DIY chips / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

DIY chips

first previous 2 3 4 5 next

Move Thread

LAN_403

IntraWinding

Sun Mar 14 2010, 12:01PM

Registered Member #2261 Joined: Mon Aug 03 2009, 01:19AM
Location: London, UK
Posts: 581

[Recovering composure]: I got a couple of bottles of Winks Rust Stain Remover when I visited the USA, just taking advantage of the opportunity as it's hard to find Hydrofluoric Acid in the UK. I'd read about it before I travelled so had a look on the shelves of the first likely looking shop and there it was, next to boot polish and kettle descaler. Presumably letting the public use it as a normal household cleaner isn't causing too much trouble over there? I'm not sure what concentration it is but it dissolves glass significantly in a few minutes (the finish is transparent rather than 'etched').

Conundrum

Sun Mar 14 2010, 12:15PM

Registered Member #96 Joined: Thu Feb 09 2006, 05:37PM
Location: CI, Earth
Posts: 4062

hmm.
Interestingly a source of small wafers is old broken PIR sensors.

these use a pretty pure Si wafer as the thermal infrared pass filter.

another idea i have is to use the backs of old Si solar cells but would this be considered cheating?

-A

IntraWinding

Sun Mar 14 2010, 12:30PM

Registered Member #2261 Joined: Mon Aug 03 2009, 01:19AM
Location: London, UK
Posts: 581

There are loads of cheap wafers on eBay, like this

hboy007

Sun Mar 14 2010, 12:56PM

Registered Member #1667 Joined: Sat Aug 30 2008, 09:57PM
Location:
Posts: 374

Conundrum wrote ...

another idea i have is to use the backs of old Si solar cells but would this be considered cheating?

Not if you manage to do the polishing down to a few nanometers

They will be unprocessed because it's cheaper.

Interesting thread, I must say. Making active semiconductor structures is something I've been dreaming of for a while now but I am still waiting for my TMP to arrive. I will start with the simple things first, like aluminium coatings. Maybe I will try my luck once I get dielectric mirrors done properly.

Jeri Ellsworth

Sun Mar 14 2010, 10:45PM

Registered Member #2737 Joined: Sat Mar 13 2010, 07:34AM
Location:
Posts: 9

I'm not sure how good devices will be if built on the backside of wafers, since they are often backside lattice damaged to make getter locations. They sometimes have nitride or other layers on the backside that would need attention.
There are companies that will sell you reclaimed wafers that are lapped down past the previously processed wafers.

has many exotic types of wafers available, but I have not purchased from them.

I've purchased from

and the owner is very nice and showed me around her facility. The clean room / container charge was more than the wafers. BTW.

I should also note that the guys at

were very nice, although the older gentleman seemed to have a hard time hearing me on the phone. I've been mostly using the solar cell kit and the dopants work well for transistors too. The flammable alcohols in them require special shipping that was costly.

-Jer

[Edit: Double post]

mikeselectricstuff

Mon Mar 15 2010, 12:21AM

Registered Member #311 Joined: Sun Mar 12 2006, 08:28PM
Location:
Posts: 253

Jeri - I recently watched the video of a presentation you gave on this, where you mentioned problems getting hold of 'proper' photoresists. I was wondering if you'd looked into resists used in the PCB industry? You can get spray-on photoresist for hobby use (although I've always found it pretty useless due to uneven thickness and dust), however on pre-coated boards you can easily get 0.01" clean track/space with contact exposure using an uncollimated UV tube, so resolution is probably useful for a "200,000nm technology node"!
Commercial PCB places use various photoimageable resists and silkscreen inks, so maybe there could be something useful out there - presumably acid-resistance is the main criterion - if none of the standard resists are suitable, I wonder if some sort of 2-stage process might be an option - use a PCB resist to coat something that is then acid-resistant.

Another thought for possible patterning methods - lasers....? (a bit of a hobby of mine!) you can get some serious power at blue from blu-ray writer diodes these days, and the optics and precise positioning actuators are right there in the drive you just took apart to pull out the laser.
Or maybe a q-switched nd-yag for ultra-short-pulse 'blasting'?

Jeri Ellsworth

Mon Mar 15 2010, 12:59AM

Registered Member #2737 Joined: Sat Mar 13 2010, 07:34AM
Location:
Posts: 9

mikeselectricstuff - I've played with several PCB photoresists, but had pretty bad results. I don't think it was the quality of the resist, but my sloppy laser printer contact method. Someday I will revisit this and make better tools for registration and exposure. The first lithography at Fairchild was done with film camera optics, so it should be possible to get decent results at home.

I've successfully made PN junctions with my 60watt laser cutter and the spin on dopant. Normally the IR will pass directly through the wafer, but the dopant absorbs the IR and heats rapidly. I think this looks promising for direct write predeposition. Doped silicon still seems to pass most of the IR, so a drive in step will be needed.

The guys at Emulsitone tell me there is research in Rapid Thermal Procssing (RTP) with their product. RTP uses high intensity lamps to rapidly heat silicon.

Silicon is an amazing material. I've pulled cherry red pieces out of the furnace with cold pliers, dropped it in water when very hot and heated slices of it unevenly, but it rarely shatters.

I've experimented with heating silicon on a graphite inside of a microwave oven and got it so hot it glowed bright orange.

-Jer

IntraWinding

Mon Mar 15 2010, 01:14AM

Registered Member #2261 Joined: Mon Aug 03 2009, 01:19AM
Location: London, UK
Posts: 581

This type of PCB photoresist should be better than the spray on stuff for fine detail, but I have no idea if it's any use DIY IC chemistries?

Has anyone investigated whether amateur holographers use something suitable, maybe dichromated gelatin, or they might have their own version of the polymer stuff used for semiconductors?

mikeselectricstuff

Mon Mar 15 2010, 02:01AM

Registered Member #311 Joined: Sun Mar 12 2006, 08:28PM
Location:
Posts: 253

Jeri Ellsworth wrote ...

mikeselectricstuff - I've played with several PCB photoresists, but had pretty bad results. I don't think it was the quality of the resist, but my sloppy laser printer contact method. Someday I will revisit this and make better tools for registration and exposure. The first lithography at Fairchild was done with film camera optics, so it should be possible to get decent results at home.

For laser PCB work, the key is to print on tracing paper - this gives good toner adhesion, and as long as you use thick stuff, it doesnt crinkle in the fuser. The lack of transparancy isn't a big deal as long as the print side is in contact with the resist.
For higher resolution, in the past I've got typesetter films done at a local print service place - these are super-high resolution (2400-9600dpi I vaguely recall) and mega contrasty. This was a few years ago and I don't know to what extent direct digital methods may have replaced film, & hence how commonly available this service is nowadays.

Something else I just remembered - as well as PCBs, similar methods are used to photochemical etch all sorts of metals (e.g. Tecan), so there may be some resists for that market that could be useful.
Out of curiosity, what sort of geometry sizes are you currently working with?

Jeri Ellsworth

Mon Mar 15 2010, 03:05AM

Registered Member #2737 Joined: Sat Mar 13 2010, 07:34AM
Location:
Posts: 9

Mikes- I worried for the longest time about feature size, because I was reading modern text book about devices and the constant work to shrink transistors. My mentor told me not to worry about size and that even very large channel lengths will work.

Most of the channel length and width are about 3-4mm and the source and drain diffusion are ~6mm square, so it's easy to open contact holes in the oxide.

-Jer

first previous 2 3 4 5 next

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob