATION) requires the special coating techniques described below (reduced spin time down to 3 s, or multi-coat with an inter-bake), not a single standard spin step, so it was not used as the stated single-coat range. [RECLASSIFIED during audit: the extraction agent labelled this \"stated\", but a printed thickness-vs-spin-speed table IS the spin curve, so its min/max is a curve span, not a manufacturer-stated achievable range. The reasoning above is unchanged and correct.]" }, "spinCurves": [ { "label": "AZ 4562 as supplied", "points": [ { "rpm": 2000, "um": 8.77 }, { "rpm": 3000, "um": 7.16 }, { "rpm": 4000, "um": 6.2 }, { "rpm": 5000, "um": 5.55 }, { "rpm": 6000, "um": 5.06 } ], "source": "numeric table \"FILM THICKNESS [µm] as FUNCTION of SPIN SPEED (characteristically)\", p.2 of AZ 4500 Series Technical Datasheet (Merck, Rev. 03/21)." } ], "spinNotes": "GENERAL INFORMATION (p.1) states AZ 4562 \"allows to spin coat 10 µm in a single step (2000 rpm)\" — this rounded prose figure disagrees with the precise tabulated value of 8.77 µm at 2000 rpm on p.2; the table value was used in spinCurves as the more precise, directly tabulated figure, but the discrepancy should be checked against the source PDF during visual QC. For thicker films the datasheet describes two special techniques specific to AZ 4562: (1) reducing the common ~30-40 s spin time to only 3 s, yielding ~20 µm, but the substrate must then be left horizontal on the spinner for an additional minute to dry; (2) multiple coating with a bake cycle in between (inter-coat bakes must not exceed 90°C or the final prebake temperature, whichever is lower), exploiting the fact that AZ 4562's high solids content (close to its dissolution limit) means the underlying coat is only minimally redissolved. No dispense volume or ramp/acceleration is given for either the standard or special techniques.", "adhesion": { "hmds": null, "notes": "Datasheet does not address substrate priming or HMDS for the AZ 4500 series; there is no SUBSTRATE PREPARATION section (unlike the AZ 1500 series datasheet from the same manufacturer)." }, "rehydration": null, "softbake": { "temp_c": 100, "time_s": 50, "method": "hotplate", "notes": "Series-wide PROCESSING GUIDELINES value, not stated as AZ 4562-specific vs AZ 4533-specific — the table applies to both grades and assumes a standard single coat, not the special thick-film techniques. For thick coatings the datasheet separately advises (p.1, GENERAL INFORMATION) leaving the resist at room temperature at least 15 minutes before prebake to let solvent evaporate, and preferring a hotplate with a ramped temperature over an oven, to avoid trapped-solvent bubbling and adhesion failure. For AZ 4562 multi-coating specifically, inter-coat bakes must not exceed 90°C or the final prebake temperature (whichever is lower).", "source": "PROCESSING GUIDELINES, p.2 (\"Prebake: 100°C, 50\\\", hotplate\"); GENERAL INFORMATION, p.1 (multi-coat inter-bake constraint)" }, "exposureDose": { "doses": [], "datasheetBasis": "PROCESSING GUIDELINES states exposure is \"broadband and monochromatic\"; PHYSICAL and CHEMICAL PROPERTIES gives spectral sensitivity as 310-440 nm for AZ 4562, with an absorptivity of 1.01 l/g·cm at 398 nm (vs 0.86 for AZ 4533). No single dose value is published for this grade.", "_note": "GENERAL INFORMATION (p.1) warns that in extreme thick-film cases 'exposure doses above 1000 mJ/cm² have to be applied,' and explains that as film thickness increases, exposure dose must be raised to fully clear the bottom of the film without over-dosing the surface (which induces crosslinking, similar to deep-UV hardening). This is a general cautionary explanation of why AZ 4562 exists (a lower-absorption PAC to reduce this problem), not a specific recommended dose for AZ 4562 at any particular thickness — not used as a value here to avoid false precision or misattribution." }, "peb": { "temp_c": null, "time_s": null, "notes": "\"PEB: not required, optional with monochromatic exposure.\" No temperature or time is given for the optional case.", "source": "PROCESSING GUIDELINES, p.2" }, "floodExposure": null, "develop": { "developer": "AZ 340 (AZ 400K also usable)", "dilution": "AZ 340 diluted 1:5 with water; no dilution stated for AZ 400K.", "time_s": null, "method": null, "rinse": null, "source": "PROCESSING GUIDELINES, p.2 (\"Development: AZ 340, 1:5, 30\\\"/µm film thickness\"); GENERAL INFORMATION, p.2", "_note": "Develop time is published as a rate — 30 s per µm of film thickness — not a fixed duration. AZ 4562's own thickness varies widely by spin speed and coating technique (5.06-8.77 µm in the tabulated single-coat range, up to 50 µm with special techniques), so collapsing this to one seconds value would require assuming a specific film thickness not otherwise specified; left null and the verbatim rate recorded here instead. GENERAL INFORMATION separately recommends targeting a development rate of about 2 µm/min. (≈30 s/µm), consistent with this rate, and notes that even heavily overexposed positive resist has a dissolution-rate ceiling around 100 nm/s. Develop method (immersion/puddle/spray) is not stated." }, "hardbake": { "temp_c": 115, "time_s": 50, "notes": "Hotplate condition (115°C, 50 s). Datasheet gives an alternative of 115°C for 60 minutes (3600 s) in an oven.", "source": "PROCESSING GUIDELINES, p.2 (\"Postbake: 115°C, 50s hotplate or 60 min. oven\")" }, "descum": null, "applications": [], "etchResistance": null, "liftoffSuitable": null, "platingSuitable": null, "stripper": "AZ 100 Remover, concentrated. Source: PROCESSING GUIDELINES, p.2.", "storage": "Store in sealed original containers between 0°C and 25°C; brief excursions do not adversely affect properties (+27°C for 10 hours, +32°C for 6 hours, or +35°C for 5 hours). Shelf life is 1 year at recommended storage conditions; the expiration date is printed on each bottle's label. Source: HANDLING ADVISES, p.3.", "notes": "AZ 4562 is the highest-viscosity grade in Merck's AZ 4500 series, formulated (per p.1) with a photoactive compound of \"low absorption and reduced nitrogen content\" so that films far thicker than the ~3 µm ceiling of standard positive resists can still be fully exposed through their depth without surface crosslinking or nitrogen-bubble lifting. Choose AZ 4562 over AZ 4533 whenever the film has to be thicker than the thinner grade's standard spin range delivers, or when only its reduced-spin-time or multi-coat routes can reach the target — for ordinary few-micron coats, AZ 4533 remains the simpler choice. It is the only grade in this datasheet with documented special coating techniques for extreme thickness: a 3 s reduced spin time (with the wafer left horizontal an extra minute to dry) for ~20 µm films, or multi-coating with inter-coat bakes capped at 90°C (or the final prebake temperature, whichever is lower) for still-thicker stacks, up to a series-wide ceiling of 50 µm. A minor internal inconsistency exists between the prose claim of '10 µm in a single step (2000 rpm)' (p.1) and the tabulated 8.77 µm at 2000 rpm (p.2) — the table value is used in the spin curve here as the more precise figure, but this is worth a human check against the source PDF. This datasheet has no dedicated APPLICATION section, does not name specific end-uses (etch mask, lift-off, plating, etc.), and does not address grayscale or 3D lithography despite being the series' thick-film flagship grade — none of those are asserted here. Develop time is a rate (30 s per µm of film thickness) rather than a fixed duration, and PEB is optional and only relevant to monochromatic exposure. No HMDS/adhesion-promotion guidance is given, unlike the AZ 1500 series datasheet from the same manufacturer.", "developerFamily": "buffered-alkaline", "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_4500_series.pdf", "datasheetVersionOrDate": "Rev. (03/21)", "accessedDate": "2026-07-10", "secondarySources": [] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "az-5214e", "name": "AZ 5214E", "manufacturer": "MicroChemicals / Merck Performance Materials GmbH", "productLine": "AZ 5200 series", "aliases": [ "AZ5214E", "AZ 5214 E", "AZ 5214-E" ], "tone": "image-reversal", "chemistry": "dnq-novolak", "grayscaleSuitable": false, "grayscaleNote": "Not addressed in the datasheet; the resist is formulated around a crosslinking image-reversal mechanism and is not marketed for grayscale/3D lithography.", "status": "active", "successorSlug": null, "summary": "AZ 5214E is a novolak/DNQ (diazonaphthoquinone) positive photoresist from Merck's AZ line that is almost exclusively used in image-reversal (IR) mode: a special crosslinking agent activated by a reversal bake makes exposed areas insoluble, so a subsequent flood exposure and standard positive development yields a negative-tone image with a re-entrant (undercut) sidewall profile ideally suited to lift-off metallization.", "thicknessRange": { "min_um": 1.14, "max_um": 1.98, "basis": "curve-span", "source": "curve-span: the AZ 5214E datasheet states no achievable-thickness range in prose; its only thickness data is the printed \"FILM THICKNESS [µm] as FUNCTION of SPIN SPEED (characteristically)\" table (p.2: 1.98/1.62/1.40/1.25/1.14 µm at 2000/3000/4000/5000/6000 rpm), which per extraction policy is the spin curve. min/max are the span of that table (matching spinCurves)." }, "spinCurves": [ { "label": "as supplied", "points": [ { "rpm": 2000, "um": 1.98 }, { "rpm": 3000, "um": 1.62 }, { "rpm": 4000, "um": 1.4 }, { "rpm": 5000, "um": 1.25 }, { "rpm": 6000, "um": 1.14 } ], "source": "numeric table \"FILM THICKNESS [µm] as FUNCTION of SPIN SPEED (characteristically)\", p.2 of MicroChemicals/Merck AZ 5214E Technical Data Sheet" } ], "spinNotes": "Datasheet gives no accel/dispense/edge-bead procedure beyond naming AZ EBR Solvent for dilution/edge-bead removal (Processing Guidelines table, p.3). 4000 rpm (1.40 µm) is bolded in the source table, suggesting it as the reference/typical spin speed, though the datasheet does not explicitly label it as such.", "adhesion": { "hmds": null, "notes": "HMDS/adhesion promotion is not mentioned anywhere in this datasheet." }, "rehydration": null, "softbake": { "temp_c": 110, "time_s": 50, "method": "hotplate", "notes": null, "source": "Processing Guidelines table, p.3: \"Prebake 110°C, 50\", hotplate\"" }, "exposureDose": { "doses": [], "datasheetBasis": "Exposure is stated only qualitatively: \"broadband and monochromatic h- and i-line\" (Processing Guidelines, p.3), with a spectral sensitivity range of 310–420 nm (Physical & Chemical Properties table, p.2). No absolute imagewise exposure dose in mJ/cm² is published.", "_note": "The datasheet gives only a relative rule of thumb (p.2): for IR-processing, the imagewise (patternwise) exposure dose should be about half that of a standard positive-resist process with the same resist, while flood-exposure energy should be about double — no absolute number or wavelength-specific dose is stated for the imagewise step, so this is left null rather than guessed or derived." }, "peb": { "temp_c": 120, "time_s": 120, "notes": "This is the image-reversal (\"reversal bake\", RB) step, not a conventional PEB — AZ 5214E's process has no separate post-exposure bake before flood exposure; the reversal bake occupies that position in the sequence. 120°C/2 min is the datasheet's example condition in the Processing Guidelines table, but the text stresses this must be individually optimized per line/equipment and always falls within 115–125°C, held constant within ±1°C (\"the most critical parameter of the IR-process\"). Above ~130°C the resist thermally crosslinks even in unexposed areas and destroys the pattern.", "source": "Processing Guidelines table, p.3 (\"Reversal bake 120°C, 2 min., hotplate (most critical step)\"); temperature-optimization procedure and ±1°C/115–125°C range, General Information, p.1" }, "floodExposure": { "dose_mJcm2": 200, "notes": "Flood exposure (no mask) is described as \"absolutely uncritical\" as long as sufficient energy is applied; > 200 mJ/cm² is called a good choice, and anywhere from 150–500 mJ/cm² has no major influence on performance.", "source": "AZ 5214E TDS, p.1–3 (General Information + Processing Guidelines table: \"Flood exposure > 200 mJ/cm² (uncritical)\")" }, "develop": { "developer": "AZ 340 (metal-ion-containing) or AZ 726 MIF (metal-ion-free)", "dilution": "AZ 340 used 1:5 diluted; AZ 726 MIF used as supplied", "time_s": null, "method": "AZ 340: tank or spray; AZ 726 MIF: puddle", "rinse": null, "source": "Processing Guidelines table, p.3: \"Development AZ 340, 1:5 (tank, spray) or AZ 726 (puddle)\"", "_note": "No develop time is given in the Processing Guidelines table. A \"1 minute\" immersion appears earlier (p.1) but only as part of the procedure for finding the correct reversal-bake temperature (test wafers at several RB temperatures, flood-exposed, then developed 1 min to see which still carry resist) — it is not labeled as the standard production develop time, so it was not carried into the `time_s` field. Rinse medium (e.g. DI water) is not stated." }, "hardbake": { "temp_c": 120, "time_s": 50, "notes": "Labeled \"Postbake\" and explicitly optional.", "source": "Processing Guidelines table, p.3: \"Postbake 120°C, 50s hotplate (optional)\"" }, "descum": null, "applications": [ "lift-off", "image-reversal" ], "etchResistance": null, "liftoffSuitable": true, "platingSuitable": false, "stripper": "AZ 100 Remover, concentrate (Processing Guidelines table, p.3: \"Removal AZ 100 Remover, conc.\")", "storage": "Store in sealed original containers, protected from light and heat, between 0°C and 25°C; brief excursions to -5°C or +30°C for up to 24 hours do not adversely affect properties. Shelf life is limited; expiration date is printed on the bottle label as [year/month/day]. (Handling Advises, p.3)", "notes": "AZ 5214E is a positive DNQ/novolak resist that is almost exclusively run in image-reversal mode for lift-off, producing a negative (re-entrant) sidewall profile instead of the ~75-85° positive slope of standard positive processing. The reversal-bake temperature is the single most process-sensitive parameter: it must be individually tuned (typically 115-125°C) and held within ±1°C, since a few degrees too high causes thermal crosslinking even in unexposed regions and destroys the pattern; the datasheet gives a explicit calibration procedure for finding it. By contrast, the flood exposure step is deliberately forgiving (150-500 mJ/cm² all work). A T-shaped (overhanging-lip) lift-off profile can also be produced with a modified sequence: light flood exposure before the reversal bake, then normal imagewise exposure and development.", "developerFamily": "tmah", "references": [ { "type": "paper", "title": "Standard AZ 5214E photoresist in laser interference and EBDW lithographies", "authors": "Škriniarová et al.", "journal": "Vacuum", "year": 2015, "doi": "10.1016/j.vacuum.2014.09.012", "url": "https://doi.org/10.1016/j.vacuum.2014.09.012", "accessedDate": "2026-07-12", "summary": "AZ 5214E run as an e-beam resist, not just UV" }, { "type": "paper", "title": "High thickness material lift-off using multi-layer photoresist", "authors": "Elshenety et al.", "journal": "Journal of Micromechanics and Microengineering", "year": 2025, "doi": "10.1088/1361-6439/adac6b", "url": "https://doi.org/10.1088/1361-6439/adac6b", "accessedDate": "2026-07-12", "summary": "Stacked multi-layer 5214E lifts off more than 6 µm of electroplated Cu — 4× its single-coat thickness" } ], "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_5214e_photoresist.pdf", "datasheetVersionOrDate": null, "accessedDate": "2026-07-10", "secondarySources": [] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "az-9245", "name": "AZ 9245", "manufacturer": "AZ Electronic Materials (Clariant AG)", "productLine": "AZ 9200", "aliases": [ "AZ9245", "AZ 9245 (220 CPS)", "AZ 9245 Photoresist" ], "tone": null, "chemistry": null, "_chemistryNote": "The document never names the underlying photoactive chemistry or tone. It states AZ 9200 'can be used as a higher resolution replacement for AZ P4000 photoresist... it is developed from the same chemistry' (p.1) — a cross-reference to a different, unread product, not a direct statement of what that chemistry is — and recommends AZ 400T/300T strippers (p.8) without characterizing them as DNQ/novolac-specific (unlike the Merck-era AZ documents processed alongside this one, which explicitly say so). Inferring 'positive DNQ-novolak' from familiarity with the wider AZ product family or from the P4000 cross-reference would be exactly the kind of guess this extraction avoids, so both fields are left null.", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed anywhere in this 1997 datasheet; grayscale/3D lithography is not a documented use case for this product.", "status": "active", "successorSlug": null, "summary": "AZ 9245 is the thinner (220 cP) of two viscosity grades in Clariant/AZ Electronic Materials' AZ 9200 high-resolution thick-resist line, coating to roughly 4.6-6.6 µm and resolving sub-micron lines and spaces at 4.6 µm film thickness, targeted at coil-plating and thin-film recording-head applications.", "thicknessRange": { "min_um": 4.6, "max_um": 6.6, "basis": "curve-span", "source": "curve-span — a printed numeric \"Film Thickness\" table (p.7, columns 2000/2500/3000/3500 rpm) gives AZ 9245 values across that range, extended by the single grade-specific anchor point in the \"Typical Process for 4.6 µm Film Thickness [AZ 9245 Photoresist (220 CPS)]\" table (p.6: Spin 3800 rpm, 60 sec → 4.6 µm). The series-wide stated figure in the intro (p.1), \"AZ 9200 photoresist is available in two viscosity grades for film thicknesses of 4 to 24 µm\", spans both AZ 9245 and AZ 9260 combined (9260 reaches the thicker end via multi-coat, per p.6), not AZ 9245 alone, so the grade's own curve-span is used instead per extraction policy." }, "spinCurves": [ { "label": "AZ 9245 (220 cP)", "points": [ { "rpm": 2000, "um": 6.6 }, { "rpm": 2500, "um": 5.8 }, { "rpm": 3000, "um": 5.2 }, { "rpm": 3500, "um": 4.8 }, { "rpm": 3800, "um": 4.6 } ], "source": "numeric table \"Film Thickness\" (columns 2000/2500/3000/3500 rpm, row \"AZ 9245 Photoresist 220 cP\"), p.7, combined with the single grade-specific anchor point from \"Typical Process for 4.6 µm Film Thickness [AZ 9245 Photoresist (220 CPS)]\" (Coat: Spin 3800 rpm, 60 sec → 4.6 µm target), p.6, of AZ 9200 Photoresist datasheet. Both are explicitly labeled AZ 9245 (220 cP) and are mutually consistent (monotonic decrease: 4.8 µm at 3500 rpm to 4.6 µm at 3800 rpm)." } ], "spinNotes": "IMPORTANT UNIT FLAG FOR QC: the p.7 table's units render ambiguously across two extraction passes of the same page — one pass shows the values with an \"Å\" (Angstrom) suffix (e.g. \"6 600 Å\"), another shows the same figures as \"6.6um\" etc. A literal Angstrom reading (0.66 µm at 2000 rpm) would be physically inconsistent with the rest of the document: it would mean the film gets THICKER at the higher 3800 rpm point (4.6 µm, from the p.6 Typical Process table) than at the lower 2000 rpm point (0.66 µm), which violates basic spin-coating physics (thickness decreases with increasing spin speed). The µm reading is internally consistent with the 4.6 µm/3800 rpm anchor point and is used here, but a QC reviewer should confirm the actual printed unit glyph against the source PDF. Coat/dispense and edge-bead removal are documented as explicit process steps for this grade (p.6): \"Coat: Dispense static or dynamic @ 300 rpm, Spin 3800 rpm/60 sec\" then \"Edge Bead Removal: Rinse 500 rpm/10 sec, Dry 1000 rpm/10 sec\" — both built into the same Typical Process table as softbake/exposure/develop. No rehydration hold is listed anywhere in this document (see the rehydration field) — despite this being a thick resist, where a rehydration hold is sometimes expected, this specific 1997 datasheet's process tables simply do not include one.", "adhesion": { "hmds": null, "notes": "Not addressed anywhere in this datasheet — no substrate priming/HMDS guidance is given at all, unlike the Merck-era AZ documents processed alongside this one." }, "rehydration": "Not mentioned anywhere in this document. The grade-specific \"Typical Process for 4.6 µm Film Thickness [AZ 9245 Photoresist (220 CPS)]\" table (p.6) lists Coat → Softbake → Edge Bead Removal → Exposure → Post Exposure Bake (not recommended) → Development, with no separate rehydration step. This may be a gap in this particular 1997 datasheet rather than confirmation the process needs none — not used as a basis to assert either way; flagged for QC.", "softbake": { "temp_c": 110, "time_s": 120, "method": "hotplate", "notes": "Grade-specific: \"Softbake 110°C, 120 sec hotplate\" — Typical Process for 4.6 µm Film Thickness [AZ 9245 Photoresist (220 CPS)], p.6. Corroborated by matching \"Softbake Hotplate 110°C, 120 sec\" conditions on the Linearity [Broadband] and Focus Latitude [Broadband] test panels for the same 4.6 µm film thickness (p.2-3).", "source": "Typical Process for 4.6 µm Film Thickness [AZ 9245 Photoresist (220 CPS)], p.6 of AZ 9200 Photoresist datasheet." }, "exposureDose": { "doses": [ { "wavelength_nm": null, "value_mJcm2": 900, "source": "Typical Process for 4.6 µm Film Thickness [AZ 9245 Photoresist (220 CPS)], p.6; corroborated by Linearity [Broadband] and Focus Latitude [Broadband], p.2-3, of AZ 9200 Photoresist datasheet." } ], "datasheetBasis": "\"Exposure (10% bias) 900 mJ/cm2, broadband stepper\" — Typical Process for 4.6 µm Film Thickness [AZ 9245 Photoresist (220 CPS)], p.6. The document separately states AZ 9200 has \"sensitivity to both h- and i-line\" and is \"capable for both broadband and i-line steppers\" (p.1), but this specific grade's documented process dose is broadband.", "_note": "900 mJ/cm² is both the Typical Process table's single stated dose AND the center point of a 780-1020 mJ/cm² broadband dose sweep (\"Linearity [Broadband]\", p.2) tested at the same 4.6 µm film thickness with matching softbake conditions (110°C/120s) — strong internal corroboration, so it is used. \"(10% bias)\" annotates a photomask CD bias compensation, not a dose range. No i-line-specific dose is published for AZ 9245: the document's only i-line dose data (\"Focus and Exposure Latitude [i-Line]\", p.5, 660-900 mJ/cm² sweep) is explicitly run at 10 µm film thickness with a different softbake (90°C/155s) and developer (AZ 300MIF spray, 360s) — that is the AZ 9260 (520 CPS, 10 µm) Typical Process (p.6), not AZ 9245's. Reusing the 9260 i-line figure for 9245 would misattribute grade-specific data across the two viscosity grades printed in this shared datasheet, so at365_mJcm2 stays null. No h-line (405 nm) dose is published for either grade despite the h-/i-line sensitivity claim." }, "peb": { "temp_c": null, "time_s": null, "notes": "\"Post Exposure Bake: not recommended in most applications\" — Typical Process for 4.6 µm Film Thickness [AZ 9245 Photoresist (220 CPS)], p.6. No PEB temperature or time is given because none is called for by default.", "source": "Typical Process for 4.6 µm Film Thickness [AZ 9245 Photoresist (220 CPS)], p.6 of AZ 9200 Photoresist datasheet." }, "floodExposure": null, "develop": { "developer": "AZ 400K Developer", "dilution": "1:4", "time_s": 120, "method": "spray", "rinse": "Spin rinse 300 rpm/20 sec, then spin dry 4000 rpm/15 sec (following the AZ 400K spray develop at 27°C dispense temperature).", "source": "Typical Process for 4.6 µm Film Thickness [AZ 9245 Photoresist (220 CPS)], p.6 of AZ 9200 Photoresist datasheet.", "_note": "AZ 400K 1:4 (buffered, potassium-hydroxide-based) is the grade-specific developer used in the Typical Process table and is described elsewhere as \"the preferred developer\" for thick films of AZ 9200 (Companion Products, p.8). AZ 300 MIF (TMAH) is offered as an alternative \"for integrated circuit applications\" (p.1, p.8) but is not the developer used in this grade's own Typical Process table, so it is not used for the scalar fields here." }, "hardbake": null, "descum": null, "applications": [ "electroplating-molding", "high-aspect-ratio", "mems-structural" ], "etchResistance": null, "liftoffSuitable": null, "platingSuitable": true, "stripper": "AZ 400T and 300T strippers are recommended for removal of AZ 9200 photoresist; AZ S-46 stripper is a non-NMP solvent stripper particularly suited to thin film recording head applications (Companion Products, p.8).", "storage": "\"Keep in sealed original container. Protect from light and heat. Store between 30 and 70°F (–1 to 24°C). Refrigerate whenever possible. Refrigeration may extend shelf life. Empty container may contain harmful residue and vapors.\" (Storage, p.8).", "notes": "AZ 9245 (220 cP) is the thinner of two viscosity grades in this 1997 Clariant AZ 9200 thick-resist line, targeting a 4.6 µm single-coat film with sub-micron resolution (<1 µm lines/spaces per p.1) and 5-7:1 aspect ratios on broadband steppers. Its Typical Process table explicitly states PEB is \"not recommended in most applications\" and does not list a rehydration-hold step at all — contrary to what might be assumed for a thick resist, this document gives no basis to include one. Edge-bead removal is documented as an explicit spin-rinse/spin-dry step (500 rpm/10s rinse, 1000 rpm/10s dry) built directly into the same process table as coat, softbake, exposure, and develop. The document states AZ 9200 is sensitive to both h- and i-line light, but only broadband exposure data (900 mJ/cm² nominal) is actually published for the 4.6 µm/AZ 9245 process specifically — the datasheet's only i-line dose data belongs to the thicker AZ 9260 (10 µm) grade and is not reused here. This document predates and differs structurally from the Merck-branded AZ datasheets processed alongside it (no explicit tone/chemistry statement, no HMDS guidance), so tone and chemistry are left unset rather than inferred from sibling documents or general familiarity with the product family.", "developerFamily": "buffered-alkaline", "provenance": { "datasheetUrl": "https://www.nanofab.utah.edu/wp-content/uploads/2022/12/AZ-9260-Thick-Positive-Photoresist-Spec-Sheet-2-1.pdf", "datasheetVersionOrDate": "July 1997 (per copyright line \"© Clariant AG, July 1997\"; no separate \"Rev.\" string is printed, unlike the Merck-era AZ documents processed alongside this one)", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "[CORRECTED during audit] The extraction agent was given the archived bytes, not a URL, and fell back to a guessed microchemicals.com path (tds_az_9200_series.pdf) which returns 404 — it named a document nobody read. The bytes it actually read are recipes-data/sources/az-9245.pdf, a byte-identical copy of the archived AZ 9200 series PDS: Clariant AG / AZ Electronic Materials, “AZ 9200 Photoresist — High-Resolution Thick Resist”, © July 1997, retrieved from a University of Utah Nanofab mirror of the original vendor PDF. That mirror is now the datasheetUrl. MicroChemicals/Merck does not host this legacy datasheet (verified 404, 2026-07-10). A university mirror of the MANUFACTURER'S OWN document is acceptable as primary; a university-authored SOP is not.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "az-9260", "name": "AZ 9260", "manufacturer": "MicroChemicals GmbH / Merck KGaA (AZ brand; originally launched by Clariant / AZ Electronic Materials)", "productLine": "AZ 9200 series", "aliases": [ "AZ9260", "AZ 9260 Photoresist", "AZ 9200 Photoresist (520 cP grade)" ], "tone": "positive", "chemistry": "dnq-novolak", "grayscaleSuitable": false, "grayscaleNote": "Not addressed in any source consulted; the datasheet frames AZ 9260 purely as a high-resolution thick binary resist (etch mask / plating mold), with no grayscale dose-response or reflow data given.", "status": "active", "successorSlug": null, "summary": "AZ 9260 (520 cP) is the high-viscosity grade of the AZ 9200 high-resolution thick positive photoresist family, sensitive to both h-line and i-line UV, used for film thicknesses of roughly 5-20 um in a single coat (up to 24 um with a double coat). It was developed under guidance from thin-film recording-head manufacturers for coil electroplating and top-pole applications, and is also used as a high-aspect-ratio etch mask and, with care, for lift-off despite being a positive resist.", "thicknessRange": { "min_um": 7.9, "max_um": 11.4, "basis": "curve-span", "source": "curve-span: the AZ 9200 datasheet states no AZ 9260-specific achievable range — its intro figure \"film thicknesses of 4 to 24 µm\" (p.1) is for the two-grade AZ 9200 series (AZ 9245 + AZ 9260) combined, the same series-vs-grade case for which the sibling grade AZ 9245 also uses curve-span. min/max are the span of AZ 9260's own published \"Film Thickness\" table (p.7, 520 cP row: 11.4/9.6/8.8/7.9 µm at 2000/2500/3000/3500 rpm). The document's \"10 µm\" process (p.6) is one demonstrated single-coat point and its \"24 µm\" process is an explicit double coat, so neither is a stated single-coat achievable range. [CORRECTED: the prior 5-20 µm was sourced from neither the prose nor the curve.]" }, "spinCurves": [ { "label": "AZ 9260 Photoresist (520 cP)", "points": [ { "rpm": 2000, "um": 11.4 }, { "rpm": 2500, "um": 9.6 }, { "rpm": 3000, "um": 8.8 }, { "rpm": 3500, "um": 7.9 } ], "source": "numeric table, 'Film Thickness' row for AZ 9260 Photoresist 520 cP, Thermal Comparison section, p.7 of AZ 9200 Photoresist -- High-Resolution Thick Resist Product Data Sheet (Clariant / AZ Electronic Materials, copyright 1997). Only 4 points are published (2000-3500 rpm); the datasheet does not plot or tabulate points below 2000 or above 3500 rpm for this grade. Cross-check: interpolating this table to 2400 rpm gives ~10.0 um, matching the same document's separately-stated 'Typical Process for 10 um Film Thickness' recipe (2400 rpm), which is a useful internal consistency check." } ], "spinNotes": "Coat by dispense (static or dynamic) at 300 rpm. Kayaku-equivalent 'Typical Process' recipes from the same datasheet: for a single-coat ~10 um film, spin at 2400 rpm for 60 s; for a double-coat ~24 um stack, first coat at 2400 rpm/60 s (target ~10 um) then a second coat at 2100 rpm/60 s (target 24 um total), with an intermediate softbake between coats. Edge-bead removal in all cases: rinse at 500 rpm for 10 s, then dry-spin at 1000 rpm for 10 s. Post-develop rinse/dry: 300 rpm for 20 s rinse, then 4000 rpm for 15 s dry. Separately, current MicroChemicals process guidance (app-note level, not the 1997 datasheet) recommends spinning highly-viscous resists such as AZ 9260 at an elevated spin speed to suppress edge-bead formation, and notes that at conventional 3000-4000 rpm speeds AZ 9260 reaches roughly 7 um in about 20 s.", "adhesion": { "hmds": false, "notes": "Not specifically addressed for AZ 9260 in the AZ 9200 datasheet (no HMDS step appears in any of its 'Typical Process' recipes). General MicroChemicals adhesion guidance (not resist-specific) recommends applying HMDS only from the vapour phase onto heated substrates, never from the liquid phase or in a spin coater also used for resist -- excess liquid-phase HMDS can release ammonia during softbake that crosslinks/scums the substrate-near resist. On noble metals (Ag, Au) organic promoters like HMDS are noted to be largely ineffective; a thin Ti or Cr adhesion layer is recommended instead." }, "rehydration": "AZ 9260 is explicitly used as the worked example in MicroChemicals' 'Rehydration of Photoresists' application note: a 22 um AZ 9260 film softbaked at 100 C for 20 minutes needed roughly 30 minutes of rehydration at 52% RH / 22 C to reach a stable, high development rate throughout the film thickness. With only 5 minutes of rehydration under the same conditions, the substrate-near resist stayed water-depleted and developed at a much lower rate, requiring an order-of-magnitude longer development time for full through-development and giving a worse resist profile (more dark erosion near the top of developed structures). General rule from the same source: required rehydration time increases from about 1 minute for a 1 um film to several hours for films beyond 30 um, and a high air humidity (roughly 45-50%, never below ~40%) is required in addition to time -- a low-humidity environment cannot be compensated for by waiting longer.", "softbake": { "temp_c": 110, "time_s": 165, "method": "hotplate", "notes": "165 s is the value for a single-coat 10 um target film (the closest standard 'Typical Process' entry to AZ 9260's nominal single-coat use). Related published points from the same document: 120 s at 110 C for a 4.6 um AZ 9245 (220 cP) film, and for a double-coat 24 um AZ 9260 stack, 80 s at 110 C after the first (~10 um) coat plus 160 s at 110 C after the second coat. Separately, MicroChemicals' general (non-resist-specific) rule of thumb for AZ/TI positive resists is ~100 C for 1 minute per micron of film thickness, but the vendor's own AZ 9260-specific numbers above deviate from that generic rule, which the same source explicitly anticipates for 'special thick resists' -- use the datasheet numbers, not the generic rule, for AZ 9260.", "source": "'Typical Process for 10 um Film Thickness [AZ 9260 Photoresist (520 CPS)]' table, p.6 of AZ 9200 Photoresist Product Data Sheet (Clariant / AZ Electronic Materials, 1997); softbake rule-of-thumb cross-reference from MicroChemicals 'Basics of Microstructuring' application note (undated, current)" }, "exposureDose": { "doses": [ { "wavelength_nm": null, "value_mJcm2": 1500, "source": "'Typical Process for 10 um Film Thickness' table (p.6) and 'Focus and Exposure Latitude [i-Line]' section (p.5), AZ 9200 Photoresist Product Data Sheet, Clariant / AZ Electronic Materials, 1997" } ], "datasheetBasis": "Multiple thickness- and tool-specific doses are published for AZ 9260, all on 'broadband' steppers unless noted: 900 mJ/cm2 (10% bias) for a 4.6 um AZ 9245 film; 1500 mJ/cm2 (10% bias) for a single-coat 10 um AZ 9260 film; 2100 mJ/cm2 (10% bias) for a double-coat 24 um AZ 9260 stack. The primary value recorded here (1500 mJ/cm2) is for the 10 um single-coat AZ 9260 'Typical Process'.", "_note": "The datasheet ALSO reports an explicitly i-line-attributed dose window of 660-900 mJ/cm2 for 3 um lines/spaces resolved in a 10 um AZ 9260 film on a Nikon i-line stepper (0.54 NA) -- a lower number than the 1500 mJ/cm2 broadband 'typical process' figure for the same nominal 10 um thickness, consistent with i-line photons being absorbed more efficiently than the full broadband spectrum. This i-line figure is recorded here as narrative context only, not as a second published dose, because it is a range from a specific focus/exposure-latitude resolution test (not the vendor's headline recommended-process dose), and picking a single number out of that range would be an invented point estimate rather than a published figure. Both numbers are reported here rather than reconciled or averaged." }, "peb": { "temp_c": null, "time_s": null, "notes": "Post exposure bake is explicitly listed as 'not recommended in most applications' in every one of the datasheet's Typical Process tables (4.6 um, 10 um, and 24 um double-coat processes). This is consistent with AZ 9260 being a standard DNQ-novolak positive resist (not chemically amplified), which does not require a PEB the way SU-8 or AZ 40XT do.", "source": "'Typical Process' tables, p.6 of AZ 9200 Photoresist Product Data Sheet, Clariant / AZ Electronic Materials, 1997" }, "floodExposure": { "dose_mJcm2": null, "notes": null, "source": null }, "develop": { "developer": "AZ 400K Developer (buffered, KOH-based)", "dilution": "1:4", "time_s": 180, "method": "spray, dispense temperature 27 C (immersion is also usable per current MicroChemicals guidance)", "rinse": "300 rpm spin rinse for 20 s, then 4000 rpm spin-dry for 15 s (from the 10 um single-coat 'Typical Process')", "source": "'Typical Process for 10 um Film Thickness [AZ 9260 Photoresist (520 CPS)]' table, p.6 of AZ 9200 Photoresist Product Data Sheet, Clariant / AZ Electronic Materials, 1997. Related published points: 120 s spray for a 4.6 um AZ 9245 film; 260 s spray for a 24 um double-coat AZ 9260 stack. Current MicroChemicals compatibility guidance (application-notes level) also lists AZ 326 MIF and AZ 726 MIF (both TMAH-based, metal-ion-free) as compatible developers for AZ 9260, and the original 1997 datasheet separately notes AZ 300 MIF (a TMAH developer) can be used for IC applications requiring metal-ion-free processing." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "No AZ 9260-specific hardbake step or numbers are given in the AZ 9200 datasheet itself (its own thermal-stability test instead shows the resist beginning to round/flow at 110-125 C, i.e. its softening point, not a recommended cure). General (non-resist-specific) MicroChemicals guidance elsewhere recommends a post-development hardbake at approximately 140-150 C for 5-10 minutes to improve wet-etch and alkaline-solution resistance and reduce underetching, and separately notes that positive/image-reversal AZ resists generally begin to round/reflow above their 110-130 C softening point during any subsequent thermal process (e.g. dry etch, evaporation/sputtering), which limits how hot a true 'hard' cure step can go without deforming the profile.", "source": "Thermal Comparison section, p.7 of AZ 9200 Photoresist Product Data Sheet, Clariant / AZ Electronic Materials, 1997 (softening behavior only); general hardbake/etch-resistance guidance from MicroChemicals 'Basics of Microstructuring' application note (undated, current) -- not AZ 9260-specific" }, "descum": null, "applications": [ "electroplating-molding", "etch-mask", "lift-off", "high-aspect-ratio" ], "etchResistance": "The datasheet does not give wet/dry etch-rate numbers for AZ 9260 directly. Related published facts: the resist begins to round/soften starting around 110-125 C (its softening point), which is also the temperature ceiling for most dry-etch and coating processes before profile rounding occurs; current MicroChemicals guidance states cresol-novolak-based AZ resists (which includes AZ 9260) are never stable enough for KOH/TMAH-based anisotropic silicon etching regardless of hardbake, and that a hardbake around 140-150 C improves resistance to HNO3-containing wet etchants and reduces underetching from marginal adhesion.", "liftoffSuitable": true, "platingSuitable": true, "stripper": "Per the 1997 AZ 9200 datasheet: AZ 400T and AZ 300T strippers are recommended for AZ 9200-family resists; AZ S-46 (a non-NMP solvent stripper) is specifically suited to thin-film recording-head applications. Current MicroChemicals-branded equivalents for Novolak-based positive resists like AZ 9260: AZ 100 Remover (amine solvent mixture, can be heated 60-80 C, but must be kept water-free on Cu/Al/ITO-bearing substrates), or the high-performance strippers TechniStrip P1316 (or P1331 as an alkaline-sensitive-material alternative) for cross-linked/hardbaked material.", "storage": "Per the 1997 datasheet: keep in the sealed original container, protected from light and heat; store between 30-70 F (-1 to 24 C); refrigeration is recommended and may extend shelf life.", "notes": "AZ 9260 is the 520 cP (higher-viscosity) grade of the AZ 9200 family, paired with the 220 cP AZ 9245 grade; it is one of the resists MicroChemicals specifically recommends when thick-film bubbling or cracking is a problem, because it has a lower photo-active-compound (PAC) concentration than most positive resists and therefore generates less N2 gas during exposure -- the N2 that IS generated still needs to diffuse out before it forms visible bubbles or stress-cracks, so thick coats, high exposure intensity, and inadequate softbake all raise bubbling risk; work-arounds given by the vendor include lowering exposure intensity (splitting exposure into steps with delays between them), increasing softbake time/temperature, and improving substrate adhesion. Separately, and just as important for thick coats: AZ 9260 needs real rehydration time after softbake before exposure (tens of minutes at 40-50% RH for a >20 um film per MicroChemicals' own study) or the substrate-near resist develops far too slowly and the profile suffers -- this is easy to overlook since PEB (a separate bake step) is explicitly NOT needed for this resist. Exposure dose also scales with target thickness rather than holding constant: the datasheet's own worked processes run roughly 900 mJ/cm2 (broadband) for a thin ~4.6 um coat, ~1500 mJ/cm2 for the 10 um single-coat target, and ~2100 mJ/cm2 for the 24 um double-coat stack, so a dose calibrated for one thickness should be re-benchmarked before reuse at another. Handling note from MicroChemicals' general troubleshooting guide: N2 can also form inside an unopened resist bottle over time from slow PAC decomposition, so a bottle that has been closed for a while (or recently shaken/moved) should be left to stand for 1-2 hours (longer for very viscous resists) before dispensing, to let bubbles rise out.", "developerFamily": "buffered-alkaline", "references": [ { "type": "paper", "title": "Potentialities of a new positive photoresist for the realization of thick moulds", "authors": "Conédéra et al.", "journal": "Journal of Micromechanics and Microengineering", "year": 1999, "doi": "10.1088/0960-1317/9/2/317", "url": "https://doi.org/10.1088/0960-1317/9/2/317", "accessedDate": "2026-07-12", "summary": "The founding AZ 9260 mold paper: aspect ratio 15-20 at 100 µm on a standard aligner" }, { "type": "paper", "title": "Thin film micro-transformers for future power conversion", "authors": "O'Donnell et al.", "journal": "Nineteenth Annual IEEE Applied Power Electronics Conference and Exposition, 2004. APEC '04.", "year": 2004, "doi": "10.1109/APEC.2004.1295935", "url": "https://doi.org/10.1109/APEC.2004.1295935", "accessedDate": "2026-07-12", "summary": "81 µm molds used for on-chip power micro-transformers, Q ~23 at 0.4 GHz" } ], "provenance": { "datasheetUrl": "https://www.nanofab.utah.edu/wp-content/uploads/2022/12/AZ-9260-Thick-Positive-Photoresist-Spec-Sheet-2-1.pdf", "datasheetVersionOrDate": "AZ 9200 Photoresist -- High-Resolution Thick Resist, Product Data Sheet, Clariant AG / AZ Electronic Materials, copyright July 1997 (accessed via a University of Utah Nanofab-hosted mirror of the original vendor PDF; the current commercial custodian of the AZ brand, MicroChemicals GmbH / Merck KGaA, does not appear to host this specific legacy datasheet on microchemicals.com, which returned a 404 for the AZ 9260/9200 filename pattern used by its other current photoresist TDS files during this session)", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://www.microchemicals.com/dokumente/application_notes/lithography_trouble_shooting.pdf", "what": "MicroChemicals GmbH 'Basics of Microstructuring' / lithography troubleshooting application note (current, undated) -- manufacturer-authored, not university content, but used here only for general (non-AZ-9260-specific) guidance: N2-bubble/cracking mechanism and mitigations for thick DNQ resists, generic softbake rule-of-thumb, generic hardbake/etch-resistance guidance, developer compatibility table listing AZ 9260's compatible developers, and lift-off suitability of positive resists including AZ 9260." }, { "url": "https://research.engineering.ucdavis.edu/cnm2/wp-content/uploads/sites/11/2014/07/photoresist_rehydration.pdf", "what": "MicroChemicals GmbH 'Rehydration of Photoresists' application note (2007-02-26, authored by Dr. Christian Koch of MicroChemicals), mirrored by UC Davis' cleanroom facility -- manufacturer-authored, not university-derived content, and used as the primary source for the AZ-9260-specific 22 um rehydration study numbers quoted in the rehydration field above." } ] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "az-eci-3007", "name": "AZ ECI 3007", "manufacturer": "Merck", "productLine": "AZ ECI 3000 series", "aliases": [ "ECI 3007", "AZ ECI 3007" ], "tone": "positive", "chemistry": "dnq-novolak", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed anywhere in the datasheet; the series is positioned as a general-purpose, high-throughput plasma/RIE and wet-etch resist, not for grayscale/3D lithography.", "status": "active", "successorSlug": null, "summary": "AZ ECI 3007 is the thinnest of three grades in Merck's AZ ECI 3000 series of general-purpose, high-throughput positive-tone cross-over (i-line/g-line) photoresists for plasma/RIE and wet-etch applications, spin-coating to roughly 0.6-1.15 µm across the datasheet's 1500-5000 rpm plotted range.", "thicknessRange": { "min_um": 0.6, "max_um": 1.15, "basis": "curve-span", "source": "curve-span — read from the plotted spin curve (\"SPIN CURVES (150mm Wafers)\", p.1), min at 5000 rpm and max at 1500 rpm, the full plotted range. The series-wide stated figure in APPLICATION (p.1), \"Spin coated thickness from 0.7 to 5.0µm\", spans all three ECI 3000 grades (3007/3012/3027) combined, not ECI 3007 alone, so the grade's own curve-span is used instead per extraction policy." }, "spinCurves": [ { "label": "AZ ECI 3007", "points": [ { "rpm": 1500, "um": 1.15 }, { "rpm": 2000, "um": 0.95 }, { "rpm": 3000, "um": 0.75 }, { "rpm": 4000, "um": 0.65 }, { "rpm": 5000, "um": 0.6 } ], "source": "read from figure (\"SPIN CURVES (150mm Wafers)\"), p.1 of AZ ECI 3000 Series datasheet — legend-labeled \"ECI 3007\" (blue circle marker), clearly separated (~2x factor) from the ECI 3012 and ECI 3027 traces across the full 1500-5000 rpm plotted range, unlike closely-spaced multi-grade charts seen elsewhere in this project.", "figureRead": true } ], "spinNotes": "Multi-grade chart plots ECI 3007 / ECI 3012 / ECI 3027 together (legend, distinct colors, p.1); unlike some other multi-grade AZ charts, the three traces stay well-separated (roughly 2x factor between adjacent grades) across the whole 1500-5000 rpm range, so the ECI 3007 trace could be identified with reasonable confidence — but the reading is still a visual estimate from a plotted curve, not a printed numeric table, and no point is extrapolated past the 1500-5000 rpm plotted range. No spin accel/dispense parameters, no edge-bead removal procedure, and no rehydration hold are published for this grade; AZ EBR Solvent/AZ EBR 70/30 are listed only as companion thinning/edge-bead products (COMPANION PRODUCTS, p.2) with no protocol given. Rehydration is not applicable — this is a thin film (<1.2 µm), not a thick-DNQ resist.", "adhesion": { "hmds": true, "notes": "\"Oxide forming substrates (Si, etc.) should be primed with HMDS (hexamethyl disilazane) or other suitable primer prior to coating AZ ECI 3000.\" (SUBSTRATE PREPARATION, p.10). No specific HMDS bake temp/time is published." }, "rehydration": null, "softbake": { "temp_c": null, "time_s": 60, "method": null, "notes": "No ECI 3007-specific reference-process table is published (all worked reference processes in this document are built around ECI 3012 or ECI 3027). Falling back to the series-wide Typical Process line: \"Soft Bake: 90° to 110°C*/60s\" — temperature is a range, time is a single stated value. Higher end of range improves adhesion to metals. Bakes may be performed on hotplates or in vented bake ovens (PROCESS CONSIDERATIONS, p.10); no default is stated, so method is left null.", "source": "TYPICAL PROCESS, p.1; SOFT BAKE, p.10 of AZ ECI 3000 Series datasheet." }, "exposureDose": { "doses": [], "datasheetBasis": "\"AZ ECI 3000 is sensitive to exposure wavelengths between 310 and 450nm. 365-436nm is recommended.\" (EXPOSURE, p.10). The series is described as \"positive tone cross-over (i-line/g-line) photoresists\" (APPLICATION, p.1).", "_note": "No process dose (nominal or otherwise) is published for ECI 3007 specifically. The only ECI 3007 figure is a Dose-to-Clear vs. Film-Thickness thin-film-interference (swing) curve (\"ECI 3007 i-line TFI Curve on Si\", p.6), which oscillates roughly 40-70 mJ/cm² with film thickness due to interference. That is a physical clearing-dose curve, not a stated process/nominal exposure dose — using a point from it as \"the\" dose would misrepresent a swing-curve artifact as a process recommendation, which is exactly the kind of guess this extraction avoided. All three dose fields are null." }, "peb": { "temp_c": 110, "time_s": null, "notes": "Series-wide Typical Process line: \"Post Expose Bake: 110°C/60-90s\" — temperature is a single stated value, time is a range. PEB is described elsewhere (PROCESS CONSIDERATIONS, p.10) as optional (\"may be employed to maximize process latitudes and mitigate standing wave effects\").", "source": "TYPICAL PROCESS, p.1 of AZ ECI 3000 Series datasheet." }, "floodExposure": null, "develop": { "developer": null, "dilution": null, "time_s": 60, "method": null, "rinse": null, "source": "TYPICAL PROCESS, p.1; DEVELOPERS list, COMPANION PRODUCTS, p.2 of AZ ECI 3000 Series datasheet.", "_note": "Typical Process states \"Develop: 60s Puddle or immersion Developer type: MIF\" — a generic developer family (metal-ion-free/TMAH), not a single named product, and either puddle or immersion, not one designated as primary. Companion developers listed are AZ 300MIF, 626MIF, 726MIF, AZ 926MIF; MIF developers are recommended over inorganic ones (DEVELOPING, p.10). No ECI 3007-specific reference process exists to disambiguate further." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Series-wide range only: \"Hard bake temperatures should be in the 100° to 115°C range to ensure minimal thermal distortion of the pattern.\" No time is published. Improves adhesion in wet-etch/plating applications and pattern stability in dry-etch processes.", "source": "HARD BAKE, p.10 of AZ ECI 3000 Series datasheet." }, "descum": null, "applications": [ "etch-mask", "electroplating-molding" ], "etchResistance": null, "liftoffSuitable": null, "platingSuitable": true, "stripper": "AZ 100 Remover, AZ 300T, AZ 400T, or AZ Kwik Strip™ (STRIPPING, p.10): \"AZ ECI 3000 strips readily in removers designed for DNQ/novolac type photoresists.\" Patterns baked above 140°C may cross-link and become harder to strip.", "storage": null, "notes": "AZ ECI 3007 is a general-purpose, i-line/g-line cross-over positive resist for thin (0.6-1.15 µm), high-throughput plasma/RIE and wet-etch masking — the thinnest of three grades (3007/3012/3027) in Merck's AZ ECI 3000 series, sharing chemistry, developer compatibility, and process temperature windows with its siblings. The datasheet resolves no grade-specific reference process for ECI 3007 itself — every worked exposure/PEB/develop example is built around ECI 3012 or ECI 3027 instead — so soft bake, PEB, and develop here fall back to the series-wide Typical Process line rather than a grade-specific table. The only ECI 3007-specific exposure data is a thin-film-interference (dose-to-clear swing) curve (p.6), which is a physical clearing-dose artifact, not a stated process dose; treat the series' 365-436 nm exposure window as a starting point and run a dose array rather than expect a single vendor number. Like the rest of the series it is a DNQ/novolac positive resist compatible with both TMAH (MIF) and inorganic developers, stripping in standard DNQ/novolac removers, with wet-etch and plating adhesion called out as target applications.", "developerFamily": "tmah", "references": [ { "type": "paper", "title": "Aperture-Controlled Fabrication of All-Dielectric Structural Color Pixels", "authors": "Lipp et al.", "journal": "ACS Applied Materials & Interfaces", "year": 2023, "doi": "10.1021/acsami.3c03353", "url": "https://doi.org/10.1021/acsami.3c03353", "accessedDate": "2026-07-12", "summary": "750 nm ECI 3007 patterned on a maskless writer to tune structural color" } ], "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_eci_3000_series.pdf", "datasheetVersionOrDate": "Rev. (03/21)", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "The archived PDF is Merck's own \"Technical datasheet AZ ECI 3000 Series\" (Merck branding and copyright throughout: \"© 2021 Merck KGaA, Darmstadt, Germany and/or its affiliates\"); the assigned datasheetUrl points to a MicroChemicals-hosted mirror of this same Merck document, not a MicroChemicals-authored sheet. The spin-curve legend and figure captions clearly label this grade (\"ECI 3007\"), with a well-separated, non-overlapping trace on the multi-grade spin chart (p.1) — see spinNotes for how it was identified.