https://nanyte.com/photoresists/az-15nxt · last updated 2026-09-28
- Manufacturer
- AZ Electronic Materials
- Tone
- negative
- Chemistry
- Chemically amplified
- Thickness
- 6.4–14.6 µm
- Exposure dose
- 400 mJ/cm² at 365 nm
- Developer
- AZ 300 MIF
- Applications
- Etch mask · Electroplating / molding · High aspect ratio
Cross-checked — two independent extractions agree on the spin curve and the single-value figures.
- Substrate
- Resist
- Exposed
Spin coating
Data points
| Series | rpm | µm |
|---|---|---|
| AZ 15nXT (450 CPS) | 1000 | 15 |
| 1500 | 11 | |
| 2000 | 9.4 | |
| 2500 | 8.3 | |
| 3000 | 7.4 | |
| 4000 | 6.4 |
Values are the manufacturer’s starting points, not a guarantee — characterize on your own tool. Series digitized from a published figure are approximate (±10 %).
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.
- 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'), so nothing is claimed either way for this grade; 'not required' cannot be inferred just because this is a chemically amplified resist.
- The document disagrees with itself on thickness: the 'Spin Speed Curve' (p.7, plotted 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.
- The curve plotted here is the one the document captions as its spin-speed curve; real coats in this deck ran measurably thicker in two other places.
- Adhesion
- Not discussed. Process examples in the document run directly on Si (photospeed testing) and Cu (imaging) wafers with no HMDS priming step mentioned.
Soft bake
- Soft bake
- 110 °C · 3 min
- 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 an oven step.
SOURCE: p.5 (Process Conditions); corroborated pp.7,8,11,16-19
Exposure dose
The manufacturer publishes 400 mJ/cm² at 365 nm. Dose scales with film thickness and depends on your optics, so treat it as a starting point and run a dose array.
- Dose at 365 nm
- 400 mJ/cm²
- Dose at 405 nm
- Not published — characterize on-tool
- As published
- 400 ± 50 mJ/cm² on an i-line stepper is the reference dose the whole document is built around, including its delay studies and exposure-latitude work.
- Post-exposure bake
- 120 °C · 60 s
SOURCE: p.5; corroborated as the boxed/highlighted reference dose in Exposure Latitude figures pp.11,22
Development
- Developer
- AZ 300 MIF
- Dilution
- 2.38% TMAH (AZ 300 MIF, ready-to-use)
- Time
- 2.5 min
- Method
- 3 × 50 s spray/puddle at 23 °C
- Rinse
- Not published — characterize on-tool
- Developer family
- TMAH-based
SOURCE: p.5 (Process Conditions); corroborated 'AZ 300 MIF/3x50sec Spray/Puddle @23°C' pp.11,16-19
Hard bake, etch & strip
- Etch resistance
- 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.
- 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')
Not published for this resist: Hard bake, Descum, Storage — characterize on-tool.
Where it's used
Practical notes from the datasheet
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 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.
Troubleshooting
Common failure modes for AZ 15nXT (450 CPS), answered from the manufacturer's datasheet and application notes. These are starting points — your substrate, tooling and environment shift the specifics, so calibrate on-tool.
How long can AZ 15nXT sit between exposure and PEB, or between coat and exposure, before CD drifts?
On 10 µm film on Cu, the datasheet's delay studies show 5 µm lines holding within roughly 4.8–5.05 µm at the reference dose for exposure-to-PEB delays up to 7 h, and within about 4.8–5.0 µm for coat-to-expose delays up to about 23.5 h. The change comes early: CD dips by roughly 0.1–0.2 µm within the first hour, then stays roughly flat. MicroChemicals' trouble shooter is stricter and advises keeping exposure-to-PEB and PEB-to-develop intervals to a few minutes, so short, consistent delays are still the safe schedule.
SOURCE: 15nXT (450 CPS) datasheet
AZ 15nXT (450 CPS) datasheet (AZ Electronic Materials, January 2009) — Coat to Exposure Delay on Copper, pp.16–17; Exposure to PEB Delay on Cu Wafers, pp.18–20; Photolithography Trouble Shooter (MicroChemicals) — Critical Negative Resists, p.9
My 5 µm AZ 15nXT lines print wider than the mask — is it dose?
