Petit Homme - 1-9-2025 at 17:29
I have modest chemistry understanding and experience, and know little about Selenium (except that it is toxic in anything but minute amounts and seems
to have a more peculiar chemistry than more common elements). I have a piece of Selenium (sold as 5N Elemental selenium from China, has a
vitreous-metallic aspect) which I'd like to further purify to 6.5N by the most simple method to carry out with little equipment. The goal being to end
up with a soluble form that very pure (trace metal basis), non-toxic in minute amounts (i.e. RDA range), little to non-oxidizing and stable with time.
I'm not sure if it's feasible but I was thinking about turning the piece I have into SeO2 with limiting amounts of H2O2 in an aqueous solution
(distilled water), and then recursively recrystallizing it between 50C and 5C to purify it, and ultimately oxidize it further with more H2O2 to H2SeO3
for long term storage. In theory (having never done it), it seems simple enough. Is this feasible? Does it sound like a good way to achieve my
purpose?
A good soul pointed out to me that the glass would leach trace amounts of alkali into solution (Na, Mg, K, Ca, Al, B, I assume?).
I'm most interested in getting rid of trace amounts of Pb, Te, Ni, Cd, Hg, As, Tl, Co, V, Cr, etc. Small traces of Na, Mg, K, Ca, Al, B that would
leach from the glass are less of a problem. I'm interested to know if the route I think is possible indeed would work (and be a good enough reasonable
mean to achieve that purpose with a method that is not an unnecessarily unsafe or complicated way to carry out the task).
j_sum1 - 1-9-2025 at 20:16
5N is pretty high.
You are not going to get anything purer than that in a home lab.
Unless you have a clean room, you will knock it back to 3N if you look at it funny.
Petit Homme - 2-9-2025 at 01:07
I believe it to be very much feasible to get >5N purity H2SeO3 with a fairly simple setup of hobby chemist equipment.
Maybe not in the sense that it would be >99.999% H2SeO3 (which is not my goal)
but in the sense of arriving at a product containing far below 10ppm of Pb, Te, Ni, Cd, Hg, As, Tl, Co, V, Cr, etc. as I tried to explain it in my
first post above.
Anyone feeling an urge to write any of all the ways this is not possible, express that urge elsewhere if you please. There's an infinity of ways of
not doing things, please keep that infinity away from me, as I have zero interest in the infinity of ways of not reaching a result I want to reach in
life.
I want to be taught and/or helped through towards a reasonably simple and efficient way (for there likely are plenty of ways to achieve this) to get
to the result I want (i.e. starting from 5N Se and ending up with H2SeO3 containing as reasonably far as possible beneath 10ppm of the sum of trace
metal impurities, more specifically, of the aforementioned trace metals).
I proposed one method that seems possible to my uneducated mind.
If this method won't work, please educate me and tell me why.
If there is something that has to be improved or changed with this method, please educate me and tell me.
If there is a better / more simple / more efficient alternative method that is likely to achieve this purpose in a better way, please share your light
and outline such a method here.
RU_KLO - 2-9-2025 at 02:21
How willl you confirm that you have less than xxx ppm of several impurities? Meassuring such low levels from some metales could be very difficult.
There are specific reagents that detect 1 ppm or less but checking for all possible impurities it's a lot of speciallized reagents.
Or pay for profesional spectroscopy/spectrometry * . There will you have your purity result.
(You could send your starting material and final material to compare)
*Inductively coupled plasma atomic emission spectroscopy
*Glow Discharge Mass Spectrometry (GDMS)
For those who dont know:
5N Selenium: means 99.999% pure selenium (five nines after the decimal point).
6N Selenium: means 99.9999% pure selenium (six nines after the decimal point).
7N Selenium: means 99.99999% pure selenium (seven nines after the decimal point)
[Edited on 2-9-2025 by RU_KLO]