Log in

View Full Version : HNO3 from air, using an electric arc.


Mr Cool
August 6th, 2002, 11:40 AM
This is about the only method for production which has not been discussed so far, so I thought I'd start a thread for it.

There are patents describing this method, but they are for producing low conc HNO3 for fertilisers. Those methods basically involve having air blown through an arc, and bubbled through water.
But we want high conc obviously, for nitrations etc. Therefore, the continuous method is unsuitable, because it produces a low conc of NO2 in the air, and since the NO2 will be in equilibrium between the air and water, this can only produce a low conc of HNO3.

I am thinking about having an arc in a sealed container of air at a few atmospheres, at a low temperature of -15*C to -20*C, with ice at the bottom. This would hopefully liquefy the NO2 (N2O4 under these conditions) as it was formed, and the still sealed container could be allowed to warm up, and the N2O4/NO2 would dissolve in the water as the ice melted. It would then be exposed to the air, or have air blown through it (if necessary, since there is a lot of air in the container) to convert the HNO2 that is also formed.

What do you think? I intend to start experimenting with this method this summer.

Alchemist
August 6th, 2002, 03:45 PM
Hello all,

Mr. Cool, I have also been experimenting with the same thing after reading all the discussion and info Mega posted on HNO3 production! I have so far tried a old Model "T" spark coil in a boiling flask through a two hole cork! This was only atmospheric pressure though. After about an hour the there was a VERY slight amount of NO2! This was with air though. I then tried pure Oxgen
via H2O2/MnO2. After only about 7 minutes the entire flask was brown/red with NO2. These where just some tests though
and I really do think a good vacuum and better arc (maybe a computer or TV flyback tranformer) would work much better.

I stopped as I do NOT have a vacuum pump, no flyback (YET), and ran out of time. I keep juming from one experiment to the
next. I do think your on the right track, so keep it up.

Bye for now then.......

Mr Cool
August 6th, 2002, 05:09 PM
Pure oxygen?? You need nitrogen!
Also, HIGH pressure is what you need, not a vaccuum, in order to shift to the right side of the equilibrium of N2 + 2O2 = 2 NO2 = N2O4
Interesting to note that NO is colourless, and IIRC gaseous N2O4 is a light brown (lighter than NO2, anyway).
My first tests will be done with a car ignition coil, 15V 5A power supply and 555 oscillator, because that's what I have at the moment. At first I'll use a glass flask and maybe only a slight pressure increase if any, so that I can see how much NO2/N2O4 is made.

Alchemist
August 6th, 2002, 07:30 PM
whoops!!!

MR.Cool is correct!

My big mistake! It is pressure and NOT a vacuum. The pure oxgen was used to in rich the air only!

*** It's been one of those days. I must think before I write! ***

<small>[ August 06, 2002, 06:34 PM: Message edited by: Alchemist ]</small>

megalomania
August 7th, 2002, 01:43 AM
I have come accross this method during my research and from what I can gather it has always just been a scientific curosity, not a means of production. However, considering that this is late 19th century science their supply of electricity would be a serious issue. Even so it seems to me to be a highly wasteful practice unless you can let this go for a long time (the very reason I only get all utilities paid apartments). They had plenty of nitrates back then and by the time they really figured this out they had Haber and Ostwald.

I read recently that this can be done to completly exhaust the supply of oxygen from the air in a closed system, so I am not sure you have to worry about this forming an equilibrium. This was a 150 year old reference, so it may be a bit suspect for accuracy.

Even so you are likely to at least get a small percentage of unreacted oxygen and nitrogen. By adding extra oxygen you could increase your yield on a batch reaction, but I am not sure it is worth the effirt. Assuming you don't have to foot the electric bill, just let it run for hours and hours, purge the container with air, flush the exhaust into water to catch the NO2 fumes, and start again.

If I really wanted to improve the process I would set up a series of spark gaps down a long tube, say a 4 foot incadesant light bulb, and I would run a very slight pressure through the system so that air would flow continuously. The process may be improved by adding a pressure release valve that opens at say 1.5-2 atm and then closes. I don't imagine such a valve would be that expensive, but I have not exactly seen one.

You could let this run all the time (continuous process) largely unattended. The air flows through, gets zapped, and the exit gas is bubbled into water which slowly becomes nitric acid.

I am not sure how you have rigged up your sparking apparatus, nor am I familiar with the technology available, so I hope my characterization of small sparking devices is not impracticle. I am thinking sparkplug tip size here. If an incandesant bulb proves to be too fragile (they can be tricky to work with) you could use several short lengths of plastic plumbing. I have a bunch of these taking up space from the cast off of various plumbing projects. These come threaded with a male and female end. With these you could drill holes into the top to install the sparker and screw them together.

You could attach any number of these together. You can even add them as you buy them if you wanted to improve your yield at a later time. They don't even have to be straight, you could spiral them or accordian them so the apparatus dosn't take up too much space as well as keep all externial wiring closer together. Using plastic would preclude using massive amounts of electricity (this is not an electric furnace), but you just need a little spark.

<small>[ August 07, 2002, 12:54 AM: Message edited by: megalomania ]</small>

PrimoPyro
August 7th, 2002, 01:45 AM
You will get much better results if you do as suggested and spike the air with excess oxygen. "Air" is only 21% available oxygen, the majority of the rest is nitrogen with some 1-2% trace gasses that will affect nothing.

Your reaction will be slow do to the slowed collision rate between N radicals and O radicals. Most of your N radicals will just be wasted by pairing up with much more available N radicals to regenerate N2 again. This is in turn going to detriment your process in two ways.

1. It's wasteful of the energy provided by the spark, and this translates to wasted time.

2.Heat will become a bigger problem than with the higher oxygen content. The energy transformed into heat will be much greater from 2N: --&gt; N2 than from N + 2O --&gt; NO2.

In the reaction N + 2O --&gt; NO2 much of the energy is needed to form the bonds, and only a fraction of the input energy is transformed into excess heat, whereas the reaction 2N --&gt; N2 takes the entire 100% of the energy placed into the molecule to make radicals and changes it all into heat.

I'd add one volume of air and one equal volume of oxygen to compensate. Again, just my $0.02

PrimoPyro

Mr Cool
August 7th, 2002, 08:22 AM
Yes, more oxygen could be added easily, so I'll try that.
If I was having a continuous process, the first thing I'd try would be to blow the air through a thin copper pipe, the end of which would be close to a copper plate, and the arc would be struck between pipe and plate. With a bit of adjustment of current, voltage and seperation I think it would be possible to have the arc form all around the openning of the pipe, so all the air would have to pass through the arc.

PrimoPyro
August 7th, 2002, 08:32 AM
Um, sorry if this ends up being k3wl or something, but how well would a Van de Graaf generator work? You know, those purple lightning ball things? They're based on mechanically induced static generation and discharge. Beyond that I really don't know.

I know the amperage is very low because they are not harmful. But what of the voltage in these things? Not the input of course, but the spark voltage that jumpts the gap from the center to the glass orb outer shell.

Fuck, I really dont know what the hell Im asking. Umm, well look at its shape. If it had inlet valves at the top for gas, the sphere represents the best shape for maximum spark coverage and even generation.

If you had a bowl or something in the bottom, and kept it in the freezer at -10C or so, any formed NO2 would condense to N2O4. And you could watch it happening too, and take pictures. :D

I wouldnt recommend using a store bought one, but rather build a device lightly based on the same principle and dimensions. What do you guys think? Good idea, or k3wl dud?

PrimoPyro

Mr Cool
August 7th, 2002, 08:52 AM
I don't think electrostatic devices have a high enough average power.

Oh, hang on a minute, that's not a VDG you're talking about! You mean the glass things you touch and get swirly purple plasma being attracted to your fingers? "Plasma Ball" or "Lighning Sphere", they're normally called something like that. A VDG is an electrostatic generator that looks like an Al sphere on a big cylinder and sparks jump off the sphere, the thing you're referring to has an induction coil as the power source.

Well, one of the patents I read had a sphere (one elctrode) with the other electrode at the centre, and air was blown in at an angle so the arc swirled round inside (don't know why it couldn't just be allowed to stay still), similar to a metal "Plasma Ball" or whatever they're called.

Don't know how well it'd work, a normal spark gap would certainly be easier though.

rikkitikkitavi
August 9th, 2002, 05:27 PM
A van de Graaf generator could work but it would take aaaages because of it low power output. In the Birkeland-Oestlund process they used up to 4 MW electrical arcs
By inducing a magnetic field the arc was spread out to form a disk, aka electrical sun and air was blow perpendicular into the disk so the residence time was as short as possible.
Still the gas leaving the reaction chamber contained about 3-4 % NO , mainly due to the fact that NO quickly decomposes at the high temperatures in the arc.

I think the better way to do it is to do like in a similar way of the
Birkeland--process , the only one comercially feasible (until Haber-Bosch kicked their asses with cheap NH3)

You can get up to 68 % acid by reacting NO2 with water.

