Sciencemadness Discussion Board
Not logged in [Login ]
Go To Bottom

Printable Version  
 Pages:  1  ..  5    7    9  ..  13
Author: Subject: Birkeland-Eyde reactor for making nitric acid.
densest
National Hazard
****




Posts: 359
Registered: 1-10-2005
Location: in the lehr
Member Is Offline

Mood: slowly warming to strain point

[*] posted on 28-10-2007 at 22:59


A random thought about some variant ways to get air enriched with oxygen up to 2000C or so. Electric arcs are a very intense heat source, but it's easy to overheat the electrodes and melt them. This has probably been tried and discarded, but.....

There are a few materials which remain solid in an oxidizing atmosphere above 2000C. Limiting the search to oxides for simplicity, there are CaO, TaO2, Ce2O3, La2O3... ThO2 is very nice at over 3000C. Would passing a N2/O2 mix through a lattice heated to 2000C be likely to produce NO in any significant quantity?

I tried producing a little fused La2O3 with a propane/oxygen torch. The flame is hot enough. At the required temperature, radiation carries away a lot of energy, so fusing the interior of a large blob of powdered oxide is much easier than working from the outside. If you try this, wear at least grade 5 welding eye protection. Grade 8 would be better. It's very bright and emits huge amounts of infrared and sufficient ultraviolet to be harmful.

A further variant to the solid phase reactor could be to heat the oxide blobs electrically once they're heated enough to be conductive. Tungsten or molybdenum electrodes protected from air with an oxide layer would be able to stand the heat.

Perhaps a set of Coleman lamp mantles would also serve... they used to be doped with thorium oxide. I forget what the generic term for the lamp is - it evaporates pressurized gasoline into a gas mantle lamp.
View user's profile View All Posts By User
12AX7
Post Harlot
*****




Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline

Mood: informative

[*] posted on 29-10-2007 at 00:48


BTW, once you heat up ceramic enough, it conducts. How much depends on the ionic mobility. You might consider an Al2O3 or mullite matrix with some SiO2 and Na2O glass phase. Na ions of course become mobile at reasonable temperatures. Then a bit of noble or refractory metal or conductive oxide could be used to maintain continuity.

After oxidation, vaporization and sputtering are probably the next things to worry about. High-Z oxides, high binding energies and high boiling points would be preferred.

In fact, WC and SiC satisfy conductivity, Z, MP and oxidation resistance, variously. A composite containing W, C, Si and SiO2 might be very practical.

Tim




Seven Transistor Labs LLC http://seventransistorlabs.com/
Electronic Design, from Concept to Layout.
Need engineering assistance? Drop me a message!
View user's profile Visit user's homepage View All Posts By User This user has MSN Messenger
densest
National Hazard
****




Posts: 359
Registered: 1-10-2005
Location: in the lehr
Member Is Offline

Mood: slowly warming to strain point

[*] posted on 29-10-2007 at 15:43


Quote:
Originally posted by 12AX7
BTW, once you heat up ceramic enough, it conducts. How much depends on the ionic mobility. You might consider an Al2O3 or mullite matrix with some SiO2 and Na2O glass phase. Na ions of course become mobile at reasonable temperatures. Then a bit of noble or refractory metal or conductive oxide could be used to maintain continuity.

After oxidation, vaporization and sputtering are probably the next things to worry about. High-Z oxides, high binding energies and high boiling points would be preferred.

In fact, WC and SiC satisfy conductivity, Z, MP and oxidation resistance, variously. A composite containing W, C, Si and SiO2 might be very practical.

Tim


All good points. According to an ancient Chem Rubber Handbook, the carbides can't stand a hot oxidizing environment, and all the silicates I looked at melt below 2000C.

High-Z materials that look good: Zirconium oxide looks good (solid to 3000C) and there are companies which fabricate it :P:P Various lanthanide oxides are solid up to 2200-2400. Thorium oxide, of course, would be ideal.

By comparison, my oxygen/propane torch which makes enough NOx to require heavy forced ventilation is about 2800C, though it might generate enough free radicals to make that figure meaningless.

