Action: azonenberg starts designing mask for fab run tomorrow | 07:27 | |
azonenberg | hi guys | 08:16 |
---|---|---|
soul-d | and added to startup join :P | 08:16 |
soul-d | well that was my point azonenberg i usaly don't understand stuff untill one works backwards and make stuff yourself | 08:17 |
azonenberg | Yeah | 08:17 |
soul-d | cool stuff so this is where digital meets chemical ? | 08:21 |
azonenberg | soul-d: digital, and mechanical | 08:21 |
azonenberg | http://i.imgur.com/wVEEB.png full wafer | 08:21 |
azonenberg | I am going to be doing the first two masks by contact lithography on a whole wafer (they dont need to be very accurate and will be the same for all dies) | 08:22 |
R0b0t1 | Which program is that? | 08:22 |
azonenberg | Glade | 08:22 |
azonenberg | free but closed source | 08:22 |
azonenberg | none of the open tools were to my liking so far, i'm still searching | 08:22 |
azonenberg | http://i.imgur.com/rTEkR.png is a single die | 08:22 |
R0b0t1 | Are you at the point where you are actually going to etch a full wafer? What's on the chips? | 08:22 |
azonenberg | not yet fully designed | 08:22 |
azonenberg | Ok, here's the process overview | 08:23 |
azonenberg | Spin coat resist, pattern "dicing" mask (blue rectangle in the background, full chip) by contact litho over a full 2 inch wafer | 08:23 |
azonenberg | This will leave lines (the space between the white and blue outlines) of photoresist with space between them wher ethe dies are | 08:23 |
azonenberg | Evaporate 1000nm of Cr on both sides of the wafer | 08:24 |
azonenberg | Lift off in acetone | 08:24 |
azonenberg | this will result in the back of the wafer being solid Cr and the front being Cr over dies and bare silicon over the spaces between them | 08:25 |
azonenberg | Etch in hot KOH for half an hour or so | 08:25 |
azonenberg | actually more like 45 mins | 08:25 |
azonenberg | Enough to go down about 100 microns | 08:25 |
azonenberg | So now i have Cr over unetched Si on the front and back of the dies and trenches 100um deep between them | 08:25 |
soul-d | sounds like fun | 08:25 |
azonenberg | Next step is to do a second contact mask | 08:25 |
azonenberg | the pink "backside" mask | 08:26 |
azonenberg | align to the edge of the dies | 08:26 |
azonenberg | and etch the Cr in HCl : H2O2 | 08:26 |
azonenberg | then return to the KOH | 08:26 |
azonenberg | now i am etching both in the center of the die and between the dies | 08:27 |
R0b0t1 | So your photoresist is positive, turn that into a negative of Cr, then etch? | 08:27 |
azonenberg | but the trenches are 100um deeper | 08:27 |
soul-d | see lots of fun chemical processes | 08:27 |
azonenberg | R0b0t1: I havent figured out all of the polarities for the masks yet | 08:27 |
azonenberg | in the CAD right now, filled areas are to not be etched | 08:27 |
R0b0t1 | Well, photo -> chrome -> etch? | 08:27 |
azonenberg | i may invert them before fab | 08:27 |
azonenberg | depending on where i do liftoff | 08:27 |
azonenberg | anyway so i then KOH a second time | 08:27 |
azonenberg | until the dies separate | 08:27 |
azonenberg | At this point, i have 76 loose chips | 08:28 |
azonenberg | 500um deep | 08:28 |
azonenberg | with 400um deep rectangular holes in the backs | 08:28 |
azonenberg | From now on, processing is done by projection litho one die at a time | 08:28 |
azonenberg | as the process is a little more experimental (and features are smaller) | 08:28 |
azonenberg | I then pattern the "fingers" mask | 08:29 |
azonenberg | Which i havent finished designing | 08:29 |
soul-d | ok so you made the "base" first ? | 08:29 |
R0b0t1 | So is this etch of the whole wafer to get you separate chips? | 08:29 |
azonenberg | R0b0t1: Yes, as well as to thin the dies in the center | 08:29 |
R0b0t1 | ah okay | 08:29 |
azonenberg | leaving a full-thickness ring around the thin part for easier handling | 08:29 |
azonenberg | Then i do the finger mask which is my actual MEMS geometry | 08:29 |
azonenberg | garden-variety comb drive | 08:29 |
azonenberg | 20um fingers at 20um spacing | 08:29 |
azonenberg | Etch this into the top surface | 08:30 |
