Proposing a test on compatibility of fluorine and chlorine chemistry

Nickolai Belov nbelov at
Thu Sep 15 09:42:53 PDT 2005

Dear PQuest users,


This week there were several discussions on influence of fluorine
etching processes on subsequent chlorine etching processes of III-V


It does not look like some solid data characterizing the influence is


I would like to propose the following test(s) in order to experimentally
characterize this influence.


1. Prepare two identical (or very similar) samples of III-V materials
for etching in Cl plasma.

2. Prepare a set of samples for etching in F plasma.

3. Do oxygen plasma cleaning and pre-conditioning of the chamber for
Cl-based etching.

4. Etch the first III-V sample in Cl-plasma.

5. Do oxygen plasma cleaning and pre-conditioning of the chamber for
F-based plasma.

6. Etch the samples prepared for F plasma processing in fluorine plasma.

7. Do oxygen plasma cleaning and pre-conditioning of the chamber for
Cl-based plasma.

8. Etch the second III-V sample in Cl-plasma using the same process as
the first III-V sample.

9. Do comparative analysis of the two III-V samples etched in Cl-plasma.


There many details that have to be worked out.

There are some suggestions related to some of the details:

-         Processing history prior to the test may affect results of the
etching of the first III-V sample. Therefore, it makes sense to do the
test after III-V Cl-etching processes which deliver good (expected)

-         Similarity of III-V samples should include size of the sample,
thickness of the film / material to be etched, pattern, and size of the
open for etching area as percentage of the sample area. Samples should
be prepared at the same time using the same masking material. If samples
are going to be etched on a carrier then the same type of carrier (e.g.,
covered with photoresist or oxidized Si wafer) and same method of
mounting should be used for both of them.

-         Cl-based etching process should be long enough to allow
accurate etch depth measurements. It should not be a self-stop process
(e.g. etch through a film and stop on a stop layer).

-         Oxygen plasma cleaning and pre-conditioning processes are not
defined. Suggested oxygen cleaning process: OXYGEN3 for 10 minutes.
Pre-conditioning might use the same process as will be used for etching.

-         Results may vary depending on parameters of the F-based
etching process, its duration and a mask used on the sample etched in
F-based chemistry. Several tests may be suggested (short etching process
with low power, long etching process with high power, etc.).  However,
it seems to be beneficial to do characterization for some processes that
are currently being used by active users who employ F-based etching.

-         It is suggested that the comparative analysis of the III-V
samples includes etching depth and surface roughness measurements as
well as profile parameters (side wall angle, undercut) evaluation.
Measurements can be done using non-contact optical profilometer first
(Zygo) and contact profilometer after that.


Hopefully, as a result of couple of the above-described tests, some
recommendations for "standard" cleaning and chamber-conditioning
processes for PQuest users will be worked out.


I will be happy to participate in the tests as I have been using both
F-based and Cl-based processes.

If the suggested test is accepted by PQuest community and details are
worked out then I'll be happy to offer some machine time for etching two
III-V samples with Cl-based processes before and after etching some of
my substrates which will be done with F-based chemistry.


Your suggestions and responses are very welcome.


With respect


Nickolai Belov


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <>

More information about the pquest mailing list