Sciencemadness Discussion Board

Preparation of elemental phosphorus

 Pages:  1  ..  3    5    7  ..  15

watson.fawkes - 3-12-2009 at 06:27

Quote: Originally posted by unome  
isn't Si above P in the table
If, by "above", you mean "next to".

Please, it is very fast to just look at the periodic table rather than making half-assertions so easily verified. Therefore, please make yourself a bookmark. Here's the periodic table at Wikipedia.

12AX7 - 3-12-2009 at 07:11

Si is even easier to purchase in powdered form, and aluminum can be alloyed with magnesium, forming an intermetallic much weaker than silicon.

Silicon will work fine as a reducing agent, simply because P is gaseous and nothing else in the reaction is.

Tim

unome - 3-12-2009 at 22:03

Quote: Originally posted by watson.fawkes  
Quote: Originally posted by unome  
isn't Si above P in the table
If, by "above", you mean "next to".

Please, it is very fast to just look at the periodic table rather than making half-assertions so easily verified. Therefore, please make yourself a bookmark. Here's the periodic table at Wikipedia.


No, in above I mean it is capable of reducing things that are reportedly able to reduce phosphates - like Mg for example...

In terms of the periodic table itself, I hadn't actually looked so you have me there - fair cop:)

Sedit - 4-12-2009 at 07:56

Sooner or later I do highly wish to produce a retort able to perform these type of operations but I can not find a single reference of anyone online doing such a thing. It there anyone here that has experiance with making a ceramic retort and suggestions on the shape and design on one.

The main issue that keeps popping up in my mind it the fact that these will be a one time use thing and it would kind of suck to put my heart into sculpting a retort just to have it destroyed on the first, more then likely failed, attempt. Can anyone suggest any designs for one please.

Im considering also about making something that I could slip into one of the port holes on my kiln and this seems like it could be ideal considering using cast iron pipes fitted correctly leading to a bucket of water at the base of the kiln. I worry about expansion with the heat though and wether or not the threads will hold up under such stress because I really do not want phosphous vapors being expelled all over the place in my ceramic workshop.

I have to go for now so ill post a few further questions a bit later.

~Sedit

blogfast25 - 6-12-2009 at 12:51

Garage Chemist:

Firstly referring to len1, there was no 'tone', it was merely a question. It was your experiments that inspired me to try the TCP (tertiary phosphate) because it's amply available as bone ash.

It does surprise me these phosphates are reduced to phosphide and not to P [0] in the absence of silica...

But I can't even get the TCP thermite to light up properly. The last formulation was:

Ca3(PO4)2.............1 mol
CaSO4....................0.5 mol
Al...............................14/3 mol

It lit up (using KClO3/Al ignition mix) but then fizzled about half way.

Now I'll try and replicate your experiment, including silica. What is a good OTC source of Ca(PO3)2? I have none available right now.

Could monoammonium (dihydrogen)phosphate be converted to ammonium metaphosphate by heating?

NH4H2PO4 ---> NH4PO3 + H2O

Alternatively maybe displacing the ammonium in monoammonium (dihydrogen) phosphate by sodium with soda:

2 NH4H2PO4 + Na2CO3 ---> Na2H2PO4 + 2 NH3 + CO2 + H2O


[Edited on 7-12-2009 by blogfast25]

blogfast25 - 8-12-2009 at 07:54

Later on today I'm gonna try and obtain NaH2PO4 by dry-distilling NH4H2PO4 with NaCl:

NH4H2PO4 (s) + NaCl (s) ---> NaH2PO4 (s) + NH4Cl (g)

And by further heating kill (hopefully) two birds with one stone:

NaH2PO4 (s) ---> NaPO3 (s) + H2O (g)

Then the metaphosphate will be mine (mwahahaha!)

garage chemist - 8-12-2009 at 09:36

You can just mix 1 mol monoammonium phosphate with 1 mol NaOH or 0,5 mol Na2CO3 in solution, and boil this down.
NaNH4HPO4 (microcosmic salt) decomposes to NaPO3 upon heating. Don't you know the analytical bead test with the magnesia stick or platinum wire and this salt, for identifying heavy metals?
http://en.wikipedia.org/wiki/Microcosmic_salt
It is used just like Borax in the http://en.wikipedia.org/wiki/Borax_bead_test .
There's only a german page on this, no english one:
http://de.wikipedia.org/wiki/Phosphorsalzperle .

blogfast25 - 8-12-2009 at 13:30

Quote: Originally posted by garage chemist  
You can just mix 1 mol monoammonium phosphate with 1 mol NaOH or 0,5 mol Na2CO3 in solution, and boil this down.
NaNH4HPO4 (microcosmic salt) decomposes to NaPO3 upon heating. Don't you know the analytical bead test with the magnesia stick or platinum wire and this salt, for identifying heavy metals?
http://en.wikipedia.org/wiki/Microcosmic_salt
It is used just like Borax in the http://en.wikipedia.org/wiki/Borax_bead_test .
There's only a german page on this, no english one:
http://de.wikipedia.org/wiki/Phosphorsalzperle .


Which is what I what I did yesterday and finished off today (I used Na2CO3). I'd never heard of NaNH4HPO4 before but found it in Holleman's 'Anorganische Chemie' yesterday. I boiled down the solution and dried the product at 250 DC. It's a bit hard to extract from the pyrex jug, so I've redissolved it and will recrystallize it more gently, tomorrow.

But the dry distillation of an equimolar mix of NH4H2PO4 and NaCl seems to be the most promising and mess-free way to convert my NH4H2PO4 into NaPO3 [(NaPO3)6, presumably?], IMHO.

I heated about 25 g of the mix in a thick stainless steel cup on medium heat on my lab gas cooker. It melted immediately, starts bubbling and thick fumes of salmiak came off for about 1/2 hour. The mass had by then started to solidify somewhat (because of the higher MP of the metaphosphate?). I cranked up the heat and give it another half hour, all gas evolution seemed to have stopped by then. After cooling I weighed the crucible again: the yield of 13.6 g of product was within 2.5 w% of the stoichio amount (13.95 g) based on:

NH4H2PO4 + NaCl + heat ---> NaPO3 + NH3 + HCl + H2O

The product is a hard, crystalline mass, slightly green due to iron pick-up. It grinds down to a fine powder easily in a granite mortar.

It dissolves fairly easily in warm water, with some turbidity (iron phosphate?) and the solution (unknown concentration) is of pH = 2.4, a bit acidic for my liking (but it would indicate there's no tertiary orthophosphate). On adding some pellets of NaOH and dissolving them, no smell of NH3 was discernable, not even a whiff.

Added to egg white, the egg white solidifies (a typical test for meta phosphate, acc. Holleman). It does look like this product is NaPO3 (or its hexamer).

So tomorrow, weather allowing, I will test:

2 NaPO3 + SiO2 + 10/3 Al ---> Na2SiO3 + 5/3 Al2O3 + 2 P

in thermite conditions.

Thanks also for the links...

blogfast25 - 9-12-2009 at 09:41

I mixed up about 10 g of the stoichiom. metaphosphate/silica/aluminium powder mixture and put it in a small, thick walled stainless steel crucible.

Two attempts at lighting it as you would a thermite, in this case using KClO3/Al mix as initiating mixture and Mg ribbon as fuse, but no joy at all: in both cases the mixture died after the initiating mixture was burned up.

Frustrated by then, I put the crucible on my gas cooker on medium-high heat direct propane flame and within a few minutes I saw clear phosphorescence occur near the walls of the crucible: a fairly bright yellow-green glow, unmistakably phosphorescence because the experiment was carried out in the dark, except for my torch light.

I called in to get my daughter to watch it but by the time I came back the crucible had caught a gentle fire: the phosphorus was being oxidised by air and fumes of presumably P2O5 could be observed. By the end of the reaction the entire mass, by now presumably mainly sodium silicate and alumina, was glowing a bright red. I'll hack into it later, when it's cooled down.

I think that also explains why lighting the mix as a thermite failed: I believe the reaction enthalpy is just borderline to make it self-sustaining. This could be remedied with small amounts of heat-booster mix.

But the mixture as such is probably very suitable for producing phosphorus in a retort with external heating to get things going. I have just the thing for that but I am clean out of aluminium powder right now...

The residue in the crucible is a hard, porous mass with a slightly unpleasant smell, reminiscent of the smell I get from dissolving reactive metals in dilute mineral acids.

%%%%%%%%%

The product from the 'microcosmic salt' route has been further treated to be redissolving it, then boiling down to a sticky, syrupy consistency. That was then transferred to a steel pan and heated on the gas cooker on high heat for a couple of hours. A hard, white-greyish crystalline mass results, which was recovered and ground down.

