https://nanyte.com/photoresists/ti-35e · last updated 2026-09-28
- Manufacturer
- MicroChemicals GmbH
- Tone
- image reversal
- Thickness
- 3–4.5 µm
- Exposure dose
- 140 mJ/cm² (broadband, calibrated at 365 nm)
- Developer
- AZ 826 MIF (metal-ion-free), or a potassium- or sodium-based developer — AZ 400K diluted 1:4, or AZ 351B
- Applications
- Lift-off · Image reversal · Etch mask
Cross-checked — the spin curve (where published) and the single-value figures were read twice, independently, from the datasheet, with any disagreement settled against it.
- Substrate
- Resist
- Exposed
Spin coating
The manufacturer does not publish a spin curve for TI 35E.
- No spin curve is plotted or tabulated.
- The datasheet gives the 3.0-4.5 µm range and one anchor inside it — 'With spin-speed of 3000 rpm a resist thickness of 3.6 µm will be achieved' (p.1) — which is a single point, not a sweep.
- The only other coating instruction is to hold the final speed for at least 30-40 s.
- Accel, dispense volume and edge-bead removal are not published.
- Note the edge effect the document does flag: on square substrates the film at the edges runs significantly thicker than 3-4 µm, which is why it tells you to double the post-exposure delay there.
- Adhesion
- HMDS recommended — Two routes are given. Dehydrate on a hotplate at a minimum of 120 °C for 10 min (or 30 min in a furnace at the same temperature) to drive off adsorbed water, or prime with HMDS — 'only from vapor phase at an optimum substrate temperature of 125°C'. For SiNx surfaces the sheet adds that using ammonia during the PECVD deposition improves resist adhesion.
Soft bake
- Soft bake
- 95 °C · 2 min · hotplate
- Notes
- 95 °C for 2 min on a hotplate. A furnace alternative of 95 °C for 20 min is given.
SOURCE: Processing the TI 35E, p.4
Exposure dose
The manufacturer publishes 140 mJ/cm² (broadband, calibrated 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 405 nm
- Not published — characterize on-tool
- As published
- 140 mJ/cm² is the typical first exposure, with 100-250 mJ/cm² the working range. The lamp is broadband or g/h/i-line; the number is metered on the 365 nm line rather than specified as a monochromatic i-line dose.
- Hold
- ≥10 min at room temperature
- Reversal bake
- 115–120 °C · 2 min · hotplate
- Flood exposure
- 540 mJ/cm²
SOURCE: Processing the TI 35E, p.4 ('at a dose of 140 mJ/cm2') and the process overview table, p.6 ('100 .. 250, 140 (typ.)')
SOURCE: p.4: 'Keep the coated substrate at room temperature after the exposure for at least 10 minutes'
Development
- Developer
- AZ 826 MIF (metal-ion-free), or a potassium- or sodium-based developer — AZ 400K diluted 1:4, or AZ 351B
- Dilution
- AZ 400K at 1:4; no dilution is stated for AZ 826 MIF or AZ 351B.
- Developer family
- TMAH or buffered alkaline
Not published for this resist: Time, Method, Rinse — characterize on-tool.
SOURCE: Processing the TI 35E, p.5, and the process overview table…
Processing the TI 35E, p.5, and the process overview table, p.6 ('Developer — Clariant AZ 351B, MIF Developers e.g. AZ 826')
Hard bake, etch & strip
- Hard bake
- 140–145 °C · 2 min · hotplate (optional — only for hard etching — avoid for lift-off)
- Etch resistance
- Built for wet chemistry. In 12.5 % HF buffered oxide etch, after a 125 °C / 2 min hardbake, the sheet reports no visible degradation over 10 min and structuring of SiNx to at least 100 nm. On thick thermal SiO2 in 20 % HF it again reports no visible degradation after 10 min, and says concentrated 50 % HF is possible for several minutes. For direct mesa grooving the TI series etches Si without an SiO2 mask, using HNO3 (70 %) : HF (50 %) : H2O at 75 : 10 : 25 to cut a groove up to 4 µm deep in 45 s.
- Stripper
- AZ 100 remover, or acetone. For lifting off the deposited film the sheet lists PGMEA, NMP, MMP, EEP, DMF, acetone and ethyl lactate.
Not published for this resist: Descum, Storage — characterize on-tool.