ZrO2 conducts electricity starting somewhere about red heat. If protected from air by (for instance) a coating of La2O3 or other convenient oxide (I have a kilo or so of it) melted onto them, tungsten and molybdenum can be used as electrodes and supports well into the 2000s. None of this equipment can stand getting wet; one must be careful about that.

Now to find the time and energy to actually try some of this :( It's fun to talk about though. Perhaps an extremely simplified version to do a quick test will come to mind.
View user's profile View All Posts By User
Twospoons
International Hazard
*****




Posts: 1147
Registered: 26-7-2004
Location: Middle Earth
Member Is Offline

Mood: A trace of hope...

[*] posted on 29-10-2007 at 16:46


If you are feeling really adventurous (and know what you are doing) you could try a microwave driven plasma.

Microwave plasma torch

Microwave coupled plasmas can achieve very high temperatures, with no electrodes to melt, or wear away. Neither simple nor cheap, unless you have access to some amazingly good surplus outlets.




Helicopter: "helico" -> spiral, "pter" -> with wings
View user's profile View All Posts By User
497
International Hazard
*****




Posts: 778
Registered: 6-10-2007
Member Is Offline

Mood: HSbF6

[*] posted on 29-10-2007 at 19:03


that is an interesting idea, using the oxides. but i really doubt it would easily achieve the temperatures needed for efficient conversion (i've read that the equilibrium is about 5% at 3000 C), and looks like it would be a considerable amount of effort to get a working setup. probably worth a try if you have the time and materials, but as far as i can see a plain old arc works pretty well. at least with a lower power one like i have, it seems to put very little heat into the electrodes at all, as i said before they have no signs of wear. the microwave plasma torch is very interesting, it looks like it could work very well indeed. unfortunately i lack the resources to experiment with it... one problem i've come to is how to get the very dilute NO2 to dissolve in water efficiently and yet make fairly concentrated acid. I had the idea of cooling the air-NO2 mixture causing it to form N2O4 which would then condense (or maybe freeze) and could be added to water in the needed proportions. would this work?
View user's profile View All Posts By User
497
International Hazard
*****




Posts: 778
Registered: 6-10-2007
Member Is Offline

Mood: HSbF6

[*] posted on 15-11-2007 at 12:10


so i've finnaly done some more testing. using a oil furnace transformer arc with an airpump pushing the gas through a bubbler, i've produced 600ml of water the reads 3500 microseimes (1750 ppm according to my meter) which gives me about a gram of HNO3 in theory. it only ran for maybe an hour. I think i'd be happy with 24 g HNO3 per day. and i can always find a bigger transformer. i am also happy the the NO2 is diffusing into the water effectively, i'm only getting a loss of maybe 5-10%. i'll run it some more later today and see how it does.
View user's profile View All Posts By User
tentacles
Hazard to Others
***




Posts: 191
Registered: 11-11-2007
Member Is Offline

Mood: No Mood

[*] posted on 15-11-2007 at 21:57


497: Interesting results, can you post pictures of your setup? I'm also interested in any results you might have for the N2O4 production - where I live the winters are long and cold, it'd be nice to have a use for it being -30C outside!

I wonder if you could run a B-E reactor made of glass continuously here, outside in the winter? Would the external cold cool the plasma arc excessively? The output gasses could easily be run inside, or into heated (above freezing anyway) jars of water. The cold would also help the water retain the NO2.
View user's profile View All Posts By User
12AX7
Post Harlot
*****




Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline

Mood: informative

[*] posted on 15-11-2007 at 22:10


I don't think outside air temperature will bother you in the least: with the plasma at circa 10kK (i.e., 10,000 K), the difference between 250 and 300K doesn't seem all that important. :)

N2O4 condensed in the environment reminds me of metal-melting friends in frigid locations... always a shame when your propane tank freezes up without even having drawn a drop of gas off it!