An approx. 5 w% solution (turbid, though) of the product in water gave a pH of 4.3. The solution also solidifies egg white (denaturation of albumin?).

Adding pellets of NaOH to the solution does not reveal any ammonia, not even the merest whiff.

Presumably this product is very similar to the one obtained via the NH4H2PO5 + NaCl route.


[Edited on 9-12-2009 by blogfast25]

garage chemist - 10-12-2009 at 07:01

Good work!
I assume you have read Strepta's method a few pages back (he attached it as a Word file)?
He obtained about 50% P4 yield by increasing the amount of aluminum above theoretical (although I would be careful with that, you don't want to bind the P as phosphide again- perhaps increase the amount of SiO2 as well) and using pyro grade "black" aluminum powder.
I would also use diatomaceous earth (calcined before use) as the SiO2 source, as it's finely divided and very reactive (but not pure SiO2).
What kind of Al and SiO2 did you use?

Strepta also used CO2 as a protective atmosphere inside the test tube, but I would advise against that, as Al vigorously reacts with it, forming CO, C and carbide. Perhaps that's why he had to increase the amount of Al?
N2 is also not an optimal choice (forms AlN).
If Ar is not available, H2 could perhaps be used.





blogfast25 - 10-12-2009 at 08:02

Quote: Originally posted by garage chemist  
Good work!
I assume you have read Strepta's method a few pages back (he attached it as a Word file)?
He obtained about 50% P4 yield by increasing the amount of aluminum above theoretical (although I would be careful with that, you don't want to bind the P as phosphide again- perhaps increase the amount of SiO2 as well) and using pyro grade "black" aluminum powder.
I would also use diatomaceous earth (calcined before use) as the SiO2 source, as it's finely divided and very reactive (but not pure SiO2).
What kind of Al and SiO2 did you use?

Strepta also used CO2 as a protective atmosphere inside the test tube, but I would advise against that, as Al vigorously reacts with it, forming CO, C and carbide. Perhaps that's why he had to increase the amount of Al?
N2 is also not an optimal choice (forms AlN).
If Ar is not available, H2 could perhaps be used.



Thanks!

I've just looked at Strepta's method and it's very neat but I'm not interested in such small amounts of P: if I'm gonna go to the trouble I want a quantity I can at least comfortably continue experimentation with: halogenation perhaps.

I was thinking CO2, Ar or CO2/Ar mix. I'm surprised Al reacts with CO2 in those particular conditions, it sounds plausible he increased the Al level to 'account' for what reacted with the CO2.

I'm not going anywhere near H2 with an open flame with a apparatus that's not certified gas tight..

I used a 200 mesh Al grade (not German black) and silica extracted from an old fashioned kitchen scouring product: the abrasive in there is merely finely ground purified sand. Get rid of the soap by leaching and the silica is left behind. I'm guessing it's about 100 mesh. I've used it many times before in SiO2/Al thermites. But I've also used finely ground desalinated and decarbonated beach sand in the past (for thermites).

As a retort I'm thinking about using an old SS teapot (500 ml) with bent copper tube soldered to the pouring spout (then the end of it submersed in iced water or leading to a cold trap). Drill a hole in the side for Ar/CO2 entry and Bob's basically your uncle.

If the amount of mixture is sufficiently large compared to the amount of air in the retort an inert atmosphere might not even be necessary (Brand - 1669! and Scheele probably didn't use one either) because the P will become its own oxygen getter: the amount of oxygen in say 250 ml of air is only about 0.002 mole of O2, equivalent to roughly about 0.06 gram of P. Of course there's the nitrogen: would it react with hot P?

I'm sure this has been discussed in this thread before but as an alternative (and very cheap) source of phosphate consider also bone ash: mainly Ca3(PO4)2. React it with the equivalent amount of sulphuric acid and partially neutralise with soda or caustic soda to obtain NaH2PO4. Convert to NaPO3.




[Edited on 10-12-2009 by blogfast25]

Nicodem - 10-12-2009 at 08:57

Maybe something too obvious, but shouldn't ammonium phosphate react with aluminium powder to give phosphorous at proper conditions? This would make the use of silica unnecessary and the reaction should proceed at lower temperatures than with any of the sodium (poly)phosphates. The gasses forming in the first stage of calcination (A; where aluminium polyphosphate forms) should purge the air from the reaction vessel protecting the mixture from oxygen. Since the reaction enthalpy of the reduction stage (B) is considerably higher than with the analogous reaction of sodium phosphates, no additional "acid" like SiO2 or Al2O3 should be necessary to smooth the reaction.


A: 3 (NH4)H2PO4 + Al => Al(PO3)3 + 3/2 H2 + 3 NH3 + 3 H2O (at < 300°C ?)

B: Al(PO3)3 + 5 Al => 3 P + 3 Al2O3 (at >900°C ?)


Or was this already tried and is not working?

blogfast25 - 10-12-2009 at 09:17

That boils down to making the metaphosphate in situ, then subsequently reacting it with the excess Al. Provided the Al wouldn't react with the orthophosphate (at modest temperatures that would be a reasonable assumption, crak up the heat to get the actual reduction started) then that's a real possibility.

Not sure whether it's been tried yet. Strepta's set up would be suitable for testing it, without CO2 blanket...

It would yield a more useful byproduct too (alumina)...

But I can see a snake in the grass too: the acidic NH4H2PO4 is likely to react with Al to form Al tertiary orthophosphate, AlPO4 and H2.

The AlPO4 is not likey to convert to metaphosphate, apparently needed to obtain elemental P and not phosphide...

But that problem could be solved by using NaNH4HPO4 instead of NH4H2PO4...

[Edited on 10-12-2009 by blogfast25]

Nicodem - 11-12-2009 at 00:40

Quote: Originally posted by blogfast25  

But that problem could be solved by using NaNH4HPO4 instead of NH4H2PO4...

Or by simply adding silica or some other solid "acid" which brings us back to the industrial solution of the problem, the exact one that I thought would be avoided.

However, I would not be so sure that AlPO4 forms at all. In the (NH4)H2PO4 + Al melt you always have an excess of the acid so no AlPO4 can form at the beginning, only a melt of aluminium dihydrogen phosphate mixed with aluminium. The question is then if this melt dehydrates faster to the solid mixture of aluminium metaphosphate + aluminium, before the aluminium gets corroded further, forming a mixture of AlPO4 + Al instead. I do not know much about inorganic reactions and much less about reactions of metals in melts of salts, so I have no idea.

entropy51 - 11-12-2009 at 06:42

Quote:
As a retort I'm thinking about using an old SS teapot (500 ml) with bent copper tube soldered to the pouring spout (then the end of it submersed in iced water or leading to a cold trap). Drill a hole in the side for Ar/CO2 entry and Bob's basically your uncle.
Sounds like a plan, but I hope you might be thinking of some kind of pressure relief valve or bursting disk, just in case.


blogfast25 - 11-12-2009 at 08:04

Quote: Originally posted by Nicodem  
However, I would not be so sure that AlPO4 forms at all. In the (NH4)H2PO4 + Al melt you always have an excess of the acid so no AlPO4 can form at the beginning, only a melt of aluminium dihydrogen phosphate mixed with aluminium. The question is then if this melt dehydrates faster to the solid mixture of aluminium metaphosphate + aluminium, before the aluminium gets corroded further, forming a mixture of AlPO4 + Al instead. I do not know much about inorganic reactions and much less about reactions of metals in melts of salts, so I have no idea.


Yes, I was a bit quick off the mark there. But in the melt:

Al + 3 NH4H2PO4 ---> Al(HPO4)3 + 3 H2 + 3 NH3

IS likely to proceed because NH4H2PO4 is quite acidic (a 1 M solution has a pH of about 3) and Al is very reactive of course. How the Al(HPO4)3 will further fare I've no idea. Much depends on how the reactions proceed in time (in what order) and perhaps the heating rate. It's worth a shot.

Quote: Originally posted by entropy51  
Sounds like a plan, but I hope you might be thinking of some kind of pressure relief valve or bursting disk, just in case.


Pressure inside the retort should be quite small because the phosphorus evolves really quite gently and because I'd be using 1/2" copper pipe and a short one at that. I'm more concerned with water being sucked up through the copper tube at the end of reaction, when the gasses start cooling and under-pressure will be created. Some simple one way valve as a precaution to prevent steam explosions may be necessary. Otherwise a water-free cold trap (or series of) may be safer still...

pip - 6-1-2010 at 20:46

What is holding back scale up? Isn't this method self controling or just at small scale? No one said that it can't scale up but no one mentioned doing so either. (I had in mind 10 gram batches as a scale up to work towards)

Method for Preparation of Small Amounts of P4.

