[labnetwork] Bubble formation when e-beam evaporating Gold on PMMA

Thu Jul 31 17:36:18 EDT 2025

You evaporator folks are kind in trying to work out the source of these problems and coming up with solutions, but from a lithography perspective the problem is the resist performance. You might consider tossing the problem back to your litho people to find a better performing resist or at least inform users of this known issue and advise them on the alternatives.  I know CSAR and ZEP are much more expensive, but as far as I know they don't have this problem.  Have the litho people experimented with different molecular weights of PMMA or the MAA copolymer? (I admit I haven't kept up and read all the literature provided).

Other thoughts: 1) those molecules outgassing from the PMMA are probably contributing to the carbon contamination of your gold source.  2) going to another resist should allow  you to use your SEM for imaging the resist patterns.

It's clear from this research that PMMA is the source of this problem.  You can replace it rather than experimenting with evaporator specific solutions.

-Cliff

Cliff Knollenberg

Processing Engineer - Lithography

Stanford Nanofabrication Facility

650-721-1274

[Image]

________________________________
From: labnetwork <labnetwork-bounces at mtl.mit.edu> on behalf of David Barth <dsbarth at seas.upenn.edu>
Sent: Thursday, July 24, 2025 2:09 PM
To: Justin C Wirth <jcwirth at purdue.edu>
Cc: labnetwork at mtl.mit.edu <labnetwork at mtl.mit.edu>
Subject: Re: [labnetwork] Bubble formation when e-beam evaporating Gold on PMMA

Thanks, Justin. I think that's a pretty complete and accurate summary. Here<https://wiki.nano.upenn.edu/wiki/images/8/8b/Dsbarth_pmma_bubbling_maebl2023.pdf> is a link to the talk Justin mentioned in case anyone is interested.

Best,
David Barth
Senior Manager
QNF, Singh Center for Nanotechnology
University of Pennsylvania

On Thu, Jul 24, 2025 at 11:28 AM Justin C Wirth <jcwirth at purdue.edu<mailto:jcwirth at purdue.edu>> wrote:

Philipp,

I want to second Jorg’s suggestion and links here, those papers are absolutely worth a read and do a great job explaining the mechanisms involved here.

A couple things stick out to me about this type of behavior, which has been seen my many, many people with PMMA under E-beam evaporation of gold (although it depends on the particular evaporator, the deposition rate, and many factors):

  1.  It’s always in an E-beam evaporator, never a thermal evaporator, and it’s *often* gold (although it can happen with other high Z metals as well)
  2.  Heat is often offered as the causal mechanism, but actually PMMA is soft baked at 180 C to evaporate the anisole, and can be left there for very long times (minutes to hours, maybe indefinitely?) without it changing the exposure characteristics (At least if anyone has measured a change in exposure characteristics at 180 C vs. time it’s not something I’ve ever seen). So heating PMMA up to 180 C on reasonable time scales is fine and shouldn’t cause the film itself to bubble (this could be proved if someone had a vacuum annealing chamber and heated a sample with PMMA and a good metal film on top absent any deposition…though I haven’t actually done that).

So, what this means to me logically is that actually heating is not the fundamental culprit. And as explained in the papers that Jorg provided, actually the fundamental cause is crosslinking of the PMMA due to charged particle irradiation. Crosslinking causes gas molecules to be generated, which if there is nothing on top of the PMMA, are simply pumped out of the chamber (which is why one would not see any issues if one doesn’t actually deposit the gold film). But if there *is* a layer of something on top of the PMMA that the gas can’t diffuse through (i.e. gold), the gas is trapped and forms bubbles between the PMMA and the gold.

The solution is to stop the crosslinking from happening, and this can be done in a myriad of ways as explained in those papers. Because a number of secondary factors are involved, changing those secondary factors can reduce or stop the effect as well, and so I assume that all of the suggestions you’ve gotten from others on how to stop or reduce the effect have indeed worked for them and may well work in your case too. But the crosslinking plus gas generation plus gas trapping is kind of the key to understanding the whole puzzle (from the sample’s perspective at least).

It’s also an effect I’ve seen myself in almost real time in an EBL with a chamber scope (JEOL JBX-8100FS) when looking at an alignment mark on a sample with PMMA and a (probably too thick) gold layer on top for anti-charging. After enough SEM dose (seconds to tens of seconds, enough to turn the PMMA negative and crosslink), a bubble starts to balloon up where the mark is being imaged because of the gas generated. It can either pop dramatically or grow or shrink down if microcracks form and let the gas out, and it looks exactly like the kind of bubbles you see in your sample at varying stages: growing, deflated, popped, etc.

David Barth at Penn (cc’d) gave a great talk about the practical aspects of this at MAEBL<https://urldefense.com/v3/__https://maebl.org/__;!!IBzWLUs!RCSqXpqaiP3UUnWSMgay109iY_ivgcdej987ixNUv33AwE6MmlTyEeBUOu1-ZOUDXxxFTDnwgjo5YoAc9GflEm8$> a few years back, hopefully he can share his talk with you and add some more details on it, he’s experimented with mitigating this effect a good amount. But this is something that’s been a repeated topic of discussion over the years and is quite confusing because it depends a lot on the particular evaporator geometry, the particulars of your deposition, how much carbon contamination there is on top of the gold, etc.