Tim




Seven Transistor Labs LLC http://seventransistorlabs.com/
Electronic Design, from Concept to Layout.
Need engineering assistance? Drop me a message!
View user's profile Visit user's homepage View All Posts By User This user has MSN Messenger
497
International Hazard
*****




Posts: 778
Registered: 6-10-2007
Member Is Offline

Mood: HSbF6

[*] posted on 16-11-2007 at 17:54


so i've found that the added pressure from the second bubbler really increases the diffusion of NO2 into H2O. with my pathetic little setup i was getting about 40 microseimen increase per minute in 500 ml of water (ran for ~3.3 hours and ended at 8000 microseimens) i then boiled off most of the water and ended up with about 2 ml of pretty concentrated acid that reacted with copper fairly quickly. so i calculated that at about .5 g per hour. not bad. but i'm just a little too impatient for that so i'm going to see if i can find a good microwave transformer at the dump. i know i'll have to add a big resistor to keep the thing from melting but thats ok, power is cheap. and maybe i'll use
the heat to do something useful like concentrate the resulting acid...

with a microwave transformer i imagine the problem will become dissipating all that heat from the arc. so this is my idea for an arc chamber: stainless steel tube maybe 2 cm dia and about 15 cm long with a cap welded on the bottom end. then have one electrode welded to the center of the cap, the other one mounted on a movable insulator (to allow adjustment of the gap.) for electrodes i'm thinking stainless rod with 5mm balls the on the hot ends. the air would travel through linearly, i think this is more effective than just having an arc in a turbulent container of air. the steel tube would have a boiling water jacket to cool it. i think it'll work, any comments/improvements?

as far as freezing the N2O4 goes, i have yet to do a good test but it seems to diffuse into water quite well at room temp, so i don't see the need.

anyone know a good way to measure the concentration of nitric acid? i have a pH/conductivity meter, but i don't trust it's accuracy above very dilute acid.

i'll post pics of my little oil furnace transformer setup as soon as i have time.

[Edited on 16-11-2007 by 497]
View user's profile View All Posts By User
497
International Hazard
*****




Posts: 778
Registered: 6-10-2007
Member Is Offline

Mood: HSbF6

[*] posted on 18-11-2007 at 23:09


so ran it for ten hours today. the first diffuser is about 120 ml volume ended up with roughly 25000 ppm (25 grams per liter) not bad... 8 grams per kw
View user's profile View All Posts By User
497
International Hazard
*****




Posts: 778
Registered: 6-10-2007
Member Is Offline

Mood: HSbF6

[*] posted on 24-11-2007 at 01:27


i just got an old microwave. perfectly good as far as i can see. i ripped it apart, the transformer is nice, works good, amazingly powerful compared to my little OFT. also got a really nice fan out of it.

the problem now is how to ballast it. with no ballast it gets hot real fast as i expected, so i hooked up 8 10 ohm 25 watt resistors but i'd need quite a few more to get them to not overheat. i've heard i can use another MOT as an inductor to ballast it but i don't have one. i suppose it's more trips to the dump then...

another problem is how i'm going to make a chamber that can handle that much power. i'm thinking water cooling.

anyone know a good use for a magnetron? i'm not sure what to do with it.
View user's profile View All Posts By User
12AX7
Post Harlot
*****




Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline

Mood: informative

[*] posted on 24-11-2007 at 03:41


Since you're probably not interested in the 10cm ameteur band, you might as well smash it. Inside, you will find a steel housing, aluminum cooling fins, copper anode, ceramic (probably pink alumina, unless it's an old one made with beryllia, in which case don't crack it!) and the cathode, which is probably "rare-earth" oxide coated tungsten.

Tim




Seven Transistor Labs LLC http://seventransistorlabs.com/
Electronic Design, from Concept to Layout.
Need engineering assistance? Drop me a message!
View user's profile Visit user's homepage View All Posts By User This user has MSN Messenger
497
International Hazard
*****




Posts: 778
Registered: 6-10-2007
Member Is Offline

Mood: HSbF6

[*] posted on 24-11-2007 at 16:30


so if i wanted to wind an inductor to ballast the arc, can anyone tell me how i could figure out the number of turns, diameter of wire, etc.? i don't have another MOT to use.
View user's profile View All Posts By User
12AX7
Post Harlot
*****




Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline

Mood: informative

[*] posted on 24-11-2007 at 20:39


If you don't have a laminated iron core, you'll be up in the 10,000 turns range, of some very substantial wire (heavier than rated, since the required length adds so much resistance). Better find another MOT methinks!

