What eats N channel silicon?

[Martin Brown]

Huh? It makes a cloud of ammonium chloride particles, aka Sal Ammoniac.

No it doesn't. That would be ammonia and hydrochloric acid.

Mixing ammonia and hypochlorite bleach and/or ammonia and free chlorine
(which is the first reaction of many possible) is very nasty. It is the
reason why they say do not mix powerful cleaning products.

NCl3 is an unstable toxic liquid high explosive like NI3 only it is
inclined to detonate when hit by light, rather than by a feather. BBC
had a short article on the main possible reactions a while back. It is a
seriously bad idea to do it outside of a fume cupboard.

http://www.bbc.co.uk/dna/h2g2/classic/A795611

One of the reaction products is carcinogenic. The others are choking
toxic gasses and one is a very tetchy high explosive.

Since when is NH4CL toxic?

A little knowledge is dangerous, but no knowledge at all masquerading as
an authoritative statement of fact could be deadly. The reaction is
fast, violent and seriously dangerous in a confined space like a toilet.

You are headed for a Darwin award.

Regards,
Martin Brown

 

[[email protected]]

Actually I have a low budget customer who may soon be my employer if
he gets the grant.  I want to thin a small spot on the backside of the
chip (and its a big easy to handle  sensor die) before blasting a
tiny via in it with a Q switched yag  or Excimer laser. This is a
patterned sensor more then a IC. We're using silicon  more or less for
the fact that the patterning can write the  5 uM electrode structure
we need.

 I have 7.5 years as a university lab tech with some stuff far worse
then HF, so safety concerns are a given.  I worked with  use Osmium
Tetroxide and some other nasties.

BYW, if anybody knows a fab that does sample quantities, that would be
interesting too. I know about MOSiS.

I'll probably clean it with H202/Piranha first.

We have been advised to avoid mechanical drilling or scribing due to
fragility of the silicon.

So thanks for the replies so far.

Steve

I've used lasers to blast passivation. Of course, hell if I know more
than just how to turn the dials. But if you are going to blast a hole
with a laser, maybe you have one that can thin out the passivation
first.

I can vouch for Acurel should you decide to go that route. Very
professional group.

 

[James Arthur]

Chlorine gas was used in the Great War, by both the Germans and
British, until even more effective agents were developed.

https://www.rkb.us/contentdetail.cfm?content_id=55256

Ironically, mustard gas, first used in warfare, led to a life-saving
chemotherapy drug used widely in cancer and autoimmune diseases:

http://en.wikipedia.org/wiki/Cyclophosphamide

AIUI someone noticed that a soldier with Wegener's Granulomatosis--
a nasty, then-fatal autoimmune vasculitis--miraculously improved
after getting gassed...

Cheers,
James Arthur

 

[Okkim Atnarivik]

[email protected] wrote:
: he gets the grant. I want to thin a small spot on the backside of the
: chip (and its a big easy to handle sensor die) before blasting a
: tiny via in it with a Q switched yag or Excimer laser. This is a
:
: I'll probably clean it with H202/Piranha first.

Piranha sounds good. In your application the trilogy etch may be too
slow. You would actually be best off with the deep RIE. The pyramide-shaped
opening created by anisotropic KOH etch (I assume your wafer is <100>
-oriented) would also be nice in this application - you could even make the
hole depth self-limiting - but creation of an etch mask which can stand KOH
would be a bit problematic in this case. Spinning the resist on a diced chip
is cumbersome, and I understood that you don't have access to a decent
microfabrication facility for hardmask deposition.

You knew of Mosis, and someone mentioned Europractice. I'm not sure
whether you share with Jim Thompson the dislike of us leftist weenies,
especially of the French, but if you can suppress that momentarily,
one more place to look is http://cmp.imag.fr . They often seem to provide
interesting processes not available through Mosis or Europractice. Even
them are unlikely to handle pre-diced customer chips, but you may want
to double-check whether there are any leads in their web pages.