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "az-eci-3012", "name": "AZ ECI 3012", "manufacturer": "Merck", "productLine": "AZ ECI 3000 series", "aliases": [ "ECI 3012", "AZ ECI 3012" ], "tone": "positive", "chemistry": "dnq-novolak", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed anywhere in the datasheet; the series is positioned as a general-purpose, high-throughput plasma/RIE and wet-etch resist, not for grayscale/3D lithography.", "status": "active", "successorSlug": null, "summary": "AZ ECI 3012 is the middle grade in Merck's AZ ECI 3000 series of general-purpose, high-throughput positive-tone cross-over (i-line/g-line) photoresists, and the best-documented grade in this datasheet — dedicated i-line and g-line reference processes, wet-etch adhesion data on ITO and thermal oxide, and PROLITH modeling parameters.", "thicknessRange": { "min_um": 1.05, "max_um": 1.9, "basis": "curve-span", "source": "curve-span — read from the plotted spin curve (\"SPIN CURVES (150mm Wafers)\", p.1), min at 5000 rpm and max at 1500 rpm, the full plotted range. The series-wide stated figure in APPLICATION (p.1), \"Spin coated thickness from 0.7 to 5.0µm\", spans all three ECI 3000 grades (3007/3012/3027) combined, not ECI 3012 alone, so the grade's own curve-span is used instead per extraction policy. (Note: reference-process coat thicknesses of 1.20/1.2/1.7/1.33 µm appear on p.3-8 for various test structures, but none are paired with a spin speed, so they cannot be folded into this range.)" }, "spinCurves": [ { "label": "AZ ECI 3012", "points": [ { "rpm": 1500, "um": 1.9 }, { "rpm": 2000, "um": 1.6 }, { "rpm": 3000, "um": 1.3 }, { "rpm": 4000, "um": 1.15 }, { "rpm": 5000, "um": 1.05 } ], "source": "read from figure (\"SPIN CURVES (150mm Wafers)\"), p.1 of AZ ECI 3000 Series datasheet — legend-labeled \"ECI 3012\" (red/pink circle marker), clearly separated (~2x factor) from the ECI 3007 and ECI 3027 traces across the full 1500-5000 rpm plotted range, unlike closely-spaced multi-grade charts seen elsewhere in this project.", "figureRead": true } ], "spinNotes": "Multi-grade chart plots ECI 3007 / ECI 3012 / ECI 3027 together (legend, distinct colors, p.1); unlike some other multi-grade AZ charts, the three traces stay well-separated (roughly 2x factor between adjacent grades) across the whole 1500-5000 rpm range, so the ECI 3012 trace could be identified with reasonable confidence — but the reading is still a visual estimate from a plotted curve, not a printed numeric table, and no point is extrapolated past the 1500-5000 rpm plotted range. No spin accel/dispense parameters, no edge-bead removal procedure, and no rehydration hold are published for this grade; AZ EBR Solvent/AZ EBR 70/30 are listed only as companion thinning/edge-bead products (COMPANION PRODUCTS, p.2) with no protocol given. Rehydration is not applicable — this is a thin film (<2 µm), not a thick-DNQ resist.", "adhesion": { "hmds": true, "notes": "\"Oxide forming substrates (Si, etc.) should be primed with HMDS (hexamethyl disilazane) or other suitable primer prior to coating AZ ECI 3000.\" (SUBSTRATE PREPARATION, p.10). The oxide-etch wet-etch adhesion test (p.8) explicitly used \"Primer: HMDS vapor\"." }, "rehydration": null, "softbake": { "temp_c": 90, "time_s": 90, "method": "hotplate", "notes": "Both grade-specific reference processes agree: \"Soft Bake: 90°C, 90 seconds, proximity hotplate\" — used for the i-line reference process (p.3) and repeated verbatim for the g-line reference process (p.5). A third reference process (Broadband Mask Aligner, p.7) uses a different condition: \"90°C, 60 seconds, contact hotplate\" for that specific tool setup — not used here since the two i-line/g-line stepper references agree with each other and are the primary documented process.", "source": "REFERENCE PROCESS tables, p.3 and p.5 of AZ ECI 3000 Series datasheet." }, "exposureDose": { "doses": [ { "wavelength_nm": 436, "value_mJcm2": 220, "source": "REFERENCE PROCESS (Dense lines and holes in AZ ECI 3012 Photoresist), p.3; REFERENCE PROCESS (Dense lines in AZ ECI 3012 Photoresist, g-line exposure), p.5 of AZ ECI 3000 Series datasheet." } ], "datasheetBasis": "\"g-line @ 220mJ/cm2 nominal (0.54NA) Nikon Stepper\" — REFERENCE PROCESS (Dense lines in AZ ECI 3012 Photoresist, g-line exposure), p.5.", "_note": "The i-line reference process for this grade (p.3) states \"i-line @ 110-136mJ/cm2 nominal (0.54NA, 0.6 s) Nikon Stepper\" — a range spanning two different feature targets: 110 mJ/cm² for dense lines (\"Dense Line Linearity @ 110 mJ/cm²\") and 136 mJ/cm² for dense holes (\"Dense Hole Linearity @ 136mJ/cm²\"), not a single per-grade nominal. Picking one of the two would be a guess, so at365_mJcm2 is left null. The g-line reference process (p.5) instead gives one unambiguous nominal, \"220mJ/cm2\", used here as value_mJcm2 since g-line (436 nm) routes to the unlabeled-wavelength field per extraction rules, not at365/at405. No h-line (405 nm) dose is published anywhere in the document." }, "peb": { "temp_c": 110, "time_s": null, "notes": "Both grade-specific reference processes agree on temperature (110°C) but differ on time and hotplate type by exposure wavelength: the i-line reference process (p.3) uses 60 seconds on a direct-contact hotplate; the g-line reference process (p.5) uses 90 seconds on a proximity hotplate. Time is left null rather than merging the two — use the pairing that matches the intended exposure wavelength/tool.", "source": "REFERENCE PROCESS tables, p.3 and p.5 of AZ ECI 3000 Series datasheet." }, "floodExposure": null, "develop": { "developer": "AZ 300MIF", "dilution": "used as supplied (ready-to-use MIF developer)", "time_s": 60, "method": "puddle", "rinse": null, "source": "REFERENCE PROCESS tables, p.3, p.5, p.7 and WET ETCH ADHESION CHARACTERISTICS, p.8 of AZ ECI 3000 Series datasheet.", "_note": "Every reference process for this grade (i-line, g-line, broadband mask aligner, and both wet-etch adhesion tests) agrees: \"AZ 300MIF, 60s single puddle @ 23°C.\" This is the most internally consistent value extracted for this grade." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Series-wide range only: \"Hard bake temperatures should be in the 100° to 115°C range to ensure minimal thermal distortion of the pattern.\" No time is published. Improves adhesion in wet-etch/plating applications and pattern stability in dry-etch processes.", "source": "HARD BAKE, p.10 of AZ ECI 3000 Series datasheet." }, "descum": null, "applications": [ "etch-mask", "electroplating-molding" ], "etchResistance": "Wet-etch adhesion data (p.8) shows AZ ECI 3012 surviving a 70s FeCl3/HCl ITO etch at 45°C (200nm ITO film) and a 6-minute thermal-oxide etch at 22°C, with acceptable pattern retention in the accompanying SEM cross-sections (8.0µm/6.0µm lines post-ITO-etch; 100µm pad edge and 7.0µm lines post-oxide-etch). No numeric etch rate or resist:film selectivity ratio is published.", "liftoffSuitable": null, "platingSuitable": true, "stripper": "AZ 100 Remover, AZ 300T, AZ 400T, or AZ Kwik Strip™ (STRIPPING, p.10): \"AZ ECI 3000 strips readily in removers designed for DNQ/novolac type photoresists.\" Patterns baked above 140°C may cross-link and become harder to remove.", "storage": null, "notes": "AZ ECI 3012 is the middle grade in Merck's AZ ECI 3000 cross-over series and the best-documented grade in this datasheet, with dedicated i-line (110/136 mJ/cm² nominal for dense lines/holes respectively) and g-line (220 mJ/cm² nominal) reference processes, PROLITH Dill and development-rate parameters, and demonstrated wet-etch adhesion on ITO (FeCl3/HCl, 70s) and thermal oxide (6 min). Its two documented PEB conditions differ by exposure wavelength (60s on a direct-contact hotplate for i-line vs. 90s on a proximity hotplate for g-line, both at 110°C), so PEB time is published as null rather than merged — a QC reviewer should pick the pairing matching the intended exposure tool. The i-line reference dose is itself split between two feature targets (110 mJ/cm² for lines, 136 mJ/cm² for holes) rather than one nominal number, which is why it is not used for the scalar dose field. Like the rest of the series it is a DNQ/novolac positive resist compatible with both TMAH (MIF, AZ 300MIF used throughout every reference process for this grade) and inorganic developers, stripping in standard DNQ/novolac removers.", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_eci_3000_series.pdf", "datasheetVersionOrDate": "Rev. (03/21)", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "The archived PDF is Merck's own \"Technical datasheet AZ ECI 3000 Series\" (Merck branding and copyright throughout: \"© 2021 Merck KGaA, Darmstadt, Germany and/or its affiliates\"); the assigned datasheetUrl points to a MicroChemicals-hosted mirror of this same Merck document, not a MicroChemicals-authored sheet. The spin-curve legend and multiple reference-process tables clearly label this grade (\"ECI 3012\"/\"AZ ECI 3012\"), with a well-separated, non-overlapping trace on the multi-grade spin chart (p.1) — see spinNotes for how it was identified.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "az-eci-3027", "name": "AZ ECI 3027", "manufacturer": "Merck (AZ Electronic Materials)", "productLine": "AZ ECI 3000 series", "aliases": [ "AZ ECI3027", "ECI 3027", "AZ ECI 3027 Photoresist" ], "tone": "positive", "chemistry": "dnq-novolak", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Datasheet does not address grayscale or 3D lithography anywhere; AZ ECI 3000 is marketed as a general-purpose, high-throughput cross-over (i-line/g-line) resist for plasma/RIE and wet-etch masking, with all resolution/linearity data framed around binary dense-line and dense-hole patterning.", "status": "active", "successorSlug": null, "summary": "AZ ECI 3027 is the thickest grade in Merck's AZ ECI 3000 series of positive-tone, i-line/g-line cross-over photoresists for plasma/RIE and wet-etch applications, coating to roughly 2.4-4.4 µm and documented in this datasheet with a full reference process at a 2.5 µm film thickness resolving 0.6 µm dense lines.", "thicknessRange": { "min_um": 2.4, "max_um": 4.4, "basis": "curve-span", "source": "curve-span: taken from the min/max of the AZ ECI 3027 curve on the p.1 spin-speed chart (1500-5000 rpm tested). The series-wide prose claim of 'Spin coated thickness from 0.7 to 5.0µm' (p.1, APPLICATION bullets) spans all three grades in the series (ECI 3007 thinnest to ECI 3027 thickest) together, not ECI 3027 alone, so it was not used here. This curve-span is consistent with the REFERENCE PROCESS on p.4, which coats AZ ECI 3027 at 2.5 µm — inside the read range." }, "spinCurves": [ { "label": "AZ ECI 3027 as supplied", "points": [ { "rpm": 1500, "um": 4.4 }, { "rpm": 2000, "um": 3.8 }, { "rpm": 3000, "um": 3.1 }, { "rpm": 4000, "um": 2.7 }, { "rpm": 5000, "um": 2.4 } ], "source": "read from figure, \"SPIN CURVES (150mm Wafers)\" chart, p.1 of AZ ECI 3000 Series Technical Datasheet (Merck, Rev. 03/21); AZ ECI 3027 identified by its purple/violet legend marker (topmost curve), distinct from AZ ECI 3007 (blue, bottom curve) and AZ ECI 3012 (red/magenta, middle curve) plotted on the same axes.", "figureRead": true } ], "spinNotes": "Datasheet does not state spin ramp/acceleration, dispense volume, or static vs dynamic dispense for the spin-curve chart. Spin coating is named as one of several compatible coating methods (spray and roller coating are also mentioned, p.10 COATING) but no method-specific parameters are published beyond that. No edge-bead-removal recipe is given, only that AZ EBR Solvent or AZ EBR 70/30 are the companion EBR products (p.2).", "adhesion": { "hmds": true, "notes": "\"Oxide forming substrates (Si, etc.) should be primed with HMDS (hexamethyl disilazane) or other suitable primer prior to coating AZ ECI 3000. Contact your products representative for detailed information on pre-treating with HMDS.\" Source: SUBSTRATE PREPARATION, p.10. (A p.8 example process for AZ ECI 3012 — a different grade in this series — separately used 'HMDS vapor' as the primer ahead of an oxide-etch adhesion test, consistent with this recommendation, but that example is not AZ ECI 3027-specific.)" }, "rehydration": null, "softbake": { "temp_c": 100, "time_s": 60, "method": "hotplate", "notes": "This is the REFERENCE PROCESS condition for AZ ECI 3027 (2.5 µm film on bare Si) on p.4, using a proximity hotplate. The series-wide TYPICAL PROCESS section (p.1) separately states a broader 90-110°C range with a fixed 60 s time for the whole ECI 3000 series (higher end of the range improves adhesion to metals); the AZ ECI 3027-specific reference-process value (100°C) sits inside that range.", "source": "REFERENCE PROCESS (Dense lines in AZ ECI 3027 Photoresist), p.4; TYPICAL PROCESS, p.1" }, "exposureDose": { "doses": [ { "wavelength_nm": 365, "value_mJcm2": 262, "source": "REFERENCE PROCESS (Dense lines in AZ ECI 3027 Photoresist), p.4" } ], "datasheetBasis": "i-line (365 nm) nominal dose stated for the REFERENCE PROCESS (Dense lines in AZ ECI 3027 Photoresist), p.4: \"i-line @ 262mJ/cm2 nominal* (0.54NA, 0.6s) Nikon Stepper\", at a 2.5 µm coated film thickness on bare Si, developed with AZ 726MIF. Series-wide TYPICAL PROCESS (p.1) separately states only \"310-450nm sensitive\" without a specific dose.", "_note": "262 mJ/cm² is the nominal i-line dose for a 2.5 µm AZ ECI 3027 film per the p.4 REFERENCE PROCESS table, paired with an AZ 726MIF develop step. That same table carries a footnote: \"Adjust nominal dose for this film thickness to ~240mJ/cm2 when using AZ 300MIF Developer\" — only the primary 262 mJ/cm² value (matching the developer actually used in that reference process, AZ 726MIF) is recorded in the single at365 field, to avoid representing two developer-dependent doses as one number. Dose at other film thicknesses will differ (the p.6 i-line thin-film-interference curve for ECI 3027 shows dose-to-clear oscillating roughly 110-150 mJ/cm² across a 2.1-2.7 µm thickness sweep under a separate, unspecified fixed test condition — not the same as the 262 mJ/cm² reference-process dose, and not substituted here). h-line (405 nm) dose is not published anywhere for this resist; the p.9 PROLITH Dill parameters table gives i-line/h-line/g-line absorption coefficients (A/B/C) for optical modeling, not a recommended exposure dose, so it was not used to fill at405_mJcm2." }, "peb": { "temp_c": 120, "time_s": 60, "notes": "REFERENCE PROCESS condition for AZ ECI 3027 (2.5 µm film), proximity hotplate. The series-wide TYPICAL PROCESS (p.1) separately states 110°C for 60-90s for the whole ECI 3000 series (single temperature value there, but a time range); the AZ ECI 3027-specific reference-process condition (120°C/60s) is recorded here in preference to the generic series line, since it applies to this exact resist and film thickness.", "source": "REFERENCE PROCESS (Dense lines in AZ ECI 3027 Photoresist), p.4; TYPICAL PROCESS, p.1" }, "floodExposure": null, "develop": { "developer": "AZ 726MIF", "dilution": null, "time_s": 60, "method": "puddle", "rinse": null, "source": "REFERENCE PROCESS (Dense lines in AZ ECI 3027 Photoresist), p.4 (\"Develop: AZ 726MIF, 60s single puddle @ 23°C\")", "_note": "This is the specific REFERENCE PROCESS develop condition for AZ ECI 3027 (2.5 µm film): AZ 726MIF, single puddle, 60 s, at a controlled 23°C (the 23°C bath temperature isn't a schema field but is recorded here since it's a real, sourced parameter). The series-wide TYPICAL PROCESS (p.1) separately allows 'puddle or immersion' generically with any MIF developer type; AZ 300MIF is the developer used in the analogous AZ ECI 3012 reference processes elsewhere in this datasheet (p.3, p.5, p.7, p.8) and is the noted alternate for AZ ECI 3027 at ~240 mJ/cm² (see exposureDose._note), but AZ 726MIF specifically is what the p.4 reference process for AZ ECI 3027 actually used. No dilution is stated for either MIF developer anywhere in this document." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Hard bake temperature should be in the 100-115°C range to minimize thermal pattern distortion; improves adhesion in wet-etch or plating applications and pattern stability in dry etch processes. No time is given, and temperature is stated as a range, not a single value. No AZ ECI 3027-specific hardbake/postbake step is documented — the p.4 REFERENCE PROCESS table for AZ ECI 3027 stops at Develop.", "source": "PROCESS CONSIDERATIONS / HARD BAKE, p.10" }, "descum": null, "applications": [ "etch-mask", "electroplating-molding", "general-prototyping" ], "etchResistance": "Datasheet demonstrates wet-etch adhesion/profile retention for AZ ECI 3012 (a different, thinner grade in the same series, not AZ ECI 3027 specifically) through an ITO etch (FeCl3/HCl, 70s @ 45°C, on a 200nm ITO film) and a thermal-oxide etch (6 min @ 22°C, on a 690nm oxide film), both shown with SEM cross-sections retaining resist profile (WET ETCH ADHESION CHARACTERISTICS, p.8). No etch-resistance data specific to AZ ECI 3027 is published; the series is marketed generally for 'plasma/RIE and wet etching applications' (APPLICATION, p.1).", "liftoffSuitable": null, "platingSuitable": true, "stripper": "AZ 100 Remover, AZ 300T, AZ 400T, or AZ Kwik Strip — removers designed for DNQ/novolac type photoresists. Strip times vary with thermal history; patterns processed above 140°C may cross-link and become harder to strip, and charred resist will not dissolve in solvent-based removers. Source: STRIPPING, p.10.", "storage": null, "notes": "AZ ECI 3027 is the thickest grade in Merck's AZ ECI 3000 series, a general-purpose, high-throughput positive DNQ/novolak cross-over resist explicitly usable at both i-line (365 nm) and g-line (436 nm), marketed for plasma/RIE and wet-etch masking (the DNQ/novolak classification comes directly from the datasheet's own stripper-compatibility statement, p.10). Unlike AZ ECI 3007/3012 (documented elsewhere in this datasheet at ~1.2 µm film with resolution down to 0.4 µm lines/0.5 µm holes), the dedicated AZ ECI 3027 reference process (p.4) is run at 2.5 µm film thickness — 100°C/60s soft bake, i-line exposure at 262 mJ/cm² nominal (0.54NA, 0.6s, Nikon stepper; ~240 mJ/cm² if developing with AZ 300MIF instead of the AZ 726MIF actually used), 120°C/60s PEB, and a 60s AZ 726MIF single-puddle develop at 23°C — and resolves dense lines down to roughly 0.60 µm at that dose, per the p.4 Dense Line Linearity chart. The series-wide 'production resolution to 0.4µm' headline claim (p.1) applies to the thinner grades' reference processes, not to AZ ECI 3027 at its documented 2.5 µm thickness. On the p.1 multi-grade spin-speed chart (ECI 3007/3012/3027 plotted together), the AZ ECI 3027 curve was identified by its distinct purple/violet legend marker as the topmost (thickest) curve; the reported points are a figure read, not a printed table, and warrant a visual QC pass against the source chart. Soft bake, PEB, and hard bake are also given as series-wide ranges (90-110°C, 105-115°C, 100-115°C respectively) in the generic TYPICAL PROCESS/PROCESS CONSIDERATIONS sections — the reference-process values above were preferred where available since they are resist- and thickness-specific.", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_eci_3000_series.pdf", "datasheetVersionOrDate": "Rev. (03/21)", "accessedDate": "2026-07-10", "secondarySources": [] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "az-nlof-2020", "name": "AZ nLOF 2020", "manufacturer": "Merck", "productLine": "AZ nLOF 2000 Series", "aliases": [ "AZ nLOF™ 2020", "nLOF 2020", "AZ nLOF 2000 Series" ], "tone": "negative", "chemistry": "car", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Datasheet does not address grayscale or 3D patterning for AZ nLOF 2020; no such use is claimed anywhere in the document.", "status": "active", "successorSlug": null, "summary": "AZ nLOF 2020 is the ~2 µm grade of AZ's chemically amplified negative-tone nLOF 2000 series, engineered for single-layer lift-off with a required post-exposure bake and no rehydration wait.", "thicknessRange": { "min_um": 1.6, "max_um": 4.6, "basis": "curve-span", "source": "curve-span — the p.1 series-level bullet ('Single coat thicknesses from 2.0 to >10µm') describes the AZ nLOF 2000 Series as a whole across all three grades (2020/2035/2070) plotted together on the p.1 spin-curve chart, not this grade alone; applying it to nLOF 2020 specifically would overstate its range, since the 2020 curve (bottom, blue) tops out well under 10 µm. The grade-specific range used here is the span of the correctly-identified nLOF 2020 curve (500-4000 rpm) on that same chart. The 2.0 µm nominal reference-process thickness (p.3, p.6-7) sits inside this span." }, "spinCurves": [ { "label": "AZ nLOF 2020", "points": [ { "rpm": 500, "um": 4.6 }, { "rpm": 1000, "um": 3.3 }, { "rpm": 1500, "um": 2.8 }, { "rpm": 2000, "um": 2.4 }, { "rpm": 2500, "um": 2.1 }, { "rpm": 3000, "um": 1.9 }, { "rpm": 3500, "um": 1.7 }, { "rpm": 4000, "um": 1.6 } ], "source": "read from figure, \"SPIN CURVES (150mm Silicon)\", p.1 of AZ nLOF 2000 Series Technical datasheet (Rev. 03/21) — chart plots three grades (nLOF 2070 red, nLOF 2035 yellow, nLOF 2020 blue) with an explicit color-coded legend; nLOF 2020 identified by its blue 'nLOF 2020' legend entry (bottom curve, lowest thickness at every speed) and cross-checked against the grade-specific 2.0 µm reference process (p.3, p.6-7, which names 'AZ nLOF 2020 (33cPs)' explicitly).", "figureRead": true } ], "spinNotes": "This grade is the bottom (lowest-thickness-at-every-speed) trace on the three-grade family chart, identified by its blue 'nLOF 2020' legend entry — and corroborated a second way, beyond the legend color alone: this datasheet's own 2.0 µm reference process names the exact resist run as 'AZ nLOF 2020 (33cPs)' (p.3, p.6-7), matching the lowest-viscosity trace on the chart. Still a figure read, not a numeric table, so visual QC is required. No dispense volume, spin ramp, or edge-bead detail is published in this datasheet.", "adhesion": { "hmds": true, "notes": "Oxide-forming substrates (e.g. Si) should be HMDS primed prior to coating (PROCESS CONSIDERATIONS > SUBSTRATE PREPARATION, p.8). The 2.0 µm reference process specifies a vapor prime of HMDS 140°C/60s (p.3, p.6-7)." }, "rehydration": "None required — p.1 TYPICAL PROCESS states 'Rehydration Hold: None', consistent with this being a chemically amplified resist, not a DNQ resist. (Source: TYPICAL PROCESS, p.1 of AZ nLOF 2000 Series Technical datasheet (Rev. 03/21))", "softbake": { "temp_c": 110, "time_s": 60, "method": "hotplate", "notes": "Consistent across every 2.0 µm nLOF 2020 reference process in this datasheet (p.3, p.6, p.7): 110°C, 60 s, direct contact hotplate. Series-level guidance (p.8) puts soft bake temperature generally in the 100-110°C range, and recommends minimizing any delay between soft bake and exposure.", "source": "EXAMPLE PROCESS (2.0µm Film Thickness on Si), p.3 of AZ nLOF 2000 Series Technical datasheet (Rev. 03/21)" }, "exposureDose": { "doses": [ { "wavelength_nm": 365, "value_mJcm2": 66, "source": "EXAMPLE PROCESS (2.0µm Film Thickness on Si), p.3 of AZ nLOF 2000 Series Technical datasheet (Rev. 03/21)" } ], "datasheetBasis": "\"Exposure: i-line @ 66mJ/cm2 nominal (0.54NA) Nikon Stepper\" — verbatim, EXAMPLE PROCESS (2.0µm Film Thickness on Si), p.3. Confirmed series-wide: \"AZ nLOF 2000 requires exposure energy at the 365nm wavelength\" (p.8).", "_note": "66 mJ/cm² is the specific nominal dose for the 2.0 µm nLOF 2020 reference process (p.3), reproduced identically in the SAMPLE PROCESS WINDOWS summary (p.6: Soft Bake 110C/60s, PEB 110C/60s, Develop AZ 300MIF 60s puddle — matches p.3 exactly). The through-dose exposure-latitude sweep on the same page (p.3) shows this process stays printable from 62-74 mJ/cm², i.e. the 66 mJ/cm² nominal sits inside its own latitude window, not at an edge." }, "peb": { "temp_c": 110, "time_s": 60, "notes": "PEB is REQUIRED for proper imaging (p.1, p.8) — the critical step for nLOF, a chemically amplified negative resist, unlike the optional or absent PEB on the DNQ positive resists elsewhere in this recipe set. Reference process (p.3, p.6-7): 110°C, 60 seconds, direct contact hotplate, matching the series-level typical-process summary (p.1: 'Post Expose Bake: 110ºC/60s'). Series-level guidance (p.8) allows 100-115°C generally, and warns CD is sensitive to PEB temperature at roughly <0.04 µm/°C (the datasheet's own measured example — slope 0.038 µm/°C over 105-115°C — was run on the related 3.5 µm nLOF 2035 grade, p.5, not nLOF 2020 itself, so it is cited here as series-level supporting evidence only).", "source": "EXAMPLE PROCESS (2.0µm Film Thickness on Si), p.3, and PROCESS CONSIDERATIONS > POST EXPOSE BAKE, p.8, of AZ nLOF 2000 Series Technical datasheet (Rev. 03/21)" }, "floodExposure": null, "develop": { "developer": "AZ 300MIF", "dilution": "undiluted (ready-to-use 0.26N / 2.38% TMAH developer; no dilution ratio stated)", "time_s": 60, "method": "puddle", "rinse": null, "source": "EXAMPLE PROCESS (2.0µm Film Thickness on Si), p.3 of AZ nLOF 2000 Series Technical datasheet (Rev. 03/21): 'AZ 300MIF, 60s single puddle'", "_note": "Consistent across every 2.0 µm nLOF 2020 reference process in this datasheet (p.3, p.6, p.7). PROCESS CONSIDERATIONS (p.8) recommends AZ 300MIF, an industry-standard 0.26N (2.38%) TMAH developer, for the whole nLOF 2000 series." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Series-level claim only, no grade-specific number for nLOF 2020: 'AZ nLOF materials are extremely thermally stable and may be hard baked at temperatures above 150°C' (p.8) — an open-ended floor, not a target temperature/time, left null rather than guessed. The datasheet's only quantified hard-bake stability demonstration (115-130°C, large pads, qualitative SEM images) was run on the 7.0 µm nLOF 2070 grade (p.8), not on nLOF 2020.", "source": "PROCESS CONSIDERATIONS > HARD BAKE, p.8 of AZ nLOF 2000 Series Technical datasheet (Rev. 03/21)" }, "descum": null, "applications": [ "lift-off", "etch-mask" ], "etchResistance": null, "liftoffSuitable": true, "platingSuitable": null, "stripper": "AZ 400T or AZ Remover 770 (solvent-based removers), recommended per PROCESS CONSIDERATIONS > STRIPPING, p.8 of AZ nLOF 2000 Series Technical datasheet.", "storage": null, "notes": "AZ nLOF 2020 is the thinnest (~2 µm) grade of AZ's chemically amplified nLOF 2000 lift-off series, and the only one of this recipe set's nLOF grades with a single stated nominal exposure dose rather than a dose-latitude sweep with no headline number: its 66 mJ/cm² i-line nominal sits comfortably inside the datasheet's own 62-74 mJ/cm² through-dose printable window, not pinned to an edge. As with the rest of the series, PEB is REQUIRED, not optional as on the DNQ positive resists elsewhere in this project, and is the step that most directly sets the negative-tone undercut lift-off profile — the datasheet's own CD-vs-PEB-temperature data (measured on the related 2035 grade) puts the sensitivity at roughly <0.04 µm/°C, so a few degrees of hotplate drift shows up as a real dimensional shift. Develop is a single 60 s AZ 300MIF puddle, simpler than the two-puddle cycle the thicker 2070 grade needs to fully clear its film. No rehydration wait is needed after softbake, unlike the thick DNQ positive resists elsewhere in this set, and the cured film's >150°C thermal stability is what lets the undercut profile survive a subsequent metal evaporation and lift-off step. The datasheet explicitly does not recommend the nLOF 2000 series for use on copper substrates.", "developerFamily": "tmah", "references": [ { "type": "paper", "title": "Performances of the Negative Tone Resist AZnLOF 2020 for Nanotechnology Applications", "authors": "Herth et al.", "journal": "IEEE Transactions on Nanotechnology", "year": 2012, "doi": "10.1109/TNANO.2012.2196802", "url": "https://doi.org/10.1109/TNANO.2012.2196802", "accessedDate": "2026-07-12", "summary": "The most-cited nLOF 2020 characterization: 50 nm lines at 100 nm pitch by e-beam" }, { "type": "paper", "title": "Spoken Digit Classification by In-Materio Reservoir Computing With Neuromorphic Atomic Switch Networks", "authors": "Lilak et al.", "journal": "Frontiers in Nanotechnology", "year": 2021, "doi": "10.3389/fnano.2021.675792", "url": "https://doi.org/10.3389/fnano.2021.675792", "accessedDate": "2026-07-12", "summary": "Lift-off Pt grid for a neuromorphic atomic-switch-network chip" } ], "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_nlof2000_series.pdf", "datasheetVersionOrDate": "Rev. (03/21)", "accessedDate": "2026-07-10", "secondarySources": [] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "az-nlof-2035", "name": "AZ nLOF 2035", "manufacturer": "Merck", "productLine": "AZ nLOF 2000 series", "aliases": [ "nLOF 2035", "AZ nLOF 2035", "AZ nLOF™ 2035" ], "tone": "negative", "chemistry": "car", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed anywhere in the datasheet; the series is positioned specifically for single-layer lift-off and RIE-etch pattern transfer, not grayscale/3D lithography.", "status": "active", "successorSlug": null, "summary": "AZ nLOF 2035 is the middle grade in Merck's AZ nLOF 2000 series of chemically-amplified, i-line-sensitive negative photoresists engineered to replace image-reversal processing for single-layer lift-off, producing an undercut sidewall directly from a standard expose/PEB/develop flow, thermally stable to >200°C.", "thicknessRange": { "min_um": 3, "max_um": 6.1, "basis": "curve-span", "source": "curve-span — read from the plotted spin curve (\"SPIN CURVES (150mm Silicon)\", p.1), min at 4000 rpm and max at 500 rpm, the full plotted range. This is corroborated by the grade-specific EXAMPLE PROCESS table (p.4), which coats \"3.5µm thick film AZ nLOF 2035 (79cPs)\" — a value that falls within the read curve-span, between the 2500 rpm (~3.9µm) and 3000 rpm (~3.6µm) points. The series-wide stated figure in APPLICATION (p.1), \"Single coat thicknesses from 2.0 to >10µm\", spans all three nLOF 2000 grades (2020/2035/2070) combined, not nLOF 2035 alone, so the grade's own curve-span is used instead per extraction policy." }, "spinCurves": [ { "label": "AZ nLOF 2035", "points": [ { "rpm": 500, "um": 6.1 }, { "rpm": 1000, "um": 5.6 }, { "rpm": 1500, "um": 4.9 }, { "rpm": 2000, "um": 4.2 }, { "rpm": 2500, "um": 3.9 }, { "rpm": 3000, "um": 3.6 }, { "rpm": 3500, "um": 3.3 }, { "rpm": 4000, "um": 3 } ], "source": "read from figure (\"SPIN CURVES (150mm Silicon)\"), p.1 of AZ nLOF 2000 Series datasheet — legend-labeled \"nLOF 2035\" (yellow/orange square marker), distinguishable from the nLOF 2070 (red diamond) and nLOF 2020 (blue diamond) traces by color and marker shape across the full 500-4000 rpm plotted range. Cross-checked against the grade-specific 3.5µm coat thickness stated in the EXAMPLE PROCESS table (p.4), which falls within this reading between the 2500 and 3000 rpm points.", "figureRead": true } ], "spinNotes": "Multi-grade chart plots nLOF 2070 / nLOF 2035 / nLOF 2020 together (legend, distinct colors/markers, p.1); the nLOF 2035 trace sits clearly between the other two across the plotted range with a stated grade-specific 3.5µm anchor point (EXAMPLE PROCESS, p.4) that agrees with the read curve, giving reasonable confidence — but the reading is still a visual estimate from a plotted curve, not a printed numeric table, and no point is extrapolated past the 500-4000 rpm plotted range. No spin accel/dispense parameters or edge-bead removal procedure are published; AZ EBR Solvent/AZ EBR 70/30 are listed only as companion thinning/edge-bead products (COMPANION PRODUCTS, p.2) with no protocol given. No rehydration hold applies to this resist — the Typical Process table explicitly states \"Rehydration Hold: None\" (p.1); see the rehydration field.", "adhesion": { "hmds": true, "notes": "Grade-specific EXAMPLE PROCESS (p.4): \"Prime: HMDS 140°C/60s (vapor)\". General guidance (SUBSTRATE PREPARATION, p.8): \"Oxide forming substrates (Si, etc.) should be HMDS primed prior to coating AZ nLOF 2000\", no generic temp/time given there. Important caveat: \"AZ nLOF 2000 series photoresists are not recommended for use on copper substrates\" (COMPATIBLE MATERIALS, p.9)." }, "rehydration": "None. The Typical Process table explicitly states \"Rehydration Hold: None\" (TYPICAL PROCESS, p.1); AZ nLOF 2000 is a chemically-amplified negative resist, not a thick-DNQ resist, and this series does not require a rehydration hold.", "softbake": { "temp_c": 110, "time_s": 60, "method": "hotplate", "notes": "Grade-specific EXAMPLE PROCESS (3.5µm Film Thickness on Si, p.4): \"Soft Bake: 110°C, 60s, direct contact hotplate\", preceded by an HMDS vapor prime and followed by \"Post Bake Delay: None\". Consistent with the series-wide range in PROCESS CONSIDERATIONS (p.8): \"Soft bake temperatures for AZ nLOF 2000 should be in the 100°-110°C range\" (this grade's example sits at the high end) and \"Delays between soft bake and exposure should be minimized for optimum performance.\"", "source": "EXAMPLE PROCESS (3.5µm Film Thickness on Si), p.4; SOFT BAKE, p.8 of AZ nLOF 2000 Series datasheet." }, "exposureDose": { "doses": [ { "wavelength_nm": 365, "value_mJcm2": 80, "source": "EXAMPLE PROCESS (3.5µm Film Thickness on Si), p.4 of AZ nLOF 2000 Series datasheet." } ], "datasheetBasis": "\"i-line @ 80mJ/cm2 nominal (0.548NA) Nikon Stepper*\" — EXAMPLE PROCESS (3.5µm Film Thickness on Si), p.4. Also: \"AZ nLOF 2000 requires exposure energy at the 365nm wavelength\" (EXPOSURE, p.8) and the Typical Process line \"Expose: 365nm sensitive\" (p.1).", "_note": "The table's own asterisk qualifies this as tunable (\"Pattern profiles can be modified by varying exposure dose and PEB temperature\"), but 80 mJ/cm² is still the document's own stated single nominal for this grade, so it is used. The adjacent \"2.0µm Lines Through Dose\" panel (p.4) sweeps 72/80/88/96 mJ/cm² around this nominal — those sweep points are not used as the scalar. A resolution image on p.2 (\"2.0µm lines and 2.0µm iso trench, 3.5µm thick AZ nLOF 2035, 72mJ/cm² i-line Exposure\") uses 72 mJ/cm² — one of the through-dose sweep points, not a competing nominal, so it does not override 80. Not used: a second Linearity/Exposure-Latitude panel on p.6 that is explicitly captioned \"Coat: AZ nLOF 2020 @ FT=3.5µm\" — despite the 3.5µm thickness and matching 120s develop time (both of which otherwise match the nLOF 2035 example), the document names that panel's resist as nLOF 2020, not 2035; this looks like a possible labeling error in the source datasheet but is not resolved here (see notes/FINAL MESSAGE — flag for QC)." }, "peb": { "temp_c": 110, "time_s": 60, "notes": "Grade-specific EXAMPLE PROCESS (p.4): \"Post Expose Bake: 110°C*, 60 seconds, direct contact hotplate\" (asterisk = same tunability caveat as exposure dose). PEB is REQUIRED, not optional: \"* PEB is required for proper imaging\" (p.1) and \"A PEB is required for proper imaging of AZ nLOF 2000\" (POST EXPOSE BAKE, p.8) — this is the critical step for this chemically-amplified negative resist. \"CD's in nLOF 2000 will exhibit some dependency on PEB temperature (< 0.04µm/°C is typical)\" (p.8). A grade-specific PEB-sensitivity study for this exact 3.5µm/nLOF 2035 film (\"EXAMPLE PEB SENSITIVITY (3.5µm Film Thickness on Si)\", p.5) tested 105/110/115°C at 60s each, reporting Top/Bottom CD of 1.734/0.726µm, 1.992/1.439µm, and 2.062/1.687µm respectively. Note an internal inconsistency in that figure: the headline states \"Slope = 0.038 µm/°C\" but the printed linear-fit equation on the same chart is \"y = 0.0328x - 1.6787\" (slope 0.0328, not 0.038) — flagged for QC, not resolved here. Series-wide PEB range: \"105° to 115°C\" (PROCESS CONSIDERATIONS, p.8; note this differs slightly from the earlier printed \"100° to 115°C\" language in the same section — both ranges appear near-verbatim in the document).", "source": "EXAMPLE PROCESS (3.5µm Film Thickness on Si), p.4; EXAMPLE PEB SENSITIVITY (3.5µm Film Thickness on Si), p.5; POST EXPOSE BAKE, p.8 of AZ nLOF 2000 Series datasheet." }, "floodExposure": null, "develop": { "developer": "AZ 300MIF", "dilution": "used as supplied (ready-to-use MIF developer; industry-standard 0.26N/2.38% TMAH per DEVELOPING section)", "time_s": 120, "method": "puddle", "rinse": null, "source": "EXAMPLE PROCESS (3.5µm Film Thickness on Si), p.4; DEVELOPING, p.8 of AZ nLOF 2000 Series datasheet.", "_note": "120s single puddle is specific to this 3.5µm nLOF 2035 example — longer than the 60s single puddle used for the thinner 2.0µm nLOF 2020 example (p.3), consistent with needing more time to clear a thicker film. The generic Typical Process line (p.1) only says \"Develop: Puddle, spray or immersion / Developer Type: MIF\" with no time or specific product, so the grade-specific table is preferred here." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "No specific temperature or time is given for nLOF 2035; the datasheet states only that \"AZ nLOF materials are extremely thermally stable and may be hard baked at temperatures above 150°C\" (HARD BAKE, p.8) and, at the series level, that \"printed features are thermally stable to >200°C\" (APPLICATION, p.1). A dedicated hard-bake stability test (115/120/125/130°C) is published, but only for AZ nLOF 2070 (7.0µm film) — not this grade — so it is not used here.", "source": "HARD BAKE, p.8; APPLICATION, p.1 of AZ nLOF 2000 Series datasheet." }, "descum": null, "applications": [ "lift-off", "etch-mask", "high-aspect-ratio" ], "etchResistance": "\"May be processed with vertical sidewalls for RIE etching\" (APPLICATION, p.1) — a qualitative capability claim; no etch rate or selectivity data is published.", "liftoffSuitable": true, "platingSuitable": true, "stripper": "AZ 400T or AZ Remover 770 (STRIPPING, p.9): \"AZ nLOF 2000 Series resists are compatible with industry standard solvent based removers.\"", "storage": null, "notes": "AZ nLOF 2035 is the middle grade in Merck's AZ nLOF 2000 series, a chemically-amplified i-line negative resist purpose-built to give an undercut lift-off sidewall from a single standard expose/PEB/develop flow, without the extra image-reversal bake/flood-expose steps older negative processes required. The PEB is not optional here — the datasheet states it is \"required for proper imaging\" and the resist's CD is directly sensitive to PEB temperature (documented at ~0.03-0.04 µm/°C for this exact 3.5µm film thickness); getting PEB temperature and time right matters more for this chemistry than for a standard DNQ positive resist. No rehydration hold is needed (explicitly \"None\" per the Typical Process table), but delays between soft bake and exposure should be minimized. It is not recommended for copper substrates. One internal inconsistency in the source document is worth flagging: a Linearity/Exposure-Latitude chart on p.6 is captioned \"AZ nLOF 2020 @ FT=3.5µm\" even though 3.5µm and a 120s develop otherwise match the nLOF 2035 example elsewhere in the same document — this recipe does not use that chart's data, since the document itself names it as the 2020 grade.", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_nlof2000_series.pdf", "datasheetVersionOrDate": "Rev. (03/21)", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "The archived PDF is Merck's own \"Technical datasheet AZ nLOF 2000 Series\" (Merck branding and copyright throughout: \"© 2021 Merck KGaA, Darmstadt, Germany and/or its affiliates\"); the assigned datasheetUrl points to a MicroChemicals-hosted mirror of this same Merck document, not a MicroChemicals-authored sheet. Grade identity for nLOF 2035 is well supported: a dedicated EXAMPLE PROCESS table (p.4) and PEB-sensitivity study (p.5) explicitly name \"AZ nLOF 2035\" and coat 3.5µm, and the spin-curve legend (p.1) separately labels an \"nLOF 2035\" trace consistent with that thickness. See exposureDose._note and peb.notes for a labeling inconsistency elsewhere in the document (p.6) that was deliberately excluded.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "az-nlof-2070", "name": "AZ nLOF 2070", "manufacturer": "Merck", "productLine": "AZ nLOF 2000 Series", "aliases": [ "AZ nLOF™ 2070", "nLOF 2070", "AZ nLOF 2000 Series" ], "tone": "negative", "chemistry": "car", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Datasheet does not address grayscale or 3D patterning for AZ nLOF 2070; no such use is claimed anywhere in the document.", "status": "active", "successorSlug": null, "summary": "AZ nLOF 2070 is the ~7 µm grade of AZ's chemically amplified negative-tone nLOF 2000 series, engineered for single-layer lift-off with a required post-exposure bake, no rehydration wait, and thermal stability above 150°C for its lift-off undercut sidewall.", "thicknessRange": { "min_um": 5.3, "max_um": 11.8, "basis": "curve-span", "source": "curve-span — same reasoning as az-nlof-2020: the p.1 series-level bullet ('Single coat thicknesses from 2.0 to >10µm') covers the whole AZ nLOF 2000 Series (2020/2035/2070 together), not the 2070 grade alone. The grade-specific range used here is the span of the correctly-identified nLOF 2070 curve (top, red) on the p.1 chart, 500-4000 rpm. The 7.0 µm nominal reference-process thickness (p.5) sits inside this span." }, "spinCurves": [ { "label": "AZ nLOF 2070", "points": [ { "rpm": 500, "um": 11.8 }, { "rpm": 1000, "um": 11 }, { "rpm": 1500, "um": 9.2 }, { "rpm": 2000, "um": 7.8 }, { "rpm": 2500, "um": 7 }, { "rpm": 3000, "um": 6.3 }, { "rpm": 3500, "um": 5.8 }, { "rpm": 4000, "um": 5.3 } ], "source": "read from figure, \"SPIN CURVES (150mm Silicon)\", p.1 of AZ nLOF 2000 Series Technical datasheet (Rev. 03/21) — chart plots three grades (nLOF 2070 red, nLOF 2035 yellow, nLOF 2020 blue) with an explicit color-coded legend; nLOF 2070 identified by its red 'nLOF 2070' legend entry (top curve, highest thickness at every speed) and cross-checked against the grade-specific 7.0 µm reference process (p.5, which names 'AZ nLOF 2070 (330cPs)' explicitly).", "figureRead": true } ], "spinNotes": "Unlike the ambiguous combined charts flagged elsewhere in this recipe set, this chart carries an explicit color-coded legend naming each of the three plotted grades individually (nLOF 2070, nLOF 2035, nLOF 2020), and the identification is cross-checked against this datasheet's own grade-specific reference-process tables (p.3, p.5), which name the exact grade and viscosity (cPs) used for each nominal film thickness. Still a figure read, not a numeric table, so visual QC is required. No dispense volume, spin ramp, or edge-bead detail is published in this datasheet.", "adhesion": { "hmds": true, "notes": "Oxide-forming substrates (e.g. Si) should be HMDS primed prior to coating (PROCESS CONSIDERATIONS > SUBSTRATE PREPARATION, p.8). The 7.0 µm reference process specifies a vapor prime of HMDS 140°C/60s (p.5)." }, "rehydration": "None required — p.1 TYPICAL PROCESS states 'Rehydration Hold: None', consistent with this being a chemically amplified resist, not a DNQ resist. The 7.0 µm reference process (p.5) likewise lists 'Post Bake Delay: None'. (Source: TYPICAL PROCESS, p.1, and EXAMPLE PROCESS (7.0µm Film Thickness on Si), p.5, of AZ nLOF 2000 Series Technical datasheet (Rev. 03/21))", "softbake": { "temp_c": 110, "time_s": 90, "method": "hotplate", "notes": "Reference process for the 7.0 µm nLOF 2070 film (p.5): 110°C, 90 s, direct contact hotplate — longer than the 60 s used for the thinner 2.0 µm (nLOF 2020) and 3.5 µm (nLOF 2035) grades, consistent with a thicker coat needing more bake time to drive off solvent. Series-level guidance (p.8) puts soft bake temperature generally in the 100-110°C range.", "source": "EXAMPLE PROCESS (7.0µm Film Thickness on Si), p.5 of AZ nLOF 2000 Series Technical datasheet (Rev. 03/21)" }, "exposureDose": { "doses": [], "datasheetBasis": "\"Expose: i-line @ various doses (0.54NA) Nikon Stepper\" — verbatim, EXAMPLE PROCESS (7.0µm Film Thickness on Si), p.5. Confirmed series-wide: \"AZ nLOF 2000 requires exposure energy at the 365nm wavelength\" (p.8).", "_note": "Unlike the 2.0 µm (nLOF 2020) and 3.5 µm (nLOF 2035) reference processes, the 7.0 µm nLOF 2070 reference process does not state one nominal dose — its process table literally says 'various doses' and the page instead shows a BOTTOM CD vs. EXPOSURE DOSE sweep at three points (mask CD = 7.0 µm dense lines, p.5): 174 mJ/cm² -> 4.45 µm bottom CD, 186 mJ/cm² -> 4.84 µm bottom CD, 198 mJ/cm² -> 5.31 µm bottom CD. None of the three is identified as 'the' nominal process dose, so no single dose figure is recorded as the process value; picking the midpoint or any one of the three would be an invented nominal the datasheet does not itself name." }, "peb": { "temp_c": 110, "time_s": 90, "notes": "PEB is REQUIRED for proper imaging (p.1, p.8) — the critical step for nLOF, a chemically amplified negative resist, unlike the optional or absent PEB on the DNQ positive resists elsewhere in this recipe set. Reference process for 7.0 µm nLOF 2070 (p.5): 110°C, 90 seconds, direct contact hotplate (scaled up from the 60 s used on the thinner grades, matching this grade's own softbake time). Series-level guidance (p.8) allows 100-115°C generally, and warns CD is sensitive to PEB temperature at roughly <0.04 µm/°C (the datasheet's own measured example — slope 0.038 µm/°C over 105-115°C — was run on the related 3.5 µm nLOF 2035 grade, p.5, not nLOF 2070 itself, so it is cited here as series-level supporting evidence only).", "source": "EXAMPLE PROCESS (7.0µm Film Thickness on Si), p.5, and PROCESS CONSIDERATIONS > POST EXPOSE BAKE, p.8, of AZ nLOF 2000 Series Technical datasheet (Rev. 03/21)" }, "floodExposure": null, "develop": { "developer": "AZ 300MIF", "dilution": "undiluted (ready-to-use 0.26N / 2.38% TMAH developer; no dilution ratio stated)", "time_s": 120, "method": "puddle", "rinse": null, "source": "EXAMPLE PROCESS (7.0µm Film Thickness on Si), p.5 of AZ nLOF 2000 Series Technical datasheet (Rev. 03/21): 'AZ 300MIF, 2 x 60 second puddles'", "_note": "120 s total across two 60-second puddle cycles, not one continuous immersion — more cycles than the single 60 s puddle used for the thinner 2.0 µm nLOF 2020 grade, consistent with a thicker film needing more develop time. PROCESS CONSIDERATIONS (p.8) recommends AZ 300MIF, an industry-standard 0.26N (2.38%) TMAH developer, for the whole nLOF 2000 series." