Sciencemadness Discussion Board

Analysis of N2O

1281371269 - 26-1-2011 at 12:45

Someone from my school was posting around on facebook asking if anyone's dad was a scientist and could help 'analyse a gas'.

I called him, and it turns out that he has a bunch of cannisters of laughing gas which don't seem 'normal' - they apparently have a different taste and seem much stronger, than the laughing gas he has used before. He's offering me £100 (yeah, right) if I can find anything out, or alternatively find a small research lab which will do some basic analysis.

N2O is not something that's ever appealed to me to use personally ( is, IMO, equally effective) but as it's not actually an illegal substance I'm not too bothered.

I've been thinking through some potential tests:

Bubble it through dilute acid / base solutions. Any changes in pH would show addition of an acidic / basic gas

Bubble it through a very strong oxidiser (piranah acid?). Any darkening of the solution / any reaction at all would suggest an organic additive (but I would need a seriously clean piece of glassware to avoid false results)

Heat it in a closed vessel (tricky) and look for evidence of carbon as a decomposition product from an organic additive

Test its solubility in water (again, tough to do accurately)

Perhaps a simple test of its colour?

Can anyone comment on these methods or suggest others?

Thanks in advance for the help,

SmashGlass - 26-1-2011 at 14:25

Oh Mossy....

Your "friend" probably has got their hands
on tanks of NOS and is looking of a way of
purifying it. Obviously to turn a buck.
Otherwise they wouldn't offer a cash for
help incentive.

Therefore you get 0 (zero) help...

Medical grade nitrous oxide is clean. And has no taste.
NOS grade is not, and would taste funny
from the impurity added for that specific reason.


1281371269 - 26-1-2011 at 14:58

I don't know, I don't think he's that intelligent if I'm honest. And I understand that it's small cannisters, whereas NOS is normally what they call it for cars (according to wiki) so that would be a large tank, no? I know of someone who used to buy the cannisters in large quantities from catering companies (used for whipping cream) and sell them at parties for a couple of quid each.

Regardless, if he wants it purified, I can't help him anyway. If he wants to know whether or not it's going to poison him, I don't see a problem in trying to find out. And it will be the first time ever that I can carry out a genuine piece of analysis in my lab.

I'll ask some questions tomorrow...I'm supposedly being given a cannister then, so I'll see if I get a brand name or something on that and start my research there.

Sedit - 26-1-2011 at 15:38

Its Sulfur Dioxide and will give your freind a bad case of sore lungs at the end of the night. It is an impurity in industrial grade N2O from what I understand. Don't go huffing that shit it will damage your lungs and you never know how much its contaminated. Sometimes its alot while other times its smooth.

SmashGlass - 26-1-2011 at 15:42

Thanks Sedit... I was trying NOT to give the plot away.

It's too easy now.


Knowledge without seeking is not knowledge.

Ozone - 26-1-2011 at 20:58

Perhaps, for that £100 I'll tell you.

Sedit - 26-1-2011 at 23:00

Don't listen to Ozone, hes only trying to ripe you off and make a quick buck...... Ill only charge you $75;)

turd - 27-1-2011 at 00:51

Quote: Originally posted by Mossydie  
Bubble it through dilute acid / base solutions. Any changes in pH would show addition of an acidic / basic gas

Forget chemical analysis. Find a lab which does atmosphere chemistry / analytics and let them put it in their GC/MS or MS/MS. You owe me £1. :P

Tiberius_C - 27-1-2011 at 01:14

In the same case as Mossy, I'd try dissolving some in dist. H2O & titrate with NaOH to get a rough idea of SO2 content. Might save the $$$ for atmospheric lab prices. Just my 2 bits, of course :-)

ps-- anyone know of something offhand, that gives a stoichiometric reaction w/ NaSO3?

ScienceSquirrel - 27-1-2011 at 03:23

Bubbling the gas through excess hydrogen peroxide would be even better.
The sulphur dioxide would be converted to the sulphuric acid.
Much stronger and easier to titrate.

1281371269 - 27-1-2011 at 10:55

ScienceSquirrel - great idea, thank you! I'll definitely use that method if I find evidence of SO2. But, having been given a cannister, I'm not convinced that SO2 is going to be in there at all (though I will test for it later). The labelling makes clear that the contents are intended for inhalation - it says 'filling kit for smelly balloon'. Everything else on it is in hebrew, so I can't look up an MSDS.

