Sciencemadness Discussion Board

Make or extract Pure Trisodium phosphate

Hilski - 29-9-2012 at 09:01

I'm attemping to duplicate the plating bath recipe from US patent 7235165. The constituents are 32 grams oxalic acid, 10 grams trisodium phosphate, and 4 grams ammonium sulfate in 1 quart of water.
The pure oxalic acid and ammonium sulfate are no problem, but all the TSP which is available to me contains 20% sodium sesquicarbonate, which ruins the plating bath.
H3PO3 + NaOH → NaH2PO3 + H2O, but no phosphate. But, I have plenty of of H3PO3, so I was wondering if anyone knew a method to synth Na3PO4 from H3PO3.
Also, I would be interested in hearing some ideas for a way to separate Na3PO4 from the sodium carbonate in the Savogran TSP product.

Thanks.

bahamuth - 29-9-2012 at 09:06

Add stoichiometric amounts of NaOH to H3PO4. Not harder than that..
If you need it "free" of acid or base, I suggest making, say 1 mole of it, add water to 1000 ml, figure out the pH of a 1 molar trisodium phosphate solution and adjust pH accordingly with either NaOH or H3PO4. Easy as that..

blogfast25 - 29-9-2012 at 09:25

How about converting the sesquicarbonate into a much more water soluble salt, e.g. sodium acetate, and leaching that out of the Savogran with boiling water. You'll lose some phosphate too, of course. Then for purity, recrystallise the remaining phosphate (it has strong dependence of solubility on temperature).

unionised - 29-9-2012 at 12:18

How about adding H3PO4 to the mixed sesquicarbonate/phosphate and boiling to drive out CO2?

Hilski - 29-9-2012 at 13:27

Quote:
Add stoichiometric amounts of NaOH to H3PO4. Not harder than that..

Yes I know that, but I already have a bunch of H3PO3 on hand, but no phosphoric acid at the moment. Thus the question about the H3PO3.
Quote: Originally posted by unionised  
How about adding H3PO4 to the mixed sesquicarbonate/phosphate and boiling to drive out CO2?

Yeah I thought about that, and it's what I'll probably end up doing. I'll just have to go find some phosphoric acid that doesn't have a bunch of other crap mixed in with it.
Quote:
How about converting the sesquicarbonate into a much more water soluble salt, e.g. sodium acetate, and leaching that out of the Savogran with boiling water. You'll lose some phosphate too, of course

I would think you would lose too much phosphate to make it worth the effort. Correct me if I'm wrong.

Hilski - 29-9-2012 at 13:37

Would converting the carbonate in the Savogran to phosphite, using H3PO3 be a good idea? I dont think the phosphites would have a negative effect on the plating bath like the carbonate does.

weiming1998 - 29-9-2012 at 22:46

Quote: Originally posted by Hilski  
Would converting the carbonate in the Savogran to phosphite, using H3PO3 be a good idea? I dont think the phosphites would have a negative effect on the plating bath like the carbonate does.


You have phosphorous acid, but no phosphoric acid? That's a bit strange. Phosphoric acid is much easier to acquire than phosphorous acid.

Anyway, you can either add a oxidizing agent like H2O2 to the phosphorous acid to oxidize it to phosphoric acid, or you can first react your TSP with vinegar to form sodium (di)hydrogen phosphate and sodium acetate. After that, just add some NaOH. It will convert the sodium hydrogen phosphates to trisodium phosphate, which precipitates out due to its low solubility compared to the extremely soluble sodium acetate and NaOH.

Hilski - 30-9-2012 at 17:00

Quote:
You have phosphorous acid, but no phosphoric acid? That's a bit strange. Phosphoric acid is much easier to acquire than phosphorous acid.

Yep. Fortunately, I haven't any problems getting small quantities of lab grade H3PO3 from lab chem suppliers, mostly for hydroponic fertilizer use. I really don't use much of it so it usually lasts a long time once I get it.

It's not like I can't go out and get some phosphoric acid-containing product, but most everything locally would contain detergents and other crap that I would have to clean out before use.

Thanks for the tip on oxidizing with H2O2. It hadn't occurred to me to try that.

In any case, I found some 99.5% Na3PO4 online for like 6 bucks, so I'll just be using that once it arrives.

Thanks.



mayko - 24-5-2013 at 17:45

I too am facing this problem. I bought a box of Dap brand TSP only to find that it contains sodium carbonate (10-30% by weight, according to the MSDS).

