Sciencemadness Discussion Board - Distilling H2SO4 from NaHSO4

Distilling H2SO4 from NaHSO4

dex - 10-10-2024 at 07:35

I spent a couple of days searching the forum for a cheap and effective way of producing concentrated H2SO4 and came across an interesting post by JJay, in which he stated that you could heat NaHSO4 to obtain H2SO4 vapors, quoting the book "Small-Scale Synthesis of Laboratory Reagents" by Leonid Lerner.

I have seen this reaction being discussed here for making oleum but not as a cheap way to make H2SO4 for the amateur chemist. In particular, the book states:

Quote:

Oleum can be obtained by the pyrolysis of NaHSO4, which in the fluid state is equiv-
alent to an equimolar mixture of H2SO4 and Na2SO4:

2NaHSO4 ≡ H2SO4 + Na2SO4. (equation 20.5)

The book then further states

Quote:

Equation 20.5 turns out to be a realizable reaction in practice. The present experi-
ment shows that H2SO4 can be distilled from NaHSO4 with almost 100% efficiency.
While collecting the distillate as a single fraction gives 100% H2SO4, unlike the SO3/
H2O system, introducing a cut in the distillate fraction yields oleum, with the lower
bp fraction being therefore a correspondingly weaker acid (because the combined
fractions must give 100% H2SO4).

If I can read this correctly, then why isn't it suggested more often as a common way of producing sulfuric acid? I may be missing something here which is why I'm making this post before attempting anything too dangerous. Honestly, reading about it in the book makes it sound so easy, but then absolutely no one mentions it in the usual H2SO4 talks, nor has anyone posted this reaction on youtube, so that makes me a little anxious about attempting it by myself right now.

Fluorite - 10-10-2024 at 08:38

I think it's more about when NaHSO₄ is heated and decomposes into H₂SO₄ and Na₂SO₄, the resultant solid sodium sulfate can form a hard block inside the glassware that would be very hard to remove. This block could also expand as it forms, potentially putting stress on the glass and causing it to crack or break.

I was wondering if there might be alternative setups to mitigate this. For example, would it be possible to heat the bisulfate on a glass plate and use vacuum suction connected to an inverted funnel placed above it to collect the H₂SO₄ vapors? That might avoid the risk of damaging enclosed glassware.

Alternatively, maybe using a quartz tube reactor with a dry inert gas flow (like nitrogen or argon) to sweep the H₂SO₄ vapors could work? Quartz can handle higher temperatures and might be less likely to suffer from the thermal stress.

dex - 10-10-2024 at 09:55

Thank you for your insight. Is the Na2SO4 product that hard to remove? Looking at Thy Labs' video on sodium sulfate it doesn't look that bad but I don't know.

The author of the book I mentioned also wrote a lab report. I can't post the whole book here because of copyright but if you can find it I invite you to read the entire section on SO3, that is very interesting. In particular he does use quartz glassware for heating, but he uses a simple distillation arm, into a 2-neck RBF immersed in a cold bath with a condenser on the 2nd neck.

bnull - 10-10-2024 at 11:17

Lerner writes on page 181 of the same book that a "result of practical significance in the present experiment is that (...) Na₂SO₄, which is solid at the maximum reaction temperature, does not attack quartz or expand on cooling", so expansion is not an issue.

For me it is both the reaction temperature and the concentration. Unless you're in the UK, you can buy a bottle of ~35% sulfuric acid for batteries, which is more than twice the concentration obtained from decomposition of bisulfate in the range 260 to 420 °C. In either case, bisulfate or battery, distillation will be necessary to remove water. And dealing with sulfur trioxide is not exactly a pleasure.

By the way, could you please share the title or DOI of the report? Lerner was probably too modest to reference his own publication in his book.

dex - 10-10-2024 at 11:51

Quote:

By the way, could you please share the title or DOI of the report? Lerner was probably too modest to reference his own publication in his book.