Tim




Seven Transistor Labs LLC http://seventransistorlabs.com/
Electronic Design, from Concept to Layout.
Need engineering assistance? Drop me a message!
View user's profile Visit user's homepage View All Posts By User This user has MSN Messenger
497
International Hazard
*****




Posts: 778
Registered: 6-10-2007
Member Is Offline

Mood: HSbF6

[*] posted on 27-11-2007 at 15:57


hmm well i calculated that about 350 feet of 20 guage magnet wire in a coil with no core will produce an impedence of ~12 ohms (plus 3 resistance) at 60 hz. that should be enough to keep my current down to 8 amps, which is no more than it would use when running the magnetron, and i know it can handle that for an extended period of time. maybe i calculated wrong. but even if i didn't i'd rather use a MOT, much easier. still havn't found another one yet though..
View user's profile View All Posts By User
12AX7
Post Harlot
*****




Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline

Mood: informative

[*] posted on 27-11-2007 at 23:42


Yeah, and 20 AWG is good for how much current?

Try 20 AWG nichrome...

Tim




Seven Transistor Labs LLC http://seventransistorlabs.com/
Electronic Design, from Concept to Layout.
Need engineering assistance? Drop me a message!
View user's profile Visit user's homepage View All Posts By User This user has MSN Messenger
497
International Hazard
*****




Posts: 778
Registered: 6-10-2007
Member Is Offline

Mood: HSbF6

[*] posted on 28-11-2007 at 12:11


well i'm not sure exactly what current its said to handle but i figured that it would have about 120 watts of resistve heating. i figure that could be handled, maybe with a fan. i suppose you could just have it submerged in water... i hope i can get some soon.
View user's profile View All Posts By User
UncleJoe1985
Hazard to Self
**




Posts: 88
Registered: 30-9-2008
Location: Sunnyvale, CA
Member Is Offline

Mood: No Mood

[*] posted on 30-9-2008 at 21:44
Increasing NO2 Formation and Dissolving


I've attempted running a Birkeland Eyde reactor with a 9kv, 30mA NST, but got very poor results: ~3-4% acid max. I'm planning to resume after stopping for 5 months. Rather than increase my power supply, which I think is sufficient, I want to improve the efficiency. Here's what I think was wrong with my previous setup (I'm pretty confident my arc configuration is good).

1. Not enough time is given for NO2 to form before bubbling it through water. I just directly lead a tube from the top of my 1.5L reactor to the bottom of the 1st water jar. I heard from someone else who had better success put a second larger container in between to allow time for the conversion, which he claimed was noticeable based on the darker color in it. I then found out that NO2 has a much higher solubility than NO, which might explain the low yield. Can anyone confirm this or know the time it takes for NO to NO2 conversion?

2. Poor NO2 absorption. The absorption columns used in the Ostwald process are 5 stories high, so that says something about the rate of dissolution. I was thinking of adding something porous to increase the time the bubbles react with water, but I can't think of anything that's acid resistant. Also, does anyone know how important temperature is on absorption rate? I heard 50C is good, which seems
to trade off solubility for reaction speed.

Thanks to everyone who shared their experience above.

[Edited on 30-9-2008 by UncleJoe1985]
View user's profile View All Posts By User
Magpie
lab constructor
*****




Posts: 5939
Registered: 1-11-2003
Location: USA
Member Is Offline

Mood: Chemistry: the subtle science.

[*] posted on 1-10-2008 at 13:13


Quote:

I was thinking of adding something porous to increase the time the bubbles react with water, but I can't think of anything that's acid resistant.


Adding inert mass (like small pieces of broken porcelain) to the absorption vessel will not increase residence time of the gas, but decrease it. It will, however, increase turbulence (and hence mixing), so might improve the absorption rate that way.