/rickard

vir sapit qui pauca loquitur
September 21st, 2002, 05:54 AM
as mega has pointed out, i should have posted a few points in THIS thread. whoops. :(

with my intended apparatus, (so that i get it clear) i have 3
chambers(formed from seamelss stainless steel) each one is larger then the last (for examples sake A=6" B=12" C=18") A and B are linked to C (at the ends of all pipes are turn valves, so i can shut off/open the flow)

i feed pure 02 into B, pure N2 into A (after flushing all tubes/pipes with helium to provide inert aptmoshpere)this pressurises A and B to about 2000 PSI. After that i open the valves to C, this pressurises C with N2 and 02. Valves are then closed.

In tube C i would have installed a small bank of car-spark plugs (due to the high pressures used i was limited in what to use) about
4.these are located at the TOP of the chamber (i was hoping that with NO2 being heavier, i would increase the yeild if i located the spark units where the substrate is lost likely to be located)

i would power the spark plugs from a car spark-gap generator, and i would power it rom the mains. power is not an issue as i also get utilities included in rent.

My aim was for the ability to set up this, and walk away for a few hours/days and to come back and to _slowly_ bleed the gases formed into water. Providing Hno3, i would know if cylinder C has been effected, as i will include a pressure guage to inform me if the pressure has gone done (a sign of formation of something) as there would be a lower number of moles of gas, hence pressure is lower
(PV=NRT, if the V=volume remains constant, and the P=pressure
lowers,and the T=temperature remains near constant
(R is the gas constant, 8.32 j/mol) then we must have a reduction of N=number of moles)

but are we talking about hours or days or even weeks ?? thats my new question, as unfortunatly i have limited knowledge of a reaction involving spark-plugs :p i'm willing to leave it for two weeks to keep on going, but i dont want to start bleeding it and lose the 99% of the gaseous aptmosphere that hasn't reacted.

i did have the idea flitter across my troglodytian brain that i shoul include an amount of water in C so that the NO2 can disslove into solution. after i coat the stainless with ptfe! and then i could drain this water from the vessel, this means that i lose virtually nothing from the reaction vessel in the way of unreacted gasses.

what do you think ?

megalomania
September 23rd, 2002, 06:10 AM
Perhaps there is an even simplier way of doing this. I found this blurb from a published reference:

</font><blockquote><font size="1" face="Verdana, Arial, Helvetica">quote:</font><hr /><font size="2" face="Verdana, Arial, Helvetica"> When strong electric sparks are passed through a confined portion of air, standing over a solution of potassium hydroxide, the volueme gradually lessens and potassium nitrate may be detected in the liquid. So when ozone acts upon the nitrogen of the air, upon ammonia, or upon the lower oxides of nitrogen, water being present, nitric acid is produced. </font><hr /></blockquote><font size="2" face="Verdana, Arial, Helvetica">I assume the presence of potassium hydroxide is intended to increase the solubility of the gas such that it reacts before it can be broken down again. You could still use a pressurized system with pure O2 and N2, but I don't see a reason to use pure gas, or necessarily high pressure. I would go with pressure if I could just because it will increase yields. I would settle for a few atm's if I did it. I would definitely modify your procedure to include the alkaline solution (I am sure sodium hydroxide would work just as good).

vir sapit qui pauca loquitur
September 23rd, 2002, 02:52 PM
thanks for that, i have just got the reaction chamber back from the company that coats my equipment with PTFE (ironically i said i was dealing with a high conc KOH solution :rolleyes: )

so in a short while i should be up and running, but i need for the money to roll in. As i bought some rather expensive investments, and being a student does not help things much.

interesting.... verrryy interesting.

megalomania
September 24th, 2002, 12:57 PM
Would you mind elaborating on your sparking equipment. I think I may try this myself, but I don't know much about electronics. Any references would also be good.

I found this pdf document about building a sparking device:
<a href="http://www.imagineeringezine.com/PDF-FILES/spark4.pdf" target="_blank">http://www.imagineeringezine.com/PDF-FILES/spark4.pdf</a>

Nitro
September 24th, 2002, 04:57 PM
Very intresting high Voltage source is an industrial Powder painter(Dont know how you call this but it used in car industrie to make an black colerd powder over some Axes parts).They use an ionisation sparc(80 000V and an curent over 50mA Deadly).If you can get one of this transformer your lucky.P.s you need an cooling defice for the Transformer i had forgett this and it was destroyed after 2 minutes(sparc range was over 15cm and 1cm thick . :)

rikkitikkitavi
September 26th, 2002, 02:25 PM
Just one notion about making NO2 from N2 and O2 from air and an electrical arc.

For the reaction
N2 + O2 =&gt; 2 NO

Heat of reaction is + 90,8 kJ / mole of NO

Since not even the best of the industrial methods utilizing an electrical arc would come over 10 % efficency
(energy bound in NO / total energy input)

this means that more than about 1000 kJ of energy is needed for one mole NO.

1 kWh then gives about 3,6 moles of NO.

so by running an electrical arc consuming 1 kW (which is quite a lot) for one hour you would at best get 3,6 moles of NO.

Then by reacting this with air to NO2 and then absorbing this into water
(3 NO2 + H2O =&gt; 2 HNO3 + NO net reaction)
you would at maximum efficency produce

2/3 * 3,6 = 2,4 moles of HNO3 = 151 grams / h .

Most likely efficency will be a lot less, lets say 50 % of above.

But still 75 grms is produced /.

I dont know the price of electricity in f e x USA, but in Sweden we pay about 0,05 € for one kWh, which means the price for one liter 68 % would be

1410*0,68 / 75 * 1 * 0,05 = 6,5 €/litre

/rickard

megalomania
September 27th, 2002, 02:56 AM
Actually all of the NO produced would react to form NO2 in this system that would then be reacted with a hydroxide to make a nitrate. From nitrate distillation, assuming you can get a 90% yield (not exactly unreasonable) your yield would be a bit higher, about 102 g of nitric acid for the amount of NO produced.

One thing I did notice is that the reaction is not the direct combination of nitrogen with oxygen. It goes to ozone first and that reacts with nitrogen. That means we have to deal with 143 kJ of energy needed to make 1 mole of ozone. Assuming 10% of that does react we are looking at 1430 kJ of energy to make one mole or about 2.517 moles per kWh. Now we are looking at about 158.6 g of nitric acid being produced assuming a 10% conversion of nitrogen. I don’t have an electric bill handy so I can’t calc that.

After a quick Internet search I came up with a rather good report of Average US electricity prices <a href="http://www.eia.doe.gov/emeu/mer/pdf/pages/sec9_14.pdf" target="_blank">here</a> which shows currently we pay about 8 cents a kWh. Last week the Euro and the dollar were almost equal. My 8 cents will get me 56.1 mL of 70% nitric acid, or about $1.43 per liter. Oh, that is at 5% efficiency by the way. To put that into perspective a bulk purchase of technical grade 70% nitric acid from Fisher (18 L) would run you $8.83 a liter. That does not even include shipping and handling.

rikkitikkitavi, shouldn't you be paying less?
75g HNO3 per kWh and 1.410 g/ml at 68% conc
75g HNO3 / 1.410g/mL = 53.19 mL 68% HNO3 per kWh
1000mL / 53.19mL = 18.8 or 18.8 kWh per L
18.8 x E0.05 ($0.05) = E0.94 ($0.94)
That comes to about 94 cents per liter, not E6.50 ($6.50).

<small>[ September 27, 2002, 02:00 AM: Message edited by: megalomania ]</small>

rikkitikkitavi
September 27th, 2002, 07:29 AM
your absolutely right about the price, mega. I missed by a decade :)

somehow I transformed it into swedish crowns (about 9-10 SEK for a €)

So it improves the situation very much, making nitric this way is so much more cheaper than purchasing.
For simplicity I assumed that any NO formed in the absorbtion-reaction step of NO2-H2O would be lost. In practice this would probably lead to large amounts of toxic NOx released into the surroundings.

NO2 + NO + 2OH- =&gt; NO2- + NO3- + H2O

as long as pH&gt; 8 the NO2- will not decompose into NO3-.
(this is a comercial route to nitrite production)
Tailgases from HNO3 production are scrubbed by lye, minimizing pollution and recovering nitrite, a useful chemical in many ways:)

Anyway , it doesnt matter if the reaction goes bu intermediates like ozone for the energy consumtion.

For the calculation, only net reaction is neccessary.

/rickard

<small>[ September 27, 2002, 06:34 AM: Message edited by: rikkitikkitavi ]</small>

Marvin
September 28th, 2002, 06:59 PM
The idea of making nitric acid out of air and water is a very attractive one. I researched this in detail in the time before NO2 chose to reveal quite how toxic it was. A very good reference for this is Mellor Volume 8, I'm not sure how common this series is in the US, if there is a problem I will scan and OCR. The book is quite dated so modern interpretations need to be applied.

If a volume of air is enclosed and heated to a high temperature around 3000K you start to get nitric oxide forming. The higher the temperature, the more is present at equilibrium and the faster equilibrium is reached. If the mix is cooled slowly the amount of NO decreases with the equilibrium until it neerly vanishes. Whats required to make nitric oxide is to cool the hot gas much faster than the reaction rate can enable it to reach equilibrium. Fast cooling is the reason both the air and the arc move in different directions in the furnace designs. Long below the temperature where rate of reaction is an issue, the nitric oxide starts to oxidise to nitrogen dioxide.
The limiting reaction is thus N2 + O2 =&gt; 2NO, and since the volumes are the same on both sides is independnt of pressure. Reports of maximum yeilds at 50% oxygen are consistant with this. At higher pressure you get more collisions per unit time in the gas and as a result higher pressure faciliates the destruction of NO during the cooling stage. High pressure also results in greater difficulty cooling the gas repidly (Surface area to specific heat capacity), and alows only much shorter arc lengths. According to Mellor the highest yeilds were obtained by Haber (Yes the same Haber) at a pressure of around 100torr. No industial furnace I know of attempted to use reduced pressure as the total nitric acid will be reduced for a given furnace design, and its total product rather than pure efficiency that matter.

