azonenberg | nathan7: ping | 04:24 |
---|---|---|
azonenberg | I ran a test of parallel lines on the 5um process | 04:25 |
azonenberg | Not entirely successful, my exposure lamp burned out halfway through the run so it underexposed :P | 04:25 |
azonenberg | but i decided to develop and etch anyway to see what i got | 04:25 |
azonenberg | Nowhere near all of the pattern came out but i did demonstrate optical resolution adequate to clearly separate the lins | 04:26 |
azonenberg | lines* | 04:26 |
azonenberg | pic following in a bit | 04:26 |
CIA-67 | homecmos r116 | trunk/lithography-tests/labnotes/azonenberg_labnotes.txt | Today's lab notes | 05:03 |
azonenberg | http://colossus.cs.rpi.edu/~azonenberg/images/homecmos/2011-08-26/5h10_004_annotated.jpg | 05:35 |
azonenberg | Latest test on the 5um process | 05:35 |
azonenberg | Not exactly successful, these are supposed to be unbroken lines | 05:35 |
azonenberg | But it demonstrates i have the optical resolution to hit 5um | 05:35 |
azonenberg | i just need to improve the develop/etch process | 05:35 |
azonenberg | A lot :P | 05:35 |
azonenberg | swkhan: check out the link above | 05:37 |
azonenberg | this is what i'm working on in my living room fab | 05:37 |
azonenberg | the 5um process is under active development but isnt yet feasible, the 20 gives reasonably good results but i still have some glitches in my exposure system | 05:38 |
nathan7 | azonenberg: pong | 07:58 |
azonenberg | http://colossus.cs.rpi.edu/~azonenberg/images/homecmos/2011-08-26/5h10_004_annotated.jpg | 07:59 |
azonenberg | today's test run | 07:59 |
nathan7 | mhm | 07:59 |
azonenberg | overetched and underexposed but does demonstrate adequate resolution | 07:59 |
nathan7 | :) | 07:59 |
azonenberg | the 5um process is clearly not ready for prime time though lol | 07:59 |
azonenberg | these are, even so, the smallest structures i've made to date | 07:59 |
azonenberg | note the scale bar | 07:59 |
azonenberg | nominally 5.29 micron half pitch | 08:00 |
azonenberg | they were theoretically 5.29 lines and 5.29 trenches but as you can see i overetched | 08:01 |
azonenberg | so the trenches got bigger and the lines got destroyed | 08:01 |
nathan7 | mhm | 08:01 |
nathan7 | It's a start | 08:01 |
nathan7 | ..don't you have this "wait, whoa, I can _MAKE_ _FIVE_ _MICROMETER_ _STRUCTURES_" | 08:02 |
nathan7 | every once in a while? | 08:02 |
nathan7 | I have that with 3D printers | 08:02 |
nathan7 | every once in a while, I go like "wait, WHOA, I can _PRINT_ _OBJECTS_" | 08:02 |
nathan7 | http://xkcd.com/354/ | 08:03 |
nathan7 | azonenberg: :D | 08:05 |
azonenberg | and lol, i have that sometimes with all kinds of projects | 08:05 |
nathan7 | (= | 08:06 |
azonenberg | This gives you a better idea of the devastation caused by the underexposure http://colossus.cs.rpi.edu/~azonenberg/images/homecmos/2011-08-26/5h10_002_annotated.jpg | 08:06 |
azonenberg | the entire circular field should have had grating in it :P | 08:06 |
azonenberg | that's the FOV of my system | 08:07 |
azonenberg | Another potential problem is that i am testing on a decently thick copper layer | 08:08 |
nathan7 | mh-hrm | 08:08 |
azonenberg | You normally dont think of a micron as being thick :P | 08:08 |
nathan7 | hehe | 08:08 |
azonenberg | But when you're trying to do 5um features in it | 08:08 |
azonenberg | using a *wet etch* | 08:08 |
azonenberg | rather than RIE | 08:08 |
azonenberg | it becomes problematic | 08:08 |
azonenberg | this is comparable to doing around a 150um PCB trace in 30um copper | 08:09 |
azonenberg | Doable, but requires careful process control | 08:09 |