azonenberg | While leaving the 400um deep hole unmasked | 08:30 |
azonenberg | So i'm now etching down and up simultaneously | 08:30 |
azonenberg | They'll meet around 50um from the top of the die | 08:30 |
azonenberg | leaving 20um wide and 50um tall fingers | 08:30 |
azonenberg | I then deposit an insulator - either spin on glass, thermal oxidation, or something TBD | 08:30 |
azonenberg | evaporate metal (aluminum most likely) | 08:31 |
azonenberg | and pattern the metal1 mask to separate the stator from the rotor | 08:31 |
azonenberg | as well as adding probe pads that i can test the device with | 08:31 |
azonenberg | Sonud like a reasonable process? | 08:32 |
R0b0t1 | . | 08:32 |
R0b0t1 | lol | 08:32 |
soul-d | sounds like fun :) | 08:32 |
R0b0t1 | It makes sense, but seems rather convoluted | 08:32 |
R0b0t1 | But right now you are only using one type of silicon? | 08:32 |
R0b0t1 | What about making transistors and having to use the P and N silicon? | 08:33 |
azonenberg | R0b0t1: I am only doing MEMS atm so doping is unimportant | 08:33 |
azonenberg | i think all of the wafers i have now are P type boron doped | 08:33 |
azonenberg | When i get to transistors, i will be using spin coated dopants and diffusion | 08:34 |
azonenberg | spin coat a layer of doped SiO2, photoresist over that, pattern | 08:34 |
azonenberg | etch oxide in 3% HF | 08:34 |
azonenberg | backfill with undoped oxide | 08:34 |
azonenberg | Cook at ~1100C to diffuse dopants into the silicon | 08:34 |
azonenberg | strip oxide | 08:35 |
soul-d | ill leave the HF to you though | 08:35 |
azonenberg | deposit field oxide | 08:35 |
azonenberg | soul-d: 3% is actually not THAT bad, though you still want to be careful | 08:35 |
azonenberg | i buy it at the local grocery store | 08:35 |
azonenberg | its an off-the-shelf rust remover | 08:35 |
soul-d | mmm ? | 08:35 |
azonenberg | whink brand rust stain remover | 08:35 |
azonenberg | brown plastic bottle | 08:35 |
azonenberg | 1.5-3% HF | 08:35 |
soul-d | ok lol cool | 08:36 |
soul-d | yeah ,y shed is chemical hazard anyhow | 08:36 |
azonenberg | At the same time, you want to wear gloves and preferably a lab coat | 08:36 |
azonenberg | as with any acid | 08:36 |
soul-d | got them and full face shield | 08:36 |
soul-d | abec1 filters | 08:36 |
azonenberg | Yep, me too | 08:36 |
azonenberg | What i am actually working on setting up is local exhaust for solvents | 08:36 |
azonenberg | i have a 6" duct fan but have to build the rest of the hood | 08:37 |
azonenberg | In any case its 4:30 in the morning and i have homework due at 1700 | 08:37 |
azonenberg | and class at 1400 | 08:38 |
azonenberg | I'd like to be somewhat awake for them | 08:38 |
soul-d | hehe | 08:38 |
azonenberg | so i'm off to get some sleep | 08:38 |
soul-d | goodnite | 08:38 |
Action: azonenberg AFKs: 503 Geek Temporarily Unavailable\r\nX-Server: azonenberg\r\nRetry-After: 21OCT2011 1030EDT\r\n\r\nThe requested geek is asleep. Your request should be repeated in six to eight hours. | 08:39 | |
azonenberg | And who said anything about night? It's the morning here :p | 08:40 |
soul-d | heh true nasty habbit i have to night time seems to work better | 08:42 |
reportingsjr | azonenberg: I would say it's /almost/ morning here, haha | 08:49 |
lekernel_ | http://www.youtube.com/watch?feature=player_embedded&v=dfNByi-rrO4 | 11:57 |
soul-d | interesting | 12:05 |
soul-d | multilayer ion paper :D | 12:11 |
soul-d | would be easy to make | 12:11 |
soul-d | could do fast prototypes if you have reliable draw surfaces probably more reliable then breadboard :) | 12:13 |
CIA-67 | homecmos r147 | trunk/mems-tests/ (22 files in 12 dirs) | First draft of new comb drive mask set | 20:44 |
azonenberg | lekernel: http://colossus.cs.rpi.edu/~azonenberg/downloads/combdrive-wafer.gds | 22:00 |
azonenberg | doing the litho for the first mask today, then the evaporation and liftoff monday or tuesday (was gonna be today but was delayed by homework) | 22:01 |
azonenberg | two contact masks on the whole wafer, then dicing and two projection masks on each die | 22:02 |
soul-d | you still where able to do homework ? :D | 22:04 |
