Aluminum Nitride

Joey Doll jcdoll at
Sun Sep 21 16:02:48 PDT 2008

Disregard the part about using aluminum to mask the TMAH; it's likely that
the TMAH formulations that don't attack aluminum wouldn't attack the
aluminum nitride. I'll need to put some more thought into that.
- joey

On Sat, Sep 20, 2008 at 9:06 PM, Joey Doll <jcdoll at> wrote:

> Thanks for the feedback re: the phosphoric acid etch, that would have been
> a major issue. I just found a paper reporting that 25% TMAH at room
> temperature etches aluminum nitride, so I will use that instead of the hot
> phosphoric acid. Then I could use aluminum to mask the TMAH etch, unless I'm
> lucky and room temperature TMAH doesn't attack photoresist like it does at
> higher temperatures. Using oxide as a phosphoric acid etch mask would have
> been tricky, because removing it afterwards with HF would have also attacked
> my titanium underlayer. Anyways though, problem solved I hope.
> I wasn't too concerned about etching metal-containing wafers on stsetch,
> because that part of my process is identical to the piezoresistive
> cantilever process used in the ME342/E341 class process and research
> projects back to Beth Pruitt's graduate work, at least. I will be careful
> about that step though. Mainly I wanted to ensure that my use of aluminum
> nitride in the lab sounds good.
> Thanks again,
> Joey
> On Fri, Sep 19, 2008 at 3:33 PM, Mary Tang <mtang at> wrote:
>> Hi Joey --
>> Thanks for the detailed process flow, as really helps...  If I read
>> correctly, your request is for etching metal-containing wafers on the
>> stsetch, with the consideration that no metal is expected to be exposed to
>> plasma.  I think we'll need a SpecMat quorum to decide that, so can't answer
>> right now.  However, I do think that you will have a problem with resist
>> surviving the hot phosphoric acid.  I would suggest that either using
>> p5000etch metal chamber to plasma etch AlN (another SpecMat request) or
>> depositing an oxide on top of the AlN, patterning that with a plasma or wet
>> oxide etch, removing the resist, then using the hard mask for the AlN etch.
>> Mary
>> Joey Doll wrote:
>>> Hi again -
>>> I have been working with a vendor in San Jose to have aluminum nitride
>>> deposited, and I would like to submit a new specmat request.
>>> The vendor is
>>> Hionix
>>> 2363 Bering Drive, San Jose, CA 95131, USA
>>> +1(408) 526-2333
>>> <>
>>> The aluminum nitride sputtering is performed in a system that is only
>>> used to aluminum and titanium sputtering, normally for backend CMOS
>>> interconnects. Sputtering is performed with an aluminum target with nitrogen
>>> ambient at ~200C. This is in contrast with the system at Berkeley which is
>>> also used for platinum.
>>> My process flow would be as follows:
>>> * start with a new DP SOI
>>> * clean wafer @ SNF
>>> * have titanium underlayer and aluminum nitride sputtered at Hionix (with
>>> an additional RCA clean there)
>>> * lithography on top of the aluminum nitride (on ASML)
>>> * etch the aluminum nitride with phosphoric acid @ 120C on wbgeneral
>>> * strip the photoresist
>>> * lithography on top of the titanium
>>> * etch the titanium using standard ti wet etch on wbmetal
>>> * strip the photoresist
>>> * blanket sputter aluminum on gryphon
>>> * lithography on top of the aluminum
>>> * etch with standard al etch on wbmetal (which doesn't attack the
>>> aluminum nitride or titanium)
>>> * strip the photoresist
>>> * frontside lithography with spr220
>>> * frontside STS etch on wafer to release front of cantilevers
>>> * strip the photoresist
>>> * backside lithography using spr220
>>> * backside STS etch to release wafers
>>> * oxide release on AMTEtcher
>>> * strip the photoresist
>>> The aluminum nitride would only be brought into the lab already deposited
>>> on wafers, and the phosphoric acid would be drained down the AWN system once
>>> returning to room temperature. Also, the aluminum nitride, aluminum and
>>> titanium will be protected by photoresist while in STS and AMTEtcher. I have
>>> been characterizing the films with XRD for the past month or so and am ready
>>> to start fabricating working devices as soon as the material is approved.
>>> Thanks for your consideration and please let me know if you have any
>>> questions.
>>> - Joey
>>> On Tue, Jan 22, 2008 at 8:52 AM, Ed Myers <edmyers at <mailto:
>>> edmyers at>> wrote:
>>>    Joey,
>>>    One of our weak areas is RF sputtering.  This is need for reactive
>>>    deposition of insulating materials.  As a result, we can not
>>>    support your request due to hardware limitations.  If you have any
>>>    questions please let me know.
>>>    Regards,
>>>    Ed
>>>    At 10:47 AM 1/19/2008, Joey Doll wrote:
>>>        Hi -
>>>        I'm considering a process which would use aluminum nitride as a
>>>        piezoelectric actuator layer. I wasn't able to find anything
>>>        mentioned
>>>        about it on the SNF website and was wondering if there is a
>>>        recipe /
>>>        it would be possible to sputter it on grypon or metallica? It is
>>>        sputtered up at Berkeley
>>>        ( and I
>>> was
>>>        hoping to use it instead of other materials like ZnO or PZT in
>>>        order
>>>        to keep my wafer semi-clean. Please let me know what you think.
>>>        Thanks!
>>>        Joey
>> --
>> Mary X. Tang, Ph.D.
>> Stanford Nanofabrication Facility
>> CIS Room 136, Mail Code 4070
>> Stanford, CA  94305
>> (650)723-9980
>> mtang at
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <>

More information about the specmat mailing list