Sorry for the confusion, I meant the report that is in the book. :')

Thank you for clearly laying out the downsides. I am not in the UK but I don't think you can buy battery acid in the EU anymore. It's such an important reagent and I just want a reliable way to get as much as I want, and NaHSO4 is so cheap, I might just take the time to replicate his entire setup with the box oven. If I do not take cuts that should avoid having to deal with the SO3 right? I'm not too sure about that, or even how to handle SO3...

Keras - 10-10-2024 at 11:53

Quote: Originally posted by dex

This reaction needs blowtorches and a quartz RBF as well as some special glass pieces (a 65° bend, an air condenser…) to be conducted effectively. There are also a lot of SO₃ fumes escaping, which makes it impractical unless you operate in a fume hood or outside.

If you want to make concentrated sulphuric acid you first have to preheat the RBF at 350/400 °C for a while in order to evaporate the major part of the water that the reaction produces, which needs a heating mantle that can reach this high.

Really, the best solution to make sulphuric acid (diluted) is to do what we devised with NurdRage: make a concentrated solution of copper sulphate and add the stoichiometric amount of oxalic acid. You’ll precipitate copper oxalate which is totally insoluble, and be left after filtration with diluted sulphuric acid. Painless, harmless and odorless.

woelen - 11-10-2024 at 01:32

I tried the reaction with NaHSO4.H2O some years ago, without success.

The solid "melts" very easily, the solid NaHSO4.H2O melts in its own water of crystallization at well be low 100 C.
On further heating, the water can be boiled away, leaving behind anhydrous NaHSO4.
The next step occurs at below 300 C. More water is lost. What remains behind is solid Na2S2O7. And here the fun stops.
In order to get free SO3 from the Na2S2O7 you need insane heating. In glass, this is not possible. Maybe in quartz, but I do not have that. I heated until my test tube became soft, but still no SO3.

I also tried with Na2S2O8. You can easily convert this to Na2S2O7 by heating, but again, at that point the fun stops.

Keras - 11-10-2024 at 02:39

Quote: Originally posted by woelen

Maybe in quartz, but I do not have that. I heated until my test tube became soft, but still no SO3.

I’m surprised. A single blowtorch suffices to get plenty of sulphur trioxide fumes, as shown in the attached picture (the test tube is made of quartz).

dex - 11-10-2024 at 10:04

Thanks everyone. Great picture. I'll try to get my hands on quartz glassware. Copper sulfate + oxalic acid are more than twice as expensive as NaHSO4 from what I can see. It'll take some time but I'll report back on my results if I do it.

Keras - 11-10-2024 at 10:19

Please do. We’ll be pleased to help

RU_KLO - 14-10-2024 at 14:11

Check Nurd Rage videos regarding oleum.

https://www.youtube.com/watch?v=hUyJ6CibhSg&t=5s&ab_...

https://www.youtube.com/watch?v=wB2zzm8VP9Y&ab_channel=N...

dex - 2-8-2025 at 10:15

I have tried that reaction twice, heating a half-filled 250 mL quartz RBF from below with a butane torch. On my first try I had the same experience as woelen: melts quickly, then boils promisingly, some water drops, then it suddenly stops boiling and nothing happens. You do get a few SO3 fumes this way but the liquid comes out dilute.

So the main issue is temp / insulation, I looked around on YouTube and saw videos of cheap DIY melting kilns / ovens, which inspired me to buy refractory bricks. I laid 4 of them around the flask and this time I was able to get a workable quantity of a strongly fuming liquid (no solid). Still, the reaction got very stagnant, very quickly. After cooling the quartz flask, I was able to dissolve the remaining solid in water (which took around 15 minutes), and by looking at how sharply a few drops affected the pH in a bucket of tap water I could see that most of the acid did not react.

(Also worth noting that instead of a 75° bend, which I do not have right now, I used a 3-way distillation adapter and sealed the thermometer inlet with PTFE tape, but the tape was strongly carbonized by the time I stopped and leaking a lot of fumes. Similarly, the PTFE tape I used on the joints was also attacked.)

So the next step is buying refractory cement, more bricks and making a "real" kiln, gapless and complete with a little lid would be perfect. Again I'll post another update if I do it. I think I'm not far from making it work and I think it would be really cool to have half a litre of the stuff.