Edit 1: In fact all industrial size absorption columns have some kind of packing, sieve trays, or bubble-cap trays for just that reason: increase interfacial contact between liquid and gas, and thereby improve the absorbtion rate.

Edit 2: I just checked my "Chemical Process Industries" book by Shreeve for making 65% nitric acid. The absorption tower for commercial units is, as you say 40 feet high. It uses bubble-cap trays with cooling coils on each tray. Column pressure is 100 psig. Column feeds are cooled water, NO, and cold air. One paper states that in the authors' opinion the following reaction is the rate limiting step (not the oxidation of NO), and that "...this is an absorption phenomenon":

3NO2 + H2O <--> 2HNO3 + NO

I think I can see why this thread is so long. It would be quite a challenge for the home chemist to simulate those conditions.

Edit3: Upon further reading of Shreeve, he implies that 55% acid can be made at atmospheric pressure. The column diameter is 5.5 ft. Production rate is 55 tons/day. So, computing a rough value for the liquid residence time:

flow = (55 tons/day)(2000 lbs/ton)(day/24 hr)/[(8.34 lb/gal)(1.2)(7.5 gal/ft3)] = 61 ft3/hr

Column volume (disregarding internals) = (3.14/4)(5.5 ft)^2(40 ft) = 950 ft3

Therefore, residence time (roughly) = 950 ft3/(61 ft3/hr) = 15.5 hr

Edit4: The assumption of a column full of liquid is erroneous for a column using bubble-cap trays. I correct this in my next post.





[Edited on 2-10-2008 by Magpie]
View user's profile View All Posts By User
UncleJoe1985
Hazard to Self
**




Posts: 88
Registered: 30-9-2008
Location: Sunnyvale, CA
Member Is Offline

Mood: No Mood

[*] posted on 1-10-2008 at 17:47


"It uses bubble-cap trays"

http://www.jaeger.com/BubbleCap.jpg - Ingenious! The cup will trap the gas until it overflows. Now if I can only buy or build one myself, I definitely would.

Thanks for looking up Magpie, but I don't think your calculations are meaningful. 15 hours would be the time it takes for liquid on the bottom of the column to rise to the top, assuming the column was initially empty and that nothing mixes. The correct calculation should involve the volume of NO2 passing:

55 ton/day => 792 K Moles HNO3 /day

Assuming, the reaction path is, 3 HNO2 + H2O => 2HNO3 + NO, the gas flow is:

2.66 * 10^7 (L NO2 /day) => 308 L/s

Using your dimensions of the absorption column, the gas will travel at
308L/s / (bottom area of column) = 14.0 cm/s

column height / 14.0 cm/s = 87.4 s

Still, that's a long time for absorption. This probably explains why so many people before have complained about low yield. Any ideas on how to increase absorption compactly? I was thinking of using a rain forest mist generator, but am worried about it not being acid resistant, assuming it was a closed cycle (i.e. respray the acid in the jar).

I proposed earlier to use hot water to increase the absorption rate. Even though they said to use cold water, I think that's just to cool the entire setup, not what the equilibrium temperature is. My chemistry is mediocre, so does anyone know if increasing the temperature improves the absorption rate?

Also, I did try using several plastic mesh pot scrubbers as a filter, but it was hard to eliminate all gaps with the jar. Large bubbles would just go through the gaps.


[Edited on 1-10-2008 by UncleJoe1985]
View user's profile View All Posts By User
Magpie
lab constructor
*****




Posts: 5939
Registered: 1-11-2003
Location: USA
Member Is Offline

Mood: Chemistry: the subtle science.