I'll report back with some test results later.

smuv - 27-1-2011 at 12:16

If it contained SO2 you would know. SO2 makes you cough at ppm levels (ppb levels for asthmatics) and does lots of nasty things to your respiratory track. Additionally, SO2 has no anesthetic effects (it is WAY too polar).

At first, I was thinking it was just placebo effect, but now that you say the canisters are meant to be used for inhalation (and have such illegitimate instructions), this changes things. I think a good test to do would be to burn the gas, most anesthetic gases (other than xenon and conveniently N2O) are fuels. Of course N2O is a great oxidizer, so you need to keep that in mind when you interpret your test results.

FYI Medical Anaesthetics (I can think of) that are gases at STP which you may want to test for: Cyclopropane, ethylene, ethane, acetylene, xenon, N2O

Also keep in mind that things like propane and butane are also anaesthetics (NOT just simple asphyxiates), it would not be totally unreasonable to suspect a shady company looking for a quick buck would sell these gases as N2O for recreational use.

SmashGlass - 27-1-2011 at 13:14

OK. OK. It seems enough people are on the right track by now.
Yes it is most likely SO2 (sulfurdioxide). This is an additive to
N2O that can be tasted at ppm levels. Normally the amount
of SO2 added to N2O is about 100ppm.

Now go find the best method for scrubbing SO2,
without destroying the N2O from your line.
Hint it's an industrialprocess and is most likely
patented long ago. Hint hint...

Persistent little buggers aren't you.


ScienceSquirrel - 27-1-2011 at 13:47

Quote: Originally posted by SmashGlass  
OK. OK. It seems enough people are on the right track by now.
Yes it is most likely SO2 (sulfurdioxide). This is an additive to
N2O that can be tasted at ppm levels. Normally the amount
of SO2 added to N2O is about 100ppm.

Now go find the best method for scrubbing SO2,
without destroying the N2O from your line.
Hint it's an industrialprocess and is most likely
patented long ago. Hint hint...

Persistent little buggers aren't you.


One of the best ways of removing and assaying sulphur dioxide on a lab scale is reaction with hydrogen peroxide. Reaction is quantitative and fast.
It will not react with nitrous oxide or a lot of other gases.
Sulphuric acid is a strong acid, unlike sulphurous acid it does not bleach indicators and is easily titrated.

ScienceSquirrel - 27-1-2011 at 14:03

I doubt that it is sulphur dioxide, but it is not that irritating.
I use sodium metabisulphite as a rinse solution and sometimes you catch a whiff of the gas.
It makes you cough but it is not that nasty.
I would describe chlorine, hydrogen chloride, nitrogen dioxide, nitrosyl chloride as far worse.

1281371269 - 27-1-2011 at 14:08

I was wrong - it's not sold for inhalation, apparently. It's sold with a balloon, the idea being that you put some aftershave or similar scented product in the balloon, fill it with the gas, and let it float around the room. After a while, the room will supposedly start to smell of this scent.

I spent an hour or so doing as many tests as I could think of with my limited reagents / equipment. Here's what I found:

It's colourless

It has no smell, but feels faintly 'sweet' if inhaled

It does not affect the pH of distilled water and it does not change the colour of damp indicator paper, nor does it reduce hot Na2Cr2O7, or cold KMnO4 (both bubbled through slowly). So there can't be any SO2 in it.

It does not decolourise bromine water. So not ethylene or acetylene. Presumably there's no chemical test for the alkanes or xenon?

If blown gently onto a bunsen, one can see a green flame - Boron? Copper?

There's no physical evidence of decomposition (e.g. carbon residue) when passed through a glass tube heated by a roaring bunsen flame.

After all this, I was about to give up, but noticed that the gas coming out of the glass tube had a much sharper smell than the gas going in. I tested it with damp indicator paper which instantly turned dark red. Even the top, dry, bit of indicator paper went red. I repeated the experiment using a different glass tube and no rubber tubing in case either of these had contaminated the gas and got the same result.

So there is presumably some sort of organic additive which breaks down at high temperatures (though fairly low high-temperatures, because the first test simply involved passing the gas through a setup like this: - with a bunsen underneath it). There's one thing I can think of that might have affected this result, which is that both heated glass setups had previously been washed with distilled water, so some of the gas coming out would have been water vapour. This shouldn't change the pH in itself, but perhaps there was a reaction with H2O?

Based on those results, does anyone have any ideas? At least, it seems that I've proved that something has been added, which might earn me that £100. Or maybe like, £15.

