[hamdani yusuf (OP)]

We are running a process that produce SO2 gas at 7% concentration as an intermediate product. Due to efficiency limit of the next stages, there is residual 0.2% SO2 wasted in flue gas. The gas is scrubbed using dillute NaOH in a wet scrubber before released to atmosphere to keep it below permitted limit.
Is there a commercially available method to collect that SO2 so it can be recycled to the process? Literatures on gas scrubber mostly direct me to Uranium enrichment. Since molecular weight difference between SO2 and air is much more contrast than Uranium hexafluoride isotopes, I think it should be feasible to use the same method here.

[evan_au]

Talk of flue gas suggests a composition of hot N₂ and CO₂ with traces of SO₂ and NO_x?
- And maybe ash, if it comes from coal (or not, if it comes from high-sulphur oil).

If the flue gas wasn't so hot after passing through the NaOH solution, you could try diffusion to slightly increase concentration of heavier molecules like SO₂?
- But the impact of increased exhaust back-pressure would need to be considered...

[Bored chemist]

Enriched uranium is very expensive.
It would be possible to strip SO2 out of the gas stream this using a gas centrifuge, but it would cost  a lot.
Even liquefying and fractionating it would be cheaper.

In principle, using Na2CO3 solution as the scrubber would make the process slightly cheaper.
Most of the NaOH you use is being converted to carbonate by CO2 in the air anyway.

If you are making lots of SO2, it may be possible to capture it as sulphite, which is a saleable by-product.

[hamdani yusuf (OP)]

The SO2 comes from burning liquid sulfur as raw material. Not much CO2 is found in the flue gas.
I want to reduce NaOH consumption because the resulting salt is discharged as waste water with TDS getting closer to allowable limit. Reducing the cost of NaOH comes as secondary goal.

[hamdani yusuf (OP)]

If you are making lots of SO2, it may be possible to capture it as sulphite, which is a saleable by-product.

We found that purifying it and drying it to meet industrial standards need significant efforts.

[Bored chemist]

Something like this
https://en.wikipedia.org/wiki/Pressure_swing_adsorption
this
https://en.wikipedia.org/wiki/Amine_gas_treating

or
https://en.wikipedia.org/wiki/Membrane_gas_separation
might work.

You might even be able to reclaim the SO2 for whatever it is you are using it for.

[chiralSPO]

You might be able to reduce it back to elemental sulfur using CO or H₂
https://pubs.acs.org/doi/abs/10.1021/acs.energyfuels.6b01006
https://www.sciencedirect.com/science/article/abs/pii/S0263876217301478

These still require stoichiometric reagent to quench the SO₂ (and a fixed bed reactor, that might not be cheap), but at least the product has positive value, not negative value.

[hamdani yusuf (OP)]

Something like this
https://en.wikipedia.org/wiki/Pressure_swing_adsorption
this
https://en.wikipedia.org/wiki/Amine_gas_treating

or
https://en.wikipedia.org/wiki/Membrane_gas_separation
might work.

You might even be able to reclaim the SO2 for whatever it is you are using it for.

Unfortunately I haven't found a system specifically designed for SO2. I thought a gas centrifuge might work because I found that gravity alone can separate SO2 from atmospheric air when I took gas measurement at various depth in an underground basin prior to confined space entry. I wonder why no such thing exists commercially.

[chiralSPO]

Of course there's also the Claus Process https://en.wikipedia.org/wiki/Claus_process

[hamdani yusuf (OP)]

You might be able to reduce it back to elemental sulfur using CO or H₂
https://pubs.acs.org/doi/abs/10.1021/acs.energyfuels.6b01006
https://www.sciencedirect.com/science/article/abs/pii/S0263876217301478

These still require stoichiometric reagent to quench the SO₂ (and a fixed bed reactor, that might not be cheap), but at least the product has positive value, not negative value.

As far as possible, I want to avoid introducing new dangerous substances such as H2S and H2. The site is already crowded by chemical processes, mechanical and electrical equipments which make it hard to prevent incidents.

[Bored chemist]

I thought a gas centrifuge might work because I found that gravity alone can separate SO2 from atmospheric air when I took gas measurement at various depth in an underground basin prior to confined space entry.