(Though after this adventure I can see how much better the salt precipitation method is. When I made this post I thought SO3 was "more concentrated H2SO4 = better / will dehydrate more" but now I realize that it will cause unwanted sulfonations, I did not see it as the entirely-something-else that it is. I think however that this oleum can be useful to further concentrate the sulfuric acid made by the salt precipitation method after boiling.)

jackchem2001 - 2-8-2025 at 16:01

Quote: Originally posted by Keras

I’m surprised. A single blowtorch suffices to get plenty of sulphur trioxide fumes, as shown in the attached picture (the test tube is made of quartz).

I would be quite worried about different thermal expansion around the joint between borosilicate and quartz, but I am not sure if this is a problem in practice. Would be safer to have any parts of the apparatus that receive heat to be quartz

MrDoctor - 3-8-2025 at 01:34

the tendency for glassware to reportedly brake through this process might actually be that its not pure sulfate but rather, sulfate soaked with bisulfate that has yet to react. and what sulfuric acid doesnt leave the system probably will react to form bisulfate again as it cools, forming a melt, and consequently, forming some sort of mass probably at the bottom of the flask that will stick to some parts, and expand in a way that potentially breaks the glass if its present in some specific percentage window.

a quartz test tube, and using a compressible and/or porous filler would resolve this. you can buy quartz test tubes with ground glass joints at that, moderately cheaply off alibaba. it would allow for rather efficient heating using a blowtorch since heat doesnt need to travel as far, so you spend less time burning gas just to keep the outside hot while you get a slow heat creep into the thick core of your mass. Batch size would be reduced but, its doable and repeatable, plus acid boiled is proportional to gas burned, so timewise your distillation rate shouldnt be any worse, and only stands to go quicker.

If you end up doing this, there is something you may consider experimenting with, which is the addition of solid polyphosphoric acid to the bisulfate melt. In short, polyphosphoric acid, and also pyrophosphoric, or whatever the dehydration below poly is, can dehydrate sulfuric to SO3, so it might be able to dehydrate the sulfuric as it attemps to form as a seperate molecule from the bisulfate, i dont know the proper term for this kind of reaction but its essentially adding a pulling element to something where two things are already pushing apart.
This might result in much lower temperature liberation of SO3 and a direct production of oleum, the only downside or risk here is that the dehydrated phosphoric acid needs to stay a solid and not really contact the glass or quartz.
to produce polyphosphoric acid you can use a blowtorch and a copper lined crucible, it wont attack it too badly, i think even electroplated copper on nickle has been reported on SM to work.

I also cant help but wonder, if maybe a steel retort might not work better for this, doing the destructive distillation normally i mean. if you can make sure its water free, dont certain kinds of steel play pretty friendly with sulfuric acid by forming passivation layers?
If simultaneous dehydration of formed sulfuric does work, at very least i know SO3 is fairly compatible with either iron or steel of alloys that arent too fancy.

this is based on things i have read and not first hand experience, so please keep this in mind. though the ability for polyphosphoric acid to dehydrate sulfuric to SO3 should be documented somewhere on the forum.

By the way since you say oxalic acid is expensive. you can make it, if you have access to glycollic, glyoxylic acid, formic acid(maybe), or ethylene glycol.
glycolic acid is available in some places pretty cheaply when bought in bulk, as a skin whitener, i see it often on aliexpress, so i would imagine a cruder grade should be available somewhere.
glyoxylic acid when heated disproportionates to oxalic and glycolic. glycolic acid and ethylene glycol are reduced forms of oxalic and glyoxylic acid. if you take any of the series of carboxylic acids here and perform a simple electrolysis, with just a rudimentary membrane like a porous membrane, at the anode it will oxidize these all the way to oxalic, and at the cathode, reduce all the way to glycolic acid, and then eventually ethylene glycol.

