[mems-talk] Ohmic Contact to p-type Si

[Nannan Chen]

Hi,
 
Jpt, about your post on ohmic contact to silicon, is your silicon p-type or n-type or just plain undoped sillicon? 

We are experiencing difficulties with getting ohmic contact to our P-type B ion implanted silicon cantilevers. 
Our surface doping concentration (required for good piezoresistivity) is on the order of 10^18 according to simulation. 
 
Currently we are using 5000A Au with 500A Ti as interfacial layer for electrical connection. Gold or copper on the top layer is a requirement for our wire-bonder.
 
We performed a quick local oxide removal before the evaporation.
No anealing is performed after metalization, to avoid diffusion of gold. 
On the I-V curve, we observe a turn-on voltage of approximately 0.2V. 

We could do a double-doping process to have a different concentration for contact than for the piezoresistors, but what doping level should we shoot for on the contact regions? 

Please correct me if am wrong, but I would suppose that the metal choice for n-type and p-type silicon would be very different, due to band structure, right? 

Any suggestions/references on choice of metal, annealing recipes, and doping would be greatly appreciated.

Nannan