I'm afraid you are mistaken.
Gravity can't (usefully) separate the gases (Air and SO2).
If the gases are separated (and they start of that way in your process because the SO2 is inside the pipework, and the air is outside) then gravity can slow down the process whereby they mix.
So, you can get "pools" of dense cases (It happened a lot in breweries and the CO2 sometimes kills people in cellars and such).

But the pools are not caused by gravity  separating the dense gases out of the air.

Now, technically, I have to say that there is some degree of separation by gravity, but it is tiny.
The air on top of everest is still pretty close to 78% nitrogen.

[hamdani yusuf (OP)]

I thought a gas centrifuge might work because I found that gravity alone can separate SO2 from atmospheric air when I took gas measurement at various depth in an underground basin prior to confined space entry.

I'm afraid you are mistaken.
Gravity can't (usefully) separate the gases (Air and SO2).
If the gases are separated (and they start of that way in your process because the SO2 is inside the pipework, and the air is outside) then gravity can slow down the process whereby they mix.
So, you can get "pools" of dense cases (It happened a lot in breweries and the CO2 sometimes kills people in cellars and such).

But the pools are not caused by gravity  separating the dense gases out of the air.

Now, technically, I have to say that there is some degree of separation by gravity, but it is tiny.
The air on top of everest is still pretty close to 78% nitrogen.

There are some factors at work. Difference in molecular weight of the gases creates difference in buoyancy, which tends to separate them. Diffusion and agitation on the other hand, tend to mix them.
In the case of everest top, the molecular weight of nitrogen is similar to air, hence the separating factor is very small compared to mixing factors which come from diffusion and agitation by wind.
If we release Helium, it will tend to go up to upper atmosphere, and finally blown up by solar wind.

Note that SO2 is more heavy than CO2.

[Bored chemist]

If you were right about dense gases settling to any meaningful extent, then you could do that "put out the candle" trick with any glass. It would be filled by CO2 "raining" out of the air.
Mind you, by the same effect, we would be dead.

You don't seem to have understood my point.
Differences in density can maintain a pool of dense gas (for a short while) but they can not create one.
(or, they can, to a tiny extent).

Difference in molecular weight of the gases creates difference in buoyancy, which tends to separate them.

If you calculate how small that effect is, you will see my point.

Even in  a sealed container with no temperature gradient, you won't get a useful degree of separation.

[hamdani yusuf (OP)]

Here is a diagram of gas centrifuge


https://www.princeton.edu/~aglaser/2008aglaser_sgsvol16.pdf

I know that the machines involve very high g-force to separate the uranium isotopes due to very similar molecular weight. With bigger difference of molecular weights in SO2 vs air, I infer that the required g-force is significantly lower.

[hamdani yusuf (OP)]

Even in  a sealed container with no temperature gradient, you won't get a useful degree of separation.

What do you think will happen if I put a mixture of Helium, Nitrogen, and SO2 with equal volume and pressure inside a 10 meters vertical pipe. Will we get the same composition between top and bottom part?

[chiralSPO]

Even in  a sealed container with no temperature gradient, you won't get a useful degree of separation.

What do you think will happen if I put a mixture of Helium, Nitrogen, and SO2 with equal volume and pressure inside a 10 meters vertical pipe. Will we get the same composition between top and bottom part?

That will depend on the temperature, pressure, and gravity.

But assuming that you mean Earth gravity (9.8 m•s^–2), 1 atm, and 300 K (27 °C), then the difference in gravitational potential energy between having 1 mole of He at the very top of the 10 m tube and 1 mole of SO₂ at the very bottom, and having 1 mole of He at the very bottom of the 10 m tube and 1 mole of SO₂ at the very top is 9.8•10•(0.064–0.04) = 5.88 J/mol. This could be viewed as the maximal enthalpy of gravitational un-mixing in a 10 m column (being favorable by 5.88 J/mol, so ΔH > –5.88 J/mol).

