Cleanliness, Residue and Contamination Verification

Photoresist Adhesion on Silicon Wafer Substrates: Verify Wafer Cleaning and HMDS Priming With Contact Angle Measurement Before Photolithography and Etch

Resist that lifts during development or wet etch usually starts with a wafer surface that was not clean or not primed right. Measure water contact angle after cleaning and after HMDS priming, against gates from your own process, before you spin the resist.

Who this is for: Process, lithography and yield engineers in semiconductor manufacturing, MEMS and university nanofabrication labs.

Where it fits: After wafer cleaning, after priming and before resist coating, on monitor wafers in the fabrication process.

What it does not do: It does not count particles, measure metal contamination or replace lithography and etch inspection.

Method

Sessile drop water contact angle at set sites; surface free energy with a second liquid

Standard

Fully compliant with SEMI/ASTM D7490-13 (reapproved 2022); complements ASTM F22

Setup

About 2 minutes

Book a 15 minute demo Request a quote

Written by

Abhimanyu Bhandankar

Holds an MBA from Schulich School of Business and a BE in IT. He joined Droplet Lab in July 2019 and now leads sales and marketing.

CEO at Droplet Lab

Why wafer surface state decides photoresist adhesion

65° to 80°

water contact angle range a university nanofab cites for the best microlithography results after HMDS priming; another targets 65° to 75°.

UT Dallas Research Cores; EPFL Center of MicroNanoTechnology, HMDS process pages

Below 5° vs 16.2°

water contact angle on silicon native oxide after a helium and oxygen plasma clean versus an RCA SC-1 clean; both angles then rose over several days.

Williams and Hicks, AVS 58th International Symposium, 2011

Half a monolayer

of carbon contamination picked up from clean room air was enough to affect gate oxide integrity on silicon wafers.

Imafuku et al., MRS Proceedings 473, 1997

Sources: UT Dallas Research Cores, HMDS process; EPFL CMi, HMDS process page; Williams T.S. and Hicks R.F., AVS 58th International Symposium, 2011; Imafuku D. et al., Organic contamination of silicon wafer in clean room air and its impact to gate oxide integrity, MRS Proceedings, 1997. Figures come from these sources, not from Droplet Lab measurements.

Quick reference

What this wafer surface check does and what it does not

A quick reference for lithography and process teams checking fit before reading further.

Evidence Box

Problem this solves

Photoresist that lifts, undercuts or peels during development or wet etch because the wafer was not clean or not primed correctly.

Standards

Fully compliant with SEMI/ASTM D7490-13 for surface free energy from two liquids; complements the ASTM F22 water break test.

Peer reviewed method

The Dropometer drop shape method was validated against a KRUSS DSA100E in two peer reviewed papers, in Review of Scientific Instruments and Colloids and Surfaces A.

Customer case study

A US medical device manufacturer checks surfaces before coating with contact angle, measuring a drop on a mandrel about 1 mm across.

Published research

Researchers used a Droplet Lab tensiometer to show vacuum plasma cut the contact angle on machined titanium from 83.1° to 24.1°.

Honest limit

Contact angle shows organic films and priming state. It does not detect particles or metals.

Who this is for

What wafer surface question are you trying to answer?

Pick the card closest to your role. Each one points to the part of this page that answers it.

Lithography process engineer

Resist lifts at the edges of fine lines during development or wet etch. Water contact angle on monitor wafers after cleaning and after priming shows whether surface preparation is the cause, before you change exposure or bake.

Measure soon after each step; clean wafers pick up hydrocarbons from the air.

Yield or quality engineer

You need a recorded check that every lot was cleaned and primed inside the process window for reliable adhesion. A contact angle gate on a monitor wafer gives each lot a pass, reclean or reprime decision.

Set gates from your own process and resist supplier guidance, not from another fab's numbers.

University nanofab manager

In a shared fabrication lab, many users share the cleaning benches and priming oven. A quick contact angle check shows when the HMDS supply runs low or a recipe has drifted.

Primed angles below your band can mean the priming source needs refilling.

