[phpBB Debug] PHP Warning: in file [ROOT]/phpbb/session.php on line 580: sizeof(): Parameter must be an array or an object that implements Countable
[phpBB Debug] PHP Warning: in file [ROOT]/phpbb/session.php on line 636: sizeof(): Parameter must be an array or an object that implements Countable
3.14.by forum • Problem: easier metallization
Page 1 of 1

Problem: easier metallization

Posted: Sat Apr 30, 2011 5:43 pm
by BarsMonster
One of the most difficult problem is IC metallization.
I see the following ways:

1. vacuum evaporation of Al, then etching it away or using lift-off lithography
2. vacuum evaporation of Cu, then etching it away or using lift-off lithography
3. Using photo-sensitive Ag-compounds - but they does not form films, just a bunch of separate Ag crystals.
4. Wet Ni metalization
5. Your idea?

Re: Problem: easier metallization

Posted: Tue May 03, 2011 11:37 am
by #E0
You should definitely chose Alu option. Copper metallization is a bad choice, as it has high diffusion rate through silicon and oxide. This can result in p-n junction poisoning and shorts from electrochemical migration. There are other contras also.

Re: Problem: easier metallization

Posted: Wed May 04, 2011 2:34 am
by BarsMonster
#E0 wrote:You should definitely chose Alu option. Copper metallization is a bad choice, as it has high diffusion rate through silicon and oxide. This can result in p-n junction poisoning and shorts from electrochemical migration. There are other contras also.
Yes, Al seems to be the most studied and reliable.

After looking into this a little more, if I can sacrifice speed and shoot for easy manufacturing & reliability, here is what we have:

Ag: require interface to SiO2 to be stable, wet metalization available
Au: require interface to Si to be stable
Cu: require interface to Si to be stable
Na: too reactive, real MOSFET poison
K: too reactive
Ca: too reactive
Li: too reactive

--------------

Al: industry proven metal, wet-metalization under question
Sb: bad resistance but very low diffusion in Si & SiO2 ( http://www.icknowledge.com/misc_technol ... fusion.pdf )
Ti: common spacer material, silicide forms ohmic contact
W: common spacer material, silicide forms ohmic contact
Zn: Terrible diffusion in Si
Fe: Terrible diffusion in Si
Pb: Not much info
Sn: bad resistance but very low diffusion in Si ( http://www.icknowledge.com/misc_technol ... fusion.pdf )
Pt: silicide form ohmic contact, should have low electro-migration, wet-metalization available, a bit expensive but acceptable
Ni: wet-metalization available

So, out of these, Nickel looks especially nice as no vacuum chamber needed to deposit it. Purity is under question (process I've seen shows phosphorus & Sodium contamination ).
If I would be able to deposit Al via wire explosion under atmospheric pressure in air - that of course would be unbeatable.

Re: Problem: easier metallization

Posted: Sun May 08, 2011 10:51 pm
by BarsMonster
Doh, they don't stop killing me:
Common chemical elements that produce deep-level defects in silicon include iron, nickel, copper, gold, and silver. In general, transition metals produce this effect, while light metals such as aluminium do not.
So, it seems that all my hopes are gone. Aluminium.