Regards,
Mikko

 

[[email protected]]

Thanks guys, it turns out that we have 2 MEMs fabs within 2 hours
drive, so unless there is a disaster in the funding process, I think
we can have it done. Who would have thunk it, Ohio, the technological
black hole, has fabs.

One is Midwest Microdevices, and it looks like what we need done is
child's play for them.

For plan B,
I will run some tests with what I have learned here, as we have a
batch that a university fab in California goofed up. They didn't put
down the SiO2 under the gold, and it diffused and ruined batch one.

The fun part will be centering a hole between 2 traces 10u Apart. In a
previous job, I did a project to center a laser on a STM tip, so I
think I can hit the spot.

BTW, I used to be a leftist weenie, then I experienced the old
joke,
"A Conservative is a liberal that got shot at a few times"

So I can get along with both sides.

A special thanks to Mr/Dr Atnarivik...

Steve

 

[Okkim Atnarivik]

[email protected] kirjoitti:

Thanks guys, it turns out that we have 2 MEMs fabs within 2 hours
drive, so unless there is a disaster in the funding process, I think
we can have it done. Who would have thunk it, Ohio, the technological
black hole, has fabs.

A pleasure. Was not aware about the black hole part, although I once
worked half a year at Mead labs in Chillicothe...

Regards,
Mikko

 

[Rich Grise]

I only remember two news stories. One from a factory in China where that
happened. Large evacuation, but IIRC 5-10 people didn't make it. Another
accident was in Chatsworth, CA. I believe the one in CA resulted in a
fire.

You snipped the line I was responding to: "Ammonia and chlorine bleach".

Clorine bleach is only sodium hypochlorite.

Thanks anyway,
Rich

 

[Rich Grise]

I heard a muffled boom south east of here.

Rich? *RICH??!*

It's OK - I only use chlorine bleach in the laundry, and seldom if ever
use ammonia; for windows I use Windex, and for cat stench I use white
vinegar, but the two are never in proximity to each other.

Cheers!
Rich

 

[Spehro Pefhany]

You snipped the line I was responding to: "Ammonia and chlorine bleach".

Clorine bleach is only sodium hypochlorite.

Thanks anyway,
Rich

http://www.bbc.co.uk/dna/h2g2/A795611


But then you're used to inhaling toxic chemicals

 

[Martin Brown]

You snipped the line I was responding to: "Ammonia and chlorine bleach".

Clorine bleach is only sodium hypochlorite.

And depending on the ratio you mix it you will make chlorine in large
quantities, NCl3 followed by a detonation or hydrazine. That ignores a
couple of other possible side reactions with other very unpleasant end
products. It is something you do *not* do if you value your life.

They obviously had ignorant people like you in mind when they set up the
About.com:First Aid site.

http://firstaid.about.com/od/hazardousmaterials/ht/bleachgas.htm

Too many would be rescuers have been overcome by fumes and died.

If you do not believe me check the MSDS for Ammonia, it is always listed
as violently incompatible with hypochlorite and for good reason.

Regards,
Martin Brown

 

[[email protected]]

[email protected] wrote:

: he gets the grant.  I want to thin a small spot on the backside of the
: chip (and its a big easy to handle  sensor die) before blasting a
: tiny via in it with a Q switched yag  or Excimer laser. This is a
:
: I'll probably clean it with H202/Piranha first.

  Piranha sounds good. In your application the trilogy etch may be too
slow. You would actually be best off with the deep RIE. The pyramide-shaped
opening  created by anisotropic KOH etch (I assume your wafer is <100>
-oriented) would also be nice in this application - you could even make the
hole depth  self-limiting - but creation of an etch mask which can stand KOH
would be a bit problematic in this case. Spinning the resist on a diced chip
is cumbersome, and I understood that you don't have access to a decent
microfabrication facility for hardmask deposition.

  You knew of Mosis, and someone mentioned Europractice. I'm not sure
whether you share with Jim Thompson the dislike of us leftist weenies,
especially of the French, but if you can suppress that momentarily,
one more place to look ishttp://cmp.imag.fr. They often seem to provide
interesting processes not available through Mosis or Europractice. Even
them are unlikely to handle pre-diced customer chips, but you may want
to double-check whether there are any leads in their web pages.