Fit check

Is this the right check for your wafer process?

It measures the wafer surface state that decides resist adhesion. It does not inspect patterns.

Good fit if

You need to verify wafer cleaning before priming or deposition.

You prime with HMDS vapor and want a check on every lot before the resist coating process.

Resist lifts or undercuts during development or wet etch.

You run several substrates, such as silicon dioxide, silicon nitride or glass.

You need surface free energy of a substrate surface from two liquids.

Less relevant if

You need particle counts or metal contamination levels.

You need inline measurement on every product wafer inside a tool.

You need film thickness or overlay data.

Your features are too small to place a drop on a test area.

Summary

How to verify wafer cleaning and HMDS priming

The answer in under a minute.

Measure water contact angle on a monitor wafer right after cleaning, where a clean oxide surface wets, and again after HMDS priming, where the angle should sit inside your process band. University nanofab pages cite about 65° to 80° after priming. Reclean or reprime outside the gates, then coat promptly.

The Dropometer contact angle kit runs the check at the bench; our contact angle measurement guide and wettability glossary entry explain the method. Surface free energy from two liquids follows SEMI/ASTM D7490. A US medical device manufacturer uses contact angle to check surfaces before coating, and researchers measured plasma cleaning of titanium with Droplet Lab equipment. Our semiconductors guide covers the wider context.

Expert Quote

Expert perspective on wafer surfaces

Because only the top few ångström decide wetting, an invisible contaminant film, even a monolayer picked up from the air, is enough to change how a liquid behaves on the surface.

The problem

Why photoresist adhesion fails on wafers

Patterning of semiconductor devices depends on resist. In the photolithography process, a photoresist layer is spin coated onto the wafer, typically at a spin speed of a few thousand rpm, exposed through optical tools to ultraviolet light, developed, and then used as a mask for the etching process or deposition, sometimes over a hard mask layer. If the resist does not adhere, developer or wet etchant creeps under the edges, lines lift or undercut, and pattern transfer fails. The adhesion of photoresist depends on the wafer surface. A freshly cleaned silicon dioxide (SiO2) surface is covered in hydroxyl groups and adsorbed water, which most resists do not stick to, so fabs prime it with HMDS (hexamethyldisilazane), an adhesion promoter applied as a vapor that makes the surface less hydrophilic. Two things go wrong. The cleaning steps leave organic residue, or the clean wafer picks up airborne hydrocarbons while it waits; half a monolayer of carbon from clean room air has been shown to affect gate oxide integrity. Or priming is off: too little because the wafer was not dehydrated or the HMDS supply ran low, or too much, for example when liquid HMDS is spun on, which MicroChemicals warns can crosslink the resist. Both show up as a change in water contact angle long before they show up as a failed pattern.

Resist lines lift or peel during development.

Undercut or lateral etching under the resist edge in wet etch.

Adhesion problems that come and go between lots or tools.

Edge of the wafer fails while the centre passes.

Resist is hard to develop or leaves residue near the substrate.

Troubleshooting

Why resist adhesion fails and what to check

Why:

Residue from resist strip, solvent or surfactant containing cleans, or handling keeps the surface from wetting, and HMDS cannot bond evenly. Our guide to sample prep and contamination control covers clean handling.

How to detect:

Water contact angle after cleaning is higher than your clean baseline.

Corrective action:

Repeat the cleaning process, for example piranha followed by RCA cleaning or an oxygen plasma, and remeasure.

Why:

Clean, hydrophilic wafers adsorb hydrocarbons from clean room air, and the angle rises with exposure time.

How to detect:

The angle rises with time between cleaning and priming.

Corrective action:

Set a maximum wait time, store wafers covered, and prime soon after cleaning.

Why:

Adsorbed water blocks HMDS from reacting with the surface.

How to detect:

The primed angle is below your band, often unevenly across the wafer.

Corrective action:

Add or lengthen the dehydration bake before vapor priming.

Why:

A depleted source or changed oven recipe gives a weaker priming process. EPFL notes angles below 65° indicate the bubbler should be refilled.

