Sciencemadness Discussion Board

Alkali ionic liquids

chornedsnorkack - 10-4-2024 at 01:16

A lot of classic reaction instructions start like "Fuse xx with sodium hydroxide", "Fuse xx with alkali", "Fuse xx with nitre", "Fuse xx with potassium chlorate..."
How miscible are various alkali ionic liquids? How much do multiple ionic pairs lower the melting points against precipitating some components?
Single components:
Alkali:
LiOH - melts at +462. In cold, poorly soluble for an alkali - 128 g/l@20C
NaOH - +323. 1000 g/l@25C. Notoriously viscous compared to all other aqueous alkali
KOH - +410
RbOH - +382
CsOH - +272

So how are the melting points for mixtures of alkali? (These would enable reactions at lower temperatures/milder conditions)
For the alkaline metals themselves, pairwise only, see himikatus.ru
Metal Li is liquid immiscible in all other alkali metals, with appreciable liquid solubility only in Na. Critical point at 303 C, eutectic freezing point lowering on Na side from 98 to 92 C. The other four alkali metals are miscible as liquids.
Na separates, sometimes as intermetallics in solid phase:
Na-K -12,6 C
Na-Rb -4,5 C
Cs-Na -31,8 C
The 3 heavy alkali metals freeze as solid solutions:
K-Rb +34 C (lowering from +39,5 of pure Rb)
K-Cs -38 C (lowering from +28 of pure Cs)
Cs-Rb +10 C
Now, himikatus has just double systems, not triple. It is said that Na-K-Cs freezes at -78 C, which is not lowered by addition of Rb
What are multiple alkali metal salt eutectics like in terms of their melting point and reactivity?

chornedsnorkack - 10-4-2024 at 01:39

https://www.google.com/url?sa=t&rct=j&q=&esrc=s&...

some data (which have different anions for same cation)

bnull - 10-4-2024 at 10:16

When you say " alkali ionic liquids", are you talking about molten alkaline hydroxides or aqueous solutions of alkaline hydroxides? The first paragraph seems to use both interchangeably.

Bedlasky - 10-4-2024 at 10:19

LiCl-KCl eutectic is pretty usefull, it melts only at 352 °C.

https://sci-hub.se/https://link.springer.com/article/10.1134...

DraconicAcid - 10-4-2024 at 18:58

If you want to use molten alkali hydroxides, be very aware that they are caustic as hell. Molten sodium hydroxide will rip through glass and ceramics with ease. IIRC (I wrote a term paper on such beasts a few decades ago), it will not attack Teflon if it's anhydrous (or was it under a certain temperature near its melting point?), but will destroy it if damp (or hot?).

Nickel crucibles should be okay.

chornedsnorkack - 10-4-2024 at 22:52

Quote: Originally posted by bnull  
When you say " alkali ionic liquids", are you talking about molten alkaline hydroxides or aqueous solutions of alkaline hydroxides? The first paragraph seems to use both interchangeably.

Mainly the molten salts. Though some remarks about aqueous solutions seemed appropriate.
Alkali metal salts are often talked of as inert spectator counterions whose identity is unimportant to reaction - no matter if you have Na or K. So I thought of for one comparing the various alkali metals if you want to pick one. For another, why even pick one? Eutectic melt of different molten salts has other properties the salts separately do not have: lower melting point allows handling the ionic liquid at lower temperatures and thus milder conditions than any single component.
And the aqueous outstanding properties are relevant in case of workup step. Say you want to separate your reaction product, such as Li2CrO4, Li2MnO4 or Li2FeO4, from the frozen reaction environment, that before the reaction was LiOH-NaOH eutectic. Then the low solubility of LiOH will matter - it is low to start with and in a saturated solution of common ion NaOH likely to be even lower. High viscosity of NaOH would also matter - it would slow down separating NaOH and something mother liquor from the various solid though highly soluble residues, correct?

bnull - 11-4-2024 at 03:10

Quote: Originally posted by chornedsnorkack  

Mainly the molten salts. Though some remarks about aqueous solutions seemed appropriate.

