Sciencemadness Discussion Board

Photo-Birkeland-Eyde reactor under pressure?

clearly_not_atara - 29-4-2020 at 10:49

Somewhat shockingly, condensed-phase N2O5 has a negative enthalpy of formation.

https://en.wikipedia.org/wiki/Dinitrogen_pentoxide
ΔfH⦵298 = −43.1 kJ/mol (s)

This suggests that if a mixture of N2 and O2 were at high enough pressure and low enough temperature, and the conversion could be catalyzed somehow, the reaction 2N2 + 5O2 >> 2N2O5 should become spontaneous.

The bond energy of dioxygen is about 5 eV which corresponds to 250 nm UV-C, so irradiation of oxygen gas with 250 nm UV-C should split oxygen into its atoms.

In total, the suggested apparatus comprises a high-pressure chamber filled with air and illuminated by 250 nm UV-C light and continuously cooled. Because the equilibrium (in theory) favors the products, much higher yields might be possible than with the conventional electric-arc method. Cooling the apparatus is essential because the UV-C illumination causes heating. Additionally, the pressure drops as N2O5 forms, making the reaction less favorable (maybe still better than 4%ish yield from B-E process).

Is this madness?

The partial reaction steps are (roughly):

N2 + O* >> N2O + ~170 kJ/mol
N2O + O* >> 2 NO + ~150 kJ/mol
2NO + O2 >> 2 NO2 + ~150 kJ/mol (no radicals required)
2 NO2 + O2 >> 2 N2O5 (s) + ~100 kJ/mol

If the apparatus were designed so that N2O5 would "fall down" away from the illuminated region you might be able to prevent the UV-C from destroying the product.

Somewhat conveniently, the bond energy of nitric oxide is 626 kJ/mol which corresponds to about 200 nm so if the UV-C is longer than this, the re-dissociation of nitric oxide is not likely.

[Edited on 29-4-2020 by clearly_not_atara]

Heptylene - 29-4-2020 at 12:41

What about the entropy of reaction? I'm a bit rusty on thermodynamics, but to form N2O5 your have convert 7 moles of gas into 2 moles gas.

The Gibbs free energy of reaction is $$\Delta_{r}G^{o} = \Delta_{r}H^{o} - T\Delta_{r}S^{o}.$$ which is negative for a spontaneous reaction. For the proposed reaction you would have (at atmospheric pressure) $$\Delta_{r}S^{o} = S^{o}_{products} - S^{o}_{reactants} = - 401 J K^{-1}mol^{-1},$$ which will make your Gibbs free energy positive unless you cool down the mixture a lot. Down to 107 K to even have $$\Delta_{r} G^{o} = 0.$$ Of course this assumed atmospheric pressure, so maybe pressure would help enough? Interesting idea anyway!

Now as I said I'm a bit rusty, if someone could check that would be nice.

EDIT: anyone knows how to format math expressions inline (i.e. not as separate equation/paragraphs?

[Edited on 29-4-2020 by Heptylene]

[Edited on 29-4-2020 by Heptylene]

FranzAnton - 30-4-2020 at 05:55

What about the strong triple binding within the nitrogen molekule? I found 945kj/mole to break it up. So it's not sufficient to break up the oxigen molekule.
So if you move from UVC into the area of cosmic radiation as it is in the upper atmosphere so you have NOx formation.

So if you can produce hard X-ray pulses in a high pressure air atmosphere ...
Seems the conversion rate depends on the irradiance of the "light" source.
If you really have to work in the X-ray area instead of UVC the efficiency is -hm- well known.

Maybe someone remember the coulor pics of atomic bomb explosions, the typical brown NO2 clouds are prominent. ---> you have a pressure front and high radiation leve of short wave length....

Heptylene - 30-4-2020 at 13:02

Quote: Originally posted by FranzAnton  
What about the strong triple binding within the nitrogen molekule? I found 945kj/mole to break it up. So it's not sufficient to break up the oxigen molekule.
So if you move from UVC into the area of cosmic radiation as it is in the upper atmosphere so you have NOx formation.

So if you can produce hard X-ray pulses in a high pressure air atmosphere ...
Seems the conversion rate depends on the irradiance of the "light" source.
If you really have to work in the X-ray area instead of UVC the efficiency is -hm- well known.

Maybe someone remember the coulor pics of atomic bomb explosions, the typical brown NO2 clouds are prominent. ---> you have a pressure front and high radiation leve of short wave length....


Really interesting idea to use x-rays. Pretty dangerous experiment though. If you have an x-ray source that has enough flux to break all the bonds in a sample in a reasonable time, then it can also do the same to your body and give you cancer, kill you, or even turn your flesh to mush.

