I want to fabricate a p-n junction device consisting of a p-type material on n-type silicon. As top electrode, I am using gold on my p-type material which is Ohmic. But for back electrode on n-type Silicon, I am not sure which metal would be nice to achieve Ohmic contact. Aluminium forms Schottky contact on n-type Silicon. But is there any way to make Al/n-Si contact Ohmic?
For an n-type semiconductor, you will need to use a material that has a low work function (less than ~1.4 eV). I can't think of any metal that meets this requirement.
Aluminum has been used in the industry for a long time. You can use silicon doped (to avoid "spiking") aluminum and sputter and sinter it. You should dope the n-type silicon more highly to reduce the width of the Schottky barrier. Most metals form a Schottky barrier and the respective metal work function has only a small influence on the barrier height. Or look for a metal that forms a silicide, where that silicide forms a small barrier to n-type. See e.g., Article Redistribution of implanted phosphorus after platinum silici...
First of all Aluminum is used by all technologists. It is cheap and easy to processing. In advanced devices (systems) could be used wolfram.
Ti schould give an ohmic contact on n-type Si (50-100 mOhm*cm2 on n-type Si with about 3 Ohm*cm). You should do an HF dip before.
I don't know if you still need help, but I have achieved ohmic AlSi1% contacts on ultra-thin N+ doped Si just recently: my procedure consisted in removing the oxide from the surface with IBE, waiting a couple of days for a thin native oxide layer to regrow (native oxide reportedly prevents Si precipitation at the interface from AlSi) and sputtering my AlSi contact, followed by annealing at 450°C for 15' in forming gas. I don't know about ageing properties of this contact, yet.
Just be careful, since if you already have gold on your P-type contacts it could easily diffuse in the Si during annealing. I have recently found sputtered Pt to give good ohmic (I assume tunneling) contacts even without anneling, so this is probably your best bet.
Adding to the above colleagues: Ohmic contacts can be achieved by two methods:
The most recommended method is to supply the semiconductor layer by heavy doped layer such that one gets the so-called high low junction. For low doped n type one has to supply it with heavily doped n type doped layer with doping greater than the effective density of state.Then any metal, say one of the most common metals is Aluminum, can build tunnel contact with the heavily doped silicon. Such tunneling contact behaves ohmic with low contact resistance.
The other technique to get ohmic contact is to make the recombination rate at the surface of the silicon very high so that there will be no chance to build excess carrier concentration at the interface and so, the interface remains at equilibrium irrespective of the injection level and the metal contact to the semicondcutor behaves ohmic. In order to increase the recombination velocity one has to build many crystallographic defects at the contact surface. In addition,The presence of heavy metals at these defects enhance the recombination rate.
One can also choose a metal such that the barrier height at the interface is very low such that it subjects low impedance for the flow carriers in both directions.
The behavior of the metal semiconductor contacts is introduced in details in the book of S M Sze, physics of semicondcutor devices.
Best wishes
I am not sure whether these two links may be helpful or not.
http://iopscience.iop.org/article/10.1088/0022-3727/40/2/031/meta
Article Ohmic Aluminum‐n‐Type Silicon Contact
Please note that Al is a P type dopant belonging to the 3rd column of the periodic table. Al contact on N silicon is not recommended, and won t give the best results. Al on P silicon is a good choice and it is usually annealed at more than 600-700 degrees to enhance diffusion and doping on p silicon. Ohmic contact on N silicon, is possible when the doping is high and using Ti/Ag for example as in solar cells
Hi Tamal Dey
I am making similar device as you. I am just wondering what contact you used at the end and if it worked out. Would you please share your recipe with me.
Thanks
Al is definitely the wrong ohmic contact material for n-type Si. The standard ohmic contact is Au-Sb (roughly 0.1%) followed by a 400°C anneal. A simpler way is to evaporate Mg. See the figure on p. 7 here: https://web.stanford.edu/class/ee311/NOTES/Contacts.pdf
For ideal MS contact formed from a metal and an n-type
semiconductor will be rectifying if the work function of the metal is less than the work function of the semiconductor.
How do you make Au:Si 1% or Au-Sb (roughly 0.1%) in the physical vapor deposition?
- Badges
- Science topic
- Similar topics
- Filing