Yeilds for most furnaces were of the order of 50 grams of nitric acid per kwh. Total power is what matters most in producing nitric acid, its pointless having a super efficient system if it only outputs a few grams a day. Electrostatic machines, induction coils etc are unlikley to output enough power for a useful system. To get something up and running try building a jacobs ladder using 2 coathangers, a bell jar and a neon sign transformer. Neon sign transformers are ideal for this sort of nitric acid research as they output reasonable power and are current limited magnetically. An arc starts small due to the breakdown of air, about 30kv/cm for dry air at atmospheric pressure, but the arc can be pulled much much wider when its started, which is exactly what happens in a jacobs ladder. Jacobs ladders are notorius for the amount of NO2 they produce while running in air, so it seems like this would be a useful start.

Its hard to produce concentrated acid from dilute nitrogen peroxide, but there are a few cunning inprovements that can be made. Silica gel will not only catalyse the production of NO2 from NO (which is much slower in dilute amounts) but will absorb the resulting nitrogen peroxide. The nitrogen peroxide can be expelled from the silica gel by heating it. Some ceramic oxides are thought to increase the rate of formation of NO in the arc area, tungston oxide I bilieve is one. The information on the net on this subject is largly rubbish, with people claiming 'vortex' reactors being cheaper at making nitric acid than large haber/ostwald plants. They may be an improvement over the old designs but reliable data wont be obtained from the designers, most of whom seem to bilieve they have solved all of the worlds problems if everyone else would just listen to them, or the increasing number of nutcases that believe science is wrong and therfore shoudlnt be learned.

The fact the internet is free for anyone to publish anything unfortunatly removes the emphasis on reproduceable experiments and unbiased trials. I suspect none of these methods produce a massive improvement, equal N2 to O2 aparently only makes a difference of about 18% at atmospheric pressure over ordinary air.

Anyone who is interested in nitric acid might want to think about electrolysis of molten sodium nitrate, which yeilds copious amounts of nitrogen oxides/oxygen at the anode, and liquid sodium at the cathode. Why the sodium is stable in the presence of the oxidising salt I wasnt able to find out, maybe the area around the cathode reduces before it forms, maybe it is more stable becuase of the electric potential. Furthur info from reliable sources needs to be obtained before attempting this. It kind of defeats the object of making nitric acid from air and water alone, but this will always be much more trouble than simply buying sulphuric acid and an alkalie nitrate and making the acid directly.

NO2 is toxic in the extreme, take all precautions to avoid breathing it in.

A-BOMB
December 20th, 2002, 12:10 PM
Sorry to bring up a old'ish topic but I have just gotten all the componets to build my own version of this type of devise, I'll be testing several differnt power sources. The first will be a stun-gun,
the second a high voltage devise I ripped out of a "Ionic Breeze" air cleaner, the thired will be a old laser power supply that I snagger from the carnegie science center, and the last an old induction coil.

I'm setting it up with a small compressor that will mix the two gasses as they are relessed from the tanks. Next the gas mixure will be pumped into a 3 foot long glass pipe with spark gaps every few inches alone the length ending in a (extemely thick) filtering flask were the mix will be bubbled through the water the first time befor it is sucked out and somewhat pressureized before being sent through the system againg.

10fingers
December 22nd, 2002, 12:50 AM
Deleted

<small>[ March 02, 2003, 04:03 PM: Message edited by: 10fingers ]</small>

10fingers
December 29th, 2002, 05:27 PM
deleted

<small>[ January 01, 2003, 07:34 PM: Message edited by: 10fingers ]</small>

Mr Cool
December 29th, 2002, 06:19 PM
"The home experimenter may want to scale this down a bit."
I rather think that a 3500*C, 1.8 metre disc of electric fire sounds rather neat :) . But I think we should scale it up :D .

Anyway, it sounds like you've had good results! Please give as many details as you can on your setup, I never got round to doing any serious experimenting with this but if you get 250mL of 70% each day, for £0.16, then I think I'll give it more thought!!

nbk2000
December 29th, 2002, 07:41 PM
I saw on a laser site how you can use a hacksaw blade, instead of a smooth wire, to provide multiple plasma arcs for a nitrogen UV laser.

This may improve efficency for your electric NOx producer since it's the arc that does the converting.

Naturally we'll want pictures or diagrams so we can build our own converters.

A-BOMB
December 29th, 2002, 10:02 PM
THAT BITCH!!! :mad: :mad: I told her to stay out of my lab, but no she just has to go down there and mess with shit :mad: Well my stupid dumb asshole that I call a mother just broke a foot off my glass tube that I had. It not totaly destroyed but still, :mad: that cost me 47$ I think I go kill her now. :mad: Well my unit didn't get that good of effcienty but that me using a couple stun guns for my spark gaps.

kingspaz
December 31st, 2002, 09:25 PM
could the mist be HNO<sub>3</sub> vapour?

10fingers
January 17th, 2003, 05:40 PM
Deleted

<small>[ March 02, 2003, 04:04 PM: Message edited by: 10fingers ]</small>

mongo blongo
January 17th, 2003, 07:17 PM
What was the max conc you were able to achieve with the water vapor? I found out that rubber can conduct the electricity too. It went straight through a rubber covering on a crocodile clip at 8K. I was working with latex gloves thinking it would prevent any shocks so I'm glad I found out sooner than later. Now I ware latex gloves under cooking gloves. That shock must have hurt dude!
I probably would use this to make nitrates but all the hydroxides I can get are very expensive.

10fingers
January 17th, 2003, 11:07 PM
Deleted

<small>[ March 02, 2003, 04:05 PM: Message edited by: 10fingers ]</small>

nbk2000
January 18th, 2003, 12:49 AM
Well, if it takes 4 days to make about a cup of 68% nitric, then how much will the electricity that it took to make have cost you? More, or less, than making it some other way?

10fingers
January 18th, 2003, 12:21 PM
Deleted

<small>[ March 02, 2003, 04:06 PM: Message edited by: 10fingers ]</small>

Marvin
January 20th, 2003, 12:38 AM
Just a few comments.

68% is only the az point, it has nothing to do with this experiment, your limitation is mainly the concentration of NO2 in the gas you produce. Should you wish to increase this, you can use silica gel as Ive mentioned before to store, and then expell by heat, the next major limit being the partial pressure of oxygen you can supply at equilibrium.

The arc should not be constricted, it should be alowed as wide as it can without breaking for any given supply voltage. This provides the highest voltage drop over the arc possible for the configureation, and this provides the best power efficiency for the total circuit. (Not max throughput, which is at 50% power efficiency). A jacobs ladder electrode arangement is very good for this, but doesnt alow for ideal air contact, which is best done with the magnet setup.

Voltage is just as important as current for power dissupation by the arc. Power = VI. Substitute ohms law and you get the other two generally useful power equations for resistive loads, Power = I^2R = V^2/R. in this case R will depend on other factors, including current flowing and arc length. The only thing you can be sure of is VI by actually measureing them, which is the voltage over the arc and the current through it, in short arcs you cripple V, and most of the power ends up being dissupated by the transformer, without actually measureing your setup, I would guess this is the major problem.

Alchemist
January 20th, 2003, 12:21 PM
Hello,

I'd also try pumping the air through concentrated hydroxide or etc. to remove as much Co2 as possible before the arcing! Good luck and keep us all updated. P.S., do you have pictures you can post?

10fingers
January 20th, 2003, 02:46 PM
Deleted

<small>[ March 02, 2003, 04:07 PM: Message edited by: 10fingers ]</small>

rikkitikkitavi
January 20th, 2003, 03:56 PM
just one note on absorbing NO2 in water.

At atmospheric pressure it is very difficult to drive the conc beyond 53-55 % , due to the kinetics. It is easy to increas conc by distillation of course.

Over azeotropic HNO3 can be made by reacting NO2,O2 and H2O in autoclaves at high pressures ~50 bar, this is a method used in some european plants. But it is of course not feasible to perform at home.

/rickard

Genital-Warfare
January 24th, 2003, 11:06 PM
Keep in mind that platinum is used as a cat for this reaction; 600*C oxygen enriched air is blow over Pt and then the procedure continues.

CTR

J
February 3rd, 2003, 07:51 PM
Not much to add, but I think the current is going to be a lot more important than the voltage. Part of my work involves experimenting with machines that ionize gas to form a plasma. Although voltage does have a bearing on the amount of plasma produced, current is far more important. I'd want to use a current of at least an amp or two, up to around 7 or 8. This would be extremely hot, and ionization would be far more efficient than at a few mA.

For an expedient power supply, a microwave oven transformer will supply about 500mA at 2000v IIRC. The spark gap would be a maximum of 2mm at atmospheric pressure, but using a jacobs ladder this could be increased.