  Regards,
           Mikko

Just ignore the wingnuts that make remarks about the left. The
wingnuts were dropped on their heads as babies, and just can't help
themselves. It is like Tourette's syndrome.

An interesting but as far as I can tell low volume chip producer was
ES2/US2. The did direct write ebeam. When I was at Exar, their gear
was in our fab.

 

[Rich Grise]

One reaction product is chlorine gas. High school chemistry or lower.

I thought I'd covered this - I misinterpreted the post. It's NH3 and
HCl that make NH4Cl. And bleach obviously releases chlorine gas
spontaneously (ever done the laundry or swum in a pool?) - the ammonia
apparently stimulates it to release even more.

What would the chemical equation be?

Thanks,
Rich

 

[Rich Grise]

Has BF3 really disappeared from silicon processing? Or was it circuit
board procesing? 20 years can really degrade memory quality.

I once read something about wafer fab where they mentioned silane, SiH4,
which is appparently pretty nasty stuff:
http://en.wikipedia.org/wiki/Chemical_compound

Cheers!
Rich

 

[[email protected]]

I once read something about wafer fab where they mentioned silane, SiH4,
which is appparently pretty nasty stuff:http://en.wikipedia.org/wiki/Chemical_compound

Yep. Nasty stuff. At the PPoE there were silane bunkers with hinged
walls and "pop tops" where they stored the stuff. Employees were
instructed to leave, immediately, if they ever smelled garlic (it was
banned from the cafeteria). Were there some bigs running around scared
when some idiot called in a bomb scare (turned out to be a
contractor).

 

[Bob Larter]

I once read something about wafer fab where they mentioned silane, SiH4,
which is appparently pretty nasty stuff:
http://en.wikipedia.org/wiki/Chemical_compound

Hm? I think you mean: <http://en.wikipedia.org/wiki/Silane>

 

[Benj]

HF is indeed used in semiconductor processing, most likely as a chemical
polish in combination with other rather agressive acids and H2O2. At least
that was the pece of kit I was refurbing. There was an emergeny exit and a
weight operated 'waterfall' in the event of getting any on you.
Fortunately the tanks were empty.

I can't imagine using it on a processed die though.

Graham

Graham please refrain from giving out your bogus gems. You obviously
have ZERO experience in semiconductor manufacturing. The etch for
silicon is the acid mix mentioned by someone else above. HF (usually
buffered HF) is used to etch silicon oxide which is (duh) glass. We
don't need to go into processes here but usually silicon wafers are
oxidized (steam) to form a layer of glass on them which is then photo-
resisted and etched down to the silicon. That then acts as a mask for
diffusion of doping impurities. Etc. Etc.

In the given case one is dealing with the back of the wafer so the
first question has to be if the back is oxide coated. If so you need
to remove it with HF etch. Although the typical process is to thin a
wafer by lapping it thin. However lapping damages the silicon crystal
to a certain depth so you need to finish the job by etching past the
damage layer. It takes a lot of care to do these things and have them
work. So expect to practice some.

And as for HF, go look it up on line. A simple neutralization or a
water wash will NOT do. The nasty stuff has a tendency to keep on
eating even though you have attempted to wash it off or neutralize it.
HF burns are VERY nasty for that reason. (Yes, I've had my share.) And
did I mention that the acid acts like an anesthetic so you tend to not
feel the burning on your skin until it's too late. But if it's just a
tiny spot or two eventually (Did I mention the wounds heal very slowly
too?) your nasty spots heal like most things in life do. Be very
careful with it. Don't be the guy who has to piss on the electric
fence to find out if all the stories are true.

I strongly recommend getting the various chemicals and acids needed
for these operations from suppliers that make commercial semiconductor
etchants rather than trying a bunch of the nonsense stuff people are
spouting at you on the internet.