#homecmos IRC log for Monday, 2012-01-09

azonenbergbarrettfriedman: But the thing is, if you sputter over the resist01:02
azonenbergOdds are pretty good that the sidewalls will be somewhat conductive01:02
azonenbergmay not be 100% covered to the same thickness as the tops, but there will be stuff there01:02
azonenbergAnd, furthermore, if you then dip it in solvent it may not separate cleanly or at all01:03
azonenbergon the other hand if you use evaporation i suppose it's possible to do what you want, but that's only if your design involves a single metal deposition step01:03
azonenbergalso, what are you trying to deposit? Liftoff is generally only used for materials that are difficult or impossible to etch01:04
azonenbergbecause it can lead to high edge roughness as well as pretty much limiting your deposition method to evaporation01:05
azonenbergwhereas conventional lithography can be used to pattern almost anything01:05
Syncneat little belljar azonenberg01:32
azonenbergSync: ty01:32
azonenbergNot mine technically, was bought by a company i did consulting with last year01:33
azonenbergBut they said i could hold onto it and use it for whatever i want01:33
azonenbergas long as they get to use it if they need it back in the future01:33
Syncbut probably not up to spec for some high vacuum work01:34
azonenbergSync: I was planning to turn it into a sputter coater01:34
azonenbergi can hit adequate vac for sputtering with the mechanical pump there01:34
Synchow low does it go?01:34
azonenbergI do need to redo some of the fittings, we totally disassembled it to find a leak01:34
azonenbergOf all the places to be leaking, the Pirani gauge :p01:34
azonenbergso i need a new one01:34
azonenbergWe had it down to high tens of mtorr, maybe 60ish? The pump is rated for 40 but we had a slow leak01:35
Syncyou can probably seal it with regular epoxy01:35
azonenbergThat's easily deep enough for DC sputtering01:35
Syncwell, I'm currently shopping for a vacuumchamber of my own01:35
azonenbergHow far do you want to go?01:36
azonenbergI eventually would like to get deep enough for evaporation01:36
azonenbergThe fittings on that chamber are def not good enough01:36
azonenbergBut the jar and plate might be01:36
Syncwell, I have a rubber sealed turbo, but can get metal gaskets for it01:36
azonenbergYou have a turbomolecular pump? o_O01:36
Syncand four dual stage rotary vanes01:36
azonenbergVery nice01:37
Syncwell okay, two,  two are fubar01:37
azonenbergThe pump i have is a dual stage rotary vane01:37
SyncI have a leybold d4b and a d8b working01:37
azonenbergMy pump is a less fancy brand, i forget what01:37
Syncthe others are ilmvac devices, used in chemical service and corroded to death01:37
azonenbergBut it hit 40mtorr when we attached it straight to the pirani gauge01:37
azonenbergAnd we got the entire bell jar assembly down to below 10001:38
azonenbergbefore the leak got worse01:38
azonenbergWhen i have $1k ish to spare i want to rip it all out and put in KF16 or KF25 fittings01:38
Synceh, that's only like 5^-2mbar ):01:38
azonenbergthats what we priced them out as01:38
azonenbergthat would allow me to go nice and deep if i had a suitable pump01:38
SyncI spent less than that on my vacuum equipment01:39
azonenbergthe base is a solid chunk of metal01:39
azonenbergSync: that was for new01:39
Syncbut the biggest problem will be the seal on the bottom01:39
azonenbergthe albany valve and fitting company is like 10 mins from my apartment01:39
azonenbergYou mean baseplate to jar?01:39
azonenbergOr baseplate to pipe01:39
Syncbaseplate to jar, you really want viton there01:39
azonenbergWe have it01:39
azonenbergviton + dow corning high vacuum grease01:39
Syncyay, I got some pfpe grease for cheap01:40
azonenbergthe limiting factor was the leak01:40
azonenbergand once i fix it, probably the pump01:40
azonenbergBut again for dc sputtering i dont *need* to go that low01:40
Synchttp://sync-hv.de/hi-vac/bpg400.jpg those are nice01:41
azonenbergwhats that?01:42
azonenbergfitting looks like a KF40 or so01:42
Syncno, it's kf2501:42
azonenbergah, and what is the device?01:43
Syncit's a bpg400 combined bayard-alpert and pirani gauge01:43
azonenbergwhats the effective range?01:43
Syncatm - 10^-901:43
Sync10^-11 on the new generation01:43
SyncI also own a mass spec01:43
azonenbergwhat'd it cost you, and what would it cost new01:43
azonenbergthe gauge, not the spec01:43
Syncthe gauges were more expensive than the spec01:44
Syncthey were 15001:44