Another idea would be to use a trigger transformer to get the arc going, and a lower voltage for the main arc. This method is used by some welders to trigger the arc, which is only a few tens of volts but many tens of amps. Perhaps a welder could be adapted for this purpose.

I'd love to build a machine of this type myself. Unfortunately I have no mains power in my garage, or safe means of getting any there :-( I have a number of very lethal power supplies that I've collected. These include a 10kv 30mA neon sign transformer, a variable voltage DC power supply that can put out 200v at 6A, and a voltage multiplier that will put out upto 80kV at 100mA when fed from a suitable PSU! This is a lot more power than I could get out of a 13A socket.

If only I had a proper lab. Maybe next year...

10fingers
February 3rd, 2003, 09:48 PM
Deleted

<small>[ March 02, 2003, 04:08 PM: Message edited by: 10fingers ]</small>

Anthony
February 3rd, 2003, 11:05 PM
If an arc welder is deemed a suitable power supply, then I would use a rewound microwave oven transformer instead of pulling apart a welder.

The secondary windings are simply hacked off and replaced with suitably thick insulated wire with the required number of turns to give the voltage desired.

Should be good for a Kw and many could be used as MOTs can be had for free.

Some people have done this to make their own arc welders BTW.

Marvin
February 8th, 2003, 02:46 PM
10fingers, the az point is not limiting in your design becuase below 100% nitric concentration, the vapour will always be a mix of acid and water. Decomposition equilibrium with the exiting gasees is as youve found. To get high concentration acid you need to concentrate the NO2 as several people have discussed viable means of, higher than az should be easily possible but a good design would add one or more low concentration scrubber(s) at the exhaust to reduce wastage.

Putting more power in the arc does mean you have to cool the aperatus much better, but then power into the arc will be proportional to the amount of nitric you can produce with the resultent gasses.

By your numbers you are a fair amount below the birkland values but considering its a first attempt Id call it a roaring success.

CTR, I havnt heard of this method, Id have thought 600C would be far too low to produce a reasonable concentration of NO, do you have any more information on this?

J, you are right about current being the limiting factor for producing plasma. This is because plasma = ions + electrons, which are your only charge carriers and therfore the greater current the more charge carriers must be carrying it, and therfore the more plasma you have. This is even more true at low pressure, where I suspect you are making your plasmas.

Unfortunatly, we dont want plasma. We dont want ions or electrons, what we are after is vast volumes of very hot air (3000K ish) being rapidly cooled. This requires having ionised gas present becuase otherwise it wont conduct the electricity that is doing the heating, but our limiting factor is power into the arc, not current through it. NO can be made without an arc, by burning fossil fuels, and this is how it ends up in a car exhaust, purely thermal.

MOTs are excellent becuase they provide high power into an arc that can be 6 inches long. The air going through it can be rapidly cooled, and relativly little power goes into the electrodes that form the arc.

A Jacobs ladder does alow an arc to start small and be pulled very wide, but the starting arc must be wide enough for the rising air to pull it up. 2mm sounds about right for a MOT but this is not wide enough to start a jacobs ladder rising, Ive tried, it sits there arcing into 2mm and most of the power is being dissupated by the MOT itself brcuase of the matched load problem.

Generally, I am unsure how far its possible to pull a 25v arc @ 50 Amps in air, but my guess would be not far enough. Arc welders are designed to put as much power as possible into the metal, not into the air around it, hense high current, low voltage.

Making the arc 'hotter' is one other obvios way of getting more NO that doesnt actually work, its impossible to cool the gas quickly enough to take advantage of the higher equilibrium concentration of NO. I just thought Id add this as an afterthought.

I'm convinced stacked MOTs, long moving arcs (magnetic field) and high total power is the way to go for nitric by arc.

10fingers
February 9th, 2003, 07:02 PM
Deleted

<small>[ March 02, 2003, 04:10 PM: Message edited by: 10fingers ]</small>

Marvin
February 20th, 2003, 02:31 AM
You are quite correct about the products of the reaction moving up with the arc, and that this is counter productive. It does stretch the arc considerably though, and due to its higher voltage drop more NO should be produced with less power wasted into the transformer. Ideally a setup of this type would have 2 main differences, firstly it would have its electrodes trimmed short enough to prevent the arc breaking, and secondly would have a vertical magnetic field to force the arc to move horizontally, perpendicular to the current flow. The two reasons I would expect the arc to break, are more obviosly that the current cant sustain the arc, and less obviosly that the voltage drop over it exceeds air breakdown for the gap at the base.

The comment that rikkitikkitavi made about concentrations over 55% being difficult to make with this process turn out to be remarkably accurate for equilibrium too. Numbers from a table I thought Id lost (or Id have posted them earlier) on practical values, just those relavent to arc processes to avoid a table formatting nightmare.

NO2 conversion to nitric acid for 2% NO2 in air by absorption in nitric acid at 10C and atmospheric pressure (higher yeilds at low temps).

5% Nitric, 65.7% conversion.
15% Nitric, 64.0% conversion.
25% Nitric, 60.5% conversion.
35% Nitric, 53.8% conversion.
45% Nitric, 38.0% conversion.
55% Nitric, 11.5% conversion.
65% Nitric, 0.4% conversion.

I will OCR the full table soon.

Some general features of the whole table Ive noticed, conversion drops a lot for higher temperatures, eg for 25% nitric acid a 60% conversion at 10C drops to 23.6% conversion at 75C. This may entirly be due to evaporation of acid, or bad absorbtion values it doesnt say, this is overall yeild for a single absorbtion stage. Rather disturbingly for the whole table, it doesnt rise to 100% conversion for a single stage at the limit of pure water, implying 40% of our possible product is lost in any single stage absorbtion. This isnt a limit due to the low concentration of NO2 in air though, as the following values show for a mere 0.1% NO2 in air.

5% Nitric, 59.8% conversion.
15% Nitric, 48.0% conversion.
25% Nitric, 31.3% conversion.
35% Nitric, 14.6% conversion.
45% Nitric, 2.2% conversion.

The secret to getting the most nitric acid from NO2 seems to be absorbing it in a series of successive wash baths at room temp, or lower.

Ok, methods of getting stronger nitric with the arc process, apart from the obvios distillation of the product acid. Use of silica gel. After cooling exit gasses to room temp, silica gel will catalyse the conversion of NO to NO2, and also absorb it. It can be expelled by heating and presumably this must be a substantial mole fraction of the silica gel used and the same sort of effect as removing water from air. Being able to produce virtually pure NO2 should alow rather more concentrated nitric acid to be produced than with 2% in air. Another idea Ive been working on, involves ozone. Ozone isnt required for the formation of NO, and isnt produced by the arc chamber. If you add it to the produced NO2 however, it oxidises it to nitrogen pentoxide rapidly and from what I'm reading, quantativly. Freezing out the nitrogen pentoxide and using it to make 99%+ acid in a single step would be very nice, but oweing to its relativly high (sublimation) vapour pressure the numbers look unfriendly. For 2% by volume NO2, mixing with an equal quantity of 1% ozone in air, producing 1% nitrogen pentoxide by volume would require a temp of about -25C to start forming solid and about -30C to produce half of the total N2O5 as solid. Simply running the 1% N2O5 into water or acid however. should prove a much easier route to neerly pure nitric acid than anything involving NO2 directly.

Ozone is very nasty stuff, but we're dealing with NO2 anyway and have allready discussed those horrors.

Tuatara
February 20th, 2003, 04:47 AM
Ozone can be readily produced by a corona discharge. This involves raising a bunch of sharp needles or a fine wire to several kV dc. The electric field intensity around the wire is sufficient to ionise the oxygen, but drops off fast enough to prevent arcing. If you were to put this ahead of your main arc you'd get the additonal benefit of reducing the breakdown voltage of the air, facilitating the use of a larger arc gap.

The ozone generation problem is one reason negative ion generators (supposed to be good for you) only run at about 3kV. A good DC supply could be built from the horizontal output transformer of a TV - that'll get you at least 25kV. They can be easily found at electronics surplus stores.

Anthony
February 20th, 2003, 12:13 PM
How much ozone would be required?

The domestic generators seem to be rather limited in output:

MorZone 30 30mg/hour ozone generator. $67.50

Red Sea 50 mg/hour ozone generator $160.99
Red Sea 100 mg/hour ozone generator $184.99
Red Sea 200 mg/hour ozone generator $248.99

megalomania
February 20th, 2003, 06:00 PM
I had envisioned using a series of water filled bottles to react my NO<sub>2</sub>, but I only knew the yield was lowered because of production of NO in the reaction. In order to react the NO back into NO<sub>2</sub> another bottle seemed logical. I wonder if this is the same principle behind your figures, Marvin? It is nice to see some data to validating this system. Industrially they use a water mist that trickels down the side of a tall chimney to react the nitric oxides as they rise up. They get a max conc. of 65% doing this. This is basicially the same thing, but in one system.

Does anyone think using a strong base (like NaOH) in solution would increase the yields of nitric conversion? The base would neutralize any acid formed thus keeping the concentration low. Of course the presence of nitrates may effect the equilbrium in other ways that limit the usefulness of this.

