Sciencemadness Discussion Board - Ammonia by the haber process--tabletop chemistry

Ammonia by the haber process--tabletop chemistry

agorot - 10-5-2010 at 19:01

Although ammonia is readily attainable from hardware stores with ammonium sulfate and a strong base, I was curious when I read this on wikipedia:

Quote:

The enormous technical problems associated with the process were first solved by German chemist Fritz Haber (with the invaluable help of Robert Le Rossignol, who developed and built the necessary high-pressure devices). They first demonstrated their success in the summer of 1909, producing ammonia from air drop by drop, at the rate of about a cup every two hours. The process was purchased by the German chemical company BASF, which assigned Carl Bosch the difficult task of scaling up Haber's tabletop machine to industrial-level production.

I've searched online and can't seem to find a schematic for this "machine." It must have been pretty intense though.

I found this book that explores the topic, but I have no idea if it actually goes into the process that haber origionally used instead of just exploring the history.
"The Alchemy of Air: A Jewish Genius, a Doomed Tycoon, and the Scientific Discovery That Fed the World but Fueled the Rise of Hitler"

The following are also good websites
http://haberchemistry.tripod.com/
http://en.wikipedia.org/wiki/Haber_process (note it seems the original hydrogen source was electrolysis of water, something easy to do)

Now everyone, lets not just shoot this down immediately. If your first inclination is to warn me, "some idiot kid that doesn't know what he must be doing", about safety, then please just don't respond. We're just exploring the chemistry. I'm aware of the safety risks/dangers of the chemicals, pressures, and temperatures involved. Be constructive.

[Edited on 11-5-2010 by agorot]

[Edited on 11-5-2010 by agorot]

roamingnome - 10-5-2010 at 19:13

Bild_Ammoniak.jpg - 24kB

I believe it is quite possible. You may have seen pictures of the original apparatus.

The French techniques have worked with higher pressures up to 10,000 psi in small diameter tubes which give a larger single pass yield of ammonia.

Technology these days is left for the large corporation that will safely deliver products to the consumer.

There is alot of details to consider of course

agorot - 11-5-2010 at 15:21

I don't suppose anyone is aware of any schematics available online?

grndpndr - 11-5-2010 at 21:47

Not here to criticize! are you considering an attempt to duplicate
the method if not the exact apparatus used? Looks fascinating but i would expect the need of a very good milling machine,shaper and lathe combined with the talent of a tool and die maker as well as heat treat equipment.Quite an expensive /difficult undertaking and possibly quite dangerous given the working pressure.I understand the appeal however.Wish i could provide concrete assitance but far beyond me.good luck in any endeavor agrogrot!

agorot - 12-5-2010 at 11:40

yeah, you may be right. I do want to find out first though if it is completely out of the question. I'd be willing to throw down a couple hundred bucks, but that may not cover it. it would be awesome to be able to make something like this, especially because it is on such a small scale

densest - 12-5-2010 at 12:40

The various web sources say 400-450C and 200bar = 3000psi ... I've done a little plumbing for 150bar O2 at 20C and that made me nervous enough!

The famous tabletop experiment had to have been fed from external high pressure and heated somehow. The process is exothermic - I don't think it's self-sustaining. There also must have been insulation covering the hot part of the process as well.

A hobbyist version using tanks of H2 and N2 from the local welding shop would take care of the pressure. 100bar would be enough to show it working. Heating the apparatus with an electric mantle would work. Cooling the output below 110C under pressure would liquefy the NH3 and allow separation, but much more cooling would be required to keep it liquid under atmospheric pressure so it could be seen.

2000psi plumbing parts can be purchased for only mildly extortionate prices. Depending on them at 400C is something else again. Keep the enclosed volume as small as feasible - 1-2mm ID tubing, for instance. The reaction (catalyst) chamber might be a piece of 2000psi iron pipe?

agorot - 12-5-2010 at 13:04

Quote: Originally posted by densest

The various web sources say 400-450C and 200bar = 3000psi ... I've done a little plumbing for 150bar O2 at 20C and that made me nervous enough!

The famous tabletop experiment had to have been fed from external high pressure and heated somehow. The process is exothermic - I don't think it's self-sustaining. There also must have been insulation covering the hot part of the process as well.

A hobbyist version using tanks of H2 and N2 from the local welding shop would take care of the pressure. 100bar would be enough to show it working. Heating the apparatus with an electric mantle would work. Cooling the output below 110C under pressure would liquefy the NH3 and allow separation, but much more cooling would be required to keep it liquid under atmospheric pressure so it could be seen.

2000psi plumbing parts can be purchased for only mildly extortionate prices. Depending on them at 400C is something else again. Keep the enclosed volume as small as feasible - 1-2mm ID tubing, for instance. The reaction (catalyst) chamber might be a piece of 2000psi iron pipe?

200 bar is what is used on the industrial scale, but if you read on, haber's machine produced only about a cup of NH3 per hour, which leads me to think that in his design, he used high pressure, but not quite that high, and not quite that hot either. I'm not sure though, that's why I'm trying to find out more about his original design.

You're right that the rxn is not self sustaining. It does need to be heated, unless haber origionally did it at room temp? Since it is exothermic, LeChatlier tells us that actually lower temperatures are more efficient, but that does not mean that they will produce more NH3. Perhaps Haber was going for efficiency in his earliest design so he did it only at room temp.

I've read that the catalyst is actually magnetite, so iron ion...

JohnWW - 12-5-2010 at 13:30

As well as high pressure, it would, ideally, require a transition-metal catalyst which absorbs large volumes of N2 and H2 at such pressures, thereby becoming the reaction substrate. Pd would be ideal for this, although at such high pressures, much cheaper Ni can be made to substitute for it, and probably others such as Fe.

[Edited on 12-5-10 by JohnWW]

watson.fawkes - 12-5-2010 at 15:25

Quote: Originally posted by agorot

I'd be willing to throw down a couple hundred bucks, but that may not cover it.

Thank you for mentioning budget relatively early in the discussion. It's critical for any sane discussion of small-scale engineering. I wish, largely in vain as of now, that this were the norm. I really doubt that $200 will be enough. It's enough to cover some personal protection gear, though, so it's enough to get started. At the very least you're going to want a respirator with an ammonia-specific cartridge.