Syncnew, uuh around 1k01:44
Synchttp://sync-hv.de/hi-vac/spectrometer.jpg small quadrupole RGA01:45
azonenbergNot familiar enough with specs to know anything about that01:45
azonenbergMy goals for tooling involve building a DC sputtering system (argon only at first, maybe add N2/O2 inlets for reactive sputtering later)01:46
azonenbergand a filament evaporator01:46
azonenbergThen in the long term maybe an electron microscope01:46
SyncI still bite my ass that I did not bid more on the varian scroll pumps here01:47
azonenbergBut in the time it'd take to build one i'm thinking i could probably save up for a real one01:47
Syncmaybe I can get one from work when it need overhaul01:47
Syncthen I could run pure o201:47
azonenbergI want a SEM01:47
azonenbergI have nowhere to put one in this apt but when i get a place of my own01:47
SyncI'd like to have the old ion implanter here that oxidizes at the university01:49
azonenberglol, ion implanter? That would be convenient all right01:49
azonenberghomemade SIMOX?01:49
Syncwell, I operate on as a job01:49
Syncor rather, I stand in front of it and scream at it that it does what I want from it01:49
azonenbergyou a cleanroom technician or something?01:49
azonenbergprocess engineer?01:50
Syncno, first semester student with some contacts01:50
azonenbergoh lol01:50
azonenbergLucky you, i didnt get into the cleanroom until i was a sophomore01:50
azonenbergAnd that was only because i was working with a third party company01:50
azonenbergthat had me as a consultant, it wasnt for school01:50
Syncwell, it's not like our process would need a clean room at all01:50
azonenbergMine doesnt :p01:51
azonenbergI would use one if i had it01:51
Syncbecause we completely fail getting things right atm01:51
azonenbergAnd i plan to build one at home for when i shrink geometry to below 20um01:51
azonenbergbut this is in my dream home / evil lair / $TERM_OF_CHOICE01:51
azonenbergWhich is five to ten years out at best01:51
Synca laminar flowbox should be enough01:51
Syncfor most stuff01:51
azonenbergI want an entire room so i can have, for exmaple, a fume hood inside it01:52
azonenbergfor spin coating and etching01:52
azonenbergand so i can have microscopes without dust covers :p01:52
azonenbergplan is for that room to contain a hood/wet bench, my SEM, a metallurgical microscope, and whatever evap/sputtering rig i build01:54
azonenbergmaybe 10-12 feet square?01:54
Syncbut I really hate the ion implanter01:54
Syncyou cannot get same results on different days because it decides that the same values on some valves produce different results01:56
Syncso you kinda have to get a feel for it01:56
azonenbergReminds me of the evaporator in the materials research center on campus01:59
azonenbergit's the only tool i use in my home fab processing that isnt actually at home01:59
azonenbergwell besides the SEMs here and there01:59
azonenbergit's  a 30-year-old CVC evaporator02:00
azonenbergfull manual02:00
azonenbergactual valves you turn by hand02:00
azonenbergBut it's so simple it almost never breaks down lol02:01
azonenbergThoguh admittedly it is hard to get reproducible thicknesses02:01
Syncyes, this is the same problem here02:01
azonenbergBut so far i havent needed to worry about exact thicknesses02:09
azonenberga few percent variation is OK02:09
azonenbergMy current short term goals are to improve lithography resolution to below 20um where it is onw02:11
azonenbergi have a roadmap down to 1.25um02:11
azonenbergbut it will use commercially printed 8000DPI masks instead of the 600DPI laser pritner i use now02:11
azonenbergAnd if i could get the 40x objective to play nice i could theoretically hit ~350nm02:12
azonenbergmore realistically i think 500-750 would be the limit02:13
berndjazonenberg, re 8000dpi, maybe not quite that res but still significantly beyond 600dpi, can you get someone with an http://en.wikipedia.org/wiki/Imagesetter to do it a little more affordably than a full mask fab?12:14
azonenbergberndj: they tend to not have the greatest results, can be noisy and have stray dots etc14:16
azonenbergsince they usually arent fabbed in cleanrooms14:16
azonenberglaserlab.com will make sub-$100 masks at 3.125um lambda and 12.5um minimum feature design rules14:17
azonenberg12x18 inches, 10x16 printable area14:17
soul-dif i din't follow preelections i would have missed that19:43
soul-dfree pony's for all american's19:44
--- Tue Jan 10 201200:00

Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!