One thing I learned in Physics (or was it Math?) is that it's bad practice to use the same symbol for different things in the same situation. Why? Because a symbol is a (sort of) shorthand for ideas. If you use the same symbol for many things at the same time, you'll have to explain what the symbol is representing every time it appears, and that defeats the purpose of using the symbol as a shorthand (I have used characters from the Latin, Greek, and Cyrillic alphabets, and was this close to using Kanji). Let's reserve "alkali ionic liquids" (or "molten alkali") only for the molten hydroxides and salts, either mixed or not, and use "aqueous solutions" for the, um, aqueous solutions.

Quote: Originally posted by chornedsnorkack  
Alkali metal salts are often talked of as inert spectator counterions whose identity is unimportant to reaction - no matter if you have Na or K. So I thought of for one comparing the various alkali metals if you want to pick one. For another, why even pick one?

It all boils down to the conditions of the reaction and properties of the product. Some reactions are more forgiving than the others. If the reaction requires large quantities of a hydroxide with isopropanol as solvent, for example, only KOH will do; LiOH is insoluble and NaOH is slightly soluble. If it is just a pinch of hydroxide, it's NaOH or KOH.

If, on the other hand, you're making perchlorate from chloride, you use sodium chloride because it's cheaper. Potassium chloride is used only after all chlorate has been converted to perchlorate. Potassium perchlorate is much less soluble than sodium perchlorate.

You see, when the choice of alkaline metal is important, it's done in a case-by-case basis.

Quote: Originally posted by chornedsnorkack  
Eutectic melt of different molten salts has other properties the salts separately do not have: lower melting point allows handling the ionic liquid at lower temperatures and thus milder conditions than any single component.

And the aqueous outstanding properties are relevant in case of workup step. Say you want to separate your reaction product, such as Li2CrO4, Li2MnO4 or Li2FeO4, from the frozen reaction environment, that before the reaction was LiOH-NaOH eutectic. Then the low solubility of LiOH will matter - it is low to start with and in a saturated solution of common ion NaOH likely to be even lower. High viscosity of NaOH would also matter - it would slow down separating NaOH and something mother liquor from the various solid though highly soluble residues, correct?

I'm not exactly sure of what you want. Correct me if I'm mistaken. You want to know the properties of molten alkali in order to choose a certain composition for a reaction so you can isolate the product more easily afterwards.

chornedsnorkack - 11-4-2024 at 09:25

Quote: Originally posted by Bedlasky  
LiCl-KCl eutectic is pretty usefull, it melts only at 352 °C.

https://sci-hub.se/https://link.springer.com/article/10.1134...

Indeed.
If you have n components that cannot react with each other then you generally have 2n-1 mixtures - the row n of Pascal triangle, except the initial 1. For example, for 5 components (all the alkali metals except the evanescent francium), you get 25-1=31 mixtures. The substracted initial 1 is "0 component mixture", then 5 pure metals ("1 component mixtures"), 10 pairwise mixtures, also 10 ternary mixtures ("2 of 5 left out"), also 5 quaternary mixtures ("1 of 5 left out") and 1 mixture of all 5 components.
What helps if you have useful generalizations to simplify. In case of the metals, note that Li is not miscible with any of the other 4, and has a limited solubility in Na alone. Since Li is essentially insoluble in the 3 heavy metals and all the 3 Na eutectics are eutectic near Na-poor end, the solubility of Li in Na-heavy metal eutectics is likely also poor, and has little effect on the melt. Leaving just 4 elements to consider (15 combinations).
Also note that the 3 heavy metals are all miscible in solid but with melting point depression. Weakest effect for K-Rb. This can give some clues as to what to expect of the 4 ternary mixes.

Now to halides...
Of the 5 chlorides, the melting points of single components are:
  1. LiCl - 607
  2. NaCl - 801
  3. KCl - 774
  4. RbCl - 718
  5. CsCl - 646

Of the 10 pairwise eutectics, there was just the mention:
2) LiCl/KCl - 352
Note the melting point depression - 255 degrees below the lower melting component (LiCl)
So what could be the other alkali chloride systems (remember, I have just 6 of the 31 here!)?
It is relevant. The first I mentioned was hydroxides, but as mentioned, molten alkali are highly aggressive. Chlorides would be relatively inert both in acid-base and redox respect... except that the 1 component chlorides have the nasty tendency to be refractory and therefore hot. Like Na: NaOH 323 C, NaCl 801 C
If LiCl/KCl gets the temperature to 352 C, could choosing a different pair or a ternary get even cooler?