The bulk of the cost of such a unit is going to be the pumps and compressors involved. The N2/H2 recycling loop is one example. You've got to provide some constant positive pressure differential across that pump, even if your flow rate isn't that high. The bigger practical problem is a pump that won't explode. You've got three relevant pressures here: one each before and after the catalyst bed and the third is atmospheric. Even if you only need (say) 1 bar difference across the catalyst bed, you've got the full 100-200 bar difference from the innards of the pump to the outside. A bottled-gas compressor, mostly easily available for scuba tanks, can do this for inert gas, but they aren't cheap. (Even then, they'd likely need new, chemically-compatible seals.)

An alternate way of getting an adequate pump would be to stick an ordinary pump inside a pressure vessel. It's all about ΔP, so if you eliminate that, a cheap pump might work. The only problem is that you need a pressure vessel, one in two parts with a pair of flanges that you can disassemble. Depending on what industry there is where you are, this might be available as scrap or surplus. Unless you get very lucky, you'll likely need to weld in this scenario.

Other costs are special materials for chemical compatibility. You've got hot hydrogen, which embrittles many common steels. Look up "hydrogen embrittlement" for more on the subject. It's a pretty good way to get a sudden explosion with flash-off of a lot of hot ammonia; not fun. Please do check that you've got compatibility at operating temperature, not just at room temperature. You might be able to get away with a material with marginal service life, but if you're willing to tolerate disassembly and inspection relatively frequently.

Personally, I'd recommend building and testing the apparatus in components. Take the separator that extracts NH3 from the exhaust from the catalyst bed. A version of this that actually worked, even if fed with a test gas mixture (bottled H2, N2, NH3, for example), would constitute progress in my book.

roamingnome - 12-5-2010 at 19:35

I've aquired a number of papers about ammonia
but this seems like it will have some real parameters to start chewing on

In reality ammonia synthesis is changing to the future.....
proton conducting ceramics coated with palladium produce ammonia at normal pressure..ill try to dig up the european patent its quite interesting and doeable..

Attachment: hn3jpdf.pdf (256kB)
This file has been downloaded 1165 times

[Edited on 13-5-2010 by roamingnome]

Attachment: EP0972855B1.pdf (82kB)
This file has been downloaded 1042 times

i hope these "chip" style generators become more common place. Im always a long way off, but i did get me some cerium oxide, but the company just kept draggin its butt on the
Yb nitrate. anyway good luck

[Edited on 13-5-2010 by roamingnome]

chief - 13-5-2010 at 02:59

I wonder how high the pressure can be driven with a refrigerator-compressor: If it's opened and the motor is substituted/hacked/better-cooled/etc. : Might it reach 50 bar ? How far above that ?

Of cource the entire experiment would have to be safely contained: Get some bricks and mortar, maybe 2 webcams ...

Piping can hold _lots_ of pressure as long as the diameter is small: Glass-tubes of 1 mm wall-strength and inner diameter of 2 or 3 mm will go up to 500 bar or above ... ; ideally quartz-glass would be used, which holds the pressure at high-enough temperature ...
==> The volume of the reaction would be distributed over lenghts of relatively thin tubing, laying side-by-side in the furnace ...
==> Of course: When a piece of the tubing blows it will take the rest with it ... !!! So: Bricks and mortar to contain it, and noone in the room since such a setup will act as a canon for some of the pieces, throwing them violently !!

The pressure-resistivity of tubing depends on the disruption-tension of the material:
==> The same wall-strenght will hold the same forces ...; with smaller-diameter tubing these forces are smaller, since the pressure then acts onto smaller surfaces ...
That's why a 5mm-tube will hold several times the pressure of a 1-cm-tube ..., at same wall-strength and material ...
==> Smaller plumbing will be aduequate for surprisingly high pressure ..., but safety first, as mentioned above ...

[Edited on 13-5-2010 by chief]

not_important - 13-5-2010 at 10:10

Well, let's look at the record

Quote:

This goal was realized by Fritz Haber in 1908[4, 5] when he was able to
produce ammonia in a substantial yield from nitrogen and hydrogen by the use of an
osmium catalyst at a pressure of about 175 bar and at 550 °C. Haber’s apparatus
produced about 90 grams of ammonia per hour with about 98 grams of the osmium
catalyst[1, 3, 6, 7]. Due to the cost of osmium, research into more economically
feasible catalysts was conducted. Then Alwin Mittasch of BASF in 1910 found that
adding alumina, Al2O3, and potash, K2O, to iron was a suitable catalyst for ammonia
synthesis.

At this point, there was a sizable gap between Fritz Haber’s apparatus and a
full-sized plant that would be able to mass-produce ammonia. The largest part of this
problem was the high pressure of reactant gasses that was needed to produce
ammonia. The technology to perform this high pressure reaction was developed by
Carl Bosch of BASF in 1913 who oversaw the first factory dedicated to ammonia
synthesis in Oppau, Germany[1, 8, 9]. This factory was able to produce 30 tons of
ammonia per day with 300 Kg of the doubly promoted iron catalyst with a total
pressure of 200 atmospheres of the nitrogen and hydrogen reactant gasses.

[1] Jennings, J.R., Catalytic Ammonia Synthesis Fundamentals and Practice.

[3] Topham, S.A., Catalysis, Science and Technology, ed. M.B. John R. Anderson. Vol. 7. 1985, Berlin: Springer-Verlag.

[4] Crookes, W., Report of the 68th meeting of the British Association for the Advancement of Science, Bristol, 1898. 1898, London: John Murray.

[5] Goran, M., The Story of Fritz Haber. 1967, Oklahoma: University of Oaklahoma Press.

[6] Haber, F., Z. Elek., 1910. 16: p. 244.

[7] Haber, F. and R.L. Rossignol, Z. Elek., 1913. 19: p. 53.

[8] Mittasch, A., Geschichte der Ammoniaksynthese. 1951, Weinheim: Verlag Chemie.

[9] Partington, J.R. and L.H. Parker, The Nitrogen Industry. 1922, London: Constable.

From the paper here

Model Catalysis of Ammonia Synthesis ad Iron-Water Interfaces - A Sum Frequency Generation
Vibrational Spectroscopic Study of Solid-Gas Interfaces and Anion Photoelectron Spectroscopic
Study of Selected Anion clusters

Ferguson, Michael James
12-15-2005

http://www.escholarship.org/uc/item/80v739hb

To me it sounds as if it should be the Haber-Mittasch-Bosch process.