Dose is the first thing to check. As a negative resist, AZ 15nXT lines grow with dose. On 10 µm film on Cu (i-line stepper, focus +1.0 µm), 5 µm lines/spaces measured roughly 4.3–4.6 µm at 150 mJ/cm² and 5.5 µm or more by 500–550 mJ/cm²; the lines crossed 5.0 µm at roughly 250–300 mJ/cm². The datasheet's headline condition is 400 ± 50 mJ/cm², and it shows linearity at both 300 and 400 mJ/cm². Run a dose series on your own substrate and pick the dose that sizes your critical feature.
SOURCE: 15nXT (450 CPS) datasheet
AZ 15nXT (450 CPS) datasheet (AZ Electronic Materials, January 2009) — Process Conditions, p.5; Linearity, Resolution @ 400 mJ/cm², p.10; Exposure Latitude, pp.11–12; L/S Linearity @ 300 mJ/cm², p.13; CD vs dose curves, pp.17, 19
Small AZ 15nXT features lift off or erode in the developer — what's wrong?
Suspect under-crosslinking first. AZ 15nXT is chemically amplified: exposure generates an acid catalyst, and the crosslinking that makes the film insoluble happens during PEB. The datasheet's process runs PEB at 120 °C for 60 s on a contact hotplate. MicroChemicals' trouble shooter warns that too low a dose, or a PEB that is too cool or too short, keeps the resist near the substrate soluble, so small structures can lift during development. Check the hotplate temperature at the wafer and the dose before you blame adhesion.
SOURCE: 15nXT (450 CPS) datasheet
AZ 15nXT (450 CPS) datasheet (AZ Electronic Materials, January 2009) — Chemically Amplified Negative Resist, p.4; Process Conditions, p.5; Photolithography Trouble Shooter (MicroChemicals) — Inferior Resist Adhesion: Image Reversal- or Negative Tone Resists, p.12
Why is my 17 µm AZ 15nXT coat so non-uniform with a heavy edge bead?
It's spun too slowly. On the datasheet's wafer maps (3 mm edge exclusion), 1000 rpm gave 17.8 µm at 4.49 % RSD, against 1.36 % at 1500 rpm and 0.70 % at 3000 rpm. In its 200 mm edge-bead study the 1000 rpm film rose from about 17.7 µm in the bulk to about 25 µm 1 mm from the edge, while at 3000 rpm the edge rise was under 2 µm. Below 1500 rpm uniformity falls off sharply, and the edge rise stays large up to 1500 rpm (about 13.5 to 21.5 µm on the same chart); only by 3000 rpm does it drop under 2 µm. Budget edge exclusion for any slow-spun coat.
SOURCE: 15nXT (450 CPS) datasheet
AZ 15nXT (450 CPS) datasheet (AZ Electronic Materials, January 2009) — Coating Uniformity Wafer Maps, p.14; Edge-Bead Coating Study (200 mm wafer), p.15
AZ 15nXT won't strip cleanly after plating — what remover and conditions?
After Ni and Cu plating, the datasheet strips AZ 15nXT completely in AZ Kwik Strip at 70 °C for 3 min. If the film has crosslinked heavily from high process temperatures, MicroChemicals recommends TechniStrip NI555 for Novolak-based negatives such as AZ 15nXT. On Cu, residue at the bottom of the features is less likely with this grade: the trouble shooter lists AZ 15nXT as Cu-compatible, where other crosslinking negatives can stop developing some 10–100 nm short of the metal.
SOURCE: 15nXT (450 CPS) datasheet
AZ 15nXT (450 CPS) datasheet (AZ Electronic Materials, January 2009) — Ni/Cu Plating Compatibility, p.21; Summary, p.23; Photolithography Trouble Shooter (MicroChemicals) — Negative Resist Residuals, p.17; Crosslinked Negative Resists, p.20
Sources & disclaimer
- AZ Electronic Materials — AZ 15nXT (450 CPS) datasheet (January 2009) · accessed 2026-07-10
- Tatikonda et al.. Sacrificial Layer Technique for Releasing Metallized Multilayer SU-8 Devices. Micromachines (2018). doi:10.3390/mi9120673AZ 15nXT is proposed as a sacrificial layer that stays chemically stable through SU-8 processing; the authors give two cross-linking routes and three release routes to free metallized multilayer SU-8 devices such as an electrospray-ionization chip.
Manufacturer datasheet values are starting points; optimal parameters depend on your substrate, equipment and environment. Product names and trademarks belong to their respective owners. NANYTE is not affiliated with the manufacturers listed. Last updated 2026-09-28.