But there is also an entropic cost. http://hyperphysics.phy-astr.gsu.edu/hbase/Therm/entropgas.html
The change of entropy of an ideal monoatomic gas with change in volume is (ideal gas is a good model for He, not so great for SO₂, but doesn't really matter here... the errors will cancel):

ΔS = N•k*ln(V_f/V_i)

so if we had 1 mole of each gas being restricted to 1/2 the volume of the tube (from having had the whole tube), then for each mole of gas we would have:
ΔS = N•k*ln(1/2) = –5.76 J•K^–1•mol^–1
and therefore ΔS = –11.52 J•K^–1•mol^–1 (because we are restricting two moles of gas, the He and the SO₂)

thus at 300 K:

ΔG_un-mixing = –5.88 J•mol^–1 – (300 K)(–11.52 J•K^–1•mol^–1 ) = +3450 J/mol

(recall that ΔG must be negative for a process to be spontaneous)

[hamdani yusuf (OP)]

and 300 K (23 °C),

300K = 26,85 °C.
My question contains 3 substances, but I can omit the nitrogen to simplify the calculation.
Can you tell me how much helium will be in the top half of the pipe, how much SO2 in bottom half of the pipe after reaching equilibrium?
How much g-force is required to get 75% SO2 in the bottom half of the pipe?

[chiralSPO]

and 300 K (23 °C),

300K = 26,85 °C.

Whoops, sorry, typo! Meant 27 °C (I'll change it).

My question contains 3 substances, but I can omit the nitrogen to simplify the calculation.
Can you tell me how much helium will be in the top half of the pipe, how much SO2 in bottom half of the pipe after reaching equilibrium?
How much g-force is required to get 75% SO2 in the bottom half of the pipe?

I don't think having more than two substances helps, and it's certainly harder to calculate.

To a first approximation, I think you would need to increase to about 600 times earth gravity (based on the calculations provided in my previous post). This isn't impossible, but I think it would be harder to scale that than, for example distilling them (He bp = 4.2 K; N₂ bp = 77 K; SO₂ bp = 263 K)

[hamdani yusuf (OP)]

I finally found the research closest to my need.
http://www.mate.tue.nl/mate/pdfs/5250.pdf

Separation of carbon dioxide and methane in continuous countercurrent
gas centrifuges
Ralph van Wissena, Michael Golombokb, J.J.H. Brouwersa,∗
aDepartment of Mechanical Engineering, Den Dolech 2, TU Eindhoven, 5600 MB Eindhoven, The Netherlands
bShell International Exploration and Production, Kessler Park 1, 2288 GS Rijswijk, The Netherlands
Received 19 November 2004; received in revised form 1 March 2005; accepted 3 March 2005
Available online 3 May 2005
Abstract
The goal of this study is to determine the order of magnitude of the maximum achievable separation for decontaminating a natural gas well using a gas centrifuge. Previously established analytical approximations are not applicable for natural gas decontamination. Numerical simulations based on the batch case show that although the separative strength of the centrifuge is quite good, its throughput is very limited. Both enrichment and throughput are only a function of length and peripheral velocity. A centrifuge with a length of 5 m and a peripheral velocity of approximately 800 m/s would have a throughput of 0.57 mol/s and a product flow of 0.17 mol/s. These numbers are calculated with the assumption that the centrifuge is refilled and spun up instantaneously. The results for the countercurrent centrifuge show how the production rate varies as a function of internal circulation, product–feed ratio, peripheral velocity and centrifuge length and radius. Under conditions similar to those of the batch case the production is approximately half compared to the batch case, i.e., 0.08 mol/s. Optimization can yield a higher production at the cost of lower enrichment. Considering the current natural gas prices and the low production rate of the centrifuge, it is certain that the gas centrifuge will not generate enough revenue to make up for the high investment costs.
2005 Elsevier Ltd. All rights reserved.

[Bored chemist]

Even in  a sealed container with no temperature gradient, you won't get a useful degree of separation.

What do you think will happen if I put a mixture of Helium, Nitrogen, and SO2 with equal volume and pressure inside a 10 meters vertical pipe. Will we get the same composition between top and bottom part?

Very Very nearly.

Fundamentally, we need to consider two terms.
How much energy does a mole of SO2 release by sinking to the bottom (and letting the air rise).
How much thermal energy does that same SO2 have.