How to detect:

Primed angles trend down across lots from the same tool.

Corrective action:

Refill or service the priming source, check the recipe, and remeasure.

Why:

A thick HMDS layer from spin coating releases ammonia that can crosslink the resist near the substrate.

How to detect:

Primed angles above your band, and resist that is hard to develop. Our experiment on treated and untreated surfaces shows how strongly treatment changes angles.

Corrective action:

Prime from the vapor phase only, and shorten priming time if angles run high.

Why:

Flow and temperature differences in tools leave the edge or centre under treated.

How to detect:

Centre, mid radius and edge sites differ. Our contact angle reference values for common surfaces give a sense check.

Corrective action:

Check tool flow, temperature and wafer position, then remeasure all sites.

Not sure whether surface preparation is the cause?

Bring a good and a failing wafer to a call, and we will compare the surfaces on screen.

Book a 15 minute demo

Process records

What a wafer surface preparation record contains

Each check produces a record you can file with the lot. These records can feed into your existing process control and surface quality records.

Lot identity

Lot, wafer, substrate film and the cleaning and priming tools used.

Timing

Time of cleaning, priming and measurement, and time to resist coating.

Method settings

Drop volume, read time and sites: centre, mid radius and edge.

Readings

Water contact angle at each site after cleaning and after priming.

Decision

Coat, reclean or reprime, with the operator and date.

Drop images

Automatic edge and baseline detection gives the same reading from the same image, so a reviewer can recheck any result.

Measurements

What to measure for surface preparation and process control

After cleaning

Water contact angle after surface cleaning

Why it matters: A clean oxide surface wets; organic residue raises the angle.

How to interpret: Values depend on the clean: near 0° after plasma and around 16° to 21° after wet cleans in published work. Set your gate from your own clean wafers.

When it is not enough: It does not show particles or metals.

After priming

Water contact angle after HMDS priming

Why it matters: It shows whether the prime took, and whether it is too weak or too strong.

How to interpret: University nanofab pages cite about 65° to 80°. Follow your resist supplier and set your band from lithography results.

When it is not enough: It does not show resist thickness or defects.

Uniformity

Centre to edge spread

Why it matters: Uneven cleaning or priming causes edge failures.

How to interpret: Compare sites on the same wafer; a large spread points to the tool.

When it is not enough: It samples sites, not every die.

Timing

Angle against waiting time

Why it matters: It sets the safe window between steps.

How to interpret: Plot the rise after cleaning and the decay after priming.

When it is not enough: Storage conditions change the rate.

Validation

Validated measurement approach for wafer surface checks

How the Dropometer itself has been validated, and where customers have used contact angle to verify cleaning.

Peer reviewed method

The Dropometer drop shape method is published in Review of Scientific Instruments and Colloids and Surfaces A. Contact angle was validated against a KRUSS DSA100E, with a published accuracy of 0.35°.

See the validation papers

Customer evidence on cleaning before coating

A US medical device manufacturer measured a water drop on a mandrel about 1 mm across during a live demo, bought a Dropometer to check surfaces before coating, and ordered a second unit within seven months.

Read the case study

QC Protocol

How to check wafer cleaning and HMDS priming

Six steps for one wafer type and one priming recipe. Repeat for each new substrate or tool.

1

Lock the method

Fix DI water source, drop volume, read time, sites (centre, mid radius, edge) and the time from each process step to measurement.

2

Baseline clean wafers

Measure monitor wafers right after your standard clean, and note how fast the angle rises while they wait.

3

Baseline primed wafers

Measure after your standard dehydration bake and vapor prime, then coat, expose, develop and etch the same wafers.

4

Set gates

Set bands after cleaning and after priming from wafers with good pattern transfer, and a maximum wait before coating. Use the trend to optimize priming time.

5

Check each lot

Measure the monitor wafer after cleaning and after priming. Our article on self assembled monolayers explains why a single molecular layer changes the angle.