SmashGlass - 27-1-2011 at 14:10

Sure... Just nip down to the local pharmacy for
some 3% or 10% peroxide solution.
To perform the initial tests.
Once that shows SO2, if you can work out titration...
Go back for a multi-liter purchase of peroxide solution.
I buy mine by the gallon normally too!

Not only will you have the cleanest N2O you will
also have shiny white teeth until your lungs are
bleached by peroxide vapour from the streaming
gas, as the obviously talented mental giants
operating the device will have it cranked
waaaay too high.

The sheer insanity of the end result is almost tangible...
Can you taste it? As opposed to the initial SO2
present that is...

garage chemist - 27-1-2011 at 14:17


If blown gently onto a bunsen, one can see a green flame - Boron? Copper?

That's an alarming sign. Halogen compounds do that. Does the bunsen flame give off pungent fumes as well when some of the gas is burned in it?
Also, the observation that the gas resulting from thermal composition is acidic strongly points towards a halogen compound. Methyl chloride?
Or perhaps one of those hydrofluoroalkanes that are ever so widely used as nonflammable refrigerants and propellants today! Tetrafluoroethane? Difluoroethane? Vinyl fluoride?
I'm guessing one of those, and that the acidic gas you observed was HF. Be careful!

[Edited on 27-1-2011 by garage chemist]

ScienceSquirrel - 27-1-2011 at 14:25

Does a balloon filled with the gas actually float?
If the gas is odourless and the balloon is lighter than air then a major component is hydrogen or helium.
Sulphur dioxide is ruled out by the absence of any smell, it has a peppery odour at even light concentrations.
Is the gas mixture flammable?
If there is a fuel gas in there it will burn.
Nitrous oxide decomposes when heated to form mainly nitrogen and oxygen but small amounts of nitrogen dioxide may also be formed and that would account for the acid reaction.

smuv - 27-1-2011 at 14:37

Did the flame color look like this ?

This is burning ethyl chloride, all chlorides burn with a flame of about the same color.

GC are you sure Fluorides burn with a green flame color? I thought fluorides were blue.

Edit: ya I check fluorides are blue, though small amounts of chloride contamination make them look green. It could be possible that early texts reported fluorides as green because of this.

[Edited on 1-27-2011 by smuv]

ScienceSquirrel - 27-1-2011 at 14:46

Ethyl chloride could be a good possibility.
Cracking at red heat would make ethene and hydrogen chloride.
An acid gas which would explain the litmus reaction.
If you could absorb the exhaust gases from your experiment on wet blotting paper or similar and test it for nitrate and chloride you might move forward.

1281371269 - 27-1-2011 at 14:54

Thank you all very much for the suggestions.

Yes, the flame was exactly that colour.

I didn't see any brown, so ruled out NO2 as causing the acidity. Also, I don't think there would be enough to have had such a noticeable effect.

I don't have any balloons at the moment to test the density in relation to air.

It doesn't seem to burn. The stream of gas had no effect on a 'storm match' at all, put out a normal match, and had the same effect on the bunsen flame as blowing it - but didn't put it out. The haloalkane suggestion sounds quite probable. Tomorrow I'll dissolve the gas given off in water and do a halide test. How can I test for a nitrate? The tests a google search suggests seem a bit beyond my means.

I wonder if there is, in fact, any N2O in there at all. I suppose the one thing to do would be to try inhaling some and see if I laugh...but I'm not very keen to do that for obvious reasons. Other than flammability, are there any tests I could do for N2O? I know that it's a strong oxidiser, so how about reaction with red Benedict's solution?

[Edited on 27-1-2011 by Mossydie]

SmashGlass - 27-1-2011 at 15:04

Getting very worried now...
It can extinguish a match.
Not N2O that's for sure.

This is probably best left in the
DON'T TOUCH IT. Category.

But it's good to see you are
determined enough to see it
through to this point.

garage chemist - 27-1-2011 at 15:12

N2O makes a glowing wood splint burst into fire, just like oxygen.
There's certainly not much, if any, N2O in there if the gas does not support combustion.

A good idea would be to pass some of the gas through a trap cooled with dry ice/acetone. That will condense the fluorohydrocarbon, while the boiling point of N2O is conveniently somewhat lower than the temperature of dry ice.
Then you can take the liquified gas out of the cooling mix, add a few boiling stones and measure the temperature it boils at, or better a boiling range.
The uncondensed gas from the dry ice trap can be tested for N2O with a glowing wood splint, to confirm or rule out the presence of N2O.