My hopes for the successive bottle system is that the first bottle would become more and more concentrated, eventually reaching fuming, while successive bottles would accept an increasing share of the nitric oxides. With enough botles in series the last should be very dilute thus getting a very high yield of acid because almost all nitric oxides are captured. Now I see that obtaining a fuming acid in the first bottle is not likely to happen. Your data, Marvin, will help me make a more accurate determiniation of how many bottles to put in the series. More will be needed, or the reaction run less, to convert all the oxides.

10fingers
February 21st, 2003, 12:25 AM
Deleted

<small>[ March 02, 2003, 04:11 PM: Message edited by: 10fingers ]</small>

Marvin
February 24th, 2003, 05:13 AM
That is doubly dissapointing. In retrospect its likley the constant rapid changes in voltage degraded the insulation. How hot was it getting while constantly on? It shouldnt have a problem thermally at least. I cant get neon sign tranformers here easily, or Id have tried something similar myself.

With regard to ozone, 10g/hour would be borderline useful. Putting it in front of the arc chamber would be counterproductive though, as the conditions that form NO, would destroy the ozone. Domestic generators are useless, as would be expected, commercial or industrial water cleaning/pool cleaning equipment seems to be much more viable. Ive found a unit being sold that produces about 30g of ozone an hour, and though I dont know what it costs (probably much more than I could justify spending) the stated power consumption values are useful. It produces 30g/hour @ 6% ozone in oxygen, (It uses an oxygen enricher from air), and consumes just over 600W, not including the enricher itself.

The standard ozone generator uses concentric glass tubes to produce a 'silent' discharge, I bilieve with high frequency AC at 20kv. This rules out use of voltage multipliers of the cockroft walton type, a single TV loptx would be capable of around 9Kv, so maybe this is possible, but would probably only supply a few 10's of watts at the most. If a fine wire corona discharge will work at power, then DC operation becomes possible and a voltage ladder with a neon transformer might be feasable. The problem I forsee with the fine wire, is this might need a very large box to be operating in without heating problems of the wire itself which would decompose the ozone as its formed.

2 moles of nitric acid require 1 mole of ozone, so the numbers are about 126g nitric, needing 48 grams of ozone. An arc chamber should be easiliy capable of 50g nitric acid per kilowatthour, so a 1 kilowatt 'furnace' needs an additional 400 watts or so to produce enough ozone to completely turn the nitrogen peroxide into pentoxide. Ignoring the cost and additional complexity of the ozone generator, its a small increase in running costs to be able to produce neerly 100% nitric in 1 step.

Multiple stages to produce 65% acid from NO2 with a nitrogen pentoxide 'booster' to turn that into 100% acid would save some power due to less ozone, but 65% odd nitric is still much more than half water on a molar basis, so the saving is small and it turns a potential tabletop generator into a small chemical plant.

A closed vessel over caustic soda will absorb oxides of nitrogen almost completely, forming mostly nitrate, with some nitrite. Argon was discovered this way by Cavandish, who added oxygen to air contained in a glass bulb, and removed nitric oxides with lye while sparks were passed through the gas. I have to say if I was going to the trouble of fixing atmospheric nitrogen, Id be very dissapointed with anything other than high concentration nitric acid. In terms of areas that cant get nitrate fertilzers, manure is much more likley to be of use for making nitrates than mains electricity.

More current than your neon tx was providing would do only fractionally worse than the extra power it provides, but not better. Very low current sparks, such as static electricity circa 100na do very well for producing nitric acid (in excess of 70g/kilowatthour IIRC), but as the total power is very small, throughput is miniscule.

Mr Cool
February 24th, 2003, 11:54 AM
"Silica gel will catalyse the conversion of NO to NO<sub>2</sub>, and also absorb it."

I wonder what it's like at removing NO<sub>2</sub> from things other than air, for example highly concentrated nitric acid :D .
It's probably just wishful thinking, but I'll try it anyway. And if it absorbs water in preference to HNO<sub>3</sub>, then it could kill two birds with one stone.
The trouble with urea is that it produces moisture (CO(NH<sub>2</sub>)<sub>2</sub> + 2 NO<sub>2</sub> --&gt; CO<sub>2</sub> + 2 H<sub>2</sub>O + 2 N<sub>2</sub> + 0.5 O<sub>2</sub> I assume, but correct me if I'm wrong), and when you warm it and buble dry air through you lose some by evapouration and you get nasty fumes. While none of these problems are very significant, it would still be nicer just to add silica gel and filter it!

10fingers
February 24th, 2003, 10:02 PM
Deleted

<small>[ March 02, 2003, 04:12 PM: Message edited by: 10fingers ]</small>

Mr Cool
February 25th, 2003, 07:04 AM
What are the advantages of using silica gel to absorb NO<sub>2</sub> rather than MgCO<sub>3</sub>, for example, if any?
It strikes me that using a basic magnesium compound slurried in water would be more efficient. Silica gel must form an equilibrium of NO<sub>2</sub> with the atmosphere, and will thus reach saturation quite easily in low partial pressures of NO<sub>2</sub> I would have thought.
MgCO<sub>3</sub> in water, however, will keep on absorbing the gas until it has all reacted.
Then you roast it like you would with silica gel (after drying it out) to get back your NO<sub>2</sub>. Calcium nitrate decomposes at around 600*C, magnesium nitrate will therefore decompose below that.
To make the NO<sub>2</sub> into HNO<sub>3</sub>, I personally would condense it with a freezing mixture, and pour it into a container with the appropriate amount of ice, below the bp of N<sub>2</sub>O<sub>4</sub>. Then seal it up and allow it to warm to room temperature. This will create the combination of higher pressures and lower temperatures that you want.

Microtek
February 25th, 2003, 01:45 PM
While working on my HMX procedure, I came across a patent about generating N2O5 in nitric acid by electrolysis. The idea was to start with essetially anhydrous HNO3 then dissolve N2O4 in it and electrolyse the solution with a carefully controlled voltage. Now, N2O4 exists in equilibrium with NO2 and low temperatures favour the dimer so maybe you could distil a portion of HNO3 90% at atmospheric pressure so there was a certain amount of NO2 dissolved in it. Then lower the temp to form N2O4 and convert this into N2O5 via electrolysis. You could then add more NO2 and continue the electrolysis until the nitric had become too viscous from the N2O5 for proper mixing. Then simply dilute with water to convert the N2O5 to HNO3, add more NO2, cool, etc. etc.

Tuatara
February 26th, 2003, 09:03 PM
I did a lot of searching yesterday and found a very interesting patent FR2549459 on <a href="http://ep.espacenet.com/" target="_blank">esp@cenet</a>. Its in French, so if theres any French speakers out there please correct me.
Summarised (from Google translator): Use a microwave powered plasma in air at a pressure of 25-75 millibars to generate NOx. Can be powered by a microwave oven magnetron (refer to <a href="http://www.belljar.net/plasma.htm" target="_blank">The Bell Jar</a> )
Notes 1/ Suitable catalysts are Tungsten trioxide or Molybdenum trioxide
2/ best energetic efficiency with 65% N2, 35% O2 , but does not alter NOx ratios
3/ with MoO3 catalyst energy use is 28MJ/kg of NO at 6% conc in gas leaving reactor
4/ without catalyst energy use is 43MJ/kg of NO at 4% conc

Saw another site where someone created a plasma is a glass container in a microwave oven at atmospheric pressure. At end of run the container was full of brown gas!

mongo blongo
February 26th, 2003, 11:49 PM
Do you mean <a href="http://www.kronjaeger.com/hv/hv/exp/no/index.html" target="_blank">this</a> site?

Tuatara
February 28th, 2003, 05:02 PM
No, actually I was talking about this <a href="http://jnaudin.free.fr/html/oa_plsm2.htm" target="_blank">web page</a>

I foolishly didn't bookmark it and it took me a while to find it again

10fingers
February 28th, 2003, 06:49 PM
Deleted

<small>[ March 02, 2003, 04:14 PM: Message edited by: 10fingers ]</small>

Tuatara
March 2nd, 2003, 02:23 AM
10fingers you will need an impedance in your arc circuit! If you strike an arc directly out of your MOT you will pull a lot more than 2kW as they are not current limited, so your transformer will die rapidly and spectacularly. Neon sign transformers are impedance limited - when you short the output (which is basically what an arc does) they become a current source, limited to 30mA or so. Your impedance should be an inductor of about 10 to 20 H for 60Hz mains(thats HUGE), or a capacitor of about 400nF, and it must be capable of handling at least 2kV.

Why not use the microwave to create the plasma? You've got 3 magnetrons, with waveguides and power supplies ...

10fingers
March 2nd, 2003, 02:48 AM
Deleted

<small>[ March 02, 2003, 04:15 PM: Message edited by: 10fingers ]</small>

Tuatara
March 2nd, 2003, 04:34 PM
Always happy to help!

Here comes Electronics 101 <img border="0" title="" alt="[Wink]" src="wink.gif" />
Had to think about the primary/secondary question - the answer is yes, you can put the impedance on the primary side but it needs to be smaller (smaller inductor or bigger capacitor). A resistor can be used but it will dissipate huge amounts of heat , 2kW. Inductors and capacitors don't do this (well, not much anyway) as they are imaginary impedances - current and voltage are 90 degrees out of phase (ideally) so when you multiply volts x amps and integrate over the complete sine wave you get ... zero!