Until recently iron-based catalysts were the only ones of interest, various such as using Rb or Cs in place of potassium, or adding some platinum, were the sort of thing that showed up. Recently a ruthenium-based catalyst has entered the field, operating at lower pressures and temperatures. It often is used in a multibed reactor with the first bed being conventional iron catalyst, which does well when the NH3 concentration is low. The Fe catalyst is also much cheaper than the Ru one, so any poisons (halogens, S, As, P) that get into the feed streams are going to trash it first.

The reaction is exothermic, reactor designs use various methods of handling the heat. The designs are such that the temperature rises from ~400 C at the injection point to ~500 C at the next cooling stage. In a very small reactor this would be less of a problem due to the decreased ratio of volume to surface.

There have been recent developments on both the traditional Haber H2-N2 method, and several electrochemical method, some of which use N2 and steam as their feedstocks. High temperature electrolysis can yield high purity H2 at lower energy costs than conventional electrolysis, in part by using the heat created by the N2-H2 reaction in steam generation.

A paper related to the European patent roamingnome supplied
http://dx.doi.org/10.1006/jcat.2000.2877

a variant of that
http://dx.doi.org/10.1016/j.ssi.2005.07.018

at 50 bar using convention iron catalyst plus proton conductor
http://pubs.acs.org/doi/abs/10.1021/jp002236v

Some non-technical overview papers on solid state NH3 synth methods

Attachment: HowardUniv1.pdf (747kB)
This file has been downloaded 1071 times

Attachment: SSAS_Oct2007_Final.pdf (210kB)
This file has been downloaded 878 times

A research paper on a plant using ruthenium catalyst at ~100 bar

Attachment: nh3syn97.pdf (168kB)
This file has been downloaded 2402 times

United States Patent Application 20090202417 is for a nanoparticle form of the iron catalyst, claimed to work as low as 10 bar.

And an oddball paper, I sense a lot of handwaving and buzzwording, using a Mn-Zn-Fe ferrite at STP in a microreactor with a magnetic field. This may be closer to running you automobile on water, than to anything useful, but hey - it's <s>Mad Science</s> Science Madness, you can try it yourself. :-)

Attachment: Electromagnetic - Microreactor Design for Ammonia Synthesis.pdf (803kB)
This file has been downloaded 1211 times

Polverone - 13-5-2010 at 11:07

not_important, it appears we've read many of the same publications about ammonia synthesis. The solid state electrochemical ammonia synthesis based on steam and N2 is the one that interests me most, but I don't think it has been demonstrated even on a pilot plant scale yet. I keep checking every 6 months or so for news.

I think the capability to use stranded renewable energy is alluring but it also raises more questions. How quickly can SSAS respond to changing availability of electrical power? What is the energy consumption including process heat and not just the electrochemical needs? How does thermal cycling from intermittent use dictated by intermittent power affect equipment lifetimes?

Another intriguing aspect of SSAS is that it appears amenable to scaling down much more than conventional ammonia synthesis. A future opportunity for opportunistic generation of NH3 at the very farm where the fertilizer is needed?

densest - 13-5-2010 at 11:10

The "Electromagnetic - Microreactor" paper leaves out one little tiny little detail - the excitation frequency. Also, the parameters for the coils just don't make sense. Assuming the coils are 10cm in diameter and 2 cm thick in both dimensions and DC excitation, the coils are not physically realizable. They show either 2 14 ohm in parallel or 2 3.5 ohm coils in series. In either case, there is no wire (except superconducting...

) which would have a low enough resistance and would fit (by a factor of 5 to 25 at least). AC power would just make things even more improbable.

The aluminum reactor body would have some interesting eddy currents to waste power if the excitation is AC.

The nanoparticle ferrite just might do something - I dunno.

Anybody want a magnet for your automobile fuel line?

Going back to the porous ceramics...

The EU patent doesn't specify thickness of the ceramic or its conductivity at the reaction temperature. The discussion implies a very high resistance since the current is very close to that required to perform the reaction alone. They don't report a voltage which would be very interesting - the power into the ceramic just might heat it enough to keep it hot. One would like to see a blank (N2 or H2 only) run. A thin platinum film (perhaps interacting with the ceramic in some unspecified way) might have substantial catalytic activity at 400-500C. They say that the rate is completely current dependent, though, but one can conceive a way that a heated catalyst constantly swept with reactants might work without any electrochemical mechanisms.

Just by inspection, it doesn't look like anything less porous than a zeolite would have enough surface area to support even the low reaction rate. I haven't calculated it. One might expect a reactor with N2 & H2 on one side under a couple of bars pressure and the product being vacuumed out the other side to make more sense.

Has anyone gotten the "Ionics" paper - reference 12 - to see how the plates were prepared?

10-8 mol/sec/cm2 is 0.6 g/hour/m2. That implies a lot of plates for even a demo version.

Looks like typical patent obfuscation.

S.C. Wack - 13-5-2010 at 11:31

Quote:

They first demonstrated their success in the summer of 1909, producing ammonia from air drop by drop, at the rate of about a cup every two hours.

Did this happen to feature osmium?
Or was it uranium?

I'm confident that relevant references to the journals including the contemporary work with le Rossignol will be found in Mellor, and many BASF patent numbers in Thorpe, and the patents will be found at espacenet.

I bet not a lot of people have made one.

EDIT: Wow, the French wikipedia is a lot better than the English one in this case. They make the one-paragraph history especially look like a joke. Do they have some form of IQ test to post on the internet?

[Edited on 13-5-2010 by S.C. Wack]

Formatik - 13-5-2010 at 11:51

The apparatus of Haber-Le Rossignol is below. Described in Fritz Haber by Margit Szöllösi-Janze, this is the device Boesch came to familiarize himself with in 1909 in Karlsruhe. There was a defective seal, so Boesch didn't see it in action then.

The reaction can be carried out at regular atmospheric pressure with basic chemistry set ups, but then the yield of ammonia will be significantly less. Haber and others have studied the effect of temperature and pressure on the reaction equilibrium. There should be some equations for this somewhere.

haber.png - 369kB

agorot - 13-5-2010 at 14:27

So I've just finished reading The Alchemy of Air. It was very informative, but sadly, it doesn't look like the Haber process is going to be performed in my house.