azonenberg | soul-d: Got everything done on time, yes | 22:04 |
azonenberg | Skipped my one class of the day but its boring | 22:04 |
azonenberg | its the EE version of stuff i already learned in CS | 22:05 |
azonenberg | But i need credits in my schedule | 22:05 |
soul-d | lol | 22:05 |
soul-d | coulnd you jsut pretend to be there and do saomthing else | 22:05 |
azonenberg | i took so many CS classes as an undergrad that i'm having problems taking enough as a grad | 22:05 |
azonenberg | all of the interesting ones i already took | 22:05 |
azonenberg | so i have to find another dozen classes to take over the next few years | 22:05 |
azonenberg | i dont think there are 12 classes offered by the department i havent taken :p | 22:05 |
azonenberg | and especially if you count ones like "numerical methods for differential equations" that are super math heavy? | 22:06 |
azonenberg | So i'm starting to branch out into other departments out of necessity | 22:06 |
azonenberg | unfortunately, this often results in overlap | 22:07 |
azonenberg | and by the time you figure out you're going to learn nothing in the class, and it's too late to add a new one and drop this one, you're stuck with it | 22:07 |
soul-d | heh hard to get motivated | 22:07 |
azonenberg | lol | 22:08 |
azonenberg | funny thing is, this is the only EE class i've taken to date | 22:08 |
azonenberg | but its all stuff i learned in CS | 22:08 |
azonenberg | Anyway so as soon as it gets dark enough that i can do litho without sunlight frying my wafer, i'm going to do a photoresist coating followed by my first attempt at full-wafer contact litho | 22:10 |
azonenberg | Good news, the feature size is huge lol | 22:11 |
azonenberg | smallest geometry on this mask is 200 um | 22:11 |
Action: azonenberg should probably look into buying professionally made masks just once to see how good they are | 22:12 | |
azonenberg | http://www.laserlab.com/plotprices.php | 22:12 |
azonenberg | 12x18 inch mask for $52 | 22:12 |
azonenberg | i could fit an entire 4+ mask set for a 2-inch wafer on that sheet | 22:13 |
azonenberg | at 8000 DPI | 22:13 |
azonenberg | that's around 3.175 micron lambda with a four pixel min-feature design rule | 22:13 |
soul-d | lol wait let me scroll back din't look for a sec | 22:13 |
azonenberg | And if i do optical reduction of say 4x on that mask | 22:14 |
azonenberg | i could do some pretty fun stuff | 22:15 |
azonenberg | but even without reduction, full contact litho would be cool | 22:15 |
cheater | hi | 22:17 |
azonenberg | hi cheater | 22:17 |
azonenberg | I think once i get the comb drive mostly working on this process | 22:18 |
azonenberg | i am going to try ordering a film mask set from laserlab | 22:18 |
azonenberg | the comb drive masks plus some test patterns | 22:18 |
azonenberg | and see how it turns out | 22:18 |
soul-d | you should make youtubes :P soem visual stuff with your explaination might make more sense for me ;) | 22:19 |
lekernel | what are the missing steps for the comb drive? | 22:19 |
azonenberg | In theory, i know how to do everything | 22:19 |
azonenberg | there are two big unknowns atm | 22:19 |
azonenberg | first, how i am going to grow dielectric over the chip before metal 1 | 22:20 |
azonenberg | Probably thermal oxide but i need to buy / borrow time in a furnace | 22:20 |
lekernel | oxidation? | 22:20 |
azonenberg | second, how to deposit metal 1 | 22:20 |
azonenberg | i did evaporation for everything so far | 22:20 |
azonenberg | but evaporated metal is tricky because it has bad step coverage | 22:20 |
azonenberg | and i'm going to want to plate the sidewalls of the fingers | 22:20 |
azonenberg | Sputtering is an option but more complex | 22:21 |
azonenberg | In either case i am hopeful that by next weekend i will have at least one die with the full mechanical structure of a comb drive | 22:21 |
azonenberg | at which point i can run the same process on another ~ten dies | 22:21 |
azonenberg | and start screwing with metal and oxidation | 22:22 |
azonenberg | I feel like i'm very close, the Cr hardmask for deep etching is working well so far | 22:22 |