6

Coat, reclean or reprime

Coat inside the gates and time limit; reclean or reprime outside them. Our guide to the best contact angle goniometer for industrial QA covers production use.

We completed our gage R&R study on the unit and it performed very well.

Brandon Barbee

Corporate Quality Engineer, Zeus Industries, Polymer Manufacturing

Download the wafer cleaning and HMDS priming SOP

A two page SOP you can adapt for your wafers, with the method lock, site map, gates after cleaning and priming, a troubleshooting table and a record sheet.

Contact angle measurement with the Dropometer

Example output

Example monitor wafer log through cleaning and priming

Example data only. Your values and gates will differ and must come from your own process.

How to read this log

Example gates: at or below 20° after cleaning, 65° to 80° after priming, and coating within 4 hours. The wafer that waited overnight picked up contamination and was recleaned. The low primed wafer pointed to a depleted priming source. The spin primed wafer read above the band and its resist was hard to develop.

Example monitor wafer log

Wafer	Step	Centre (°)	Mid radius (°)	Edge (°)	Result
W01	After RCA clean	15	16	17	Pass
W01	After vapor prime	71	70	68	Coat
W02	After clean, 16 h wait	27	29	31	Reclean
W03	After vapor prime	58	57	52	Reprime, check supply
W04	After spin on HMDS	86	85	84	Strip and reclean

Example data, not customer data. 150 mm silicon wafers with native oxide; 2 µL DI water drops; three sites per wafer; reading at 5 s.

Signal order

The order to check signals when resist lifts

Check these in order. Each one rules out a cause before you look at the next.

1

Angle after cleaning

Rules out organic residue on the substrate surface.

2

Wait time

Rules out recontamination between steps.

3

Angle after priming

Shows a weak or excessive prime.

4

Centre to edge spread

Points to tool uniformity.

5

Resist process

Only then look at bake, exposure and develop; our page on surface cleanliness verification covers other surfaces.

FAQ

Common questions about wafer cleaning and photoresist adhesion

Questions lithography and process engineers ask about surface preparation.

University nanofab pages give about 65° to 80° (UT Dallas) and 65° to 75° (EPFL). MicroChemicals suggests a different test: water on the bare substrate and on the resist film should be equal and between 45° and 60°. Some photoresist materials, such as SU-8 epoxy resists, are often used with other adhesion treatments. Follow your resist supplier, then set your own band from lithography results.

A clean silicon dioxide surface carries hydroxyl groups and wets with water. Organic residue or airborne hydrocarbons raise the angle. Published values depend on the clean: below 5° after an oxygen containing plasma and about 16° after an RCA SC-1 clean (Williams and Hicks, 2011), and 20.7° in another study (Bryk and colleagues, Materials, 2020).

It depends on your clean room air and storage. Studies show the angle on clean oxide rises over hours to days as hydrocarbons adsorb, at a rate that depends on the cleaning method. Measure your own curve, then set a maximum wait between cleaning, priming and resist coating.

Vapor. MicroChemicals advises never to apply HMDS in a spin coater, because the thick layer it leaves releases ammonia that can crosslink the resist and make development difficult. Vapor priming on a heated, dehydrated wafer forms a thin, even layer for better adhesion, and a contact angle check can confirm it.

No. Contact angle shows thin organic films and priming state across the area under the drop. The IRDS yield enhancement roadmap calls particles more critical than other contaminants for yield, so keep your particle and metal inspections. Use contact angle alongside other measurement techniques as the fast check for the surface chemistry that decides resist adhesion.

SEMI/ASTM D7490-13 covers surface free energy of solid substrates from two liquid contact angles, and the Dropometer is fully compliant with it. ASTM F22 is a water break test for hydrophobic films; contact angle complements it with a number instead of a visual call.

Yes, but each substrate needs its own baseline and gates, because clean and primed angles differ by material, from a bare silicon substrate to nitride or glass. Some films, such as noble metals, may need a different adhesion promoter than HMDS. Measure each new substrate before setting gates.