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "General series claim: 'AZ nLOF materials are extremely thermally stable and may be hard baked at temperatures above 150°C' (p.8) — an open-ended floor, not a specific temperature/time, so left null rather than guessed. This datasheet's only quantified hard-bake demonstration is specific to this grade: large pads in a 7.0 µm nLOF 2070 film were hard baked at 115°C, 120°C, 125°C and 130°C (p.8, HARD BAKE STABILITY for Large Pads in AZ nLOF 2070) with SEM images showing the pattern edge profile across that range; no hold time is stated for the test and no single temperature is identified as the recommended process value, so this is reported as a demonstrated range, not a chosen recipe.", "source": "PROCESS CONSIDERATIONS > HARD BAKE, and HARD BAKE STABILITY for Large Pads in AZ nLOF™ 2070 (7.0µm Film Thickness), p.8 of AZ nLOF 2000 Series Technical datasheet (Rev. 03/21)" }, "descum": null, "applications": [ "lift-off", "etch-mask" ], "etchResistance": null, "liftoffSuitable": true, "platingSuitable": null, "stripper": "AZ 400T or AZ Remover 770 (solvent-based removers), recommended per PROCESS CONSIDERATIONS > STRIPPING, p.8 of AZ nLOF 2000 Series Technical datasheet.", "storage": null, "notes": "AZ nLOF 2070 is the ~7 µm grade of AZ's nLOF 2000 series, a chemically amplified negative photoresist purpose-built to replace older image-reversal and multi-layer lift-off processes with a single expose/PEB/develop flow and a clean undercut sidewall. As with the whole series, the post-exposure bake is required, not optional, unlike the DNQ positive resists elsewhere in this recipe set, and CD is sensitive to PEB temperature (roughly <0.04 µm/°C per the datasheet's series-level guidance). Unlike thick DNQ positive resists it needs no rehydration wait after softbake, and softbake/PEB both scale up to 90 s at this thickness versus 60 s for the thinner 2020/2035 grades. It develops in AZ 300MIF (0.26N/2.38% TMAH) via two 60-second puddle cycles rather than one. Printed features are thermally stable well above 150°C — the datasheet specifically demonstrates hard-baking large 7 µm nLOF 2070 pads at 115-130°C with stable edge profiles — which is what lets nLOF's negative-tone undercut survive a subsequent metal evaporation and lift-off step. The datasheet explicitly does not recommend nLOF 2000 series resists for use on copper substrates.", "developerFamily": "tmah", "references": [ { "type": "paper", "title": "Reduced Etch Lag and High Aspect Ratios by Deep Reactive Ion Etching (DRIE)", "authors": "Gerlt et al.", "journal": "Micromachines", "year": 2021, "doi": "10.3390/mi12050542", "url": "https://doi.org/10.3390/mi12050542", "accessedDate": "2026-07-12", "summary": "7.2 µm nLOF 2070 DRIE mask; documents its slightly angled sidewalls" }, { "type": "paper", "title": "Fabrication of metal air bridges for superconducting circuits using two-photon lithography", "authors": "Huang et al.", "journal": "Applied Physics Letters", "year": 2025, "doi": "10.1063/5.0271788", "url": "https://doi.org/10.1063/5.0271788", "accessedDate": "2026-07-12", "summary": "7 µm nLOF 2070 air-bridge lift-off for superconducting qubit circuits, direct-write patterned" } ], "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_nlof2000_series.pdf", "datasheetVersionOrDate": "Rev. (03/21)", "accessedDate": "2026-07-10", "secondarySources": [] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "az-p4620", "name": "AZ P4620", "manufacturer": "AZ Electronic Materials", "productLine": "AZ P4000 Series", "aliases": [ "P4620", "AZ P4620 Photoresist" ], "tone": "positive", "chemistry": "dnq-novolak", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Datasheet does not address grayscale/3D patterning for AZ P4620; no grayscale suitability is asserted or implied by the document.", "status": "active", "successorSlug": null, "summary": "AZ P4620 is a thick, single-coat DNQ-novolak positive photoresist from the AZ P4000 series, demonstrated in this datasheet for copper and gold electroplating molds at film thicknesses from roughly 12 to 28 µm.", "thicknessRange": { "min_um": 6.9, "max_um": 14.8, "basis": "curve-span", "source": "curve-span — the dedicated p.26 spin-speed curve (1000-4000 rpm, single series) is the only P4620-specific numeric thickness range in the document. The P4000-Series spec table (p.3: 2-55 µm, 25 µm max single coat) covers the whole product family (P4620/P4903/P4400/P4330-RS/P4210/P4110, see the multi-grade chart on p.27) rather than P4620 alone, and is directly contradicted by this same document's own P4620 gold-plating process (p.7-9), which reports 28 µm achieved in a single coat — 3 µm over the series table's stated 25 µm max single coat. Individual demonstrated processes elsewhere in the document span roughly 12-28 µm (p.4, p.7, p.10, p.18, p.22) at speeds/bakes outside the p.26 curve's 1000-4000 rpm window, so the curve-span reported here is narrower than the full achievable envelope — treat it as a documented floor, not the whole range." }, "spinCurves": [ { "label": "AZ P4620 as supplied", "points": [ { "rpm": 1000, "um": 14.8 }, { "rpm": 1500, "um": 12.1 }, { "rpm": 2000, "um": 10.1 }, { "rpm": 2500, "um": 8.9 }, { "rpm": 3000, "um": 8 }, { "rpm": 3500, "um": 7.4 }, { "rpm": 4000, "um": 6.9 } ], "source": "read from figure, \"AZ® P4620 Spin Speed Curve\", p.26 of AZ P4620 Photoresist Data Package — single-product chart with one plotted series (unambiguous), softbake 110°C/180 s hotplate on 150 mm Si", "figureRead": true } ], "spinNotes": "Curve conditions per p.26: 150 mm silicon substrate, softbake 110°C/180 s hotplate. A separate multi-grade comparison chart (p.27, a lower-resolution scanned image) plots P4620 alongside P4903, P4400, P4330-RS, P4210 and P4110; at matching speeds it reads noticeably higher (~16.5 µm at 1000 rpm vs. 14.8 µm on the dedicated p.26 chart, an ~11% disagreement between the two figures in the same document). The p.26 chart was used as the published curve because it is the higher-precision, single-product source with unambiguous curve identity; the p.27 chart is noted here only as an internal-consistency flag for QC. No dispense volume, spin ramp, or edge-bead detail is published anywhere in this document.", "adhesion": { "hmds": null, "notes": "No HMDS or other adhesion-promoter step is mentioned anywhere in this datasheet." }, "rehydration": "A 60-minute rehydration wait is specified after the two-stage softbake (100°C/500 s hotplate, then 90°C/180 min oven) in the 28 µm single-coat gold-plating process, before exposure. (Source: AZ® P4620 Gold Plating Process, p.7 of AZ P4620 Photoresist Data Package)", "softbake": { "temp_c": 110, "time_s": 180, "method": "hotplate", "notes": "This is the most consistently repeated single-recipe softbake in the document (15 µm copper-plating process, p.4-6, and matches the spin-curve conditions on p.26). Other film thicknesses in this same datasheet use different bakes: 12 µm at 110°C/80 s (p.10, p.13-17); a two-layer 24 µm coat at 110°C/80 s then 115°C/180 s (p.18-21); 17 µm at 120°C/240 s (p.22); and the 28 µm single-coat process at 100°C/500 s hotplate then 90°C/180 min oven, followed by a 60-minute rehydration wait (p.7-9, see rehydration field). Softbake is thickness- and tool-dependent; no single value applies to all P4620 processes.", "source": "p.4-6 of AZ P4620 Photoresist Data Package (AZ P4620 Copper plating process conditions)" }, "exposureDose": { "doses": [], "datasheetBasis": "ghi-line / gh-line exposure (broadband, includes g-, h- and i-line together) — Canon PLA-501F ghi-line contact aligner (p.4-9, p.22), Ultratech 1500 gh-line stepper (p.10-21). Not restricted to i-line 365 nm or h-line 405 nm alone.", "_note": "Dose is thickness- and tool-dependent, not a single canonical value: 630 mJ/cm² at 17 µm (Canon PLA-501F, p.22); 893-927 mJ/cm² at 12 µm (Ultratech 1500 stepper, dense lines / contact holes, p.11); 1574-1742 mJ/cm² at 24 µm (Ultratech 1500 stepper, dense lines / contact holes, p.19). Reporting a single value_mJcm2 would be false precision, so all numeric dose fields are left null; the individual figures are recorded here instead." }, "peb": null, "floodExposure": null, "develop": { "developer": "AZ 400K", "dilution": "1:4", "time_s": 300, "method": "immersion", "rinse": null, "source": "AZ P4620 Copper plating process conditions, p.4-6 of AZ P4620 Photoresist Data Package (AZ 400K 1:4, immersion 300 s, 23°C)", "_note": "AZ 400K is the datasheet's own bolded/most-compatible developer choice (p.3 table). Other processes in the same document use different developer/dilution/time/temperature combinations: AZ 400K 1:3, immersion 300 s at 21.5°C for the 28 µm gold-plating process (p.7); AZ 400K 1:4, continuous spray 260 s at 27°C for the 24 µm two-layer process (p.18-21); and AZ 300 MIF (TMAH, used undiluted), continuous spray 200 s at 23°C for the 12 µm stepper process (p.10, p.13-17). Develop conditions scale with thickness/tool; the values above are the most consistent single recipe (15 µm copper-plating process)." }, "hardbake": null, "descum": null, "applications": [ "electroplating-molding" ], "etchResistance": null, "liftoffSuitable": null, "platingSuitable": true, "stripper": null, "storage": null, "notes": "AZ P4620 is a DNQ-novolak (diazonaphthoquinone/novolak) positive resist from AZ's thick-film P4000 series, positioned in this datasheet for copper and gold electroplating molds rather than lift-off or etch masking. It develops in AZ 400K (a buffered, metal-ion-containing alkaline developer — the datasheet's own preferred choice) or, for finer-pitch stepper lithography, in AZ 300 MIF (TMAH, metal-ion-free). Softbake, exposure dose and develop time all scale with film thickness rather than following one fixed recipe: a 12 µm coat baked at 110°C/80 s takes roughly 900 mJ/cm² on a gh-line stepper, while a 24 µm two-layer coat (110°C/80 s then 115°C/180 s) needs roughly 1650-1950 mJ/cm². The 28 µm single-coat gold-plating process is the process worth flagging for a first-time user: it uses a two-stage softbake (100°C/500 s hotplate, then 90°C/180 min oven) followed by a 60-minute rehydration wait before exposure — a classic thick-DNQ step that is easy to skip and, if skipped, tends to cause scumming or adhesion failure at develop. As a non-chemically-amplified DNQ resist it needs no post-exposure bake.", "developerFamily": "buffered-alkaline", "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/datenblaetter/tds/merck/en/tds_az_p4620_photoresist.pdf", "datasheetVersionOrDate": null, "accessedDate": "2026-07-10", "secondarySources": [] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "hmds", "name": "HMDS (Hexamethyldisilazane)", "manufacturer": "MicroChemicals GmbH (application-note publisher; HMDS itself is a commodity chemical sold by many suppliers, not a single manufacturer's proprietary product)", "productLine": null, "aliases": [ "HMDS", "Hexamethyldisilazane", "Hexamethyl disilazane", "(NH)[Si(CH3)3]2" ], "tone": null, "chemistry": "ancillary", "photoimageable": false, "grayscaleSuitable": false, "grayscaleNote": "Not applicable — HMDS is a vapor-phase adhesion promoter, not a photoimageable resist. Grayscale/3D lithography is not addressed anywhere in this application note.", "status": "active", "successorSlug": null, "summary": "HMDS (hexamethyldisilazane) is a vapor-phase adhesion promoter used ahead of photoresist coating: it chemically converts a hydrophilic, oxidized substrate surface (native/thermal SiO2, quartz, glass, most base metals) into a hydrophobic, resist-wetting surface by bonding a monolayer of trimethylsilyl groups, releasing ammonia as a byproduct.", "thicknessRange": null, "spinCurves": [], "spinNotes": "HMDS must not be spin-coated as a liquid — the only correct application is vapor-phase, via a nitrogen-carrier 'bubbler' held at room temperature feeding a heated (75-120°C) substrate, where HMDS bonds chemically as a monolayer (\"Correct Use of HMDS\", Fig. 47-48, p.2). If liquid HMDS is spun on instead, the resulting layer is only physically (not chemically) bound; ammonia released from it during the resist softbake diffuses into and cross-links the resist near the substrate, suppressing development and degrading resolution and profile (\"Incorrect Application of HMDS\", Fig. 49, p.2-3) — this failure mode is explicitly documented, not inferred. For the same reason, the note advises against applying HMDS in the same spin coater used for resist coating (slow HMDS evaporation can contaminate later coating runs); if spin-application is unavoidable, it recommends a 100-120°C water-desorption bake beforehand, a 100-120°C thermal-activation bake afterward, and strict spatial separation from resist-coating equipment (p.3). Because there is no legitimate spin-speed-vs-thickness relationship for a vapor-deposited chemisorbed monolayer, spinCurves is empty by design, not because data is missing.", "adhesion": { "hmds": null, "notes": "Not applicable — this entry describes HMDS itself (the adhesion promoter), not a photoresist that would require HMDS pretreatment." }, "rehydration": null, "softbake": { "temp_c": null, "time_s": null, "method": null, "notes": "This field is repurposed to record the bake/process temperatures this note actually publishes for HMDS, since the schema has no dedicated vapor-prime field. Correct (vapor-phase) HMDS application: HMDS vapor is carried by dry nitrogen from a room-temperature bubbler to a substrate heated to 75-120°C, where it bonds chemically as a monolayer (p.2). No dwell time is stated for this step, and temperature is given only as a range (75-120°C), so both temp_c and time_s are left null here rather than fabricated as a single value; the range is recorded in this note instead. If HMDS can only be applied by spin-coating (the note's documented 'incorrect' fallback), it separately recommends a 100-120°C bake to desorb water before HMDS application and a further 100-120°C activation bake afterward (p.3) — again stated as ranges. A general, non-HMDS-specific substrate dehydration bake of 'approx. 120°C for a few minutes' (optionally above 140°C on oxidized surfaces, to maximize adhesion) is described as a preceding step independent of which adhesion promoter is used (\"Adsorbed Water\", p.1).", "source": "Correct Use of HMDS / Incorrect Application of HMDS, p.2-3; Adsorbed Water, p.1" }, "exposureDose": null, "peb": null, "floodExposure": null, "develop": null, "hardbake": null, "descum": null, "applications": [], "etchResistance": null, "liftoffSuitable": null, "platingSuitable": null, "stripper": null, "storage": null, "notes": "HMDS (hexamethyldisilazane) is a vapor-phase adhesion promoter, not a photoimageable resist: it converts a hydrophilic, OH-terminated oxide surface (native or thermal SiO2, quartz, glass, most base metals) into a hydrophobic, resist-wetting surface by chemically bonding a monolayer of non-polar trimethylsilyl groups, releasing ammonia as a byproduct (p.2). The only correct application is vapor-phase, via a nitrogen-carrier bubbler at room temperature feeding a substrate heated to 75-120°C; spin-coating liquid HMDS instead yields only a physically bound layer whose ammonia, released during the subsequent resist softbake, diffuses into and cross-links the resist near the substrate, suppressing development and degrading resolution — a failure mode this note documents explicitly as 'Incorrect Application' (p.2-3). On noble metals without native oxide (gold, platinum), HMDS and other organic adhesion promoters show little to no adhesion benefit because they cannot chemically bond to the surface; base metals such as Al, Cr and especially Ti already adhere well without a promoter (p.3-4). This document is a general MicroChemicals application-note chapter ('01 Chapter — Basics of Microstructuring: Substrate Preparation'), not a manufacturer technical datasheet for a specific HMDS product or grade — HMDS itself is a commodity chemical sold by many suppliers, and this note is the correct place to attribute process guidance about it rather than any single vendor's TDS.", "developerFamily": null, "provenance": { "datasheetUrl": "https://www.microchemicals.com/dokumente/application_notes/substrate_cleaning_adhesion_photoresist.pdf", "datasheetVersionOrDate": null, "accessedDate": "2026-07-10", "secondarySources": [] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "apol-lo-3200", "name": "KemLab APOL-LO 3200", "manufacturer": "KemLab", "productLine": "APOL-LO 3200 series", "aliases": [ "APOL-LO 3200", "APOL-LO3200", "KemLab APOL-LO 3200" ], "tone": "negative", "chemistry": null, "_chemistryNote": "The TDS describes APOL-LO 3200 only as a 'negative tone Advanced Photoresist with a Lift-Off profile' for i-line and broadband use, developed in 0.26N TMAH, with PEB 'necessary to crosslink the photoresist.' It never names the underlying resin/photoinitiator chemistry in writing (the 'Advanced Photoresist' branding is marketing language, not a chemistry class), so chemistry is left null rather than guessed.", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed in the datasheet. APOL-LO 3200 is presented as a negative lift-off resist for conventional binary patterning, not grayscale lithography.", "status": "active", "successorSlug": null, "summary": "A series of negative-tone, high-resolution lift-off photoresists (APOL-LO 3202, 3204, 3207) covering 2-10+ µm film thickness, developed in 0.26N TMAH for i-line and broadband exposure, with customizable lift-off undercut angle and photospeed.", "thicknessRange": { "min_um": 2, "max_um": 10, "basis": "stated", "source": "stated — TDS bullet (p.1): 'Film Thickness range of 2 – 10+ µm'; corroborated by the per-grade 'Film Thickness Range' table on p.2 (APOL-LO 3202: 2-4, APOL-LO 3204: 3-6, APOL-LO 3207: 5-10+ µm). max_um is recorded as 10 because '10+' is an open-ended stated bound, not a precise number." }, "spinCurves": [ { "label": "APOL-LO 3202", "points": [ { "rpm": 1000, "um": 3.8 }, { "rpm": 2000, "um": 2.7 }, { "rpm": 3000, "um": 2.3 }, { "rpm": 4000, "um": 1.9 } ], "source": "read from figure 'Spin Curve: APOL-LO 3200 Series', p.2 of APOL-LO 3200 TDS — identified by legend marker/color (blue diamonds = APOL-LO 3202) on the combined three-curve chart; bottom curve, consistent with the stated 2-4 µm range. The chart itself only plots 1000-4000 rpm (4 marked points per curve) — no 5000/6000 rpm points exist in the source to report.", "figureRead": true }, { "label": "APOL-LO 3204", "points": [ { "rpm": 1000, "um": 6.2 }, { "rpm": 2000, "um": 4.5 }, { "rpm": 3000, "um": 3.7 }, { "rpm": 4000, "um": 3.2 } ], "source": "read from figure 'Spin Curve: APOL-LO 3200 Series', p.2 of APOL-LO 3200 TDS — identified by legend marker/color (red squares = APOL-LO 3204) on the combined three-curve chart; middle curve, consistent with the stated 3-6 µm range. Chart plots only 1000-4000 rpm.", "figureRead": true }, { "label": "APOL-LO 3207", "points": [ { "rpm": 1000, "um": 10.2 }, { "rpm": 2000, "um": 7.2 }, { "rpm": 3000, "um": 5.7 }, { "rpm": 4000, "um": 5 } ], "source": "read from figure 'Spin Curve: APOL-LO 3200 Series', p.2 of APOL-LO 3200 TDS — identified by legend marker/color (purple triangles = APOL-LO 3207) on the combined three-curve chart; top curve, consistent with the stated 5-10+ µm range. Chart plots only 1000-4000 rpm.", "figureRead": true } ], "spinNotes": "Coat program includes a 5-10 second spread cycle; spin time at final speed is 45 seconds. Spin curves determined on 6-inch Si with static dispense of ~3 mL of photoresist (TDS p.2, Spin Coat) — note this is a smaller dispense volume than the ~4 mL used for KemLab's other series datasheets. No edge-bead-removal step is published.", "adhesion": { "hmds": true, "notes": "APOL-LO adheres to gold, glass, aluminum, chromium and copper without a stated primer requirement; for silicon specifically, HMDS primer 'can increase adhesion' (TDS p.2, Substrate)." }, "rehydration": null, "softbake": { "temp_c": 110, "time_s": 60, "method": "hotplate", "notes": "110°C for 60 seconds is the default softbake, stated both in the narrative Soft Bake section (p.2) and in three of four columns (2, 4, 6 µm) of the Lift-Off Process Guide table (p.2). The datasheet gives an explicit, clearly labeled exception: 'For films over 7 microns: Soft-bake on hotplate: 110°C for 90 seconds' — used for the 10 µm (APOL-LO 3207) process point in the same table.", "source": "Lift-Off Process Guide table and Soft Bake section, p.2 of APOL-LO 3200 TDS" }, "exposureDose": { "doses": [], "datasheetBasis": "Broadband (Lift-Off Process Guide table header: 'Expose (broadband) on Si'); the narrative Exposure & Optical Parameters section states the resist is 'Sensitive at i-Line and broadband exposures' generally, but the quantified doses are all captioned broadband-on-Si.", "_note": "Left null because the Lift-Off Process Guide table (p.2) gives four different thickness-specific broadband doses on Si — 140 mJ/cm2 (2 µm, APOL-LO 3202), 145 mJ/cm2 (4 µm, APOL-LO 3204), 150 mJ/cm2 (6 µm, APOL-LO 3204 or 3207), 200 mJ/cm2 (10 µm, APOL-LO 3207) — none labeled typical/series-wide, so no scalar is asserted. All four are recorded verbatim in `notes`. A resolution demo box on p.1 additionally notes an APOL-LO 3202 exposure was done on a 'Nikon i9c stepper' (no dose given) and an APOL-LO 3207 exposure was 'Broadband' (no dose given) — neither adds a usable number." }, "peb": { "temp_c": 110, "time_s": 60, "notes": "PEB is explicitly necessary (unlike most positive resists in this manufacturer's catalog): 'PEB is necessary to crosslink the photoresist. PEB can be changed to modify performance' (p.2). 110°C for 60 seconds is the default, stated in both the narrative Post-Exposure Bake section and three of four columns (2, 4, 6 µm) of the process table. Explicit exception: 'For films over 7 microns: PEB on contact hotplate: 110°C for 90 seconds' — used for the 10 µm (APOL-LO 3207) process point.", "source": "Lift-Off Process Guide table and Post-Exposure Bake section, p.2 of APOL-LO 3200 TDS" }, "floodExposure": null, "develop": { "developer": "0.26N TMAH", "dilution": null, "time_s": null, "method": null, "rinse": null, "source": "Lift-Off Process Guide table, p.2, and Develop section, p.3 of APOL-LO 3200 TDS", "_note": "Develop time is left null because the table gives four different thickness-specific times (40 s, 60 s, 75 s, 120 s for the 2/4/6/10 µm points respectively), none labeled typical. Method is left null because neither the table (just 'Develop (TMAH 0.26N)') nor the narrative Develop section ('APOL-LO Photoresists are optimized for use with 0.26N TMAH developers. They are also compatible with other industry developers.') names a specific immersion/puddle/spray method — unlike the sibling KL5300/KL6000/KL IR datasheets, this document simply doesn't state one." }, "hardbake": null, "descum": null, "applications": [ "lift-off", "general-prototyping" ], "etchResistance": null, "liftoffSuitable": true, "platingSuitable": null, "stripper": "NMP, DMSO or similar solvent-based removers at 50-80°C (TDS p.3, Photoresist Removal). No two-bath-process guidance is given for this product (unlike the KL5300/KL6000/KL IR datasheets, which explicitly recommend one for thick films).", "storage": "Store upright in tightly closed containers at 40-70°F (4-21°C), away from oxidizers, acids, bases and ignition sources (TDS p.3, Storage).", "notes": "APOL-LO 3200 is a series of three grades (APOL-LO 3202, 3204, 3207) explicitly engineered for a lift-off undercut profile, with customization offered to adjust the lift-off angle or photospeed. Unlike most conventional positive resists, PEB here is a required crosslinking step, not an optional one — the datasheet flags that PEB conditions 'can be changed to modify performance,' making it a tunable process lever rather than a fixed step. Softbake and PEB share one explicit thickness break point: films over 7 µm (i.e. the APOL-LO 3207 10 µm process point) get 90 s instead of 60 s at the same 110°C, while exposure dose and develop time scale continuously with thickness across all four demonstrated process points (2/4/6/10 µm). An n,k optical-dispersion curve (200-500 nm) is published for lithography simulation. As with any lift-off resist, the achieved undercut profile — not just critical dimension — should be verified on-tool, since it is the property this product is specifically sold on.", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://www.kemlab.com/_files/ugd/5b8579_a8dd77c4036e4f199a9a6c899311b7c8.pdf", "datasheetVersionOrDate": "not stated — no revision or copyright date is printed anywhere in this TDS; the only certification mark present is an ISO 9001:2015 seal on the final page.", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "First-party kemlab.com TDS PDF (hosted at kemlab.com/_files/ugd/), discovered via the APOL-LO 3200 product page (kemlab.com/apollo-3200-resist) and fetched/read directly.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": true }, { "slug": "hare-sq", "name": "KemLab HARE SQ", "manufacturer": "KemLab Inc.", "productLine": "SQ Series", "aliases": [ "HARE SQ", "SQ", "SQ Series", "KemLab SQ" ], "tone": "negative", "chemistry": "epoxy", "_chemistryNote": null, "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed in the datasheet. SQ is presented as a high-aspect-ratio negative epoxy for permanent structural/microfluidic patterning with vertical sidewalls, not grayscale lithography (SU-8-class resists are used for grayscale in industry generally, but this document makes no such claim).", "status": "active", "successorSlug": null, "descum": null, "summary": "A high-aspect-ratio SU-8-class epoxy negative photoresist covering 2-100 µm film thickness in a single coat, for polymeric MEMS, microfluidics and micromachining, with vertical sidewalls suitable for permanent structures.", "thicknessRange": { "min_um": 2, "max_um": 100, "basis": "stated", "source": "stated — TDS subtitle/p.1 header: 'Film Thickness 2-100 microns'; corroborated by the FEATURES box: 'Film Thickness: Up to 100 µm single coat.'" }, "spinCurves": [ { "label": "SQ 2", "points": [ { "rpm": 1000, "um": 3.8 }, { "rpm": 2000, "um": 2.1 }, { "rpm": 3000, "um": 1.6 }, { "rpm": 4000, "um": 1.4 } ], "source": "read from the left-hand spin-curve chart (unlabeled figure, square-marker solid line = SQ 2), p.2 of SQ Series TDS; legend explicitly lists 'SQ 2' and 'SQ 5' as the two curves on this chart. The 2000 rpm read (~2.1 µm) is consistent with the Processing Guidelines table's stated reference point 'SQ 2, 2 µm @ 2000 rpm.'", "figureRead": true }, { "label": "SQ 5", "points": [ { "rpm": 1000, "um": 11.3 }, { "rpm": 2000, "um": 5.7 }, { "rpm": 3000, "um": 4.3 }, { "rpm": 4000, "um": 3.5 } ], "source": "read from the left-hand spin-curve chart (triangle-marker dashed line = SQ 5), p.2 of SQ Series TDS; legend explicitly lists 'SQ 5'. The 2000 rpm read (~5.7 µm) is consistent with the table's reference point 'SQ 5, 5 µm @ 2000 rpm.'", "figureRead": true }, { "label": "SQ 10", "points": [ { "rpm": 1000, "um": 23 }, { "rpm": 2000, "um": 12 }, { "rpm": 3000, "um": 9 }, { "rpm": 4000, "um": 7 } ], "source": "read from the right-hand spin-curve chart (diamond-marker solid line = SQ 10), p.2 of SQ Series TDS; legend explicitly lists 'SQ 10', 'SQ 25', 'SQ 50'. The 2000 rpm read (~12 µm) is reasonably consistent with the table's reference point 'SQ 10, 10 µm @ 2000 rpm' (figure-read tolerance).", "figureRead": true }, { "label": "SQ 25", "points": [ { "rpm": 1000, "um": 48 }, { "rpm": 2000, "um": 23 }, { "rpm": 3000, "um": 17 }, { "rpm": 4000, "um": 13 } ], "source": "read from the right-hand spin-curve chart (circle-marker dashed line = SQ 25), p.2 of SQ Series TDS; legend explicitly lists 'SQ 25'. The 2000 rpm read (~23 µm) matches the table's reference point 'SQ 25, 25 µm @ 2000 rpm' closely.", "figureRead": true }, { "label": "SQ 50", "points": [ { "rpm": 1000, "um": 98 }, { "rpm": 2000, "um": 48 }, { "rpm": 3000, "um": 33 }, { "rpm": 4000, "um": 24 } ], "source": "read from the right-hand spin-curve chart (triangle-marker dashed line = SQ 50), p.2 of SQ Series TDS; legend explicitly lists 'SQ 50'. The 2000 rpm read (~48 µm) matches the table's reference point 'SQ 50, 50 µm @ 2000 rpm' closely.", "figureRead": true } ], "spinNotes": "Coat program uses a 5-10 second spread cycle; spin time at final speed is 30 seconds (TDS p.2, Coat). Both spin-curve charts plot 1000-4000 rpm only (tick marks extend to 5000 but no data point is plotted there), so no 5000/6000 rpm value is reported for any grade. Edge-bead removal (EBR) is explicitly called out as necessary: 'It is necessary to remove this thick edge bead to reduce hotplate contamination... By removing the edge bead, the photomask can be positioned closer to the wafer, improving aspect ratio and resolution' (p.3) — KemLab names its own 'KL EdgeClean EBR' solvent for this step. The Processing Guidelines table's last row is labeled 'SQ 50 | 100 µm' — this repeats the 'SQ 50' product name from the row above it (which is 50 µm), rather than a distinct 'SQ 100' designation; quoted verbatim as printed, since the right-hand spin-curve chart shows only three curves (SQ 10, SQ 25, SQ 50) with no separate 100 µm curve, consistent with this being a second process recipe for the same SQ 50 material coated thicker at lower spin speed, not a sixth SKU.", "adhesion": { "hmds": null, "notes": "Not addressed. Substrate Preparation (p.2) states only that 'SQ Series adheres to variety of substrates; including silicon, gold, aluminum, glass, and chromium. For maximum adhesion, substrates should be clean and dry prior to applying SQ epoxy photoresist' — no HMDS or other primer is mentioned, unlike KemLab's other (non-epoxy) resist datasheets which explicitly recommend HMDS. hmds is left null rather than false, since the document is silent rather than stating a primer is unnecessary." }, "rehydration": null, "softbake": { "temp_c": null, "time_s": null, "method": "hotplate", "notes": "Softbake is a two-step contact-hotplate bake (65°C step then 95°C step) 'to minimize film stress and adhesion issues' (p.3), with both step times scaling by thickness. Left null because the process is fundamentally two-step, not a single temp/time pair. Per-grade values from the Processing Guidelines table (p.2): SQ2 (2 µm) 65°C/1min + 95°C/1min; SQ5 (5 µm) 65°C/1min + 95°C/3min; SQ10 (10 µm) 65°C/2min + 95°C/5min; SQ25 (25 µm) 65°C/3min + 95°C/7min; SQ50 (50 µm) 65°C/5min + 95°C/15min; SQ50 (100 µm process point) 65°C/10min + 95°C/30min.", "source": "Processing Guidelines table, p.2, and Softbake section, p.3 of SQ Series TDS" }, "exposureDose": { "doses": [], "datasheetBasis": "Broadband on Si with a 360 nm cutoff filter (Processing Guidelines table column header, p.2). The narrative Exposure & Optical Parameters section (p.3) states SQ is 'designed for near UV (350-400nm) exposure wavelengths' and that 'Exposure dose will vary depending on the exposure tool set, film thickness, and process conditions' — explicitly warning against treating the table doses as universal. Sensitivity is separately listed as 'NUV, Broadband, i-line' in the FEATURES box, but no dose is attributed specifically to i-line/365 nm.", "_note": "Left null because the Processing Guidelines table gives thickness-specific doses: 200 mJ/cm2 for SQ2 (2 µm) and 180 mJ/cm2 for every other listed point (SQ5, SQ10, SQ25, SQ50 at both 50 µm and 100 µm). None is labeled typical/series-wide, and the narrative text explicitly cautions the values are 'nominal' and tool/process-dependent, so no scalar is asserted. Both values are recorded verbatim above." }, "peb": { "temp_c": null, "time_s": null, "notes": "PEB is a two-step contact-hotplate bake (65°C step then 95°C step), 'adjusted according to film thickness in order to ensure sufficient crosslinking' with 'a two-step PEB... recommended to reduce film stress which can lead to cracking and/or adhesion loss' (p.3). Left null for the same reason as softbake — fundamentally two-step. Per-grade values from the table (p.2): SQ2 65°C/1min + 95°C/2min; SQ5 65°C/1min + 95°C/2min; SQ10 65°C/1min + 95°C/2min; SQ25 65°C/1min + 95°C/3min; SQ50 (50 µm) 65°C/1min + 95°C/5min; SQ50 (100 µm process point) 65°C/2min + 95°C/10min.", "source": "Processing Guidelines table, p.2, and Post-Exposure Bake section, p.3 of SQ Series TDS" }, "floodExposure": null, "develop": { "developer": "KemLab SQ Developer (SU-8 PGMEA)", "dilution": null, "time_s": null, "method": "immersion", "rinse": "Isopropyl alcohol (IPA) rinse and dry (TDS p.3, Develop).", "source": "Processing Guidelines table (column header 'Develop Immersion'), p.2, and Develop section, p.3 of SQ Series TDS", "_note": "Develop time is left null because the table gives six thickness-specific times: 1 min (SQ2), 1 min (SQ5), 2.5 min (SQ10), 3.5 min (SQ25), 6 min (SQ50 @ 50 µm), 15 min (SQ50 @ 100 µm process point) — none labeled typical. Method is set to 'immersion' because that is the table's explicit column header for this product; the narrative Develop section separately notes 'It can be developed using immersion, puddle or spray puddle. Thicker films benefit from refreshing developer during the develop step; such as with a double puddle' — recorded here as an alternate, not used to override the table's stated method." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Hardbake is optional, for permanent applications: 'Bake at > 120°C for at least 5 minutes (hot plate). A short hardbake can fuse cracks caused by film stress. For permanent structures, temperatures above 150°C are recommended. Oven bake will increase crosslinking with minimal increase in stress' (p.3). Left null because both figures are open-ended minimum thresholds ('>120°C', 'at least 5 minutes', 'above 150°C'), not single recommended set-points — quoted verbatim rather than converted to a false-precision scalar.", "source": "Hardbake section, p.3 of SQ Series TDS" }, "applications": [ "mems-structural", "microfluidics", "high-aspect-ratio" ], "etchResistance": null, "liftoffSuitable": false, "platingSuitable": null, "stripper": null, "storage": "Avoid light; store in an upright airtight container at 4-21°C or room temperature. If refrigerated, bring up to room temperature before opening. Keep away from oxidizers, acids, bases and ignition sources (TDS p.4, Storage).", "notes": "This datasheet's own title and running header are 'SQ Series' throughout — it never uses the name 'HARE SQ' anywhere in the body text. The KemLab product webpage this PDF was found on (kemlab.com/haresq) is titled 'HARE SQ Negative Epoxy,' and a second webpage (kemlab.com/sq-su8-epoxy-photoresist) titled simply 'SQ' links to this identical PDF. 'HARE' most plausibly stands for 'High Aspect Ratio Epoxy,' matching the TDS's own subtitle 'High Aspect Ratio SU-8 Epoxy Photoresist' — treated here as the same product under two marketing names, not as two distinct products, since no second, differently-numbered datasheet was found. The chemistry is explicitly stated as 'SU-8 polymer epoxy' (FEATURES box, p.1) — an SU-8-class thick-film epoxy negative resist, not merely SU-8-like. Both softbake and PEB are recommended as deliberate two-step (65°C then 95°C) bakes specifically to manage film stress and cracking in thick films, and the process explicitly requires edge-bead removal to protect hotplates and improve achievable aspect ratio (see spinNotes). Notably, unlike every other KemLab TDS read for this library, this document publishes no resist-removal/stripper section at all — consistent with the general difficulty of stripping thick, heavily-crosslinked SU-8-class epoxy negatives (especially once hardbaked above 150°C for permanent use), so no stripper value is asserted here; process/tooling should plan around that difficulty rather than assume a solvent strip will work as with the manufacturer's other product lines. This is a permanent structural resist, not a lift-off resist (liftoffSuitable is set false; KemLab markets its separate APOL-LO 3200 series for lift-off profiles).", "developerFamily": "solvent", "provenance": { "datasheetUrl": "https://www.kemlab.com/_files/ugd/5b8579_fe35828f146544b997e518176bdf6c58.pdf", "datasheetVersionOrDate": "Copyright 2025 © KemLab Inc., Rev 3-2025", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "First-party kemlab.com TDS PDF (hosted at kemlab.com/_files/ugd/), discovered independently via both the 'HARE SQ' product page (kemlab.com/haresq) and the 'SQ' product page (kemlab.com/sq-su8-epoxy-photoresist) — both link to this identical PDF, whose own title is 'SQ Series Technical Data Sheet.' Fetched and read directly; no distributor fallback was needed since the first-party document was found. (The verified elexansci mirror mentioned in the assignment brief was not used.)", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "kl-5300", "name": "KemLab KL 5300 series", "manufacturer": "KemLab Inc.", "productLine": "KL 5300 series", "aliases": [ "KL5300", "KemLab KL-5300" ], "tone": "positive", "chemistry": "dnq-novolak", "_chemistryNote": "The TDS describes KL5300 only as a 'positive photoresist' for i-line/g-line/broadband use, developed in 0.26N TMAH, and lists it as competing with Shipley S1805/S1808/S1811/S1813/S1818. It never names the photoactive-compound chemistry (e.g. DNQ-novolak) in writing, so chemistry is left null rather than inferred from the competitor cross-reference or product category.", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed in the datasheet. KL5300 is presented as a standard i-line/g-line/broadband positive resist for conventional binary patterning (0.55 µm resolution claimed), not grayscale lithography.", "status": "active", "successorSlug": null, "summary": "A series of thin-film positive photoresists (KL5315, KL5310, KL5305, KL5302-HiRes) spanning 0.15-2.5 µm single-coat thickness, developed in 0.26N TMAH for i-line, g-line and broadband exposure.", "thicknessRange": { "min_um": 0.15, "max_um": 2.5, "basis": "stated", "source": "stated — the TDS's 'Film thickness Range (microns)' table (p.1/p.2) gives four grade-specific stated ranges (KL5302-HiRes 0.15-0.30, KL5305 0.4-1.0, KL5310 0.7-1.5, KL5315 1.2-2.5 µm); min/max here is the span across all four stated grade ranges, not a curve span." }, "spinCurves": [ { "label": "KL5315", "points": [ { "rpm": 1000, "um": 2.35 }, { "rpm": 2000, "um": 1.68 }, { "rpm": 3000, "um": 1.4 }, { "rpm": 4000, "um": 1.2 }, { "rpm": 5000, "um": 1.05 } ], "source": "read from figure 3, p.2 of KL 5300 series TDS — identified by legend color (blue = KL5315) on the combined 'KL5300 Spin Curve' chart; the topmost curve, consistent with KL5315 having the largest stated film-thickness range (1.2-2.5 µm).", "figureRead": true }, { "label": "KL5310", "points": [ { "rpm": 1000, "um": 1.42 }, { "rpm": 2000, "um": 0.98 }, { "rpm": 3000, "um": 0.8 }, { "rpm": 4000, "um": 0.7 }, { "rpm": 5000, "um": 0.63 } ], "source": "read from figure 3, p.2 of KL 5300 series TDS — identified by legend color (red = KL5310) on the combined 'KL5300 Spin Curve' chart; second curve from top, consistent with the stated 0.7-1.5 µm range.", "figureRead": true }, { "label": "KL5305", "points": [ { "rpm": 1000, "um": 0.92 }, { "rpm": 2000, "um": 0.62 }, { "rpm": 3000, "um": 0.53 }, { "rpm": 4000, "um": 0.47 }, { "rpm": 5000, "um": 0.43 } ], "source": "read from figure 3, p.2 of KL 5300 series TDS — identified by legend color (green = KL5305) on the combined 'KL5300 Spin Curve' chart; third curve from top, consistent with the stated 0.4-1.0 µm range.", "figureRead": true }, { "label": "KL5302-HiRes", "points": [ { "rpm": 1000, "um": 0.27 }, { "rpm": 2000, "um": 0.19 }, { "rpm": 3000, "um": 0.16 }, { "rpm": 4000, "um": 0.14 }, { "rpm": 5000, "um": 0.125 } ], "source": "read from figure 4, p.2 of KL 5300 series TDS — a dedicated single-curve 'KL5302 HI-RES Spin Curve' chart (purple, unambiguous, no other grade plotted), consistent with the stated 0.15-0.3 µm range.", "figureRead": true } ], "spinNotes": "Spin curves determined on 6-inch Si with static dispense of ~4 mL of KL5300 photoresist (TDS p.2). No spin ramp/accel profile or edge-bead-removal step is published for KL5300.", "adhesion": { "hmds": true, "notes": "HMDS primer recommended with oxide-forming substrates (Si, etc.) for maximum adhesion; KL5300 also adheres to copper, gold, glass, aluminum and chromium (TDS p.1, Substrate Preparation)." }, "rehydration": null, "softbake": { "temp_c": 105, "time_s": 60, "method": "hotplate", "notes": "Processing-guidelines table (p.1) gives a single softbake condition, 105°C for 60 seconds, applied uniformly across all four grades. Separately, the narrative Softbake section (p.3) gives a broader recommended range: 'contact hotplate temperature is 90-105°C. Typical bake time is 60 seconds' — 105°C is the top of that range and matches the table value, used here as the single recipe value.", "source": "Processing Guidelines table, p.1 of KL 5300 series TDS; corroborated by Softbake section, p.3." }, "exposureDose": { "doses": [], "datasheetBasis": "Broadband, i-line, g-line (exposure figure caption on p.1 additionally cites 'Exposure: 60mJ @ 365 nm, Nikon I9 Stepper, NA=0.54' for the resolution figure only).", "_note": "Left null because the Processing Guidelines table gives four DIFFERENT grade-specific doses (KL5315 40 mJ/cm2, KL5310 35 mJ/cm2, KL5305 30 mJ/cm2, KL5302-HiRes 60 mJ/cm2), all under the umbrella 'Exposure (Broadband, i-line, g-line)' — none is a single series-wide value, and none is labeled typical/representative, so no scalar is asserted here. All four are recorded verbatim in `notes`. Separately, the p.1 resolution-demo caption states 60 mJ @ 365 nm specifically, but that is tied to the KL5302-HiRes resolution figure, not a series dose, so it is not promoted to at365_mJcm2 either." }, "peb": { "temp_c": 115, "time_s": 60, "notes": "Processing-guidelines table gives a single PEB condition (115°C/60s) applied uniformly across all four grades; confirmed verbatim in the Post-Exposure Bake section (p.4): 'Bake on contact hotplate at 115ºC for 60 seconds.'", "source": "Processing Guidelines table, p.1, and Post-Exposure Bake section, p.4 of KL 5300 series TDS" }, "floodExposure": null, "develop": { "developer": "0.26N TMAH", "dilution": null, "time_s": null, "method": null, "rinse": null, "source": "Processing Guidelines table, p.1, and Development section, p.4 of KL 5300 series TDS", "_note": "Develop time is left null because the table gives four grade-specific times (KL5315 30 s, KL5310 25 s, KL5305 20 s, KL5302-HiRes 30 s), not a single series value. Method is left null because the Development section (p.4) names three interchangeable options without singling one out: 'It can be developed with immersion, puddle or spray puddle' — 'spray puddle' does not map cleanly to a single enum key either." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "The optional Hard Bake row in the Processing Guidelines table (p.1) lists three conditions — 105°C for 60 seconds, 115°C for 60 seconds, 110°C for 60 seconds — under four grade columns (KL5315, KL5310, KL5305, KL5302-HiRes). The column-to-value alignment is not legible from the extracted table (three values, four columns), so no scalar is asserted for any grade; all three values are recorded here verbatim rather than guessed into a mapping.", "source": "Processing Guidelines table, p.1 of KL 5300 series TDS" }, "descum": null, "applications": [ "general-prototyping", "etch-mask" ], "etchResistance": null, "liftoffSuitable": null, "platingSuitable": null, "stripper": "NMP, DMSO or similar solvent-based removers at 50-80°C (TDS p.1/p.4, Resist Removal); thicker films may benefit from a two-bath process — first bath removes bulk resist, second bath cleans thoroughly.", "storage": "Avoid light; store in an upright airtight container at 4-21°C, away from oxidizers, acids, bases and ignition sources (TDS p.4, Storage).", "notes": "KL5300 is a series of four grades (KL5315, KL5310, KL5305, KL5302-HiRes) sharing one process family but with distinct spin curves, doses and develop times scaled to their target thickness band, from ~0.15 µm (KL5302-HiRes, high-resolution thin coat) up to 2.5 µm (KL5315). The datasheet's optical characterization (refractive index vs. wavelength, absorbance, i-line/g-line swing curves) is series-wide rather than grade-specific. As with any DNQ-class positive resist family, softbake and PEB times should be re-verified on-tool since the datasheet's single softbake/PEB condition is applied uniformly across grades of very different viscosity and coat thickness. HMDS priming is recommended on oxide-forming substrates; no adhesion promoter step beyond that is specified. Chemistry classified as dnq-novolak from the KemLab KL5300 MSDS composition table (mixed cresol novolak resin + diazo photoactive compound) (chemistry classified 2026-07-12 from manufacturer SDS/TDS).", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://www.kemlab.com/_files/ugd/5b8579_b2a2f2276aa74224a5ef58ed804848ea.pdf", "datasheetVersionOrDate": "Copyright 2022 © KemLab Inc., Rev 1-2022", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://www.nanofab.utah.edu/wp-content/uploads/2018/08/Positive-PhotoResist-%E2%80%93-KL5300-KemLab-27Feb15_v1.pdf", "what": "KemLab KL5300 MSDS states 'Mixed cresol novolak resin <25.0%, Diazo Photoactive Compound <4.0%'; the basis for classifying KL5300 as dnq-novolak." } ] }, "_provenanceNote": "First-party kemlab.com TDS PDF (hosted at kemlab.com/_files/ugd/), fetched and read directly.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "kl-6000", "name": "KemLab KL 6000 series", "manufacturer": "KemLab Inc.", "productLine": "KL 6000 series", "aliases": [ "KL6000", "KemLab KL-6000" ], "tone": "positive", "chemistry": null, "_chemistryNote": "The TDS describes KL6000 only as a 'positive photoresist' for i-Line/g-Line/broadband applications, developed in 0.26N TMAH, with Dill A/B/C parameters at 365 nm provided for optical modelling. It never names the underlying photoactive-compound chemistry (e.g. DNQ-novolak) in writing, so chemistry is left null rather than inferred from product category or Dill-parameter presence.", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed in the datasheet. KL6000 is presented as a thick-film positive resist for conventional binary patterning (1-3 µm line resolution demonstrated at various thicknesses), not grayscale lithography.", "status": "active", "successorSlug": null, "summary": "A series of thick-film positive photoresists (KL6008, KL6005, KL6003) covering 2.5-12+ µm in a single coat, developed in 0.26N TMAH for i-line, g-line and broadband exposure; PEB is not required for most applications.", "thicknessRange": { "min_um": 2.5, "max_um": 12, "basis": "stated", "source": "stated — the TDS bullet on p.1 states 'Cover 2.5 – 12 microns in a single coat'; corroborated by the p.2 'Film Thickness Range' table's per-grade stated ranges (KL6003 2.5-4.5, KL6005 4-7, KL6008 5-12+ µm), whose combined span matches." }, "spinCurves": [ { "label": "KL6008", "points": [ { "rpm": 1000, "um": 10 }, { "rpm": 2000, "um": 7 }, { "rpm": 3000, "um": 5.7 }, { "rpm": 4000, "um": 5 }, { "rpm": 5000, "um": 4.4 } ], "source": "read from figure 'KL6000 Spin Curve', p.2 of KL 6000 series TDS — identified by legend marker/color (blue diamonds = KL6008) on the combined three-curve chart; topmost curve, consistent with KL6008 having the largest stated film-thickness range (5-12+ µm).", "figureRead": true }, { "label": "KL6005", "points": [ { "rpm": 1000, "um": 7.3 }, { "rpm": 2000, "um": 5.2 }, { "rpm": 3000, "um": 4.2 }, { "rpm": 4000, "um": 3.6 }, { "rpm": 5000, "um": 3.2 } ], "source": "read from figure 'KL6000 Spin Curve', p.2 of KL 6000 series TDS — identified by legend marker/color (red squares = KL6005) on the combined three-curve chart; middle curve, consistent with the stated 4-7 µm range.", "figureRead": true }, { "label": "KL6003", "points": [ { "rpm": 1000, "um": 5.1 }, { "rpm": 2000, "um": 3.6 }, { "rpm": 3000, "um": 2.9 }, { "rpm": 4000, "um": 2.5 }, { "rpm": 5000, "um": 2.2 } ], "source": "read from figure 'KL6000 Spin Curve', p.2 of KL 6000 series TDS — identified by legend marker/color (purple triangles = KL6003) on the combined three-curve chart; bottom curve, consistent with the stated 2.5-4.5 µm range.", "figureRead": true } ], "spinNotes": "Coat program includes a 5-10 second spread cycle (longer for thicker films); spin time at final speed is 45 seconds. Spin curves determined on 6-inch Si with static dispense of ~4 mL of KL6000 photoresist (TDS p.2, Spin Coat). The TDS also publishes a fine-tuning formula for thickness under 10 µm: New Spin Speed = Spin Speed x (measured film thickness / desired film thickness)^2. No explicit edge-bead-removal step is published.", "adhesion": { "hmds": true, "notes": "HMDS (hexamethyldisilazane) primer is recommended and 'will increase adhesion of KL6000 to most substrates' (TDS p.2, Substrate); KL6000 adheres to gold, glass, aluminum, chromium and copper." }, "rehydration": null, "softbake": { "temp_c": 105, "time_s": 120, "method": "hotplate", "notes": "Narrative Soft Bake section (p.2): 'The recommended soft-bake by hotplate is 105°C +/- 5°C. Typical bake time is 120 seconds; longer bake times can help to drive the casting solvent out of thicker films.' 120 s is used here as the labeled-typical value. The p.1 Process Guide table instead gives four thickness-specific times at the same 105°C — 150 s (11 µm and 8 µm points on KL6008), 120 s (5 µm point on KL6005), 90 s (3 µm point on KL6003) — reflecting the 'longer bake for thicker films' guidance; these are recorded here rather than promoted to the scalar since they are per-thickness, not typical/series-wide.", "source": "Soft Bake section, p.2 of KL 6000 series TDS ('Typical bake time is 120 seconds'); per-thickness table values on p.1 Process Guide." }, "exposureDose": { "doses": [], "datasheetBasis": "Broadband on Si (Process Guide table header: 'Expose (broadband) on Si'); the narrative Exposure & Optical Parameters section states KL6000 is 'suitable for i-Line, broadband or g-Line exposure' generally, and separately publishes Dill A/B/C parameters specifically 'at 365 nm' for optical modelling, but the quantified doses in the process table are captioned broadband, not i-line.", "_note": "Left null because the Process Guide table (p.1) gives four different thickness-specific broadband doses on Si — 210 mJ/cm2 (11 µm), 180 mJ/cm2 (8 µm), 120 mJ/cm2 (5 µm), 90 mJ/cm2 (3 µm) — none labeled typical/series-wide, so no scalar is asserted. All four are recorded verbatim in `notes`. The Dill parameters (A=0.371 µm-1, B=0.075 µm-1, C=0.036 cm2/mJ, all 'at 365 nm', TDS p.3) are real published i-line optical-modelling constants but do not map to a dose in mJ/cm2, so they are not used to fill at365_mJcm2 and are recorded in `notes` instead." }, "peb": { "temp_c": 90, "time_s": 90, "notes": "PEB is not necessary for most applications (TDS p.1 table: '90°C for 90 sec if needed'; p.3 Post-Exposure Bake section: 'PEB is not necessary for most applications. If PEB is preferred for a particular process, bake on a contact hotplate at 90°C for 90 seconds.'). Unlike softbake/exposure/develop, this single condition is constant across all three grades — it is an optional step, not a required one.", "source": "Process Guide table, p.1, and Post-Exposure Bake section, p.3 of KL 6000 series TDS" }, "floodExposure": null, "develop": { "developer": "0.26N TMAH", "dilution": null, "time_s": null, "method": null, "rinse": null, "source": "Process Guide table, p.1, and Develop section, p.3 of KL 6000 series TDS", "_note": "Develop time is left null because the table gives four different thickness-specific times (90 s, 75 s, 45 s, 60 s). Method is left null because the process differs by grade — 'double spray puddle' for the KL6008/KL6005 example points, 'single spray puddle' for the KL6003 example point — and the narrative Develop section separately lists immersion, puddle or spray puddle as interchangeable general options; 'spray puddle' does not map cleanly to a single enum key. Full detail in `notes`." }, "hardbake": null, "descum": null, "applications": [ "etch-mask", "general-prototyping" ], "etchResistance": "Wet chemical etchants for Au, Cu, Cr, Al, etc. do not degrade patterns made with KL6000 (TDS p.3, Etch Resist).", "liftoffSuitable": null, "platingSuitable": null, "stripper": "NMP, DMSO or similar solvent-based removers at 50-80°C (TDS p.3, Photoresist Removal); thicker films may benefit from a two-bath process — first bath removes bulk resist, second bath cleans thoroughly.", "storage": "Store upright in tightly closed containers at 40-70°F (4-21°C), away from oxidizers, acids, bases and ignition sources (TDS p.3, Storage).", "notes": "KL6000 is a series of three grades (KL6008, KL6005, KL6003) sharing one thick-film process family, with the Process Guide table demonstrating four representative thickness/process points: 11 µm and 8 µm on KL6008, 5 µm on KL6005, and 3 µm on KL6003 (softbake 90-150 s at 105°C, broadband dose 90-210 mJ/cm2, develop 45-90 s by spray puddle, scaling with thickness). PEB is explicitly optional and not required for most applications — a notable process simplification versus thinner i-line resists. As a thick single-coat resist, edge-bead buildup and softbake solvent retention are the process risks most worth re-verifying on-tool for the thickest (KL6008) grade; the datasheet itself flags that longer softbake helps drive solvent out of thicker films. Dill optical parameters at 365 nm (A=0.371 µm-1, B=0.075 µm-1, C=0.036 cm2/mJ) are published for lithography simulation. Etch resistance to common wet metal etchants (Au, Cu, Cr, Al) is explicitly claimed by the manufacturer. KemLab family SDS data (e.g. the sibling KL5300 MSDS) suggests a DNQ-novolak chemistry consistent with KL6000's product category, but no directly-verifiable KL6000-specific source could be opened during this pass, so the classification is withheld (chemistry review 2026-07-12).", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://www.kemlab.com/_files/ugd/5b8579_d8f91ac120ea462bbe675ec4dcc00526.pdf", "datasheetVersionOrDate": "not stated — no revision or copyright date is printed anywhere in this TDS; the only certification mark present is an ISO 9001:2015 seal on the final page.", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "First-party kemlab.com TDS PDF (hosted at kemlab.com/_files/ugd/), fetched and read directly.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "kl-ir", "name": "KemLab KL IR 15", "manufacturer": "KemLab, Inc.", "productLine": "KL IR series", "aliases": [ "KL IR", "KL IR 15", "KemLab KL-IR" ], "tone": "image-reversal", "chemistry": "dnq-novolak", "_chemistryNote": "The TDS describes KL IR 15 only as an 'image reversal' resist usable as either positive or negative photoresist in i-line/g-line/broadband applications, developed in 0.26N TMAH. It never names the underlying photoactive-compound / crosslinking chemistry in writing, so chemistry is left null rather than inferred from the image-reversal product category. (chemistry classified 2026-07-12: re-checked the full archived 3-page PDF — no chemistry-mechanism wording anywhere; null confirmed, not a gap.)", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed in the datasheet. KL IR 15 is presented as a dual-tone (image-reversal) resist for conventional binary patterning, not grayscale lithography.", "status": "active", "successorSlug": null, "summary": "A dual-tone image-reversal photoresist, processed as either positive or negative in i-line, g-line and broadband applications, 1.2-2.6 µm film thickness. In negative mode the reversal-bake temperature is the most process-critical parameter.", "thicknessRange": { "min_um": 1.2, "max_um": 2.6, "basis": "stated", "source": "stated — the TDS spec box (p.1): 'Film Thickness: 1.2 – 2.6' (microns)." }, "spinCurves": [ { "label": "KL IR 15", "points": [ { "rpm": 1000, "um": 2.6 }, { "rpm": 2000, "um": 1.75 }, { "rpm": 3000, "um": 1.45 }, { "rpm": 4000, "um": 1.28 }, { "rpm": 5000, "um": 1.15 } ], "source": "read from figure captioned 'Spin Curve' (referenced in text as Figure 3), p.2 of KL IR 15 TDS. Single unambiguous curve — only one grade/SKU is plotted, no multi-curve legend to disambiguate. Axis range (1.00-2.80 µm, 0-6000 rpm) and curve shape are consistent with the stated 1.2-2.6 µm film-thickness range.", "figureRead": true } ], "spinNotes": "Spin curve determined on 6-inch Si with static dispense of ~4 mL of KL IR 15 resist (TDS p.2, Coat). No spin ramp/accel profile or edge-bead-removal step is published. A closely related sibling SKU, KL IR LO 15 (lift-off variant, separate TDS, Rev 10-2023), publishes a similarly-shaped spin curve on the same axis scale but is a distinct product literature and was not used to source any value here.", "adhesion": { "hmds": true, "notes": "HMDS primer is recommended with oxide-forming substrates (Si, etc.); KL IR adheres to silicon, copper, gold, glass, aluminum and chromium (TDS p.1, Substrate Preparation, in both the Negative and Positive Resist Mode tables)." }, "rehydration": null, "softbake": { "temp_c": 105, "time_s": 90, "method": "hotplate", "notes": "Single softbake condition (105°C, 90 sec) is common to both positive-mode and negative-mode processing guidelines (TDS p.1-2).", "source": "Negative Resist Mode Processing Guidelines table, p.1, and Positive Resist Mode Processing Guidelines table, p.2 of KL IR 15 TDS" }, "exposureDose": { "doses": [], "datasheetBasis": "Broadband, i-line, g-line (all quantified doses in this document are captioned 'Broadband'; the doc states KL IR 15 is used 'in i-line, g-line and broadband applications' but does not attribute a dose specifically to i-line or g-line).", "_note": "Left null because the two tone modes carry different, non-interchangeable exposure conditions: the Negative Resist Mode Processing Guidelines table (p.1) states a specific formal value, Broadband 120 mJ/cm2, for the image exposure preceding reversal bake. The Positive Resist Mode Processing Guidelines table (p.2), by contrast, states exposure only qualitatively as 'Broadband, i-line, g-line' with no number. A separate worked demo box on p.1 (Figure 2, positive tone process, 1.5 µm film) gives an approximate value, '~70 mJ/cm2 at 1.5 µm FT (broadband)' — explicitly marked with a tilde as approximate, not a formal guideline figure, and tied to one example film thickness. None of these can honestly be reduced to a single series-wide scalar, so all three are recorded here rather than merged." }, "peb": { "temp_c": 115, "time_s": 60, "notes": "PEB applies only in Positive Resist Mode (115°C, 60 sec); Negative Resist Mode has no PEB step — it instead has a reversal bake (see notes / floodExposure).", "source": "Positive Resist Mode Processing Guidelines table, p.2 of KL IR 15 TDS" }, "floodExposure": { "reversalBake": { "temp_c": 130, "time_s": 120, "criticalTolerance": "±1°C — the TDS states: 'The most critical factor of the image reversal process is the reversal bake temperature. This critical temperature must be kept within ± 1°C to maintain stable processes' (p.2, Reversal Bake section). This is the single most process-sensitive parameter in the whole recipe.", "source": "Negative Resist Mode Processing Guidelines table, p.1, and Reversal Bake section, p.2 of KL IR 15 TDS" }, "flood": { "value_mJcm2": 150, "notes": "Flood exposure is explicitly NOT critical: 'The flood exposure is not critical to the process. 