microelectrofluidic module idea

Once you start milling or laser cutting a glass surface, you'll create a lot of damage unless you use something fancy like a femtosecond or excimer laser. The damage and the accompanying swarf will make hydroxyl bonding harder even after cleaning. (I don't know how much harder.)

Since you'll need to remove the silvering from the parts you want to bond, litho followed by HF etching would be my first choice for cutting the channels. (Not if it's being done in the basement, of course.)

HF is nasty, nasty stuff. Besides the vapour eating your lungs, if you spill it on your finger it can migrate through your skin and dissolve the bones inside. Plus it has equally nasty longer term health effects. So really really don't use it in your basement.

Anodic bonding is another approach.

You can also do chemical silvering, as is commonly done by amateur telescope makers. It would be fun to see if you can make circuits on glass that way, using a liftoff process with PC board type litho.

The silver would need to be pretty thin, because it would sit above the glass surface, and so would prevent the substrate and cover from making intimate contact, which of course they need to if you're going to bond them together. The glass surface will bend a bit, so contact will be restored some distance away from the metal lines. The distance over which that occurs goes like the square root of the layer thickness, so it can be surprisingly far unless the layer and the cover glass are both very thin.

Very thin film wiring is vulnerable to scratches and surface irregularities. (I had a very difficult time trying to make the first sensor films for Footprints by etching nickel-coated PVDF films, but that was mostly because when the film stretched even a little, the wiring cracked. Carbon ink was a biiig improvment.)

Getting wires onto the silver might be an issue as well, since it almost certainly won't stick as well as Ti:Al.

Process development is an absorbing task. (It also absorbs time, money, enthusiasm, management patience, ....)

Cheers

Phil Hobbs

Odd. The surface quality of chip wafers are ground pretty flat and smooth before they start building chips.

A place called Ball "Electronics" experiments (or actually makes) high end, hybrid circuits and chips on spherical shaped substrates.

LCDs are circuits so thin we cannot see them until the liquid gets 'oriented'.

We should make circuits the size of the etchings we make on the aprons of diamonds to ID them by.

I am sure that RFID technology should be what drives this.

Won't be long, we'll all be getting wifi implants.

Found a possibly useful paper on bonding soda lime glass with dilute HF:

Cheers

Phil Hobbs

Back when I was graduate student in chemistry, I got exposed to the fluorine lab's safety kits, which were basically a bunch of big hypodermic syringes loaded with a calcium compound that could be injected under the skins to convert any fluoride present to extremely insoluble calcium fluoride.

The students and staff involved weren't supposed to self-administer, but to take the kit over to the university teaching hospital across the road and get the doctors there to do it.

The kit included colour reprints of the relevant article in the Lancet, with gruesome images of what happened if the doctors were dilatory about making the injections; fluoride poisoning isn't all that common in the general population, and the doctors needed to be motivated to get on with the injections, whence the - then expensive - colour reprints.

Thanks for all the info, I don't think I am ever interested in using HF luckily!

cheers, Jamie