A contact angle instrument that is quick to set up near the clean room, records drop images, and calculates surface free energy with two liquids. The Dropometer weighs under 2.5 kg and sets up in about 2 minutes. Our pages on plasma treatment for adhesion and PCB surface preparation cover related checks.

Business impact

What changes when you verify wafer surface preparation

Typical changes in how teams work. Benchmarks are given only where a published source exists.

Before and with wafer surface checks

Metric	Before Dropometer	With Dropometer	Indicative Benchmark
Cleaning verification	Water break by eye	Water contact angle at set sites	Below 5° after plasma, 16.2° after SC-1 (Williams and Hicks, 2011)
Priming check	Assumed from recipe	Primed angle on a monitor wafer per lot	65° to 80° cited for best microlithography (UT Dallas)
Finding adhesion failures	After develop or etch	Before resist coating	No published benchmark; track on your own lots
Tool drift	Found from yield loss	Trend of primed angles by tool	Below 65° signals a bubbler refill (EPFL)

See the check on your own wafers

Bring a cleaned and a primed monitor wafer to a call. We will show the measurement on screen.

Book a 15 minute demo

Estimate the cost of resist rework

Enter your own numbers. The result is an estimate, not a Droplet Lab claim.

Instrument cost ($) Each Dropometer unit is US$5,000; the default models one unit.

Resist rework events per month

Average rework cost per event ($) Your own average cost to strip, reclean, reprime and recoat one wafer or lot.

Failures preventable (%) Use your own failure analysis; start low if unsure.

Scrap cost per event ($) Scrap cost per event from adhesion related failures only, from your own records.

Other monthly savings ($)

Result

~0

Monthly savings

~0

Payback period

~0

First year net benefit

Monthly savings = preventable rework cost + preventable scrap cost + other monthly savings.

Limits

What contact angle cannot tell you about wafers

Honest limits, so you know when to reach for another test.

It does not detect particles or metal contamination.

It samples sites, so use monitor wafers or set test areas.

The drop itself contacts the surface; measure monitor wafers, not product die.

Angles change with time after each step; fix the wait time.

Priming targets differ by resist and supplier; on resist films, measure away from edge beads.

It is a bench instrument, not an inline tool inside a track, and it does not replace defect inspection.

Water contact angle shows organic contamination and priming state at the sites measured. Particles, metals, film thickness, pattern defects and electrical results need their own tests.

Book a 15 minute demo Request a quote

How this page was created

Editorial and technical transparency notes for this page.

Transparency Details 4 checklist items

01

Drafting assistance

Drafted with Claude Opus 5.5 (Anthropic) using web search for sources, then edited by the Droplet Lab team.

02

Technical review

Reviewed and edited for technical accuracy by the Droplet Lab Team.

03

Verification steps

Standard identifiers, units, thresholds and key procedural claims are checked against cited sources before publication.

04

Updates

Reviewed every 12 months or when the underlying method changes.

Report a correction

Spotted an issue in this summary? Send a correction request and our team will review it.

Correction Request

We work hard to keep this page accurate and up to date. If you spot an error (wrong revision/year, missing requirement, incorrect interpretation, or broken link), tell us and we'll review it.

Contact us to report a correction

Sources

References

1.

UT Dallas Research Cores. HMDS process. https://cores.research.utdallas.edu/?p=21312

2.

EPFL Center of MicroNanoTechnology (CMi). HMDS process page. https://www.epfl.ch/research/facilities/cmi/?p=4602

3.

Williams, T.S., Hicks, R.F. Conference abstract, session SE+PS-WeA2, AVS 58th International Symposium, 2011. https://www2.avs.org/symposium2011/Papers/Paper_SE+PS-WeA2.html

4.

Imafuku, D. et al. Organic contamination of silicon wafer in clean room air and its impact to gate oxide integrity. MRS Proceedings 473:161, 1997. https://doi.org/10.1557/PROC-473-161

5.

Bryk, P. et al. What is the value of water contact angle on silicon? Materials 13(7):1554, 2020. https://doi.org/10.3390/ma13071554

6.