150 mJ/cm2 (broadband) is the processing guideline. Exposures between 150 - 300 mJ/cm² will not have a major effect on performance' (p.2). This document is internally consistent (table value 150 mJ/cm2 matches the narrative guideline exactly). Note for QC: the sibling KL IR LO 15 (lift-off variant) TDS states a DIFFERENT table value (200 mJ/cm2) for the same step while its own narrative text repeats the identical '150 mJ/cm2... 150-300 mJ/cm² will not have a major effect' language — i.e. that sibling document is internally INCONSISTENT (200 vs 150). That number was not used here; this recipe uses only the self-consistent KL IR 15 (non-LO) document.", "source": "Negative Resist Mode Processing Guidelines table, p.1, and Flood Exposure section, p.2 of KL IR 15 TDS" } }, "develop": { "developer": "0.26N TMAH", "dilution": null, "time_s": null, "method": "puddle", "rinse": null, "source": "Negative Resist Mode Processing Guidelines table, p.1, and Positive Resist Mode Processing Guidelines table, p.2 of KL IR 15 TDS", "_note": "Develop time is left null: both tone-mode tables state '45 – 60 second puddle' as a RANGE, not labeled typical/recommended in the formal guidelines table, so no scalar is asserted. (Separately, the sibling KL IR LO 15 TDS demo boxes label specific times as 'recommended' — 45 s for a negative-mode 1.5 µm demo — but those are LO-variant demo figures, not this document's, and were not imported here.) Method is set to 'puddle' since both tables state that consistently and unambiguously (unlike the KL 5300/6000 TDS's ambiguous 'spray puddle' phrasing)." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Hardbake is optional and differs by tone mode: Positive Resist Mode table (p.2) gives 115°C, 60 sec; Negative Resist Mode table (p.1) gives 130°C, 60 sec. Left null because a single scalar cannot represent both modes honestly; both values are recorded here verbatim.", "source": "Negative Resist Mode Processing Guidelines table, p.1, and Positive Resist Mode Processing Guidelines table, p.2 of KL IR 15 TDS" }, "descum": null, "applications": [ "image-reversal", "general-prototyping" ], "etchResistance": null, "liftoffSuitable": null, "platingSuitable": null, "stripper": "KL Photoresist Remover, or industry-standard removers such as NMP or DMSO-based strippers, at 50-80°C (TDS p.1 spec box; p.3 Resist Removal); thicker films may benefit from a two-bath process — first bath removes bulk resist, second bath cleans thoroughly.", "storage": "Avoid light; store in an upright airtight container at 4-21°C, away from oxidizers, acids, bases and ignition sources (TDS p.3, Storage — the document's own wording says 'Keep developer away from oxidizers...', apparently a boilerplate carryover referring to the resist itself).", "notes": "KL IR 15 is a dual-tone image-reversal resist: exposed through a mask, it can be processed as either a positive resist (softbake -> exposure -> PEB at 115°C/60s -> develop -> optional 115°C/60s hardbake) or, via an added reversal bake and flood exposure, as a negative resist with the opposite tone (softbake -> exposure -> reversal bake at 130°C/120s -> flood exposure -> develop -> optional 130°C/60s hardbake). The reversal-bake temperature is explicitly called out by the manufacturer as the single most process-sensitive parameter (±1°C tolerance), while the flood exposure is explicitly non-critical over a 150-300 mJ/cm2 window — the opposite of the usual intuition that exposure dose is the sensitive knob. KemLab also sells a separate, distinct SKU, KL IR LO 15, whose own TDS (same Rev 10-2023 date) markets it specifically for lift-off-profile negative processing and states it 'can replace AZ 5214E'; the plain KL IR 15 datasheet used here does not make an equivalent lift-off-profile claim, so liftoffSuitable is left null rather than asserted either way. Negative mode is described as having 'excellent thermal stability' and being 'optimized for metallization processes.' Chemistry classified as dnq-novolak from the KemLab KL IR TDS's classic DNQ image-reversal process (reversal bake + flood exposure) and family SDS composition data (mixed cresol novolak resin + diazo photoactive compound) (chemistry classified 2026-07-12 from manufacturer SDS/TDS).", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://www.kemlab.com/_files/ugd/5b8579_400cc3c492104ef283eb7ae601468096.pdf", "datasheetVersionOrDate": "Copyright 2023 © KemLab, Inc., Rev 10-2023", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://www.kemlab.com/_files/ugd/5b8579_45ff748fc0894fa094a12c113747f6ef.pdf", "what": "KemLab KL IR TDS shows the classic DNQ image-reversal process (reversal bake + flood exposure); the family SDS lists mixed cresol novolak resin + diazo photoactive compound, the basis for classifying KL IR as dnq-novolak." } ] }, "_provenanceNote": "First-party kemlab.com TDS PDF (hosted at kemlab.com/_files/ugd/), discovered via the KL IR product page (kemlab.com/image-reversal-photoresist) and fetched/read directly. A second, distinct first-party PDF for the sibling 'KL IR LO 15' lift-off variant (https://www.kemlab.com/_files/ugd/5b8579_45ff748fc0894fa094a12c113747f6ef.pdf) was also fetched and read for comparison but is NOT the source of any value in this recipe — see notes and the floodExposure.flood.notes internal-consistency flag.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "kmpr-1050", "name": "KMPR 1050", "manufacturer": "Kayaku Microchem (MicroChem)", "productLine": "KMPR 1000 series", "aliases": [ "KMPR® 1050", "MicroChem KMPR 1050", "Kayaku Microchem KMPR 1050" ], "tone": "negative", "chemistry": "epoxy", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed by the datasheet. The document markets KMPR 1000 exclusively for binary, high-aspect-ratio, vertical-sidewall structures (MEMS, electroplating molds, DRIE masks); it makes no mention of grayscale or partial-exposure profiling.", "status": "active", "successorSlug": null, "summary": "KMPR 1050 is the highest-viscosity grade of Kayaku Microchem/MicroChem's KMPR 1000 series, a chemically amplified, TMAH-developable, epoxy-based negative photoresist built for thick, high-aspect-ratio MEMS, electroplating-mold, and DRIE-mask structures.", "thicknessRange": { "min_um": 34, "max_um": 115, "basis": "curve-span", "source": "curve-span: no single stated range is given for KMPR 1050 specifically (the intro text gives only a family-wide 4-120 µm range spanning all four KMPR 1000 viscosities). The range here is the min/max of the KMPR 1050 traces across BOTH published spin charts — Figure 1 (21°C US & EU: 44-101 µm, 4000-1000 rpm) and Figure 2 (23°C Japan & Asia: 34-115 µm, 4000-1000 rpm) — widened 2026-07-12 when the Figure 2 series was digitized into this file." }, "spinCurves": [ { "label": "KMPR 1050", "points": [ { "rpm": 1000, "um": 101 }, { "rpm": 2000, "um": 68 }, { "rpm": 3000, "um": 51 }, { "rpm": 4000, "um": 44 } ], "source": "read from figure 1 \"Spin speed vs. Thickness for KMPR 1000 resists (21°C US & EU)\", p.2 of the Kayaku Microchem/MicroChem \"KMPR 1000 Chemically Amplified Negative Photoresist\" datasheet (Ver. 4.2, UPenn nanoSOP mirror). Chart plots four curves (KMPR 1050/filled square, 1025/filled diamond, 1010/filled circle, 1005/open square), each with 4 markers at 1000/2000/3000/4000 rpm. The KMPR 1050 curve was identified as the filled-square series and the topmost (thickest-film) curve at every rpm, consistent with its legend position (listed first) and Table 1's viscosity ordering (1050 = 13,000 cSt, the highest of the four, so it should coat thickest at a given speed). Read at high resolution (4x page render, cropped) so the four data-point markers were individually resolvable; still an eyeball figure read, not a numeric table.", "figureRead": true }, { "label": "KMPR 1050 (23°C Japan & Asia)", "points": [ { "rpm": 1000, "um": 115 }, { "rpm": 2000, "um": 67 }, { "rpm": 3000, "um": 47 }, { "rpm": 4000, "um": 34 } ], "source": "read from figure 2 \"Spin speed vs. Thickness for KMPR® 1000 resists (23°C Japan & Asia)\", p.2 of the Kayaku Microchem/MicroChem \"KMPR 1000 Chemically Amplified Negative Photoresist\" datasheet (Ver. 4.2, UPenn nanoSOP mirror). Chart plots four curves (KMPR 1050/filled square, 1025/filled diamond, 1010/filled circle, 1005/open square), each with 4 markers at 1000/2000/3000/4000 rpm. The KMPR 1050 curve was identified as the filled-square series and the topmost (thickest-film) curve at every rpm, consistent with its legend position (listed first) and Table 1's viscosity ordering (1050 = 13,000 cSt, the highest of the four). Points were read by extracting the marker vector-path pixel rectangles and the axis tick-label pixel positions directly from the PDF page content stream (not a manual on-screen eyeball estimate), then converting via the resulting pixel-to-value linear scale (0-120 µm y-axis span; 1000 rpm x-axis grid spacing); cross-checked against the rendered chart image (kmpr-1050-p2-hires.png). digitized 2026-07-12", "figureRead": true } ], "spinNotes": "Recommended program: dispense 1 ml of resist per inch (25 mm) of substrate diameter; spin at 500 rpm for 5-10 s at 100 rpm/s acceleration, then spin at the target speed for 30 s at 300 rpm/s acceleration. The document publishes TWO spin curves for the same four resists at two different ambient conditions: Figure 1 (21°C, US & EU) and Figure 2 (23°C, Japan & Asia) — the curves are not identical (e.g. KMPR 1050 at 1000 rpm reads ~101 µm in Figure 1 vs. ~115 µm in Figure 2), a reminder that spin results are sensitive to coat-bowl ambient temperature/humidity and should be recharacterized on-tool. Both curves are now captured above: Figure 1 (21°C US & EU) as the primary curve, Figure 2 (23°C Japan & Asia) as a second series digitized by extracting marker vector-path pixel positions and axis tick-label pixel positions directly from the PDF page content stream. Source: \"Coat\" / \"Recommended Program\", p.1, and Figures 1-2, p.2.", "adhesion": { "hmds": false, "notes": "\"Adhesion promoters are typically not required.\" HMDS pretreatment (MCC Primer 80/20) is recommended \"for applications that require electroplating\" — and the document's own Plating process recipe lists HMDS as its first step. Source: \"Substrate Preparation\", p.1, and \"Plating\", p.3." }, "rehydration": null, "softbake": { "temp_c": 100, "time_s": null, "method": "hotplate", "notes": "Recommended bake temperature is 100°C (95-105°C also usable, per text) — this scalar temperature is explicit in the document. Time is published as a THICKNESS-BINNED table only up to 80 µm: 5-11 µm → 5 min; 12-20 µm → 7 min; 21-30 µm → 12 min; 31-55 µm → 15 min; 56-80 µm → 20 min. KMPR 1050's own spin curve (Figure 1) reaches 101 µm at 1000 rpm, i.e. above the table's published range — the datasheet does not give a soft-bake time for film thicker than 80 µm at all, so no time is assumed for the thick end of this grade's range. Convection ovens are explicitly not recommended. A cool-down/re-heat 'wrinkle' check is described to confirm the film is fully dry.", "source": "Table 2 \"Soft Bake Times\" and \"Soft Bake\" section text, p.2" }, "exposureDose": { "doses": [], "datasheetBasis": "\"KMPR® 1000 is most commonly exposed with conventional UV (350-400 nm) radiation, although i-line (365 nm) is recommended. It may also be exposed with e-beam or x-ray radiation.\" (Processing Guidelines, p.1); the front-page banner separately calls the product a \"KMPR®1000 i-Line photoresist.\" The dose table (Table 3) is not separately re-attributed to a specific wavelength beyond this general statement.", "_note": "Dose is published only as a THICKNESS-BINNED range, and only up to 80 µm: 5-11 µm → 235-335 mJ/cm²; 12-20 µm → 355-485 mJ/cm²; 21-30 µm → 500-645 mJ/cm²; 31-55 µm → 665-1055 mJ/cm²; 56-80 µm → 1070-1465 mJ/cm². KMPR 1050 reaches 101 µm (Figure 1, 1000 rpm), above this table's coverage — no dose is published for the thick end of this grade's range. Relative dose also scales with substrate (Table 4): silicon 1X, glass/Pyrex/ITO 1.5X, most metals (Au, Al, NiFe, Cu, Ni, Ti) and silicon nitride 1.5-2X. A >350 nm long-pass filter (recommended for vertical sidewalls) requires ~40% more exposure time for the same effective dose. Left null rather than picking a bin midpoint, per the range rule." }, "peb": { "temp_c": 100, "time_s": null, "notes": "Recommended PEB temperature is 100°C (95-105°C also usable, per text). Time is given as a THICKNESS-THRESHOLD rule rather than a table: ≤25 µm → 2 minutes; >25 µm → 3 minutes; >50 µm → 4 minutes. KMPR 1050's own thickness range (44-101 µm, Figure 1) straddles the 25 µm and 50 µm thresholds, so no single time applies across the whole grade — left null rather than picking one threshold's value. After 1 minute of PEB a latent mask image should already be visible; none appearing means insufficient exposure and/or heat.", "source": "\"Post Exposure Bake (PEB)\" section text, p.3" }, "floodExposure": null, "develop": { "developer": "2.38% TMAH (0.26N) aqueous alkaline developer (primary); SU-8 Developer (solvent-based) is also usable as an alternative", "dilution": "2.38% TMAH (0.26N), used at this standard concentration (not diluted from a stock in the document)", "time_s": null, "method": "immersion", "rinse": "DI water, ~20 s spray rinse, then filtered air/N2 dry (TMAH path). If the optional SU-8-developer alternative is used instead, rinse is ~10 s fresh developer spray then ~10 s IPA spray, per the SU-8-developer note.", "source": "\"Develop\" and \"Rinse and Dry\" sections plus Table 5 (TMAH) and Table 6 (SU-8 developer), p.3", "_note": "Development is designed for immersion, spray, or spray-puddle processing, primarily with 2.38% TMAH; SU-8 developer is offered as an alternative. Immersion times are thickness-binned (published only up to 80 µm, same coverage gap as soft bake/exposure): TMAH — 5-11 µm: 3 min; 12-20 µm: 5 min; 21-30 µm: 6 min; 31-55 µm: 6 min; 56-80 µm: 8 min. SU-8 developer — 5-11 µm: 2 min; 12-20 µm: 2 min; 21-30 µm: 2 min; 31-55 µm: 3 min; 56-80 µm: 4 min. No time is given above 80 µm, which is below KMPR 1050's own published thickness range (up to 101 µm)." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "The datasheet contains no general Hard Bake process step at all — no recommended temperature or time is published anywhere in the document (unlike the sibling SU-8 2000 datasheets, which have a dedicated Hard Bake section). It explicitly states, for the electroplating process specifically: \"Hard bake is NOT REQUIRED OR RECOMMENDED for plating resistance.\"", "source": "\"Plating\" section note, p.3" }, "descum": null, "applications": [ "high-aspect-ratio", "mems-structural", "electroplating-molding", "etch-mask" ], "etchResistance": "Listed as a Feature: \"Excellent dry etch resistance.\" Demonstrated in an application photo (\"Etched Trenches\", 10 µm features, 65 µm deep, credited to ULVAC) consistent with use as a DRIE (deep reactive ion etch) mask. No quantitative etch rate or selectivity numbers are published. Source: \"Features\" list and application photos, p.1.", "liftoffSuitable": false, "platingSuitable": true, "stripper": "MicroChem Remover PG (NMP): heat bath to 80°C, immerse substrate 10-20 minutes (actual time depends on resist thickness and agitation, e.g. ultrasound). For fully electroformed metal structures, a stronger sequence is given: Remover PG 10 min @80°C → DIW rinse → XP Remover K (epoxy stripping chemistry) 10 min @80°C → DIW rinse → XP Neutralizer K 3 min @25°C. Plasma removal: RIE 200 W, 80 sccm O2, 8 sccm CF4, 100 mTorr, 10°C. Source: \"Removal\" / \"Process Recommendation\" / \"Plasma Removal\", p.3.", "storage": "Store frozen, in tightly closed, upright containers, at 14°F (-10°C), away from light, heat, acids, and ignition sources. Shelf life is twelve months at 14°F (-10°C), but typically only one to two months at room temperature. Defrost at room temperature for 24 hours before use. Source: \"Storage\", p.4.", "notes": "KMPR 1050 is the thickest-coating grade in Kayaku Microchem/MicroChem's KMPR 1000 line and, unlike the solvent-developed SU-8 2000 family, is a chemically amplified epoxy resist designed to develop in aqueous TMAH — a meaningful process difference worth flagging for anyone assuming all thick epoxy negative resists behave like SU-8. Its published process tables (soft bake, exposure dose, TMAH develop time) only cover film thicknesses up to 80 µm, while KMPR 1050's own spin curve reaches roughly 101 µm at 1000 rpm; process engineers working at the thick end of this grade's range will need to extrapolate or characterize on-tool, since the datasheet simply does not publish numbers there. PEB time follows a thickness-threshold rule (2/3/4 minutes at 25/50 µm cutoffs) rather than a continuous table. A distinctive storage gotcha: KMPR 1000 resists must be stored frozen (14°F/-10°C) and require a full 24-hour room-temperature thaw before use — treating it like a room-temperature-stable resist will produce inconsistent films. The resist strips cleanly with Remover PG when only lightly cross-linked, but a fully electroformed/plated structure needs the stronger Remover PG → XP Remover K → XP Neutralizer K sequence to fully dissolve.", "developerFamily": "tmah", "references": [ { "type": "paper", "title": "Fabrication of thick electroforming micro mould using a KMPR negative tone photoresist", "authors": "Lee et al.", "journal": "Journal of Micromechanics and Microengineering", "year": 2008, "doi": "10.1088/0960-1317/18/5/055032", "url": "https://doi.org/10.1088/0960-1317/18/5/055032", "accessedDate": "2026-07-12", "summary": "180 µm, 18:1 KMPR molds — strippable, unlike SU-8." }, { "type": "paper", "title": "UV Lithography and Molding Fabrication of Ultrathick Micrometallic Structures Using a KMPR Photoresist", "authors": "Shin et al.", "journal": "Journal of Microelectromechanical Systems", "year": 2010, "doi": "10.1109/JMEMS.2010.2045880", "url": "https://doi.org/10.1109/JMEMS.2010.2045880", "accessedDate": "2026-07-12", "summary": "KMPR molds for mm-wave traveling-wave-tube structures." } ], "provenance": { "datasheetUrl": "https://www.seas.upenn.edu/~nanosop/documents/KMPRDataSheetver4_2a.pdf", "datasheetVersionOrDate": "Ver. 4.2 (printed in the footer of every page: \"Ver. 4.2\")", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "Fetched directly from the assigned URL (seas.upenn.edu/~nanosop mirror) and confirmed to be the correct document: Kayaku Microchem / MicroChem's own \"KMPR® 1000 Chemically Amplified Negative Photoresist\" processing datasheet, Ver. 4.2. This is a university (UPenn nanoSOP) mirror of the manufacturer's own document, not a university-authored SOP.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "lor-3a", "name": "LOR 3A", "manufacturer": "Kayaku Advanced Materials", "productLine": "LOR / PMGI lift-off resists (LOR-A series)", "aliases": [ "LOR3A", "LOR-3A", "MicroChem LOR 3A" ], "tone": null, "toneNote": "LOR 3A is a non-photoimageable lift-off underlayer, not an imaging photoresist — it is never itself exposed, so 'tone' does not apply. It is left null rather than set to 'positive' (a common but incorrect shorthand, since LOR is not a dissolution-inhibition/exposure-based system at all).", "chemistry": "ancillary", "grayscaleSuitable": false, "grayscaleNote": "Not applicable — LOR is a non-photoimageable underlayer; grayscale/continuous-tone relief is an imaging-resist property that has no meaning for a layer that is never exposed.", "status": "active", "successorSlug": null, "summary": "LOR 3A is a polydimethylglutarimide (PMGI)-based lift-off underlayer from the low-dissolution-rate 'LOR-A' series, originally developed by MicroChem Corp. and now manufactured and sold by Kayaku Advanced Materials. It is spin-coated beneath a conventional UV/DUV/e-beam imaging resist; only the top imaging layer is exposed, and on development LOR 3A dissolves isotropically to form a controllable re-entrant (undercut) profile that enables clean, discontinuous metal-film lift-off.", "thicknessRange": { "min_um": 0.28, "max_um": 0.567, "basis": "curve-span", "source": "curve-span: the LOR/PMGI datasheet (Rev. A) states only a whole-product-line figure (\"Film thicknesses for depositions from <20nm - >5µm\", Benefits, p.1) spanning every LOR/PMGI grade, not an LOR 3A-specific achievable range. min/max are the span of this recipe's own digitized LOR 3A spin curve, pixel-calibrated 2026-07-12 off the 'Spin Speed vs Thickness - Intermediate Films' chart, p.5 (the LOR 3A/3B trace only plots 4 markers, 1000-4000 rpm; the prior 500/1500/2500/4500/5000 rpm points were unsupported eyeball extrapolations and have been dropped). Adjudicated 2026-07-12: max_um 0.567 confirmed correct against check's own pixel re-extraction; check's top-level thicknessRange (0.27-0.90) was stale and inconsistent with its own re-extracted spin-curve points." }, "thicknessRangeNote": "Approximate range read off manufacturer spin-curve figures over roughly 1000-5000 rpm (see spinCurves). No numeric thickness-vs-speed table specific to LOR 3A was found in either primary datasheet obtained — only plotted curves.", "spinCurves": [ { "label": "LOR 3A", "points": [ { "rpm": 1000, "um": 0.567 }, { "rpm": 2000, "um": 0.402 }, { "rpm": 3000, "um": 0.335 }, { "rpm": 4000, "um": 0.28 } ], "source": "re-extracted 2026-07-12, pixel-calibrated read of the combined 'LOR 3A, LOR 3B' trace (black diamond markers) in the 'Spin Speed vs Thickness – Intermediate Films' chart, p.5 of the Kayaku/MicroChem 'LOR and PMGI Resists' datasheet (Rev. A). The chart plots exactly 4 markers for this trace (1000/2000/3000/4000 rpm) — no data exists at 500, 1500, 2500, 4500 or 5000 rpm; supersedes the earlier eyeball read (7 fabricated points spanning 1000-5000 rpm, cross-referenced against MicroChem's 2002 flyer Figure 1). Neither document publishes a numeric table isolating LOR 3A alone — this remains a figure read (±10% uncertainty), just a pixel-calibrated one rather than an eyeball estimate. Treat as approximate; scrutinise before relying on for a critical process.", "figureRead": true } ], "spinNotes": "Recommended baseline coating parameters (Table 1, applies across the LOR/PMGI line, not LOR 3A-specific): dispense 5 mL for a 150 mm wafer, dynamic dispense 3-5 s at 300-500 rpm, acceleration 10,000 rpm/s, terminal spin speed 3,000 rpm held for 45 s, edge-bead removal with EBR PG. Spin speeds of 2,500-4,500 rpm give maximum coating uniformity (higher speeds for smaller substrates, lower for larger/topographic ones). Acetone and conventional resist edge-bead removers are explicitly NOT recommended with LOR (causes precipitation) — use EBR PG.", "adhesion": { "hmds": false, "notes": "HMDS priming is explicitly stated as typically NOT required to promote adhesion with LOR/PMGI products. LOR exhibits excellent inherent adhesion to Si, glass, NiFe, GaAs, InP and other III-V/II-VI materials, and Au. Substrate prep: solvent clean or dilute-acid rinse followed by DI water rinse, then a dehydration bake at 200°C for 5 min (contact hotplate) or 30 min (convection oven) immediately before coating." }, "rehydration": null, "softbake": { "temp_c": 180, "time_s": 180, "method": "hotplate", "notes": "180°C for 3 min is the specific bake condition the manufacturer states was used to generate its own published spin-curve/optical-constant data ('Products were soft-baked at 180°C for 3 min', Technical Data section). It is not published as 'the' single recommended production bake. The datasheet's general recommended prebake range is 150-200°C (hotplate preferred, oven-compatible; some PMGI grades up to 250°C) — no single fixed time is prescribed for LOR 3A, because prebake temperature is the dominant control lever for undercut rate and is meant to be tuned experimentally (a temperature/time matrix) against the target undercut, developer choice, and develop time. An earlier (2002) MicroChem flyer for the LOR-A series quotes a narrower 150-190°C recommended range.", "source": "Technical Data section note (p.5) and 'Soft-bake/Prebake Process' section (p.3), Kayaku/MicroChem 'LOR and PMGI Resists' datasheet Rev. A" }, "exposureDose": { "doses": [], "datasheetBasis": "Not applicable", "_note": "Not photoimageable. LOR 3A is never itself exposed — only the top imaging resist in the bilayer stack is patterned, and that resist's exposure dose is set by its own manufacturer datasheet, entirely independent of the LOR layer beneath it." }, "peb": { "temp_c": null, "time_s": null, "notes": "Not applicable. The datasheet states explicitly: 'LOR/PMGI does not require post-exposure baking... Refer to patterning resist manufacturer process recommendations to determine whether a PEB step is required' (i.e. any PEB is for the top imaging resist, not LOR).", "source": "'Post-Exposure (PEB) Process' section, p.4, Kayaku/MicroChem 'LOR and PMGI Resists' datasheet Rev. A" }, "floodExposure": { "dose_mJcm2": null, "notes": "Not required for the standard LOR bilayer lift-off process (no extra flood exposure, develop, amine treatment, or toxic soak steps needed). A deep-UV (240-290 nm) flood-exposure 'Cap-On' process exists in the same datasheet for straighter sidewalls, but it is described specifically for PMGI, not for LOR/LOR 3A.", "source": "'Benefits' list and 'Application and Processing the Patterning Resist Layer' section, Kayaku/MicroChem 'LOR and PMGI Resists' datasheet Rev. A" }, "develop": { "developer": "0.26N (2.38%) TMAH metal-ion-free developer — e.g. Shipley/Rohm and Haas CD-26 or TOK NMD-3 — is the developer class the LOR-A series (including LOR 3A) is optimized for; the LOR-B series is instead optimized for lower-normality/metal-ion-bearing developers such as AZ 400K 1:4 or Shipley MF-319.", "dilution": "ready-to-use (as supplied)", "time_s": null, "method": "immersion (spray development is recommended instead for thick, >2 µm LOR/PMGI stacks, for straighter sidewalls)", "rinse": "DI water", "source": "'Development Process' section (p.4) and Product Selection Guide (p.6), Kayaku/MicroChem 'LOR and PMGI Resists' datasheet Rev. A. Develop time is explicitly not published as a single fixed number — the datasheet states it depends on the combined thickness of the LOR/PMGI layer and the patterning resist layer, and on the desired undercut (see undercut-rate figures); it must be set experimentally, not read off a table." }, "hardbake": null, "hardbakeNote": "No hardbake step is described for LOR in either primary datasheet — the process proceeds directly from development to metal deposition to lift-off.", "descum": "Not required. The datasheet states no intermixing occurs between LOR and the imaging resist above it, so the bilayer stack does not need a plasma descum step between coating the LOR layer and the patterning resist.", "applications": [ "lift-off" ], "etchResistance": null, "liftoffSuitable": true, "platingSuitable": null, "stripper": "MicroChem/Kayaku Remover-PG is the recommended stripper; baseline two-tank process at 60°C for 30 min per tank, optionally ultrasonic-assisted to improve strip efficiency (actual time varies with prebake temperature, step coverage, and resist profile). Acetone will NOT dissolve or remove LOR — keep acetone (and PGMEA, ethyl-lactate) waste streams separate, as LOR precipitates in these solvents and can clog lines.", "storage": "Store upright in original sealed containers in a dry area between 4-27°C (40-80°F), away from ignition sources, light, heat, oxidants, acids, and reducers. Do not use past the expiration date (1 year from date of manufacture). Recommended processing environment: 20-25°C ± 1°C, 35-45% ± 2% relative humidity.", "notes": "LOR 3A is not a photoresist in the imaging sense — it is a non-photoimageable PMGI underlayer that is coated and soft-baked beneath a conventional imaging resist, which alone is exposed and developed. During development the LOR layer dissolves isotropically (laterally undercutting beneath the imaged top resist), producing the re-entrant sidewall profile a clean bilayer lift-off requires. Undercut amount is set primarily by LOR soft-bake temperature — higher bake temperature lowers the dissolution rate and reduces undercut for a given develop time; the manufacturer's own bar-chart data (0.26N TMAH/CD-26 developer) give LOR-A-class undercut rates of 42 Å/s at 190°C, 67 Å/s at 170°C, and 111 Å/s at 150°C (recommended bake range 150-200°C). This rate is published for the 'LOR A' class as a whole (which LOR 3A belongs to), not broken out for the 3A sub-grade specifically. Secondary levers are prebake time, developer choice/normality, develop mode, and develop time (increasing develop time increases undercut for otherwise fixed bake conditions). HMDS priming is explicitly not required. For clean lift-off, LOR film thickness should exceed the deposited metal thickness, by roughly 25-33% per the two primary datasheets (guidance differs slightly: 'typically 1.2 to 1.3 times' the metal thickness in the 2002 flyer vs. 'typically by 25%' in the later Rev. A datasheet).", "developerFamily": "tmah", "references": [ { "type": "paper", "title": "A lift-off process for high resolution patterns using PMMA/LOR resist stack", "authors": "Chen", "journal": "Microelectronic Engineering", "year": 2004, "doi": "10.1016/j.mee.2004.02.053", "url": "https://doi.org/10.1016/j.mee.2004.02.053", "accessedDate": "2026-07-12", "summary": "Sub-100 nm lift-off geometry demonstrated with a PMMA/LOR resist bilayer." } ], "provenance": { "datasheetUrl": "https://apps.mnc.umn.edu/pub/pds/lor.pdf", "datasheetVersionOrDate": "Kayaku/MicroChem 'LOR and PMGI Resists', Rev. A (undated on the document itself); cross-checked against the earlier MicroChem 'LOR™ Lift-Off Resists' flyer, © MicroChem Corp. 2002, mirrored at https://amolf.nl/wp-content/uploads/2016/09/datasheets_LOR_datasheet.pdf. The current kayakuam.com-hosted PDF (https://kayakuam.com/wp-content/uploads/2023/06/KAM-LOR-PMGI-Datasheet-4.30.24-final-1.pdf, dated 4/30/24 per its filename) returned HTTP 403 and could not be fetched directly; the two documents actually read are university-hosted mirrors of the same manufacturer-authored content.", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://bionium.miami.edu/_assets/pdf/lor-3a-photoresist-process.pdf", "what": "University of Miami cleanroom process recipe for LOR 3A — used only to corroborate typical practical parameters (180°C/5min softbake, 3000rpm/35s spin, CD-26 60-90s or MF-319 45-60s develop), not as a primary numeric source" }, { "url": "https://cns1.rc.fas.harvard.edu/facilities/docs/SOP112_r1_1_%20LOR.pdf", "what": "Harvard CNS SOP112 lift-off processing procedure — used only to corroborate typical bake (180°C/4min) and CD-26 develop (75s) practice, and to confirm HMDS is not used with LOR; not a primary numeric source" } ] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "photoimageable": false, "humanVerified": false }, { "slug": "lor-5a", "name": "LOR 5A", "manufacturer": "MicroChem (MCC) — now Kayaku Advanced Materials", "productLine": "LOR A series", "_productLineNote": "The document never states in words that 'LOR 5A belongs to the LOR A series.' The connection is inferred from (a) the product's own name (the 'A' suffix, mirroring LOR 3A/3B and LOR 10A/10B), and (b) the datasheet's own use of 'LOR_A' / 'LOR A' as a printed family label — Table 1 (Optical Constants) and Table 2 (Cauchy Parameters), p.5, both have a 'LOR_A' row, and the Product Selection Guide, p.6, has a 'LOR A' column — but none of those tables lists LOR 5A by name as a member. Recorded as a naming-convention inference, not a verbatim statement.", "aliases": [ "LOR5A", "MicroChem LOR 5A", "Kayaku LOR 5A" ], "tone": null, "chemistry": "ancillary", "photoimageable": false, "grayscaleSuitable": false, "grayscaleNote": "Not applicable and not addressed by the document. LOR 5A is not photoimageable and carries no exposure dose; the datasheet states 'LOR or PMGI does not require an exposure step' in the standard bi-layer lift-off process (p.4), so dose-modulated grayscale profiling — which requires a photoimageable resist — cannot apply to this material.", "status": "active", "successorSlug": null, "summary": "LOR 5A is a mid-thickness grade in MicroChem/Kayaku's LOR series of PMGI-type (polydimethylglutarimide) lift-off underlayers: it is spin-coated beneath a conventional photoresist, is never itself exposed, and its develop-time dissolution — governed chiefly by pre-bake temperature — produces the re-entrant, undercut sidewall profile that enables clean metal lift-off.", "thicknessRange": { "min_um": 0.47, "max_um": 0.98, "basis": "curve-span", "source": "curve-span: the document states only a whole-product-line thickness range ('Film thicknesses for depositions from <20nm - >5μm', Benefits, p.1) spanning every LOR/PMGI grade, not LOR 5A specifically. The range recorded here is the min/max of LOR 5A's own plotted curve (1000-4000 rpm) in the 'Spin Speed vs Thickness - Intermediate Films' figure, p.5." }, "spinCurves": [ { "label": "LOR 5A", "points": [ { "rpm": 1000, "um": 0.98 }, { "rpm": 2000, "um": 0.67 }, { "rpm": 3000, "um": 0.55 }, { "rpm": 4000, "um": 0.47 } ], "source": "read from figure \"Spin Speed vs Thickness - Intermediate Films\", p.5 of MicroChem \"LOR and PMGI Resists\" datasheet. This figure plots five grades: LOR 7B (blue square), LOR 5A (red circle), LOR 5B (green x), a COMBINED 'LOR 3A, LOR 3B' trace (black diamond), and SF6 (purple triangle). LOR 5A was identified and read as the red-circle series specifically because it has its OWN distinct legend entry ('LOR 5A', not grouped with any other grade) and its own separate curve, clearly offset above the neighboring LOR 5B (green) and well above the combined LOR 3A/3B (black) trace in the same chart — i.e. this is NOT the combined multi-grade series that caused a misread on a sibling LOR 3A recipe in this project. Only 4 markers are plotted for LOR 5A (1000/2000/3000/4000 rpm); no 1500/2500/3500 rpm points exist for this series (unlike SF6, which has more points in the same figure). Read from a rendered PNG crop of the chart at native resolution; an eyeball read, ±10-15% plausible per-point error, not a numeric table. Chart footnote (p.5): 'Products were soft-baked at 180 ºC for 3 min' — the measurement condition for the plotted films, not a stated LOR-5A-specific process bake recommendation.", "figureRead": true } ], "spinNotes": "Only 4 spin-speed points (1000/2000/3000/4000 rpm) are plotted for LOR 5A; extrapolation below 1000 or above 4000 rpm is not supported by the chart. The page's boxed 'RECOMMENDED COATING PARAMETERS' (dispense 5 ml on a 150 mm Si wafer, dynamic dispense 3-5 s at 300-500 rpm, acceleration 10,000 rpm/s, terminal spin speed 3,000 rpm, spin time 45 s, EBR PG for edge-bead removal) appears once for the whole Technical Data page (p.5) and is not stated to be grade-specific, so it is recorded here as general page-level guidance rather than a confirmed LOR-5A parameter. Separately, general text (p.2) states spin speeds of 2,500-4,500 rpm give 'maximum coating uniformity' across the LOR/PMGI line, with lower speeds favored for larger or irregular/topographic substrates.", "adhesion": { "hmds": false, "notes": "\"Primers such as HMDS (hexamethyldisilazane) are typically NOT required to promote adhesion with PMGI/LOR products when used as recommended.\" LOR/PMGI is described as exhibiting 'excellent adhesion to most semiconductor, GaAs, and thin-film head substrates.' A dehydration bake (200°C, 5 min contact hot plate, or 30 min convection oven) is recommended before coating if maximum process reliability is needed. Source: \"Substrate preparation\", p.2." }, "rehydration": null, "softbake": { "temp_c": null, "time_s": null, "method": "hotplate", "notes": "THE BAKE TEMPERATURE IS THE UNDERCUT CONTROL, and the document gives only ranges, not an LOR-5A-specific scalar. General guidance (p.3): 'The recommended bake temperature range is 150°C - 200°C, although some PMGI products may be baked to 250°C... Hot plates are the preferred tool for the pre-bake; however, LOR/PMGI resists are also compatible with convection oven processes... a matrix design varying pre-bake temperature and time is recommended for process fine-tuning.' Separately, the Technical Data chart footnote (p.5) that includes LOR 5A's own spin curve states those plotted films 'were soft-baked at 180 ºC for 3 min' (180 s) — this is the measurement condition for the chart, not a stated process recommendation specific to LOR 5A. No grade-specific bake-temperature-vs-undercut-rate curve is published for LOR 5A: Figures 5a/5b (undercut rate vs. bake temperature and vs. bake time) are explicitly labeled 'LOR 10B' only, and Figure 6's dissolution-rate comparison groups grades only at the family level (LOR_A / LOR_B / SF / SF Slow bars, no per-grade numbers) at a single condition (180°C).", "source": "p.3 (\"Soft-bake/Prebake Process\") and p.5 chart footnote, MicroChem \"LOR and PMGI Resists\" datasheet" }, "exposureDose": null, "peb": null, "floodExposure": null, "develop": { "developer": "Document does not name a specific commercial developer product for LOR 5A; general text states LOR/PMGI is compatible with both metal-ion-free (MIF) and metal-ion-bearing (MIB) developer chemistries.", "dilution": null, "time_s": null, "method": null, "rinse": null, "source": "p.1 (\"Compatible with TMAH and metal-ion bearing developers\"), p.4 (\"Development Process\"), and p.6 Product Selection Guide (Developer Compatibility row), MicroChem \"LOR and PMGI Resists\" datasheet", "_note": "The Product Selection Guide (p.6) gives family-level developer-compatibility stars: 'LOR A' → 0.26N MIF only; 'LOR B' → 0.24N MIF and MIB. Figure 7 (p.4) separately reports dissolution rate vs. temperature for 2.38% TMAH (0.26N) and 2.2% TMAH w/surfactant (0.24N) developers, but only for product 'SF11' — not LOR 5A. Because the document never explicitly states which family (A or B) LOR 5A belongs to (see _productLineNote), and because no developer, dilution, develop time, or rinse step is published by name for LOR 5A itself, all of those fields are left null rather than assigned from the family table or from another grade's entry." }, "hardbake": null, "descum": null, "applications": [ "lift-off" ], "etchResistance": null, "liftoffSuitable": true, "platingSuitable": null, "stripper": "Remover PG (MicroChem/Kayaku). Baseline two-tank process: 60°C for 30 minutes in the first tank, then a 60°C rinse in the second tank; ultrasonic agitation improves removal efficiency. Actual time depends on pre-bake conditions, step coverage, and resist profile. Removal rate is reported as a function of soft-bake temperature and remover-bath temperature (Figure 9, p.4), but that figure is labeled for the 'SF Series', not LOR 5A specifically. Source: \"Lift-Off Process\", p.4.", "storage": "\"Store upright in original sealed containers in a dry area between 4 and 27°C (40-80°F). Keep away from sources of ignition, light, heat, oxidants, acids, and reducers. Do not use after the expiration date (1 year from date of manufacture).\" Source: \"LOR/PMGI Storage\", p.7 (general to all LOR/PMGI products; not stated to be LOR-5A-specific, but no grade-specific storage guidance exists in the document).", "notes": "LOR 5A is an ancillary, non-photoimageable PMGI-type (polydimethylglutarimide) underlayer: it is spin-coated beneath a conventional photoresist and never exposed itself, so it carries no dose, tone, or lithographic pattern of its own — only the imaging resist above it is exposed and developed. The undercut geometry that enables clean metal lift-off is a develop-time and pre-bake-temperature-controlled dimension, not a lithographic one: the datasheet states pre-bake temperature has 'the greatest influence on undercut rate,' with pre-bake time, the imaging resist's own exposure dose, developer choice, develop mode, and develop time as secondary factors. No grade-specific bake-temperature-vs-dissolution-rate curve is published for LOR 5A — the only such curves in this document (Figures 5a/5b) are for LOR 10B — so LOR 5A's own undercut rate cannot be read off this datasheet and should be characterized on-tool. In practice, too little undercut leaves an insufficient re-entrant profile, so evaporated or sputtered metal bridges over the sidewall and lift-off fails or leaves ragged edges; too much undercut can collapse the unsupported span of imaging resist over the gap, degrading pattern fidelity — both failure modes are tuned via pre-bake temperature/time and develop time, not exposure.", "developerFamily": null, "references": [ { "type": "paper", "title": "A lift-off process for high resolution patterns using PMMA/LOR resist stack", "authors": "Chen", "journal": "Microelectronic Engineering", "year": 2004, "doi": "10.1016/j.mee.2004.02.053", "url": "https://doi.org/10.1016/j.mee.2004.02.053", "accessedDate": "2026-07-12", "summary": "Sub-100 nm lift-off geometry demonstrated with a PMMA/LOR resist bilayer." } ], "provenance": { "datasheetUrl": "https://apps.mnc.umn.edu/pub/pds/lor.pdf", "datasheetVersionOrDate": "Rev. A (printed at the bottom of p.7; no other date is printed anywhere in the document)", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "kayakuam.com (the current rights-holder for the MicroChem/Kayaku LOR and PMGI product line) returns 403 to automated fetches on every path as verified 2026-07-10, so this University of Minnesota Nano Fabrication Center public PDS mirror (apps.mnc.umn.edu/pub/pds/lor.pdf) is used as the primary source instead. The mirrored file is MicroChem's own 'LOR and PMGI Resists' datasheet (MICRO•CHEM letterhead, Rev. A) — a manufacturer-authored document hosted by a university, not a university-authored SOP — which is why it is acceptable as datasheetUrl rather than only a secondarySources entry.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "ma-n-1410", "name": "ma-N 1410", "manufacturer": "micro resist technology GmbH", "productLine": "ma-N 1400 series", "aliases": [ "maN 1410", "ma-N1410" ], "tone": "negative", "chemistry": "bisazide-novolak", "_chemistryNote": "This document ('Processing guidelines — Negative Tone Photoresist Series ma-N 1400') never uses the words 'bisazide' or 'novolak' anywhere in its text; it describes ma-N 1400 only as 'a negative tone photoresist series designed for the use in microelectronics and microsystems technology.' Some other ma-N 1400-series documents reportedly state an 'aromatic bisazide / novolak' chemistry, but that wording is not present in the specific document read for this extraction, so chemistry is left null rather than assumed from the product family.", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Document does not market ma-N 1400 for greyscale or 3D lithography; it targets microelectronics pattern transfer, etch masking, and PVD lift-off with nearly vertical (standard) or optionally undercut (lift-off) binary profiles.", "status": "active", "successorSlug": null, "summary": "ma-N 1410 is the third-thinnest grade in micro resist technology's ma-N 1400 negative-tone photoresist series, nominally coating 1.0 µm at 3000 rpm/30 s, used as a high-etch-resistance mask or, with an extended develop time, for undercut lift-off patterning.", "thicknessRange": { "min_um": 1, "max_um": 1, "basis": "stated", "source": "stated — the only achievable-thickness figure this document gives for ma-N 1410 specifically is the single nominal value 1.0 ± 0.1 µm at 3000 rpm/30 s (Physical properties of the resist solution table, p.1). No separate min-max coating range is stated in prose for this grade; see spinNotes for why the fuller multi-rpm curve is not converted into a range here." }, "spinCurves": [ { "label": "ma-N 1410", "points": [ { "rpm": 1000, "um": 1.75 }, { "rpm": 2000, "um": 1.25 }, { "rpm": 3000, "um": 1 }, { "rpm": 4000, "um": 0.87 }, { "rpm": 5000, "um": 0.77 }, { "rpm": 6000, "um": 0.7 } ], "source": "read from figure, \"Fig. 1: Spin curves of the ma-N 1400 series, 30 s spin time\", p.2 of ma-N 1400 series Processing guidelines (micro resist technology, rev. ls.05.11.25.02); trace identified as the red line, second from the top, per the chart's own inline legend (ma-N 1420 black / ma-N 1410 red / ma-N 1407 green / ma-N 1405 blue), matching the series' viscosity ranking (1420 highest > 1410 > 1407 > 1405 lowest); the curve carries no individual point markers, so points were read at its 6 labelled x-axis gridlines (1000/2000/3000/4000/5000/6000 rpm); anchor check: the chart's own vertical reference line at 3000 rpm crosses the red trace at ~1.0 µm, matching the numeric-table anchor (Film Thickness = 1.0 ± 0.1 µm at 3000 rpm/30s, 'Physical properties of the resist solution' table p.1, corroborated by the 'Processing conditions - STANDARD PROCESS' table p.2) essentially exactly; digitized 2026-07-12", "figureRead": true } ], "spinNotes": "Fig. 1 (p.2, 'Spin curves of the ma-N 1400 series, 30 s spin time') plots a full thickness-vs-spin-speed curve for ma-N 1410 alongside ma-N 1405/1407/1420 (four individually legended, distinctly colored lines, no individual point markers, 1000-6000 rpm, ranked by viscosity: 1420 highest/topmost black, 1410 second red, 1407 third green, 1405 lowest blue). The red (ma-N 1410) trace has now been digitized at its 6 labelled x-axis gridlines — see spinCurves; the chart's own vertical reference line at 3000 rpm crosses the trace at ~1.0 µm, matching the numeric-table anchor below essentially exactly, which is the strongest evidence the correct (red) trace was isolated in a chart carrying no point markers. The datasheet additionally publishes a single verified numeric ANCHOR POINT — 3000 rpm → 1 µm (numeric table 'Physical properties of the resist solution', p.1 of ma-N 1400 series processing guidelines (footnote 1: 'Spin coated at 3000 rpm for 30 s'); Film Thickness = 1.0 ± 0.1 µm for ma-N 1410. Corroborated by the 'Processing conditions - STANDARD PROCESS' table, p.2, which also lists 1.0 µm for ma-N 1410 at the same 3000 rpm/30 s condition.). A prior extraction pass declined to digitize this figure at all (citing two unrelated prior misreads on other resists, SU-8 2050 and LOR 3A); this pass re-attempted the read directly against the source chart, anchored and cross-checked against the two numeric tables above, rather than repeating that blanket refusal. 'Uniform coatings are obtained by spin coating of ma-N 1400 solutions in the thickness range indicated in the spin curves' (p.2); no dispense volume, acceleration ramp, or edge-bead-removal step is described anywhere in this document.", "adhesion": { "hmds": true, "notes": "'For improving resist film adhesion to Si and SiO2 substrates it is advisable to apply an adhesion promoter such as HMDS' (Substrate preparation, p.2). Standard process substrate preparation is 'Oven: 200 °C, 30 min (HMDS for Si and SiO2 substrates)' (Processing conditions table, p.2)." }, "rehydration": null, "softbake": { "temp_c": 100, "time_s": 90, "method": "hotplate", "notes": "Value shown (100 °C, 90 s) is the ma-N 1410 column of the hotplate prebake row; other grades in the same table use different times at the same 100 °C (1405/1407: 60 s, 1420: 120 s). An oven alternative is also given: 100-105 °C for 15-30 min (shared across all grades). Recommended prebake temperature also varies by substrate: Si 100-120 °C, Au 120 °C, Si3N4 120 °C, GaAsP/GaAs 100 °C (p.3) — the 100 °C/90 s value applies to the standard (unspecified-substrate) process. A higher prebake (up to 160 °C) or longer time increases etch resistance/thermal stability but also increases developing time.", "source": "Processing conditions - STANDARD PROCESS table, p.2; Recommended prebake temperatures table, p.3" }, "exposureDose": { "doses": [ { "wavelength_nm": null, "value_mJcm2": 450, "source": "Processing conditions - STANDARD PROCESS table, p.2" } ], "datasheetBasis": "'Exposure dose [mJ cm-2]... broadband exposure, intensity measured at λ=365 nm' (footnote 1, Processing conditions table, p.2). 'The resists are effective for broadband or i-line exposure' (Exposure section, p.3).", "_note": "[RECLASSIFIED during audit] The extraction agent filed 450 mJ/cm² as at365_mJcm2. The datasheet says \"broadband exposure, intensity measured at λ=365 nm\" — that is a BROADBAND dose whose intensity was METERED at 365 nm, not a monochromatic i-line dose. Those are different claims and the distinction is the whole point of datasheetBasis, so the value now lives in value_mJcm2 with both at365 and at405 null. Original agent note follows. 450 ± 30 mJ/cm² is explicitly attributed to intensity measured at 365 nm (i-line), even though the exposure itself is described as broadband. Fig. 3 (p.3) plots UV/vis absorption of unexposed/exposed ma-N 1400 out to 450 nm and marks both the i-line (365 nm) and h-line (405 nm) wavelengths on the axis, but no dose VALUE at 405 nm is published anywhere in the document, so at405_mJcm2 is left null rather than derived from the 365 nm number or the absorption curve." }, "peb": null, "floodExposure": null, "develop": { "developer": "ma-D 533/S", "dilution": null, "time_s": 30, "method": "immersion", "rinse": "Developed resist films are thoroughly rinsed with deionized water and then dried.", "source": "Processing conditions - STANDARD PROCESS table (footnote 2: immersion development), p.2; Develop section, p.4", "_note": "Manufacturer states 30 ± 10 s for ma-N 1410 specifically (other grades: 1405 = 20±5 s, 1407 = 25±5 s, 1420 = 60±10 s). Ready-to-use developer, no dilution given. Developer temperature should be 20-25 °C." }, "hardbake": { "temp_c": 100, "time_s": 1800, "notes": "Optional, series-wide (not per-grade) recommendation: 'Hardbaking of the developed resist patterns is suggested in an oven at 100 °C for approximately 30 min. A temperature ramp is beneficial in order to reduce pattern reflow.'", "source": "Hardbake (optional) section, p.4" }, "descum": null, "applications": [ "etch-mask", "lift-off", "mems-structural" ], "etchResistance": "\"Well suitable as an etch mask exhibiting high dry and wet etch resistance\" (Characteristics, p.1).", "liftoffSuitable": true, "platingSuitable": null, "stripper": "mr-Rem 660 (solvent based) and ma-R 404/S (strongly alkaline) recommended; acetone, N-methylpyrrolidone (NMP) or oxygen plasma also suitable for residue-free removal (Removal section, p.4). For lift-off specifically: mr-Rem 660 or NMP at 40-60 °C assisted by ultrasonics, or acetone assisted by ultrasonics (p.5).", "storage": "\"Storage at temperatures of 18 – 25 °C is recommended. Do not store ma-N 1400 resists in a refrigerator. Resists and unprocessed resist films have to be stored under yellow light. Keep the bottle closed when not in use. Under these conditions a shelf life of 6 months from the date of manufacture is ensured.