I was watching a YT video on the recrystallization of a substance to purify it. One of the commenters said that stirring during crystallization
produces a purer product. I can’t copy from the YT video comments, so I’ll repeat here some of what he said.
“….stirring continuously is necessary in the absolute majority of cases to get a purer end product. Large crystals are a sign of decent purity,
but they can actually trap mother liquor and make it harder to wash out impurities.”
He goes on to say that the amount of stirring versus resultant crystal size is usually worked out experimentally, as smaller crystals catch too much
mother liquor on their surface and have to be washed more during cleaning.
Is this right? This is the first I’ve ever heard of stirring during recrystallization to enhance purity of the final product. Later he claims that
in industry that purities can be improved from 99% to 99.5% simply by improving the stirring/crystallization ratio, and that stirring is the default
method, versus non stirring. He says that ‘not stirring’ is almost never done. If this is true, then why do we not do this in the lab? At the
very least, why have I not ever seen this mentioned before?Texium - 13-3-2024 at 06:49
It sounds like that commenter is referring to large industrial scale processes. In those cases, smaller, more uniform crystals are also more desirable
from a product handling perspective, so it may not be merely a purity consideration.
On a lab scale, especially when you only have a gram or two, there’s typically no need to stir and your crystals are not going to be able to grow
large enough to encounter the problems they describe. But ultimately, you have to see for yourself what the crystallization behavior of your substance
is. If it slowly forms large, clear crystals on cooling, it’s unlikely to entrain mother liquor and should be quite pure. If it’s something that
is liable to entrain mother liquor, you could stir it during crystallization, or you could just crush up the crystals and wash with a little more cold
solvent. If you’re recrystallizing >100 g of something that crystallizes rapidly as the solution cools, though, it might be helpful to stir the
solution so that you get more uniform crystals and it doesn’t all just clump up at the bottom of the container.
There’s also the consideration of balancing purity and yield. Do you NEED 99.5%, or is 99% good enough? The smaller your crystals are, the more
product you’ll lose when you wash them, since the surface area is higher and more will dissolve. Generally I’d rather get, say, an 80% yield at
99% purity than a 50% yield at 99.5%.Sulaiman - 13-3-2024 at 07:00
When the desired substance crystallises out
the ratio of impurities to product in the immediately surrounding liquor will increase,
stirring will reduce this effect.
I think charley1957 - 13-3-2024 at 11:25
Well he did mention industrial a couple of times, but made it seem that it applied in the lab just as well, he said that people almost never stir
during recrystallization in the lab. You make some valid points about yield versus purity. I agree. As hobbyists, often that 99% is more than
enough purity for what we are doing.Texium - 13-3-2024 at 12:41
Well he did mention industrial a couple of times, but made it seem that it applied in the lab just as well, he said that people almost never stir
during recrystallization in the lab.
I've noticed that industry-minded people frequently assume that their
methods are automatically superior to those employed in small-scale research. Not saying that's definitely the case here, but I'd take any absolute
statements with a grain of salt.Precipitates - 13-3-2024 at 18:58
Or a crystal of salt in this case! Sorry couldn't help myself!! j_sum1 - 14-3-2024 at 18:19
Or a crystal of salt in this case! Sorry couldn't help myself!!
Wins the username-comment matching game.BromicAcid - 14-3-2024 at 18:59
As someone who has done quite a few industrial crystalizations, one of the benefits of stirring at scale is that your heat exchange is completely
different in large vessels vs small vessels as a ratio of surface area to volume so without stirring to allow something to cool 'naturally' takes
forever. If you look at a cooling curve the majority of the heat is lost quickly (relatively) then tapers off. In industry with a jacketed vessel
you watch for initiation of the crystallization by a baffle temperature and then stop the cooling and hold until crystallization is no longer actively
occurring then cool at a set rate to maintain uniform crystal formation.
With these sorts of controls in place you can produce very pure crystals that are of a uniform size and are easy to transfer via process hoses / pumps
as Texium pointed out and you can do so in a reasonable amount of time. I think that both processes can make crystals of a high purity but if you're
trying to grow crystals for XRD then you definitely want larger crystals.
[Edited on 3/15/2024 by BromicAcid]SnailsAttack - 15-3-2024 at 16:18
“Large crystals are a sign of decent purity, but they can actually trap mother liquor and make it harder to wash out impurities.”
I don't see why larger crystals would entrain more solution than smaller ones. Any entrained solution will result in visible defects, for example the
X's or other orthorhombic shapes that sometimes manifest in sodium chloride cubes. These patterns are comprised of water trapped between the grain
boundaries, and are visible because its index of refraction is less than that of the salt crystal.
...smaller crystals catch too much mother liquor on their surface and have to be washed more during cleaning.
Yes. Large crystals have a much lower surface area to volume ratio, which means that less mother liquor adheres to them and less
mass is lost on washing.Sulaiman - 16-3-2024 at 05:59
I've noticed that 'crashing out'/rapid crystallisation is very likely to trap/entrain liquid
in my case especially with
Potassium nitrate that rapidly crystallises to look a bit like Superman's ice fortress,
Salycylic acid forming hollow square needlesAlkoholvergiftung - 16-3-2024 at 06:31
The Saltpeter makers done the first 2 Crystallisations normal the third under stiring so the Potassiumnitrate falls out as an very fine Powder. So no
dirt inside the Crystalls. Sulaiman - 16-3-2024 at 07:17
Crystallization/recrystallization has been discussed a few times here,
due to my early experience (mostly just crystal growing for it's own sake)
I was CERTAIN that bigger = purer
Now I realise that it is certainly not a simple topic,
which is interesting for me, and reminds me how little I actually KNOW.
My apologies to anyone that I argued with.
[Edited on 16-3-2024 by Sulaiman]zed - 20-3-2024 at 17:31
Ummm. Back when we rode around on dinosaurs, I once crystalized a nitro-olefin via rapid stirring.
My comrades hated this. Very pure crystals were perhaps produced, but these fine crystals had a huge surface area and they were coated by a gooey
crud. Rinsing this crud off of fine crystals, insured a lot of product loss.
So yes, small, pure crystals are produced by rapid crystallization, but this may/or may not prove desirable, in actual practice. Very large crystals,
that can be easily rinsed or even "wiped off", may in some cases be more desirable.
