OT: homemade microfluidics modules

I'm not a big fan of Macor- it likes to chip and flake etc., but physicists seem to love it. At high temperature in a vacuum, maybe a good choice. It can be lapped flat & polished. It's a bit softer than glass (4.5 vs. 5.5)-- carbide tools are okay, but might not last long in that small a size.

Perhaps the advantage of glass in this application is that you can easily source glass sheets that are very smooth and flat and which also can be anodically bonded, so laminating two pieces together and getting a seal is relatively easy.

Maybe a diamond-coated microfine grain solid carbide end mill to get down to those dimensions. Very, very high RPM, very slow feed and lots of coolant, I would guess, but have not done it.

Best regards, Spehro Pefhany

--
"it's the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

Most microfluidics devices are fabricated on microscope slides by etching, not machining, so they have good optical properties for detection and can just use a second slide with through holes as the cover. The translume (interesting web site, thanks) ones sure look that way. I think you would have better luck cutting with a laser than machining. A CO2 laser in the

10-50W range set up for marking should be able to make channels in glass pretty well, and CNC cutting/marking machines with tables in the 12"x12" range are in the $2-4K range new.

Hi,

I am thinking about how to make a microfluidic module using a 3axis benchtop CNC milling machine and small milling bits or diamond grinding bits perhaps. Commercial microfluidic modules seems to be made out of glass like these ones:

I think glass might be too difficult to machine or grind with a milling machine, the required cutting width and depth of the microchannels would be from 0.1mm to 0.3mm. What would a good material be to use? I would like to use a hard material that is fairly chemically stable. The idea is to take two plates, and cut the micro channels into one and then laminate them together.

I found this machinable ceramic/glass which might work:

Thanks for any ideas!

cheers, Jamie

...

ld

.au/MachinGlass.shtml

At a meeting this summer I met a couple of guys from vanderbilt who were making microchannels out of some polymer. They would machine a 'negative' (on stainless?) polyermize a film on top of that, then peal it off and stick it to another flat piece of the same material. I think they were part of John Wikswo's group (empire) at vanderbilt.. but poking around the website I couldn't find any details. If this sounds at all useful I could see if I can dig up their names and you can get more details.

George H.

Is the microscope slide glass opaque to the CO2 IR wavelength?

Almost all oxides are opaque past about 5 um, including soda lime glass. See e.g.

.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

yes

work:

Hi,

Ya that has the advantage of allowing for smaller channels than the cutter diameter used too. I was thinking of that idea already but if you can send me their names that would be cool too! I made a test one yesterday with a 0.38mm diameter end mill in aluminum and it turned out quite well. I like the CO2 laser and microscope slide idea best though, I think I should mount a 50watt CO2 laser to my milling machine!

cheers, Jamie

Yes, glass and quartz are opaque to 10.6 um light, which is why a co2 laser does such a good job of marking, drilling and cutting glass. It's also the cheapest in terms of watts of laser light out per unit of electricity in, so best bang for the buck. Biggest problem is that the minimum spot size (gaussian tem00 beam) is equal to wavelength * f# of the focusing optic, so for something reasonable (lots of assumptions just went into that word :-)) like f2 or f3 the minimum spot size is 20-30 um which isn't all that much smaller than the channels you want to make. If you want tiny features go with an excimer laser like ArF at 193 nm in the uv so the same f3 focusing gets you a 0.58 um spot (but the beam quality is usually not that great with low end lasers, so 1 um is realistic for quick and dirty setups), where glass is also opaque and it does a fine job of drilling and cutting. I once drilled a couple of 200 um diameter gas ports through the side of a precision glass capillary 1 mm ID 6 mm OD where the inside bore was coated with gold to make a lightpipe for an IR spectrometer, using an ArF laser. Impressed the prof who asked if I could do it :-). Usually they made end caps with the flow connections that also held sodium chloride windows but he wanted to minimize all the dead volume so I drilled the holes and they glued the windows directly to the ends of the lightpipe. (It's always good to do favors for profs who will be on your dissertation committee :-). Ah, the good old days of grad school.)

----- Regards, Carl Ijames "Spehro Pefhany" wrote in message news: snipped-for-privacy@4ax.com...

On Mon, 10 Sep 2012 23:41:08 -0400, "Carl Ijames" wrote:

Is the microscope slide glass opaque to the CO2 IR wavelength?

Hi,

What would be the best way to bond glass to glass or glass to metal? I checked the anodic bonding and also glass frit bonding but not sure how the glass frit bonding will affect optical properties. For prototyping I guess it is probably simpler to use a thin layer gasket and compress it to prevent leakage from the microchannels. Any good ideas on the best ways to prevent leakage from the microchannels?

cheers, Jamie

Frit is opaque, so your best bet may be to put a thin Teflon gasket on the flat side, and not press too hard.

You can do good glass-to-glass bonding using dilute KOH on both sides and then squeezing them together--google for "hydroxyl bonding". (I don't know whether the patent on that technique has expired yet.) The surfaces have to be clean as a whistle--use collodion or Opticlean polymer to get rid of the particles, and put the KOH through an ultrafilter (0.015 um). Don't try squeegeeing it off, just press the surfaces together and work the solution out from the centre to the edges.

You can get syringe-mounting ultrafilters inexpensively, and they're pretty resistant to aqueous solutions.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058

hobbs at electrooptical dot net
http://electrooptical.net

g

of

=3DC...

g

ld

would

en

com.au/MachinGlass.shtml

OK I tried the 'reply to author' feature in google groups. If you don't get his name and email then drop me an email and I'll send it to you (gherold at teachspin dot com) (goolge groups mungles your email so I can't send it to you directly.)

George H.

...

ld

.au/MachinGlass.shtml

idics.asp

Cheers, James Arthur

work:

Awesome.

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510
845-480-2058

hobbs at electrooptical dot net
http://electrooptical.net

Yep, neat. It's right up there with making graphene with DVD drives.

supercapacitor.asp

-- Cheers, James Arthur

As Phil said, frits are opaque. It has been awhile since I've read up on this, but I think in the early days they just pressed on a second slide or cover slip after cleaning everything. One memory says someone was spin coating a very thin film of polydimethylsiloxane on the cover to act as glue and seal. Most of the early stuff was published in the journal "Analytical Chemistry" - go to

and surf to it and search. You can read the abstracts for free, then either pay by the article or go to a college library and read them there.

Hi,

What would be the best way to bond glass to glass or glass to metal? I checked the anodic bonding and also glass frit bonding but not sure how the glass frit bonding will affect optical properties. For prototyping I guess it is probably simpler to use a thin layer gasket and compress it to prevent leakage from the microchannels. Any good ideas on the best ways to prevent leakage from the microchannels?

cheers, Jamie

C&Category_Code=3DMC2

My first idea was to go to chemical etching, with resist and = dimensionally stable master films. But you have the mill platform and lasers are not all that expensive. 100 to 300 micron work should not be all that bad, the semiconductor industry was doing 3 micron 50 years ago.

?-)

grinding

of

...

milling

would

would

then

work:

luidics.asp

Bloody hell right. This one looks to be a possible world changer. =20

YMMV

?-)

rn-supercapacitor.asp

Another technology changer, a little harder to fully commercialize perhaps.

YMMV

?-)

Wow!

Could be a fairly large thing. Might be hyperbolic, but making graphene tech. widely accessible might be as big as polymerase chain reaction was to genetics.

Replicability ( dCost/dUnit ) is the key.

-- Les Cargill