\" (Storage section, p.5)", "notes": "ma-N 1410 is the third grade (of four) in micro resist technology's ma-N 1400 negative-tone series, nominally coating 1.0 ± 0.1 µm at 3000 rpm/30 s and developing in the metal-ion-free ma-D 533/S developer recommended for microelectronics work. The standard process yields nearly vertical sidewalls; the datasheet describes a separate lift-off recipe in which undercut profile is tuned primarily by extending development time (holding exposure dose and prebake constant) rather than by underexposing — a worked 2.0 µm-thick example (not this grade specifically) shows undercut growing from 0.6 to 2.1 µm as development time increases from 65 to 120 s at a fixed 550 mJ/cm² dose and 100 °C/120 s prebake. For clean lift-off the datasheet recommends a resist film 1.5-2x the metal deposition thickness, plus — for sputtered metals in particular — a higher prebake and/or a deep-UV (200-300 nm) flood exposure at 2-5x the patterning dose to thermally stabilize the pattern before deposition. The exposure dose (450 ± 30 mJ/cm²) is explicitly measured at 365 nm on a broadband tool; the document plots UV/vis absorption out past 405 nm but never gives a dose at that wavelength, so the h-line field is left null. The datasheet's Fig. 1 plots a full ma-N 1400-series spin curve (four grades, 1000-6000 rpm) that would let a human read additional rpm points for ma-N 1410, but only the single verified 3000 rpm/1.0 µm numeric-table anchor is published here as structured data — see spinNotes. Chemistry classified as bisazide-novolak from microresist.de's statement that the ma-N 1400 series is an 'aromatic bisazide/novolac based resist series' (chemistry classified 2026-07-12 from manufacturer SDS/TDS).", "developerFamily": "tmah", "references": [ { "type": "paper", "title": "Combined ultraviolet- and electron-beam lithography with Micro-Resist-Technology GmbH ma-N1400 resist", "authors": "Thoen et al.", "journal": "Journal of Vacuum Science & Technology B", "year": 2022, "doi": "10.1116/6.0001918", "url": "https://doi.org/10.1116/6.0001918", "accessedDate": "2026-07-12", "summary": "one ma-N 1400 layer exposed by UV and 100 kV e-beam for an on-chip THz spectrometer" } ], "provenance": { "datasheetUrl": "https://www.nanophys.kth.se/nanolab/resists/vh_man_1400_en_05112502_ls.pdf", "datasheetVersionOrDate": "ls.05.11.25.02", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://www.microresist.de/wp-content/uploads/2022/08/NegativeResists_Flyer_Aug22.pdf", "what": "fetched and read in this same session (for the mr-dwl-40 recipe); used only to corroborate that ma-D 533/S is TMAH-based (that flyer's Developer table, p.2, lists 'ma-D 533/S (TMAH based)' for the ma-N 1400 series), since the primary datasheet referenced above states only 'aqueous alkaline development' and does not itself name the alkali type for ma-D 533/S." }, { "url": "https://microresist.de/en/produkt/ma-n-1400-series/", "what": "microresist.de states the ma-N 1400 series is an 'aromatic bisazide/novolac based resist series'; the basis for classifying ma-N 1410 as bisazide-novolak." } ] }, "_provenanceNote": "Document is micro resist technology's own 'Processing guidelines — Negative Tone Photoresist Series ma-N 1400' technical datasheet (revision ls.05.11.25.02, 5 pages), mirrored by KTH Nanolab (nanophys.kth.se) at a stable institutional URL rather than hosted on microresist.de/microresist.com directly. This is a full multi-page processing datasheet (substrate prep, bake, exposure, develop, hardbake, lift-off, storage), not a marketing flyer — closer to a true TDS than the ma-P 1275G or mr-DWL 40 documents in this extraction batch. Extracted independently, without reference to the ma-N 1420 entry already in this repository (which reportedly used the phrase 'aromatic bisazide / novolak' from a document not seen in this extraction).", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "ma-n-1420", "name": "ma-N 1420", "manufacturer": "micro resist technology", "productLine": "ma-N 1400 series", "aliases": [ "maN 1420", "ma N 1420", "ma-N1420" ], "tone": "negative", "chemistry": "bisazide-novolak", "chemistryNote": "Aromatic bisazide/novolak per the manufacturer's processing guidelines: a novolak resin cross-linked by a photolyzed bisazide. Deliberately NOT tagged dnq-novolak — diazonaphthoquinone is a positive-tone dissolution-inhibition mechanism, and this is a negative-tone cross-linking resist.", "grayscaleSuitable": false, "grayscaleNote": "Not addressed in the datasheet either way. ma-N 1400 is marketed as a high-contrast negative resist for standard lithography and lift-off, not for grayscale/3D relief work; no dose-to-thickness linearity data (which grayscale processing requires) is published.", "status": "active", "successorSlug": null, "summary": "ma-N 1420 is a negative-tone photoresist in micro resist technology's ma-N 1400 series, based on an aromatic bisazide/novolak chemistry with aqueous alkaline development. It is used both as a high dry/wet-etch-resistant mask for pattern transfer and, with a tunable vertical-to-undercut profile, as a single-layer lift-off resist for physical vapour deposition (evaporation/sputtering) of metals.", "thicknessRange": { "min_um": 1.41, "max_um": 3.35, "basis": "curve-span", "source": "curve-span: the ma-N 1400 datasheet states no achievable-thickness range for ma-N 1420 — only a 2.0 ± 0.1 µm nominal at 3000 rpm/30 s (Physical properties table, p.1) and the Fig. 1 spin curves (p.2-3). min/max are the span of this recipe's own digitized ma-N 1420 spin curve, re-extracted 2026-07-12 via PyMuPDF vector-path centroid read of the native vector line (1.41 µm at 6000 rpm to 3.35 µm at 1000 rpm). Adjudicated 2026-07-12: supersedes both the prior eyeball read (min 1.5 µm) and check's stale eyeball estimate (min 1.2 µm) — neither matched the exact vector-line crossing at 6000 rpm." }, "thicknessRangeNote": "Approximate achievable range for the ma-N 1420 grade specifically, read off the datasheet's spin-curve figure over 1000-6000 rpm (see spinCurves). The nominal/reference-condition thickness is 2.0 ± 0.1 µm at the standard 3000 rpm / 30 s spin (this exact value comes from a numeric table, not the figure).", "spinCurves": [ { "label": "ma-N 1420", "points": [ { "rpm": 1000, "um": 3.35 }, { "rpm": 2000, "um": 2.39 }, { "rpm": 3000, "um": 1.97 }, { "rpm": 4000, "um": 1.71 }, { "rpm": 5000, "um": 1.54 }, { "rpm": 6000, "um": 1.41 } ], "source": "re-extracted 2026-07-12, pixel-calibrated via PyMuPDF vector-path extraction (the chart is drawn as native PDF vector line segments, so the trace's true (x,y) coordinates were pulled directly from the page's vector drawing commands rather than estimated; axes calibrated from the tick-label text bounding boxes). Fig. 1 ('Spin curves of the ma-N 1400 series, 30 s spin time'), p.3 (numbered p.2 on the printed page) of micro resist technology's 'Processing guidelines — Negative Tone Photoresist Series ma-N 1400' datasheet (doc code ls.05.11.25.02). The curve is a continuous line with no point markers, so values are exact line-crossings read only at the labelled gridlines (every 1000 rpm from 1000 to 6000) — never interpolated between them or extrapolated past the line's plotted range. The 3000 rpm point (1.97 µm) is cross-confirmed by the numeric table on p.1 ('Physical properties of the resist solution': Film Thickness 2.0 ± 0.1 µm, footnoted 'spin coated at 3000 rpm for 30 s'). Supersedes the earlier eyeball read (which also guessed intermediate 1500/2500 rpm points not distinguishable on the chart).", "figureRead": true } ], "spinNotes": "Substrates should be free of impurities/moisture, baked at 200°C and cooled immediately before coating (or O2/ozone plasma cleaned); HMDS is advised for adhesion to Si/SiO2. No dispense-volume/acceleration/edge-bead protocol is published for ma-N 1400 (unlike some other resist datasheets); equipment used to generate the datasheet's own data: Convac or Suss RC5 spin coater without cover, 3000 rpm / 30 s reference condition.", "adhesion": { "hmds": true, "notes": "HMDS recommended as an adhesion promoter for Si and SiO2 substrates. Substrates should be free of impurities and moisture, baked at 200°C and cooled to room temperature immediately before coating; oxygen or ozone plasma cleaning is an accepted alternative." }, "rehydration": null, "softbake": { "temp_c": 100, "time_s": 120, "method": "hotplate", "notes": "100°C / 120 s is the standard hotplate prebake for the 2.0 µm ma-N 1420 film (thinner grades in the same series use shorter times at the same 100°C: ma-N 1405/1407 = 60 s, ma-N 1410 = 90 s). Oven alternative: 100–105°C for 15–30 min. Prebake temperature may be raised (max 160°C) or time extended to further increase etch resistance/thermal stability, at the cost of a longer required develop time. Recommended prebake temperature varies by substrate: 100–120°C (Si), 120°C (Au), 120°C (Si3N4), 100°C (GaAsP/GaAs).", "source": "Processing conditions - STANDARD PROCESS table + 'Prebake' section, p.2 and p.3, micro resist technology ma-N 1400 datasheet ls.05.11.25.02" }, "exposureDose": { "doses": [ { "wavelength_nm": null, "value_mJcm2": 550, "source": "Processing conditions - STANDARD PROCESS table, p.2, micro resist technology ma-N 1400 datasheet ls.05.11.25.02" } ], "datasheetBasis": "Broadband exposure with the dose measured/specified at λ = 365 nm (i-line); footnote 1 of the Processing Conditions table explicitly states 'broadband exposure, intensity measured at λ=365 nm'.", "_note": "Datasheet states 550 ± 30 mJ/cm² for ma-N 1420 in the standard process (2.0 µm film), explicitly attributed to 365 nm. The resist's spectral sensitivity spans 300–410 nm (broadband/i-line/h-line all usable per the manufacturer's overview literature) and Fig. 3 shows absorption at both i-line (365) and h-line (405), but no separate 405 nm dose number is ever published — at405_mJcm2 is left null rather than derived from the 365 nm value or the absorption curve." }, "peb": { "temp_c": null, "time_s": null, "notes": "No post-exposure bake step is specified anywhere in the datasheet. Consistent with a classical bisazide-crosslinking negative resist, which crosslinks directly on exposure rather than via an acid-catalyzed post-bake reaction (as in chemically amplified resists).", "source": null }, "floodExposure": { "dose_mJcm2": null, "notes": "For lift-off/PVD stabilization, an optional deep-UV flood exposure of the developed pattern (200–300 nm) at 'twice to fivefold' the 365 nm patterning dose is recommended — i.e. roughly 2–5× 550 = 1100–2750 mJ/cm² for the standard ma-N 1420 process, but the datasheet only publishes the multiplier, not an absolute number, so dose_mJcm2 is left null. Applied after development and before PVD metal deposition to increase thermal/dissolution stability of the resist pattern during sputtering.", "source": "'Processing conditions - LIFT-OFF' section, p.4, micro resist technology ma-N 1400 datasheet ls.05.11.25.02" }, "develop": { "developer": "ma-D 533/S", "dilution": "ready-to-use (undiluted)", "time_s": 60, "method": "immersion", "rinse": "DI water", "source": "Processing conditions - STANDARD PROCESS table, p.2, micro resist technology ma-N 1400 datasheet ls.05.11.25.02. Development time for ma-N 1420 is 60 ± 10 s; developer temperature should be 20–25°C." }, "hardbake": { "temp_c": 100, "time_s": 1800, "notes": "Optional. Oven bake at 100°C for approximately 30 min to further increase etch resistance and thermal stability of developed patterns. A temperature ramp is recommended to reduce pattern reflow.", "source": "'Hardbake (optional)' section, p.4, micro resist technology ma-N 1400 datasheet ls.05.11.25.02" }, "descum": null, "applications": [ "etch-mask", "lift-off" ], "etchResistance": "High wet and dry etch resistance (a headline characteristic of the ma-N 1400 series). Can be further increased by raising prebake temperature (up to 160°C max) or prebake time, or by an optional post-develop hardbake (100°C, ~30 min); developing time increases correspondingly.", "liftoffSuitable": true, "platingSuitable": null, "stripper": "mr-Rem 660 (solvent-based) or ma-R 404/S (strongly alkaline) ready-to-use removers recommended; acetone, N-methylpyrrolidone (NMP), or O2 plasma also give residue-free removal. For lift-off specifically after PVD: mr-Rem 660 or NMP at 40–60°C with ultrasonic assist (acetone with ultrasonics also recommended).", "storage": "Store at 18–25°C; do not refrigerate. Store resist and unprocessed films under yellow light, bottle closed when not in use. Shelf life 6 months from date of manufacture under these conditions. Best patterning results at 20–25°C ambient and 40–46% RH.", "notes": "ma-N 1420 gives a nominal 2.0 ± 0.1 µm film at the reference 3000 rpm / 30 s spin condition. Standard processing (100°C/120s prebake, 550 mJ/cm² dose, 60s develop) yields near-vertical sidewalls. Undercut (lift-off) profiles are produced primarily by INCREASING develop time (the manufacturer's preferred lever) and/or reducing exposure dose, while holding prebake conditions at standard values — the datasheet's own worked example (2.0 µm film, 100°C/120s prebake, 550 mJ/cm² dose) shows undercut of 0.6/0.8/1.7/2.1 µm for ma-D 533/S develop times of 65/80/100/120 s respectively. Raising prebake temperature (up to 160°C) and/or extending prebake time reduces undercut for a given develop time and improves thermal stability for sputtering, but increases the develop time needed. For clean PVD lift-off, resist thickness should be 1.5–2× the metal deposition thickness. Exposure dose must be de-rated for reflective/absorbing substrates (e.g. ~0.5× the Si dose on Au).", "developerFamily": "tmah", "references": [ { "type": "paper", "title": "Combined ultraviolet- and electron-beam lithography with Micro-Resist-Technology GmbH ma-N1400 resist", "authors": "Thoen et al.", "journal": "Journal of Vacuum Science & Technology B", "year": 2022, "doi": "10.1116/6.0001918", "url": "https://doi.org/10.1116/6.0001918", "accessedDate": "2026-07-12", "summary": "one ma-N 1400 layer exposed by UV and 100 kV e-beam for an on-chip THz spectrometer" } ], "provenance": { "datasheetUrl": "https://litho.nano.cnr.it/wp-content/datasheets/man_1400.pdf", "datasheetVersionOrDate": "micro resist technology 'Processing guidelines — Negative Tone Photoresist Series ma-N 1400', document code ls.05.11.25.02 (no separate calendar date printed on the document; also mirrored as vh_man_1400_en_05112502_ls.pdf)", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://www.microresist.de/en/produkt/ma-d-533-s/", "what": "used only to confirm that the ma-D 533/S developer is TMAH-based (the primary ma-N 1400 datasheet only says 'aqueous alkaline development' without naming the base) — fills the developerFamily field, not a numeric process source" } ] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "photoimageable": true, "humanVerified": false }, { "slug": "ma-p-1215", "name": "ma-P 1215", "manufacturer": "micro resist technology GmbH", "productLine": "ma-P 1200 series", "aliases": [ "ma-P1215" ], "tone": "positive", "chemistry": "dnq-novolak", "_chemistryNote": "Document does not state the resin/PAC chemistry (e.g. novolak, DNQ) anywhere in the text; not inferred from generic knowledge of positive g/i-line resists.", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "This document (standard ma-P 1200 series flyer, grades 1205/1210/1215/1225/1240/1275 + 1275 HV) never mentions grayscale or 3D lithography. Grayscale capability is associated with a separately-branded 'ma-P 1200G' sheet (see the ma-p-1275g recipe already in this library, extracted from a distributor mirror of that different document) — this ma-P.pdf does not confirm or deny any relationship between the two lines.", "status": "active", "successorSlug": null, "summary": "Positive-tone resist from micro resist technology's ma-P 1200 series; ma-P 1215 is the 1.5 µm nominal-thickness grade, coated at 3000 rpm for 30 s, part of a six-grade family spanning 0.3–40 µm.", "thicknessRange": { "min_um": 1.5, "max_um": 1.5, "basis": "stated", "source": "stated — the family table (p.1) prints one film-thickness value per grade at one shared spin condition (3000 rpm / 30 s); for ma-P 1215 that value is 1.5 µm. No achievable min–max spread is stated for this specific grade, and the multi-grade 'Spin curves, 30 s spin time' figure on the same page was not read for individual points (see spinNotes) — 1.5 µm is therefore recorded as a single documented anchor, not a genuine range." }, "spinCurves": [], "spinNotes": "The 'Positive Photoresist Series' table (p.1) lists ma-P 1205/1210/1215/1225/1240/1275 side by side, each with exactly one film-thickness value, all coated at the SAME condition: 3000 rpm, 30 s spin time. For ma-P 1215 that is 1.5 µm — a single anchor point, not a spin-speed sweep, so per protocol no curve is published (a one-point series is not a curve). A separate figure titled 'ma-P 1200 series — Spin curves, 30 s spin time' (p.1) plots film thickness vs. spin speed (1000–6000 rpm) for all six grades on one chart with a color-coded legend (ma-P 1275/1240/1225/1215/1210/1205, in that order, thickest to thinnest). This extraction did NOT attempt to read numeric points off that figure: the PDF's text layer returns only the axis tick labels (0–16 µm, 1000–6000 rpm) and the legend strings, not the underlying vector-curve geometry, so any per-grade value would be an unverifiable eyeball guess on a multi-grade chart — exactly the failure mode flagged by the SU-8 2050 (20–42% error) and combined-trace LOR 3A precedents. A human with the rendered PDF open could read this figure directly for a tighter thicknessRange. No accel, dispense volume, edge-bead removal, or rehydration guidance is published anywhere in this document. The manufacturer's own feature list states 'No post exposure bake' for the whole ma-P 1200 series.", "adhesion": { "hmds": null, "notes": "Not mentioned in this document." }, "rehydration": null, "softbake": null, "exposureDose": { "doses": [ { "wavelength_nm": null, "value_mJcm2": 45, "source": "table \"Positive Photoresist Series\", p.1 of ma-P.pdf, ma-P 1215 column" } ], "datasheetBasis": "Dose @ 365 nm (broadband exposure)", "_note": "Table column header reads 'Dose @ 365 nm (broadband exposure)' — this is a broadband-lamp dose merely metered/reported at the 365 nm line (the same pattern as the documented 'broadband exposure, intensity measured at λ=365 nm' trap), not an i-line-only dose. Routed to value_mJcm2, not at365_mJcm2, per the wavelength-routing rule. at405_mJcm2 is not published anywhere in this document." }, "peb": null, "floodExposure": null, "develop": { "developer": null, "dilution": null, "time_s": null, "method": null, "rinse": null, "source": "bullet \"Aqueous alkaline development\", p.1 of ma-P.pdf", "_note": "Document states development is aqueous-alkaline as a series-wide feature but names no specific developer product, dilution, method, or time for any grade." }, "hardbake": null, "descum": null, "applications": [ "etch-mask", "electroplating-molding" ], "etchResistance": "Manufacturer states \"Outstanding pattern stability in wet etch processes and acid and alkaline plating baths\" and \"Highly stable in dry etch processes e.g. CHF3, CF4, SF[x]\" for the ma-P 1200 series (p.1).", "liftoffSuitable": null, "platingSuitable": true, "stripper": null, "storage": null, "notes": "ma-P 1215 is the 1.5 µm nominal-thickness member of micro resist technology's ma-P 1200 positive-tone series, a six-grade family (1205 through 1275) spanning 0.3–40 µm, all coated at the same 3000 rpm / 30 s condition per the family table. The manufacturer calls out \"no post exposure bake\" as a series-wide feature, which simplifies the process relative to resists that require a critical PEB step. Development is aqueous-alkaline, but this flyer names no specific developer product, dilution, or time — only a general process-flow diagram (coat, expose, develop) is shown. A multi-grade spin-speed curve (1000–6000 rpm, 30 s) is published for the whole family on one chart, but this extraction did not read individual trace values off it (see spinNotes) — a human QC pass with the rendered figure could tighten thicknessRange. The document also markets ma-P 1275 / ma-P 1275 HV as separate high-viscosity grades for electroplating molds up to 60 µm, but does not mention any 'ma-P 1275G' or grayscale variant. Chemistry classified as dnq-novolak from kayakuam.com's statement that ma-P 1200 is a 'positive tone DNQ/novolac based resist series' (chemistry classified 2026-07-12 from manufacturer SDS/TDS).", "developerFamily": null, "provenance": { "datasheetUrl": "https://www.nanophys.kth.se/nanolab/resists/mrt-pdfs/ma-P.pdf", "datasheetVersionOrDate": "20 June 2008", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://kayakuam.com/products/ma-p-1200/", "what": "kayakuam.com states 'ma-P 1200 is a positive tone DNQ/novolac based resist series'; the basis for classifying ma-P 1215 as dnq-novolak." } ] }, "_provenanceNote": "Hosted by KTH Nanolab (nanophys.kth.se) as a mirror of micro resist technology GmbH's own product-information flyer (micro resist technology's own site serves this TDS behind a non-guessable ?jet_download= URL, so the KTH mirror is the retrievable copy of the manufacturer's document). The document is branded 'micro resist technology GmbH' throughout, not a university-authored SOP.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "ma-p-1275g", "name": "ma-P 1275G", "manufacturer": "micro resist technology GmbH", "productLine": "ma-P 1200G series", "aliases": [ "maP 1275G", "ma-P1275G" ], "tone": "positive", "chemistry": "dnq-novolak", "_chemistryNote": "Document ('ma-P 1200G — Positive Greyscale Photoresist Series' product-information sheet) never names a chemistry mechanism (no 'novolak', 'DNQ', or 'diazonaphthoquinone' anywhere in the text) — only 'Positive tone photoresist series specifically designed for the requirements of greyscale lithography'. Left null rather than assumed from ma-P's typical DNQ-novolak family membership, which is not stated here.", "photoimageable": true, "grayscaleSuitable": true, "grayscaleNote": "Document markets the series explicitly for greyscale lithography: 'Positive tone photoresist series specifically designed for the requirements of greyscale lithography... Reduced contrast... Film thickness up to 60 µm and higher... 50 - 60 µm depth range of the patterns possible in greyscale lithography.' The ma-P 1275G example figure shows '~53 µm pattern depth in ~58 µm thick ma-P 1275G'. Also states 'An application in standard binary lithography is also possible.'", "status": "active", "successorSlug": null, "summary": "ma-P 1275G is the thickest grade in micro resist technology's ma-P 1200G positive-tone greyscale photoresist series, coating from roughly 9.3 to 60 µm and used for 3D micro-optic, MEMS/MOEMS and display structures written by dose-modulated laser direct writing or a greyscale mask.", "thicknessRange": { "min_um": 9.3, "max_um": 60, "basis": "stated", "source": "stated — per-grade 'Film thickness' table for ma-P 1275G (p.1) lists four achievable coatings (9.3, 15, 30, 60 µm), each reached with a different spin speed AND a different spin time (30 s, 30 s, 60 s, 4 s respectively)." }, "spinCurves": [], "spinNotes": "No spin curve is published here. The per-grade film-thickness table for ma-P 1275G gives four (rpm, time, thickness) triples — 3000 rpm/30 s→9.3 µm, 1500 rpm/30 s→15 µm, 500 rpm/60 s→30 µm, 1000 rpm/4 s→60 µm — but spin TIME differs across the four points (30/30/60/4 s), so they are four discrete target-thickness recipes, not points on one continuous rpm-only spin curve; sorted by rpm alone the thickness rises from 500→1000 rpm (30→60 µm) before falling again at 1500 and 3000 rpm, which would fail a monotonic-curve sanity check and misrepresent the data as a single spin curve. A separate figure ('Film thickness [µm] vs Spin speed [rpm]', 1000–6000 rpm, 0–25 µm axis) plots ma-P 1275G, 1225G and 1215G together at a FIXED 30 s spin time, which is a genuine spin curve for this grade, but that 30 s curve's y-axis (max 25 µm) does not reach the 60 µm value from the table, so it evidently covers a different/narrower thickness regime than the table's 4-14 s recipes. Given the risk of misreading a multi-grade figure (flagged explicitly for this project) and the confound between the table and the figure, no curve is published; a human QC reviewer should read the figure directly against the source PDF, p.1.", "adhesion": { "hmds": null, "notes": "Not addressed in this product-information sheet." }, "rehydration": null, "softbake": null, "exposureDose": { "doses": [], "datasheetBasis": "'Spectral sensitivity 350…450 nm' (Characteristics, p.1); example SEM captions cite tool wavelengths of 390 nm (µPG301), 405 nm (DWL66+) and 355 nm (VPG400), but these are exposure-TOOL wavelengths for demo patterns, not a published dose.", "_note": "No dose value in mJ/cm² is published anywhere in this document for any ma-P 1200G grade — only a spectral sensitivity range and an absorption-coefficient-vs-wavelength curve (unexposed/exposed). All dose fields left null rather than estimated." }, "peb": null, "floodExposure": null, "develop": { "developer": null, "dilution": null, "time_s": null, "method": null, "rinse": null, "source": "Characteristics, p.1", "_note": "Document states only 'Aqueous alkaline development, for greyscale lithography with TMAH based developers, for standard binary lithography also with metal ion bearing developers' — no specific developer product name, dilution, time or method is given." }, "hardbake": null, "descum": null, "applications": [ "grayscale-3d", "mems-structural", "electroplating-molding", "etch-mask" ], "etchResistance": "\"Suitable for dry etch processes e.g. with CHF3, CF4, SF6\" (Characteristics, p.1).", "liftoffSuitable": null, "platingSuitable": true, "stripper": null, "storage": null, "notes": "ma-P 1275G is the thickest grade of micro resist technology's ma-P 1200G positive-tone greyscale series, rated by the manufacturer for coatings from roughly 9.3 to 60 µm depending on spin speed and time. Greyscale (analog) lithography needs a resist whose exposed thickness responds smoothly to dose so a continuous 3D relief can be written by a modulated laser or a grey-level mask; the datasheet markets the series as having 'reduced contrast' and states 50-60 µm pattern depths are achievable, but it does not publish a dose-vs-remaining-thickness (contrast) curve, so the actual linearity of that dose response is not characterized here and must be measured on-tool. The four film-thickness/spin-speed pairs the datasheet lists for this grade each use a different spin time (30 s, 30 s, 60 s and 4 s), so they describe four discrete target-thickness recipes rather than points on one continuous spin curve — see spinNotes for why no curve is published. No softbake, PEB, hardbake, developer product name, or dose values are given in this flyer-style product-information sheet; a full technical datasheet would be needed to fill those in. Chemistry classified as dnq-novolak from microresist.de's ma-P 1200G series page, the greyscale variant of the same DNQ/novolak ma-P family (chemistry classified 2026-07-12 from manufacturer SDS/TDS).", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://www.ostech.co.jp/wp/wp-content/uploads/2020/11/map1200g_pi_1.pdf", "datasheetVersionOrDate": "ls.18.02.06", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://microresist.de/en/produkt/ma-p-1200g-series/", "what": "microresist.de's ma-P 1200G series page (greyscale variant of the same DNQ/novolak family as ma-P 1200); the basis for classifying ma-P 1275G as dnq-novolak." } ] }, "_provenanceNote": "Document is micro resist technology GmbH's own 'ma-P 1200G — Positive Greyscale Photoresist Series' product-information sheet (manufacturer letterhead, Köpenicker Straße 325, 12555 Berlin, Germany), retrieved via an OSTECH (Japanese micro resist technology distributor) mirror URL rather than microresist.de/microresist.com directly. Content and formatting match a manufacturer-authored product-information sheet (single page, 'ls.18.02.06' revision code, standard microresist layout), not an OSTECH-authored document — OSTECH's own branding does not appear anywhere in the retrieved PDF.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "spr220-3", "name": "Megaposit SPR 220-3.0", "manufacturer": "Rohm and Haas Electronic Materials", "productLine": "MEGAPOSIT SPR 220 series", "aliases": [ "SPR220-3.0", "MEGAPOSIT SPR220-3.0", "Shipley SPR220-3.0" ], "tone": null, "chemistry": "dnq-novolak", "_chemistryNote": "The datasheet does not state the resin/photoactive-compound chemistry anywhere in its text (no mention of novolak, DNQ, epoxy, or acrylate) — only the coating solvents are listed (ethyl lactate, anisole, n-amyl acetate, under Handling Precautions). Tone (positive/negative) is likewise never stated explicitly in this document. Table 5 publishes Dill A/B optical parameters, a model conventionally associated with positive-tone DNQ/PAC resists, but that is an inference from the choice of characterization method, not a textual statement — left unresolved for a human reviewer rather than assumed. (chemistry classified 2026-07-12: re-checked the full archived 4-page PDF — no chemistry-mechanism wording anywhere; null confirmed, not a gap.)", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed by the datasheet. The document positions SPR220 for dense lines/spaces resolution, thick-film etch masking, and electroplating molds — no mention of grayscale or partial-exposure profiling.", "status": "active", "successorSlug": null, "summary": "SPR220-3.0 is the 3 µm-nominal grade of Rohm and Haas's MEGAPOSIT SPR 220 series, a general-purpose, broadband/g-line/i-line multiwavelength photoresist covering roughly 2-5 µm depending on spin speed, used for MEMS, plating and etch-mask applications.", "thicknessRange": { "min_um": 2.28, "max_um": 4.62, "basis": "curve-span", "source": "curve-span: the datasheet does not state an achievable thickness range for the SPR220-3.0 grade specifically (only a family-wide '1-30 µm' range spanning all five grades in the Description — that family-wide figure is not this grade's range and must not be used here). The range here is the min/max of the SPR220-3.0 curve in Figure 3 (4-inch wafer spin chart) between 6000 rpm and 1500 rpm, pixel-calibrated 2026-07-12 (10 points at 500 rpm intervals, self-consistency-checked against a t∝1/√rpm fit within 1-3%): 4.62 µm at 1,500 rpm to 2.28 µm at 6,000 rpm. Adjudicated 2026-07-12: supersedes the earlier 6-point eyeball read (2.2-4.65 µm) and check's stale family-wide 1-30 µm grab. The '3.0' in the product name denotes its nominal reference thickness (used as the thickness label in Table 2 and Table 6), not a range." }, "spinCurves": [ { "label": "SPR220-3.0", "points": [ { "rpm": 1500, "um": 4.62 }, { "rpm": 2000, "um": 4 }, { "rpm": 2500, "um": 3.59 }, { "rpm": 3000, "um": 3.23 }, { "rpm": 3500, "um": 3.02 }, { "rpm": 4000, "um": 2.81 }, { "rpm": 4500, "um": 2.66 }, { "rpm": 5000, "um": 2.5 }, { "rpm": 5500, "um": 2.41 }, { "rpm": 6000, "um": 2.28 } ], "source": "re-extracted 2026-07-12, pixel-calibrated (rendered-pixel method — Figure 3 is an embedded raster image, not vector data); Figure 3 (p.2), \"Spin Speed Curves on 4\\\"\", of Rohm and Haas \"MEGAPOSIT SPR220 Series Photoresists\" datasheet (ME04N097, Rev. 2, Sept. 2004; UC Davis CNM2 mirror). Chart plots five curves (SPR220-7.0/circle, 4.5/square, 3.0/triangle, 1.5/diamond, 1.2/inverted-triangle), each with 10 markers from 1,500-6,000 rpm. SPR220-3.0 identified via its filled-triangle legend marker (3rd of 5 traces), cross-checked against the series-naming convention where each grade's value near 3000 rpm approximates its nominal designation (this trace reads 3.23 µm at 3000 rpm ≈ 'SPR220-3.0'). Page rendered at 10x zoom, axis calibrated from gridline pixel positions, all 10 marker centers found via per-column dark-run detection; self-consistency confirmed via a t∝1/√rpm fit (predicted-vs-read within ~1-3% across all 10 points). Supersedes the earlier 6-point eyeball read.", "figureRead": true } ], "spinNotes": "Figure 3 is measured on 4-inch substrates. Nominal film thickness may vary slightly with process, equipment, and ambient conditions, per the datasheet's own caveat under \"Coat\", p.2. Coat uniformity is separately reported for the SPR220-7.0 grade only (7.31 µm, standard deviation 0.036 µm across 33 points) — no uniformity figure is published for SPR220-3.0 specifically.", "adhesion": { "hmds": true, "notes": "\"A hexamethyldisilizane (HMDS)-based MICROPOSIT primer is recommended to promote adhesion with substrates that require such treatment. Vacuum vapor priming at 120°C for 30 seconds with concentrated HMDS is recommended.\" Source: \"Substrate\", p.2." }, "rehydration": "Not applicable at SPR220-3.0's nominal 3.0 µm reference thickness. The datasheet's exposure-to-PEB moisture-diffusion hold requirement applies only to \"films above 4 µm\" (minimum 35-minute hold) and films \">12 µm\" (minimum 120-minute hold) — both thicker than this grade's 3.0 µm nominal coat, so no hold time is called for at the reference process point. (See the SPR220-7.0 entry, where this hold is required.) (Source: \"Post-Exposure Bake\" section, p.3)", "softbake": { "temp_c": 115, "time_s": 90, "method": "hotplate", "notes": "\"The recommended softbake process for SPR220 for films up to 4.0 µm is 115°C for 90 seconds on a contact hotplate.\" SPR220-3.0's 3.0 µm nominal thickness falls within this \\u201cup to 4.0 µm\\u201d bin, so this is a direct (non-ranged) match, not a picked bin midpoint. Table 1 independently confirms the same value for the 1.1-4.0 µm thickness class: \"115°C/90 sec. Contact Hotplate.\"", "source": "\"Softbake\" section, p.2, and Table 1 \"Recommended Process Conditions\" (1.1-4.0 µm column), p.1" }, "exposureDose": { "doses": [ { "wavelength_nm": 365, "value_mJcm2": 310, "source": "Table 2 \"Photospeed and Linearity of Dense Lines/Spaces at Various Thicknesses\", p.1" } ], "datasheetBasis": "SPR220 is described as \"Broadband, g-Line and i-Line capable\" (Advantages, p.1). Table 2 gives photospeed at both g-Line (436 nm) and i-Line (365 nm) for several dense-line/space film thicknesses; Table 1 specifies the reference exposure tool as an \"ASML PAS 5500/200 i-Line (0.48 NA, 0.50σ)\" stepper.", "_note": "at365_mJcm2 = 310 mJ/cm² is an exact-thickness match from Table 2's \"i-Line 3.0 µm\" row (0.90 µm linearity at that dose/thickness), matching SPR220-3.0's nominal 3.0 µm reference thickness exactly — not a bin midpoint. Table 2 also lists a g-Line dose at the same 3.0 µm thickness (320 mJ/cm², 0.90 µm linearity); not recorded in a scalar field here because this schema has no g-Line (436 nm) slot, only i-Line (365) and h-Line (405). General thick-film guidance on p.2 (\"For film thickness greater than 12 µm the resist is exposed to an energy dose between 700 and 1,300 mJ/cm² [...] measured using standard radiometer @ 365 nm wavelength\") does not apply to this 3.0 µm grade." }, "peb": { "temp_c": 115, "time_s": 90, "notes": "Table 1 gives a single PEB value spanning both thickness classes (1.1-4.0 µm and 4.0-10.0 µm): \"115°C/90 sec. Contact Hotplate.\" PEB runs at the same temperature as softbake, per the \"Post-Exposure Bake\" section text. Separately from the PEB bake itself, thicker films (>4 µm) require a moisture-diffusion hold BEFORE PEB — not applicable to this 3.0 µm grade; see the \"rehydration\" field.", "source": "Table 1 \"Recommended Process Conditions\", p.1, and \"Post-Exposure Bake\" section, p.3" }, "floodExposure": null, "develop": { "developer": "MF-24A (0.24N MIF — metal-ion-free TMAH developer)", "dilution": "0.24N (MF-24A is the 0.24N-normality developer in Rohm and Haas's MF line; SPR220 is \"optimized for 0.24N developers\", with 0.26N (MF-26A) offered as an alternative for thicker films or high-throughput processes)", "time_s": 60, "method": "puddle", "rinse": null, "source": "Table 6 \"Recommended Develop Conditions\" (3.0 µm FT column) and Table 1, p.1 and p.3", "_note": "Table 6's \"3.0 µm FT\" column is an exact-thickness match for SPR220-3.0's nominal reference thickness: MF-24A → 60 sec. single spray puddle (SP); MF-26A → 60 sec. SP; M452 (a metal-ion-bearing/MIB developer) → 3 min. immersion. MF-24A is recorded here as it is also Table 1's family-wide recommended developer (\"MF-24A @ 21°C, 60 sec. single spray puddle\"). No rinse step or rinse chemistry is specified anywhere in the document for the develop step (unlike the SU-8/KMPR datasheets, which both specify a DI-water or IPA rinse) — left null rather than assumed." }, "hardbake": null, "descum": null, "applications": [ "etch-mask", "electroplating-molding", "mems-structural" ], "etchResistance": "\"Excellent wet and dry etch adhesion\" (Advantages, p.1). Figure 8 (p.4) reports 100:1 etch selectivity in a Bosch DRIE process, demonstrated patterning 2.5-10 µm features to 200 µm deep and 5-20 µm features to 100 µm deep (these etch-depth examples are shown at the SPR220 series level, not attributed to a specific grade). A 1:5 HF wet-etch example (2 µm feature) is also shown on p.1.", "liftoffSuitable": null, "platingSuitable": true, "stripper": "MICROPOSIT REMOVER 1165, two-bath process, each bath at 80°C (176°F): the first bath removes the bulk of the photoresist, the second removes residual traces. Source: \"Photoresist Removal\", p.4.", "storage": "\"Recommended storage for SPR220 is in an upright position in a dry area at 40-60°F (4-15°F). Keep away from oxidizers, acids, and bases. Keep container sealed when not in use.\" (Quoted verbatim, including an apparent unit inconsistency in the source: 40-60°F does not correspond to 4-15°F — the parenthetical is likely a typo for °C in the original document, but is transcribed as printed rather than silently corrected.) Source: \"Storage\", p.4.", "notes": "SPR220-3.0 is the 3 µm-nominal member of the MEGAPOSIT SPR 220 line, a broadband resist usable at g-line (436 nm) or i-line (365 nm) that Rohm and Haas positions for thick-film MEMS, plating-mold, and DRIE-mask work rather than fine-pitch IC lithography. Unlike the thicker SPR220-7.0 grade, films at this 3.0 µm reference thickness stay under the datasheet's 4 µm threshold for the exposure-to-PEB rehydration hold, so processing is comparatively straightforward: softbake, PEB and (per Table 1) develop all run at essentially the textbook single-step conditions. The lineage of this resist family is worth noting for provenance purposes: Shipley Company originated the SPR line, was acquired into Rohm and Haas Electronic Materials (the entity printed on this 2004 datasheet), which was in turn acquired by Dow Chemical (2009), then spun into DuPont Electronics & Imaging, and the MEGAPOSIT/MICROPOSIT/MF brand family is now sold by Kayaku Advanced Materials — the same corporate family that owns MicroChem/KMPR and SU-8. Tone and base chemistry are not stated anywhere in this particular datasheet and are left unset rather than assumed from the product family's general reputation. Chemistry classified as dnq-novolak from the Dow MEGAPOSIT SPR220-3.0 MSDS composition table (cresol novolak resin + diazo photoactive compound) (chemistry classified 2026-07-12 from manufacturer SDS/TDS).", "developerFamily": "tmah", "references": [ { "type": "paper", "title": "A comparative study of different thick photoresists for MEMS applications", "authors": "Koukharenko et al.", "journal": "Journal of Materials Science: Materials in Electronics", "year": 2005, "doi": "10.1007/s10854-005-4977-2", "url": "https://doi.org/10.1007/s10854-005-4977-2", "accessedDate": "2026-07-12", "summary": "Benchmarks SPR220 against SU-8 and AZ 9260 as thick-film MEMS resists." } ], "provenance": { "datasheetUrl": "https://research.engineering.ucdavis.edu/cnm2/wp-content/uploads/sites/11/2013/05/SPR220_Data_Sheet.pdf", "datasheetVersionOrDate": "ME04N097, Rev. 2, September 2004 (printed on p.1 and in the footer)", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://asrc.gc.cuny.edu/wp-content/uploads/media/global-assets/Megaposit-SPR-220-3.0-Positive-Photoresist-MSDS.pdf", "what": "Dow MEGAPOSIT SPR220-3.0 MSDS states 'Cresol novolak resin 25.0-35.0%, Diazo Photoactive Compound 1.0-10.0%'; the basis for classifying SPR220-3.0 as dnq-novolak." } ] }, "_provenanceNote": "Fetched directly from the assigned URL (UC Davis CNM2 mirror) and confirmed to be the correct document: Rohm and Haas Electronic Materials' own \"MEGAPOSIT SPR220 Series Photoresists\" datasheet (ME04N097, Rev. 2, September 2004). This is a university (UC Davis) mirror of the manufacturer's own document, not a university-authored SOP.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "spr220-7", "name": "Megaposit SPR 220-7.0", "manufacturer": "Rohm and Haas Electronic Materials", "productLine": "MEGAPOSIT SPR 220 series", "aliases": [ "SPR220-7.0", "MEGAPOSIT SPR220-7.0", "Shipley SPR220-7.0" ], "tone": null, "chemistry": "dnq-novolak", "_chemistryNote": "The datasheet does not state the resin/photoactive-compound chemistry anywhere in its text (no mention of novolak, DNQ, epoxy, or acrylate) — only the coating solvents are listed (ethyl lactate, anisole, n-amyl acetate, under Handling Precautions). Tone (positive/negative) is likewise never stated explicitly in this document. Table 5 publishes Dill A/B optical parameters, a model conventionally associated with positive-tone DNQ/PAC resists, but that is an inference from the choice of characterization method, not a textual statement — left unresolved for a human reviewer rather than assumed.", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed by the datasheet. The document positions SPR220 for dense lines/spaces resolution, thick-film etch masking, and electroplating molds — no mention of grayscale or partial-exposure profiling.", "status": "active", "successorSlug": null, "summary": "SPR220-7.0 is the 7 µm-nominal grade of Rohm and Haas's MEGAPOSIT SPR 220 series, a general-purpose, broadband/g-line/i-line multiwavelength photoresist that spin-coats from roughly 5 µm up to 30-50+ µm depending on speed, used for thick-film MEMS, plating and Bosch-etch-mask applications.", "thicknessRange": { "min_um": 5.45, "max_um": 52, "basis": "curve-span", "source": "curve-span: the datasheet does not state an achievable thickness range for the SPR220-7.0 grade specifically (only a family-wide '1-30 µm' range spanning all five grades in the Description, though this grade's own curve exceeds that at low rpm). The range combines the two SPR220-7.0-specific curves published: the thin end (5.45 µm) is from Figure 3 (4-inch wafer) at 6000 rpm; the thick end (52 µm) is from Figure 4 (8-inch wafer, grade-dedicated chart) at 150 rpm. The '7.0' in the product name denotes its nominal reference thickness (used as the thickness label in Table 2 and Table 6), not this full spin-achievable range." }, "spinCurves": [ { "label": "SPR220-7.0 (8\" wafer, Figure 4)", "points": [ { "rpm": 150, "um": 52 }, { "rpm": 250, "um": 39.3 }, { "rpm": 500, "um": 22 }, { "rpm": 750, "um": 16.5 }, { "rpm": 1000, "um": 13.8 }, { "rpm": 1500, "um": 10.5 }, { "rpm": 2000, "um": 9.3 }, { "rpm": 2500, "um": 8.3 } ], "source": "read from figure 4 \"Spin Speed Curve, SPR220-7.0 on 8\\\"\", p.2 of Rohm and Haas \"MEGAPOSIT SPR220 Series Photoresists\" datasheet (ME04N097, Rev. 2, Sept. 2004; UC Davis CNM2 mirror). This figure plots a single curve for SPR220-7.0 only, so there is no grade-identification ambiguity. Read at high resolution (4x page render, cropped); all 8 published markers reported. Cross-check: the text states \"a 375 RPM spin will yield a film thickness of approximately 30 µm,\" which interpolates consistently between this curve's 250 rpm (39.3 µm) and 500 rpm (22 µm) points, supporting the accuracy of this figure read.", "figureRead": true }, { "label": "SPR220-7.0 (4\" wafer, Figure 3)", "points": [ { "rpm": 1500, "um": 10.1 }, { "rpm": 2000, "um": 9 }, { "rpm": 3000, "um": 7.5 }, { "rpm": 4000, "um": 6.55 }, { "rpm": 5000, "um": 5.9 }, { "rpm": 6000, "um": 5.45 } ], "source": "read from figure 3 \"Spin Speed Curves on 4\\\"\", p.2 of the same datasheet. Chart plots five curves; SPR220-7.0 identified as the filled-circle series, the topmost (thickest-film) curve, consistent with legend position (listed first) and its numeric name being the highest-viscosity grade shown. Read at high resolution; 6 of 10 available markers reported (1500, 2000, 3000, 4000, 5000, 6000 rpm). Cross-check against Figure 4 at overlapping speeds: 1500 rpm reads 10.1 µm here vs. 10.5 µm on the 8-inch chart, and 2000 rpm reads 9.0 µm here vs. 9.3 µm on the 8-inch chart — the two independently-read curves agree within ~4%, supporting both readings.", "figureRead": true } ], "spinNotes": "Figure 3 (4-inch wafer) and Figure 4 (8-inch/200 mm wafer, SPR220-7.0 only) are separate empirical curves for the same grade; both are reported above rather than merged, since they come from different substrates/tool runs. Coat uniformity is reported at 7.31 µm: standard deviation 0.036 µm across 33 points (source figure/wafer size for this uniformity measurement is not stated). Nominal film thickness may vary slightly with process, equipment, and ambient conditions, per the datasheet's own caveat under \"Coat\", p.2.", "adhesion": { "hmds": true, "notes": "\"A hexamethyldisilizane (HMDS)-based MICROPOSIT primer is recommended to promote adhesion with substrates that require such treatment. Vacuum vapor priming at 120°C for 30 seconds with concentrated HMDS is recommended.\" Source: \"Substrate\", p.2." }, "rehydration": "\"With thicker films (above 4 µm), a hold time is used between exposure and PEB to allow water (which is necessary to complete the photo-reaction) to diffuse back into the photoresist film. Thick films should use a minimum hold time of 35 minutes.\" SPR220-7.0's 7.0 µm nominal reference thickness is above the 4 µm threshold, so this minimum 35-minute ambient hold between exposure and PEB applies. (The document's 12 µm threshold for a 120-minute hold does not apply at the 7.0 µm nominal thickness, though it would for a thicker coat of this same grade at low rpm.) This is a genuinely easy step to omit by mistake since it sits between two other steps (expose, then PEB) rather than being a bake step itself. (Source: \"Post-Exposure Bake\" section, p.3)", "softbake": { "temp_c": 115, "time_s": 90, "method": "hotplate", "notes": "\"For films greater than 4.0 µm, use a 30 second ramp in temperature (step-down to hotplate) to 115°C and hold for a minimum of 90 seconds.\" SPR220-7.0's 7.0 µm nominal thickness falls in this bin (a direct match, not a picked bin midpoint) — a 30-second temperature step-down ramp precedes the 90-second hold at 115°C. (Films >12 µm instead require a 300-second minimum hold — not applicable at the 7.0 µm nominal thickness.) Table 1 independently confirms for the 4.0-10.0 µm thickness class: \"30 sec. step-down to 115°C/90 sec. Contact Hotplate.\"", "source": "\"Softbake\" section, p.2, and Table 1 \"Recommended Process Conditions\" (4.0-10.0 µm column), p.1" }, "exposureDose": { "doses": [], "datasheetBasis": "SPR220 is described as \"Broadband, g-Line and i-Line capable\" (Advantages, p.1). Table 2 gives photospeed at both g-Line (436 nm) and i-Line (365 nm) for several dense-line/space film thicknesses; Table 1 specifies the reference exposure tool as an \"ASML PAS 5500/200 i-Line (0.48 NA, 0.50σ)\" stepper.", "_note": "No exact-thickness i-Line dose is published for a 7.0 µm film — Table 2's i-Line rows stop at 5.0 µm (380 mJ/cm²). The closest published values are: g-Line 7.0 µm → 470 mJ/cm² (Table 2 — an exact thickness match, but g-Line/436 nm, a wavelength this schema has no field for) and i-Line 5.0 µm → 380 mJ/cm² (Table 2 — an exact wavelength match, but not the exact thickness). Neither is populated into at365/at405/value_mJcm2 to avoid misrepresenting either the wavelength or the thickness basis. General thick-film guidance (p.2): for films >12 µm, exposure dose is 700-1,300 mJ/cm² measured @365 nm using a broadband 350-400 nm source — informative context but not a 7.0 µm-specific number, and the 7.0 µm nominal thickness is below that 12 µm threshold anyway. Separately, the Advantages-bullet claim \"Fast photospeed: 210 mJ/cm2 for 1.1 µm lines/spaces @ 4.0 µm film thickness (i-Line)\" does not match any row in Table 2 (Table 2's 210 mJ/cm² entry is g-Line at 1.2 µm, not i-Line at 4.0 µm) — an apparent inconsistency between the marketing summary and the technical table in the source document itself; not used for any grade." }, "peb": { "temp_c": 115, "time_s": 90, "notes": "Table 1 gives a single PEB value spanning both thickness classes (1.1-4.0 µm and 4.0-10.0 µm): \"115°C/90 sec. Contact Hotplate.\" PEB runs at the same temperature as softbake, per the \"Post-Exposure Bake\" section text. This 90-second figure is the PEB bake itself — it is separate from and follows the mandatory 35-minute exposure-to-PEB moisture-diffusion hold required for this grade; see the \"rehydration\" field.", "source": "Table 1 \"Recommended Process Conditions\", p.1, and \"Post-Exposure Bake\" section, p.3" }, "floodExposure": null, "develop": { "developer": "MF-24A (0.24N MIF — metal-ion-free TMAH developer)", "dilution": "0.24N (MF-24A is the 0.24N-normality developer in Rohm and Haas's MF line; SPR220 is \"optimized for 0.24N developers\", with 0.26N (MF-26A) offered as an alternative for thicker films or high-throughput processes)", "time_s": 120, "method": "puddle", "rinse": null, "source": "Table 6 \"Recommended Develop Conditions\" (7.0 µm FT column), p.3", "_note": "Table 6's \"7.0 µm FT\" column is an exact-thickness match for SPR220-7.0's nominal reference thickness: MF-24A → 60/60 sec. double spray puddle (DP) — two sequential 60-second puddles, recorded here as 120 s total; MF-26A → 60/60 sec. DP; M452 (MIB developer) → 3 min. immersion; M453 (MIB developer) → 3 min. immersion. MF-24A is recorded here as it is also Table 1's family-wide recommended developer. For thick-film application ≥12 µm using MF-26A, the text separately notes overall development time behaves similarly to the thin-film time cited in Figure 7 — not directly relevant to this 7.0 µm grade. No rinse step or rinse chemistry is specified anywhere in the document for the develop step (unlike the SU-8/KMPR datasheets, which both specify a DI-water or IPA rinse) — left null rather than assumed." }, "hardbake": null, "descum": null, "applications": [ "etch-mask", "electroplating-molding", "mems-structural" ], "etchResistance": "\"Excellent wet and dry etch adhesion\" (Advantages, p.1). Figure 8 (p.4) reports 100:1 etch selectivity in a Bosch DRIE process, demonstrated patterning 2.5-10 µm features to 200 µm deep and 5-20 µm features to 100 µm deep. The p.1 application photo (\"Etched Trenches (Bosch Process), 4-10 µm Features up to 100 µm deep\") shows the same class of result. (These etch-depth examples are shown at the SPR220 series level, not attributed to a specific grade.)", "liftoffSuitable": null, "platingSuitable": true, "stripper": "MICROPOSIT REMOVER 1165, two-bath process, each bath at 80°C (176°F): the first bath removes the bulk of the photoresist, the second removes residual traces. Source: \"Photoresist Removal\", p.4.", "storage": "\"Recommended storage for SPR220 is in an upright position in a dry area at 40-60°F (4-15°F). Keep away from oxidizers, acids, and bases. Keep container sealed when not in use.\" (Quoted verbatim, including an apparent unit inconsistency in the source: 40-60°F does not correspond to 4-15°F — the parenthetical is likely a typo for °C in the original document, but is transcribed as printed rather than silently corrected.) Source: \"Storage\", p.4.", "notes": "SPR220-7.0 is the thickest-nominal grade in the MEGAPOSIT SPR 220 line pictured in this datasheet, spin-coating from roughly 5 µm up to 50+ µm depending on speed and wafer size (two separately measured curves, on 4-inch and 8-inch wafers, are both captured above and agree within a few percent where they overlap). Its most easily-missed processing step is the mandatory minimum 35-minute ambient hold between exposure and PEB — water must diffuse back into the film to complete the photoreaction, and skipping this wait is a well-known way to get an incomplete or inconsistent PEB result with this resist family. The exposure-dose table only publishes i-Line numbers up to a 5.0 µm reference thickness, so there is no clean i-Line dose match at this grade's 7.0 µm nominal thickness; the nearest published numbers (g-Line 470 mJ/cm² at exactly 7.0 µm, or i-Line 380 mJ/cm² at 5.0 µm) each mismatch on either wavelength or thickness and are reported only as context, not as a scalar dose. The lineage of this resist family is worth noting for provenance purposes: Shipley Company originated the SPR line, was acquired into Rohm and Haas Electronic Materials (the entity printed on this 2004 datasheet), which was in turn acquired by Dow Chemical (2009), then spun into DuPont Electronics & Imaging, and the MEGAPOSIT/MICROPOSIT/MF brand family is now sold by Kayaku Advanced Materials — the same corporate family that owns MicroChem/KMPR and SU-8. Tone and base chemistry are not stated anywhere in this particular datasheet and are left unset rather than assumed from the product family's general reputation. Chemistry classified as dnq-novolak from the Dow MEGAPOSIT SPR220-3.0 MSDS composition table (same MEGAPOSIT SPR220 line: cresol novolak resin + diazo photoactive compound) (chemistry classified 2026-07-12 from manufacturer SDS/TDS).", "developerFamily": "tmah", "references": [ { "type": "paper", "title": "Realization of electroplating molds with thick positive SPR 220-7 photoresist", "authors": "Kukharenka et al.", "journal": "Journal of Materials Science: Materials in Electronics", "year": 2003, "doi": "10.1023/A:1023923911921", "url": "https://doi.org/10.1023/A:1023923911921", "accessedDate": "2026-07-12", "summary": "34-54 µm multi-coat SPR220-7 films used as electroplating molds." }, { "type": "paper", "title": "A comparative study of different thick photoresists for MEMS applications", "authors": "Koukharenko et al.", "journal": "Journal of Materials Science: Materials in Electronics", "year": 2005, "doi": "10.1007/s10854-005-4977-2", "url": "https://doi.org/10.1007/s10854-005-4977-2", "accessedDate": "2026-07-12", "summary": "Benchmarks SPR220 against SU-8 and AZ 9260 as thick-film MEMS resists." } ], "provenance": { "datasheetUrl": "https://research.engineering.ucdavis.edu/cnm2/wp-content/uploads/sites/11/2013/05/SPR220_Data_Sheet.pdf", "datasheetVersionOrDate": "ME04N097, Rev. 2, September 2004 (printed on p.1 and in the footer)", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://asrc.gc.cuny.edu/wp-content/uploads/media/global-assets/Megaposit-SPR-220-3.0-Positive-Photoresist-MSDS.pdf", "what": "Dow MEGAPOSIT SPR220-3.0 MSDS (same MEGAPOSIT SPR220 line) states 'Cresol novolak resin 25.0-35.0%, Diazo Photoactive Compound 1.0-10.0%'; the basis for classifying SPR220-7.0 as dnq-novolak." } ] }, "_provenanceNote": "Fetched directly from the assigned URL (UC Davis CNM2 mirror) and confirmed to be the correct document: Rohm and Haas Electronic Materials' own \"MEGAPOSIT SPR220 Series Photoresists\" datasheet (ME04N097, Rev. 2, September 2004). This is a university (UC Davis) mirror of the manufacturer's own document, not a university-authored SOP.