Here's what I was thinking. According to Merck, the solubility of sodium phosphate is:

1 part in 3.5 water
1 part in 1 water @ 100C

So it should recrystallize nicely on cooling. However, sodium carbonate has a solubility of:

1 part in 3.5 water @ room temp
1 part in 2.2 water @ 35C

This looks like an even steeper solubility curve than the phosphate, so I would expect it to be even more liable to crystallize!

Perhaps one could use this fact to recover the phosphate from the mother liquor, but I had another idea. Sodium chloride has a solubility of:

1 part in 2.6 water @ 35C
1 part in 2.5 water @ 100C

Thus, sodium chloride has a much shallower solubility curve, and a higher solubility at low temperatures than the phosphate. So, what if one were to add HCl to a solution of the phosphate/carbonate mix, turning the Na2CO3 into NaCl and CO2, prior to recrystallization?

The only problem I'd forsee is a side reaction with the phosphate; would the addition of HCl convert the tribasic phosphate into the di/mono/acid forms of the salt? If so, could this be prevented, eg by monitoring pH or by adding just enough acid so that fizzing ceases?

Edit: Actually, why not do another recrystallization from NaOH solution?

[Edited on 25-5-2013 by mayko]

watson.fawkes - 25-5-2013 at 05:20

Quote: Originally posted by mayko  
The only problem I'd forsee is a side reaction with the phosphate; would the addition of HCl convert the tribasic phosphate into the di/mono/acid forms of the salt?
If you're going to add something, use Ca(OH)2 and precipitate the carbonate as chalk.

mayko - 6-6-2013 at 10:54

Good suggestion; I tried the HCl method but it was messy, time consuming, and had very low yields. The calcium hydroxide method was much more effective.

blogfast25 - 6-6-2013 at 13:53

Quote: Originally posted by watson.fawkes  
If you're going to add something, use Ca(OH)2 and precipitate the carbonate as chalk.


How is that supposed to work, when Ca3(PO4)2 is also incredibly insoluble?

Going by solubility products, Ca3(PO4)2 (Ks = 2.1 10-33) makes CaCO3 (Ks = 6 10-9) look positively soluble!

[Edited on 6-6-2013 by blogfast25]

aliced25 - 6-6-2013 at 13:58

Hang onto the H3PO3, it is a source of dry phosphine which burns in an atmosphere of Cl2 to give PCl5. There are better routes to Trisodium Phosphate (the reaction of sodium phosphate with ammonium phosphates being one).

blogfast25 - 6-6-2013 at 14:03

For H3PO4, turn the skeletal remains of your last chicken dinner into bone ash and treat with conc. sulphuric acid. Seriously... :D

watson.fawkes - 6-6-2013 at 16:09

Quote: Originally posted by blogfast25  
Going by solubility products, Ca3(PO4)2 (Ks = 2.1 10-33) makes CaCO3 (Ks = 6 10-9) look positively soluble!
Hmm. Should have looked that up first.

Added: I should have also looked up the other aqueous phosphate species. Calcium monohydrogen phosphate, CaHPO4, K ~ 1 × 10–6. Calcium dihydrogen phosphate Ca(H2PO4)2, K ~ 1 × 10–4. Calcium bicarbonate doesn't exist as a solid. So assuming the pH isn't too high, the monohydrogen phosphate species is available to keep phosphate solubility above that of the carbonate.

I'm not claiming this is a definitive answer; it's not. There's also the Ka for all the acid species involved. I don't have good intuition for what the phase diagram looks like.

[Edited on 2013-6-7 by watson.fawkes]

mayko - 7-6-2013 at 10:33

Ack, how embarrassing. I analyzed the 'purified' and commercial TSP by measuring the CO2 given off on acidification. It seems that I enriched the material of carbonate (42% by mass compared to 30%) rather than depleting it!

[Edited on 8-6-2013 by mayko]

blogfast25 - 7-6-2013 at 12:39

Quote: Originally posted by watson.fawkes  
Added: I should have also looked up the other aqueous phosphate species. Calcium monohydrogen phosphate, CaHPO4, K ~ 1 × 10–6. Calcium dihydrogen phosphate Ca(H2PO4)2, K ~ 1 × 10–4. Calcium bicarbonate doesn't exist as a solid. So assuming the pH isn't too high, the monohydrogen phosphate species is available to keep phosphate solubility above that of the carbonate.


[Edited on 2013-6-7 by watson.fawkes]


Trust me, that also occurred to me but the Ka2 and Ka3 of H3PO4 are really small. I wouldn't like to bet that selective precipitation of CaCO3 is possible here.