SOURCE: Processing the TI 35E, p.5, the fields-of-application section, p.3, and the process overview table, p.6
Where it's used
Practical notes from the datasheet
TI 35E is run in image-reversal mode, which is why the exposure mask has to be inverted: the areas exposed through the mask are the ones that survive. The sequence adds two steps to an ordinary positive process — the reversal bake and a maskless flood exposure — and needs no new equipment beyond that. The payoff is the negative wall profile, and the first exposure is the knob that sets it: 140 mJ/cm² suits most work, and lowering it opens the undercut further, at the cost of eroding the areas that are supposed to clear if taken too far. At 3.6 µm coated from 3000 rpm the sheet rates the undercut for lifting off evaporated films up to 6 µm thick, with typical feature aspect ratios of 1.6 to 2.0; sputtered films reach roughly 1 µm instead, because the deposition is not directional, and may need ultrasonic help. One step is easy to skip and shouldn't be: after the first exposure the wafer rests at room temperature for at least 10 minutes so the nitrogen generated during exposure can diffuse out, doubled for square substrates whose edge bead runs thicker. The other trap is the reversal bake temperature, which the datasheet flags as the critical one and the reason it steers you off furnace baking.
Troubleshooting
Common failure modes for TI 35E, 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.
My TI 35E image-reversal results change from run to run — how tight is the reversal bake window?
Tight. The datasheet calls the reversal bake very temperature-critical: 115–120 °C for 2 min on a hotplate, and it advises against a furnace. MicroChemicals' image-reversal note asks for ±1–2 °C. Too cool, and even the exposed surface does not convert, so the film erodes in the developer and loses dimensional accuracy. Too hot converts the weakly exposed resist near the substrate, which shrinks the undercut and slows development. Thick, massive or poorly conducting substrates need their own bake optimisation.
SOURCE: TI 35E technical data sheet
TI 35E technical data sheet (MicroChemicals, revised 08/2016) — Processing the TI 35E, p.4; Overview, p.6; Image Reversal Resists and Their Processing (MicroChemicals) — The Image Reversal Bake Step, Influence of Substrate and Equipment, p.2
Not enough undercut on TI 35E for lift-off — what do I change first?
The first (masked) exposure dose. The datasheet says it sets the undercut: lower it for more undercut, with 140 mJ/cm² (i-line calibrated) a good choice for most applications. Don't go too low, though: an underdosed first exposure dramatically increases erosion of the resist that should stay. Then use over-development. Once the pattern has cleared, keep developing for another 10–30 % of the total time to finish the sidewall profile. Check narrow resist lines don't lift as the undercut grows.
SOURCE: TI 35E technical data sheet
TI 35E technical data sheet (MicroChemicals, revised 08/2016) — Processing the TI 35E, pp.4–5; Image Reversal Resists and Their Processing (MicroChemicals) — The First Exposure, p.1; Formation of the Undercut, p.3
Crater-like defects show up in TI 35E after development — where do they come from?
Nitrogen bubbles. TI 35E is DNQ-based and releases N2 on exposure; if the reversal bake softens the film before the gas has diffused out, bubbles form and later appear as craters. The datasheet asks for at least 10 min at room temperature between the first exposure and the reversal bake, doubled on square substrates where the edges coat much thicker than 3–4 µm. If a later deposition heats the film above 80 °C, or a plasma supplies UV, use a high flood dose to avoid N2 bubbles then.
SOURCE: TI 35E technical data sheet
TI 35E technical data sheet (MicroChemicals, revised 08/2016) — Processing the TI 35E, p.4; Image Reversal Resists and Their Processing (MicroChemicals) — Formation of Bubbles in the Resist Film, p.2
TI 35E areas that should clear stay partly undeveloped after the flood exposure — is the flood dose too low?
Possibly, though the flood is forgiving. The datasheet's typical flood is 540 mJ/cm² without a mask, and anything from 200 to 800 mJ/cm² makes no dominant difference. MicroChemicals' image-reversal note adds that overexposure does not damage the process, since the reversal bake has made the first-exposed areas insensitive, though a high flood dose can release enough nitrogen to form bubbles or cracks. The film is also anhydrous after the reversal bake and must rehydrate before the flood, especially above about 3 µm. Develop in AZ 826 MIF, or AZ 400K diluted 1:4.
SOURCE: TI 35E technical data sheet
TI 35E technical data sheet (MicroChemicals, revised 08/2016) — Processing the TI 35E, pp.4–5; Overview, p.6; Image Reversal Resists and Their Processing (MicroChemicals) — The Flood Exposure, p.3
After hard baking TI 35E, lift-off stopped working — why?
The hard bake removed the undercut. The datasheet reserves the hard bake (140–145 °C, 2 min on a hotplate) for harsh etch masks and mesa grooving, and warns the sidewall loses its undercut, which makes lift-off more problematic. For lift-off, skip it. If you need a hardened film and lift-off, or very high deposition temperatures, UV-cure the resist after development instead. Lift off in PGMEA, NMP, MMP, EEP, DMF, acetone or ethyl lactate.
SOURCE: TI 35E technical data sheet (MicroChemicals, revised 08/2016) — Processing the TI 35E, p.5; Overview, p.6
Sources & disclaimer
- MicroChemicals GmbH — TI 35E datasheet (revised 08/2016) · accessed 2026-09-10
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.