The attached is a description of a technique used to prepare small amounts (< .5g at a time) of white (yellow) phosphorus from sodium hexametaphosphate (calgon). The chemistry for reduction of calgon using aluminum and silica was published by Franck more than a century ago and first mentioned near the beginning of this thread by Polverone 6 years ago.

6NaPO3 + 10AI + 3Si02 = 3Na2Si03 + 5Al203 + 3P2

The reaction, properly implemented, is self -sustaining and initiates at temperatures which can be obtained with a lab (Meker) burner and conducted in a test tube (18mmX150mm, pyrex), which can be re-used.

[Edited on 7-1-2010 by pip]

Strepta - 7-1-2010 at 04:48

pip: Of course it can be scaled up, albeit with somewhat different apparatus. I would suggest proceeding cautiously and incrementally; ie, do not mix 100 grams of reactants and confine them in a small space. Perhaps a length of pipe (or quartz tube if you wish to observe the reaction) with the reactants laid in a linear train through the length of the pipe to help avoid a thermal runaway.

blogfast25 - 7-1-2010 at 07:57

Quote: Originally posted by Strepta  
pip: Of course it can be scaled up, albeit with somewhat different apparatus. I would suggest proceeding cautiously and incrementally; ie, do not mix 100 grams of reactants and confine them in a small space. Perhaps a length of pipe (or quartz tube if you wish to observe the reaction) with the reactants laid in a linear train through the length of the pipe to help avoid a thermal runaway.


I don't think there is much risk of a thermal runaway: thermochemical estimates of mine show that in adiabatic conditions (all heat retained) the metaborate reduction reaction would reach about 800 C end-temperature and that is in line with observations by you, Garage Chemist and myself. 800 C is much lower than a typical 'successful' thermite reaction and most of those run very contained too. If achieved temperature had been much higher your test tube would of course have melted. It's exothermic all right but not terribly so. The 'slag' left behind was sintered, but hadn't melted either...

Before any scale up I want to more or less replicate your test tube experiment but with some modifications. But the weather here is against me... :(


[Edited on 7-1-2010 by blogfast25]

pip - 8-1-2010 at 11:15

Of course everyone here knows that scale ups can be dangerous and small increments are the safe thing to do. If not they do now. Seriously for those like me who have minor chemistry knowledge but a vast willingness to learn and experiment never just multiply the "ingredients" to make the amount of whatever you want, we're not baking cakes we are changing matter at one of its most basic levels. Best case senerio is it works, next best is it fails to produce the product you want, and most likely it will blow in your face. P will kill you very painfully. I've had 3rd degree burns form gasoline and faced with that again or a "P" mishap bring on the gas.

pip - 10-1-2010 at 16:49

what would the best choice of metal be for the reaction vessel? I don't want to accidentially want to use something that is going to react with the phosphorus since it is at a high temp and all.

Taoiseach - 11-1-2010 at 00:44

@pip

I think you forgot to attach the calgon paper or at least I didn't see a link...

blogfast25 - 11-1-2010 at 07:38

Quote: Originally posted by pip  
what would the best choice of metal be for the reaction vessel? I don't want to accidentially want to use something that is going to react with the phosphorus since it is at a high temp and all.


Steel or copper would be fine.

Myfanwy - 7-2-2010 at 07:38

could phosphine and hydrogen peroxide yield some P4?

2PH3 + 3H2O2 -> 6H2O + 1/2 P4


entropy51 - 7-2-2010 at 08:06

There's a book on phosphorus in the forum library. Why don't you read that and get back to us?

blogfast25 - 7-2-2010 at 08:51

Quote: Originally posted by Myfanwy  
could phosphine and hydrogen peroxide yield some P4?

2PH3 + 3H2O2 -> 6H2O + 1/2 P4



Possibly but why go to the trouble of making a highly toxic, highly inflammable gas when you can make P from Calgon tablets, some fine sand and some Al powder?

densest - 7-2-2010 at 09:35

@blogfast25 - depending on where one buys it, Calgon(tm) may contain no phosphorous :( - some other products may - one has to look at the fine print on the package. Algal blooms & such are a p.i.t.a....

Myfanwy - 7-2-2010 at 11:44

i tested it today with small amounts in a test tube. Even without sand P4 is produced.

you ever smelled the vapours, when white P is stored under water. This odor is really nice and exotic.
Dont know how to describe it better, but its not phosphine.

some strange Phosphorus oxides/acids..

12AX7 - 7-2-2010 at 11:49

Mmmmm! I can feel my jaw softening already! :D

blogfast25 - 7-2-2010 at 12:55

Quote: Originally posted by densest  
@blogfast25 - depending on where one buys it, Calgon(tm) may contain no phosphorous :( - some other products may - one has to look at the fine print on the package. Algal blooms & such are a p.i.t.a....


If you scroll up, you'll see I made my own metaphospate, as have others here. It can be made from just about any source of orthophosphate...

[Edited on 7-2-2010 by blogfast25]

JohnWW - 10-2-2010 at 02:57

In the olden days when matches were made from white P4, the sort that could be struck on any rough surface, "phossy jaw" was common among the factory workers who were exposed to its vapor. Its symptoms were loss of bone density of the jaws and other soft bones in the head, causing loss of teeth eventually. Elemental arsenic is at least as bad in this regard, also being a cumulative poison.

Attempted Reduction of AlPO4 with Carbon

Strepta - 28-2-2010 at 13:55

Introduction

Intrigued by an article posted by SC Wack (Preparation of Elemental Phosphorus), concerning reduction of AlPO4 with carbon @ 1100 C yielding 72% of available P (IEC vol. 21, no. 11, page 1130) I did an experiment to verify the claim and explore an approach.

Methodology and Fixturing

I used a temperature controlled quartz tube furnace to heat the reactants, fed by a stream of inert or nearly inert gas to sweep the reaction products toward the exit portion of the tube. This is an improvement of a furnace design I’ve been using for about 5 years which utilizes a ni-chrome element from a cheap toaster oven re-formed over a ¾” (1.9 cm) threaded steel rod and heated to red-heat to anneal. After cooling, the element is “screwed off” the mandrel and fitted directly onto a 56 cm section of 18mm OD quartz tubing. The heating element portion of the tube (23 cm in length and near the input end) is then wrapped with 5 turns of ½” (1.25 cm) kaowool blanket and fitted into a section of 10 cm dia. galvanized pipe. The quartz tube is cut from a longer section with a dremel diamond dust cutoff wheel and its ends are fire-polished with a MAPP gas-oxygen torch to anneal the quartz and reduce the chance of cracks. The torch is barely able to fuse the quartz ends. Rubber stoppers are used in each end—2-hole at the entrance for temp probe and gas inlet tube and single hole at the other for gas exit. A 90 deg tube from this stopper terminates just under the surface of H2O in a 100 ml beaker. The reaction tube temperatures at each end do not go much beyond 100C when operating at full temperature. “00” stoppers fixed to the ends of dowel rods will fit snugly into the quartz tube and are used to position and contain the reactant mixture in the area of the heating coil during assembly. After assembly and final positioning of the furnace, the dowels and stoppers are removed and the tube ends fitted with the aforementioned 1 and 2 hole stoppers.

The temperature controller is an old design, originally used for an oil bath but now re-biased for control of much higher temperatures. The temperature probe is a k-type thermocouple (Omega TJ36-CAXL-18G-12) the leads of which terminate at the input to an AD 595. The 595 output is differenced with a reference voltage and that result input to a SG3524 variable proportional controller which is synchronized to the power line. The output of the 3524 triggers a triac via an opto-coupler. The triac drives the heating element.

The sweep gas used was “balloon grade” helium stated to be 94-96% He. The tank is a low pressure disposable type pressurized to 18 atmospheres (260psig) when new. I’m sure these tanks are neither back flushed nor evacuated prior to filling, so I’ll assume they contain 1 atmosphere of air and 18 of He. This would make the residual O2 content ~ 1%. Because I could not effectively control the flow from the tank using only a needle valve, I adapted a single stage acetylene regulator to the tank fitting. This involved removing the CGA (Compressed Gas Association) 510 nut and nipple from the regulator body and replacing with a ¼” flare fitting. All connections to the regulator body are ¼” NPT so commonly available plumbing hardware was used. The photo shows the tank pressure to be about 40 psig and the output gage pointer is just off the stop at about 2 psig. With the output needle valve, this arrangement maintains a uniform flow throughout the experiment- set it and forget it. Flow rate used was about 1.5cc/sec.