It seems that Haber was actually using 100-200 atm in his "little tabletop machine." The book says that this is pressure enough to crush most modern submarines. That seems believable to me.

The reaction was carried out in "thick-walled quartz tubing encased in an iron jacket." This tubing was made from a full quartz crystal, and a drill was used to hollow out a narrow cavity for the inner part of the tube, and then the outer part of the tube was made more cylindrical, and then it was surrounded with the iron.

When Haber built is very first prototype, it ran at 1000C. In 1908 he made several improvements to the design to allow it to run at higher pressure, so he only had to run it at 600C.

As for the catalyst:

Quote:

Haber tried powdered nickel, manganese, and platinum; none did the trick. So he turned to more exotic elements. One of the many consulting jobs Haber had done involved looking for ways to make better filaments for electric lights, and the company he had worked with had supplied him with a number of rare elements. He still had samples of many of them, and tried them in his ammonia machine.
In March 1909, one year after signing with BASF, Haber had a breakthrough. His team put a bit of one of his light bulb elements, a bluish black brittle metal called osmium, in the high-pressure chamber, heated it, and ran hot nitrogen and hydrogen over it. The yield of ammonia shot up.

The only problem is that osmium was way too rare. BSAF could not have economically produced synthetic ammonia in this form. Eventually though, Haber and Le Rossignol (another member of Haber's team) found that uranium also worked suitably well. It wasn't quite as good as osmium, but it was close enough.

That's what I've found about Haber's original machine. I think to build something like that it would take a couple of thousands of dollars, not a couple of hundred.

There is something else. Apparently, there was another chemist working for BASF, Ostwald (you may know the name from the Ostwald Process), that thought he had beat Haber to it. He was using a strait iron catalyst, or so he thought. In reality, he was using some very high purity iron wire for his experiments to make ammonia that had been used, and contaminated by, another chemist. His catalyst contained significant amounts of Iron Nitride, and actually he had just been heating iron nitride in the presence of hydrogen to make ammonia.

Does anyone know how to make/obtain Iron Nitride--either the (II) or (III) salts? Incidentally, I think it can be made if you react iron with, you guessed it, Ammonia.

[Edited on 13-5-2010 by agorot]

[Edited on 13-5-2010 by agorot]

agorot - 13-5-2010 at 14:36

Quote: Originally posted by S.C. Wack

Quote:

They first demonstrated their success in the summer of 1909, producing ammonia from air drop by drop, at the rate of about a cup every two hours.

Did this happen to feature osmium?
Or was it uranium?
[Edited on 13-5-2010 by S.C. Wack]

From my sources, Haber was using Uranium at that point.

Another chemist named Mittasch, working with Bosch, found in late 1909 that a catalyst of magnetite and aluminum oxide and calcium. Now other catalysts are available.

12AX7 - 13-5-2010 at 15:14

Steel is nitrided by a plasma/ion process. Not much better than Birkeland-Eyde, I'm afraid.

Tim

Nitrides

agorot - 13-5-2010 at 15:55

Lithium, specifically.

Lithium reacts with nitrogen in the air to form Lithium Nitride very quickly. It is a black compound. Lithium Hydroxide and Lithium Oxide are both white compounds. You know that lithium nitride forms much more quickly than the latter two compounds if you've ever cut a piece of lithium in half. It turns black quickly, although some lithium oxide and eventually hydroxide is formed. Then this happens:

(A) Li3N (s) + 3 H2O (l) → 3 LiOH (aq) + NH3 (g)

If only there was some way for this to happen:

(B) 2 Li3N (s) + 3H2 (l) → 6 Li (s) + 2 NH3 (g) <--NOTE: doesn't occur

I got really excited just a minute ago because I was thinking that you could possibly use equation (B), but then I did some research and supposedly, this one happens:

(C) Li3N (s) + 2 H2 (g) → LiNH2 (s) + 2 LiH (s)

Theoretically, if you could get equation (B) to work, instead of equation (C), your lithium would not be used up except for efficiency losses (your lithium would react with oxygen in the air or something if you had no way to purify nitrogen first to make lithium nitride).

The only other problem would be to deal with this equation:

(D) 2 Li + 2 NH3 → 2 LiNH2 + H2

Quote:

Lithium amide reacts vigorously with water to generate gaseous NH3. In experiments at Argonne National Laboratory, in which it was mixed with water and stirred at room conditions, about 23 percent of the theoretical yield of NH3 evolved as a gas in the first 0.6 minutes

from http://cameochemicals.noaa.gov/chemical/991

(E) LiNH2 + H2O → LiOH + NH3

(F) LiH + H2O → LiOH + H2

Equations (E) and (F) could be used if equations (C) or (D) were to occur so you would end up with lithium hydroxide.

Unless you could find a way to very efficiently decompose lithium hydroxide (melting point ~460C) into lithium metal, this process would not work well for a home chemist.

So there are two ways to make lithium work efficiently:

1) Find a way to reduce lithium hydroxide to metallic lithium efficiently
2) Find another metal that is CHEAP that acts as lithium does in that it reacts with nitrogen at atmospheric pressures and temperatures (or slightly above within reason), and you don't care if you have to use a lot of it because you are just going to end up with it's hydroxide and concentrated ammonia. That way you just use up a lot of that metal. Lithium is too expensive in my opinion to be used for obtaining large amounts of ammonia, but perhaps small amounts.... You will get less than a gram of lithium from a lithium battery, and they are not too cheap.

Sedit - 13-5-2010 at 19:22

"Lithium is too expensive in my opinion to be used for obtaining large amounts of ammonia, but perhaps small amounts.... You will get less than a gram of lithium from a lithium battery, and they are not too cheap. "

You get more then a gram of Li from batteries. Ill have to look at my notes but I think its around 1.2grams per battery. And any amount of ammonia from Lithium is to expensive considering Ammonia is in no way hard to make.

Quote:

The only other problem would be to deal with this equation:

(D) 2 Li + 2 NH3 → 2 LiNH2 + H2

This is no problem. Your reacting Li3N with H2 to make LiNH2 an LiH, there is no Li present at the formation of Ammonia. Your reacting this with H2O to generate the NH3. All extremely inefficient. Not to mention that Li + NH3 only forms LiNH2 after reacting with an alkene or heated, all requiring added expense.