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "s1805", "name": "Microposit S1805", "manufacturer": "Rohm and Haas Electronic Materials", "productLine": "MICROPOSIT S1800 G2 Series", "aliases": [ "S1805", "S1805 G2", "MICROPOSIT S1805 G2" ], "tone": "positive", "chemistry": "dnq-novolak", "_chemistryNote": "The document never names a resist chemistry (no 'novolak', 'DNQ', or 'diazonaphthoquinone' anywhere in its six pages) for any grade, including S1805. Left null rather than assumed from general industry knowledge of the S1800 family.", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not discussed. The document only describes single-layer binary IC masking properties (Figure 1 'Masking Linearity SEMS', Figure 7 masking-linearity plot); no grayscale or gray-tone lithography capability is claimed anywhere in the document.", "status": "active", "successorSlug": null, "summary": "MICROPOSIT S1805 G2 is the thin-film grade of the MICROPOSIT S1800 G2 positive photoresist family — the one to reach for when a design needs the thinnest single-layer coat in the series and the finest linewidth control rather than film thickness. It is the thinnest of four undyed grades named in Rohm and Haas Electronic Materials' MICROPOSIT S1800 G2 positive photoresist datasheet (ME06N041, Rev. 0, October 2006), a g-line-optimized single-layer resist for IC device fabrication; it is the lowest trace on the series' film-thickness-vs-spin-speed chart, alongside S1811, S1813 and S1818 G2.", "thicknessRange": null, "spinCurves": [ { "label": "S1805 G2", "points": [ { "rpm": 2000, "um": 0.67 }, { "rpm": 3000, "um": 0.55 }, { "rpm": 4000, "um": 0.5 }, { "rpm": 5000, "um": 0.47 }, { "rpm": 6000, "um": 0.44 }, { "rpm": 7000, "um": 0.41 } ], "source": "read from figure, 'MICROPOSIT S1800 G2 Photoresist Undyed Series Spin Speed Curves' (Figure 2), p.2 of MICROPOSIT S1800 G2 Series Photoresists datasheet (Rohm and Haas Electronic Materials, ME06N041, Rev. 0, October 2006); identified as the lowest of four plotted traces (cross marker, legend order S1818/S1813/S1811/S1805 top-to-bottom by thickness — the four curves do not cross across the plotted range); read at gridline/marker crossings every 1,000 rpm from 2,000–7,000 rpm off a linear-linear chart (thickness 0–40,000 Å, spin speed 1,000–8,000 rpm); no numeric table accompanies the figure so no independent point-value anchor exists in the document — visual chart-reading only, typical uncertainty ~10–15%; digitized 2026-07-12", "figureRead": true } ], "spinNotes": "Figure 2 ('MICROPOSIT S1800 G2 Photoresist Undyed Series Spin Speed Curves', p.2) plots four traces — S1818 G2, S1813 G2, S1811 G2, S1805 G2 — on one axis (spin speed 1,000–8,000 rpm vs photoresist thickness 0–40,000 Å), each carrying a distinct legend marker (diamond, triangle, square, cross respectively, in that order top-to-bottom by thickness) and, because the four curves do not cross across the plotted range, S1805 G2 (cross marker) is identified as the lowest trace at every spin speed. The trace was digitized 2026-07-12 by reading gridline/marker crossings at 1,000 rpm intervals from 2,000–7,000 rpm (see spinCurves); no numeric thickness-vs-rpm table accompanies the figure, so there is no independent point-value anchor and the digitized values carry typical chart-reading uncertainty (~10–15% — see spinCurves[0].source). Table 2 (p.2) records that Figure 2's coatings used an SVG 81 coater and a 115°C/60 sec hotplate softbake on silicon, measured on a Nanometrics 210 — that condition applies to this grade's trace and is recorded under softbake below. The document states maximum coating uniformity is 'typically attained between the spin speeds of 3,500–5,500 rpm' (p.2, series-wide statement, not S1805-specific). No rehydration hold or edge-bead handling is discussed anywhere in the document.", "adhesion": { "hmds": true, "notes": "\"MICROPOSIT S1800 G2 Series Photoresist work well with the hexamethyldisilazane-based MICROPOSIT Primers. Concentrated MICROPOSIT Primer is recommended when vacuum vapor priming. Diluted primer is recommended for liquid phase priming applications.\" (SUBSTRATE PREPARATION, p.2 — series-wide statement, not grade-specific)." }, "rehydration": null, "softbake": { "temp_c": 115, "time_s": 60, "method": "hotplate", "notes": "From Table 2 ('Process Conditions, Refer to Figure 2'): coat tool SVG 81, softbake 115°C/60 sec hotplate, measured on a Nanometrics 210. This is the condition used to generate the Figure 2 spin-speed-curve trace that includes S1805 G2; it is not a separately stated softbake recommendation outside that figure.", "source": "Table 2, p.2 of MICROPOSIT S1800 G2 SERIES PHOTORESISTS, ME06N041, Rev. 0, October 2006" }, "exposureDose": { "doses": [], "datasheetBasis": "\"MICROPOSIT S1800 G2 Series Photoresists can be exposed with light sources in the spectral output range of 350–450 nm. The exposure properties have been optimized for use at 436 nm.\" (EXPOSURE section, p.3 — series-wide statement; no wavelength-specific dose given for this grade).", "_note": "No dose number is published for S1805 anywhere in this document. The only quantified dose — 150 mJ/cm² 'Sizing Energy' (Table 9, p.4) — plus the entire Dill-parameter table (Table 6: A/B at both 365 nm and 436 nm, p.3), the contrast curve (Figure 6, E0 = 82 mJ/cm²), and the exposure/focus-latitude plots (Figures 7–9) are all explicitly attributed to MICROPOSIT S1813 G2 (Tables 1, 4–9 name S1813 G2 as the tested photoresist for every one of those figures) and must not be borrowed for S1805." }, "peb": null, "floodExposure": null, "develop": null, "hardbake": null, "descum": null, "applications": [], "etchResistance": null, "liftoffSuitable": null, "platingSuitable": null, "stripper": "\"Residue-free photoresist removal using standard MICROPOSIT removers\" (ADVANTAGES section, p.1) — no specific remover product is named.", "storage": "\"Store products in tightly closed original containers at temperatures recommended on the product label.\" (STORAGE section, p.5) — no specific temperature is printed in the document itself.", "notes": "This grade appears exactly once in the document: as the lowest of four traces (alongside S1811, S1813 and S1818 G2) on the undyed-series film-thickness-vs-spin-speed chart (Figure 2, p.2). Choose S1805 G2 over the thicker S1811, S1813 and S1818 grades when the process needs a thin resist layer and tight resolution rather than film thickness — on the same g-line process and develop chemistry it lays down the least film of the four. No dose, develop time, or lithographic-performance data (contrast curve, Dill parameters, absorbance spectrum, masking linearity, exposure/focus latitude — Figures 4–9, Tables 4–9) is published for S1805 specifically; every one of those figures and tables explicitly names MICROPOSIT S1813 G2 as the tested grade, and none of that data may be borrowed here. The document states the S1800 G2 system broadly is optimized for g-line (436 nm) exposure and usable across 350–450 nm, is compatible with HMDS-based MICROPOSIT primers, and works with both metal-ion-free (MF-319 family) and metal-ion-bearing MICROPOSIT developers, but none of those statements carry an S1805-specific number. This recipe replaces an earlier extraction made from the superseded circa-1993 Shipley 'MPR S1800 1093' edition, per a WL decision on 2026-07-10; the S1800 line traces from Shipley through Rohm and Haas Electronic Materials (publisher of this G2 edition) to Dow, DuPont, and today's Kayaku Advanced Materials. Chemistry classified as dnq-novolak from the Dow/Rohm and Haas S1818 MSDS composition table (mixed cresol novolak resin + diazo photoactive compound) and Kayaku's S1800 G2 series-wide statement (chemistry classified 2026-07-12 from manufacturer SDS/TDS).", "developerFamily": null, "provenance": { "datasheetUrl": "https://www.nanophys.kth.se/nanolab/resists/S1813/Microposit_S1800_G2_Serie.pdf", "datasheetVersionOrDate": "ME06N041, Rev. 0, October 2006", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://amolf.nl/wp-content/uploads/2016/09/datasheets_S1800.pdf", "what": "AMOLF-hosted mirror of the superseded ~1993 Shipley 'MPR S1800 1093' edition of the S1800 datasheet, archived here per the 2026-07-10 WL decision to re-extract S1805 from the current G2 edition instead; retained as a reference, not a data source for this recipe." }, { "url": "https://kayakuam.com/products/microposit-s1800-g2-series-photoresists/", "what": "Kayaku's MICROPOSIT S1800 G2 series page states the series-wide formulation (mixed cresol novolak resin + diazo photoactive compound); used as the basis for classifying S1805 as dnq-novolak, corroborated by the Dow/Rohm and Haas S1818 MSDS composition table." } ] }, "_provenanceNote": "kayakuam.com (the current manufacturer's own site) 403s automated fetches (verified 2026-07-10); this URL is a KTH Nanolab (Royal Institute of Technology, Sweden) mirror of the manufacturer's own MICROPOSIT S1800 G2 datasheet (Rohm and Haas Electronic Materials, ME06N041 Rev. 0, October 2006). This recipe REPLACES an earlier extraction made from the superseded ~1993 Shipley edition ('MPR S1800 1093'), per a WL decision on 2026-07-10. [AUDIT NOTE, 2026-07-10] This recipe was re-extracted against the CURRENT “S1800 G2 Series” edition (ME06N041 Rev. 0, October 2006) on WL's decision, replacing an earlier extraction from the superseded ~1993 Shipley edition (document code “MPR S1800 1093”, archived at the secondarySources URL). Consequence a QC reviewer must know: the G2 edition's undyed spin curves (Figure 2) could not be confidently digitized, so this recipe publishes NO spin curve, whereas the 1993 edition plots traces that a reader could read. The curve must be digitized from G2 Figure 2 at visual QC — do NOT copy points from the legacy edition into a recipe that cites G2. Also note: the G2 edition publishes exactly one quantified dose in the entire document (150 mJ/cm² “Sizing Energy”), attributed only to S1813 on g-line (436 nm) steppers. No per-grade dose exists for this grade, and no i-line (365 nm) or h-line (405 nm) dose appears anywhere in it for any grade — Table 6's Dill A/B coefficients at 365 nm are an absorption-model parameter, not an exposure dose.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "s1818", "name": "Microposit S1818", "manufacturer": "Rohm and Haas Electronic Materials", "productLine": "MICROPOSIT S1800 G2 Series", "aliases": [ "S1818", "S1818 G2", "MICROPOSIT S1818 G2" ], "tone": "positive", "chemistry": "dnq-novolak", "_chemistryNote": "The document never names a resist chemistry (no 'novolak', 'DNQ', or 'diazonaphthoquinone' anywhere in its six pages) for any grade, including S1818. Left null rather than assumed from general industry knowledge of the S1800 family.", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not discussed. The document only describes single-layer binary IC masking properties (Figure 1 'Masking Linearity SEMS', Figure 7 masking-linearity plot); no grayscale or gray-tone lithography capability is claimed anywhere in the document.", "status": "active", "successorSlug": null, "summary": "MICROPOSIT S1818 G2 is the thick-film grade of the MICROPOSIT S1800 G2 positive photoresist family — chosen when a single S1800 coat must be as thick as the series allows, for deeper etch masking or taller step coverage than the thinner S1805, S1811 and S1813 grades provide. It is the thickest of four undyed grades named in Rohm and Haas Electronic Materials' MICROPOSIT S1800 G2 positive photoresist datasheet (ME06N041, Rev. 0, October 2006), a g-line-optimized single-layer resist for IC device fabrication; it is the top trace on the series' film-thickness-vs-spin-speed chart, alongside S1805, S1811 and S1813 G2.", "thicknessRange": null, "spinCurves": [ { "label": "S1818 G2", "points": [ { "rpm": 2000, "um": 2.65 }, { "rpm": 3000, "um": 2.15 }, { "rpm": 4000, "um": 1.85 }, { "rpm": 5000, "um": 1.7 }, { "rpm": 6000, "um": 1.6 }, { "rpm": 7000, "um": 1.5 } ], "source": "read from figure, 'MICROPOSIT S1800 G2 Photoresist Undyed Series Spin Speed Curves' (Figure 2), p.2 of MICROPOSIT S1800 G2 Series Photoresists datasheet (Rohm and Haas Electronic Materials, ME06N041, Rev. 0, October 2006); identified as the highest of four plotted traces (diamond marker, legend order S1818/S1813/S1811/S1805 top-to-bottom by thickness — the four curves do not cross across the plotted range); read at gridline/marker crossings every 1,000 rpm from 2,000–7,000 rpm off a linear-linear chart (thickness 0–40,000 Å, spin speed 1,000–8,000 rpm); no numeric table accompanies the figure so no independent point-value anchor exists in the document — visual chart-reading only, typical uncertainty ~10–15%; digitized 2026-07-12", "figureRead": true } ], "spinNotes": "Figure 2 ('MICROPOSIT S1800 G2 Photoresist Undyed Series Spin Speed Curves', p.2) plots four traces — S1818 G2, S1813 G2, S1811 G2, S1805 G2 — on one axis (spin speed 1,000–8,000 rpm vs photoresist thickness 0–40,000 Å), each carrying a distinct legend marker (diamond, triangle, square, cross respectively, in that order top-to-bottom by thickness) and, because the four curves do not cross across the plotted range, S1818 G2 (diamond marker) is identified as the highest trace at every spin speed. The trace was digitized 2026-07-12 by reading gridline/marker crossings at 1,000 rpm intervals from 2,000–7,000 rpm (see spinCurves); no numeric thickness-vs-rpm table accompanies the figure, so there is no independent point-value anchor and the digitized values carry typical chart-reading uncertainty (~10–15% — see spinCurves[0].source). Table 2 (p.2) records that Figure 2's coatings used an SVG 81 coater and a 115°C/60 sec hotplate softbake on silicon, measured on a Nanometrics 210 — that condition applies to this grade's trace and is recorded under softbake below. The document states maximum coating uniformity is 'typically attained between the spin speeds of 3,500–5,500 rpm' (p.2, series-wide statement, not S1818-specific). No rehydration hold or edge-bead handling is discussed anywhere in the document.", "adhesion": { "hmds": true, "notes": "\"MICROPOSIT S1800 G2 Series Photoresist work well with the hexamethyldisilazane-based MICROPOSIT Primers. Concentrated MICROPOSIT Primer is recommended when vacuum vapor priming. Diluted primer is recommended for liquid phase priming applications.\" (SUBSTRATE PREPARATION, p.2 — series-wide statement, not grade-specific)." }, "rehydration": null, "softbake": { "temp_c": 115, "time_s": 60, "method": "hotplate", "notes": "From Table 2 ('Process Conditions, Refer to Figure 2'): coat tool SVG 81, softbake 115°C/60 sec hotplate, measured on a Nanometrics 210. This is the condition used to generate the Figure 2 spin-speed-curve trace that includes S1818 G2; it is not a separately stated softbake recommendation outside that figure.", "source": "Table 2, p.2 of MICROPOSIT S1800 G2 SERIES PHOTORESISTS, ME06N041, Rev. 0, October 2006" }, "exposureDose": { "doses": [], "datasheetBasis": "\"MICROPOSIT S1800 G2 Series Photoresists can be exposed with light sources in the spectral output range of 350–450 nm. The exposure properties have been optimized for use at 436 nm.\" (EXPOSURE section, p.3 — series-wide statement; no wavelength-specific dose given for this grade).", "_note": "No dose number is published for S1818 anywhere in this document. The only quantified dose — 150 mJ/cm² 'Sizing Energy' (Table 9, p.4) — plus the entire Dill-parameter table (Table 6: A/B at both 365 nm and 436 nm, p.3), the contrast curve (Figure 6, E0 = 82 mJ/cm²), and the exposure/focus-latitude plots (Figures 7–9) are all explicitly attributed to MICROPOSIT S1813 G2 (Tables 1, 4–9 name S1813 G2 as the tested photoresist for every one of those figures) and must not be borrowed for S1818." }, "peb": null, "floodExposure": null, "develop": null, "hardbake": null, "descum": null, "applications": [], "etchResistance": null, "liftoffSuitable": null, "platingSuitable": null, "stripper": "\"Residue-free photoresist removal using standard MICROPOSIT removers\" (ADVANTAGES section, p.1) — no specific remover product is named.", "storage": "\"Store products in tightly closed original containers at temperatures recommended on the product label.\" (STORAGE section, p.5) — no specific temperature is printed in the document itself.", "notes": "This grade appears exactly once in the document: as the highest of four traces (alongside S1805, S1811 and S1813 G2) on the undyed-series film-thickness-vs-spin-speed chart (Figure 2, p.2). Pick S1818 G2 over its thinner S1805/S1811/S1813 siblings when the process needs the most film a single S1800 G2 coat can give — it sits at the top of the series' spin-speed chart, coating thickest of the four undyed grades at any given speed. No dose, develop time, or lithographic-performance data (contrast curve, Dill parameters, absorbance spectrum, masking linearity, exposure/focus latitude — Figures 4–9, Tables 4–9) is published for S1818 specifically; every one of those figures and tables explicitly names MICROPOSIT S1813 G2 as the tested grade, and none of that data may be borrowed here. The document states the S1800 G2 system broadly is optimized for g-line (436 nm) exposure and usable across 350–450 nm, is compatible with HMDS-based MICROPOSIT primers, and works with both metal-ion-free (MF-319 family) and metal-ion-bearing MICROPOSIT developers, but none of those statements carry an S1818-specific number. This recipe replaces an earlier extraction made from the superseded circa-1993 Shipley 'MPR S1800 1093' edition, per a WL decision on 2026-07-10; the S1800 line traces from Shipley through Rohm and Haas Electronic Materials (publisher of this G2 edition) to Dow, DuPont, and today's Kayaku Advanced Materials. Chemistry classified as dnq-novolak directly from the Dow/Rohm and Haas MICROPOSIT S1818 MSDS composition table (mixed cresol novolak resin + diazo photoactive compound) (chemistry classified 2026-07-12 from manufacturer SDS/TDS).", "developerFamily": null, "provenance": { "datasheetUrl": "https://www.nanophys.kth.se/nanolab/resists/S1813/Microposit_S1800_G2_Serie.pdf", "datasheetVersionOrDate": "ME06N041, Rev. 0, October 2006", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://amolf.nl/wp-content/uploads/2016/09/datasheets_S1800.pdf", "what": "AMOLF-hosted mirror of the superseded ~1993 Shipley 'MPR S1800 1093' edition of the S1800 datasheet, archived here per the 2026-07-10 WL decision to re-extract S1818 from the current G2 edition instead; retained as a reference, not a data source for this recipe." }, { "url": "https://aggiefab.tamu.edu/wp-content/uploads/2021/04/Microposit-S1818-G2-Positive-Photoresist-MSDS.pdf", "what": "Dow/Rohm and Haas MICROPOSIT S1818 MSDS states 'Mixed cresol novolak resin 15.0-25.0%, Diazo Photoactive Compound 1.0-10.0%'; the basis for classifying S1818 as dnq-novolak." } ] }, "_provenanceNote": "kayakuam.com (the current manufacturer's own site) 403s automated fetches (verified 2026-07-10); this URL is a KTH Nanolab (Royal Institute of Technology, Sweden) mirror of the manufacturer's own MICROPOSIT S1800 G2 datasheet (Rohm and Haas Electronic Materials, ME06N041 Rev. 0, October 2006). This recipe REPLACES an earlier extraction made from the superseded ~1993 Shipley edition ('MPR S1800 1093'), per a WL decision on 2026-07-10. [AUDIT NOTE, 2026-07-10] This recipe was re-extracted against the CURRENT “S1800 G2 Series” edition (ME06N041 Rev. 0, October 2006) on WL's decision, replacing an earlier extraction from the superseded ~1993 Shipley edition (document code “MPR S1800 1093”, archived at the secondarySources URL). Consequence a QC reviewer must know: the G2 edition's undyed spin curves (Figure 2) could not be confidently digitized, so this recipe publishes NO spin curve, whereas the 1993 edition plots traces that a reader could read. The curve must be digitized from G2 Figure 2 at visual QC — do NOT copy points from the legacy edition into a recipe that cites G2. Also note: the G2 edition publishes exactly one quantified dose in the entire document (150 mJ/cm² “Sizing Energy”), attributed only to S1813 on g-line (436 nm) steppers. No per-grade dose exists for this grade, and no i-line (365 nm) or h-line (405 nm) dose appears anywhere in it for any grade — Table 6's Dill A/B coefficients at 365 nm are an absorption-model parameter, not an exposure dose.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "mr-dwl-40", "name": "mr-DWL 40", "manufacturer": "micro resist technology GmbH", "productLine": "mr-DWL series", "aliases": [ "mrDWL40", "mr-DWL40" ], "tone": "negative", "chemistry": "epoxy", "_chemistryNote": "The flyer lists mr-DWL only as one of several 'Negative Photoresists' series (alongside ma-N 400/1400/2400, mr-EBL 6000, mr-UVL 6000, EpoCore/EpoClad); it never states a chemistry mechanism (e.g. epoxy, acrylate) for mr-DWL specifically. EpoCore/EpoClad's 'Epo' branding suggests epoxy chemistry for THAT pair but is also not explicitly labeled 'epoxy' anywhere in this document, and in any case mr-DWL is a separate series. Left null rather than guessed.", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Document never uses the word 'greyscale' for mr-DWL. It markets mr-DWL for 'Direct Laser Writing (DLW) @ 405 nm & Two Photon Polymerization (2PP)' — arbitrary focused-beam 3D structuring — which is a related but distinct capability from continuous dose-modulated greyscale lithography, and is not described as such here.", "status": "active", "successorSlug": null, "summary": "mr-DWL 40 is the mid-thickness grade of micro resist technology's mr-DWL negative-tone resist series, formulated for 405 nm direct laser writing (DLW) and two-photon polymerization (2PP), sold ready-to-use for coatings from 20 to 100 µm.", "thicknessRange": { "min_um": 20, "max_um": 100, "basis": "stated", "source": "stated — the flyer's mr-DWL table (p.4) lists 'mr-DWL 40: 20 µm → 100 µm' as the ready-to-use film-thickness range for this grade specifically (alongside mr-DWL 5: 3-12 µm and mr-DWL 100: 20-150 µm)." }, "spinCurves": [], "spinNotes": "No spin-speed-vs-thickness table or figure is published for mr-DWL anywhere in this flyer — only the stated achievable film-thickness range per grade (see thicknessRange). No accel, dispense, or edge-bead guidance is given either. This is expected: the source is a multi-series marketing flyer, not a technical datasheet.", "adhesion": { "hmds": null, "notes": "Not addressed in this flyer." }, "rehydration": null, "softbake": null, "exposureDose": { "doses": [], "datasheetBasis": "'Spectral sensitivity: High sensitivity > 400 nm, DLW @ 405 nm' (mr-DWL table, p.4). mr-DWL is explicitly marketed for 405 nm direct laser writing.", "_note": "Despite mr-DWL being marketed for 405 nm DLW (the one documented exception noted in this project's extraction protocol), this flyer gives no numeric dose value in mJ/cm² or mW/cm² for mr-DWL at any wavelength — only the qualitative 'high sensitivity > 400 nm' descriptor and an absorption-coefficient-vs-wavelength figure (unexposed / exposed / exposed+PEB) with no axis calibration to a dose. Both dose fields left null rather than estimated from the qualitative claim or the absorption curve." }, "peb": null, "floodExposure": null, "develop": { "developer": "mr-Dev 600", "dilution": null, "time_s": null, "method": null, "rinse": null, "source": "mr-DWL table, p.4 of Negative Photoresists flyer", "_note": "Flyer states only the developer product name ('mr-Dev 600 (solvent based)'); no dilution, time, or method is published for mr-DWL." }, "hardbake": null, "descum": null, "applications": [ "etch-mask", "electroplating-molding", "general-prototyping", "mems-structural" ], "etchResistance": "\"Etch mask for wet and dry etch processes\" (Main applications, mr-DWL section, p.4).", "liftoffSuitable": null, "platingSuitable": true, "stripper": null, "storage": null, "notes": "mr-DWL 40 is the mid-thickness grade of micro resist technology's mr-DWL negative-tone resist series, formulated for 405 nm exposure in direct laser writing (DLW) and two-photon polymerization (2PP) tools, and sold ready-to-use for coatings from 20 to 100 µm. The only source retrievable for this SKU is a multi-series marketing flyer rather than a full technical datasheet, so process parameters a TDS would normally publish — softbake, exposure dose, post-exposure bake, and a spin-speed-vs-thickness curve — are not stated here and are left null rather than estimated; mr-DWL's own TDS is served behind a non-guessable microresist.de download link and was not retrievable. The flyer names mr-Dev 600 as the developer (solvent-based) but gives no dilution, time, or method. Because mr-DWL is marketed specifically for 405 nm direct-write exposure, its h-line dose could in principle be the operationally relevant one, but no dose value in mJ/cm² is published for this SKU in this document, so exposureDose stays fully null rather than being backfilled from the 365 nm conventions used elsewhere in this library. Chemistry classified as epoxy from microresist.de's description of mr-DWL as chemically amplified negative tone sharing SU-8's advantageous properties, corroborated by peer-reviewed literature grouping SU-8 and mr-DWL as epoxy-based photoresists (chemistry classified 2026-07-12 from manufacturer SDS/TDS).", "developerFamily": "solvent", "provenance": { "datasheetUrl": "https://www.microresist.de/wp-content/uploads/2022/08/NegativeResists_Flyer_Aug22.pdf", "datasheetVersionOrDate": "August 2022", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://microresist.de/en/produkt/mr-dwl-series/", "what": "microresist.de describes mr-DWL as 'chemically amplified negative tone... shares several advantageous properties with SU-8', and peer-reviewed literature groups 'epoxy-based photoresists SU-8 and mr-DWL' together; the basis for classifying mr-DWL 40 as epoxy." } ] }, "_provenanceNote": "Document is micro resist technology GmbH's own product FLYER 'Negative Photoresists for UV, Laser & Electron Beam Lithography' (August 2022, 4 pages), a multi-series marketing overview covering ma-N 400/1400/2400, mr-EBL 6000, mr-UVL 6000, mr-DWL, and EpoCore/EpoClad — NOT a per-SKU technical datasheet (TDS). mr-DWL's TDS is served behind a non-guessable microresist.de '?jet_download=' URL and could not be retrieved. As instructed, this recipe is deliberately mostly null: of the numeric/process fields in the schema, only the stated per-grade thickness range and the developer product name are populated from this document; everything else (bake schedule, dose, spin curve) is genuinely absent from the only retrievable manufacturer document.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "nr7-1500py", "name": "NR7-1500PY", "manufacturer": "Futurrex, Inc.", "productLine": null, "aliases": [ "NR7 1500PY", "NR7-1500 PY" ], "tone": "negative", "chemistry": null, "_chemistryNote": "The retrieved document never names a chemistry mechanism (no 'novolak', 'bisazide', 'epoxy', etc.) — only 'Negative Resist NR7-1500PY is a negative tone photoresist designed for 365nm wavelength exposure.' Left null rather than assumed. (chemistry classified 2026-07-12: re-checked the full archived PDF, including the embedded UPenn SOP prose — no chemistry-mechanism wording anywhere; null confirmed, not a gap.)", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not marketed for greyscale or 3D lithography; the datasheet targets binary lift-off patterning with a tunable undercut sidewall profile adjustable by exposure energy.", "status": "active", "successorSlug": null, "summary": "NR7-1500PY is Futurrex's negative-tone, 365 nm lift-off resist with a negative-sloping (undercut) sidewall profile tunable by exposure energy, coating roughly 1.2-3.0 µm across the manufacturer's published 800-5000 rpm spin range.", "thicknessRange": { "min_um": 1.2, "max_um": 3, "basis": "curve-span", "source": "curve-span — min/max of the published spin curve (Properties table, p.1 of the Futurrex NR7-1500PY Technical Information excerpt) over its stated 800-5000 rpm range; no separate achievable-range statement is made in prose. Points are the midpoint of each rpm's published nm range (library convention — see NR71-3000P). Adjudicated 2026-07-12: confirmed correct against the archived PDF; check's 1.14-3.15 µm used the outer tolerance bounds of the published ranges (5000 rpm's lower bound and 800 rpm's upper bound) rather than the per-point midpoints this library's spin-curve convention uses." }, "spinCurves": [ { "label": "NR7-1500PY", "points": [ { "rpm": 800, "um": 3 }, { "rpm": 1000, "um": 2.7 }, { "rpm": 2000, "um": 1.9 }, { "rpm": 3000, "um": 1.5 }, { "rpm": 4000, "um": 1.3 }, { "rpm": 5000, "um": 1.2 } ], "source": "numeric table 'Film thickness after 150°C hotplate bake for 60 s. / Coating spin speed, 40 s spin (rpm): (nm)', p.1 of Futurrex NR7-1500PY Technical Information (the vendor's own product datasheet page, embedded as p.3 of the retrieved PDF). The datasheet publishes each point as an nm RANGE, not a single value (e.g. 800 rpm -> 2850-3150 nm); each point above is the exact midpoint of that stated range, converted nm->um. Raw ranges: 800rpm=2850-3150nm, 1000rpm=2565-2835nm, 2000rpm=1805-1995nm, 3000rpm=1425-1575nm, 4000rpm=1235-1365nm, 5000rpm=1140-1260nm. NOTE FOR QC: these six ranges are numerically IDENTICAL, rpm-for-rpm, to the spin table already recorded in this library's nr9-1500py.json (also a Futurrex '...1500PY' grade) — plausibly Futurrex reusing one master curve across product lines formulated to the same nominal reference thickness, but a human should re-verify this against the vendor page image before qc:pass, since two independently-read identical tables is also consistent with one of the two readings having been contaminated by memory of the other. The 3000 rpm midpoint (1500 nm = 1.5 um) matches the '1500' in the product name NR7-1500PY, corroborating the midpoint reading." } ], "spinNotes": "The retrieved PDF is a University of Pennsylvania INRF lab SOP (Richard Chang / Ngoc Thanh Pham, Summer 2008) that embeds one page of Futurrex's own 'NEGATIVE RESIST NR7-1500PY' Technical Information sheet as a reference appendix (p.3 of the PDF = the vendor's own page '1'). Only that one vendor page was retrievable; the vendor's usual second 'Processing' page (with step-by-step spin/EBR/softbake/PEB/develop instructions, present in this library's NR9-1500PY and NR71-3000P entries) was not part of this document, so no Futurrex-stated dispense volume, acceleration, or edge-bead-removal procedure is recorded here. Separately, the UPenn SOP itself (pp.1-2 of the PDF, a lab procedure, NOT a Futurrex-published figure) reports its own campus recipe: HMDS primer spun at 3000 rpm/30 s (425 rpm/s accel) then baked 90°C/3 min hotplate before coating; NR7-1500PY spun at 800 rpm/40 s, softbake 150°C/60 s (hotplate) — this line matches the vendor Properties-table bake condition exactly — exposure '2 mins (MA6)' aligner, post-exposure bake 100°C/2 min hotplate, develop with MF-319 for 10 s, yielding 1.5 µm; hard-bake 90°C oven/20 min is called optional. These lab-specific numbers (PEB temp/time, develop time, developer choice, exposure dose-as-time-on-a-specific-tool) are recorded here in prose only, NOT copied into the structured peb/develop/exposureDose fields, because they are one university lab's own process on one specific aligner (Karl Suss MA6) rather than a Futurrex-published recommendation — conflating the two would misattribute a secondary source's numbers to the manufacturer.", "adhesion": { "hmds": null, "notes": "Not addressed on the retrieved vendor page (the vendor's Substrate Preparation guidance, if any, would be on the missing Processing page). The University of Pennsylvania INRF SOP embedding this document applies an HMDS primer (3000 rpm/30 s spin, 90°C/3 min hotplate bake) before coating NR7-1500PY as its own lab practice — a secondary source, not confirmed as Futurrex's own recommendation for this SKU." }, "rehydration": null, "softbake": { "temp_c": 150, "time_s": 60, "method": "hotplate", "notes": "Sourced from the Properties-table heading itself ('Film thickness after 150°C hotplate bake for 60 s.'), which functions as the bake condition Futurrex measured its spin-curve thickness against — the same pattern used on this library's sibling NR9-1500PY and NR71-3000P sheets, where the identical phrase is explicitly corroborated by a numbered Processing step. That corroborating Processing step is not present in this retrieved document, so this figure rests on the Properties-table heading alone. Independently, the University of Pennsylvania SOP embedding this document reports the identical 150°C/60 s softbake as its own lab practice for NR7-1500PY, which corroborates (but does not itself constitute) the vendor figure.", "source": "Properties table heading, p.1 of Futurrex NR7-1500PY Technical Information (embedded as p.3 of the retrieved PDF)" }, "exposureDose": { "doses": [ { "wavelength_nm": 365, "value_mJcm2": 390, "source": "Properties, p.1 of Futurrex NR7-1500PY Technical Information (embedded as p.3 of the retrieved PDF)" } ], "datasheetBasis": "'Sensitivity at 365 nm exposure wavelength (mJ/cm² for 1 µm thick film) 390' (Properties, p.1 of the Futurrex NR7-1500PY Technical Information excerpt); the retrieved page also states the resist is 'designed for 365nm wavelength exposure, using tools such as wafer steppers, scanning projection aligners, proximity printers and contact printers.'", "_note": "390 mJ/cm² is explicitly normalized to a 1 µm-thick film — the datasheet excerpt does not give an absolute dose for other process thicknesses or a scaling formula. No h-line (405 nm) dose is published on the retrieved page. The vendor's exposure-tool/equipment guidance (if any, beyond the wavelength statement above) would be on the missing Processing page." }, "peb": null, "floodExposure": null, "develop": { "developer": null, "dilution": null, "time_s": null, "method": null, "rinse": null, "source": null, "_note": "The vendor Description bullet states only that 'development of NR7-1500PY is accomplished in a basic water solution' (p.1) — no specific developer product, dilution, method, or time is named on the retrieved vendor page (that detail would be on the missing Processing page). Left entirely null rather than filling from the University of Pennsylvania SOP embedding this document, which reports its own lab practice of developing with Shipley MF-319 (also listing Futurrex's own Resist Developer RD6 as an accepted alternative in its Materials list) for 10 s — that is one lab's process choice on one tool, not a Futurrex-published recommendation for this SKU, so it is not copied into this field (see spinNotes)." }, "hardbake": null, "descum": null, "applications": [ "lift-off", "etch-mask" ], "etchResistance": "\"Superior selectivity in RIE process\" (Description, p.1 of the Futurrex NR7-1500PY Technical Information excerpt).", "liftoffSuitable": true, "platingSuitable": null, "stripper": "Resist Remover RR4 (Description, p.1: 'easy resist removal in Resist Remover RR4') — note this differs from the RR5 remover specified for Futurrex's NR9-1500PY and the RR41 remover specified for NR71-3000P elsewhere in this library; the three SKUs' strip chemistries should not be conflated.", "storage": "\"Guaranteed shelf life at 25°C storage (years) 3\" (Properties, p.1); \"shelf life exceeding 3 years at room temperature storage\" (Description, p.1) of the Futurrex NR7-1500PY Technical Information excerpt. Full storage-condition guidance (temperature range, light/heat/ignition precautions), if published, would be on the missing Processing page.", "notes": "NR7-1500PY is Futurrex's negative-tone, 365 nm lift-off resist, formulated in cyclohexanone (24-28% solids) and developed in an unspecified 'basic water solution'; the datasheet's headline advantage is 'easy adjustment of the degree of resist undercut as a function of exposure energy', which is what makes its negative-sloping sidewall profile useful for lift-off. Only one page of Futurrex's own two-page Technical Information sheet was retrievable — embedded as a reference appendix inside a 2008 University of Pennsylvania INRF cleanroom SOP rather than fetched as a standalone vendor PDF — so the vendor's own Processing-page guidance (PEB temperature/time, develop time, edge-bead removal, storage conditions) is not available here and those fields are left null; the SOP's own campus recipe for those steps is recorded in spinNotes/adhesion.notes/develop._note but explicitly NOT copied into the structured fields, since it is one lab's process on one aligner (Karl Suss MA6), not a Futurrex-published figure. The spin-thickness table that IS available reproduces exactly the same six rpm/nm-range data points already recorded for the sibling grade NR9-1500PY elsewhere in this library — flagged in spinNotes for a human QC pass rather than silently assumed correct. The 390 mJ/cm² sensitivity at 365 nm is normalized to a 1 µm-thick film, and resist removal uses Resist Remover RR4 — a third distinct remover code, alongside RR5 (NR9-1500PY) and RR41 (NR71-3000P), that should not be conflated across the three Futurrex SKUs in this library.", "developerFamily": null, "provenance": { "datasheetUrl": "https://www.seas.upenn.edu/~nanosop/documents/NR7-1500PY.pdf", "datasheetVersionOrDate": null, "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://www.seas.upenn.edu/~nanosop/documents/NR7-1500PY.pdf", "what": "Pages 1-2 of this SAME PDF (as distinct from page 3, the appended vendor sheet used as datasheetUrl basis): a University of Pennsylvania INRF ('Negative Resist NR7-1500PY photolithography', Richard Chang / Ngoc Thanh Pham, Summer 2008) cleanroom lab SOP. Used only for practical/lab-specific process numbers not present on the retrieved vendor page — HMDS pretreatment step, PEB temp/time, develop time and developer product, hard-bake — each explicitly flagged in the relevant field's notes as one lab's own recipe, not a Futurrex-published recommendation. Not used for any headline vendor spec (spin table and 365 nm dose both come from the appended Futurrex Technical Information page itself)." } ] }, "_provenanceNote": "[FLAGGED during audit — WL must adjudicate before qc:pass] The cited URL is a 2008 UPenn INRF LAB SOP that EMBEDS one page (Description + Properties) of Futurrex's own two-page Technical Information sheet. Our standing rule is that a university-authored SOP is never a datasheetUrl — only a secondarySources entry. The extraction used ONLY the embedded vendor page (spin table, 365 nm sensitivity) and explicitly refused the SOP's own campus recipe (100 °C/2 min PEB, MF-319 10 s develop, HMDS pretreat, MA6 aligner), which is why peb and develop are null. Futurrex publishes no per-SKU PDF on futurrex.com and UMN mirrors no NR7 sheet. Either accept this as a mirror of the manufacturer's page (and say so), or treat it as fetchFailed. SEPARATE QC FLAG: this recipe's spin table is numerically IDENTICAL rpm-for-rpm to nr9-1500py's. That is consistent with Futurrex reusing one master curve across its \"…1500PY\" grades (both are 1.5 µm at 3000 rpm, matching the SKU naming), and the two extractions were independent and disagree on dose (390 vs 190 mJ/cm²), so contamination is unlikely — but it must be confirmed against the source page image before either recipe passes.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "nr71-3000p", "name": "NR71-3000P", "manufacturer": "Futurrex, Inc.", "productLine": null, "aliases": [ "NR71 3000P", "NR71-3000 P" ], "tone": "negative", "chemistry": null, "_chemistryNote": "Datasheet never names a chemistry mechanism (no 'novolak', 'bisazide', 'epoxy', etc.) — only 'Negative Resist NR71-3000P is a negative tone photoresist'. Left null rather than assumed.", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not marketed for greyscale or 3D lithography; the datasheet describes a 'straight resist sidewall profile' for binary patterning, high thermal stability (up to 180°C) and RIE selectivity.", "status": "active", "successorSlug": null, "summary": "NR71-3000P is Futurrex's high-photospeed negative-tone 365 nm resist with a straight sidewall profile and superior thermal stability (up to 180°C), coating roughly 2.2-6.0 µm across the manufacturer's published 800-5000 rpm spin range without requiring an HMDS adhesion promoter.", "thicknessRange": { "min_um": 2.233, "max_um": 6, "basis": "curve-span", "source": "curve-span — min/max of the published spin curve (Properties table, p.1) over its stated 800-5000 rpm range; no separate achievable-range statement is made in prose." }, "spinCurves": [ { "label": "NR71-3000P", "points": [ { "rpm": 800, "um": 6 }, { "rpm": 3000, "um": 3 }, { "rpm": 4000, "um": 2.59 }, { "rpm": 5000, "um": 2.233 } ], "source": "numeric table 'Film thickness after 150°C hotplate bake for 60 s (nm) / Coating spin speed, 40 s spin (rpm)', p.1 of Futurrex NR71-3000P Technical Information. The datasheet publishes each point as an nm RANGE, not a single value: published as 5700-6300 nm at 800 rpm, 2850-3150 nm at 3000 rpm, 2460-2720 nm at 4000 rpm, 2140-2326 nm at 5000 rpm; midpoint used for each (every range is symmetric, approx. nominal +/-4-5%), converted nm->um. The 3000 rpm midpoint (3000 nm = 3.0 um exactly) matches the '3000' in the product name NR71-3000P, corroborating the midpoint reading. Only 4 spin speeds are published (no 1000 or 2000 rpm rows, unlike the NR9-1500PY datasheet). human-adjudicated 2026-07-12 from the archived PDF." } ], "spinNotes": "Spin coating is at a selected speed for 40 s (Processing step 1). Edge Bead Remover EBR2 is applied to the bottom and edge of the coated wafer for 10 s, stopping 5 s before spin cycle completion (Processing step 2). Bake schedules differ by substrate thermal conductivity — see softbake/peb notes.", "adhesion": { "hmds": false, "notes": "Datasheet lists 'elimination of a need for application of adhesion promoters such as HMDS' as one of NR71-3000P's advantages over other resists (Description, p.1)." }, "rehydration": null, "softbake": { "temp_c": 150, "time_s": 60, "method": "hotplate", "notes": "150°C/60 s applies to good thermal conductors (Si, GaAs, InP). For a 1 mm-thick glass substrate the datasheet instead specifies 165°C for 240 s.", "source": "Processing step 3, p.2" }, "exposureDose": { "doses": [ { "wavelength_nm": 365, "value_mJcm2": 21, "source": "Properties, p.1" } ], "datasheetBasis": "'Sensitivity at 365 nm exposure wavelength (mJ/cm² for 1 µm thick film) 21' (Properties, p.1); Processing step 4: 'Resist exposure with a tool emitting 365 nm wavelength.'", "_note": "21 mJ/cm² is explicitly normalized to a 1 µm-thick film — no absolute dose for other thicknesses or scaling formula is given. No h-line (405 nm) dose is published anywhere in the document." }, "peb": { "temp_c": 100, "time_s": 60, "notes": "100°C/60 s applies to good thermal conductors (Si, GaAs, InP). For a 1 mm-thick glass substrate the datasheet instead specifies 110°C for 240 s.", "source": "Processing step 5, p.2" }, "floodExposure": null, "develop": { "developer": "Resist Developer RD6", "dilution": null, "time_s": 30, "method": null, "rinse": "Deionized water rinse until water resistivity reaches prescribed limit (Processing step 7).", "source": "Processing steps 6-7, p.2", "_note": "30 s is the develop time example given 'for 3 µm thick film'. Datasheet states development is 'by spray or immersion' without choosing one, so method is left null. No dilution variant is mentioned for this SKU (unlike NR9-1500PY's documented 3:1 RD6:water option). developerFamily tmah: RD6 is 2-3% TMAH in water per Futurrex MSDS (classified 2026-07-12)." }, "hardbake": null, "descum": null, "applications": [ "etch-mask" ], "etchResistance": "\"Superior selectivity in RIE process\" (Description, p.1).", "liftoffSuitable": null, "platingSuitable": null, "stripper": "Resist Remover RR41 (Processing step 9, p.2) — note this differs from the RR5 remover specified for Futurrex's NR9-1500PY.", "storage": "\"Guaranteed shelf life at 25°C storage (years) 3\" (Properties, p.1); \"shelf life exceeding 3 years at room temperature storage\" (Description, p.1).", "notes": "NR71-3000P is Futurrex's high-photospeed negative-tone 365 nm resist, formulated in gamma-butyrolactone and rated for 180°C thermal stability — notably higher than the 100°C rating on Futurrex's NR9 series — and the datasheet lists elimination of an HMDS adhesion-promoter step as a specific advantage. Unlike NR9-1500PY, which is explicitly marketed for lift-off with a tunable undercut, this datasheet describes NR71-3000P's developed sidewall as 'straight' and makes no lift-off claim, so liftoffSuitable is left null rather than assumed either way. The datasheet gives two full bake schedules for both softbake and post-exposure bake — 150°C/60 s (softbake) and 100°C/60 s (PEB) for good thermal conductors (Si, GaAs, InP), versus 165°C/240 s and 110°C/240 s respectively for 1 mm-thick glass — a substrate-dependent difference easy to lose when copying a recipe across substrates. The 21 mJ/cm² sensitivity at 365 nm is explicitly normalized to a 1 µm-thick film, and resist removal uses Resist Remover RR41 (not the RR5 used for NR9-1500PY — the two SKUs' strip chemistries should not be conflated).", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://apps.mnc.umn.edu/pub/photoresists/nr71_3000p_pds.pdf", "datasheetVersionOrDate": null, "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "Document is Futurrex, Inc.'s own 'Negative Resist NR71-3000P / TECHNICAL INFORMATION' product data sheet (2 pages, Futurrex letterhead, Franklin NJ), mirrored by the University of Minnesota Nanofabrication Center (apps.mnc.umn.edu/pub/photoresists/) — futurrex.com does not publish per-SKU PDFs directly (verified 2026-07-10), so this UMN mirror is the primary document read. Content matches the requested SKU (NR71-3000P) exactly throughout — no filename/content mismatch. No revision code or date is printed anywhere on the document.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "nr9-1500py", "name": "NR9-1500PY", "manufacturer": "Futurrex, Inc.", "productLine": null, "aliases": [ "NR9 1500PY", "NR9-1500 PY" ], "tone": "negative", "chemistry": null, "_chemistryNote": "Datasheet never names a chemistry mechanism (no 'novolak', 'bisazide', 'epoxy', etc.) — only 'Negative Resist NR9-1500PY is a negative tone photoresist'. Left null rather than assumed. (chemistry classified 2026-07-12: re-checked the full archived 2-page PDF — no chemistry-mechanism wording anywhere; null confirmed, not a gap.)", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not marketed for greyscale or 3D lithography; the datasheet targets binary lift-off patterning with a tunable undercut sidewall.", "status": "active", "successorSlug": null, "summary": "NR9-1500PY is Futurrex's negative-tone, 365 nm lift-off resist with a negative-sloping (undercut) sidewall profile tunable by exposure dose, coating roughly 1.1-3.0 µm across the manufacturer's published 800-5000 rpm spin range.", "thicknessRange": { "min_um": 1.2, "max_um": 3, "basis": "curve-span", "source": "curve-span — min/max of the published spin curve (Properties table, p.1) over its stated 800-5000 rpm range; no separate achievable-range statement is made in prose. Points are the midpoint of each rpm's published nm range (library convention — see NR71-3000P). Adjudicated 2026-07-12: confirmed correct against the archived PDF; check's 1.14-3.15 µm used the outer tolerance bounds of the published ranges (5000 rpm's lower bound and 800 rpm's upper bound) rather than the per-point midpoints this library's spin-curve convention uses." }, "spinCurves": [ { "label": "NR9-1500PY", "points": [ { "rpm": 800, "um": 3 }, { "rpm": 1000, "um": 2.7 }, { "rpm": 2000, "um": 1.9 }, { "rpm": 3000, "um": 1.5 }, { "rpm": 4000, "um": 1.3 }, { "rpm": 5000, "um": 1.2 } ], "source": "numeric table 'Film thickness after 150°C hotplate bake for 60 s. / Coating spin speed, 40 s spin (rpm): (nm)', p.1 of Futurrex NR9-1500PY Technical Information. The datasheet publishes each point as an nm RANGE, not a single value (e.g. 800 rpm -> 2850-3150 nm); each point above is the exact midpoint of that stated range (every range is symmetric, approx. nominal +/-5%), converted nm->um. Raw ranges: 800rpm=2850-3150nm, 1000rpm=2565-2835nm, 2000rpm=1805-1995nm, 3000rpm=1425-1575nm, 4000rpm=1235-1365nm, 5000rpm=1140-1260nm. The 3000 rpm midpoint (1500 nm = 1.5 um) matches the '1500' in the product name NR9-1500PY, corroborating the midpoint reading." } ], "spinNotes": "Spin coating is at a selected speed for 40 s (Processing step 1). Edge Bead Remover EBR2 is dispensed simultaneously onto top and bottom surfaces through nozzles 0.5-1.0 cm from the substrate edge, starting as soon as edge bead forms (3-5 s after resist dispense ends) and stopping 5 s before spin cycle completion (Processing step 2). Bake times throughout this datasheet assume a good thermal conductor substrate (Si, GaAs); the datasheet instructs multiplying bake times by 3.5 for poor thermal conductors such as glass.", "adhesion": { "hmds": null, "notes": "Not addressed in this datasheet." }, "rehydration": null, "softbake": { "temp_c": 150, "time_s": 60, "method": "hotplate", "notes": "Applies to good thermal conductors (Si, GaAs, etc.); bake times must be multiplied by 3.5 for poor thermal conductors such as glass (no explicit alternate schedule given for this SKU, unlike NR71-3000P's dual schedule).", "source": "Processing step 3, p.2; corroborated by Properties table heading, p.1" }, "exposureDose": { "doses": [ { "wavelength_nm": 365, "value_mJcm2": 190, "source": "Properties, p.1" } ], "datasheetBasis": "'Sensitivity at 365 nm exposure wavelength (mJ/cm² for 1µm thick film) 190' (Properties, p.1); Processing step 4: 'Resist exposure with a tool emitting 365 nm wavelength.'", "_note": "190 mJ/cm² is explicitly normalized to a 1 µm-thick film — the datasheet does not give a single absolute dose for an arbitrary process thickness, nor a scaling formula for other thicknesses, so this is reported as-is. No h-line (405 nm) dose is published anywhere in the document." }, "peb": { "temp_c": 100, "time_s": 60, "notes": "Applies to good thermal conductors (Si, GaAs, etc.); no alternate schedule for glass is given for this SKU.", "source": "Processing step 5, p.2" }, "floodExposure": null, "develop": { "developer": "Resist Developer RD6", "dilution": null, "time_s": 12, "method": null, "rinse": "Deionized water rinse until water resistivity reaches prescribed limit (Processing step 7).", "source": "Processing steps 6-7, p.2", "_note": "12 s is the standard develop time example given 'for 1.5 µm thick film... including overdevelopment'. Datasheet states development is 'by spray or immersion' without choosing one, so method is left null rather than picking one. To extend develop time to 60 s, the datasheet specifies diluting RD6:water 3:1 — a distinct, slower process not modeled as a separate array entry here. developerFamily tmah: RD6 is 2-3% TMAH in water per Futurrex MSDS (classified 2026-07-12)." }, "hardbake": null, "descum": null, "applications": [ "lift-off", "etch-mask" ], "etchResistance": "\"Superior selectivity in RIE process\" (Description, p.1).", "liftoffSuitable": true, "platingSuitable": null, "stripper": "Resist Remover RR5 at room temperature (Processing step 9, p.2).", "storage": "\"Guaranteed shelf life at 25°C storage (years) 3\" (Properties, p.1); \"shelf life exceeding 3 years at room temperature storage\" (Description, p.1).", "notes": "NR9-1500PY is Futurrex's negative-tone, 365 nm lift-off resist, formulated in cyclohexanone and developed in the basic aqueous RD6 developer; degree of undercut is tuned by exposure dose (per the datasheet, 'easy adjustment of the degree of resist undercut as a function of exposure energy') as well as by extending development time via RD6/water dilution. The product code encodes its reference film thickness: '1500' corresponds to the 3000 rpm/40 s coating point (1425-1575 nm, midpoint 1.5 µm) in the published spin table. Bake times given here (150°C softbake, 100°C PEB, both 60 s) assume a substrate that conducts heat well (silicon, GaAs); the datasheet explicitly instructs multiplying bake times by 3.5 on poor thermal conductors such as glass — a detail easy to miss when porting a Si recipe to a transparent substrate. Sensitivity is published as 190 mJ/cm² at 365 nm normalized to a 1 µm-thick film rather than as an absolute dose for whatever process thickness is actually chosen, and no h-line (405 nm) dose is given, so that field is left null.", "developerFamily": "tmah", "references": [ { "type": "paper", "title": "Micromachining on and of Transparent Polymers for Patterning Electrodes and Growing Electrically Active Cells for Biosensor Applications", "authors": "Karnati et al.", "journal": "Micromachines", "year": 2017, "doi": "10.3390/mi8080250", "url": "https://doi.org/10.3390/mi8080250", "accessedDate": "2026-07-12", "summary": "NR9 lift-off gold MEAs on flexible PEN for biosensors" } ], "provenance": { "datasheetUrl": "https://apps.mnc.umn.edu/pub/pds/nr9-1500py.pdf", "datasheetVersionOrDate": null, "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "Document is Futurrex, Inc.'s own 'NEGATIVE RESIST NR9-1500PY / TECHNICAL INFORMATION' product data sheet (2 pages, Futurrex letterhead, Franklin NJ), mirrored by the University of Minnesota Nanofabrication Center (apps.mnc.umn.edu/pub/pds/) — futurrex.com does not publish per-SKU PDFs directly (verified 2026-07-10), so this UMN mirror of Futurrex's own PDS is the primary document read. Content matches the requested SKU (NR9-1500PY) exactly throughout — no filename/content mismatch. No revision code or date is printed anywhere on the document.