[labnetwork] Some info on HfO2 precursors

Thu Aug 20 15:31:00 EDT 2026

I'd love to be on ALD mailing list.

Also, does anybody have 30 min or so to talk to me about specifying and ordering ALD precursors?

We're buying and installing ALD precursors for the first time right now. It would really help to get advice on how to go about this, precursor handling, approximate pricing, what to watch for when ordering, which ones decompose or don't like to sit too long, how to know we need a bubbler and how it works getting those filled, etc. We don't have a glove box, so assuming we'd need everything filled elsewhere (also no idea how to do the fill myself).

I know the chemistry in the chamber pretty well from my Oxford days, but this side of it is new to me. Our system is the new Oxford ASP. I have some guidance from them, but fab manager advice would be very helpful.

Russ

________________________________
From: labnetwork <labnetwork-bounces at mtl.mit.edu> on behalf of Jing Guo <jg78 at rice.edu>
Sent: Thursday, August 20, 2026 7:42 AM
To: Martin, Michael <michael.martin at louisville.edu>
Cc: Yu <ywu387 at gatech.edu>; Stefan Myhajlenko <myhajlenko at asu.edu>; Gary <gary.spinner at ien.gatech.edu>; Network Fab <labnetwork at mtl.mit.edu>; Morris, JD <james.morris.3 at louisville.edu>; a-dhote <a-dhote at northwestern.edu>; Chen, Zhi D. <zhi.chen at uky.edu>; dleber <dleber at vt.edu>; Bill Mitchell <mitchell at ece.ucsb.edu>; Tony Whipple <whipp003 at umn.edu>
Subject: Re: [labnetwork] Some info on HfO2 precursors

[EXTERNAL EMAIL]

Hi Michael,

We are using TDMAHf on Savannah 200 ALD, 25g in a 50ml swage lock cylinder. It’s been working very stable for a long time until recently (Our system is relatively old). Usually we keep the cylinder at 75C for the idle status. Nowadays the standard recipe pulse for Hf is much lower than it is supposed to be. But the weight of the whole cylinder shows it only used about 5 grams.

I have talked to Veeco engineers about this problem. They actually suggested not to keep the cylinder hot as idle. Because the manifold was hot, TDMAHf could decompose at the valve position then cause some deposition/accumulation. The redeposition will cause some clogs.  They definitely suggest to have a scheduled maintenance to clean the manifold. But it is a very hard job to disassemble the manifold and reinstall it back on Savannah. All these accumulated materials will take years to happen.

We still keep the TDMAHf cylinder 75C and suggest our users to use a little bit longer pulse time to keep a stable Hf pulse.

Thanks

Best
Jing
Jing Guo Ph.D.
Cleanroom Manager/Research Scientist
SEA Cleanroom (SST 017)
Rice University
Houston, TX
jeanne.guo at rice.edu
713-348-8227

On Aug 18, 2026, at 16:35, Martin, Michael <michael.martin at louisville.edu> wrote:

Hi Labnetwork:
   I wanted to summarize some advice from our informal ALD working group (formerly NNCI ALD group I think)  about HfO2 precursors so that this info will be in the labnet archive and available in the Labnetwork chatbot.

"I see there are something like 4 Hf precursors that folks have used in the literature.

  1.
TEMAHf   (Tetrakis(ethylmethylamido)hafnium(IV),(EtMeN)4Hf)
  2.
Tris(dimethylamido)cyclopentadienyl Hafnium, HfCp(NMe2)3 (HyALD)
  3.
Tetrakis(dimethylamido)hafnium(IV) [TDMAH]
  4.
And a chlorinated Hf compound.

We have some limited experience with TDMAH but Beneq suggests TEMAHf or HyALD is more stable.  What is your go-to Hf precursor? "

Tony Whiple from Minnesota pointed out that they:
"use Tetrakis(dimethylamido)hafnium(IV) [TDMAH]  in all of our ALD systems, Savannah, Lesker, Fiji. We continue to keep with the same precursors as users have set processes for that.

We have had a few issues with using this, mainly the random end of service of the precursor.
The causes of the cylinder going bad ( not able to pulse material, while the cylinder is only half empty ) can at times be assigned to the cylinder heat being turned off and going to room temperature.  After reheating to the correct temperature, there is a chance the cylinder will go bad.  We have had this happen only a few times.  To keep this from happening, we avoid having the cylinders cool off."

Evidently others (Mac from Harvard and Bangzhi from PSU) have had similar problems with TDMAH- incidentally this is a big problem with the SnO2 precursor TDMASn.  Mac suggests loading only a small amount of TDMAH at a time, and Bangzhi suggested that sometimes cycling the manual valve to the bottle may free up the delivery line.

My inclination based on this advice is to try the TEMAHf   (Tetrakis(ethylmethylamido)hafnium(IV),(EtMeN)4Hf) option given that we may not have super high throughput so a stable precursor might be important in our case and Beneq has provided a starting recipe.  Generally, my impression is that the chlorinated precursors can cause problems with the plumbing and pumps since the reaction yields HCl .

I'd love to hear any other advice people may want to share on this process and precursors.

Regards,
   Michael Martin, Ph.D.

Manager, Micro/Nano Technology Center
University of Louisville
(502) 552-1945
Web page<https://urldefense.com/v3/__https://louisville.edu/micronano__;!!PcititFMkQu_!upn3uk9uoqSskH0MJwvAN1OIH_2D53f-ECBUpfdvB3rJ1M4bapgWjy-r9OIu43QJMMbC1C4r1KaQ9XyDTQbPS1RLHlQUJg$> Instagram<https://urldefense.com/v3/__https://www.instagram.com/micronanolouisville/__;!!PcititFMkQu_!upn3uk9uoqSskH0MJwvAN1OIH_2D53f-ECBUpfdvB3rJ1M4bapgWjy-r9OIu43QJMMbC1C4r1KaQ9XyDTQbPS1SsNvvcLQ$> Facebook<https://urldefense.com/v3/__https://www.facebook.com/MNTCUOFL/__;!!PcititFMkQu_!upn3uk9uoqSskH0MJwvAN1OIH_2D53f-ECBUpfdvB3rJ1M4bapgWjy-r9OIu43QJMMbC1C4r1KaQ9XyDTQbPS1RmBn-Ouw$>
_______________________________________________
labnetwork mailing list
labnetwork at mtl.mit.edu<mailto:labnetwork at mtl.mit.edu>
https://urldefense.com/v3/__https://mtl.mit.edu/mailman/listinfo.cgi/labnetwork__;!!PcititFMkQu_!upn3uk9uoqSskH0MJwvAN1OIH_2D53f-ECBUpfdvB3rJ1M4bapgWjy-r9OIu43QJMMbC1C4r1KaQ9XyDTQbPS1TtYIQwPw$

This email originated outside of the University of Nevada, Reno. Do not click on links or attachments unless you recognize the sender and know the content is safe. Report suspicious emails to the Office of Information Technology (OIT) at abuse at unr.edu.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mtl.mit.edu/pipermail/labnetwork/attachments/20260820/165146b9/attachment.html>