you only need to make it once since i believe you regenerate the oxalic acid from oxalate salt, also, the same would go of copper, though a multitude of sulfate salts can be used. the idea being either buy the sulfate, like magnesium if it works, or iron sulfate if its cheap for you, or, buy a convenient anion, and react your bisulfate with it, seperate, then use it in the oxalate precipitation method instead of just electrolyzing the sulfate salt, from experience iron sulfate is very easy to electrolyze, i found that when trying to make electrolytic iron, the anode would dissolve much slower than iron plated out, causing me enormous problems since too much sulfuric acid formed and would rather dissolve high surface iron near the cathode, than dissolve the anode. you can probably strip the iron from the solution, then, i think adding a touch of peroxide will oxidize the iron sulfate remaining into a solid which will slowly react with iron to reform normal iron sulfate again, but in the mean time it forms an insoluble annoying to filter solid from which rather pure acid should be obtainable. In place of a sacrificial steel anode, a lead anode worked too, i produced 500g of iron with no visible corrosion, but no doubt lead went into solution and formed solid lead sulfate

[Edited on 3-8-2025 by MrDoctor]

Radiums Lab - 4-8-2025 at 12:19

Lab Coatz posted a video on this synthesis recently go check it out. He used a furnace and some quartz glass. Actually he made some oleum by using this method.

[Edited on 4-8-2025 by Radiums Lab]

MrDoctor - 4-8-2025 at 22:18

Quote: Originally posted by Radiums Lab

Lab Coatz posted a video on this synthesis recently go check it out. He used a furnace and some quartz glass. Actually he made some oleum by using this method.

[Edited on 4-8-2025 by Radiums Lab]

i misread and thought you said a tube furnace, ive been wanting to see someone pull off a nice clean benchtop contact process for ages.
I watched the video, was only like 16 hours ago it was posted and he got 40% yield of oleum claiming 60% was possible, which is pretty amazing given starting from bisulfate. the use of a furnace probably lets this happen reliably assuming the speculated cause of broken glassware is what i said before.

for anyone disuaded by the need for a furnace to hit 550C, let me tell you. that temperature is easily reached using an ordinary paint can and an induction coil, exploiting the fact that you can insulate the crap out of the steel tin, wheras with flame, you need an exposed surface that also ends up being burned actively, and electrically with heating wire, theres electrocution hazards, plus you have to cast the damn thing so a thermal insulator is holding the resistive wires up, but you can make a paint-can furnace that could easily reach over 600C using a 400W 48VDC power supply and a cheap "1000W" ZVS driver from ali, just place them in something like a terracotta clay pot and pour vermiculite in to fill all the gaps like its packing peanuts. most expensive thing will probably be the quartz flask.

once pyrosulfate has been produced, I wonder, could some sort of steel, or steel with an electroplated interior, like pure iron, nickle, or something else, be used to further break down the pyrosulfate? Because wether induction, flame or whatever really, it would make the second part of the process vastly more simple if you didnt have to use quartz.
I only know that pure iron will tolerate dry SO3 gas, which is how steel pipes can be used in the contact process at scale.

Alkoholvergiftung - 5-8-2025 at 00:01

Interesting old patent of making SO3 from water free sodiumbisulfate and water free Magnesiumsulfate. Its in German.
On bottom there are drawings of the plant.

Attachment: DE000000003110A_all_pages.pdf (152kB)
This file has been downloaded 119 times

MrDoctor - 5-8-2025 at 02:50

ah yes, in the true spirit of patentry, it defines the temperature the reaction will be conducted at, as the temperature at which the bonds between the sodium salt and magnesium salts are most fluid. the use of magnesium would decrease the temperature needed for SO3 to form, but, at what temp that is? who can say, just heat to a molten state, then heat even more until it reacts.

Now my technical german is not great, have i misunderstood that the double-salt of sodium-magnesium would form?
Also, its talking about the theoretical yield, does it mean to say this process easily hits the theoretical yield? So, you get a mol of SO3 per bisulfate used? or is it still half and the product is the sulfates of either?

Alkoholvergiftung - 5-8-2025 at 05:40

For temperature its writen significantly under darkred glowing. Darkred is around 600C , an mol of SO3 per bisulfate thats my understanding.