During a test run, I placed a layer of aluminum oxide (assumed to be inert) throughout the heated length of the tube so that it filled the tube about halfway (half of tube diameter) and then passed a current of air (aquarium air pump) through the tube. I stepped the heating up incrementally to check performance. It appeared to work fine up to and including 1250 deg. C was as far as I needed or wanted to go as I was concerned with burning out the element. Also, although the probe is calibrated to 1335 C, it is not recommended to use it over about 1150 C continuously. The flow rate of air through the tube was varied to check out the effect on the temperature controller duty cycle. There was not a perceptible change for the modest flow rates used.

Preparation of AlPO4

I prepared AlPO4 (intended yield 5 g) from 200cc solutions of Al2(SO4)3 (15.5 g) and Na3PO4.12H2O (14g):

Al2(SO4)3 + Na3PO4.12H2O ==> Na2SO4 + AlPO4


When these solutions are mixed, the insoluble AlPO4 immediately precipitates. As I could not be sure of the exact water of hydration in the Al2(SO4)3, I heated it first to drive off most of the H2O, and for purposes of calculating a weight to use, assumed an anhydrous product. I added an additional 200 ml of H2O to reduce the concentration of the remaining Na2SO4 solution to .1M. The resultant solution pH was highly basic so I incrementally added Al2(SO4)3 until the pH was lowered to 6.1, the published pH of a .1 M sol. of Na2SO4. The excess liquid was poured off and the remaining liquid and precipitate were filtered through a Buchner funnel. Yield after drying was 4.8 g out of 5g theoretical.

Reduction

The reaction to be tried was: 2AlPO4 + C ==> 2Al2O3 + CO2 + 2P. The authors used equal weights of AlPO4 and C although this results in a large excess of C by stoichiometry, possibly to ensure that all of the AlPO4 is reduced.

I added 4.5 g of dried AlPO4 to 4.5 g of carbon black and mixed this in a coffee grinder for two minutes. I was only able to get 4.7 g (of the 9g total) of this mix into the reaction tube as I wanted it no more than half full. More could have been added by reducing the headspace above the mix and/or by tightly packing the mix.

4.7 g of reactants would represent .6g available P at 100% yield. With a yield of 70% this amount would be reduced to about .4g.

Experimental

When finally assembled and ready, I started the gas flow and let it run about 10 minutes before ramping the temperature to 650 C. After allowing the initial transient to settle, I continued to ramp the temperature in 100 deg increments every 5 minutes. As the temp transitioned from 1050 to 1150 C, the exit portion of the tube darkened and the exhaust gas bubbles burst into flame as they surfaced in the beaker. I continued the temp ramp to 1250 C. After a few minutes at this temp, the probe readings (monitored at the 595 output) became erratic, dropping to as low as 600 and to as high as 1400. I suspected a poor connection in the circuitry or a failure of the 595. After a few minutes, I shut the power off to the controller but left the gas flowing and allowed the tube to cool. As it cooled the readings became steady again and at 250 C I removed the insulation and quartz tube from the galvanized pipe section and unwrapped the kaowool. As I unwrapped the tube it broke into two pieces. (Fig 7) The area under the heating element was extensively cracked, and through handling, another section broke off. I removed the tc probe and found that the Ni-Chrome sheathing had melted and the melt had largely gathered into three globules. One of these had a vitreous solid adhering to it –this appeared to be a piece of melted quartz.

Some phosphorus was evident in the exit section of the tube (Fig 9) but this was not recovered. About 2.7 g of the original 4.7 g of reactants was recovered (Fig 8) and this had not fused but was still a loose powder as described in the article.

Conclusions

It appears that AlPO4 is reduced by carbon in the vicinity of 1100-1150 C as described, although I was not abIe to confirm the % released (claimed as 72-83% after 1 hr) as the apparatus itself was also reduced to junk. It also appears that a carbon reaction, presumably with O2, is responsible for the extreme temperature excursion experienced. I had not thought of it at the time, but the He tank I used had been sitting in the garage for over a year since it’s purchase. In that time the pressure had dropped from 255 psig to 50 psig. If that represented He preferentially escaping vs air, the O2 content could have re-enriched by a factor of 5 to ~ 5%. That could have been enough for a low-level charcoal fire to have taken place in the tube.

I will probably try this again after I have replaced the quartz tube, perhaps with CO2 as the sweep gas. CO2 will nearly certainly be reduced to CO at temp in the excess carbon environment and suitable precautions must be taken to ensure safety. Also, this gives me an idea for a very hot tube furnace, with temperature modulated by air flow. Materials (alumina tubes, type R thermocouple) would be expensive.



Fig. 1 - Reaction loaded and captured under heating element



Fig. 2 - Assembled, ready to begin



Fig. 3 - Temperature controller



Fig. 4 - He tank and gas regulator



Fig. 5 - Inlet end of furnace at temperature



Fig. 6 Exit end of furnace and gas beaker at temperature



Fig. 7 - Broken tube, temp probe and residual reactants



Fig. 8 - Carbon glow



Fig. 9 - Some P condensed in exit section of tube



[Edited on 28-2-2010 by Strepta]

[Edited on 28-2-2010 by Strepta]

entropy51 - 28-2-2010 at 14:55

Strepta, I think this is very nice work for a first attempt, especially if that is really P condensed in the exit section.

Might it perhaps be better to embed or otherwise surround the heating element with some type of refractory that would give a more uniform temperature distribution in the quartz and avoid hot spots from direct contact with the hot wire? I guess refractories would have low thermal conductivities and might not work as I am thinking.

Better luck on your next try, which I hope is soon!

Magpie - 28-2-2010 at 15:09

Very interesting experiment. I'm intrigued by the gas bursting in flame after passing through water in the beaker. Surely this is CO, right? I never saw this with my ZnO reduction with carbon.

I suggest you look into buying an argon cylinder: not too expensive, refillable, and readily available at your compressed gas dealer.

I agree with entropy that it would be better to wrap your heating element around a ceramic tube such as one made of Al2O3 or mullite. I presume you have seen the designs of garage chemist and myself. I have taken my Al2O3 tube up to 1300C many times with no apparent problems.

hissingnoise - 28-2-2010 at 15:40

The spontaneously flammable gas may have been phosphine but that would beg the question - where did it get the hydrogen?
The breaking element could have produced a hot arc which would melt quartz and possibly account for the fluctuating temperature.
I'm guessing though!
But you sure came close - next time (or the one after that?) could be entirely successful if you decide to go through it one more time.


12AX7 - 1-3-2010 at 00:43

There could be moisture in the tank, but not very much. The charcoal could adsorb a lot, though. What source of charcoal, was it calcined before use?

Tim

IrC - 1-3-2010 at 01:43

Strepta

"The sweep gas used was “balloon grade” helium stated to be 94-96% He. The tank is a low pressure disposable type pressurized to 18 atmospheres (260psig) when new. I’m sure these tanks are neither back flushed nor evacuated prior to filling, so I’ll assume they contain 1 atmosphere of air and 18 of He. This would make the residual O2 content ~ 1%"

I think you have much more O2 than you realize. To keep children enticed by the high pitched voice from suffocating if they inhale the gas every balloon grade He I have seen contains much oxygen. I ran into this trying to build a lifting device where I was experimenting with carrying a wire high into the air in an experiment measuring the voltage on the wire. To save money I tried balloon gas from more than one vendor eventually having to buy a tank of pure He. The amount of oxygen was so high it was negating the lift I needed without resorting to filling an unrealistic number of surplus weather balloons joined together. Meaning of course the entire 5 or 6 percent is oxygen not the 1 percent you think it is.


[Edited on 3-1-2010 by IrC]

watson.fawkes - 1-3-2010 at 04:53

Quote: Originally posted by Magpie  
I suggest you look into buying an argon cylinder: not too expensive, refillable, and readily available at your compressed gas dealer.

I agree with entropy that it would be better to wrap your heating element around a ceramic tube such as one made of Al2O3 or mullite. I presume you have seen the designs of garage chemist and myself. I have taken my Al2O3 tube up to 1300C many times with no apparent problems.
I'll second both of these suggestions. Argon is a good choice of carrier gas. It's cheaper that helium and doesn't leak out of every nanoscopic pore.

The reason to use a ceramic tube is to mechanically support the heating element. The strength of heating wire goes down at high temperature, eventually enough that it can't support itself. My guess is that you had a hot spot on your reaction tube that made a weak spot in the wire. That in itself wouldn't have been a problem, unless the wire were subject to mechanical forces. As a wire loosely wrapped around a tube, it was free to move. The hot spot would have caused differential thermal expansion at that point, which is enough to generate such forces. By cementing heating wire to a fixed ceramic tube, it constrains the motion of the wire, greatly reducing the need to rely on the mechanical strength of the wire itself.