To add more insult to injury wouldn't the theoretical reaction B occure on electrolysis of the melt(If you can get a melt) formed from the Nitride and H2. Sorry to stray further off topic but would Li Dissolved in Liquid NH3 and electrolysis performed cause the Lithium to act as a catalyst when N2 and H2 is the feed? Just curious as to why or why not.

[Edited on 14-5-2010 by Sedit]

JohnWW - 13-5-2010 at 23:57

Why use rare and expensive Li for such a purpose? You could just as easily use Na, although it does not slowly combine with N2 like Li, but instead reacts very quickly with it, and may possibly be induced to burn in it. You could also use K, which spontaneously ignites in N2, and besides K3N, may also form a pernitride K2N2 and supernitride KN2. However, with Na and K, you would have to safely contain the reaction somehow because of the large amount of heat rapidly generated.

not_important - 14-5-2010 at 02:18

Given that the quote I included from Model Catalysis of Ammonia Synthesis..., and references papers by Haber and Mittasch, I'll go with osmium being Haber's catalyst used in his demonstration in 1909, and Mittasch's development of the iron catalyst an outgrowth of BASF's interest in Haber's work and Ostwald's early discovery that hadn't panned out. The demo ran at 175 bar.

The ceramic in the Euro patent isn't porous, it's a Proton Conducting Ceramic - a subset of ionic conductors. In the device hydrogen is transferred through the ceramic as ions - protons - and delivered to the other side where it is converted back to hydrogen; depending on the electrode on the surface it can be effectively a lightly adsorbed hydrogen atom.

The use of lithium as a nitrogen to ammonia converter is close to the electrolytic method in one of those papers I uploaded. In that process a molten salt bath (NaCl-KCl-Li3N) is used, N2 is reduced to N3- at the cathode, the nitride migrates through the salt electrolyte to the anode where it reacts with the hydrogen migrating through the porous metal electrode. It looks to be at least as energy efficient as the H-M-B process using iron catalysts, doesn't involve messing about with elemental lithium outside of the original Li3N formation to make the electrolyte, at higher temperatures can use nickel electrodes.

Two patents on the process, IMHO PPP examples, for much detail one would have to chase down the journal papers referenced in the patents.

Attachment: 6881308_Electrochemical_synthesis_of_amm.pdf (122kB)
This file has been downloaded 1545 times

Attachment: 7314544_Electrochemical_synthesis_of_amm.pdf (217kB)
This file has been downloaded 1019 times

An updated paper to one of the solid state NH3 synth papers I included before is at http://www.energy.iastate.edu/Renewable/ammonia/ammonia/2008... at 4+ MB too large to attach here.

--------------------------------------------

From Technology and Manufacture of Ammonia (Strelzoff):

Code:



At 300 C using promoted iron catalyst
pressure(atm)   % NH3 in exit gases
100    8%
300    18%
500    27%
1000    47%
2000   65%

at 450 C
atm  % NH3 in equilibrium
1    0,209%
10    ~2%
31.6    ~6%
100   ~16%

so it looks as if using iron based catalyst at pressures below 200 atmospheres gives pretty low conversion, and pressures much under 100 atm too little NH3 production to be practical. Ruthenium based catalysts are more active, by that ref I gave earlier down to 100 bar and maybe lower. There's also hope that variants of iron catalysts might work that low.

---------------------------------------------

@Polverone - yeah, been into it for a few years I was Skypeing with an old friend, who brought in a chum of theirs who turned out to be connected withe several people very active in the stranded resources activities in North America.

Unfortunately, like far too many other renewable energy projects in the USA, the players seem to be more into PR and playing their cards close to their vests, rather than actually publishing real data. Nocera and the magic Co electrolysis catalyst, NanoSolar's undisclosed details, EEstore and the eternally slipping deadline, Aptera's not-yet-in-production-but-gamechanging BEV, there's so many 'breakthrough' companies out there that seemed to be able to make it through another round of ventural capital raising even as another deadline goes by or promissed specifications fail to appear.

There's a lot of interesting stuff out there that does get published. Using the nitrogen fixing system from cyanobacteria in a stabilizing matrix for NH3 synth, feeding eletricity to Methanobacterium palustre to convert CO2 to CH3; possible routes to making fuels and storing peak power. Wind power keeps creeping down towards the cost of coal power, photovoltaics are seeing developments on a number of fronts.

As for interfacing the newer ways of making NH3 to variable supplies of power - the problem is a bit like using SOFC for power generation What I've heard includes making the reactive elements in very small form to decrease thermal stresses, and either "banking back" subsets of the entire system to run at/use fuel at the lowest level in order to keep the cells hot enough, or cycle through running subsets of cells while the remainder slowly cools down, but becomes the active suset often enough to avoid needing a warm-up sstage.

agorot - 14-5-2010 at 05:11

Quote: Originally posted by Sedit

And any amount of ammonia from Lithium is to expensive considering Ammonia is in no way hard to make.

That's why I specifically said the only way to make this process efficient is to use a very cheap metal that is easy to obtain. At the very beginning of the thread, I specifically acknowledged that ammonia is easy to obtain from a strong base and an ammonium salt, but the whole purpose of this thread is to explore other options and (maybe) find another way that is also somewhat cheap to use the nitrogen in the air. Because of the availability of cheap ammonium salts, I don't think it will be easy to find a substitute for just making it with strong bases, but it is getting more and more difficult to find raw alkali-metal hydroxide. It is almost always mixed with something.

Quote: Originally posted by Sedit

To add more insult to injury

I'm not at all insulted Sedit, and I hope you wouldn't think that I would be. The whole purpose of this forum is for people like us to come together to talk about these ideas. I posted that equation you were referring to just so others could have it in mind when thinking about a possible lithium (or other metal) based system.

@JohnWW, you said:

Quote:

Why use rare and expensive Li for such a purpose? You could just as easily use Na, although it does not slowly combine with N2 like Li, but instead reacts very quickly with it, and may possibly be induced to burn in it. You could also use K, which spontaneously ignites in N2, and besides K3N, may also form a pernitride K2N2 and supernitride KN2. However, with Na and K, you would have to safely contain the reaction somehow because of the large amount of heat rapidly generated.