Quote:

Some reactions are more forgiving than the others. If the reaction requires large quantities of a hydroxide with isopropanol as solvent, for example, only KOH will do; LiOH is insoluble and NaOH is slightly soluble. If it is just a pinch of hydroxide, it's NaOH or KOH.

Here you´re forgetting the "obvious" alternatives of RbOH and CsOH.
I was inspired by a recent post about an (aqueous) reaction which is known to be possible with Cs but impossible with K. Since Cs has some interesting chemistry not shared with K, it is important to remember checking up Rb.

Quote:

You see, when the choice of alkaline metal is important, it's done in a case-by-case basis.

Here I think that systematically organized data and filling the gaps would make sense. Just to make sure you don´t overlook "obvious" alternatives.
One example of the issues with hydroxides: single component hydroxides are hot. Over 320 C (CsOH actually has new corrected data over 320 too)... Seem useless for a lot of interesting and sensitive chemistry...
except NaOH/KOH binary alone already gives +170. A lot of organics that crack at +320 hold up to +170. And considering just Na and K gives you just 1 mixture out of the 26! Some of them are not as good (LiOH/NaOH and LiOH/KOH are at 220...225 C), but others (that is the Cs, Rb and multiples) I have not looked up.
If you have a systematic (simplified with useful insights) dataset for each of several steps, you could look for where the combinations with "no objection in this step" line up...

bnull - 11-4-2024 at 11:44

Quote: Originally posted by chornedsnorkack  
Here you´re forgetting the "obvious" alternatives of RbOH and CsOH.

They were not forgotten. I ignored them because (1) that was just an example and (2) the other three hydroxides are more accessible, affordable, and can perform the all tasks usually ascribed to generic "alkaline hydroxides". If, and only if, the reaction requires rubidium hydroxide or caesium hydroxide--or I want to use them because I have a ten pound pot almost full just taking up space and getting dusty, or for curiosity's sake--I'll use rubidium hydroxide or caesium hydroxide in the examples. It'll be either sodium hydroxide or potassium hydroxide, and lithium hydroxide when I'm in the mood. Again, caesium and rubidium hydroxides were deliberately left out, not overlooked.

Quote: Originally posted by chornedsnorkack  
Here I think that systematically organized data and filling the gaps would make sense.
[...]
If you have a systematic (simplified with useful insights) dataset for each of several steps, you could look for where the combinations with "no objection in this step" line up...
[...] but others (that is the Cs, Rb and multiples) I have not looked up.

Compile the data and use a flowchart to choose the mixtures. You can use a spreadhseet to do that. Fill the gaps by performing the experiments.

And for Chrissake, karl, be careful. Eutectic mixtures of alkaline hydroxides are still alkaline hydroxides. Melting at 320 °C or at 170 °C makes no difference when you're saponifying yourself.

bnull - 13-4-2024 at 15:15

(Edit: I was a little cheesed off the other day. Anyway.)

You can find more references on alkaline eutectic (which I keep writing as euctetic for no apparent reason) mixtures within these references:
  1. https://www.sciencedirect.com/science/article/abs/pii/004060...
  2. https://www.frontiersin.org/articles/10.3389/fenrg.2021.6661...
  3. https://patents.google.com/patent/US20050072837
  4. https://www.sciencedirect.com/science/article/abs/pii/002207...

The Journal of Physical and Chemical Reference Data seems to be the best place where you can find relevant data on what you want (https://pubs.aip.org/aip/jpr/issue/1/1).

By the way, there are more than 2n-1 mixtures. Some (maybe all; I haven't verified the hundreds of possibilities) pairings of alkaline compounds form more than one euctetic mixture. And there's more, sometimes the euctetic is not the best composition for a reaction. It's here that the "useful insights" enter.

And because there are hundreds, if not thousands, of alkaline inorganic compounds--and some "hard-boiled" organic salts--it seems advisable to start with the most common (Na, K, Li) metals, and expand the database to the others as need arises. That's one of the reasons I wrote
Quote:
You see, when the choice of alkaline metal is important, it's done in a case-by-case basis.


[Edited on 13-4-2024 by bnull]