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "pmgi-sf-6", "name": "PMGI SF6", "manufacturer": "MicroChem", "productLine": "PMGI SF series", "aliases": [ "SF6", "MicroChem SF6", "LOR/PMGI SF6" ], "_seriesNote": "The document's own product name for this grade, as printed in every chart legend and the Product Selection Guide, is the bare code \"SF6\" (no space, no 'PMGI' prefix attached). 'PMGI' is used throughout the document as the umbrella family/chemistry name for the SF-numbered grades (SF2, SF3, SF5, SF6, SF9, SF11, plus a slower-dissolving 'SF Slow' variant), parallel to 'LOR' as the umbrella name for the LOR-numbered grades (LOR 1A/3A/3B/5A/5B/7B/10A/10B/20B/30B) - both families share the same base polymer. 'PMGI SF6' (the assigned name) is therefore a defensible combination of the document's own family name and grade code, not an invented SKU.", "tone": null, "chemistry": "ancillary", "photoimageable": false, "grayscaleSuitable": false, "grayscaleNote": "Not applicable / not addressed. PMGI SF6 is a non-photoimageable ancillary underlayer used for lift-off undercut control - it is never itself exposed or patterned by light, so grayscale/3D-relief processing does not apply to it.", "status": "active", "successorSlug": null, "summary": "PMGI SF6 is a non-photoimageable, polydimethylglutarimide-based ancillary resist from MicroChem's PMGI SF series, used as the undercut/sacrificial layer beneath a conventional imaging resist in bi-layer lift-off processing. It is never exposed; its dissolution rate (and the resulting undercut geometry) is instead controlled by soft-bake temperature.", "thicknessRange": { "min_um": 0.26, "max_um": 0.485, "basis": "curve-span", "source": "curve-span - min/max of the SF6 trace read from the \"Spin Speed vs Thickness - Intermediate Films\" chart, p.5 of the Technical Data section, over its plotted 1000-4000 rpm range." }, "spinCurves": [ { "label": "PMGI SF6", "points": [ { "rpm": 1000, "um": 0.485 }, { "rpm": 1500, "um": 0.39 }, { "rpm": 2000, "um": 0.335 }, { "rpm": 2500, "um": 0.3 }, { "rpm": 3000, "um": 0.28 }, { "rpm": 3500, "um": 0.265 }, { "rpm": 4000, "um": 0.26 } ], "source": "read from figure (\"Spin Speed vs Thickness - Intermediate Films\"), p.5 of the \"LOR and PMGI Resists\" Technical Data section. Five traces share this chart (legend: LOR 7B - open square; LOR 5A - open circle; LOR 5B - x; \"LOR 3A, LOR 3B\" combined - open diamond; SF6 - open triangle). The SF6 trace was identified unambiguously by its own dedicated triangle-marker legend entry, distinct from the explicitly-combined 'LOR 3A, LOR 3B' diamond series (which was NOT used for this recipe per the no-combined-trace rule) and from the other individually-named LOR grades. SF6 is the bottommost (thinnest) of the five traces at every plotted speed and is also the only one of the five sampled at 500 rpm intervals (1000-4000 rpm, 7 points) rather than 1000 rpm intervals (4 points) used for the other four traces on the same chart.", "figureRead": true } ], "spinNotes": "General coating guidance (p.2, applies to the LOR/PMGI line as a whole, not SF6-specific numbers): spin speeds between 2,500 and 4,500 rpm give maximum coating uniformity; higher speeds for smaller substrates, lower for larger or topographically irregular substrates. Recommended Coating Parameters box (p.5, generic to the line): dispense volume 5 ml (150 mm Si wafer), dispense mode dynamic 3-5 s, dispense spin speed 300-500 rpm, acceleration 10,000 rpm/second, terminal spin speed 3,000 rpm, spin time 45 seconds, edge-bead remover MicroChem EBR PG (acetone and conventional-resist edge-bead removers are explicitly NOT recommended with LOR/PMGI). None of this coating-parameter box is stated as SF6-specific - it is a line-wide recommendation.", "adhesion": { "hmds": false, "notes": "\"Primers such as HMDS (hexamethyldisilazane) are typically NOT required to promote adhesion with PMGI/LOR products when used as recommended.\" (Substrate preparation, p.2). Recommended substrate prep instead: solvent clean or dilute-acid rinse, followed by DI water rinse, then a dehydration bake at 200°C for 5 minutes on a contact hotplate or 30 minutes in a convection oven. LOR/PMGI is stated to have superior adhesion to Si, NiFe, GaAs, InP and other III-V materials (p.1 Benefits list)." }, "rehydration": null, "softbake": { "temp_c": null, "time_s": null, "method": "hotplate", "notes": "Recommended bake TEMPERATURE RANGE is 150-200°C (some PMGI products may be baked up to 250°C) - a range, not a single value, so temp_c is left null. Pre-bake (soft-bake) temperature is stated as having the single greatest influence on undercut rate, more so than pre-bake time, the patterning resist's exposure dose, developer choice, develop mode, or develop time. Hot plates are preferred; convection ovens are also compatible. No single recommended temperature or time is stated for SF6 specifically - the document instead recommends a matrix design varying pre-bake temperature and time for process fine-tuning per grade. Separately, Table 1's optical-constant (n/k) measurements for the SF family (which includes SF6) were taken on samples soft-baked at 180°C for 3 minutes, but this is stated as the optical-measurement condition, not a general process recommendation, so it is not used to populate temp_c/time_s either.", "source": "Soft-bake/Prebake Process section, p.3 (\"The recommended bake temperature range is 150°C - 200°C...\"); Table 1 footnote, p.5 (\"Products were soft-baked at 180°C for 3 min\")" }, "exposureDose": null, "peb": { "temp_c": null, "time_s": null, "notes": "\"LOR/PMGI does not require post-exposure baking. Refer to patterning resist manufacturer process recommendations to determine whether a PEB step is required.\" - stated generically for the whole LOR/PMGI line, including the PMGI SF series that SF6 belongs to; consistent with PMGI/LOR never itself being exposed in the standard bi-layer lift-off flow.", "source": "Post - Exposure (PEB) Process section, p.4" }, "floodExposure": null, "develop": { "developer": "Metal-ion-free (MIF) developers, specifically 0.26N and 0.24N MIF per the Product Selection Guide's \"SF\" column (SF6's category); the SF category is NOT starred for metal-ion-bearing (MIB) compatibility in that same guide (only LOR B is). General text elsewhere states the line is \"optimized for use with various metal ion free and metal ion containing developers,\" a broader claim than the SF-specific selection-guide row.", "dilution": "0.26N MIF (2.38% TMAH) and 0.24N MIF (2.2% TMAH w/surfactant) are both marked compatible for the \"SF\" category in the Product Selection Guide (p.6); no single dilution is singled out as SF6's own specific recommendation.", "time_s": null, "method": null, "rinse": null, "source": "Development Process section, p.4; Product Selection Guide (\"Developer Compatibility\" rows, \"SF\" column), p.6", "_note": "No SF6-specific develop time, method, or rinse step is published. The document states develop time depends on the combined thickness of the LOR/PMGI layer AND the overlying patterning (imaging) resist layer (not separable into an SF6-only number), and recommends spray development specifically for straighter sidewalls when the LOR/PMGI layer exceeds 2 µm - a conditional recommendation, not a fixed method assignment. Figure 7 (\"The Effect of Developer Type on Dissolution Rate\", p.4), which plots TMAH-concentration-dependent dissolution rate vs. bake temperature, is explicitly labeled \"SF11\" - a different SF-series grade - so its numbers are not attributed to SF6 here." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Not addressed. LOR/PMGI is coated and soft-baked, then covered by an overlying imaging resist which is exposed and developed (developing both layers together); the process then proceeds directly to metal/dielectric deposition (Deposition Process, p.4) with no separate hardbake step described for the LOR/PMGI layer itself.", "source": null }, "descum": null, "applications": [ "lift-off", "mems-structural" ], "etchResistance": null, "liftoffSuitable": true, "platingSuitable": null, "stripper": "MicroChem Remover PG. \"Use MicroChem's Remover PG to remove the bi-layer resist stack. Removal rate of LOR/PMGI is dependent upon soft-bake temperature of the LOR/PMGI product and remover bath temperature. As a baseline process, use Remover PG in two tanks: at 60°C for 30 minutes in the first tank and rinse at 60°C in the second tank. Ultrasonic action will improve the resist removal efficiency.\" (Lift-Off Process, p.4). A separate category-level chart (Figure 9, p.4, labeled \"SF Series\" - not SF6 individually) shows removal rate in Remover PG rising with bath temperature (40°C bath removes faster than 25°C) and falling as the PMGI's own soft-bake temperature increases (150 -> 180 -> 200°C); exact rates were not extracted here since the chart is not SF6-specific and its bar-scale precision is limited.", "storage": "\"Store upright in original sealed containers in a dry area between 4 and 27°C (40-80°F). Keep away from sources of ignition, light, heat, oxidants, acids, and reducers. Do not use after the expiration date (1 year from date of manufacture).\" (LOR/PMGI Storage, p.7)", "notes": "PMGI SF6 belongs to MicroChem's PMGI SF series (SF2/SF3/SF5/SF6/SF9/SF11, plus a slower-dissolving 'SF Slow' variant), a polydimethylglutarimide-based ancillary layer that is never itself exposed for patterning: in the standard bi-layer lift-off flow it is coated and soft-baked first, then a conventional imaging resist is coated, exposed and developed on top, and the same develop step dissolves an undercut into the PMGI beneath the imaging-resist pattern. Its dissolution rate - and therefore the achievable undercut geometry - is controlled primarily by SOFT-BAKE TEMPERATURE rather than by exposure or develop time; this datasheet's own quantified bake-temperature-vs-undercut-rate curves (Figures 5a/5b) are published only for LOR 10B, not for SF6, so no grade-specific undercut-rate number is reported here - a matrix design varying pre-bake temperature and time is explicitly recommended instead for fine-tuning any given grade. HMDS priming is explicitly NOT required for PMGI/LOR adhesion, and the document notes the overlying imaging resist can be applied directly over PMGI without barrier layers or a plasma descum step. As a PMGI-branded (not LOR-branded) product, SF6 is also compatible with an optional 'Cap-On' process in which the PMGI layer is separately deep-UV (240-290 nm) flood exposed to obtain straighter sidewall profiles - the document ties this option to PMGI generically rather than to SF6 specifically, so it is noted here as available but not confirmed SF6-specific. Stripping uses MicroChem Remover PG (baseline two-tank 60°C/30 min process); actual removal rate depends on the layer's own soft-bake temperature and the remover bath temperature.", "developerFamily": "tmah", "references": [ { "type": "paper", "title": "Robust shadow-mask evaporation via lithographically controlled undercut", "authors": "Cord et al.", "journal": "Journal of Vacuum Science & Technology B: Microelectronics and Nanometer Structures Processing, Measurement, and Phenomena", "year": 2006, "doi": "10.1116/1.2375090", "url": "https://doi.org/10.1116/1.2375090", "accessedDate": "2026-07-12", "summary": "The Dolan-bridge PMMA/PMGI undercut process used worldwide for superconducting-qubit Josephson junctions." } ], "provenance": { "datasheetUrl": "https://apps.mnc.umn.edu/pub/pds/lor.pdf", "datasheetVersionOrDate": "Rev. A (printed bottom-right of p.7)", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "kayakuam.com (the current manufacturer/brand owner's own site) returns HTTP 403 to automated fetches (verified 2026-07-10); this document is a MIRROR of the manufacturer's own \"LOR and PMGI Resists\" datasheet (MCC / MicroChem, Rev. A), hosted as a PDF by the University of Minnesota's Minnesota Nano Center (apps.mnc.umn.edu), an academic nanofabrication facility, not by MicroChem/Kayaku themselves.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-10", "humanVerified": false }, { "slug": "s1813", "name": "S1813", "manufacturer": "Kayaku Advanced Materials (Microposit brand; datasheet originally published by Rohm and Haas Electronic Materials / later Dow Electronic Materials)", "productLine": "MICROPOSIT S1800 G2 series", "aliases": [ "Microposit S1813", "S1813 G2", "Microposit S1813 G2 Photoresist" ], "tone": "positive", "chemistry": "dnq-novolak", "grayscaleSuitable": false, "grayscaleNote": "Not addressed in this datasheet; it is presented as a standard single-tone (binary) positive photoresist for IC masking, with no grayscale/3D process information given.", "status": "active", "successorSlug": null, "summary": "MICROPOSIT S1813 is the mid-film, general-purpose workhorse of the MICROPOSIT S1800 G2 series — thicker-coating than S1805, thinner than S1818, and the one grade this datasheet fully characterizes, which is why most cleanrooms keep it as their default g-line positive resist. MICROPOSIT S1813 (S1800 G2 series) is a classic positive g-line photoresist for general IC device microlithography, coating to roughly 1.2-1.4 µm and resolving sub-micron lines/spaces (down to 0.48 µm demonstrated). It is one of the most widely used general-purpose cleanroom resists and is compatible with both metal-ion-free (MIF, e.g. MF-319/MF-321) and metal-ion-bearing developers.", "thicknessRange": { "min_um": 1.03, "max_um": 1.9, "basis": "curve-span", "source": "curve-span: adjudicated 2026-07-12 — the prior 'stated' basis (1.23-1.37 µm) misrepresented the datasheet. 12,300 Å (Table 1, p.1; also Tables 5/7/8) and 13,675 Å (Table 3, p.2) are two UNRELATED single reference coat thicknesses used as the fixed film for other test figures (masking linearity SEMs, the dispersion-curve measurement) — not a stated achievable thickness range, and not even paired with a spin speed. The datasheet publishes no prose achievable-range statement for S1813 G2. The real achievable range is the S1813 G2 trace (triangle markers) in Figure 2, 'MICROPOSIT S1800 G2 Photoresist Undyed Series Spin Speed Curves', p.2 — min/max are the span of this recipe's own spin curve, pixel-calibrated 2026-07-12 (PyMuPDF vector-path marker centroid extraction, axes calibrated from tick-label text baselines): 1.90 µm at 2,000 rpm to 1.03 µm at 7,000 rpm — the trace's actual plotted x-range, confirmed by vector extraction to run 2,000-7,000 rpm only, NOT 1,000-8,000 rpm (that wider span is the chart's axis range, not the S1813 trace's own plotted extent — check's original 0.75-2.7 µm mistakenly used the axis range)." }, "spinCurves": [ { "label": "S1813 G2", "points": [ { "rpm": 2000, "um": 1.9 }, { "rpm": 3000, "um": 1.55 }, { "rpm": 4000, "um": 1.33 }, { "rpm": 5000, "um": 1.21 }, { "rpm": 6000, "um": 1.11 }, { "rpm": 7000, "um": 1.03 } ], "source": "re-extracted 2026-07-12, pixel-calibrated via PyMuPDF vector-drawing extraction (the chart's markers and axis tick-labels are native PDF vector/text objects; the S1813 G2 triangle marker centroid at each of the 6 rpm columns was located directly from the page's vector drawing commands and identified by its vertical rank — 2nd of 4 stacked series, diamond/S1818 above, square/S1811 and x/S1805 below, matching the legend order — then converted to Å via a y-axis calibration fit from the gridline tick-label text baselines: 280.11 Å/pt, exact and linear across all 9 gridlines). 'MICROPOSIT S1800 G2 Photoresist Undyed Series Spin Speed Curves' (Figure 2), p.2 of MICROPOSIT S1800 G2 Series Photoresists datasheet (Rohm and Haas Electronic Materials, ME06N041, Rev. 0, October 2006). The trace's plotted markers run 2,000-7,000 rpm only (6 points) — confirmed by vector extraction, not the chart's wider 1,000-8,000 rpm axis range. No rpm-paired numeric anchor exists in the document — the two printed reference thicknesses (12,300 Å and 13,675 Å) are coat conditions used for OTHER test figures (masking linearity, dispersion curve), not points on this spin curve. Supersedes the earlier eyeball read, which under-read the mid-to-low-rpm points by 5-12%.", "figureRead": true } ], "spinNotes": "The datasheet publishes a spin-speed FIGURE (Figure 2, p.2, \"MICROPOSIT S1800 G2 Photoresist Undyed Series Spin Speed Curves\") plotting film thickness (0-40,000 Å) vs. spin speed (1,000-8,000 rpm axis range) for four grades (S1805, S1811, S1813, S1818 G2) on one chart, coated on an SVG 81 with a 115°C/60s softbake. The S1813 G2 trace (triangle marker, second-highest of the four, which do not cross across the plotted range) was pixel-calibrated 2026-07-12 by vector-extracting each marker centroid at its 6 plotted rpm columns (2,000-7,000 rpm — the trace's own plotted extent, narrower than the chart's 1,000-8,000 rpm axis range) and converting to Å via a linear fit off the axis tick-label baselines (see spinCurves). RESOLVED 2026-07-12 (was previously flagged for QC): thicknessRange is now the span of this corrected curve (1.03-1.90 µm), not the two unrelated reference coat thicknesses (12,300 Å / 13,675 Å) previously used, which are single fixed-film conditions for OTHER test figures (masking linearity, exposure latitude, dispersion curve) and were never a stated achievable-range statement — see thicknessRange.source. The datasheet does state maximum coating uniformity is typically attained between 3,500-5,500 rpm. The two reference thicknesses still recur across the process-condition tables: 12,300 Å in Tables 1, 4, 5, 7, 8, and 13,675 Å in Table 3.", "adhesion": { "hmds": true, "notes": "Works well with hexamethyldisilazane (HMDS)-based MICROPOSIT Primers. Concentrated MICROPOSIT Primer is recommended for vacuum vapor priming; diluted primer is recommended for liquid-phase priming applications." }, "rehydration": null, "softbake": { "temp_c": 115, "time_s": 60, "method": "hotplate", "notes": "Consistent across all process-condition tables in the datasheet (Tables 1-8).", "source": "e.g. Table 1, p.1: \"Softbake 115°C/60 sec. Hotplate\"" }, "exposureDose": { "doses": [ { "wavelength_nm": 436, "value_mJcm2": 150, "source": "Table 9 (\"Lithographic Responses Summary\"), p.4: \"Sizing Energy 150 mJ/cm2 (1.3 E0)\"; E0 = 82 mJ/cm2 from Figure 6 contrast curve, p.4" } ], "datasheetBasis": "Sizing (\"E-size\") exposure dose measured on g-line (436 nm) exposure tools (GCA 8500 g-Line 0.35 NA; Nikon 1505 G6E g-Line 0.54 NA), reported as 150 mJ/cm² = 1.3 x E0, where E0 = 82 mJ/cm² is the resist's clearing dose from the contrast curve (Figure 6, Table 9). Resist is exposable over 350-450 nm broadly, with exposure properties optimized for 436 nm. A separate Table 6 gives Dill optical parameters (A, B absorption coefficients, not a dose) at both 365 nm and 436 nm for use in lithography simulators — these are not exposure doses and were not used to populate at365_mJcm2/at405_mJcm2.", "_note": "The 150 mJ/cm² sizing energy is explicitly tied to g-line (436 nm) exposure equipment in Table 9 and the surrounding text, not to i-line (365 nm) or h-line (405 nm), so at365_mJcm2 and at405_mJcm2 are left null rather than assumed equal or scaled." }, "peb": { "temp_c": null, "time_s": null, "notes": "Not addressed anywhere in this datasheet. The described process flow is softbake -> expose -> develop with no post-exposure bake step mentioned.", "source": null }, "floodExposure": { "dose_mJcm2": null, "notes": "Not applicable / not addressed - S1813 is a standard single-exposure positive photoresist in this datasheet, with no image-reversal or flood-exposure step described.", "source": null }, "develop": { "developer": "MICROPOSIT MF-319 metal-ion-free (MIF) developer family (stated as optimized for); also compatible with metal-ion-bearing (MIB) MICROPOSIT developers. Individual test figures cite MF-31 (i.e. MF-319) and MF-321 by name.", "dilution": "Test conditions in Tables 4, 7 and 8 use MF-319/MF-321 diluted 1:10 (\"MF-31/10\", \"MF-321/10\"); Table 1 does not name a specific developer or dilution.", "time_s": null, "method": "Double Spray Puddle (DSP) at 21°C in all cited process-condition tables", "rinse": null, "source": "e.g. Table 1, p.1: \"Develop 15 +50 sec. Double Spray Puddle (DSP) @ 21°C\"; Tables 4, 7, 8, p.3-4: \"MF-31/10 +30 DSP @ 21°C\" / \"MF-321/10 +30 DSP @ 21°C\"", "_note": "Develop time varies between the datasheet's own example process conditions (15+50 sec in Table 1 vs. +30 sec in Tables 4/7/8) and no single time is presented as THE standard develop time, so time_s is left null rather than picking one arbitrarily. Rinse medium (e.g. DI water) is not stated anywhere in this document." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Not addressed in this datasheet.", "source": null }, "descum": null, "applications": [ "general-prototyping", "etch-mask" ], "etchResistance": null, "liftoffSuitable": false, "platingSuitable": false, "stripper": "Datasheet states only that \"residue-free photoresist removal\" is achieved \"using standard MICROPOSIT removers\", without naming a specific product.", "storage": "\"Store products in tightly closed original containers at temperatures recommended on the product label\" (no specific temperature is given in the datasheet text itself). Shelf life/expiry is not otherwise quantified here.", "notes": "S1813 is a standard positive DNQ/novolak-class g-line resist (chemistry classification based on well-established industry knowledge of this product family; this specific TDS does not itself restate the chemical composition). Reach for S1813 when you want the series' balanced, best-documented option rather than pushing to the thin (S1805) or thick (S1818) ends of the family — it is the one grade whose exposure, contrast and develop behaviour the manufacturer actually measured in this datasheet. Its ~85 deg wall angle (Table 9) and single-tone positive process (no reversal/undercut step described) make it unsuited to lift-off without an added process (e.g. bilayer or image-reversal), and this datasheet makes no lift-off or electroplating claims. The published spin-speed figure (Figure 2) could not be digitized reliably from text extraction, so no spinCurves points are reported - only the two explicit reference coating thicknesses (12,300 A and 13,675 A) used across its process-condition tables. Develop time and post-exposure/hardbake steps are largely unaddressed or inconsistent across the datasheet's own example tables, and are left null rather than guessed.", "developerFamily": "tmah", "references": [ { "type": "paper", "title": "Mechanical Properties of Microposit S1813 Thin Layers", "authors": "Nikolaev et al.", "journal": "Advanced Structured Materials", "year": 2020, "doi": "10.1007/978-3-030-48161-2_9", "url": "https://doi.org/10.1007/978-3-030-48161-2_9", "accessedDate": "2026-07-12", "summary": "Nanoindentation of S1813 films (a book/proceedings chapter, not a standalone journal article)." }, { "type": "paper", "title": "Reduced Etch Lag and High Aspect Ratios by Deep Reactive Ion Etching (DRIE)", "authors": "Gerlt et al.", "journal": "Micromachines", "year": 2021, "doi": "10.3390/mi12050542", "url": "https://doi.org/10.3390/mi12050542", "accessedDate": "2026-07-12", "summary": "1.4 µm S1813 used as a DRIE etch mask, reporting a Si:resist selectivity of roughly 22:1." }, { "type": "paper", "title": "Fabrication of Large-Area Silicon Spherical Microlens Arrays by Thermal Reflow and ICP Etching", "authors": "Wu et al.", "journal": "Micromachines", "year": 2024, "doi": "10.3390/mi15040460", "url": "https://doi.org/10.3390/mi15040460", "accessedDate": "2026-07-12", "summary": "Reflowed S1813 microlenses transferred into a 128x128 silicon microlens array with roughly 1.1 nm surface roughness." } ], "provenance": { "datasheetUrl": "https://www.nanophys.kth.se/nanolab/resists/S1813/Microposit_S1800_G2_Serie.pdf", "datasheetVersionOrDate": "ME06N041, Rev. 0, October 2006 (Rohm and Haas Electronic Materials)", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://kayakuam.com/products/microposit-s1800-g2-series-photoresists/", "what": "Confirms Kayaku Advanced Materials currently lists/sells the MICROPOSIT S1800 G2 series (used only to support 'status: active'); the page itself returned HTTP 403 to automated fetch during this session, so its content was not used for any numeric field." } ] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "s1822", "name": "S1822", "manufacturer": "Shipley Company", "productLine": "MICROPOSIT S1800 series", "aliases": [ "MICROPOSIT S1822", "MICROPOSIT S1822 PHOTO RESIST" ], "tone": "positive", "chemistry": "dnq-novolak", "_chemistryNote": "The document identifies the whole S1800 series only as a 'positive photoresist system' and names the casting solvent (propylene glycol monomethyl ether acetate); it never states the photoactive-compound / resin chemistry class (e.g. novolak, DNQ) anywhere in this edition, so chemistry is left null rather than assumed from general S1800-family knowledge.", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not discussed anywhere in the document. S1800 is positioned as a standard positive resist for fine line/space IC masking (down to 0.48 µm in the worked S1813 example); no grayscale or 3-D patterning claim is made for the series or for S1822 specifically.", "status": "discontinued", "successorSlug": null, "summary": "MICROPOSIT S1822 was the highest-viscosity, thickest-coating member of Shipley's original (1993) undyed MICROPOSIT S1800 series of positive photoresists, optimized for G-line (436 nm) exposure with broadband compatibility. Within that five-grade lineup it sat at the very top of the spin-speed chart — coating thicker than even S1818 — so it was the grade to reach for when a single S1800 coat had to be as heavy as the family allowed; the later G2 edition drops it, which is why its current availability needs a human check.", "thicknessRange": null, "spinCurves": [ { "label": "S1822", "points": [ { "rpm": 2000, "um": 3.2 }, { "rpm": 3000, "um": 2.5 }, { "rpm": 4000, "um": 2.1 }, { "rpm": 5000, "um": 1.85 }, { "rpm": 6000, "um": 1.65 }, { "rpm": 7000, "um": 1.5 } ], "source": "read from figure, 'MICROPOSIT S1800 PHOTO RESIST UNDYED SERIES, Spin Speed Curves' (Figure 1), p.2 of MICROPOSIT S1800 Series Photo Resists datasheet (Shipley Company, document code MPR S1800 1093, 1993 edition); identified as the topmost, thickest of five plotted traces (open-square marker, shared shape with S1811 — disambiguated by curve position, since S1822 is listed first in the legend and, consistent with it being the highest-viscosity/thickest-film grade in the family, plots as the uppermost curve at every spin speed with no crossings against S1818/S1813/S1811/S1805); read at gridline/marker crossings every 1,000 rpm from 2,000–7,000 rpm off a linear-linear chart (thickness 0–40,000 Å, spin speed 1,000–8,000 rpm); no numeric table or rpm-paired anchor is published anywhere else in the document for S1822 — visual chart-reading only, typical uncertainty ~10–15%; digitized 2026-07-12", "figureRead": true } ], "spinNotes": "Figure 1 ('MICROPOSIT S1800 PHOTO RESIST UNDYED SERIES, Spin Speed Curves', p.2) plots five undyed grades — S1822, S1818, S1813, S1811, S1805 — photoresist thickness (Å) vs spin speed (1000–8000 rpm). S1822 is identified as the topmost, thickest trace of the five. The legend reuses the same open-square marker for both S1822 and S1811, so marker shape alone cannot distinguish them; disambiguation here is by curve position — S1822 is listed first in the legend and, consistent with it being the highest-viscosity/thickest-film grade in the family (the curve ordering top-to-bottom is S1822 > S1818 > S1813 > S1811 > S1805), it plots as the uppermost, thickest curve on the chart. The trace was digitized 2026-07-12 by reading gridline/marker crossings at 1,000 rpm intervals from 2,000–7,000 rpm (see spinCurves), spanning roughly 3.20 µm (2,000 rpm) down to 1.50 µm (7,000 rpm); no numeric table or rpm-paired anchor is published anywhere else in the document for S1822, so the digitized values carry typical chart-reading uncertainty (~10–15% — see spinCurves[0].source). No single-point anchor (e.g. 'X µm at Y rpm') is stated anywhere else in the document for S1822. Softbake for the wafers used to generate Figures 1 and 2 was 115°C/60 s hotplate (Process Parameters box, p.2, keyed 'Refer to Figures 1 and 2') — this is the test condition tied to the figure that includes S1822, not a grade-specific recommendation, and is recorded separately below under softbake.", "adhesion": { "hmds": true, "notes": "\"MICROPOSIT S1800 SERIES PHOTO RESISTS work well with the hexamethyldisilazane based MICROPOSIT PRIMERS. Concentrated MICROPOSIT PRIMER is recommended when vacuum vapor priming. Diluted PRIMER is recommended for liquid phase priming applications.\" (Substrate Preparation, p.2 — series-wide statement covering S1822.)" }, "rehydration": null, "softbake": { "temp_c": 115, "time_s": 60, "method": "hotplate", "notes": "This is the coat/softbake condition printed in the 'Process Parameters (Refer to Figures 1 and 2)' box on p.2, i.e. the condition used to generate the undyed-series spin-speed chart that includes S1822 (Substrate: Silicon, Coat: SVG 81, Measure: Nanometrics 210). It is a documented test condition for the chart, not an explicit grade-specific recommendation for S1822.", "source": "Process Parameters box (\"Refer to Figures 1 and 2\"), p.2" }, "exposureDose": { "doses": [], "datasheetBasis": "\"MICROPOSIT S1800 SERIES PHOTO RESISTS can be exposed with light sources in the spectral output range of 350 nm - 450 nm. The exposure properties have been optimized for use at 436 nm.\" (Exposure section, p.3 — series-wide statement, g-line/broadband, not a specific dose number.)", "_note": "No S1822-specific dose is published anywhere in this document. The only numeric sizing energy in the document — 150 mJ/cm² (1.3 E0), Table 2, p.4, and the identical 150 mJ/cm² in the p.1 process box — is explicitly attributed to MICROPOSIT S1813, not S1822, and must not be borrowed. Table 1's Dill parameters (365 nm and 436 nm A/B coefficients, p.3) likewise cover only S1813, S1813 D1, S1811 J2 and S1818 J1 — S1822 is absent from that table. The contrast-curve E0 = 82 mJ/cm² (Figure 7, p.4) is also S1813-specific." }, "peb": null, "floodExposure": null, "develop": { "developer": "MICROPOSIT MF-319 Metal-Ion-Free (MIF) DEVELOPER family", "dilution": null, "time_s": null, "method": null, "rinse": null, "source": "Develop Properties, feature list, p.1 (\"Optimized for use with the MICROPOSIT MF-319 Metal-Ion-Free DEVELOPER family\" / \"Compatible with Metal-Ion-Bearing MICROPOSIT DEVELOPERS\"); DEVELOP section, p.4", "_note": "Series-wide developer family statement only. No S1822-specific dilution, develop time, or method is published — every numeric develop example in this document (\"MF-321 /10 + 30 DSP @ 21°C\", \"MF-321 /15 + 50 DSP @ 21°C\") is tied explicitly to MICROPOSIT S1813 process-parameter boxes (pp.1, 3–4), not S1822, and is not carried over here. The DEVELOP section itself (p.4) also just refers readers to a Shipley technical sales representative rather than stating a number. developerFamily tmah: MF-319 is 2.2% TMAH in water per Rohm & Haas MSDS (classified 2026-07-12)." }, "hardbake": null, "descum": null, "applications": [], "etchResistance": null, "liftoffSuitable": null, "platingSuitable": null, "stripper": "Standard MICROPOSIT REMOVERS — \"Residue-free photoresist removal using standard MICROPOSIT REMOVERS\" (Removal Property, feature list, p.1; series-wide, no specific remover product named).", "storage": "Store MICROPOSIT S1800 PHOTO RESISTS only in upright, original containers in a dry area at 50°-70°F (10°-21°C). Store away from light, oxidants, heat, and sources of ignition. Do not store in sunlight. Keep container sealed when not in use. (Storage section, p.5 — series-wide statement covering S1822.)", "notes": "S1822 was originally sold by Shipley Company — later Rohm and Haas Electronic Materials, then Dow Electronic Materials, then a Dow/DuPont-linked joint venture, and eventually Kayaku Advanced Materials, as the industry consolidated — as the thickest-film member of the undyed MICROPOSIT S1800 series. In grade-selection terms it was the pick when even S1818 could not lay down enough film in one coat; anyone specifying that film thickness today should confirm current availability and look to the surviving S1800 G2 grades, since the 2006 G2 edition omits S1822 entirely. This 1993 edition (document code \"MPR S1800 1093\") is, as far as this project has located, the only Shipley/Rohm-and-Haas S1800 datasheet that documents S1822 at all: the later 2006 \"S1800 G2 Series\" edition (ME06N041 Rev. 0) lists only S1805/S1811/S1813/S1818 in its undyed-series legend and omits S1822 entirely — which is why this superseded edition is retained here, and why S1822's status is flagged discontinued pending a human check of current availability. Nearly every quantitative process parameter in this document — exposure dose, Dill optical parameters, contrast curve, masking-linearity, exposure-latitude and focus-latitude data — is explicitly tied to worked examples for MICROPOSIT S1813, not S1822; those numbers must not be borrowed into S1822's own process window. Only series-wide statements (G-line/broadband exposure range, HMDS-based priming, MF-319-family MIF developer compatibility, storage conditions) and S1822's position as the topmost, thickest trace in the undyed spin-speed chart (Figure 1) can be attributed to this grade from this source. Like other thick-film DNQ-family positive resists of this era, coatings at this viscosity are the ones most likely to need an edge-bead removal step and a rehydration/relaxation wait after spin-coating before bake — but neither is mentioned in this document for S1822, so both are left null rather than assumed. Chemistry classified as dnq-novolak from Kayaku's S1800 G2 series-wide statement and the Dow/Rohm and Haas S1818 MSDS composition table (mixed cresol novolak resin + diazo photoactive compound) (chemistry classified 2026-07-12 from manufacturer SDS/TDS).", "developerFamily": "tmah", "provenance": { "datasheetUrl": "https://amolf.nl/wp-content/uploads/2016/09/datasheets_S1800.pdf", "datasheetVersionOrDate": "MPR S1800 1093 (printed at the foot of p.1, bottom-right corner)", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://www.nanophys.kth.se/nanolab/resists/S1813/Microposit_S1800_G2_Serie.pdf", "what": "The newer 2006 Rohm and Haas 'S1800 G2 Series' datasheet (ME06N041 Rev. 0), checked and found to NOT list S1822 in its undyed-series legend (only S1805/S1811/S1813/S1818 appear); cited for comparison to justify retaining this superseded 1993 edition as S1822's only located source, not as a data source for S1822 itself." }, { "url": "https://kayakuam.com/products/microposit-s1800-g2-series-photoresists/", "what": "Kayaku's MICROPOSIT S1800 G2 series page states the series-wide formulation (mixed cresol novolak resin + diazo photoactive compound); used as the basis for classifying S1822 as dnq-novolak, corroborated by the Dow/Rohm and Haas S1818 MSDS composition table." } ] }, "_provenanceNote": "kayakuam.com (the current rights-holder's own site) 403s automated fetches (verified 2026-07-10). This document is an AMOLF-hosted mirror of the manufacturer's own S1800 datasheet, not a third-party summary. It is the ~1993 Shipley edition (\"MPR S1800 1093\"), which is superseded by the 2006 Rohm and Haas \"S1800 G2\" edition (ME06N041 Rev. 0). The newer G2 edition was checked and does not mention S1822 at all — its undyed-series legend lists only S1805/S1811/S1813/S1818 — so this legacy edition is retained here as the only located document that characterizes this grade. A QC reviewer must decide whether S1822 is still a current product before this page ships.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "su-8-2025", "name": "SU-8 2025", "manufacturer": "MicroChem", "productLine": "SU-8 2000 series", "aliases": [ "SU8 2025", "MicroChem SU-8 2025" ], "tone": "negative", "chemistry": "epoxy", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed by the datasheet. The document presents SU-8 2000 exclusively for binary, high-aspect-ratio, vertical-sidewall permanent structures (\"high aspect ratio imaging\", \"vertical sidewalls\"); it makes no mention of grayscale or partial-exposure profiling.", "status": "active", "successorSlug": null, "summary": "SU-8 2025 is the thin-film choice within the SU-8 2000 epoxy family — reached for when a high-aspect-ratio permanent structure needs a coat in the tens-of-microns range rather than the far thicker films its higher-viscosity siblings (SU-8 2050 and 2100) deliver. SU-8 2025 is a mid-viscosity member of MicroChem's SU-8 2000 epoxy negative photoresist series, coating roughly 20-80 µm in a single pass for high-aspect-ratio, permanent microstructures.", "thicknessRange": { "min_um": 21.7, "max_um": 79.2, "basis": "curve-span", "source": "curve-span: the document states no single achievable-thickness range for SU-8 2025 specifically (only thickness-binned process tables shared across all four grades in this doc); the range here is the min/max of the SU-8 2025 curve in Figure 1 (4000 rpm to 1000 rpm), pixel-calibrated 2026-07-12 (PyMuPDF axis-tick and vector marker-path extraction). Adjudicated 2026-07-12: confirmed against the archived high-resolution chart image; check's stale 0.5/null was not derived from this grade's curve and has been corrected to match." }, "spinCurves": [ { "label": "SU-8 2025", "points": [ { "rpm": 1000, "um": 79.2 }, { "rpm": 2000, "um": 40.8 }, { "rpm": 3000, "um": 28.2 }, { "rpm": 4000, "um": 21.7 } ], "source": "re-extracted 2026-07-12, pixel-calibrated (PyMuPDF: axis-tick word coordinates for calibration, vector marker path rects for data points); Figure 1 \"SU-8 2000 Spin Speed versus Thickness\", p.2 of MicroChem \"SU-8 2000 Permanent Epoxy Negative Photoresist Processing Guidelines for SU-8 2025, SU-8 2035, SU-8 2050 and SU-8 2075\" (AMOLF mirror). Chart plots four curves (SU-8 2075/circle, 2050/triangle, 2035/diamond, 2025/square) each with only 4 markers at 1000/2000/3000/4000 rpm. The SU-8 2025 curve was identified as the square-marker series and the lowest (thinnest-film) of the four curves at every rpm, consistent with (a) its legend position (listed last: 2075, 2050, 2035, 2025) and (b) Table 1's viscosity ordering (2025 = 4500 cSt, the lowest of the four, so it should coat thinnest at a given speed — matching the bottom curve). Supersedes the earlier eyeball read (80/41/28/22), which was already close but imprecise.", "figureRead": true } ], "spinNotes": "Recommended program (same for all grades in this document): dispense 1 ml of resist per inch (25 mm) of substrate diameter; spin at 500 rpm for 5-10 s at 100 rpm/s acceleration, then spin at the target speed (per Figure 1) for 30 s at 300 rpm/s acceleration. Edge bead removal (EBR) with MicroChem's EBR PG solvent stream at the wafer edge is recommended before soft bake, both to limit hotplate contamination and to let the photomask reach close contact with the wafer. Source: \"Coat\" / \"Recommended Program\" / \"Edge Bead Removal (EBR)\", p.2.", "adhesion": { "hmds": false, "notes": "\"Adhesion promoters are typically not required.\" HMDS pretreatment (MCC Primer 80/20) is recommended only \"for applications that include electroplating.\" Source: \"Substrate Preparation\", p.2." }, "rehydration": null, "softbake": { "temp_c": null, "time_s": null, "method": "hotplate", "notes": "Published only as a THICKNESS-BINNED table shared by all four grades covered in this document (SU-8 2025/2035/2050/2075), not a single value per grade: 25-40 µm → 0-3 min @65°C then 5-6 min @95°C; 45-80 µm → 0-3 min @65°C then 6-9 min @95°C; 85-110 µm → 5 min @65°C then 10-20 min @95°C; 115-150 µm → 5 min @65°C then 20-30 min @95°C; 160-225 µm → 7 min @65°C then 30-45 min @95°C. SU-8 2025's own spin curve spans roughly 22-80 µm (1000-4000 rpm), i.e. mostly the 25-40 and 45-80 µm bins. Convection ovens are explicitly not recommended (can skin over and trap solvent). A cool-down/re-heat 'wrinkle' check is described to confirm the film is fully dry.", "source": "Table 2 \"Soft Bake Times\", p.2 of the MicroChem SU-8 2000 (2025-2075) Processing Guidelines" }, "exposureDose": { "doses": [], "datasheetBasis": "\"SU-8 2000 photoresist is most commonly exposed with conventional UV (350-400 nm) radiation, although i-line (365 nm) is the recommended wavelength. SU-8 2000 may also be exposed with e-beam or x-ray radiation.\" (Processing Guidelines, p.1). The dose table itself (Table 3) is not separately re-attributed to a specific wavelength beyond this general statement.", "_note": "Dose is published only as a THICKNESS-BINNED range shared across all four grades, not a single scalar for SU-8 2025: 25-40 µm → 150-160 mJ/cm²; 45-80 µm → 150-215 mJ/cm²; 85-110 µm → 215-240 mJ/cm²; 115-150 µm → 240-260 mJ/cm²; 160-225 µm → 260-350 mJ/cm². SU-8 2025's own process (22-80 µm) falls mostly in the 150-160 and 150-215 mJ/cm² bins. Relative dose also scales with substrate (Table 4): silicon 1X, glass/Pyrex/ITO 1.5X, most metals (Au, Al, NiFe, Cu, Ni, Ti) and silicon nitride 1.5-2X. Using a >350 nm long-pass filter (recommended for vertical sidewalls) requires ~40% more exposure time to reach the same effective dose. Left null rather than picking a bin midpoint, per the range rule." }, "peb": { "temp_c": null, "time_s": null, "notes": "Also thickness-binned across the whole grade family: 25-40 µm → 1 min @65°C (optional stress-reduction step) then 5-6 min @95°C; 45-80 µm → 1-2 min @65°C then 6-7 min @95°C; 85-110 µm → 2-5 min @65°C then 8-10 min @95°C; 115-150 µm → 5 min @65°C then 10-12 min @95°C; 160-225 µm → 5 min @65°C then 12-15 min @95°C. After 1 minute of PEB at 95°C a latent mask image should already be visible; if not, exposure and/or heating was insufficient.", "source": "Table 5 \"Post Exposure Bake Times\", p.3" }, "floodExposure": null, "develop": { "developer": "SU-8 Developer (MicroChem)", "dilution": null, "time_s": null, "method": "immersion", "rinse": "IPA", "source": "\"Development\" and \"Rinse and Dry\" sections plus Table 6 \"Development Times for SU-8 Developer\", p.3-4", "_note": "Development is designed for immersion, spray, or spray-puddle processing with SU-8 Developer (other solvent developers such as ethyl lactate and diacetone alcohol are also usable per the text, but no dilution ratio is given for any of them). Immersion times are thickness-binned, not a single value: 25-40 µm → 4-5 min; 45-75 µm → 5-7 min; 80-110 µm → 7-10 min; 115-150 µm → 10-15 min; 160-225 µm → 15-17 min. Rinse: ~10 s fresh developer spray/wash, then ~10 s IPA spray/wash, then dry with filtered N2/air. A white film after IPA rinse indicates underdevelopment and calls for another develop+rinse cycle." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Hard bake (cure) is optional and generally only needed if the finished device will see thermal processing in use; recommended final bake temperature is 10°C above the maximum expected device operating temperature. Typical range: 150-250°C for 5-30 minutes depending on degree of cure required. Separately, a short 150°C bake \"for a couple of minutes\" is recommended (all thicknesses) specifically to anneal any surface cracks seen after development.", "source": "\"Hard Bake (cure)\" section, p.4" }, "descum": null, "applications": [ "high-aspect-ratio", "mems-structural", "electroplating-molding" ], "etchResistance": null, "liftoffSuitable": false, "platingSuitable": true, "stripper": "MicroChem Remover PG swells and lifts only minimally cross-linked SU-8 2000. A fully cured/hard-baked film cannot be removed with Remover PG alone — it requires an OmniCoat (30-100 nm) sacrificial underlayer (heat Remover PG to 50-80°C, immerse 30-90 min) or an oxidizing strip (piranha etch, plasma ash, RIE, laser ablation, or pyrolysis). RIE recipe given: 200 W, 80 sccm O2, 8 sccm CF4, 100 mTorr, 10°C. Source: \"Removal\" / \"Plasma Removal\", p.5.", "storage": "Store upright in tightly closed containers, cool and dry, away from direct sunlight, at 40-70°F (4-21°C); away from light, acids, heat, and ignition sources. Shelf life is twelve months from date of manufacture. Source: \"Storage\", p.5.", "notes": "SU-8 2025 is the lowest-viscosity grade covered by this processing document (4500 cSt vs. up to 22,000 cSt for SU-8 2075) and spin-coats to roughly 20-80 µm depending on speed. Pick 2025 when your target film sits in the low tens of microns and you want the shortest bakes and best feature fidelity the 2000 family offers at that thickness; step up to SU-8 2050 or 2100 for progressively thicker single coats. Like every SU-8 2000 grade, it cross-links in two stages — exposure generates acid, and the post-exposure bake thermally drives the epoxy cross-linking — so PEB is a required processing step, not an optional cure. Soft-bake, PEB, dose and develop times are published only as thickness-binned ranges shared across the whole 2025-2075 family; treat the low end of the matching bin as a starting point and use the datasheet's cool-down/re-heat 'wrinkle' test to confirm the soft bake is complete. Once fully cross-linked, SU-8 is notoriously hard to strip: plain solvent remover only works on minimally exposed/baked film, and a hard-baked structure needs either a sacrificial OmniCoat layer beneath it or an oxidizing strip (piranha, plasma ash, RIE, laser ablation, pyrolysis) to remove. Edge bead removal before soft bake is recommended for every grade in this series to keep the photomask in close contact with the wafer and preserve resolution and aspect ratio in thick films.", "developerFamily": "solvent", "references": [ { "type": "paper", "title": "Negative photoresists for optical lithography", "authors": "Shaw et al.", "journal": "IBM Journal of Research and Development", "year": 1997, "doi": "10.1147/rd.411.0081", "url": "https://doi.org/10.1147/rd.411.0081", "accessedDate": "2026-07-12", "summary": "The IBM origin paper of SU-8." }, { "type": "paper", "title": "SU-8: a low-cost negative resist for MEMS", "authors": "Lorenz et al.", "journal": "Journal of Micromechanics and Microengineering", "year": 1997, "doi": "10.1088/0960-1317/7/3/010", "url": "https://doi.org/10.1088/0960-1317/7/3/010", "accessedDate": "2026-07-12", "summary": "15:1 aspect ratios in 1997 — the paper that made SU-8 a MEMS resist." }, { "type": "paper", "title": "SU-8: a photoresist for high-aspect-ratio and 3D submicron lithography", "authors": "del Campo et al.", "journal": "Journal of Micromechanics and Microengineering", "year": 2007, "doi": "10.1088/0960-1317/17/6/R01", "url": "https://doi.org/10.1088/0960-1317/17/6/R01", "accessedDate": "2026-07-12", "summary": "The canonical SU-8 review." }, { "type": "paper", "title": "Microfluidic probes for use in life sciences and medicine", "authors": "Qasaimeh et al.", "journal": "Lab Chip", "year": 2013, "doi": "10.1039/C2LC40898H", "url": "https://doi.org/10.1039/C2LC40898H", "accessedDate": "2026-07-12", "summary": "Free-standing SU-8 microfluidic probes for live-cell chemistry." } ], "provenance": { "datasheetUrl": "https://amolf.nl/wp-content/uploads/2016/09/datasheets_SU-82000DataSheet2025thru2075Ver4.pdf", "datasheetVersionOrDate": "No revision/date string is printed in the document body (a 5-page processing-guidelines PDF with no header/footer revision code). The mirror's filename indicates 'Ver4'.", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "The originally-assigned URL (nanofab.utah.edu/wp-content/uploads/2018/08/SU-8-2000-Series-Resists-MicroChem-19Mar131.pdf) was fetched successfully as a PDF but turned out to be MicroChem's Safety Data Sheet (SDS acc. to ISO/DIS 11014, printing date 04/17/2013, reviewed 03/19/2013) for 'SU-8 2000 Series Resists' — a hazard/GHS document with no spin-curve, bake, exposure-dose, or develop-time content. It was not used for any process data. A web search located the correct document — MicroChem's own 'SU-8 2000 Permanent Epoxy Negative Photoresist PROCESSING GUIDELINES FOR: SU-8 2025, SU-8 2035, SU-8 2050 and SU-8 2075' — mirrored by AMOLF (amolf.nl), which is the document actually read and cited above. This is a university mirror of Kayaku/MicroChem's own processing datasheet, not a university-authored SOP.