I didn't say we use lithium, in fact, this is what I said the following in my last post

Quote:

So there are two ways to make this work efficiently:

1) Find a way to reduce lithium hydroxide to metallic lithium efficiently
2) Find another metal that is CHEAP that acts as lithium does in that it reacts with nitrogen at atmospheric pressures and temperatures (or slightly above within reason), and you don't care if you have to use a lot of it because you are just going to end up with it's hydroxide and concentrated ammonia. That way you just use up a lot of that metal. Lithium is too expensive in my opinion to be used for obtaining large amounts of ammonia

Sodium and potassium aren't much cheaper though... I was hoping for something like zinc or aluminum or iron or copper or nickel, but I haven't done any research on these metals' nitrides and how difficult they are to make.

[Edited on 14-5-2010 by agorot]

chief - 14-5-2010 at 05:20

Where's the pressure-source ? What compressor ?
==> Maybe something can be constructed by using a high-pressure waterjet-cleaner, letting it pump something that displaces the gases in a cylinder ...
==> 200 bar means: 1 liter of air at standard-conditions get's compressed to 5 or 6 ccm ...

200 Bar will be held by a small-diameter glass-tubing; anyone can try that out by melting in some water and heat the ampoule safely in a furnace, measuring the temperature ... until ist burst's (or not) : From this temperature the pressure can be calculated ...

Mineralogists do high-pressure exp. regularly: The ampoule holds the pressure, but is inserted in a steel-autoclave for safety ...
==> From the filling-level of the ampoule (%) and the temperature the pressure is known ...
==> Thats how eg. hydrothermal conditions are simulated ...
==> The steel-autoclave gets some water-filling too, to build up a counter-pressure, so the ampoule does get some assistance for containing the 500 Bar it regularly will have ...

==================

[Edited on 14-5-2010 by chief]

agorot - 14-5-2010 at 05:27

What about this:

quoting wikipedia:

Quote:

Magnesium nitride can be produced by burning magnesium metal in a pure nitrogen atmosphere.
3 Mg + N2 → Mg3N2

Mg3N2 + 6 H2O → 3 Mg(OH)2 + 2 NH3

magnesium isn't that hard to find, and you could use the waste hydroxide if you have a stomach ache

chief - 14-5-2010 at 06:19

Fort this reaction Mg is either ignited or at least glowed:
==> A thicker layer of Mg-powder is used ...

The O2 is consumed first, so after the burning
==> the upper layers of the educt contain oxide,
==> while the lower layers are the nitride: Quasi the O2 is filtered out of the air through burning with the upper layers of the Mg ...

S.C. Wack - 14-5-2010 at 06:28

http://books.google.com/books?id=0ekNIaJX3-YC&pg=PA88

This book states here that 100 grams of osmium powder were used for a demonstration on July 2.

I wonder if this is the low hanging fruit.

watson.fawkes - 14-5-2010 at 07:06

Quote: Originally posted by S.C. Wack

http://books.google.com/books?id=0ekNIaJX3-YC&pg=PA88

This book states here that 100 grams of osmium powder were used for a demonstration on July 2.

I wonder if this is the low hanging fruit.

Thanks for posting this book reference. The beginning of the section you linked to, the one on the creation of the ammonia synthesis, stresses the importance of the high-pressure technology (pump, seals and values explicitly), to the extent that Haber thanked his engineer and machinist in his Nobel lecture. (Good for him.)

The importance of this high-pressure technology means that, for the amateur chemist, the apparatus itself is going to be the dominant expense involved in doing this. In contrast, in an industrial context, the operating costs (predominantly energy inputs) of a reactor dominate. The critical ratio is that between capital costs (the apparatus) and operating costs. These are different for amateurs and industry. Now the effect of a less efficient catalyst has an effect on the operating cost, but that has to be weighed against how much that machine is ever going to produce. A less efficient catalyst means that the recirculation ratio is higher, which means more pumping. So the trade-off is between the energy cost for the output over the life of the machine and the cost of the catalyst.

Therefore, I can't recommend osmium, expensive, over the iron-based (and other) catalysts for a small-scale machine, at least for a first catalyst. I make this recommendation on economic grounds, that you don't need to spend the money for the desired output. On the other hand, if you want to study catalysis itself, knock yourself out.

chief - 14-5-2010 at 08:09

The link to the book is very interesting ...
==> Steel embrittles from the H2, and vessels burst after a while ...
==> lining with silver or copper doesn't work ...

So maybe thin glass-pipes, within a strong containment (in case something fails), are the way to go ...
==> 1 kg of Quartz-tubes is at 20 € ..., various diameters; _hot_ welding-flame (with oxyen) is needed for handling ... ; good UV-shield is essential ...

==============

Still the pressure-source is needed ...

densest - 14-5-2010 at 18:33

@not_important - thanks for the clarification about the ceramic. Is it correct that the proton emerges as a single atom and thus is reactive enough to disturb the N-N triple bond?

If so, it's a real shame that passing air over an oxygen-conducting ceramic doesn't produce NOx. I need to learn how to calculate the thermodynamics again

mr.crow - 14-5-2010 at 19:03

It depends on how much ammonia you want to make. Maybe just for shits and giggles so you can sniff it.

A reaction tube, heat exchanger, radiator and circulation pump should be good enough. The ammonia will turn into a liquid at that pressure and can be separated to drive it forward.

According to wikipedia (yeah I know...) hydrogen embrittlement can be prevented by baking the hydrogen out of the metal within 4 hours.

And cheif, whats with the horrible grammar?

chief - 15-5-2010 at 03:35

Horrible grammar ? Thats worthy german grammar, refined version ...
==> ... german invention here too

===================

Maybe some sort of "periodic" pumping might do it:
==> The pressure built up to as much bar as the compressor can handle, filling a "tank" ...
==> Then closing the valve to the compressor, and heating ... : Gives extra-pressure

From 300 Kelvin to maybe 600 the pressure should double, to 900 K it should triple ...

Required would be then a compressor that can make 60-90 bar, a good valve, and a tank that can store the gases at 200 bar ..., and another valve that connects the tank to the conversion-unit ...

Would be a batch-wise compressing ... ..., batches as large as the tank ... ; tank would have to be somewhat larger than the conversion-tubing with the catalyst, so the pressure would not drop too much upon opening valve Nr. 2 (into the conversion-unit)

The pressure-rise upon heating could be made higher by cooling first, as long as the compressor works: Thereby more gas would fit into the tank, that could be heated for pressure-buildup ...

