Vipul, 1000 Ang peak to peak across a few mm long scan sounds very challenging. We typically see about 2000 ang peak to peak using KOH (with a very clean tank and agitation). Adding IPA to the soltution may help, athough we have not characterized it. We suspect the majority of the roughness to be the KOH attacking some of the secondary planes in the lattice (?) EDP may perform a bit better if this is the case, as it has better selectivity to the (111) plane. I would suspect TMAH would be worse. Eric Miller Laboratory Manager Washington Technology Center Ph: 206 616-3855 www.watechcenter.org -----Original Message----- From: mems-talk-bounces+em3=u.washington.edu@memsnet.org [mailto:mems-talk-bounces+em3=u.washington.edu@memsnet.org] On Behalf Of Vipul Patel Sent: Wednesday, June 16, 2004 11:22 AM To: mems-talk@memsnet.org Subject: [mems-talk] KOH Etching - surface roughness of v-groove surfaces Dear MEMS experts, I have a need to have a mirror-like surface of the etched V-grooves ( macroscopic) using an-isotropic wet silicon etch. When I used conventional KOH etching ( I think it was 40% sol. at 80 deg. C and a stirrer was used), the resultant surface has several features larger than 2 micron.( I measured it by cleaving the V-groove and measuring the surface roughness of the etched surface by dektek profilometer.) The crystal plane was aligned to better than 0.05 deg. during lithography process using an additional mask. I came across several papers related to improving the surface roughness, however being an Electrical Engineer, I do not understand chemical etching well. Our goal is to have a surface with less than 0.1 micron peak-to-peak roughness on macroscopic level ( across a few mm long facet) Can anyone suggest on what changes should I make to etch process and share their experience? Do I have a spec. that is not doable? Eventhough I have no experience, I can probably also try TMAH or EDP solutions. Also, does anyone recommend a MEMS foundry with experience in this specific area. If they can guarantee the spec. , I can probably give them the entire device/wafer fabrication. With best regards, Vipul Patel.. _______________________________________________ MEMS-talk@memsnet.org mailing list: to unsubscribe or change your list options, visit http://mail.mems-exchange.org/mailman/listinfo/mems-talk Hosted by the MEMS Exchange, providers of MEMS processing services. Visit us at http://www.memsnet.org/