MrDoctor - 5-8-2025 at 06:03

if a mol of SO3 per bisulfate, that means the byproduct is NaMgSO4, is that possible?
if so though, thats extra incredible, also i did read the dark-red part, though thats reffering to how hot the heat source should be, but i guess it probably means thats about the reaction temp.

given this drastic yield improvement though, it makes it more viable to just put a bunch of this stuff in something like plumbing pipe as a retort, because of the yield relative to the mass. I think ill give this a try at some point and report back how it goes, to see if its 1 mol or 0.5. a critical issue is, if it really is the sodium double salt of magnesium sulfate, thats going to probably be alkaline enough to damage quartz, which iirc is etched a lot easier than borosilicate is.

dex - 8-8-2025 at 08:30

I did it again on the same butane bunsen burner, except this time I had more insulation (more bricks and a glass wool lid). The cost of the furnace setup without the quartz glassware is just below a hundred euros. I took a cut later than I did before, and though I couldn't see anything during the process due to the thick fumes produced inside the glass and thought it was going nowhere, this time I was left with quite a lot of white solid. (Sorry for the qualitative description but I'm not taking any risk handling that, see attachment, it's in a 500 mL flask).

I melted the solid with a hot water bath and managed to pour it into a glass bottle without *too much* trouble. (When pouring, I placed a fan behind my back because the air outside was very thick and I wanted to make sure the vapors would go the other way. Instead it blew a TON of smoke in my garden, which reminded me how Leonid mentionned in his own report it was once used as a smoke agent. Scary! But worth noting that I did not have any fuming when the glassware was connected.)

Now I'd rather not have to deal with such profusely fuming acid every time I need 50 mL of acid. So I thought about it and one thing that makes the CuSO4/Oxalic acid precipitation method unattractive to me is the price of the copper sulfate. If you could do it with the much cheaper aluminum sulfate, whose oxalate salt is also insoluble in water, from what I can see the cost per mol of H2SO4 would be almost identical to that of this method (but would produce a less valuable diluted acid).

With careful temp control and consistency you can certainly estimate how much SO3 is going to come over, and distill that into ~80% dilute sulfuric acid to neutralize the free SO3. This should be a much less dangerous, easier and more predictable way to obtain plain 100% sulfuric acid. This is what I'm going to try next, I'll try to make a kiln with temp control, etc. However I now have enough sulfuric acid as it is and it will probably have to wait until next year.

Quote: Originally posted by MrDoctor

This might result in much lower temperature liberation of SO3 and a direct production of oleum, the only downside or risk here is that the dehydrated phosphoric acid needs to stay a solid and not really contact the glass or quartz.

That would be a problem yes, I don't want to risk damaging the flask, maybe using a cheaper vessel. If you take a look at Leonid's thread you'll see how he uses a really cheap vinegar dispenser. About using a steel retort, I think getting it completely water-free would be hard, even if some type of steel managed to handle it.

Quote: Originally posted by Radiums Lab

Lab Coatz posted a video on this synthesis recently go check it out. He used a furnace and some quartz glass. Actually he made some oleum by using this method.

Thanks, it inspired me to use glass wool, with great success :').

Quote: Originally posted by MrDoctor

for anyone disuaded by the need for a furnace to hit 550C, let me tell you. that temperature is easily reached using an ordinary paint can and an induction coil, exploiting the fact that you can insulate the crap out of the steel tin, wheras with flame, you need an exposed surface that also ends up being burned actively, and electrically with heating wire, theres electrocution hazards, plus you have to cast the damn thing so a thermal insulator is holding the resistive wires up, but you can make a paint-can furnace that could easily reach over 600C using a 400W 48VDC power supply and a cheap "1000W" ZVS driver from ali, just place them in something like a terracotta clay pot and pour vermiculite in to fill all the gaps like its packing peanuts. most expensive thing will probably be the quartz flask.

Well keep in mind that, at 500, even 600°C, you are not getting that much "bang for your buck", most of the juice comes over much later. Leonid notes that a 500°C cut = 90% yield, and what 10% SO3 was lost probably came along with the water.

Anyway, thank you everyone in this thread for all your valuable thoughts.