Si contact I-V

n that I have.

100 ohm resistor and DUT.

s both contacts.

5 V (peak to peak)

going to futz around with the contacts some more.

Hmm Well I first realized I had the I-V backwards and upside down from what is 'normal'. So here it is corrected, I along x-axis (voltage across 100 ohm resistor) and V along X axis.

Reheatng the contacts and making a new contact didn't change anything. (We ll i learned I've got to let it cool before doing the IV :^) Still searching the web for bits of wisdom.

George H.

Three guesses:

1: the contacts are giving you some sleazy combination of rectifying and ohmic.

2: the Silicon is very lightly doped, and you're somehow saturating it with carriers. This is a huge guess for me -- I have never, ever, had to know the details of conduction in Silicon.

3: or maybe it's just a measurement artifact? Go back and double-check?

--

Tim Wescott 
Wescott Design Services 
http://www.wescottdesign.com

This measures a dynamic conductance. Well maybe since we don't know what funny stuff is going on in the transformer. Is there some reason you haven't done a simple DC test?

In a chip, contact is made to a heavily doped area to insure ohmic contact with so-called metal. ["Metal" is pretty complicated these days.] Your experiment is more along the lines of material research. I would research how they do wafer profiling.

Did you do anything to insure you got past residual oxide on the silicon?

Grin.. it's certainly sleazy! It doens't look rectifying though... well at least not to my untrained eye.

Yeah.. I have no idea what the silcon wafer is. I'm guessing lightly doped n type, but it could be anything. I used it years ago as an optical filter for IR. It's been kicking around in my junk drawer ever since.

Yeah I made a thrid contact and then measured 1-2, 2-3 and 3-1.. they all looked prety much the same. I'm very confused by the non-symetric nature.. It seems like I should have two 'ohmic' contacts in series... that should be symetric.

George H.

n't

Yeah I started by floating my 'scope and doing it DC. I had a bunch more n oise than I liked so I tried the transformer. (Didn't help) and then when I reheated the indium contacts the noise went down.. so one of the contacts was a bit flaky I guess. After that I just left the transformerin there.. . I'll take it out tomorrow.

t

Nope, just soldering iron with fresh tip and indium. 700F, I let it sit th ere for maybe a minute or so. Tomorrow I want to try contacting the edge o f a freshly cleaved hunk of Si. That would give me a clean surface and I s hould also be able to get a guesstimate of the conductivity... Well if you can define any conductivity with such a weird I-V.

George H.

Driving home I realized the above is wrong, I only measured 1-2, 2-3, so maybe it's contact #2. (the one with my fingerprint under it :^)

I was also thinking the I-V looked more like a tunneling thing. At even higher voltages >10V it becomes more conductive again. Weird... George H.

Hi all, well second attempt at ohmic contacts. (I?m doing this at ~DC now with a floating ?scope.) I cleaved off a new piece of Silicon. Then made contact all along the short side. And two little ?point? contacts about in the middle of the long side.

(width = 0.33? (8.4mm) length = 1.7? (43mm) thickness = 0.025? (0.63mm) The I-V looks a bit better than before, But still something?s wrong. Here?s what I get if I use one of the sid e contacts as a voltage probe, So If you?ll forgive me for thinking out loud. (and please correct me if I say something stupid.) For the negative current region the voltage (at a given current) goes down by the expected factor of two. (I?ve got 1/2 the length of silicon all the resistance is in the Si wafer.) But for positive currents there is not much of a change and most of the voltage drop is across the contact. So it seems like I still have one bad contact.

George H.