On 1/15/2013 9:00 PM, Andrew Sarangan wrote: > Having done interference lithography for more years than I'd like to admit, > I can offer some of these suggestions: > > - what presents itself as sloped sidewalls, under-developed or underexposed > can often be entirely due to poor resist/substrate interface. Lines that > are fully developed with straight walls can often float away during > development, making diagnostics very difficult. I've had much better > success since switching to vapor priming the substrates (assuming silicon). I tend to see scalloped resist profile, almost like an inverted cycloid. With long enough development time I am left with the cusps of the inverted cycloids. I don't think the resist is underexposed as I am working at an exposure dose where the development rate plateaus. As I understand it, this should correspond to the regime where the DNQ has been fully broken down, giving the highest ratio between development rate between unexposed and exposed regions. Currently, I am spin coating an 80/20 primer. I have a standing offer to vapor prime some wafers with Bill Moffatt's company, so I should probably pursue that. > - I have had better success with i-line resists than with g-line, even as > far away as 488nm (Ar ion laser). The sensitivity will be low, so you have > to increase the dose. Interesting. I wouldn't have considered trying that. Currently, I don't have any i-line resists. > - You can get even better sidewall results with an anhydrous ammonia image > reversal. Yes, I'm working on a proposal to acquire an oven capable of carrying out this process. > - as for resistance to fluorine plasma etching, it depends on what > substrate you are trying to etch. Most resists are more sensitive to ion > energy than to the fluorine radicals. Reducing the ion energy and > increasing the plasma density would give better results, although it will > produce some undercut. An ICP would be useful here to control these > parameters. Thank you for the information. We have been able to obtain anisotropic etching so far. We can also implement hard bake (which hasn't been necessary so far) to improve selectivity. > - For even greater etch resistance, I have used negative-acting chemically > amplified resist in the deep-UV (266nm). We are currently limited in our operating wavelength. It looks like those DUV resists don't absorb at our wavelength. If we have selectivity problems we have some ideas to get around that. > - As for dilution, I am surprised to hear the difficulty getting consistent > dilutions. I have found the variations due to spin & bake steps to be more > significant than the dilution step. If you use a sensitive weighing scale, > and make large batches, that should not be much different than the premixed > bottles. This was probably due to working in small batches and running into snags related to mixing vs. degassing. Thank you for your advice! Justin -- Justin M. Hannigan, Ph.D. Process Development Engineer LightSmyth Technologies,Inc. 875 Wilson St, Unit C Eugene, OR 97402 Tel: 541-431-0026 www.lightsmyth.com !DSPAM:50f6e31d205389100839904! _______________________________________________ Hosted by the MEMS and Nanotechnology Exchange, the country's leading provider of MEMS and Nanotechnology design and fabrication services. Visit us at http://www.mems-exchange.org Want to advertise to this community? See http://www.memsnet.org To unsubscribe: http://mail.mems-exchange.org/mailman/listinfo/mems-talk