We contact Au nanowires, 100 nm across and 18 nm deep (20-30 nm long, with wider contact wires), in a vacuum probe station. These wires are very sensitive to electrostatic discharge - even a couple volts is enough to blow them.
To mitigate that, we take the following steps:
- We are grounded at all times with a wrist strap.
- We set our electronics, SMUs, to output zero volts as we contact the device. Other times we try contacting the devices with them connected directly to ground through a 1 megaohm resistor (since the ground itself could experience shocks from the environment).
- We place a TVS diode with a sub-1-V cutoff across the leads when contacting.
Nevertheless we have a 20-40% failure rate due to ESD. Is there anything else anyone might recommend trying?
Thank you in advance!
Hello Darin,
I find that SU-8 has good dielectric properties for this type of application. I use it to prevent charge build-up and to reduce fringe capacitance.
Some properties of SU-8:
http://www.microchem.com/pdf/SU-8-table-of-properties.pdf
Thanks for the responses!
When you say place polymer, where were you thinking - coating on top of the nanowires? Unfortunately the nanowires are coated in a layer of a molecule for testing that I am concerned would be disturbed by the SU-8.
Darin,
Apologies, didn't see your follow-up question.
I originally meant that you could use a SU-8 layer on top of the structures to provide passivation. It's difficult to provide a perfect answer without understanding the limitations of geometry and the fabrication process, but I will try to provide some information.
If you wish to keep an 'open' geometry (for well/droplet sensing) you would want to cover the top surface of Au with SU-8 (see attached images of a device I am working on at the moment). SU-8 can easily be dry-etched in O2 using RIE or ICP (ICP leaves less residue). Undercut is ~15 nm for RIE (80 nm SU-8) and
Thanks Jules, awesome in-depth response. If the nanowire itself weren't the predominant culprit in ESD failure, I think a two-step SU-8 process as you propose to protect the contact wire might be effective. Sadly the nanowire is the issue, so for now I don't think that this capping would satisfy our needs. Still, I hope it helps someone else!
Not sure you can ever avoid ESD, but as mentioned above, you might apply an insulating material around ESD area, such as low dielectric constant BCB.
Hi Qijin - when you say 'increase the distance between' what exactly do you mean? We have one Au nanowire for one device, about 80 nm long contacted by 10 um wide contact wires.
I don't really understand what you are doing so it is difficult to answer. However I do know that hard drive manufacturers routinely handle magnetoresistive read heads that have sensitivity in the volt region and it is possible to control ESD to very low levels. You have to know what you are doing though. The key is to understand the ESD threats and how they are applied to the victim devices. It's likely that you will need to carefully eliminate non-essential insulators from the surrounding area. All insulating materials can charge up and induce voltage differences on conductors that can lead to ESD. Insulators that are an essential part of your process will need to be evaluated for their contribution to ESD risk, and if necessary some way of mitigating that risk must be found. oltage differences btween the nanowires and any other conductor they make contact with would probably need to be minimised or eliminated. I don't know what this means in practice in your case.
Grounding a person with an ordinary wrist strap may not keep their body voltage at a sufficiently low level. a lower resistance to ground that a standard wirst strap cord gives may be needed. If your personnel are touching the nanowires this is the first thing I would look at.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Graphs