Picric-A - 1-3-2010 at 05:49

Balloon grade helium is 94% He and the rest being O2 to stop suffocation of poeple who breathe it soley to get a high voice :O
Maybe use N2 carrier gas? Cheap and more effective!

Picric-A - 1-3-2010 at 05:50

Edit: deleted double post.

[Edited on 1-3-2010 by Picric-A]

JohnWW - 1-3-2010 at 06:13

Frankly, He is too rare (on Earth) and valuable to be used just for party balloons and unmanned advertizing blimps (for which H2 could be used), or for inert-gas applications for which the much more common Ar can be used. Because the only economic source of it on Earth is from certain deposits (as in Texas) of natural gas contained in sedimentary rocks derived from granite, which contains U-238 and U-235 and K-40, the alpha-decay of which (and of daughter isotopes) gives rise to He nuclei, its availability (by separation from these natural gas deposits which are likely to be exhausted by 2100 at the present rate of consumption) is quite finite. While there is a small amount of it in the atmosphere, it is too small for economic extraction by fractionation of liquid air; and once in the atmosphere it is slowly lost by diffusion (like gaseous H2 unless oxidized) into space due to its low atomic weight.

Strepta - 1-3-2010 at 17:57

I should clarify my objectives a bit. In the above experiment with AlPO4 I wanted to verify that P was reduced with carbon at or near 1100 C in the yields described (72%) and stimulate some discussion of this as an approach. My longer term interest is in exploring paths for the preparation of elemental P at the amateur level. This assumes no extreme temperature capability such as carbon resistance tubes, MoSi2 heating elements, arc furnaces, etc, but limited to the temperatures attainable in resistance wire tube furnaces such as those nicely documented by garage_chemist and Magpie or other readily available means. The chemistries exlored to date include:

1) Reduction of Pb3(PO4)2 with H2 @ 700 C

2) Reduction of NaPO3 with pyro Al

3) Reduction of AlPO4 with C @ 1100 C

I've also conducted some experiments atempting to reduce H3PO4 by adsorbing it onto activated charcoal and heating in a microwave but had no success with that approach.

@ Magpie: The flames may well be hot CO oxidizing to CO2 as it emerges from the water. I do not recall seeing any "smoke".

As far as the constuction of the apparatus goes, I've used this before to 1200 C without incident. I don't believe it was an issue with the resistance wire on the quartz (which did not burn out by the way) so much as a "charcoal fire" in the tube. The inconel sheathing of the thermocouple probe melted off implying temps on the order of 1400 C. The vitreous deposit on one nodule of sheathing would likely have been from either the quartz tube or the AlPO4 --either would imply temps > 1500 C.

@12AX7: I preheated the C to 450C but it was not truly "calcined" and did contain some moisture as was evidenced in the exit portion of the tube when first heated to 650 C. I allowed this to dry out with the sweep gas before further ramping the temperature.

[Edited on 2-3-2010 by Strepta]

densest - 1-3-2010 at 18:38

@strepa - Your careful description of your experiment is a model of clarity - congratulations!

If you get Ar from a welding supply house (usually by far the least expensive place to get it!) be sure to ask for pure Ar since welding Ar usually has 1% O2 in it because it improves the weld quality.

For such low pressures and low rates of flow from a high pressure (140 bar/2000 PSI or so), a "two stage" regulator would perhaps be beneficial. They are quite expensive new but can be had for $40 or so on EBay or other surplus sources.

Nichrome heating element wire is surprisingly(?) inexpensive from companies who deal with it in quantity. Arklay S. Richards www.asrichards.com sells standard thermocouples and bulk resistance wire at reasonable prices. I had them fabricate 25 simple type K thermocouples for me at about $1 each - 100 wouldn't have cost much more. They sell bulk nichrome and if I remember their web site, they will wind it into heating element form for a not excessive cost. I don't know if they sell Kanthal or other super high temperature elements.

Mullite/Al2O3 tubes show up as surplus or overstock occasionally at industrial or scientific suppliers - I got two 1" x 36" x 3/16" wall mullite tubes for $12 each a while back.

I -think- the AD595 has a "broken thermocouple" indication on some output. I've always bought complete temperature controllers. The Omron 1/16 DIN size (universal temp sensor input, relay outputs for 1A or so) used to go for $40 or $50 on EBay. They come calibrated and have a lot of features like proportional-integral-differential control which make operating the furnace more consistent.

Perhaps some of the furnace details should be archived for reference?


Magpie - 1-3-2010 at 20:10

@strepa: In looking at your pictures again there appears to be a considerable amount of yellow flame within the quartz tube. GC says that P burns with a characteristic yellow flame. Perhaps that is P burning. But I'm not sure where the O2 would be coming from other than the He tank or a leak of some kind. If it was just a small amount of O2 coming in with the He you would think that your large excess of carbon would take that up as CO.



[Edited on 2-3-2010 by Magpie]

Sedit - 2-3-2010 at 13:28

Quote: Originally posted by Magpie  
@strepa: But I'm not sure where the O2 would be coming from other than the He tank or a leak of some kind.
[Edited on 2-3-2010 by Magpie]


Commercial He used for filling balloons is not pure He but is cut with air which is the source if thats where he got his He from.

Strepta - 3-3-2010 at 17:31

Sorry about that last post (removed). I'll edit the photos down in size and re-post later.

blogfast25 - 4-3-2010 at 13:46

Not quite sure why you didn't go with the NaPO3/Al/SiO2 reaction. With this very interesting set up (but using Argon welding gas instead of He), you should have had some nice phosphorus by now.

At those temperatures, consider also the original method: bonemeal (calcium orthophosphate), sand and carbon...

Strepta - 4-3-2010 at 17:22

I took some photos of the quartz tube from my AlPO4 experiment near the point where the tube broke. There is a white, porcelain-like circumferential deposit inside the tube which is fuzed to the quartz wall. Everywhere this deposit is present, the quartz is shattered. It's likely that the tube failure was due to the difference in coefficients of thermal expansion between the coating and the quartz as the tube cooled. This white coating is probably either Al2O3 or AlPO4. I measured the thickness of a sliver of the coating and found it to be .05 mm; the tube wall is 1.5 mm. The last photo is of the temp probe beside the same model of a new probe. Things got pretty toasty in there. I will probably repeat the experiment in the near future using CO2 as the sweep gas.












blogfast25 - 5-3-2010 at 07:50

@ Strepta:

The coating must be alumina, possibly with some unreacted phosphate. I've been fooling around with trying to coat things with alumina as a by-product of reactions, you've inadvertently succeeded!

CO2 is likely to be an oxidiser for Al in these 'toasty' conditions but whether the Al will favour being oxidised by the CO2 or by the metaphosphate depends on the chemical equilibria reigning.

Cheap argon welding gas is still your best bet as an inert atmosphere but have you considered not using an inert atmosphere at all? The amount of oxygen contained in your assembly is really quite small: at STD conditions 22.4 L of O2 is still only about 32 g (1 mole). Without inert gas, some P would be lost to oxidation by O2 but once the small amount of O2 has been scavenged, your P is good to go!

Which type of thermocouple did you use and where did you get them?

Strepta - 5-3-2010 at 10:51

@blogfast:

I've used the CO2 before even with elemental Al in the tube. While some may have been consumed, the amount did not appear to be significant.

The gas is used solely for sweeping the product away from the reactants and making recovery/harvesting of any P4 easier.

I favor CO2 because of ease of fixturing, availability and cost. A reducing gas such as H2 could also be used, and I did use it (produced from an electrolysis cell)in a reaction at 700 C described somewhat upthread. Some are concerned with using flammable gas at elevated temps but the volume in these experiments is so small that even a tube failure should not produce more than a loud pop.

The probe is an Omega TJ36-CAXL-18G-12.

jgourlay - 5-3-2010 at 11:12

Wow...this thread is a LOT to absorb. From what I've read so far, it appears this thread has a global contributorship. For this reason, I'd appreciate if you all would indulge me in one clarifying question.

IN THE UNITED STATES, what are the readily available sources of raw material for one of the proven methods detailed in the thread? For example, I read "calgon tablets" but then also "have none depending on where you are".

Magpie - 5-3-2010 at 17:50

Today I made a small amount of P according to the following reaction:

6NaPO3 + 10Al + 3SiO2 --> 3Na2SiO3 + 5Al2O3 + 6P

Stoichiometric ratios of the reactants were mixed in a mortar. The Al was 100-200 mesh, the SiO2 200 mesh pottery grade, and the NaPO3 was technical grade. My basis was 5g of P.

