Annealing Al/Si contact increases resistance?

[Hyo X]

I'm trying to make an ohmic contact with p-type doped silicon. Using Aluminum.
Under what condition of an Al/Si contact would annealing increase the resistance of the contact? or the silicon?
two regions are defined using photolithography. A brief exposure to buffered oxide etch (6:1 H2O to HF) is followed by placing the sample in a vacuum chamber and, several hours later, with aluminum depostion.
Initially, 2-pt probe measurements from two separated aluminum regions were somewhat non-linear.
After annealing the sample, for 1 hr, the resistance increased by ~3 orders of magnitude!
I would have thought that annealing would decrease the resistance and increase the linearity of the i-v curve.
What is the problem? is there contamination or some other issue? thanks for your help or opinion.

 

[Vagn]

I'm trying to make an ohmic contact with p-type doped silicon. Using Aluminum.
Under what condition of an Al/Si contact would annealing increase the resistance of the contact? or the silicon?
two regions are defined using photolithography. A brief exposure to buffered oxide etch (6:1 H2O to HF) is followed by placing the sample in a vacuum chamber and, several hours later, with aluminum depostion.
Initially, 2-pt probe measurements from two separated aluminum regions were somewhat non-linear.
After annealing the sample, for 1 hr, the resistance increased by ~3 orders of magnitude!
I would have thought that annealing would decrease the resistance and increase the linearity of the i-v curve.
What is the problem? is there contamination or some other issue? thanks for your help or opinion.

You could have diffusion of aluminium into the silicon. Aluminium would be a donor in silicon.