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "su-8-2050", "name": "SU-8 2050", "manufacturer": "Kayaku Advanced Materials", "productLine": "SU-8 2000 series", "aliases": [ "SU8 2050", "MicroChem SU-8 2050" ], "tone": "negative", "chemistry": "epoxy", "grayscaleSuitable": false, "grayscaleNote": "Not addressed in the datasheet. SU-8 2000 is a chemically-amplified negative epoxy resist optimized for binary high-aspect-ratio structures (near-vertical sidewalls, high contrast); the datasheet gives no partial-crosslink / grayscale dose-response data, so grayscale suitability cannot be confirmed from this source.", "status": "active", "successorSlug": null, "summary": "SU-8 2050 is the mid-range thick-film workhorse of the SU-8 2000 family — the grade most MEMS and microfluidics processes default to when they need tall, robust structures without stepping up to the very thickest SU-8 2100/2150 coats. SU-8 2050 is a high-viscosity (12,900 cSt, 71.65% solids) grade in Kayaku Advanced Materials' SU-8 2000 epoxy photoresist series, formulated for thick, high-aspect-ratio, permanent MEMS and micromachining structures. It crosslinks via UV-generated photoacid followed by thermally-driven epoxy crosslinking during post-exposure bake, yielding a chemically and thermally stable, mechanically robust film that is normally left on the device rather than stripped.", "thicknessRange": { "min_um": 40.1, "max_um": 171, "basis": "curve-span", "source": "curve-span: the SU-8 2000 datasheet (April 2021, UBC mirror) states only a family-wide \"0.5 to >200 µm single coat\" range covering all twelve viscosities (Description/Features, p.1), not an SU-8 2050-specific achievable range; the only 2050 thickness data is the multi-grade Figure 1 (Thickness vs. Spin Speed, p.2). min/max are the span of this recipe's own figure-read SU-8 2050 curve (40.1 µm at 4000 rpm to 171 µm at 1000 rpm). ADJUDICATED 2026-07-12 (third pixel read, WL directive: figure over table): min/max corrected from 44/165 to 40.1/171 to match the corrected curve span (see spinCurves[0].source) — the datasheet's own printed nominal viscosity/thickness anchor is not published per-grade, so curve-span remains the correct basis, just recomputed from the corrected points." }, "spinCurves": [ { "label": "SU-8 2050", "points": [ { "rpm": 1000, "um": 171 }, { "rpm": 2000, "um": 75.5 }, { "rpm": 3000, "um": 53 }, { "rpm": 4000, "um": 40.1 } ], "source": "read from Figure 1 'SU-8 2000 Thickness vs. Spin Speed', p.2 of SU-8 2000 (2025-2100) Technical Data Sheet, Kayaku Advanced Materials, April 2021 — the chart plots four family curves (2025/2035/2050/2075) together with axis gridlines every 20 µm (0-240) and every 500 rpm (500-4500); no numeric table is published for this figure. ADJUDICATED 2026-07-12 (third pixel read, WL directive: figure over table): this figure is VECTOR-DRAWN, not a raster image — re-extracted directly from the PDF's vector marker paths (PyMuPDF get_drawings(), the small filled+stroked rects at each data point) with axes calibrated from the y-axis and x-axis tick-LABEL TEXT coordinates (exact, not eyeballed): y=0 at PDF-space y=619.5, y=240 at y=464.95 (1.5529 µm/pt); x=500rpm at x=48.9, x=4500rpm at x=249.2 (0.02 rpm/pt-inverse, i.e. 20.03 pt per 1000 rpm). The chart plots exactly FOUR markers per series, at 1000/2000/3000/4000 rpm only — the crawl-era raw read's 1500/2500/3500 rpm points were fabricated (not present in the source figure) and are dropped. SU-8 2050 identified as the second-from-top series (triangle marker, filled+stroked rect) by y-order at each x-column, consistent with its rank between SU-8 2075 (top) and SU-8 2035 (third); vector-derived values (171.17, 75.49, 53.03, 40.08 µm, rounded here to 3 sig figs) match the independent cross-check pixel read (171, 75.5, 53.0, 40.1 µm) to better than 0.2% at every point, confirming the crawl-era raw read (165/90/58/44, plus the fabricated 115/72/50 points) erred and the cross-check extraction was correct. Uncertainty on the vector-derived read is well under ±2% (exact marker geometry, not an eyeball estimate).", "figureRead": true } ], "spinNotes": "Kayaku's stated 'Recommended Program' (family-wide, not thickness-specific): dispense 1 ml resist per inch (25 mm) of substrate diameter; spin at 500 rpm for 5-10 s at 100 rpm/s acceleration (spread step); then spin at 2000 rpm for 30 s at 300 rpm/s acceleration as a generic starting point — the actual final rpm/time should be taken from Figure 1 for the desired thickness. Edge-bead removal (EBR) with a solvent stream (Kayaku EBR PG) at the wafer edge is recommended so the mask can seat in close contact. Secondary/practical (Cornell Nanoscale Facility SOP, university source, not the vendor): pour SU-8 2050-or-thicker from stock into small working bottles at least 24 h before spinning to let entrained bubbles dissipate; for layers ≥150 µm, let the wafer rest on the spinner after spin-up so hanging edge resist can retract back onto the wafer before wiping the edge bead, since a fully removed edge bead will just reflow into the gap during softbake.", "adhesion": { "hmds": false, "notes": "Adhesion promoters are stated as 'typically not required' for standard use. Substrates should simply be clean and dry — piranha wet etch (H2SO4 + H2O2) + DI rinse, or RIE / O2 barrel-ashing, is recommended for best results. The one exception: for applications involving electroplating, a substrate pre-treatment with HMDS IS recommended." }, "rehydration": null, "softbake": { "temp_c": 95, "time_s": 450, "method": "hotplate", "notes": "Datasheet times are given per THICKNESS bracket, not per grade. For a nominal SU-8 2050 film in the 45-80 µm range (matching ~3000 rpm per the spin curve above), Table 2 specifies 0-3 min at 65°C followed by 6-9 min at 95°C (450 s = midpoint of the 95°C range used here; the 65°C step is optional/short). A level hotplate is required — convection ovens are explicitly not recommended, as a skin can form on the resist and trap solvent. Vendor's own optimization method: after the prescribed bake, remove the wafer and let it cool to room temperature, then return it to the hotplate; if the film wrinkles, continue baking a few more minutes and repeat the cool/reheat cycle until wrinkles disappear.", "source": "Table 2 'Soft Bake Times', p.3 of SU-8 2000 (2025-2100) Technical Data Sheet, Kayaku Advanced Materials, April 2021" }, "exposureDose": { "doses": [], "datasheetBasis": "Table 3 'Exposure Dose' gives dose vs. FILM THICKNESS (not grade): 25-40 µm -> 150-160 mJ/cm2; 45-80 µm -> 150-215 mJ/cm2; 85-110 µm -> 215-240 mJ/cm2; 115-150 µm -> 240-260 mJ/cm2; 160-225 µm -> 260-350 mJ/cm2; 230-270 µm -> 350-370 mJ/cm2. For a SU-8 2050 film in its typical ~45-80 µm working range (per the spin curve above, roughly 2500-4000 rpm) the applicable dose is 150-215 mJ/cm2.", "_note": "Left null rather than picking one number: the correct dose depends on which thickness within SU-8 2050's usable range is actually coated, and the datasheet publishes a range per bracket rather than a single figure, so any single value here would be an invented point estimate. The text states 'i-Line (365 nm) is the recommended wavelength' for exposure, but Table 3's doses are not explicitly labeled as i-line-only figures (the intro also says SU-8 2000 is 'most commonly exposed with conventional UV (350-400 nm) radiation') — so attributing the table doses specifically to 365 nm would be an inference the source doesn't make explicit. A long-pass filter (350 nm cutoff, e.g. Omega PL-360-LP) is recommended for vertical sidewalls, which requires ~40% more exposure time to reach the same effective dose. Also note per-substrate dose multipliers are published separately (Table 4): 1x on silicon, 1.5x on glass/Pyrex/ITO, 1.5-2x on silicon nitride, gold, aluminum, NiFe, copper, nickel, titanium." }, "peb": { "temp_c": 95, "time_s": 390, "notes": "Datasheet times are per THICKNESS bracket. For the 45-80 µm bracket (SU-8 2050's typical working range): 1-2 min optional PEB at 65°C for stress reduction, followed by 6-7 min at 95°C (390 s = midpoint of that range used here). PEB should start directly after exposure. A visible latent mask image should appear in the resist within 5-15 s of starting the 95°C PEB step if exposure/heating were adequate; if no image forms, exposure and/or bake was insufficient.", "source": "Table 5 'Post Exposure Bake Times', p.4 of SU-8 2000 (2025-2100) Technical Data Sheet, Kayaku Advanced Materials, April 2021" }, "floodExposure": { "dose_mJcm2": null, "notes": null, "source": null }, "develop": { "developer": "SU-8 Developer (Kayaku's proprietary PGMEA-based developer); other solvent developers such as ethyl lactate or diacetone alcohol are also stated to work", "dilution": "undiluted", "time_s": 360, "method": "immersion (spray or spray-puddle also usable per datasheet)", "rinse": "Fresh SU-8 developer spray/wash ~10 s, optionally repeated, then air/N2 dry with filtered pressurized gas. A white film during an IPA rinse indicates under-development of the unexposed resist -- treat with more SU-8 developer and repeat, rather than relying on IPA to finish the job. Ultrasonic/megasonic agitation is recommended for high-aspect-ratio or tight-pitch structures.", "source": "Table 6 'Development Times for SU-8 Developer' (45-75 µm bracket: 5-7 min, 360 s used as midpoint), p.4 of SU-8 2000 (2025-2100) Technical Data Sheet, Kayaku Advanced Materials, April 2021" }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Optional and application-dependent, so no single figure is given: recommended final bake temperature is 10°C above the maximum expected device operating temperature; typical range is 150-250°C for 5-30 min depending on cure degree required. Separately, a short 150°C bake for 'a couple of minutes' is recommended specifically to anneal any surface cracks visible after development, and this applies to all film thicknesses.", "source": "Hard Bake (cure) section, p.5 of SU-8 2000 (2025-2100) Technical Data Sheet, Kayaku Advanced Materials, April 2021" }, "descum": null, "applications": [ "mems-structural", "high-aspect-ratio", "microfluidics", "electroplating-molding" ], "etchResistance": "Described as chemically and thermally very stable once crosslinked: thermal stability to 315°C (5% wt. loss), and the Removal section states it is 'extremely difficult to remove... with conventional solvent based resist strippers.' This robustness makes cured SU-8 usable as a durable etch/plating mold in many wet-chemical processes, at the cost of being hard to strip afterward (see stripper field).", "liftoffSuitable": false, "platingSuitable": true, "stripper": "Minimally-crosslinked SU-8: Kayaku Remover PG, 50-80°C bath, 30-90 min immersion (swells and lifts off partially-cured material; if OmniCoat 30-100 nm sacrificial layer was applied first, Remover PG gives a clean full lift-off). Fully cured/hard-baked SU-8 CANNOT be removed with Remover PG at all unless OmniCoat was used underneath. To rework fully crosslinked SU-8: oxidizing acid solutions (piranha etch), plasma ash, RIE (200 W, 80 sccm O2, 8 sccm CF4, 100 mTorr, 10°C), laser ablation, or pyrolysis.", "storage": "Store upright in tightly closed containers, in a cool dry environment away from direct sunlight, light, acids, heat, and ignition sources, at 40-70°F (4-21°C). Shelf life is 13 months from date of manufacture.", "notes": "SU-8 2050 sits in the middle of the SU-8 2000 family by viscosity (12,900 cSt vs. 4,500 for 2025 and 22,000 for 2075) and covers roughly 45-165 µm in a single coat over the datasheet's plotted 1000-4000 rpm range (the family description states thicknesses down to 0.5 µm and beyond 200 µm are achievable across the whole SU-8 2000 line, implying lower speeds than plotted would push a single 2050 coat higher still, but that isn't shown numerically for this grade). Choose 2050 over the thinner SU-8 2025 when a structure needs more height and mechanical robustness, and over the much thicker SU-8 2100/2150 grades when their hour-long bakes and very deep develops would otherwise dominate the process. PEB is the step where crosslinking actually completes (thermally-driven, acid-catalyzed epoxy reaction) -- a visible latent mask image appearing within 5-15 s of starting the 95°C PEB is the datasheet's own go/no-go check for adequate exposure and heating. Thick films are prone to cracking and delamination from thermal stress; besides the softbake wrinkle-check cycle, a University of Cornell Nanoscale Facility SOP (secondary source, not the vendor) recommends gradual heating/cooling for any layer >=50 µm, since silicon cools faster than SU-8, and a ~10 minute wait between exposure and PEB start for full latent-image (photoacid) formation. Because fully cross-linked SU-8 is essentially unstrippable with ordinary solvents, any process that will need to remove or release the structure later should plan for an OmniCoat sacrificial layer from the start, not as an afterthought.", "developerFamily": "solvent", "references": [ { "type": "paper", "title": "Negative photoresists for optical lithography", "authors": "Shaw et al.", "journal": "IBM Journal of Research and Development", "year": 1997, "doi": "10.1147/rd.411.0081", "url": "https://doi.org/10.1147/rd.411.0081", "accessedDate": "2026-07-12", "summary": "The IBM origin paper of SU-8." }, { "type": "paper", "title": "SU-8: a low-cost negative resist for MEMS", "authors": "Lorenz et al.", "journal": "Journal of Micromechanics and Microengineering", "year": 1997, "doi": "10.1088/0960-1317/7/3/010", "url": "https://doi.org/10.1088/0960-1317/7/3/010", "accessedDate": "2026-07-12", "summary": "15:1 aspect ratios in 1997 — the paper that made SU-8 a MEMS resist." }, { "type": "paper", "title": "SU-8: a photoresist for high-aspect-ratio and 3D submicron lithography", "authors": "del Campo et al.", "journal": "Journal of Micromechanics and Microengineering", "year": 2007, "doi": "10.1088/0960-1317/17/6/R01", "url": "https://doi.org/10.1088/0960-1317/17/6/R01", "accessedDate": "2026-07-12", "summary": "The canonical SU-8 review." }, { "type": "paper", "title": "Microfluidic probes for use in life sciences and medicine", "authors": "Qasaimeh et al.", "journal": "Lab Chip", "year": 2013, "doi": "10.1039/C2LC40898H", "url": "https://doi.org/10.1039/C2LC40898H", "accessedDate": "2026-07-12", "summary": "Free-standing SU-8 microfluidic probes for live-cell chemistry." } ], "provenance": { "datasheetUrl": "https://kayakuam.com/wp-content/uploads/2020/08/KAM-SU-8-2000-2025-2075-Datasheet.8.19.20-final.pdf", "datasheetVersionOrDate": "SU-8 2000, 2025-2100, Technical Data Sheet, April 2021 (per footer on every page; fetched via a university-hosted mirror of the same document, https://nanofab.sites.olt.ubc.ca/files/2026/01/KAM-SU-8-2000-2025-2100-Datasheet-4.9.21-final-2.pdf, since kayakuam.com returned HTTP 403 to automated fetches during this session)", "accessedDate": "2026-07-10", "secondarySources": [ { "url": "https://www.cnfusers.cornell.edu/sites/default/files/Equipment-Resources/SU8%20processing%20suggestions.pdf", "what": "Cornell Nanoscale Facility 'SU-8 Processing Suggestions' (v2, July 2013) -- used only for practical process tips not in the vendor datasheet: pre-spin bubble-dissipation wait time for SU-8 2050+, the ~10 min exposure-to-PEB delay for latent image formation, and thermal-stress/cracking guidance for thick and multi-layer coats. Not used for any headline numeric spec (spin curve, dose, bake time/temp all come from the Kayaku datasheet)." } ] }, "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "su-8-2100", "name": "SU-8 2100", "manufacturer": "MicroChem", "productLine": "SU-8 2000 series", "aliases": [ "SU8 2100", "MicroChem SU-8 2100" ], "tone": "negative", "chemistry": "epoxy", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed by the datasheet. The document presents SU-8 2000 exclusively for binary, high-aspect-ratio, vertical-sidewall permanent structures (\"high aspect ratio imaging\", \"vertical sidewalls\"); it makes no mention of grayscale or partial-exposure profiling.", "status": "active", "successorSlug": null, "summary": "SU-8 2100 is the very-thick-film grade of the SU-8 2000 family — the choice when a single coat must reach into the hundreds of microns, well beyond what SU-8 2025 or 2050 can lay down, at the cost of hour-scale bakes. SU-8 2100 is a very high-viscosity member of MicroChem's SU-8 2000 epoxy negative photoresist series, coating roughly 100-270 µm in a single pass for thick, permanent, high-aspect-ratio microstructures.", "thicknessRange": { "min_um": 103, "max_um": 269, "basis": "curve-span", "source": "curve-span: the document states no single achievable-thickness range for SU-8 2100 specifically (only thickness-binned process tables shared with SU-8 2150); the range here is the min/max of the SU-8 2100 curve in Figure 1 (3000 rpm to 1000 rpm), pixel-calibrated 2026-07-12 (PyMuPDF axis-tick and vector marker-path extraction, blue-fill color isolation from the red SU-8 2150 trace). Adjudicated 2026-07-12: confirmed against the archived high-resolution chart image; check's stale 0.5/null was the SU-8 2000 product-LINE's general '0.5->200 µm' description, not this grade's own curve — corrected to match." }, "spinCurves": [ { "label": "SU-8 2100", "points": [ { "rpm": 1000, "um": 269 }, { "rpm": 2000, "um": 137 }, { "rpm": 3000, "um": 103 } ], "source": "re-extracted 2026-07-12, pixel-calibrated (PyMuPDF: axis-tick word coordinates for calibration, vector marker path rects for data points, colored (0,0,1)=blue to isolate from the red 2150 trace); Figure 1 \"SU-8 2000 Spin Speed versus Thickness\", p.2 of MicroChem \"SU-8 2000 Permanent Epoxy Negative Photoresist Processing Guidelines for SU-8 2100 and SU-8 2150\" (CNR-Nano mirror). Chart plots only two curves, unambiguously distinguished by both color and marker shape: SU-8 2150 = red triangles (top curve), SU-8 2100 = blue circles (bottom curve), matching Table 1's viscosity ordering (2100 = 45,000 cSt < 2150 = 80,000 cSt, so 2100 should coat thinner at a given speed). Each curve has only 3 markers, at 1000/2000/3000 rpm — the blue trace does not extend to 3500 rpm despite the chart's wider axis range. Supersedes the earlier eyeball read (270/137/105), which was already close but imprecise.", "figureRead": true } ], "spinNotes": "Recommended program (same for both grades in this document): dispense 1 ml of resist per inch (25 mm) of substrate diameter; spin at 500 rpm for 5-10 s at 100 rpm/s acceleration, then spin at the target speed (per Figure 1) for 30 s at 300 rpm/s acceleration. Edge bead removal (EBR) with MicroChem's EBR PG solvent stream at the wafer edge is recommended before soft bake, both to limit hotplate contamination and to let the photomask reach close contact with the wafer. Source: \"Coat\" / \"Recommended Program\" / \"Edge Bead Removal (EBR)\", p.2.", "adhesion": { "hmds": false, "notes": "\"Adhesion promoters are typically not required.\" HMDS pretreatment (MCC Primer 80/20) is recommended only \"for applications that include electroplating.\" Source: \"Substrate Preparation\", p.2." }, "rehydration": null, "softbake": { "temp_c": null, "time_s": null, "method": "hotplate", "notes": "Published only as a THICKNESS-BINNED table shared by both grades covered in this document (SU-8 2100 and 2150), not a single value per grade: 100-150 µm → 5 min @65°C then 20-30 min @95°C; 160-225 µm → 5-7 min @65°C then 30-45 min @95°C; 230-270 µm → 7 min @65°C then 45-60 min @95°C; 280-550 µm → 7-10 min @65°C then 60-120 min @95°C. SU-8 2100's own spin curve spans roughly 105-270 µm (1000-3000 rpm), i.e. mostly the 100-150 and 230-270 µm bins. Convection ovens are explicitly not recommended (can skin over and trap solvent). A cool-down/re-heat 'wrinkle' check is described to confirm the film is fully dry.", "source": "Table 2 \"Soft Bake Times\", p.2 of the MicroChem SU-8 2000 (2100-2150) Processing Guidelines" }, "exposureDose": { "doses": [], "datasheetBasis": "\"SU-8 2000 photoresist is most commonly exposed with conventional UV (350-400 nm) radiation, although i-line (365 nm) is the recommended wavelength. SU-8 2000 may also be exposed with e-beam or x-ray radiation.\" (Processing Guidelines, p.1). The dose table itself (Table 3) is not separately re-attributed to a specific wavelength beyond this general statement.", "_note": "Dose is published only as a THICKNESS-BINNED range shared across both grades, not a single scalar for SU-8 2100: 100-150 µm → 240-260 mJ/cm²; 160-225 µm → 260-350 mJ/cm²; 230-270 µm → 350-370 mJ/cm²; 280-550 µm → 370-600 mJ/cm². SU-8 2100's own process (105-270 µm) falls mostly in the 100-150 and 230-270 µm bins. Relative dose also scales with substrate (Table 4): silicon 1X, glass/Pyrex/ITO 1.5X, most metals (Au, Al, NiFe, Cu, Ni, Ti) and silicon nitride 1.5-2X. Using a >350 nm long-pass filter (recommended for vertical sidewalls) requires ~40% more exposure time to reach the same effective dose. Left null rather than picking a bin midpoint, per the range rule." }, "peb": { "temp_c": null, "time_s": null, "notes": "Also thickness-binned across both grades: 100-150 µm → 5 min @65°C (optional stress-reduction step) then 10-12 min @95°C; 160-225 µm → 5 min @65°C then 12-15 min @95°C; 230-270 µm → 5 min @65°C then 15-20 min @95°C; 280-550 µm → 5 min @65°C then 20-30 min @95°C. After 1 minute of PEB at 95°C a latent mask image should already be visible; if not, exposure and/or heating was insufficient.", "source": "Table 5 \"Post Exposure Bake Times\", p.3" }, "floodExposure": null, "develop": { "developer": "SU-8 Developer (MicroChem)", "dilution": null, "time_s": null, "method": "immersion", "rinse": "IPA", "source": "\"Development\" and \"Rinse and Dry\" sections plus Table 6 \"Development Times for SU-8 Developer\", p.3", "_note": "Development is designed for immersion, spray, or spray-puddle processing with SU-8 Developer (other solvent developers such as ethyl lactate and diacetone alcohol are also usable per the text, but no dilution ratio is given for any of them). Immersion times are thickness-binned, not a single value: 100-150 µm → 10-15 min; 160-225 µm → 15-17 min; 230-270 µm → 17-20 min; 280-550 µm → 20-30 min. Rinse: ~10 s fresh developer spray/wash, then ~10 s IPA spray/wash, then dry with filtered N2/air. A white film after IPA rinse indicates underdevelopment and calls for another develop+rinse cycle." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Hard bake (cure) is optional and generally only needed if the finished device will see thermal processing in use; recommended final bake temperature is 10°C above the maximum expected device operating temperature. Typical range: 150-250°C for 5-30 minutes depending on degree of cure required. Separately, a short 150°C bake \"for a couple of minutes\" is recommended (all thicknesses) specifically to anneal any surface cracks seen after development.", "source": "\"Hard Bake (cure)\" section, p.4" }, "descum": null, "applications": [ "high-aspect-ratio", "mems-structural", "electroplating-molding" ], "etchResistance": null, "liftoffSuitable": false, "platingSuitable": true, "stripper": "MicroChem Remover PG swells and lifts only minimally cross-linked SU-8 2000. A fully cured/hard-baked film cannot be removed with Remover PG alone — it requires an OmniCoat (30-100 nm) sacrificial underlayer (heat Remover PG to 50-80°C, immerse 30-90 min) or an oxidizing strip (piranha etch, plasma ash, RIE, laser ablation, or pyrolysis). RIE recipe given: 200 W, 80 sccm O2, 8 sccm CF4, 100 mTorr, 10°C. Source: \"Removal\" / \"Plasma Removal\", p.4.", "storage": "Store upright in tightly closed containers, cool and dry, away from direct sunlight, at 40-70°F (4-21°C); away from light, acids, heat, and ignition sources. Shelf life is thirteen months from date of manufacture (note: the sibling 2025-2075 processing document states twelve months — the two documents disagree on shelf life). Source: \"Storage\", p.4.", "notes": "SU-8 2100 is the second-highest-viscosity grade in MicroChem's SU-8 2000 line (45,000 cSt) and is built for very thick, single-coat films — its own spin curve runs roughly 105-270 µm over 1000-3000 rpm. Reach for 2100 only when the device genuinely needs that film height; its long soft-bake and develop times and thick-film handling make the thinner SU-8 2025 and 2050 grades the better default whenever they can meet the thickness target. Like every SU-8 2000 grade, it cross-links in two stages — exposure generates acid, and the post-exposure bake thermally drives the epoxy cross-linking — so PEB is a required processing step, not an optional cure, and at this thickness the bake times stretch into hours (up to 60-120 minutes soft bake and 20-30 minutes PEB at 95°C for the thickest bin in this document). Soft-bake, PEB, dose and develop times are published only as thickness-binned ranges shared with SU-8 2150; treat the low end of the matching bin as a starting point and use the datasheet's cool-down/re-heat 'wrinkle' test to confirm the soft bake is complete before proceeding. Once fully cross-linked, SU-8 is notoriously hard to strip: plain solvent remover only works on minimally exposed/baked film, and a hard-baked structure needs either a sacrificial OmniCoat layer beneath it or an oxidizing strip (piranha, plasma ash, RIE, laser ablation, pyrolysis) to remove. Edge bead removal before soft bake is especially important at this film thickness to keep the photomask in close contact with the wafer.", "developerFamily": "solvent", "references": [ { "type": "paper", "title": "Negative photoresists for optical lithography", "authors": "Shaw et al.", "journal": "IBM Journal of Research and Development", "year": 1997, "doi": "10.1147/rd.411.0081", "url": "https://doi.org/10.1147/rd.411.0081", "accessedDate": "2026-07-12", "summary": "The IBM origin paper of SU-8." }, { "type": "paper", "title": "SU-8: a low-cost negative resist for MEMS", "authors": "Lorenz et al.", "journal": "Journal of Micromechanics and Microengineering", "year": 1997, "doi": "10.1088/0960-1317/7/3/010", "url": "https://doi.org/10.1088/0960-1317/7/3/010", "accessedDate": "2026-07-12", "summary": "15:1 aspect ratios in 1997 — the paper that made SU-8 a MEMS resist." }, { "type": "paper", "title": "SU-8: a photoresist for high-aspect-ratio and 3D submicron lithography", "authors": "del Campo et al.", "journal": "Journal of Micromechanics and Microengineering", "year": 2007, "doi": "10.1088/0960-1317/17/6/R01", "url": "https://doi.org/10.1088/0960-1317/17/6/R01", "accessedDate": "2026-07-12", "summary": "The canonical SU-8 review." }, { "type": "paper", "title": "Microfluidic probes for use in life sciences and medicine", "authors": "Qasaimeh et al.", "journal": "Lab Chip", "year": 2013, "doi": "10.1039/C2LC40898H", "url": "https://doi.org/10.1039/C2LC40898H", "accessedDate": "2026-07-12", "summary": "Free-standing SU-8 microfluidic probes for live-cell chemistry." } ], "provenance": { "datasheetUrl": "https://litho.nano.cnr.it/wp-content/datasheets/SU-82000DataSheet2100and2150Ver5.pdf", "datasheetVersionOrDate": "No revision/date string is printed in the document body (a 5-page processing-guidelines PDF with no header/footer revision code). The mirror's filename indicates 'Ver5' (a byu.edu mirror of the apparently same document is filenamed 'Ver5-1').", "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "The originally-assigned URL (nanofab.utah.edu/wp-content/uploads/2018/08/SU-8-2000-Series-Resists-MicroChem-19Mar131.pdf) was fetched successfully as a PDF but turned out to be MicroChem's Safety Data Sheet (SDS acc. to ISO/DIS 11014, printing date 04/17/2013, reviewed 03/19/2013) for 'SU-8 2000 Series Resists' — a hazard/GHS document with no spin-curve, bake, exposure-dose, or develop-time content. It was not used for any process data. A web search located the correct document — MicroChem's own 'SU-8 2000 Permanent Epoxy Negative Photoresist PROCESSING GUIDELINES FOR: SU-8 2100 and SU-8 2150'. A first attempt to fetch this document from a BYU cleanroom mirror (cleanroom.byu.edu) failed with a curl TLS/connection error (exit code 60); the identical document was then fetched successfully from a CNR-Nano (Italy) mirror (litho.nano.cnr.it), which is the document actually read and cited above. This is a university mirror of Kayaku/MicroChem's own processing datasheet, not a university-authored SOP.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "su-8-3025", "name": "SU-8 3025", "manufacturer": "Kayaku MicroChem / MicroChem Corp.", "productLine": "SU-8 3000 series", "aliases": [ "SU8 3025" ], "tone": "negative", "chemistry": "epoxy", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed in the datasheet; SU-8 3000 is described as a chemically-amplified negative epoxy resist for binary, high-aspect-ratio, permanent structures, and no partial-crosslink/grayscale dose-response data is given.", "status": "active", "successorSlug": null, "summary": "SU-8 3025 is the mid-viscosity grade of the newer SU-8 3000 series — the one to choose over an equivalent SU-8 2000 resist when a thick, high-aspect-ratio film keeps cracking or lifting, since the 3000 formulation was reworked to fix exactly that. SU-8 3025 is the third-lowest-viscosity member (4,400 cSt, 72.3% solids) of Kayaku MicroChem's five-grade SU-8 3000 permanent epoxy series (3005/3010/3025/3035/3050), an 'improved formulation' of SU-8 2000 offering better adhesion and lower coating stress for thick (family-wide 4-120 µm single-coat), high-aspect-ratio (>5:1), permanent MEMS structures that are imaged, cured, and left on the device rather than lifted off.", "thicknessRange": { "min_um": null, "max_um": null }, "spinCurves": [ { "label": "SU-8 3025", "points": [ { "rpm": 1000, "um": 58 }, { "rpm": 2000, "um": 37 }, { "rpm": 3000, "um": 25 }, { "rpm": 4000, "um": 23 } ], "source": "read from figure 1, \"Spin speed vs. Thickness for SU-8 3000 resists (21°C US & EU)\", p.1 of the Kayaku Microchem/MicroChem \"SU-8 3000 Permanent Epoxy Negative Photoresist\" datasheet (BYU cleanroom mirror; no printed revision code). Chart plots five grades (3050/filled square, 3035/filled triangle, 3025/filled diamond, 3010/filled circle, 3005/open square), each with 4 markers at 1000/2000/3000/4000 rpm. The SU-8 3025 curve was identified as the filled-diamond series, 3rd from the top of five at every rpm, consistent with its legend position (3rd of 3050/3035/3025/3010/3005) and Table 1's viscosity ordering (3025 = 4400 cSt, the middle of the five grades). Points were read by extracting the marker vector-path pixel rectangles and the axis tick-label pixel positions directly from the PDF page content stream (not a manual on-screen eyeball estimate), then converting via the resulting pixel-to-value linear scale (0-110 µm y-axis span; 1000 rpm x-axis grid spacing); cross-checked against the rendered chart image. No numeric per-rpm table exists for SU-8 3025 alone in this datasheet (Tables 2/3/5/6 are thickness-binned ranges shared across the whole 3000 series, not per-grade spin data). digitized 2026-07-12", "figureRead": true } ], "spinNotes": "Recommended Program (family-wide, not grade-specific): dispense 1 ml resist per inch (25 mm) of substrate diameter; spin at 500 rpm for 5-10 s at 100 rpm/s acceleration (spread step); then spin at 3000 rpm for 30 s at 300 rpm/s acceleration as a generic starting point — the actual final rpm/time for a target SU-8 3025 thickness should be read from Figure 1 ('Spin speed vs. Thickness for SU-8 3000 resists', 21°C US & EU, p.1) or Figure 2 (23°C Japan & Asia, p.2), each of which plots all five grades (3005/3010/3025/3035/3050) as separate traces vs. 1000-5000 rpm with NO accompanying numeric table. Figure 1 (21°C US & EU) has now been digitized above by extracting the marker vector-path pixel positions and axis tick-label pixel positions directly from the PDF page content stream (not a manual on-screen eyeball estimate); the SU-8 3025 trace was identified as the filled-diamond series, 3rd from the top of five, consistent with its legend position and Table 1's viscosity ordering (3025 = 4400 cSt, the middle of the five grades). Figure 2 (23°C Japan & Asia) was not digitized — only Figure 1 is captured as the primary curve, matching the convention used elsewhere in this library for datasheets publishing two ambient-condition charts. No numeric table exists for SU-8 3025 alone — Table 1 (p.2) only ranks the five grades by viscosity. No edge-bead-removal step is described anywhere in this SU-8 3000 datasheet, and no resist rehydration hold is mentioned either.", "adhesion": { "hmds": false, "notes": "Adhesion promoters are stated as 'typically not required' for substrates cleaned per the recommended piranha (H2SO4+H2O2) wet etch or RIE/O2-barrel-ash prep. Exception: for applications requiring electroplating, the datasheet recommends pre-treating the substrate with MCC Primer 80/20 (HMDS)." }, "rehydration": null, "softbake": { "temp_c": 95, "time_s": null, "method": "hotplate", "notes": "95°C is fixed across all thickness brackets; only bake TIME varies by film thickness per Table 2: 4-10 µm -> 2-3 min; 8-15 µm -> 5-10 min; 20-50 µm -> 10-15 min; 30-80 µm -> 10-30 min; 40-100 µm -> 15-45 min. time_s is left null because no confident SU-8 3025-specific thickness/bracket could be established (see spinNotes). Convection ovens are explicitly NOT recommended — a skin can form on the resist and trap solvent, incompletely drying the film. Vendor's own optimization check: after the prescribed bake, cool the wafer to room temperature, then return it to the hotplate; if the film wrinkles, continue baking and repeat the cool/reheat cycle until wrinkles no longer appear.", "source": "Table 2 'Soft Bake Times', p.2 of SU-8 3000 Permanent Epoxy Negative Photoresist datasheet, Kayaku MicroChem/MicroChem Corp." }, "exposureDose": { "doses": [], "datasheetBasis": "'SU-8 3000 is most commonly exposed with conventional UV (350-400 nm) radiation, although i-line (365 nm) is recommended' (Processing Guidelines, p.1). Table 3's dose values are not explicitly labeled as i-line-only.", "_note": "Table 3 'Exposure Dose' (p.2) publishes energy vs. FILM THICKNESS bracket, not per-grade: 4-10 µm -> 100-200 mJ/cm²; 8-15 µm -> 125-200; 20-50 µm -> 150-250; 30-80 µm -> 150-250; 40-100 µm -> 150-250. Left entirely null rather than picking a single value or bracket: (1) no confident spin curve was established for SU-8 3025 specifically (see spinNotes), so the thickness bracket a real SU-8 3025 process would fall into cannot be determined from this document, and (2) the datasheet does not explicitly attribute Table 3's doses to i-line — it states broadband UV (350-400 nm) is 'most common' with i-line merely 'recommended' — so even a chosen bracket could not be routed into at365_mJcm2 without over-interpreting the source. Table 4 (p.2) gives substrate dose multipliers relative to silicon (1x): glass/Pyrex/indium-tin-oxide 1.5x; silicon nitride, gold, aluminum, nickel-iron, copper, nickel, titanium 1.5-2x. Separately, a long-pass filter (Omega PL-360-LP, or Asahi Technoglass V-42+UV-D35) to cut UV below 350 nm is recommended for vertical sidewalls, requiring ~40% more exposure time to reach the same effective dose." }, "peb": { "temp_c": 95, "time_s": null, "notes": "An optional 1-minute 65°C stress-reduction pre-bake step precedes the main PEB across ALL thickness brackets (65°C time is a constant 1 min; only the main 95°C step's time varies by thickness) per Table 5: 4-10 µm -> 1-2 min; 8-15 µm -> 2-4 min; 20-50 µm -> 3-5 min; 30-80 µm -> 3-5 min; 40-100 µm -> 3-5 min. time_s is left null (no confident SU-8 3025-specific thickness bracket — see spinNotes). PEB should start directly after exposure. A visible latent mask image should appear within 1 minute of starting the 95°C step if exposure and bake were adequate; no visible image means insufficient exposure and/or temperature.", "source": "Table 5 'Post Exposure Bake Times', p.3 of SU-8 3000 Permanent Epoxy Negative Photoresist datasheet, Kayaku MicroChem/MicroChem Corp." }, "floodExposure": null, "develop": { "developer": "SU-8 Developer (MicroChem's proprietary solvent-based developer); ethyl lactate and diacetone alcohol are also stated to work", "dilution": null, "time_s": null, "method": null, "rinse": "Spray/wash with fresh SU-8 developer for ~10 s, then a second spray/wash with isopropyl alcohol (IPA) for ~10 s, then air/nitrogen dry with filtered pressurized gas. A white film appearing during the IPA rinse indicates under-development — immerse or spray with more SU-8 developer and repeat the rinse. Ultrasonic or megasonic bath agitation is recommended for developing out vias/holes in high-aspect-ratio structures.", "source": "Develop and Rinse-and-Dry sections, p.3 of SU-8 3000 Permanent Epoxy Negative Photoresist datasheet, Kayaku MicroChem/MicroChem Corp.", "_note": "Table 6 'Development Times for SU-8 Developer' (p.3) publishes immersion develop time vs. FILM THICKNESS bracket, not per-grade: 4-10 µm -> 1-3 min; 8-15 µm -> 4-6 min; 20-50 µm -> 5-8 min; 30-80 µm -> 6-12 min; 40-100 µm -> 7-15 min ('approximate, since actual dissolution rates can vary widely as a function of agitation'). time_s is left null (no confident SU-8 3025-specific bracket — see spinNotes). Method left null: the datasheet states development works 'in immersion, spray or spray-puddle processes' without recommending one for these times. Dilution is not addressed anywhere in this document (no 'undiluted' statement is printed, unlike some sibling SU-8 datasheets), so left null rather than assumed." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Optional, for applications where the imaged resist is left as part of the final device: 'the resist may be ramp/step hard baked between 150-200°C on a hot plate or in a convection oven to further cross link the material. Bake times vary based on type of bake process and film thickness.' No single temperature or time is given, and unlike softbake/PEB/develop, no thickness-bracket table is published for hardbake either.", "source": "Hard Bake (cure) section, p.3 of SU-8 3000 Permanent Epoxy Negative Photoresist datasheet, Kayaku MicroChem/MicroChem Corp." }, "descum": null, "applications": [ "mems-structural", "high-aspect-ratio", "electroplating-molding" ], "etchResistance": "Listed under Features as 'Excellent dry etch resistance' (p.1, no numeric selectivity given). The Plasma Removal section (p.4) gives an RIE recipe (200 W, 80 sccm O2, 8 sccm CF4, 100 mTorr, 10°C) specifically for STRIPPING crosslinked SU-8 — i.e. cured SU-8 requires an aggressive dedicated RIE recipe to remove at all, consistent with strong etch resistance during normal device processing.", "liftoffSuitable": false, "platingSuitable": true, "stripper": "MicroChem's Remover PG, 50-80°C bath, 30-90 min immersion — swells and lifts off minimally-crosslinked SU-8 3000; achieves a clean, thorough lift-off of fully-crosslinked SU-8 3000 only if a 30-100 nm OmniCoat sacrificial layer was applied first. Will NOT remove fully-cured/hard-baked SU-8 3000 without OmniCoat underneath. To rework fully cross-linked material with no OmniCoat: oxidizing acid solutions (piranha etch), plasma ash, RIE (200 W, 80 sccm O2, 8 sccm CF4, 100 mTorr, 10°C), laser ablation, or pyrolysis.", "storage": "Store upright, in tightly closed containers, in a cool, dry environment away from direct sunlight, at 40-70°F (4-21°C). Store away from light, acids, heat, and sources of ignition. Shelf life is twelve months from date of manufacture.", "notes": "SU-8 3025 is the third-lowest-viscosity member (4,400 cSt, 72.3% solids) of Kayaku MicroChem's five-grade SU-8 3000 permanent epoxy series (3005/3010/3025/3035/3050), an 'improved formulation' of SU-8 2000 offering better adhesion and lower coating stress for thick (family-wide 4-120 µm single-coat), high-aspect-ratio (>5:1), permanent MEMS structures that are imaged, cured, and left on the device rather than lifted off. Within the 3000 series it is the mid-viscosity option — thicker-coating than 3005 and 3010, thinner than 3035 and 3050 — and it is generally preferred over the matching SU-8 2000 grades when a process suffers adhesion loss or stress cracking, since the 3000 line was reformulated specifically to reduce both. Figures 1-2 plot all five grades' film thickness against 1000-5000 rpm spin speed on one shared axis with no accompanying numeric table; because a prior recipe already in this library (SU-8 2050) was built from a misread multi-grade SU-8 chart with errors of 20-42% at every point, no spin-curve points are published here for SU-8 3025 — spinNotes records the viscosity-ranking argument that was available (a physically sound bound on which curve is 3025, but not a substitute for digitizing it) and why that fell short of a confident pixel-level read. Softbake time, PEB time, exposure dose, and develop time are all published as THICKNESS-BINNED ranges common to the whole 3000 series (not per-grade single values), so every corresponding scalar field here is null with the full bracket table quoted in its own notes — reporting a single number would be false precision without first knowing which bracket a real SU-8 3025 process actually lands in. As with the rest of the SU-8 family, fully cross-linked SU-8 3025 is notoriously difficult to strip: MicroChem's Remover PG only lifts minimally-crosslinked resist (or fully-crosslinked resist sitting over a sacrificial OmniCoat layer); a hard-baked film with no OmniCoat underneath requires piranha etch, plasma ashing, RIE, laser ablation, or pyrolysis to remove.", "developerFamily": "solvent", "provenance": { "datasheetUrl": "http://cleanroom.groups.et.byu.net/su8.parts/SU-8%203000%20Data%20Sheet-1.pdf", "datasheetVersionOrDate": null, "accessedDate": "2026-07-10", "secondarySources": [] }, "_provenanceNote": "Document is Kayaku MicroChem / MicroChem Corp.'s own 'SU-8 3000 Permanent Epoxy Negative Photoresist' product data sheet (4 pages, dual KAYAKU MICROCHEM / MICRO•CHEM letterhead on every page), mirrored by Brigham Young University's Cleanroom group (cleanroom.groups.et.byu.net/su8.parts/) — kayakuam.com returns HTTP 403 to automated fetches (verified three times per this assignment), so this BYU-hosted mirror of Kayaku/MicroChem's own SU-8 3000 series datasheet is the primary document read. This is a manufacturer datasheet, not a university-authored SOP; BYU's role here is purely as file host/mirror. Content covers the whole SU-8 3000 series (3005/3010/3025/3035/3050) including SU-8 3025 by name throughout Table 1 and Figures 1-2 — no slug/content mismatch. No revision code or printed date appears anywhere in the document.", "tier": "datasheet", "dateAdded": "2026-07-10", "dateModified": "2026-07-12", "humanVerified": false }, { "slug": "suex", "name": "SUEX", "manufacturer": "DJ MicroLaminates", "productLine": "SUEX Dry Film Sheets (TDFS)", "aliases": [ "SUEX TDFS", "SUEX Dry Film Sheets", "DJ MicroLaminates SUEX" ], "tone": "negative", "chemistry": "epoxy", "photoimageable": true, "grayscaleSuitable": false, "grayscaleNote": "Not addressed by either datasheet. Both the Thick and Thin SUEX Data Sheets describe SUEX only as a negative-tone, permanent structural/plating epoxy for 'plating, wafer level packaging and MEMS applications'; neither mentions grayscale or partial-exposure profiling.", "status": "active", "successorSlug": null, "summary": "SUEX is DJ MicroLaminates' cationically-cured modified-epoxy dry-film photoresist, laminated — not spin-coated — onto a substrate as pre-cut sheets, spanning a thin-film line (20-75 µm) and a thick-film line (100 µm to 1 mm), for plating, wafer-level-packaging and MEMS structural applications.", "thicknessRange": { "min_um": 20, "max_um": 1000, "basis": "stated", "source": "Thin SUEX: 'Sheets are available in thicknesses from 20µm to 75µm...' (PRODUCT AVAILABILITY, p.1 of the Thin SUEX Data Sheet, Rev 6/2020) — Standard Thicknesses: 20, 25, 30, 50 and 75 µm (Table 1, p.2 additionally tabulates a 40 µm condition not listed among 'Standard Thicknesses'). Thick SUEX: 'Sheets are available in thicknesses from 100µm to 1mm...' (PRODUCT AVAILABILITY, p.1 of the Thick SUEX Data Sheet, June 2020) — Standard Thicknesses: 100, 125, 150, 200, 225, 250, 300, 350, 400 and 500 µm, with 'Custom thicknesses up to 1mm...available upon special request' stated in the same section. min_um=20 is the Thin sheet's stated floor; max_um=1000 is the Thick sheet's own stated ceiling (1 mm), taken verbatim from its opening availability sentence rather than the 500 µm standard-catalog subset." }, "coatingMethod": "dry-film-lamination", "spinCurves": [], "spinNotes": "SUEX is laminated onto the substrate with a heated roll laminator, not spin-coated — there is no spin curve, and thickness is fixed per sheet (see thicknessRange). Recommended hot-roll lamination conditions, stated identically in both the Thick and Thin datasheets: roller/plate temperature 60-70°C for all rolls (Thin datasheet: 'for all rolls and plates'); pressure 5-10 psi (30-65 kPa); speed 0.5-1.5 ft/min (0.15-0.5 m/min), with thicker films requiring slightly slower speeds. Source: 'LAMINATION' / 'Recommended conditions for hot roll lamination', p.1 of both the Thick SUEX Data Sheet (June 2020) and the Thin SUEX Data Sheet (Rev 6/2020). Pre-lamination substrate prep (both datasheets, 'SUBSTRATE PREPARATION', p.1) calls for the substrate to be free of organic contamination and metal oxides and cleaned/dried immediately before lamination, and mentions a 'dehydration bake' as sometimes needed before further cleaning/surface activation — but no dehydration-bake temperature or time is published anywhere in either datasheet, so no numeric pre-lamination substrate temperature is recorded. The clear PET cover sheet is removed immediately before lamination; sheets must not touch the substrate until <1 cm before the rollers; vacuum lamination is recommended over topography for thick films.", "adhesion": { "hmds": false, "notes": "'Adhesion promoters are typically not useful.' Source: 'SUBSTRATE PREPARATION', p.1 of both the Thick and Thin SUEX Data Sheets." }, "rehydration": null, "softbake": { "temp_c": null, "time_s": 300, "method": "hotplate", "notes": "Both datasheets call this step the 'post lamination bake' (PLB), not a softbake in the classic solvent-drying sense — the laminated sheet arrives solvent-free. Both state it is 'normally not needed or recommended', but that 'for improved adhesion and surface quality, the laminated article may be baked on a hotplate at 80 - 85°C for 5 minutes.' Temperature is published only as an 80-85°C range and is left null here to avoid false precision; the 5-minute duration is a single stated value.", "source": "'BAKE', p.1 of both the Thick SUEX Data Sheet (June 2020) and the Thin SUEX Data Sheet (Rev 6/2020)" }, "exposureDose": { "doses": [], "datasheetBasis": "'The solvent developed negative working photoresist is sensitive to UV radiation in the range of 350 – 395nm' (both datasheets, product description, p.1). 'For optimum resolution i-line filters or soda lime masks are highly recommended to remove wavelengths below 350nm for improved resolution and sidewall acuity' ('EXPOSURE', p.1). Both dose tables (Table 1, p.2) explicitly label their two exposure-dose rows 'mJ/cm2 @ 365 nm' — one for an i-line/UV-filtered exposure, one for an unfiltered exposure.", "_note": "Dose is published only as a THICKNESS-BINNED table (Table 1) per product line, never as a single scalar for 'SUEX' overall, so no single value_mJcm2 is recorded here (false-precision risk, same shape as the SU-8 2000 family). Thick SUEX (100-500 µm), @365nm: UV-filtered dose 1000 mJ/cm² (100 µm), 1200 (150 µm), 1350 (200 µm), 1500 (250 µm), 1900 (350 µm), 2500 (500 µm); no-filter dose 450 (100 µm), 515 (150 µm), 575 (200 µm), 675 (250 µm), 900 (350 µm), 1150 (500 µm). Thin SUEX (20-75 µm), @365nm: UV-filtered dose 720 mJ/cm² (20 µm), 725 (25 µm), 760 (30 µm), 790 (40 µm), 825 (50 µm), 920 (75 µm); no-filter dose 325 (20 µm), 320 (25 µm), 335 (30 µm), 348 (40 µm), 370 (50 µm), 430 (75 µm). All figures are 'estimated doses on silicon' per the EXPOSURE section and will vary with substrate/process/mask. Source: Table 1, p.2 of both the Thick SUEX Data Sheet (June 2020) and the Thin SUEX Data Sheet (Rev 6/2020)." }, "peb": { "temp_c": null, "time_s": null, "notes": "Published only as THICKNESS-BINNED tables (Table 1), and the two product lines differ. Thick SUEX (100-500 µm) bakes at a constant 85°C with time scaling by thickness: 30 min (100 µm), 30 min (150 µm), 35 min (200 µm), 40 min (250 µm), 45 min (350 µm), 45 min (500 µm). Thin SUEX (20-75 µm) bakes for a constant 5 min across all thicknesses at 85-95°C. The general PEB prose (worded identically in both datasheets) separately recommends 'a 85°C oven bake for 30-60 minutes for lower stress or a 95°C bake for 5–10 minutes for speed', then to 'cool slowly over 3-5 hours for minimum cracks and best adhesion.' The PET coversheet must be removed before PEB.", "source": "Table 1 'Process conditions for Thick/Thin SUEX TDFS on Silicon Wafers', p.2; 'POST EXPOSURE BAKE (PEB)', p.1 — both the Thick SUEX Data Sheet (June 2020) and the Thin SUEX Data Sheet (Rev 6/2020)" }, "floodExposure": null, "develop": { "developer": "PGMEA (propylene glycol methyl ether acetate), two-bath system", "dilution": null, "time_s": null, "method": "immersion", "rinse": "IPA", "source": "'DEVELOPMENT' and 'RINSE/DRY', p.1; Table 1 'Devl Time Face Down', p.2 — both the Thick SUEX Data Sheet (June 2020) and the Thin SUEX Data Sheet (Rev 6/2020)", "_note": "Develop at room temperature in a two-bath PGMEA system, face down with mild agitation; no dilution ratio is published (PGMEA appears to be used as supplied). Develop time is thickness-binned (Table 1), not a single value: Thick SUEX — 35 min (20+15) at 100 µm, 50 min (30+20) at 150 µm, 60 min (40+20) at 200 µm, 75 min (50+25) at 250 µm, 100 min (70+30) at 350 µm, 140 min (100+40) at 500 µm. Thin SUEX — 12 min (10+2) at 20 µm, 15 min (10+5) at 25 µm, 15 min (10+5) at 30 µm, 20 min (15+5) at 40 µm, 20 min (15+5) at 50 µm, 25 min (20+5) at 75 µm. Rinse: wash the developed wafer in isopropyl alcohol, then immerse in clean IPA for 1-5 minutes." }, "hardbake": { "temp_c": null, "time_s": null, "notes": "Optional. The datasheet gives discrete named options rather than a single condition: 125°C/60 min, 150°C/30 min, or 150°C/15-20 min. Separately, 'a 200°C hard bake for 30 – 60 minutes is required to remove all volatile components from the film' if full outgassing is needed.", "source": "'HARD BAKE (Optional)', p.1 of both the Thick SUEX Data Sheet (June 2020) and the Thin SUEX Data Sheet (Rev 6/2020)" }, "descum": null, "applications": [ "mems-structural", "electroplating-molding" ], "etchResistance": null, "liftoffSuitable": null, "platingSuitable": true, "stripper": "SUEX is not intended to be removed once processed: 'SUEX is generally used as a permanent highly cross-linked film and is not intended to be removed.' For film that has not been hard baked, an NMP-based remover may lift it from the substrate; hard-baked film is generally removed only with CO2 laser ablation equipment. Source: 'REMOVAL', p.1 of both the Thick SUEX Data Sheet (June 2020) and the Thin SUEX Data Sheet (Rev 6/2020).", "storage": "Store in the original black packaging in a standard, temperature-controlled environment between 18°C (65°F) and 25°C (77°F); shelf life is up to 2 years from date of manufacture under those conditions. Source: 'STORAGE', p.1 of both the Thick SUEX Data Sheet (June 2020) and the Thin SUEX Data Sheet (Rev 6/2020).", "notes": "SUEX is DJ MicroLaminates' cationically-cured epoxy dry-film photoresist, supplied as thin (20-75 µm) and thick (100 µm-1 mm) laminate sheets rather than spin-coated, making it a common choice for high-aspect-ratio MEMS structures, wafer-level-packaging molds, and electroplating templates where single-pass spin-coat thickness limits become impractical. Because it laminates onto the substrate via a heated roller rather than a spindle, its process is defined by roller temperature, speed and pressure instead of a spin curve, and its exposure dose, post-exposure bake and develop time are all published as thickness-binned tables rather than single values — treat the row matching the target sheet thickness as the starting point. The datasheet explicitly identifies the cured film as a permanent, highly cross-linked structural layer 'not intended to be removed', strippable only with an NMP-based remover before hard bake or by CO2 laser ablation after, so it is suited to permanent-structure and plating-mold applications rather than a sacrificial lift-off role.", "developerFamily": "solvent", "provenance": { "datasheetUrl": "https://djmicrolaminates.com/wp-content/uploads/2020/06/Thick-SUEX-Data-Sheet-June-2020.pdf", "datasheetVersionOrDate": "June 2020 (Thick SUEX Data Sheet, Rev June 2020; sibling Thin SUEX Data Sheet, Rev 6/2020)", "accessedDate": "2026-07-11", "secondarySources": [ { "url": "https://djmicrolaminates.com/wp-content/uploads/2020/06/Thin-SUEX-Data-Sheet-June-2020.pdf", "what": "Thin SUEX (20-75 µm) Data Sheet, Rev 6/2020 — the source of the thin-film Table 1 process conditions and the 20-75 µm stated thickness range; read jointly with the Thick datasheet since 'SUEX' is one product family spanning both sheets." } ] }, "_provenanceNote": "This recipe combines TWO sibling DJ MicroLaminates datasheets covering one product family under one brand name: the Thick SUEX Data Sheet (100 µm-1 mm, Rev June 2020) and the Thin SUEX Data Sheet (20-75 µm, Rev 6/2020). Both are cationically-cured modified-epoxy SUEX with identical prose in nearly every processing section (lamination, PEB guidance, hard bake, removal, plating, storage); only the per-thickness Table 1 values and the covered thickness band differ. The Thick datasheet is recorded as the primary datasheetUrl (it closes the >100 µm dry-film gap named in DRY-FILM-RESEARCH.md §5); the Thin datasheet is a secondarySource, not a lesser document.", "tier": "datasheet", "dateAdded": "2026-07-11", "dateModified": "2026-07-11", "humanVerified": false } ] }