Microwave isn't that hard. The microwave energy is absorbe in the plasma. Heres a couple of good links
<a href="http://www.tpub.com/neets/book11/index.htm" target="_blank">Microwave theory</a>
<a href="http://www.psfc.mit.edu/plasmatech/plasma_technology.html" target="_blank">MIT plasma research</a>

Putting your transformers in series is no problem - if you get it wrong you will get no volts out the end.

Marvin
March 2nd, 2003, 06:25 PM
I would think the big advantage to silica gel would be a vastly reduced temperature for driving off virtually all of the NO2. I would think substantially below 200C. Silica gel decomposes itself so that limits the temperature it must work at.

My understanding of the urea reaction is it is the diazotising of an aliphatic amine, which promply decomposes to nitrogen gas. I would not expect oxygen to be in the products, and NO2 to be effrctivly exhausted by the equilibrium shift.

MOTs are in another level entirly to neon sign transformers, you need to do your EE homework. <img border="0" title="" alt="[Wink]" src="wink.gif" />
Taking 80% of the max power of the transformer wont get you even within a factor of 10 of whats really going on if you dont get the design of the electronics and arc sorted.

A single MOT, with a peak output of about 2kv will need an external starting system. The ones Ive used would stand being shorted long enough to start the arc, but would get very hot very quickly. When a MOT is shorted it pulls its max power, and dissupates all the energy entirly within intself. A short arc is indeed much the same as a direct short, but a long arc - close to the maximum the transformer can produce is usually enough of a load to prevent the transformer overheating (in the case of MOTs). I am unsure what rating the transformers I had were, so your experience may differ.

My best 2kv MOT developed a stable, curved arc in excess of 6 inches total length. Putting 2 transformers in series should not be required or advantagous. The semiconductor diode the doublers use wont be happy with your total power, the short current at arcing, or the high peak transients formation of a sparc produces.

In purely theoretical grounds, using microwaves has quite a lot going for it, dispite the large reduction in overall power, dispite only half of the energy ending up in the plasma, the electronics is allready expertly designed and working, the transfomer wont overheat, nothing will be overloaded, and if the arc is at low pressure, its a good way of solving the arcing problem.

You dont have a waveguide to use however, the connection between the magnetron and the oven is short and stubby with the magnetron grafted on so that the oppasit of a waveguide happens to the oven. The waves bounce around randomly so the food is evenly irradiated. Leaks are the main worry, not just to simply seal the gas cavity, but also to get the gas in, arced, cooled and out, and the cooling water youd need to shift potentially 2kw of thermal power out of the system. The first thing youd know about a minor leak would probably be not being able to read this forum anymore. Not using this method is probably a very good plan overall.

If you do still decide to wire 2 of the transformers in series, remeber one side of the secondary is grounded to the core, so the second transformer needs to be connected backwards (and therefore the primary also backwards) to avoid being shorted. Even if you isolate both cores, putting them both forwards potentially overstresses the insulation even without the voltage doublers you were thinking of leaving on, remebering these produce DC from AC, if you thought the electronics/math was hard for neons, you aint seen nothing yet :D . Let us know how long the max arc you can pull is with one transformer. Try using 2 gloves, one for insulation on the inside, one for padding. The padded one cant conduct electricity (obvoisly), but you need the electrical insulation on the inside in case what your touching is hot.

<small>[ March 02, 2003, 05:33 PM: Message edited by: Marvin ]</small>

Tuatara
March 5th, 2003, 03:48 AM
How could you possibly be worried about microwave leakage when your trying to make nitric acid, in order to make chemicals that will gleefully remove your limbs at the slightest provocation :p ?

Seriously though, do check the MIT plasma link above - they built a kW atmospheric pressure plasma torch using microwave oven magnetrons, they got 95% coupling into the plasma from the magnetron (with a simple 3 screw tuner). All I need now is the circulator to avoid magnetron destruction while I tune the waveguide...

BTW microwave don't fly randomly inside the oven - its actually a multimode cavity, and will develop hot spots, or even destroy the magnetron if not loaded (with food).

All that aside (no circulator remember?) I gather from earlier posts that it is best to cool the reacted air as fast as possible to favour the NO2 over NO. So how about putting one electrode partialy into a quartz tube ( from a halogen lamp?), insert the tube endon into the water so the terminal is within arc strike distance of the surface, then make the water the other terminal. When the arc strikes, the air pump can be turned on, forcing the water out of the quartz tube and lengthening the arc to the desired operating point. The air from the arc then bubbles directly into the water and should cool very fast. Only problem is your water is going to get very hot very quickly. Also the ideal terminal material is probably the thoriated tungsten electrodes used in plasma welding torches - I've no idea what they cost though.

Mr Cool
March 5th, 2003, 11:51 AM
1mm diameter TIG electrodes, 150mm long, cost about 20p each, but are sold in packs of ten usually.

photonic
March 5th, 2003, 11:44 PM
I think this question can be made to fit in this topic..
Do any of you know of a source that has rules or information describing reactions involving electricity/electrolysis? I have an experiment in mind, but it's derived solely from curiousity, and I want to find out whether or not its completely assanine before I try it.

Marvin
March 6th, 2003, 02:34 PM
Nitric acid, oxides of nitrogen, and the type and quantity of anything made are the sort of things anyone seriously interested in chemistry are going to have to learn to control sooner or later, and the sooner the better. Building power microwave devices requires specific training and safety equipment. If someone was building a microwave device becuase they wanted to build a microwave device and understand the principles then my attitude would be different than to people that simply want nitric acid - virtually everyone reading this thread I think.

Sometimes 'multimode cavity' is just a fancy way of saying 'box in which the waves fly randomly around'. In the case of domestic microwave ovens, this is exactly what it is. Domestic ovens have no active mode switching, they rely on a combination of the food moving through any hotspots, and the position of the food changing the operating mode, and thus the position of any hotspots to evenly cook the food. The path of the microwaves is random overall, becuase the food is random.

Ive read the MIT paper, it was interesting but I'm not exactly impressed. Some of the ideas they seem to want to claim, have been in use for decades, for example short coupling distances and off the shelf microwave componants in argon plasma cutters. The artical looks like an undergrad project, and reads like an advert. The 95% coupling they are so proud of for example, is achieved at the expense of using 2.45GHz systems that only produce slightly more than half the microwave power they consume, this should have been included and discussed, they clearly know about this problem from the discussion of moving to lower frequency microwaves. (2.45GHz magnetron alone is typically 70% , add power supply and coupling factor, its down to 50-60% in a well engineered system). In the specific case of nitric acid production, the output temperature is too high, reducing this to about 3000C, will almost certainly reduce the coupling factor due to reduced ion concentration. In a low pressure system microwaves have advantages, and the low power efficiency of the microwave system would be offset by the higher nitric acid yeilds. One might ask though, if the low tech system and the high tech system are going to produce roughly the same amounts of nitric acid, why bother with the extra complexity?

Fast cooling isnt about NO2, its about 2NO &lt;-&gt; N2 + O2. Formation of NO is endothermic, so at a high enough temperature it has a reasonable equilibrium concentration in air (about 4.5% at 3000C). Cooling this rapidly to around 1000C or below will reduce the rate of reaction, which is negligable at 1000C, before the equilibium has had a chance to change much. The output gas at this point typically has 2-3% NO in it. When the temperature drops below about 600C the NO starts to oxidise to NO2, max concentration of NO2 occurs (IIRC around 120C), below which the NO2, which has an unpaired electron, making it a radical, paramagnetic and strongly coloured, dimerises to colourless N2O4. How much NO/NOx you think you have in the gas is thus highly dependant on temperature, and this is worth bearing in mind for anyone experimenting.

Tungsten is fantastic stuff for high temperature applications, provided oxygen be excluded as in lightbulbs, inert gas welding etc. I have no idea if this will be a problem for this application, if well cooled the tungsten might not oxidise at all compaired to say copper, used in several furnace designs.

I like the idea for the water electrode chamber, but I'm not sure how well it would work if the water, which would be boiling at the electrode, would interfere with the process. I'm currently trying to understand why the yeilds of existing processes in the french patent, fail to match the birkeland eyed's values, which seem pretty consistant with most arc methods. The best way for atmospheric pressure air arc furnaces seems to be widening the arc magnetically, and blowing air directly through it.

photonic, unless you plan to calculate the redox potentials for every possible reaction in the system, its easier just to ask if its likley to work, I suggest modifying your question to include your exact idea, and if its not directly nitric acid related, you might get away with it anyway. :)

rikkitikkitavi
March 6th, 2003, 03:20 PM
marvin, just a point about the NO2/N2O4 equlibrium.

When it comes to absorbing into water, they are similar.
However the lower the temperature the faster and further the absorbtion goes. The reaction between NO2(aq) and H2O =&gt; HNO3 is strongly exothermic.

I have some pages from Ullmans encyclopedia describing the kinetics
of NO oxidation to NO2 and absorbtion mechanism.

I will scan them with my companys new printer/scanner/copy machine (3 s scanning time for 600 dpi!!!, and it even sends it as an email to my work computer!Isnt technology great!!) and upload them to the FTP. Very interesting reading for anyone seriously interested in designing a HNO3 absorber.