Leading the gases from pyrolysis through water is also a good idea. Keep in mind that silver nitrate does not precipitate fluoride- test with CaCl2 solution too. CaF2 precipitate is slimy and may not be clearly visible when it forms.

ScienceSquirrel - 27-1-2011 at 15:14

Nitrous dioxide is not an oxidiser at room temperature and pressures.
It is stable at room temperature and pressure but falls apart at combustion temperatures to form a mixture containing about 30% oxygen so it is an ideal way to pump up an engine.
Room temperature oxidations with nitrous oxide and a catalyst would be clean, green and use an industrial byproduct.
It could be worth lots of cash to the inventor and might score you a Dalton medal or the Nobel prize!

1281371269 - 30-1-2011 at 08:16

I've done some further tests and am pretty much stumped.

I passed the gas (x) through a heated glass tube as before, then the exhaust gasses (y) were bubbled through distilled water to form a solution (n). Any gases (z) that didn't dissolve in water were passed through another length of tubing and bubbled through Bromine water.

The exhaust gasses (y) didn't seem particularly soluble - after a short while I noted dense white fumes sitting above the solution (n) in the gas washing bottle and also coming out of the last length of tubing as z. I remember seeing similar looking fumes when we cracked longer chain hydrocarbons in school last year, so they would seem to fit with the HX + alkene idea. However, they were still turning indicator paper dark red and didn't seem to decolourise Br2 water (though I would need to repeat this test to be sure)

The solution (n) didn't give a product with the silver nitrate halide test. It was also only slightly acidic (pH ~3-4). Both of these again match the HF idea. However, n also doesn't seem to give a precipitate when mixed with warm CaCl2 solution. Do I need to heat it? How difficult is it to see the precipitate?

I don't really know what other tests to do or what conclusions to draw from this...

Thanks again for the help,


1281371269 - 30-1-2011 at 08:53

Also, I've double checked and he would still like to pass the stuff on to a proper lab for further testing. Does anyone here (around London, UK) have the machinery to carry out that kind of analysis (for a reasonable sum) or know of somewhere which would do it?

Ozone - 30-1-2011 at 08:59

Don't listen to Ozone, hes only trying to ripe you off and make a quick buck...... Ill only charge you $75


The green/blue flame color is the tell. I agree with GC, as he beat me to it (I though the same thing when you mentioned the green flame test). Excellent testing, by the way; it's amazing how many forget about flame tests, these days.

Also, if the collar on the bunsen is copper (or copper alloy), the resulting HCl can give CuCl2 which is volatile and gives a nice green flame color; this is the basis of a reliable test for organic halogen--flame test on copper.


smuv - 30-1-2011 at 10:05

Yes, definitely flame tests can tell a lot but are not used much these days...

There used to be a copper test, where the tip of a copper wire was allowed to stand in a flame until it no longer colored it (covered by non-volitile oxide). Then the passivated wire was dipped into the halocarbon so the tip was moistened. Finally the tip was put back into the flame, Cl, I and Br would give distinct flame colors as they all yield volatile copper salts, however Fluorocarbons do not color the flame because there copper salts are non-volatile. I don't know exactly how this can be modified to work on a gas but I am sure something could be worked out.

It also might be a good idea, to do a gas density test. This could help immediately rule out N2O.

Based on what you have indicated, I think it is highly likely that you are dealing with a fluorocarbon, probably either di, tri, or tetrafluoroethane.

Also, inhaling the stuff will probably tell you little; all anesthetics make people a little silly. This includes, fluoroalkanes, choloroalkanes and N2O.

[Edited on 1-30-2011 by smuv]

1281371269 - 1-2-2011 at 10:59

I'll definitely try that copper wire test, thanks.

I also had another idea about testing for HF: I measure the pH of the solution (I can only do this roughly, with UI indicator paper) before addition of CaCl2, and then again afterwards. If HCl has been formed, I should see a notable increase in pH - not only because HCl is a better acid, but because two moles are formed for every mole of HF.
However, CaCl2 is basic in solution - will this significantly affect my pH values?

1281371269 - 9-2-2011 at 12:39

I've pretty much found out all I can with they equipment available to me - and I've been paid (not £100, £40, but I'll get more if I can find out precise details).

Does anybody know of somewhere in the UK which will run the gas through a mass spec for me? (I think that should be enough, but if it's inconclusive then I guess GC / NMR). How much is it likely to cost?