A question for chemists

[ffutures]

One of the nuisances of being an educational technician is the way that many products aimed purely at schools have incredibly high prices. Case in point is a kit of chemicals used to simulate blood group testing which costs 40-odd pounds for a total of about 100ml of various solutions, all of which I suspect are fairly basic compounds. What you get is vials of "antibody A", "Antibody B", and (in the deluxe kit that costs much more) "anti-Rh", some samples of fake blood, plus a few mixing sticks and bits of laminated card to mix them on. I suspect I could make the lot for a pound or two if I knew what the chemicals were. The type A blood clots when mixed with Antibody A, type B with antibody B, AB with both, O with neither, and so forth.

Needless to say the chemicals are not identified, and in any case I would prefer to develop something independently and publish it on the web without having to worry about the manufacturer saying that I've ripped off their product.

So what I need is three distinct blood chemicals, each producing a precipitate when mixed with the appropriate "antibody", but not precipitating when mixed with each other or the other "antibodies". Harmlessness would be a plus, and they need to be miscable with food dye and water to look vaguely like blood. The precipates should look like little blobs of clotting if possible.

Any suggestions? Or anyone know of a web site I've missed that describes a suitable set of chemicals?

Comments

[parakkum.livejournal.com]

Okay, I have have a possible solution, with the caveat that I have been doing molecular biology for the last five years and haven't had an inorganic chemistry course since 1998.

Calcium chloride, magnesium chloride and manganese chloride are all fairly soluble and reasonably nontoxic (at the very least, we played with them in my own high school chemistry class, way back when). However, the carbonates and hydroxides of each of these metals are very insoluble:

MgCO3 has a Ksp of 6.82x10^-6, for example, and that's the most soluble one.

So at least one of your "blood types" could be one of these chlorides (CaCl2, MgCl2, MnCl2), and the "antibody" for that chloride can be NaCO3 or NaOH (probably better to go wih NaCO3).

Unfortunately, NaCO3 and NaOH both cause precipitates in all these compounds, so you'd need to think of another "antibody-antigen" mix for your work. But hopefully that's a start. Maybe I can find a more thorough table of solubility products from which to work.

[ffutures.livejournal.com]

Thanks - that is indeed a start.