I have been wanting to try this for some time but have not been able to find a suitable luting compound to join a ceramic retort to a glass adaptor. I finally settled upon the best candidate that would give me a truly positive seal, yet was releaseable following the experiment. This is Permatex high temperature RTV silicone. Previous testing with a ceramic tile/glass slide showed that an RTV seal can be destroyed in about 2 hours with con sulfuric acid at 125C, thereby allowing recovery of the glass piece.

After loading the retort the 24/40 glass adapter was attached using the RTV and allowed to cure overnight. Today the retort was backfilled with argon, placed in a tube furnace, and a 400 mL beaker of water located to provide a water seal for the adaptor outlet.

Over a period of 3 hours the temperature of the furnace was brought up to a maximum of 1300C. Most of the time there was just a periodic large bubble evolved indicating expansion of the gas in the retort. However, at times the bubbling would stop, be erratic, or even form a vacuum of about 1/2" water. During the last 100C or so it seemed like no gas was formed, or any vacuum either.

P never did drop into the receiver as I had intended. When I removed the insulation from the adaptor I found a small pool of solidified P, tainted red from the RTV. Using a bunsen burner I melted the P, picked up the furnace and drained the 4 or 5 drops of waxy, heavy P into the receiver.

In an attempt to scavenge the remaining P I removed the hot retort from the furnace and tried to drain any P into the receiver. Even with heating from a bunsen burner no more was obtained. My overzealous use of the bunsen burner resulted in cracking of the glass adaptor.

One of the drops of P was placed in a small crucible and lit with a match. It burned vigorously. I was slow to get my camera but you can see the last of it burning in one of the pictures.

I'm still curious about the rest of the P. So when I figure out a good place to do it I will open up the retort with a pair of channel locks and report back.

luting curing.jpg - 60kB ready to heat.jpg - 66kB furnace full view.jpg - 58kB pooled P.jpg - 64kB drops of P.jpg - 102kB burning P.jpg - 61kB scavenging attempt.jpg - 96kB

[Edited on 6-3-2010 by Magpie]

Jor - 5-3-2010 at 18:07

Nice work! What are you going to use the P for? PCl3, PCl5 or anything else?

How expensive was the broken piece of glassware?

Magpie - 5-3-2010 at 19:24

Quote: Originally posted by Jor  
Nice work! What are you going to use the P for? PCl3, PCl5 or anything else?

How expensive was the broken piece of glassware?


I would like to make some PCl3 and PCl5. But I'll need to collect a larger quantity first. And to do that I've got some bugs to work out.

The glass adaptor was $21 including shipping. I won't make that stupid mistake again.

blogfast25 - 6-3-2010 at 06:52

Interesting run. I'm surprised you didn't get more phosphorus: my own, much simpler experiment

http://www.sciencemadness.org/talk/viewthread.php?tid=65&...

seemed to indicate that once you get ignition the reaction becomes self-sustaining because of the reaction heat.

In your experiment, most of the phosphorus must have condensed in a cooler part...

But why use a ceramic retort, which by now will be rendered useless by the slag: a mixture of alumina and sodium silicate?

I'd suggest a steel tube of similar dimensions and capped off with a threaded cap or welded shut. Heat with Bunsen or propane flame torch. A higher charge would also help: 5 g of phosphorus isn't much...

Magpie - 6-3-2010 at 08:24

Quote: Originally posted by blogfast25  
Interesting run. I'm surprised you didn't get more phosphorus: my own, much simpler experiment

http://www.sciencemadness.org/talk/viewthread.php?tid=65&...

seemed to indicate that once you get ignition the reaction becomes self-sustaining because of the reaction heat.

In your experiment, most of the phosphorus must have condensed in a cooler part...

But why use a ceramic retort, which by now will be rendered useless by the slag: a mixture of alumina and sodium silicate?

I'd suggest a steel tube of similar dimensions and capped off with a threaded cap or welded shut. Heat with Bunsen or propane flame torch. A higher charge would also help: 5 g of phosphorus isn't much...


>I'm in the process of learning just how high I have to take the heat to get a decent yield - likely not the full 1300C capability of my furnace.

> The cooler part was essentially the glass adaptor. When it broke off some phosphorus did ignite near the top of the retort. When I open it up today I'll know if there is anymore inside. I doubt if there is as I think I would have seen it burning.

> Yes, the ceramic tube is now worthless - it is intended to be sacrificial. They cost very little in materials to make but do take some time. Have you seen how much heat is required to use a steel tube? If not, see the work of Gruson, this forum.

> Yes, 5g is small. This was my first attempt.

blogfast25 - 6-3-2010 at 08:51

@ Magpie:

In my case, the reaction started well before my 25 ml SS 'crucible' (a measure for liquor shots, actually) started to actually glow. By the end of the reaction (10 g total charge) the whole thing, crucible included, was glowing a bright red-orange...

How about a copper (cooking gas) tube? 1 inch diameter, fairly thin walled (about 1 mm I'd say), quite cheap, conducts heat like mad...

Strepta - 6-3-2010 at 09:15

@Magpie-- Nice work as usual! Was your ceramic retort made of alumina? You may have success trying to clean it with HCl. That worked well for me when cleaning pyrex or quartz after a calgon reduction with Al. Fill it about half full with HCl and close the ends with neoprene stoppers and let stand for an hour or so. Then remove the slag with a suitable wooden dowel rod. The glass always recovered nicely, but of course I wasn't taking the reaction temp to 1300 C!

On a different note, at some point in the future you may want to try boron oxide in place of SiO2. I got my best yield of P (52%) using B2O3, although it's not a huge improvement over SiO2. Good luck and keep those pictures coming!

Magpie - 6-3-2010 at 11:55

Quote: Originally posted by Strepta  

@Magpie-- Was your ceramic retort made of alumina?


The retort was made of about 2/3's mullite and 1/3 feldspathic to vitrify to 0% porosity. Also, thanks for your suggestions.

I just finished cutting open the retort. I decided to use a Dremel diamond cutoff wheel so that the content integrity could be better maintained. I made two cuts.

The exit end of the retort was virtually plugged with a somewhat foamy grey slag. The slag front was coated with a red powder. I don't know what this could be other than red P. Is that possible?

Because this slag is so voluminous this may not be the best approach for making P. I will be trying other reactants eventually.



1st Dremel cutoff.jpg - 59kBslag front with red coating.jpg - 69kB2nd Dremel cut.jpg - 58kB

blogfast25 - 6-3-2010 at 13:01

@ Magpie:

Surely that must be impure Red P. Surely you will test it?

The slag resembles what I obtained. I doubt if other reactants would make much difference: your gaseous P has to evade from a half molten/half sintered mass of silicate/alumina. No wonder it 'puffs up'! I saw that happening in my crucible...

A higher temperature at the end and slightly longer charge and the P would possibly have come gushing out.

Bear also in mind that due to mechanical entrainment, your P will still need purifying: sublimation, I'm guessing...



[Edited on 6-3-2010 by blogfast25]

DJF90 - 6-3-2010 at 14:00

Blogfast25: From your reasoning it would be sensible not to pack the tube, but to have an air space above the reaction mass when the tube is in the horizontal position. Using a packed tube seems kinda illogical in the first place but perhaps that's just my personal view.

blogfast25 - 6-3-2010 at 14:16

Quote: Originally posted by DJF90  
Blogfast25: From your reasoning it would be sensible not to pack the tube, but to have an air space above the reaction mass when the tube is in the horizontal position. Using a packed tube seems kinda illogical in the first place but perhaps that's just my personal view.


@ DJF90:

Yes, something like that: where you have reaction the P cannot condense (too hot) but it does condense easily where it's a little cooler, as both Strepta's test tube as well as Magpies's run clearly show. A packed tube is probably not advisable: that will block the path of the evolving phosphorus gas...

Well observed...

Magpie - 6-3-2010 at 16:35

I might add that a full retort (up to the taper) would require 77mL of reactants. For 5g of P the charge volume was 25mL. Before placing the retort in the furnace I tapped the tube to level the charge with a slight bias toward the closed end. Of course the end result is just a guess.

len1 - 7-3-2010 at 04:24

Nice work! But I must agree with blfst25 that steel is a more convenient material in this reaction. It gives you more even heat distribution, and you can control heat of your receptacle tube with water cooling, that way you wont get clogging. Excessive heat consumption only occurs if you expose much un-insulated metal surface outside the furnace. Back when I was still doing home experiments and my practical guides I was going to this, but phosphorus unfortunately has a bad press, so I never did.

Panache - 7-3-2010 at 23:57

Consider pressing small tablets out of your starting reagents using a kbr die used to press IR plates, (however make the tablet thicker than you would a KBr plate), this would seem like a suitable way to overcome your volume issues, as the tabletted form can be more easily dispersed.
As for retort materials, have you considered tupperware, if you're interested perhaps you would like to host a party. There are great benefits to hosting!!