===========================

Ultimately a periodic thing, maybe at last even some car-engine could be abused: Compression only 10 bar, but many cycles per second ...
==> of a 4-cylinder-engine maybe 3 for the work, and 1 for the NOx/NH3/whatever-generation ...
==> lot's of people like hacking car-engines, someone could succeed, many would follow, could "democratize" the NH3-production, would be much safer for the lower pressures ...
==> Temperatures in the car-engine upon compression anyhow reach the high required values ...

[Edited on 15-5-2010 by chief]

watson.fawkes - 15-5-2010 at 06:43

Quote: Originally posted by densest

@not_important - thanks for the clarification about the ceramic. Is it correct that the proton emerges as a single atom and thus is reactive enough to disturb the N-N triple bond?

If so, it's a real shame that passing air over an oxygen-conducting ceramic doesn't produce NOx. I need to learn how to calculate the thermodynamics again

In an ionic-proton conductor, the proton does indeed conduct through the ceramic lattice as a "naked" ionic proton. Some of the proton-only conductors have lattice spacing small enough that atomic H can't form inside the lattice, not even as a metastable state. Other materials have a predominance of naked protons, but (almost certainly) with some population partition that includes atomic H. The ionization energy of H is 13.6 eV, just enormous compared to most chemical bond energies, and so it's very reactive. Indeed, I have to imagine that will react with the very first thing it encounters, which is almost always something adsorbed on the crystal surface. The problem, then, is that to disturb the N2 bond in any quantity, that N2 needs to be an adsorbed species. You can do this; there are, after all, the zeolites used for N2 separation. The bigger problem is that there's an electric field across this ceramic and, barring clever artifice, the predominant adsorbed species are going to be ionic, outcompeting the N2 for surface area. Presumably you want this transient N2(+) species to react with something, and that something is likely to ionize more readily than N2. It's not that this is all impossible, but it's also anything but obvious.

Furthermore, what you need is a material that adsorbs N2 preferentially and is also a proton conductor. So what you have is a materials science problem. In my alternate life that I am not living, I would be researching this class of topics; it's completely fascinating. As for research problems, I would not doubt that there exists some ceramic material that can produce NOx out of air. On the other hand, the search space of potential materials is, um, rather large.

(Alternately, you can use high pressure to force a higher adsorption population. The only problem there is that most of these ionic conductors only conduct ionically at rather elevated temperatures, say 600° C. That means you've now got mechanical problems just like the Haber process itself has.)

peach - 18-6-2011 at 17:44

Is that Swagelok he's got on there?

Nice thread.

I was reading around about it just out of interest, not really because I need ammonia. It would indeed be fun to copy his table top machine for the sake of seeing it again. It's a shame we didn't start the thread earlier on, then we could have possibly had one working by 2009, to mark the century on from his demonstration of it.

A fridge compressor will go over 500psi - I think I've seen one I took apart running at about 600 or 700. But you'd need filters in there to get all the oil out of the stream.

The pressure washer idea is good, but the cylinder would be the way to go; particularly if the washer is only going to be used to produce one charge of gas (not recirculate it). It does look like he's running a belt driven pump to charge / recirculate the gas in there. If one was doing it for the historical accuracy, a cylinder might be cheating.

I'd be happy with a cup per day, or less. Which means it could possibly be done with lengths of normal diameters of Swagelok. That would also lower the safety issues. As someone pointed out early on, that stuffs still expensive, but it is rated above those requirements.

{edit} In the wiki article, it says they demonstrated it working from air. So the pump is just running straight into that (not recirculating) and it's injecting air? Are they using zeolite to separate out the oxygen perhaps?

[Edited on 19-6-2011 by peach]

IndependentBoffin - 18-6-2011 at 23:13

Quote: Originally posted by roamingnome

Easy way to get high pressures at home is to scavenge a Diesel engine from an old car. Diesel engines have compression ratios (CR) of 14 - 23 to 1.

10,000 psi = 68.9 MPa

Assuming adiabatic compression:
P2/P1 = (V1/P2

γ

The following table therefore gives the theoretical ideal cylinder pressures for a diesel engines:
CR _Pressure
14 _40.2MPa
15 _44.3MPa
16 _48.5MPa
17 _52.8MPa
18 _57.2MPa
19 _61.7MPa
20 _66.3MPa
21 _71.0MPa
22 _75.8MPa
23 _80.6MPa

As you can see any Diesel Engine with a CR > 20 will get you the desired pressures. I wonder what happens if you take a 4 cylinder engine, modify the fuel intakes of 1 - 2 cylinders to accept a N2 + H2 feed and collect the ammonia at the exhaust?

For better efficiency you could even feed the exhaust from one cylinder into the adjacent one to give your N2 + H2 mixture two chances to produce ammonia

jimmyboy - 19-6-2011 at 00:42

I think your biggest problem will be getting the pressure - the heat can be supplied by nichrome and insulation - why waste all that money on compressors? an ordinary fridge compressor could never attain the pressures you need - an airtight steel piston and a hydraulic jack would work though - find a steel cylinder drill a hole for a schrader valve and one for a release and two more to pass the heat element inside - tap some threads and seal with brass parts/solder - inject the gas mix and crush with the jack - open the release for the ammonia - this will give you a small amount of ammonia at a time but not continuous - considering your budget this may be the only feasible way unless you are skilled in the art of metalworking - I would probably start looking for an old engine block for your piston and cylinder and test the rings/seal with a strong pressure guage - bore and adjust as needed..
with an old block the ports are already drilled

just some ideas

experimenter - 16-7-2011 at 09:53

What if you react nitrogen and hydrogen together with the help of an electric arc, like in the Birkeland–Eyde process?