Thanks,

Justin

Justin C. Wirth, PhD

Senior Research Engineer – EBL Processes & Nanofab Process Consultation

Birck Nanotechnology Center | Room 2287A
765.494.8203 | jcwirth at purdue.edu<mailto:jcwirth at purdue.edu>

BNCWiki<https://urldefense.com/v3/__https://purdue.atlassian.net/wiki/spaces/BNCWiki__;!!IBzWLUs!RCSqXpqaiP3UUnWSMgay109iY_ivgcdej987ixNUv33AwE6MmlTyEeBUOu1-ZOUDXxxFTDnwgjo5YoAc-hdVzOg$> | Purdue iLab<https://urldefense.com/v3/__https://purdue.ilabsolutions.com/__;!!IBzWLUs!RCSqXpqaiP3UUnWSMgay109iY_ivgcdej987ixNUv33AwE6MmlTyEeBUOu1-ZOUDXxxFTDnwgjo5YoAcAeW4cEY$> | Book time on my calendar<https://urldefense.com/v3/__https://outlook.office.com/bookwithme/user/5ad94572a9c04a4e8aa9159479128868@purdue.edu?anonymous&ep=plink__;!!IBzWLUs!RCSqXpqaiP3UUnWSMgay109iY_ivgcdej987ixNUv33AwE6MmlTyEeBUOu1-ZOUDXxxFTDnwgjo5YoAcTXdaA_c$>

From: labnetwork <labnetwork-bounces at mtl.mit.edu<mailto:labnetwork-bounces at mtl.mit.edu>> On Behalf Of Jorg Scholvin
Sent: Tuesday, July 22, 2025 7:34 PM
To: Philipp Altpeter <philipp.altpeter at lmu.de<mailto:philipp.altpeter at lmu.de>>; labnetwork at mtl.mit.edu<mailto:labnetwork at mtl.mit.edu>
Subject: Re: [labnetwork] Bubble formation when e-beam evaporating Gold on PMMA

---- External Email: Use caution with attachments, links, or sharing data ----

Hi Philipp:

Two papers of interest that may help explain:

https://pubs.aip.org/avs/jvb/article/39/5/052601/1070034/Role-of-electron-and-ion-irradiation-in-a-reliable<https://urldefense.com/v3/__https://pubs.aip.org/avs/jvb/article/39/5/052601/1070034/Role-of-electron-and-ion-irradiation-in-a-reliable__;!!IBzWLUs!RCSqXpqaiP3UUnWSMgay109iY_ivgcdej987ixNUv33AwE6MmlTyEeBUOu1-ZOUDXxxFTDnwgjo5YoAcMjLAsrU$>

https://iopscience.iop.org/article/10.1088/1361-6463/abe89b<https://urldefense.com/v3/__https://iopscience.iop.org/article/10.1088/1361-6463/abe89b__;!!IBzWLUs!RCSqXpqaiP3UUnWSMgay109iY_ivgcdej987ixNUv33AwE6MmlTyEeBUOu1-ZOUDXxxFTDnwgjo5YoAcZ2c2vFY$>

An insightful experiment, inspired by Sun’s paper, to have a plain wafer with PMMA spend time in the evaporator, with the power level just below the point where you get deposition.  Maybe 15-30 mins is sufficient. That will expose your sample to all the things you don’t want (temperature, electrons, …) without putting down a film. After that treatment, develop the PMMA (assuming there was no deposition) and you may find some interesting patterns that may suggest electron exposure. See attached image of a simple experiment we ran a few years ago (these were quick qualitative tests only).  Using a shim under the crucible helped fix things for our users (less cooling = less current = less electron radiation).

Best,

-Jorg

From: labnetwork <labnetwork-bounces at mtl.mit.edu<mailto:labnetwork-bounces at mtl.mit.edu>> On Behalf Of Philipp Altpeter
Sent: Tuesday, July 22, 2025 4:23 AM
To: labnetwork at mtl.mit.edu<mailto:labnetwork at mtl.mit.edu>
Subject: [labnetwork] Bubble formation when e-beam evaporating Gold on PMMA

Dear all,

In recent months, we installed a new UHV e-beam evaporator and have since encountered significant issues with our standard lift-off process using 3 nm Chromium followed by 50 nm Gold. Specifically, we are observing the formation of large bubbles beneath the PMMA, which severely damage the PMMA layer (see image below). The rate is decently low, 0.6 A/s at around 40mA emission current, 10 kV. After around 30 nm of thickness, bubbles become clearly visible.

Interestingly, deposition on bare silicon, silicon dioxide, or other (photo)resists appears to proceed without problems.

We have already tried adjusting various parameters — including cooling conditions, beam wobbling, throw distance, and acceleration voltage. We also modified the PMMA baking protocol and tested PMMA dissolved in different solvents — all without success.

If anyone has experienced similar issues or has suggestions for troubleshooting, your input would be greatly appreciated.

Thank you in advance for your help!

Best regards,
Philipp

[cid:ii_1983e4386764cff311]

--

Philipp Altpeter

Fakultät für Physik der LMU

LS Prof. Efetov

Geschwister-Scholl-Platz 1

D-80539 München

T. +49 (0)89 2180-3733
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20250731/a2ef8be7/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 383866 bytes
Desc: image001.png
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20250731/a2ef8be7/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Outlook-Image.png
Type: image/png
Size: 83498 bytes
Desc: Outlook-Image.png
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20250731/a2ef8be7/attachment-0001.png>