/rickard

Tuatara
March 6th, 2003, 04:00 PM
I suggest thoriated tungsten due to the high melt point, and improved electron emission. Every time I play with my neon txfr I melt the ends of what ever wire is supporting the arc (copper, steel, stainless). Next time I'm near a welding shop I'm going to get some electrodes.

I wasn't sure what the water vapour would do to the rxn in the water electrode system I decribed - can water react with NO2 in vapour phase? It might not matter anyway as the air stream blowing down into the water should provide good local stirring, and carry away any steam.

Sorry about my fixation with microwaves - I'm an electronics engineer and I can't help it :rolleyes: . Your right marvin, most here would not want to play with that stuff so I'll shutup.

<small>[ March 06, 2003, 03:01 PM: Message edited by: Tuatara ]</small>

photonic
March 6th, 2003, 05:55 PM
Well, I'm afraid some chemistry knowledge I don't possess may make this a stupid question. At that risk, here it goes. I was wondering if you dissolved/mixed some KNO3 in water and then ran a current through the water if there would be anyway to obtain HNO3. All the right elements are there but they normally won't react. I was thinking it might be possible to put a basin full of the H2O/KNO3 solution at the bottom of a column and then pull HNO3 vapor off the column at a certain point. LIke they do in industrial gas plants. Any ideas or info that would help?

Thanks

mongo blongo
March 7th, 2003, 11:16 AM
If you can get KNO3 then it will much simpler to just mix it with an acid.

rikkitikkitavi
March 7th, 2003, 02:22 PM
No, it wont work because you will form OH- and H2 at the cathode, and H+ and O2 at the anode. The gases escapes and the ions wanders about in the solution, immedieately reacting, forming water. Thus you will just decompose water into H2+O2.
This reaction is much faster than the evaporation of HNO3, giving a extremely low yield.

A few caveats though:
First, NO3- can be reduced cathodically (it has been used to decompose NO3- containing nuclear waste) giving NOx and NO. Yield is very low though.

Second, more important, if you put the correct ion exchange membrane bewteen the electrodes, creating two separatable chambers, it will prevent the OH- and H+ ions from reacting, thus you will get
KOH at the cathode, and HNO3 at the anode. These liquids can be sepatated. This kind of methods are patented in numerous ways , with various components.

Getting hands on the ion exchanger, let me know if you find any!!

/rickard

10fingers
March 7th, 2003, 05:59 PM
I got my 3 MOTs out and have almost finished a power supply. I must say I like MOTs. They put out some power! The main drawback with them, as others have noted is that they are not current limited and they have a tendencey to get hot very fast with an arc, i.e. smoke in 10 secs.
The one I am using I believe to be 1 kilowatt but am not sure since there is no rating on the oven it came out of, however it is the biggest of the three and I know one of the smaller ones is 500 watt.
I took this xfmr and put a 8 uF 200 Vac capacitor in parallel with the primary, this keeps it from popping the circuit breaker when it is initially powered up. I then took one of the smaller xfmrs and used it as a ballast on the secondary. I shorted its primary and wired the secondary in series with the arc gap. This seems to work pretty well. It can run fairly long without getting too hot. It does get warm though and I think some cooling is necessary, I may try putting it in a container of oil. What would be the best type of oil for this? It needs to be readily available and cheap. Will motor oil work? It must not dissolve the varnish on the wire.
The finished power supply must be able to run for days without overheating.
Initially I wanted to put 2 MOTs in series to double the output voltage but I think this is unnecessary. You were right Marvin. The way I have it configured right now it will draw a horizontal arc of about 1 inch and a vertical arc of about 3 inches. The electrodes get pretty hot though, which is another reason I decided not to use two xfmrs in series, I think it would cause a lot of problems in keeping the electrodes cool. I'm going to try using electrodes that have a lot of copper attached to act as a heat sink, if that doesn't work I'll try using copper tubing with air blown through it. Also, I have some pieces of tungsten carbide that I could braze to the tubing to keep the copper from melting.

xyz
March 7th, 2003, 07:05 PM
I think that you will have no problems using motor oil to cool the small transformer, it shouldn't eat the insulation or anything.

photonic
March 7th, 2003, 07:12 PM
Motor oil will be fine. Just make sure there's no bubbles when you pour it. I used motor oil in a salt water bottle cap and it worked fine until it broke(after about 30 seconds of use). You can use vegetable oil too, but it rots and will start to smell after a while. Also, if you're going to have it in an airtight container you might want to consider some sort of pressure release valve. A perfect container for this sort of thing can be had by going to your grocery store and asking them for their old icing buckets.

Tuatara
March 8th, 2003, 05:40 PM
You might want to use a synthetic motor oil - there should be fewer volatile compounds in it. I've used it successfully to insulate a high voltage transformer for an electric fence - (had some trouble with micro-discharges inside the secondary causing EMI problems - oil fixed it beautifully).

rikkitikkitavi
March 8th, 2003, 07:05 PM
I have uploaded a file, scanned from Ullmans describing the reactions and their kinetics (rate expressions, rate konstants) and the mechanisms involved from NO oxidised to NO2 and furthter reacting NO2 with water .

Very serious reading ( 7 pages) but very useful for anyone seriously interested in designing a maximum yield NO2 reactor and NO2 absorber.
The file is called "Kinetics of HNO3 formation.pdf"

/rickard

Marvin
March 8th, 2003, 09:24 PM
photonic, in fact it does work, provided you have a diaphram of some description, salt bridge, clay pot, specific ion membranes would be better. Heres the catch though, you need obscene currents to make useful amounts of acid, and the concentration is deeply dissapointing, think a few percent at the very most. Ive tried this with a few hundred milliamps (a bridge of any sort rather limits current flow), and it was barely detectable after several hours. I had more luck, though not much more, electrolysing copper into sodium nitrate with a salt bridge, also of sodium nitrate between partitions, forming sodium nitrate with sodium hydroxide in it on one side, and sodium nitrate with copper nitrate in it on the other side. The reaction produced more copper hydroxide than anything else, but the copper nitrate could be fractionally crystalised out of the anolyte, which was incidently, light blue when dilute/cold, and dark green when concentrated/hot.

I came to the conclusion that these methods wernt at all useful. A good ion specific membrane might change that, but would probably cripple current flow for a given area, thus in terms of efficiancy its a big improvement, but in terms of throughput, isnt.

10fingers, MOTs will pull a lot more than the oven was rated for, this is becuase theres no magnetic limiting and becuase low resistance primaries/secondaries reduce the power wasted in the transformer. A transformer that is capable of 500W would only be able to power an oven up to 250W, into a matched load, and would waste half of the power put into the equipment doing it. A tranformer capable of several kilowatts (In terms of peak power, not thermal limitations) would be able to power a 500W microwave while wasting much much less power. The limiting cap should not be in parallel with the primary, this is very bad, and might hike up your electricity bill somewhat though its interesting it stops your breakers blowing, power breakers are not friendly to this research compaired with fuse wire. Find a cap small enough to do the job, and put it in series with the primary, or use the microwave ovens own doubling cap, and put it in series with the secondary. Another transformer secondary in series with the secondary is good, it will be a big impedence, though a lower power one might saturate. Do not short the primary of the ballest, or it starts dissupating power becuase you effectivly have a resistor in the circuit magnified by the transformer turns ratio. Done well, only one or the other should be required, Id put my money on a low enough valued, high voltage cap in series with the secondary. The one in the doubling circuit might be much too big, but try it anyway. An identical transformer secondary in series only doubles the impedence of the high voltage side, you may need a factor of 3 or 4 to reduce the power down to something the transformer can cope with thermally, hense my preference for a series cap, which can be done much more easily. If your arc goes out when your electrodes are only 1 inch apart horizontally, I think you are probably crippling the power too much. So long as the power limiting doesnt waste power, resitivly, or by shorting the system out from mains, as the parallel cap on the primary does (a gross simplification, but it does draw current which you might be charged for depending on how the meter works, and the power gets dissupated not in the cap, but in the lines of the power company), so long as the system you make doesnt waste power, you are only left with a rate of production problem, and not a unit cost problem.

(generally to readers). EE people might be forgiven for thinking what is formed is a tank circuit, but it wont work as one becuase it isnt decoupled from the mains. I seem to recall saying the EE was hard in MOT systems, so no apologies for how difficult any of this might sound, or how badly Ive explained it.

I will need liquid insulation myself for another project, but oweing to the high voltages I need, over 100kV, I'll be using a fluorinated hydrocarbon, like fluorinert. If anyone plans to try stuff over 20kV or so (not for nitric acid), they might like to consider the same thing.

Tuatara, I think you breed patriotic scottish canaries in cages and look to see if theyre trying to put their wings in their ears.

10fingers
March 9th, 2003, 12:05 AM
Ok, thanks a lot for your help Marvin. I'll give your ideas a try.
Right now I just have a test setup to get everything working right and then I have to make an airtight, heat resistant container to put the arc in. Also, I put the transformer in a container of oil but it still gets too hot after 10 minutes or so. I think I have to make the arc gap wider so that there is a little less current flowing.
I did a little test by putting a glass jar over the arc and after a few minutes it was intense with NO2 vapor, I had to turn it off and air out my workshop. It is producing a lot more NO than the neon sign xfmr.

Tuatara
March 9th, 2003, 12:15 AM
It will be interesting to see which is more efficient - neon tfr or MOT. Somebody do some measurements please?