MicroChemicals GmbH. Substrate pretreatment: cleaning and adhesion promotion. Application note. https://www.microchemicals.com/dokumente/application_notes/substrate_cleaning_adhesion_photoresist.pdf

7.

MicroChemicals GmbH. HMDS, adhesion promotion. Product information. https://www.microchemicals.com/products/adhesion_promotion/hmds.html

8.

IEEE International Roadmap for Devices and Systems, 2024 Edition. Yield Enhancement. https://irds.ieee.org/images/files/pdf/2024/2024IRDS_YE.pdf

9.

ASTM D7490-13(2022). Standard test method for measurement of the surface tension of solid coatings, substrates and pigments using contact angle measurements. https://dropletlab.com/industry-standards/semi-astm-d7490/

10.

Ho, K., Shiba, T., Chen, C.Y., Kim, D.M. Plasma treatment to remove titanium surface contaminants and improve implant biocompatibility: an in vitro study. Biomimetics, 2025. See our analysis of this paper. https://dropletlab.com/validation/citations/analysis/plasma-treatment-to-remove-titanium-surface-contaminants-and-improve-implant-biocompatibility-an-in-vitro-study/

11.

Chen, X. et al. Contact angle measurement with a smartphone. Review of Scientific Instruments 89(3):035117, 2018. https://pubs.aip.org/aip/rsi/article-abstract/89/3/035117/368179/Contact-angle-measurement-with-a-smartphone

12.

Surface tension measurement with a smartphone using a pendant drop. Colloids and Surfaces A: Physicochemical and Engineering Aspects. https://www.sciencedirect.com/science/article/abs/pii/S0927775717307744

13.

Droplet Lab validation papers

14.

Customer publications that cite the Dropometer

15.

US medical device manufacturer case study

16.

Semiconductors guide

Photoresist Adhesion on Silicon Wafer Substrates: Verify Wafer Cleaning and HMDS Priming With Contact Angle Measurement Before Photolithography and Etch

Abhimanyu Bhandankar

Droplet Lab Team

Featured Studies & Publications

Why wafer surface state decides photoresist adhesion

65° to 80°

Below 5° vs 16.2°

Half a monolayer

What this wafer surface check does and what it does not

Evidence Box

What wafer surface question are you trying to answer?

Lithography process engineer

Yield or quality engineer

University nanofab manager

Is this the right check for your wafer process?

Good fit if

Less relevant if

How to verify wafer cleaning and HMDS priming

Expert perspective on wafer surfaces

Why photoresist adhesion fails on wafers

Why resist adhesion fails and what to check

Not sure whether surface preparation is the cause?

What a wafer surface preparation record contains

Lot identity

Timing

Method settings

Readings

Decision

Drop images

What to measure for surface preparation and process control

Water contact angle after surface cleaning

Water contact angle after HMDS priming

Centre to edge spread

Angle against waiting time

Validated measurement approach for wafer surface checks

How to check wafer cleaning and HMDS priming

Lock the method

Baseline clean wafers

Baseline primed wafers

Set gates

Check each lot

Coat, reclean or reprime

Download the wafer cleaning and HMDS priming SOP

Example monitor wafer log through cleaning and priming

Example monitor wafer log

The order to check signals when resist lifts

Angle after cleaning

Wait time

Angle after priming

Centre to edge spread

Resist process

Common questions about wafer cleaning and photoresist adhesion

What changes when you verify wafer surface preparation

Before and with wafer surface checks

See the check on your own wafers

Estimate the cost of resist rework

Estimate avoided strip, reclean and rework cost from resist adhesion failures.

What contact angle cannot tell you about wafers

Standards, evidence and related cleanliness workflows

SEMI/ASTM D7490-13

ASTM F22

US medical device manufacturer

Plasma cleaning of titanium

Dropometer

Cleanliness, residue and contamination

PCB surface preparation

Catheter hydrophilic coating

Semiconductors guide

How this page was created

Drafting assistance

Technical review

Verification steps

Updates

Report a correction

References