[NewCandle] crystallization weight change transients

[Keith Nagel]

Hehehe....

Good find Nick, that site's nice. But I just can't resist doubling up on it,
I'm just that kind of an a-hole (grin).

Ready?

Let's say that the supersaturated thing is a bit of a red
herring, and in fact, the key is just a fast phase transition.
Let's also assume that the more dense, the better.

So here's the idea. Melt some woods metal, you know the stuff,
it's quite common for modeling purposes. Put it on the scale,
and let it cool. Should be quite revealing. A chill plate
would make it go faster, if speed is important. Put the whole
thing in a thermos if thermals are a problem. 

How about it?

K.



-----Original Message-----
From: newcandle-bounces at ipdiscover.com
[mailto:newcandle-bounces at ipdiscover.com]On Behalf Of Nick Reiter
Sent: Thursday, October 05, 2006 4:23 PM
To: New energy for the new world.
Subject: Re: [NewCandle] crystallization weight change transients


Update for 10-05

Back on Saturday, I had been talking with Sam Faile,
who pulled out one of his old chemistry texts and
looked up heats of solution as we chatted.  Sam scared
up a potentially good candidate for an endothermic
crystallizer - aluminum sulfate hydrate... otherwise
known as alum.  I bought some of the powdered stuff at
the grocery store, but I think it's cut with some
manner of white foofoo powder to prevent caking.  I
couldn't get a good solution or crystals from it.  I
did order some pure Al2(SO4)3.18H2O though, and will
give that a whirl soon.

Tried a couple of other chemicals - thorium nitrate,
ammonium acetate, and thiourea.  All pretty slow
crystallizers and very fussy.  Thiourea makes some
really damned COOOLLLL crystals, though.  Lookin' like
da back of a stegosaurus they was.  Weight changes
either null or indeterminately small.

So then yesterday, I'm seeking crystal lore online,
and I find this web page from a German
krystallmeister!

http://www.crystalgrowing.com/recipes/lightspeed/lightspeed.htm

Ah, now I see!  Sodium thiosulfate / hyposulphite
pentahydrate can simply be heated past about 48C and
it goes to a melt.  No extra water is necessary...
maximum booster go-go warp speed crystals upon
seeding!  He has some video that is wonderful.

So I tried this earlier today, made a crystallizing
run, and got about the same results as I had last
week.  hmph.  What was slowing it down?  Aha!  The
heat itself evolved during crystallization slows up
the crystallizing past a couple of minutes, as the
temperature climbs above about 40C!  So I re-melted,
cooled, then chilled about an extra 7 or 8 degrees C
in the refrigerator before seeding.

This time, the speed with which the crystals grew was
nigh surreal... literally millimeters per second, much
like the videos in Udo's web site.  The drop in weight
was very rapid too - in fact during the time it took
me to stopper the flask, and let the balance pan
settle in upon starting, I probably lost a milligram
or so even there.  Total loss was about 6 mg at least.
 

Now in all of this there was an interesting clue.  The
crystallizing had been so rapid, that the weight loss
had maximized and leveled off... even though the
evolved heat was still being transfered to the glass
wall of the flask!  The temperature maxima didn't show
up until a good 3 minutes or so after the peak weight
loss.  There was probably another good 10 to 12 degree
C rise after the weight drop had leveled off with
maximum crystal volume being achieved.

This might be the ticket for analysis.  A set of runs
to examine the typical weight change profile, then
another set to look at the rate of heat exchange
through the flask.  So far this is hinting that the
weight loss may be be outstripping the flask warming,
speed wise.  Which is very interesting...

More to come...

NR

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com 

_______________________________________________
NewCandle mailing list
NewCandle at ipdiscover.com
http://ipdiscover.com/mailman/listinfo/newcandle_ipdiscover.com