Magpie - 8-3-2010 at 08:34

Quote: Originally posted by Panache  
Consider pressing small tablets out of your starting reagents using a kbr die used to press IR plates, (however make the tablet thicker than you would a KBr plate), this would seem like a suitable way to overcome your volume issues, as the tabletted form can be more easily dispersed.
As for retort materials, have you considered tupperware, if you're interested perhaps you would like to host a party. There are great benefits to hosting!!


I'm not sure what the tableting, or briqueting, would accomplish other than facilitating the passage of gaseous P. I know this has been done on a commercial scale, for that reason I believe.

I have considered Tupperware. But there is a small seal on this product and I have made a pledge that no animals will be harmed in my experiments.

entropy51 - 9-3-2010 at 11:09

An experiment along these lines is described on page 74 of the July 1938 Popular Science.

Magpie - 9-3-2010 at 12:54

Very interesting. Thanks for posting this!

It seems like all home chemists wore ties in the old days. I suppose it gives the aura of competence. I also like the references to being able to easily obtain the needed chemicals, like hypophosphite, at most any old chemical distributor.

These home chemistry articles always seem so casual about it all. At least here they recognize phosphorus as "treacherous."

len1 - 9-3-2010 at 15:22

A beautiful demonstration of how people have lost their marbles since those days.

Now if you submitted a way to make P to an amateur magazine they'll

1) Start screaming OHS, wheras in fact people care about each other not one iota more than in those days

2) Say you are encouraging terrorism and drug making

3) People are not so "naive" to bother to do work just for fascination in science.

[Edited on 9-3-2010 by len1]

Taoiseach - 10-3-2010 at 04:29

Quote: Originally posted by Magpie  
It seems like all home chemists wore ties in the old days. I suppose it gives the aura of competence


His index finger looks as if he used to fiddle with even more dangerous compounds than P tough :o

[Edited on 10-3-2010 by Taoiseach]

entropy51 - 10-3-2010 at 05:59

Quote: Originally posted by Magpie  

It seems like all home chemists wore ties in the old days. I suppose it gives the aura of competence.
That was just the city boys. If you find the March 1939 Popular Science on Google Books, on page 206 the chemist is a farm boy making acetamide. He is dressed accordingly in what we used to call "bib overalls", but I suppose some call coveralls.

If you go to the link I posted and enter "Wailes" in the search box and check "search all issues" you will find all the other home chemistry articles by Raymond Wailes. Some are quite interesting. Unfortunately there is no PDF download.

Magpie - 10-3-2010 at 09:25

Quote: Originally posted by entropy51  

...a farm boy making acetamide.


There's something I really like about that. A love of science even in the sticks, in an age of innocence.

AlPO4 substrate

Magpie - 20-3-2010 at 12:01

Here's the results of another try at making the elusive P, this time using AlPO4 as substrate.

As proposed by Strepta et al:

2AlPO4 + 5C --> Al2O3 + 2P + 5CO

The AlPO4 was made by reacting Na3PO4 (T.S.P.--the pure stuff in the milk cartons as made in China and repackaged in Canada) and Bonide Al2(SO4)3 from the garden store. It was calcined at 1000C in a muffle furnace for 1 hour to get rid of hydrate water. It was then ground in a mortar followed by a coffee grinder to get something like 200 mesh. I noted that the mix seemed hygroscopic as it became somewhat non-flowable just sitting in an unsealed beaker.

The carbon was coconut shell activated charcoal ground in a coffee grinder to ~ 200 mesh.

The reactants were mixed in equal proportions by weight and again ground together in the coffee grinder. Sufficient powder (36g) for a theoretical 5g of P was then placed in a ceramic retort. A Pyrex adaptor was affixed to the retort with Permatex high temperature RTV silicone. After curing the retort was placed in a tube furnace as shown below.

The temperature was slowly brought up to a maximum of 1258C over a 3hr period. Condensate (water) began forming in the glass adaptor right away. Even though I dispeled some of it with a heat gun there was always some remaining.

At about 950C I heard the glass adaptor crack. I think I had shoved it on too hard leaving no elastomer gap to take up the expansion of the ceramic. The seal appeared to have maintained integrity, however, so I just kept on heating.

Around 1100C the bubbling became more rapid and I could see a white film deposit, that did not look like P, forming in the adaptor. Around 1160C puffs of white smoke began coming out of the water receiver and then I heard regular cracking pops, then saw bursts of light. Phophine was burning. This was very impressive. I turned out the lights and began taking pictures hoping to capture one of these burst of light. I was successful as can be seen below. At no time did I see any burning within the glass adaptor.

No drops of P formed. Just a lot of phosphine. When I cracked open the retort there was just a lot of black powder, looking like the original charge. A minor amount of P began to smoke as it was exposed to the air.

The last picture shows my setpup for a gentle purge flow of argon to prevent suckback as the retort was cooling.

It should be noted that this experiment was conducted in my efficient hood with the fan on. Phosphine is deadly.



AlPO4 setup.jpg - 75kBcondesation & white powder.jpg - 58kBPH3 burning.jpg - 60kBmore deposits.jpg - 64kB

blogfast25 - 20-3-2010 at 12:45

Are you sure you saw phosphine (PH3 + P2H4) burning and not gaseous P? There's no source of hydrogen in your mix!

Also, you're not allowing for much cooling: gaseous P could be escaping from the adapter (through that 1 cm or so of water).

I would try this again but with silica present, as is done industrially with bone meal:

2 AlPO4 + 3 SiO2 --->Al2(SiO3)3 + P2O5
P2O5 + 5/2 C ---> 2P + 5/2 CO2

Making AlPO4 react with carbon without the 'silicating step' may prove very hard... Are you getting significant yield with AlPO4 alone, is the question...

Magpie - 20-3-2010 at 13:40

Quote: Originally posted by blogfast25  
Are you sure you saw phosphine (PH3 + P2H4) burning and not gaseous P? There's no source of hydrogen in your mix!

Also, you're not allowing for much cooling: gaseous P could be escaping from the adapter (through that 1 cm or so of water).

I would try this again but with silica present, as is done industrially with bone meal:

2 AlPO4 + 3 SiO2 --->Al2(SiO3)3 + P2O5
P2O5 + 5/2 C ---> 2P + 5/2 CO2

Making AlPO4 react with carbon without the 'silicating step' may prove very hard... Are you getting significant yield with AlPO4 alone, is the question...


>There was apparently quite some H2O in the mix as evidenced by the significant condensate forming in the adaptor and the large amount of phosphine. If the gas was P4 I believe it would have just settled to the bottom of the receiver as white P.

> Back about one inch from the end of the adaptor flange you can touch the adaptor with your hand --it's not hot at all.

>I do plan to try the well known industrial reactants for making P. I will substitute pottery grade Ca3(PO4)2 for apatite, however:

Ca3(PO4)2 + 8C + 3SiO2 ----> 3CaSiO2 + 8CO + 2P

>I don't think I got a good yield of P. I did get a lot of phosphine. However, this likely did not account for much consumption of P.

blogfast25 - 20-3-2010 at 14:20

Seems to me you didn't produce much P and that some of it got converted to phosphine.

Another theory could be that the orthophosphate, in the absence of silica, was reduced to phosphide [-III], rather than to elemental phosphorus:

AlPO4 + 2 C ---> AlP (or similar) + 2 CO2

AlP + 3 H2O ---> Al(OH)3 + PH3

You seem rather hell bent on trying a very elegant set up on methods (mixtures) that seem unproven... :(

DO try Strepta's metaphosphate + silica + Al: salvation lies therein!

Magpie - 20-3-2010 at 16:05

What does this look like?

Quote: Originally posted by Magpie  
Today I made a small amount of P according to the following reaction:

6NaPO3 + 10Al + 3SiO2 --> 3Na2SiO3 + 5Al2O3 + 6P

[Edited on 6-3-2010 by Magpie]

Strepta - 21-3-2010 at 07:15

@blogfast: Magpie knows full well about the reduction of metaphosphates. He and I are exploring different chemistries for the lab prep of P4. The following excerpt is from Industrial and Engineering Chemistry, V 21, No 11, p 1130, originally posted by SC_Wack upthread several weeks ago:

“To determine whether addition of silica to the charge
increases the rate of reduction of aluminum phosphate,
10-gram mixtures containing equal weights of aluminum
phosphate and carbon and sufficient silica to give a 1 : 1 mol
ratio of alumina to silica were heated for 1 hour at 1050"
and at 1100" C. The percentages of phosphorus volatilized
were 5.3 and 73.7, respectively, showing that silica has very
little, if any, effect in accelerating the reduction of aluminum
phosphate at these temperatures.”

