[labnetwork] Query regarding quartz tube cleaning station

[Bob Hamilton]

The University of California Microlab made a decision to obviate the use 
of a tube wash circa 1987. We have not regretted that decision. Having 
said this we do have a provision in one of our wetsinks to remove the 
decking and do horizontal cleaning of fused silica tubes in the plenum 
area of that sink. We fabricated vee-shaped supports to hold a furnace 
tube, if needed.

Our decision to obviate cleans was based on our proximity to our quartz 
vendor, inspection and a high confidence factor in their 
post-fabrication cleaning facility, their use of polyethylene bags when 
packing and shipping quartzware and our experience with process.

Your choice of HF and HNO3 infers, amongst other things, you will be 
removing polysilicon from fused silica LPCVD tubes. In the Microlab and 
now the Marvell NanoLab, which supersedes the Microlab, we do not clean 
poly from our tubes.  We have chosen to run out tubes to failure, i.e. 
"run-to-failure" This means poly tube replacements at somewhere between 
350 - 500 u's of deposition. We do not see dep-thickness tube failures 
in LTO or SiN dep processes. I'll explain how we pull this off.

Cleaning poly tubes requires cooling them off. To prevent 
stress-cracking from cooldoan this means tubes must be pulled at ~50-100 
u's of deposition. The benefits of cleaning are supposed to be lower 
particle counts and longer tube life. Given our MEMS polysilicon needs 
this would mean pulling tubes every 25-50 runs. Cleaning a tube requires 
downtime, handling and a fair amount of acid.

In lieu of cleans we created "liners", i.e. sacrificial fused silica 
cylinders designed to take most of the deposition. We also use a 
split-liner to lay over our thermocouple. Liners crack from deposition; 
however, the cracking of liners does not result in tube (vacuum) 
failures. We have not noticed significant increases in particle counts 
when running beyond 50-100 u's in our poly tubes (note, we have 
dedicated MOS, MEMS, SiC, SiGe-MOS and SiGe-MEMS tubes).

In addition to these LPCVD tubes we run LPCVD LTO and SiN. In terms of 
cleaning we install and treat these tubes the same way, from the vendors 
polyethylene bagging directly to use. These tubes do not break from 
deposition stress; therefore, we do not use liners. We do clean the rear 
transitions of the nitride and LTO tubes as this area occludes from 
depositions. A wet cleanroom towel placed in a cooled SIN rear 
transition overnight loosens the NH4Cl deposit enough to clean it out.


How did our decision affect the quality of our films, particularly 
atmospheric-oxide? Significant process monitoring confirms our furnaces 
produce high quality films. Like LPCVD tubes atmospheric tubes are 
placed right into service and then cleaned in situ using a classic 
Trichloroethane bubbler and O2. We produce superb gate-ox.

In terms of particle management, cleaning quartzboats, dummy wafers and 
pump forelines has been more important for particle control than tube 
cleaning.

A final comment, if you use a free standing tube wash make sure it is 
sited within secondary containment to prevent an acid spill. When these 
things leak you have a frightful mess to clean up which may cause 
future, serious damage. An anecdote; we had such an experience with a 
set of cryopump line which ran across a floor that experienced a dilute 
HCl spill. In spite of spill cleanup and HCl being a volatile acid the 
lines corroded and began to leak about a year after exposure.

Regards,
Bob Hamilton

-- 
Robert Hamilton
University of California at Berkeley
Marvell NanoLab
Equipment Eng. Mgr.
Room 520 Sutardja Dai Hall
Berkeley, CA 94720-1754
bob at eecs.berkeley.edu
Phone: 510-809-8600
Mobile: 510-325-7557
e-mail preferred





On 3/10/2014 3:54 AM, vanaparthy at ee.iitb.ac.in wrote:
> Hello,
>
> We are planning to install a new quartz tube cleaning station / bench for
> Furnace tubes in our lab at IIT Bombay. Do you have such tube cleaning
> station at your facility. If yes or if you could let us know following:
>
> Quartz tube cleaning requires HF & HNO3.
>
> 1. Does such tube cleaning require a clean room? or is semi-clean area is
> also OK?
>
> 2.Is the Quartz tube cleaning station kept in the same room / area where
> furnaces are there.
>
> 3.If furnaces are in different area, how do you transport quartz tubes
> from Quartz tube cleaning station to furnace area.
>
> The following are the process tubes at present in IITB:
>
> A. Dry oxidation
> B. Pyrogenic Oxidation
> C. Nitride deposition
> D. Poly Deposition (doped & undoped),
> E. Solid source diffusion furnaces
>
>