I have attached a metallic thin sample in Si substrate using FIB. After that, I have made the connection by depositing platinum in FIB with the sample and the pre-pattern platinum electrode. The prepattern platinum electrode is made by photolithography which has contact resistance around 10 ohms. But after making the connection with the pre-pattern platinum electrode by depositing the platinum in FIB the resistance become very high.
Can anyone suggest how can I improve my connection?
Actually, GIS-deposited "platinum" is not a pure metal. Typically, It is pretty spongy and contains some non-metal inclusions. Thus, poor electrical parameters are not surprising.
Possible solutions:
1. Just increase the thickness of the deposited layer;
2. Coat the sample with Pt using a normal sputter (perhaps with a mask), and then remove the excess of metal with FIB.
Good luck!
Typical FIB-deposited "Platinum" is in reality a composite of Pt particles with dimensions on the scale of ~5nm - ~8nm suspended in the matrix of amorphous carbon with some Ga contamination. Exact composition is heavily depent on the deposition conditions (precursor gas flux and beam parameters) but you can expect that less then 50% of the deposited material is actual Pt, while rest being mostly C and Ga in single percentage. FIB Pt deposits have very high resistance.
To get low-resistance contacts you can try:
1. FIB deposition of Cu from Cu(hfac)2 precursor. Good luck getting hands on only 5 or 6 FIB instruments with such capability that were ever made and sold commercially, but if you are lucky to find one of them then purely metallic depo is straight-forward.
2. FIB deposition of W from W(CO)6 or Mo from Mo(CO)6 precursors. These are commercially available, and Mo precursor can be charged into any system designed for W deposition. Resistivity is still fairly high, but you can expect metal content ~>60% and with properly designed deposition recipe resistivity 1/2 to 1/3 of Pt deposits.
3. You didn't mention lateral dimensions of your sample, but on the scale of >~20um you can try using micro-manipulator to position droplets of Ga to wet your contacts and establish low-resistance contact.
4. Increase area and thickness of FIB deposition to lower contact and bulk resistance.
5. For W and Mo FIB deposits - create a recipe working in near-sputtering deposition regime, to increase Ga content in the deposited material. Not sure if this approach would work for Pt - haven't tried.
6. If the dimensional scale of your sample is large enough, then use laser-induced Pt deposition. It is commercially available now from Varioscale Inc., or you may be able to find one of the older systems from former Revise Inc.
7. Do a hybrid lift-off: dispense thin layer of photoresist on your sample with ultrasonic nozzle, selectively expose areas of contacts with laser ablator at low power or with just OM light by reducing illuminated area with field aperture, use ordinary sputter coater to deposit metal, and finally dissolve resist in solvent.
8...Multiple other approaches are possible, depending on material and dimensions of your sample, purpose of the whole exercise, availability of equipment, chemicals, funding, etc....
Good luck :)
Valery
Hello Sk,
the contact resistance by FIB (Focused Ion Beam) is very high due to the following reasons:
- The contact material is a mixing of Pt and carbon (due to the cracking of a hydrogen-carbon- molecule)
- the ion beam destroys the surface of silicon
- you have an intermediate layer between Si and C (deep centers)
- a evaporated contact (if well prepared) is more clean and pure than a FIB contact
FIB is a nice tool but it is unsufficient if you need a low resistance. After our experience I recommend to try another supplier for the precursor material. The resistance of deposited material can differ some order in magnitude. It is not recommended to buy the cheapest material.
With Regard
R. Mitdank
Hello Sk,
the contact resistance by FIB (Focused Ion Beam) is very high due to the following reasons:
- The contact material is a mixing of Pt and carbon (due to the cracking of a hydrogen-carbon- molecule)
- the ion beam destroys the surface of silicon
- you have an intermediate layer between Si and C (deep centers)
- a evaporated contact (if well prepared) is more clean and pure than a FIB contact
FIB is a nice tool but it is unsufficient if you need a low resistance. After our experience I recommend to try another supplier for the precursor material. The resistance of deposited material can differ some order in magnitude. It is not recommended to buy the cheapest material.
With Regard
R. Mitdank
Hello Sk,
besides the methods mentioned above:
Some authors used annealing in oxygen or water vapour to reduce the carbon content of the deposits leading to better conductivity, see e.g.:
Thesis Dr. A.P.J.M. Botman (2009): "Toward high purity nanostructures from electron beam induced deposition of platinum"
https://repository.tudelft.nl/islandora/object/uuid%3A86e39f88-dfdb-46c3-b69d-b2d06aded1b5
Regards
Stefan
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Probability
- Probability Theory
- Reliability