Thanks for that canary suggestion Marvin - I'm off to save the world a whole lotta pain...

Dammit I need a new signature now!

photonic
March 10th, 2003, 09:38 PM
I'm a little confused here on even the basic idea. Is the premise to bubble NOx through water to form HNO3? If this is so, could a small/medium tesla coil be employed to generate the NOx? I have a medium(900 watt) tesla coil that will probably affixiate you after a few minutes in a non-ventilated area. I was thinking(if I even have the basic idea right) that I could enclose the toroid in a rubbermaid bin or something and bubble that bin's output through water.
Forgive my ignorance.

Tuatara
March 10th, 2003, 10:02 PM
You have the idea right! Get a good arc off your coil and your bound to produce NO<sub>2</sub> - try it and see if your container fills with brown gas. Is that 900W wall-plug or output? How long can you run your coil for before it melts?

You'd be in a good position to try an atmospheric pressure glow discharge (APGD) with a tesla coil. Here's a link that explains the concept <a href="http://optics.phys.spbu.ru/~golub/english/barrier.html" target="_blank">Barrier discharge</a>
A quick search with google will provide more info. I don't have the gear to hand, or I'd try it myself.

photonic
March 10th, 2003, 10:20 PM
The 900 watts is the input power(2 30ma 15kv trannies in parallel). I built a pi filter to protect the trannies from the rf current and after I build a safety gap for the capacitor I should be able to run it for quite some time. Right now I only give it about 30-60 second runs. I'd guess I could run it for about 5-10 minutes. I guess I'll have to get a container big enough that the sparks won't strike it or else it probably will melt the bin. The sparks are about 36" long.

Tuatara
March 11th, 2003, 04:05 AM
36" sparks? Excellent fun! Sounds like you need a piece of plastic drain pipe to contain your spark, wont melt if you've got enough air flow round the outside.

Here's a caveat for all who are contemplating putting an arc inside glass - glass becomes quite conductive if you get it hot enough. As I remembered tonight, after putting an arc inside a Pyrex ground joint flask - arc too close to side, heated glass, arc jumps to glass, panic switch off, 'ping' goes the flask with a nice little stress crack where the arc melted the glass :(

Gots lots of lovely NO<sub>2</sub> in about 30sec, well the gas was visibly brown anyway!

10fingers
March 11th, 2003, 03:30 PM
To get any usable amout of NO2 you need a device that can run for days without trouble.
I would not use plastic, but pyrex glass. Even then you do not want the arc too close to the inside wall of the glass chamber. It will break as Tuatara said. One time I tried a quart glass chamber which was very small and the arc start flowing along the inside of it and it broke in less than a minute.
But I would think a tesla coil with that much power would be able to produce NO fairly well. Most Tesla coils have high voltage but little current so it is questionable that the arc gets hot enough to produce NO. The temp ref I have says you need over 3000*K. The more current through the arc the higher the temperature and the greater the amount of NO produced.

photonic
March 11th, 2003, 05:41 PM
Hmm, I'm glad you told me that. Well, I guess that's a days work down the drain. I didn't realize the NOx would eat away at the plastic or does it? Currently my setup is two electrodes inside a pvc chamber with a microwave fan feeding it air. I was going to take the output of that chamber and bubble it through some water in a flask or jar. I'll take pictures of it later today. I was going to feed it with an NST but that will only provide about 30ma of current. I guess I'll put a couple of MOTs in parralel and ballast them with a roll of wire. Also, what's the highest concentration anybody's achieved with this personally?

10fingers
March 11th, 2003, 08:33 PM
I don't think the NO will affect the plastic, it's the heat that I was thinking about. I don't know exactly how your setup works but if your dissipating 900 watts into the arc then the plastic might melt.

Tuatara
March 11th, 2003, 09:14 PM
You'd probably want to pick a plastic with a lower dielectric dissipation factor, PVC isn't terribly good IIRC. Polystyrene or polyethylene are better, but the melt point is rather low. Teflon would be perfect, but where does one find teflon pipe?

photonic
March 11th, 2003, 10:22 PM
This setup isn't for the tesla coil. This is simply two electrodes in a pvc pipe connected to a fan. The electrodes will be connected to either an NST or a MOT. I haven't decide yet. Did somebody say a microwave oven capacitor will be a sufficient ballast for the MOTs? I guess I'll go take pictures of it.

EDIT:
You can see pictures of everything at <a href="http://www.sensorystatic.com/explosives/" target="_blank">http://www.sensorystatic.com/explosives/</a>

<small>[ March 11, 2003, 09:48 PM: Message edited by: photonic ]</small>

Tuatara
March 11th, 2003, 11:58 PM
It really is true that a picture is worth a thousand words! Heres a couple of ideas to stop your pipe melting.

1/ deflect your air flow tangentially to the pipe, so the air swirls. This should keep a layer of nice cold air next to the wall, and hold the plasma in the centre of the pipe.

2/ All that spare thread on your electrodes could be covered with alternating fender washers and nuts, to create a heatsink for the electrodes. In my experience the electrodes will get damn hot, and brass conducts heat far too well.
BTW a 'fender washer' is a big, thin, washer with a little hole - for those unfamiliar with the term.

photonic
March 12th, 2003, 08:03 PM
Hmm, things just aren't going my way this week. I've hooked two different trannies up to the electrode system and neither of them will arc. One was 12kv so it definately should have. The other(approx. 1.5 kv) probably should have as well. Especially when I adjusted the spacing down to about 5 mm. So, I'm not really sure what to do. Has anybody ever seen a MOT that the core wasn't the second output of the secondary? Would shorting the filament winding on the MOT help at all?

10fingers
March 13th, 2003, 02:15 AM
I have three MOTs and they all have one side of the secondary winding connected to the core. The MOT I am trying to use will only arc about 2mm but then you can move the electrodes apart to maybe 40 or 50mm. Shorting the filament winding will not be beneficial, just cut it off.
A 12 Kv NST should be able to arc at least 15mm.

Tuatara
March 13th, 2003, 02:58 AM
Try making your electrodes pointy. The electric field concentration your get around a sharp point aids ionisation, and hence arc strike.

I think with MOTs being only 2kV or so, some form of ignitor circuit will be required to get the long arc started. I'll have a think and post anything I come up with

photonic
March 13th, 2003, 03:43 AM
What about a voltage doubling circuit? Also, couldn't you create a marx generator hooked to a high current power supply. That would provide the voltage and current and if you could get a high current PSU then the charge/fire rate wouldn't be too bad.

On a separate note, I have a friend who has done lots of research on using magnets to form electric arcs. It might be possible to form the arc into a disk with a larger surface area using rare earth magnets or a powerful(about 1 tesla) electromagnet. It probably isn't worth the effort however. I will try again tommorrow and hopefully we'll have some pictures after that.

BTW, I added a slurpee lid to the top of the pvc pipe and duct taped it. Now it looks like a proper miniature industrial model.

rikkitikkitavi
March 13th, 2003, 01:14 PM
using a magnetic field to extend the surface of the arc was used in the original Birkeland process, hence the name "electrical sun" since the arc was disc shaped.

A mot can to start to arc over 1-2 mm, since the breakdown voltage is about 2500 V/mm. However when the air is ionized, it is a completely different matter. Then the arc can be extended further, compare it to a welding machine, where you have an arc of a few mm @ 20-30V .

/rickard

Tuatara
March 13th, 2003, 10:30 PM
A Cockroft-Walton multipier would do for starting. 10 stages would give you 20kV strike voltage at least. Caps would not need to be a large value, just rated at at least 5kV. Could use 1N5408 diodes - series 4 to get 4kV breakdown. These are 1kV 3A rated. Better still use 6A10 diodes - 1kV 10A.

Once the arc is started the current just runs through the diode chain. This trick is often used for starting HeNe lasers.

photonic
March 14th, 2003, 01:02 AM
First trial run is a failure I would say. I ran the "generator" for about 30 seconds and I could smell ozone(or whatever it is, my tesla coil makes the same smell) but that was about it. It did succeed in melting the slurpee lid I have on it.

I have taken pictures and video of it. If a larger video is needed I can just export another one. I was trying to keep the filesize small.

Pictures:
<a href="http://www.sensorystatic.com/explosives/NOxGenerator%20001.jpg" target="_blank">http://www.sensorystatic.com/explosives/NOxGenerator%20001.jpg</a> - Picture of Arc inside Tube

<a href="http://www.sensorystatic.com/explosives/NOxGenerator%20002.jpg" target="_blank">http://www.sensorystatic.com/explosives/NOxGenerator%20002.jpg</a> -
Picture with No Arc

<a href="http://www.sensorystatic.com/explosives/NOxGenerator%20003.jpg" target="_blank">http://www.sensorystatic.com/explosives/NOxGenerator%20003.jpg</a> -
Entire Setup

Video:
<a href="http://www.sensorystatic.com/explosives/NOxGenerator.mpg" target="_blank">http://www.sensorystatic.com/explosives/NOxGenerator.mpg</a> -
Setup Running for 35 Seconds. The arc gets kind of wild when the fan is on full power.

I ran the setup for about 30-60 seconds normally. The pictures and video are without the top capped. During tests I put a box over the lid to try and keep gases in. Any suggestions?