azonenberg | http://colossus.cs.rpi.edu/~azonenberg/images/homecmos/2011-08-26/5h10_005_annotated.jpg | 08:09 |
azonenberg | the shiny areas are the only part of the lins that weren't at least somewhat attacked | 08:09 |
azonenberg | lines* | 08:09 |
azonenberg | again, severe overetch | 08:09 |
azonenberg | http://colossus.cs.rpi.edu/~azonenberg/images/homecmos/2011-08-26/5h10_008_annotated.jpg is the side of the exposure field showing the edge of the etched area | 08:10 |
azonenberg | http://colossus.cs.rpi.edu/~azonenberg/images/homecmos/2011-08-26/5h10_012_annotated.jpg is the edge of the die | 08:10 |
azonenberg | the copper doesnt quite go all the way to the edge because some of it chipped off when i cleaved the die off the wafer (coating was done before separation of dies) | 08:10 |
azonenberg | http://colossus.cs.rpi.edu/~azonenberg/images/homecmos/2011-08-26/5h10_014_annotated.jpg is another corner shot | 08:11 |
azonenberg | That last pic is cool because it's showing submicron features lol | 08:12 |
azonenberg | not created by me, but present on the edges nonetheless | 08:12 |
nathan7 | :o | 08:13 |
azonenberg | i've been looking at some of the rough bumps along the edge and it looks like i'm able to clearly resolve points around 600nm apart | 08:14 |
azonenberg | i'm looking to see if there are any smaller features i can find | 08:14 |
azonenberg | to enahance resolution i turned off the red/blue channels which eliminates any chromatic aberration | 08:15 |
azonenberg | hmm, one of the pinholes near the bottom left of the film looks to be around 350nm wide | 08:15 |
azonenberg | the crack on the middle left is around 400ish | 08:16 |
azonenberg | So do you know what that means? | 08:16 |
azonenberg | It means my optics are good enough for submicron resolution if i had a mask that was fine enough | 08:18 |
azonenberg | i could potentially hit say 500nm with these lenses if i had a fine enough mask | 08:18 |
nathan7 | azonenberg: ..awesomesauce | 08:19 |
azonenberg | There are of course some downsides | 08:19 |
azonenberg | First off, it'd be immersion lithography | 08:19 |
azonenberg | Which would mean i'd have to clean oil off the die after exposure before developing | 08:20 |
azonenberg | without scratching anything | 08:20 |
azonenberg | so i'd need a solvent/surfactant that dissolves or emulsifies my immersion medium without harming photoresist etc | 08:20 |
azonenberg | second is that the FOV of my 100x objective is around 160 um :P | 08:20 |
azonenberg | so that'd be the maximum die size lol | 08:21 |
azonenberg | Realistically though, i do intend to attempt that soon | 08:21 |
azonenberg | i have a selection of immersion oils from Ted Pella inbound that i'm gonna play with and see which gives good results | 08:21 |
azonenberg | And lol, are any of you guys among the 8 fans of immersion litho on facebook other than myself? | 08:22 |
azonenberg | https://www.facebook.com/pages/Immersion-lithography/113536905362750 | 08:22 |
nathan7 | I sould | 08:23 |
azonenberg | Most semiconductor people use water for immersion but i'd need a water immersion objective so for now i'll be using normal microscope immersion oil on my oil lens | 08:23 |
azonenberg | there are a lot of highly technical topics on fb that have zero or a handful of fans lol | 08:24 |
azonenberg | like Ta2O5 | 08:24 |
azonenberg | i'm the only fan https://www.facebook.com/pages/TantalumV-oxide/116628238385003 | 08:24 |
Viper-7 | lol | 20:50 |
--- Sun Aug 28 2011 | 00:00 |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!