A rough calculation: if you need one 10 eV photon to break one bond, and you want to break about 1 mol of bonds to make your N2O5, that's 963 kJ worth of x-ray photons.

To put that into perspective, a lethal full-body dose of radiation is a few Grays (J/kg of matter), so a normal human would need to absorb only a few hundred joules of radiation to die. So this mole-scale experiment would deal about a thousand lethal doses worth of x-ray photons.

An electric arc is much simpler and much less dangerous and essentially does the same using electron impacts I think? But I do not know whether the bonds are actually broken in an electric arc.

Sulaiman - 30-4-2020 at 14:25

AFAIK the Birkeland-Eyde process is just a method of exposing the nitrogen to >3000oC

maybe a large parabolic reflector or Fresnel lens could be used to focus sunlight
at a point inside a large transparent container ?
(possibly with external mirroring of the container, except for the light acceptance angle)

FranzAnton - 30-4-2020 at 14:46

uh... that sounds nearly as dangerous as the covid19 eh? Better not to deal with. We sould set up a price for an "old school" arc design which will yield over 10% NO in the exhaust air after the arc chamber :)

clearly_not_atara - 1-5-2020 at 06:50

Quote:
Down to 107 K to even have delta-G = 0


Yes, that is a problem. But as you mentioned, higher pressures change the thermochemistry. In about a week I will have time to study the thermochemistry but I don't remember off the top of my head how to incorporate pressure in a delta-G equation.

As a simple guess, maybe the ideal gas law applies? We can decrease the temperature (in Kelvin) by a factor of 3, or instead we can triple the pressure. If the rxn becomes spontaneous at 3 atmospheres that would be quite achievable. However, this estimate is probably at least a little wrong.
Quote:
What about the strong triple binding within the nitrogen molekule?


The reaction of dinitrogen with a free oxygen atom to give nitrous oxide is favorable to the tune of 170 kJ/mol. So the initiation of reactions with nitrogen does not require splitting nitrogen directly.

To that point, nitrous oxide formation is commonly observed in air-fed ozone generators, even though these are carefully designed so that the discharge will split only oxygen and not nitrogen. Oxygen radicals do all the work.
https://www.tandfonline.com/doi/abs/10.1080/0191951870855233...
Quote:
So if you can produce hard X-ray pulses in a high pressure air atmosphere .


That's a bit overkill. One very useful bit of information is that 1 kJ/mol is about 0.1 eV/molecule. So the bond energy of N2 is about 10 eV, which would have a breaking frequency of about 120 nm aka "far UV" (UV-C ends at 200).

I suppose it's convenient that my day job involves radiation therapy. Unfortunately radiation in the range 10-1000 eV is strongly attenuated in all media and furthermore is not very easy to produce. X-rays at higher energies, by contrast, will pass through almost anything, including the human body (which is their purpose), and so only a small fraction of the energy will be deposited in the gas-filled chamber.

FranzAnton - 12-5-2020 at 07:03

ok, so that brings me to ja quite "simple" solution. With a microwave induced plasmaball burning in a sheric reactor feed tangentially with the N2, O2 mixture, an external magentic field compresses the plasma until the reaction takes place. The N2O5 as fallout will be catapulted towards the äquator of the spere. For controlling the 757 electromagnets around the spere we can set up a small and fancy arduino uno project so that all the hackers here can also have fun.
If it does not work from the beginning, we finally can plunge it all in liquid nitrogen (for the poor of us) the more advanced members will do thar with He II.
With a little bit of good will we can drive the price of 1g HNO3 in the range of 1g antimatter :D:D:D

clearly_not_atara - 12-5-2020 at 08:24

None of that is necessary.

The primary loss mechanisms would be recombination of oxygen radicals, the reaction N2O + hv >> N2 + O, and UV escape from the apparatus. What would be ideal is to coat the inside of the reactor with a very thin layer of gold to confine the UV-C without attenuation. Recombination of O* might be prevented by using a higher nitrogen admixture but the standard 4:1 in air is probably fine.

Energy efficiency is probably bad, but I thought it was interesting. It's far beyond my construction ability though.

FranzAnton - 12-5-2020 at 08:38

Hm, the ability and the time... but here (on paper) it looks quite easy to get a stable plasma flame in air out of a modified household microwave oven. Pls. see that document. Would be really interesting to build this, but there are the time and then the ability. By the way this kind of assembly has alredy a patent pending.:D

Attachment: design-and-construction-of-2-45ghz-microwave-plasma-source-at-atmospheric-pressure.pdf (769kB)
This file has been downloaded 524 times

[Edited on 12-5-2020 by FranzAnton]