AndersHoveland - 16-7-2011 at 11:23

Quote: Originally posted by agorot

Unfortunately this reaction doesn't occur:
(B) 2 Li3N (s) + 3H2 (l) → 6 Li (s) + 2 NH3 (g)
I did some research and supposedly, this is the reaction that happens instead:
(C) Li3N (s) + 2 H2 (g) → LiNH2 (s) + 2 LiH (s)

Theoretically, if you could get equation (B) to work, instead of equation (C), your lithium would not be used up

Reaction (C) is reversible at 270 °C.
When heated, lithium amide (LiNH2) decomposes to lithium imide (Li2NH) and gaseous ammonia (NH3).
(for what scarce information could be found about "sodium imide" you can read https://sites.google.com/site/ecpreparation/sodium-amide-imi... )

One would think that, with heat, nitrogen could be reacted with lithium hydride to form ammonia
(6)LiH + (2)N2 --> (2)Li2NH + NH3
and the lithium imide could be recycled by heating with hydrogen.
Li2NH + H2 --> LiNH2 + LiH

Quote: Originally posted by JohnWW

This is actually a common misconception. Interestingly, lithium is the only alkali metal which burns in nitrogen. The others, including sodium and potassium, are unreactive towards nitrogen, even when heated. This is somewhat paradoxical because sodium, and especially potassium, are generally much more reactive than lithium. True nitride ions (with three extra electrons and not covalently bonded) are not very stable. It is not favorable for three extra electrons to occupy the smaller outer orbital of a nitrogen atom. Lithium ions are slightly acidic and can covalently bond to nitrogen. In the crystal structure of lithium nitride the nitrogen atoms have only one extra electron, while being covelently bonded to two other lithium atoms. One of the reasons why oxygen and nitrogen prefer to exist as diatomic elements is the repelling from the lone pairs. With only two atoms bonded together, this gives the lone pairs more space. Other elements have few lone pairs, or a biger atomic orbital, which means more space for the lone pairs. As far as I am aware, compounds of pernitride and supernitride have never been prepared. The potassium ion would likely attract both electrons towards the closer nitrogen atom of a pernitride anion, leading to two of the anions dimerizing into (-2)N--N=N--N(-2). This intermediate would likely spontaneously decay into nitride ions and azide ions, if not elemental nitrogen. Because of the great stability of diatomic nitrogen, it is unlikely that nitrogen can form anions which would correspond to the peroxide and superoxide anions of oxygen. Azide anions are somwewhat of an exception because there are three electron-withdrawing nitrogen atoms to attract the extra electron, which resonates around.
(-)N=N(+)=N(-) <--> NΞN(+)--N(-2)

[Edited on 16-7-2011 by AndersHoveland]

Aperturescience27 - 19-4-2012 at 13:47

I apologize if someone's already suggested this, but there's one simple (yet not necessarily efficient) solution: an electrolytic cell with molten lithium hydroxide (M.P. 462 C) producing metallic lithium, which reacts with nitrogen to form lithium nitride, which reacts with water to form ammonia and lithium hydroxide. Consumes only nitrogen and water, produces ammonia and oxygen (oxygen at the anode). Design would be difficult, but not expensive (unless you had to use an expensive material for the anode, which would of course have to be resistant to corrosion by both oxygen and lithium hydroxide, and you couldn't use carbon because carbon dioxide would react with the lithium hydroxide and produce lithium carbonate), and you wouldn't need high pressures. Air could probably be used as a nitrogen source, because the lithium oxide produced would just end up as lithium hydroxide, and lithium doesn't produce peroxides or superoxides in air. Hopefully the lithium nitride would dissolve in the lithium hydroxide, because then the air or nitrogen input, which would go into the cathode compartment, and the water input/ammonia output could be separate. The oxygen would come off separately from the anode compartment. You might get some hydrogen at the cathode, from water reacting with lithium metal, which could be dangerous.

Also, AndersHoveland is right, other alkali metals don't form nitrides. Alkaline earth metals do, but their hydroxides have really high melting points.

bbartlog - 19-4-2012 at 17:19

Quote: Originally posted by Aperturescience27

there's one simple (yet not necessarily efficient) solution:

If the topic is the Haber process (as the thread title states), this is OT. If you want to restate the problem as 'how do I get ammonia', your solution is notable mainly for being *even less practical* than the tabletop Haber process idea.

Aperturescience27 - 21-4-2012 at 20:48

I'm not the only one who's OT. Just saying. Anyway, the Haber process is really better suited to large-scale production. You have to circulate the N2 and H2 a whole bunch of times, getting a little bit of ammonia out each cycle, because of equilibrium, and shifting the equilibrium to the right causes slower reaction rate.

Fluorite - 18-11-2020 at 03:38

Quote: Originally posted by experimenter

What if you react nitrogen and hydrogen together with the help of an electric arc, like in the Birkeland–Eyde process?

https://m.youtube.com/watch?v=8LriocEzpC0 They tried this but the arc can split ammonia again to N2 and H2
But this can be limited if you add acid like hcl to make nonvolatile salt?

symboom - 18-11-2020 at 05:41

Here is a small reactor
It uses iron wool as a catalyst
Nitrogen and hydrogen

https://www.youtube.com/watch?v=S3j8dCweN4A

[Edited on 18-11-2020 by symboom]

Fluorite - 18-11-2020 at 06:43

Quote: Originally posted by symboom

Here is a small reactor
It uses iron wool as a catalyst
Nitrogen and hydrogen

https://www.youtube.com/watch?v=S3j8dCweN4A

[Edited on 18-11-2020 by symboom]

Can I use stainless steel wool?

itsallgoodjames - 18-11-2020 at 08:55

Quote: Originally posted by Fluorite

Quote: Originally posted by symboom

Here is a small reactor
It uses iron wool as a catalyst
Nitrogen and hydrogen

https://www.youtube.com/watch?v=S3j8dCweN4A

[Edited on 18-11-2020 by symboom]

Can I use stainless steel wool?

Most likely not, and even if it did, it'd probably be incredibly inefficient. Iron acts as a catylist, and it's one of the few metals to do so. The only other that I'm aware of are ruthenium, uranium and osmium. It might be worth a try, but I doubt it'd work all that well.

clearly_not_atara - 18-11-2020 at 10:21

Quote: Originally posted by Fluorite

https://m.youtube.com/watch?v=8LriocEzpC0 They tried this but the arc can split ammonia again to N2 and H2
But this can be limited if you add acid like hcl to make nonvolatile salt?

big revival energy :/

symboom is on the right track. The easiest way to mimic Haber isn't to literally copy him unless you have a professional machinist on call (he did!). Instead you can probably find some lab-scale Haber reactors designed more recently that take advantage of the last 100 years of research in chemistry.

If you ask me the future is artificial nitrogenase:
https://onlinelibrary.wiley.com/doi/pdf/10.1002/tcr.20160002...

gerrockium - 18-11-2020 at 16:34

recently there was a paper about using manganese powder at high temperature, manganese forms nitride, then manganese is recovered venting ammonia.

https://doi.org/10.1002/ceat.202000154