Say the pipe is a metre high and initially full of 50:50 SO2 in air.
If all the SO2 "settled" then it would (at the same pressure)  move down from (on average) 0.5 metres to (again, on average) 0.25 metres
a Mole of SO2 has a bass of about 64 grams.
And it falls (on average) by 0.25 metres.
In doing so it releases  potential energy equal to Mgh
0.064 *9.81*0.25
That's 0.157 J
And, in doing so it lifts the air- which takes (by a similar calculation)
0.029 *9.81*0.025
That's 0.071 J
So the net energy released by letting it settle is
0.157-0.071 =0.085 J/ mole

And the (thermal) kinetic energy of a mole of SO2 is 3/2 RT
where R is the gas constant- about 8.31 J/mol/K

So, with a temperature of about 300K, the energy is about 2500 J/ mole
So the thermal energy is roughly 2500/ 0.085 i.e. about 30,000 times bigger than the gravitational energy. We can see there's not going to be much separation.

We can now use the Boltzmann distribution to see what the proportions of the material will be in the upper and lower energy states (corresponding to the upper and lower halves of the tube).

The Botzman factor is given here
https://en.wikipedia.org/wiki/Boltzmann_distribution

Boltz.png (2.68 kB . 157x92 - viewed 11580 times)

e_j and  e_i are the gravitational energies in the upper and lower states.
And the difference between them is 0.085 J/mole

KT is 2500

So the Boltzmann factor is exp(0.085/2500) which is 1.000034
For every molecule in the upper half of the tube, there will be 1.000034 in the bottom half.

It's slightly less bad with a 10 metre pipe. Exp((0.85/2500) roughly 1: 1.00034
So, 50.0085% of the SO2 would be in the lower half of the tube and 49.9915 % in the top.

Speaking as an analytical chemist, there's no way you could measure that difference.

Typical gas centrifuge rotors are of the order of 50 millimetres in radius and spin at something like 60,000 revolutions per minute.
The calculation here
https://en.wikipedia.org/wiki/Centrifuge
gives the "acceleration" as
1.1118 * r/1000000 * n^2
where r is the radius and n is the number of revolutions per minute.
In the case of a gas centrifuge that gives something like
1.118 * (50 /1000000) * 60000^2
About 200,000 g

It's tricky to build stuff that will survive forces  200,000 times its weight.

Even with those sorts of figures, the gas centrifuges used for isotope enrichment  are not good enough.
You need cascades of them in series.
But, as you say, the SO2/ air case is much easier than U₂₃₅F6 vs U₂₃₈F6

If we had one of those centrifuges then the energy difference between "top" and "bottom" would be 20 times less- because of the difference in "height"- 50mm vs 1 metre)  But 200,000 times more because of the increased acceleration.
So that's a 10,000 fold improvement overall.
So the exponential factor is improved from about 30,000 to about 3
exp(0.333) is about 1.4
So your gas stream containing 0.2%  would be split into a rich stream containing about 0.28% and a "clean" stream containing about  0.14%.


I don't know what the pollution control limits are where you are, but I will guess that they might let you vent a waste stream with 100 ppm v/v
And your current effluent is 0.2% which is 2000 ppm
So you need a 20 fold reduction.
You can get that by cascading centrifuges. each one gives a reduction  of SO2 by  a factor of about 1.4 So, about 9 stages should do it.

Then we need to look at running costs.
In order to work, the gas gas to get spun up to high speed. and then slowed down again when it leaves as either  the enriched or depleted stream.The tangential speed of a centrifuge like this is about mach 2
So, you need to figure the cost of raising all your effluent gas to mach 2 (then slowing it down again) 9 times.
Each tonne of gas will need to be whizzed up to 700 m/s- which takes 1/2 *1000*700^2 Joules of energy
That's 245 MJ per tonne.
Nine times
2.2 GJ per tonne.
So one tonne per second would take 2.2 GW
Or 1 tonne per day would need 25.5 KW
24 Hrs at that rate, if it was on my electricity bill would cost about £100

Is your product worth £100 per tonne?



Almost anything is a better option than a gas centrifuge