[*] posted on 1-10-2008 at 20:38


Quote:

Thanks for looking up Magpie, but I don't think your calculations are meaningful. 15 hours would be the time it takes for liquid on the bottom of the column to rise to the top, assuming the column was initially empty and that nothing mixes. The correct calculation should involve the volume of NO2 passing:


You are right about my calculation not really being meaningful. In thinking about how a bubble-cap tray really works, the tray (of which there are many) is the only place there is liquid holdup, ie, each tray holds up enough liquid to overflow its downcomer weir. The rest of the column is gas. The liquid flows in from the top then cascades down through the trays. The gas enters at the bottom and flows up through the bubble cap slits, sparging through the liquid held up on the tray. So a more reasonable estimate of the liquid volume in the column might be 1/10th of the column volume. This would reduce the residence time to 1.5 hrs. I will see if I can look this up, or maybe someone else can give a better estimate.

A bubble-cap tray is not practical for small scale work. I'm thinking that a packed column might be about the best a home chemist could make. Again, the residence time is going to be the problem. There's no way you are going to have a 40 ft column (right?).

The usual chemists' expedient to gas absorption is a sparge tube in a glass column, ie, a "gas absorber." Shown below is a picture of mine. Gas residence time is a problem here. You just have to keep sparging away until you get an acceptable concentration of acid.

The cold temperatures are necessary for efficient gas absorbtion. The colder the solvent, the more soluble the gas.

All this theory and technology of gas absorption can be found in Perry's Chemical Engineers' Handbook, available in any technical library.

gas absorber.jpg - 95kB
View user's profile View All Posts By User
jarynth
Hazard to Self
**




Posts: 76
Registered: 12-8-2008
Member Is Offline

Mood: No Mood

[*] posted on 18-10-2008 at 04:58


What materials would be best suitable for the electrodes in an amateur setting?

US1462987 suggests a 50:50 mixture of O2/N2 for better yield... In a oxygen-poorer atmosphere (such as air), the equilibrium should be more strongly shifted towards the monoxide. What catalysts cold be used to reverse the disproportionation? Maybe V2O5 as in the lead chamber process could do the job.

[Edited on 18-10-2008 by jarynth]
View user's profile View All Posts By User
watson.fawkes
International Hazard
*****




Posts: 2793
Registered: 16-8-2008
Member Is Offline

Mood: No Mood

[*] posted on 18-10-2008 at 09:20


Quote:
Originally posted by jarynth
What materials would be best suitable for the electrodes in an amateur setting?
I suggest tungsten electrodes for TIG (GTAW) welding. They're readily available, modestly priced, and have well-proved resistance against degrading in ordinary atmosphere. (The shielding gas protects the weld, not the electrode.)
View user's profile View All Posts By User
12AX7
Post Harlot
*****




Posts: 4803
Registered: 8-3-2005
Location: oscillating
Member Is Offline

Mood: informative

[*] posted on 18-10-2008 at 09:37


I find that suspect. Tungsten filaments *burn* in air. In fact, they literally produce a flame!

Filaments are a whole lot thinner than rods, much like powders are pyrophoric when the bulk material isn't, but still...

Tim




Seven Transistor Labs LLC http://seventransistorlabs.com/
Electronic Design, from Concept to Layout.
Need engineering assistance? Drop me a message!
View user's profile Visit user's homepage View All Posts By User This user has MSN Messenger
dann2
International Hazard
*****




Posts: 1523
Registered: 31-1-2007
Member Is Offline

Mood: No Mood

[*] posted on 18-10-2008 at 12:33
Atomizer for absorbing NO, yea!!


Hello Folks,

I am totally new in this thread.
Was just wondering about the problem (perhaps it's not a problem) of getting the NO to dissolve and not go to the atmosphere.
There are devices sold in stores that atomise water for to give a display somewhat like what you get with a bowl of liquid Nitrogen in humid conditions. (ie. a creeping mist flowing out of the bowl).
If this device were incorporated into the Birkeland-Eyde it would surely capture all the NO produced via the water mist.
The membrane of the atomizer would need to be acid resistant. Have they got ceramic membranes?, I think they have, or you could pipe in the mist from a seperate chamber.
Perhaps the whole idea is a bit too 'Heath Robinson' (ridicoulsy complicated) with the bubbling system doing the job.

Dann2
View user's profile View All Posts By User
 Pages:  1  ..  5    7    9  ..  13

  Go To Top