azonenberg | that was my last big problem | 22:22 |
cheater | what is step coverage? | 23:23 |
azonenberg | cheater: How conformal the coating is | 23:24 |
cheater | ok | 23:25 |
azonenberg | Good step coverage means that you'll cover sidewalls of vertical holes, etc | 23:25 |
cheater | yeah | 23:25 |
azonenberg | poor step coverage will cover the top and the floor but not the walls | 23:25 |
azonenberg | This is, however, good for lift-off processes | 23:25 |
cheater | why? | 23:25 |
azonenberg | where you need exposed sidewalls of the resist to get your solvent into | 23:25 |
cheater | hm | 23:25 |
azonenberg | With liftoff, you put photoresist on a bare wafer | 23:25 |
azonenberg | evaporate metal | 23:25 |
azonenberg | then dip in a solvent | 23:25 |
azonenberg | wherever you had resist, the metal is on top of it | 23:25 |
azonenberg | so it floats off when you put it in the solvent | 23:26 |
cheater | interesting | 23:26 |
azonenberg | Generally speaking, spin coating and sputtering are conformal (good step coverage, bad liftoff profile) | 23:26 |
azonenberg | so is CVD | 23:26 |
azonenberg | while evaporation is a good liftoff profile | 23:26 |
cheater | mhm | 23:26 |
cheater | can you mix these techniques at all? | 23:27 |
azonenberg | Certainly | 23:27 |
azonenberg | I mostly use spin coating at home and evaporation on campus | 23:27 |
azonenberg | i plan to build evaporation and sputtering rigs at home eventually | 23:27 |
azonenberg | CVD is more complex and dangerous as it requires compressed cylinders of some distinctly unpleasant gases | 23:27 |
azonenberg | i think i cna avoid it entirely | 23:27 |
azonenberg | can* | 23:27 |
cheater | no reason to die quickly | 23:32 |
azonenberg | Lol | 23:32 |
azonenberg | Well, worse than CVD would be RIE | 23:33 |
azonenberg | One of the nicer chemicals they use for RIE is chlorine gas | 23:33 |
azonenberg | But they've used stuff as nasty as chlorine trifluoride | 23:33 |
azonenberg | the two biggest baddest halogens all in one molecule | 23:33 |
cheater | haha | 23:33 |
cheater | yeah | 23:34 |
azonenberg | the stuff ignites on contact with ANYTHING | 23:34 |
azonenberg | including glass and asbestos | 23:34 |
cheater | wouldn't want to breathe that in | 23:34 |
azonenberg | And produces hydrofluoric acid in the smoke :p | 23:34 |
cheater | doesn't chlorine trifluoride get used as a rocket fuel too | 23:34 |
azonenberg | cheater: they tried a few times | 23:35 |
azonenberg | it's extremely effective | 23:35 |
azonenberg | if you can manage to handle it | 23:35 |
cheater | wasn't there that funny blog post about it | 23:35 |
azonenberg | yeah | 23:35 |
cheater | heh, is it difficult to fabricate? | 23:36 |
azonenberg | From what i understand, get a cylinder of chlorine and fluorine | 23:36 |
azonenberg | and mix them up lol | 23:37 |
azonenberg | If anything within a quarter mile of you survives the resulting reaction, you now have a tank of ClF3 | 23:37 |
cheater | i wonder if you need to do this under specific pressure/temperature | 23:38 |
azonenberg | I have no intention of attempting the reaction lol | 23:39 |
azonenberg | Or even going anywhere near the precursors | 23:39 |
azonenberg | the strongest thing i use in my fab work is 1-3% HF | 23:39 |
cheater | in what | 23:42 |
azonenberg | Water | 23:42 |
cheater | that sounds like mouth wash | 23:42 |
azonenberg | lol, no | 23:43 |
cheater | why? | 23:43 |
azonenberg | you're thinking NaF and its around 0.01% | 23:43 |
azonenberg | maybe less | 23:43 |
cheater | yeah, one digit less or more.. who cares | 23:43 |
cheater | as long as it keeps your breath fresh | 23:43 |
azonenberg | http://www.whink.com/index.cfm?fuseaction=cEcommerce.Product&ProductID=4328&CategoryID=1541 | 23:43 |
cheater | hahaha | 23:43 |
azonenberg | http://www.whink.com/cmssites/ws0811www.whink.com/uploads/Documents/MSDS/Moonglum/Rust%20%20Stain%20Remover%2010062008.pdf | 23:44 |
azonenberg | most definitely not mouthwash :p | 23:44 |
cheater | :) | 23:44 |
--- Sat Oct 22 2011 | 00:00 |
Generated by irclog2html.py 2.9.2 by Marius Gedminas - find it at mg.pov.lt!