The importance of silica in the reduction of tricalcium phosphate is acknowledged and contrasted with this result. One result is that the reactants do not fuse together as they do with the metaphosphate/SiO2/Al reaction, making P4 recovery and cleanup much more straightforward.

blogfast25 - 21-3-2010 at 08:32

Touché! :o

Melgar - 1-4-2010 at 05:40

I made elemental phosphorus today. I was surprised how easy it was. I got a 6-inch long, 1-inch wide alumina tube off ebay for $20, and a roll of nichrome wire for like $5. I wound about ten feet of nichrome wire around the tube. Then I lined a big tomato can with fiberglass insulation, then made an inner layer of rockwool purchased from a hydroponics store. I nested the tube inside that, and packed both ends with fireclay plugs. I had a stainless steel tube coming out of the top plug and into a beaker of water. I hooked up alligator clips to the wire, then plugged the whole thing directly into the wall. A bad idea if you don't know what you're doing, but I do. I had it hooked up to one of those kill-a-watt things so I could see how much power I was drawing, which turned out to be like 500 watts.

The next step was packing it with reactants. I mixed ground up charcoal, phosphate fertilizer, and powdered fiberglass together and filled the tube with that.

Then I fired it up again with the reactants in it. At first a bunch of smoke came out. Then that stopped. Then some other gas came out, then that stopped. Then a bunch of white smoke came out, which caught fire as soon as the bubbles came out of the water. The bursts of flame continued for a while, then stopped. This was all done outside, btw. So I unplugged the device and gave it some time to cool off. Then I went and looked at the beaker. It had a white scum on top, that when I swirled it around, a bunch of flecks fell to the bottom. It also had that matches/fireworks phosphorus smell to it. I didn't take a huge whiff of it, but I couldn't help noticing that.

I was surprised at how simple it was, especially considering the lack of firsthand accounts when it comes to preparing elemental phosphorus.

[Edited on 4/1/10 by Melgar]

halogenstruck - 1-4-2010 at 10:51

it`s easier if u use microwave.
i mixed equal weights of charcoal powder and 85% H3PO4 very well then put in test tube and cotton wool and then salt layer==>microwave for 4 minutes.
a gas comes out which burns and at the end u can see green phosphorescence of P inside tube and if u put end of tube in water upside down u see it stops.
red spots in cotton wool and salt is red phosphorus
i did not have glass wool-glass tube,... to separate P.
it smells also like white P ,garlic.
huge fumes of P4O10 comes almost immediately after turning microwave on

halogenstruck - 7-4-2010 at 23:10

P preparation in microwave from H3PO4/C

Attachment: P prep in microwave.pdf (1.3MB)
This file has been downloaded 1075 times

halogenstruck - 8-4-2010 at 13:26

I MIXED 100% extra mole/mole charcoal powder and 85% H3PO4 then i put it at the bottom of a test tube,covered by a thin glass wool layer,upside-down inside a cup of water.
in first 4 or 5 minute,a lot of gas was evolved.but 16min was necessary to allow P ring comes down the tube.
because P releases very fast but immediately because of heat turns to red/violet P.
red/violet P melting point or sublimation temperature based on wikipedia is between 416 to 590 °C.
therefore it does not come out easily as mixture does not warm very well and needs lengthy heating in microwave to get warm enough all the area inside.
as another idea,it`s not a bad way just to heat in microwave for 5min then using the black mixture as a phosphorus containing mixture: Red P/active C in reactions
result was yellow P containing little Red P,waxy.
white fume in air due to auto-oxidation in air,greenish phosphorescence[chemiluminescence] especially when boiling with water
starting with 1.5g C resultin just a little bit.
taking out P from mixture is not easy.even in us patent related to microwave reduction,they calculated yield based on weight reduction
[Edited on 8-4-2010 by halogenstruck]

[Edited on 8-4-2010 by halogenstruck]

condennnsa - 7-5-2010 at 10:49

Quote: Originally posted by Melgar  
It also had that matches/fireworks phosphorus smell to it. I didn't take a huge whiff of it, but I couldn't help noticing that.

[Edited on 4/1/10 by Melgar]


Fireworks contain NO phosphorous. I think you confuse the smell of P2O5 with that of sulfur dioxide. SO2 is the smell of matches and fireworks. The only pyrotechnic composition containing phosphorous is armstrong's mixture (KClO3/P) and it is extremely sensitive and never used. Matches however do rely on phosphorous, but only at the instant when they are lit, the KClO3 from the match head and the red phosphorous from the striking surface react to give enough heat to ignite the rest of the sulfur/KClO3 mixture in the matchhead.

Jimmymajesty - 11-6-2010 at 06:14

I also made some phosporous from the strike surface of a matchbook, scaped off the red P with a pocket knife, then took a small glass tube sealed at one end, and poured the red P inside it, then I filled up the small tube with butane and heated with an open flame, the red P immediately turned into white.. It took about a second!

I went into a dark room and shook the white P drop out of the tube, it immediately started to glow upon contact with air:) It is a really nice experiement and easily carried out by anyone @ home.

entropy51 - 11-6-2010 at 06:22

Quote: Originally posted by Jimmymajesty  
I also made some phosporous from the strike surface of a matchbook, scaped off the red P with a pocket knife, then took a small glass tube sealed at one end, and poured the red P inside it, then I filled up the small tube with butane and heated with an open flame, the red P immediately turned into white.....It is a really nice experiement and easily carried out by anyone @ home.


I don't get it. You think white P is a good home experiment, but you just said:
Quote: Originally posted by Jimmymajesty  
I saw a video on youtube (elementary videos IIRC) in which the dude handled uranium and its salts, but in a glove box to avoid poisoning himself to death.. Who wants to bring chems like this to his home?
It is much easier to become poisoned by the fumes of white P than by U.

Jimmymajesty - 11-6-2010 at 06:34

I also do not get it what kind of illness can you get from H3PO4 possibly P2O5?

Random - 25-6-2010 at 23:43

Quote: Originally posted by Jimmymajesty  
I also do not get it what kind of illness can you get from H3PO4 possibly P2O5?



Quote:

Phosphorus pentoxide is not flammable. It reacts vigorously with water and water-containing substances like wood or cotton, liberates much heat and may even cause fire. It is corrosive to metal and is very irritating – may cause severe burn to the eye, skin, mucous membrane, and respiratory tract even at concentrations as low as 1 mg/m3.


From wikipedia.

not_important - 26-6-2010 at 00:53

Quote: Originally posted by Jimmymajesty  
I also do not get it what kind of illness can you get from H3PO4 possibly P2O5?


It's not the oxidation products, it's actual vapour from the white phosphorous - P4 molecules.

Do a Web search on the term "phossy jaw", you'll like the pictures.


Anders Hoveland - 27-6-2010 at 14:16

If the elemental phosphorous is being made just to obtain P2O5 (or P4O10) as a dehydrating agent, I have a suggestion that can avoid the need for distillation.
2Na3PO4 + 12S --> 3Na2S + P2S5 + 4SO2gas
obviously this is fused and heated together to drive off the SO2.

The Phosphorus pentasulfide could be used to make H2S, which otherwise would require metal powder to make.
P4S10 + 16 H2O → 4 H3PO4 + 10 H2S (from wikipedia)
Or the P2S5 could slowly burned, giving off SO2, and the P4O10 smoke would quickly crystallize after cooling.

Acetone reacts with P2S5 to make the unstable thioketone, which is extremely stinky, far more so than H2S. You can cause a small town to be evacuated with this reaction.

[Edited on 27-6-2010 by Anders Hoveland]

blogfast25 - 28-6-2010 at 08:24

Have you tried this in practice? This might take some serious temperatures, possibly above the BP of S...

Also I'm not convinced that ΔG < 0 for the first reaction. Of course if SO2 starts coming off it would drive the equilibrium to the right...

Random - 1-7-2010 at 13:29

I am interested in getting red phosphorus from matchboxes. What is the best way of doing it to make it almost pure?

Sedit - 1-7-2010 at 17:00

Locate a forum called wetdreams and then get to reading!

JohnWW - 1-7-2010 at 17:11

Red P is very soluble in CS2. If you can get that solvent, try it. It is also soluble in liquid NH3. (I am not sure about ".88" NH3 solution, though).

Or alternatively, you could find a solvent that dissolves the binder (I am not sure what it is - it could be ordinary water-soluble paste) used to glue the pyrophoric abrasive mixture to the matchboxes, leaving the red P and abrasive unchanged as solids. But then, the problem would be how to separate the red P from the abrasive.

[Edited on 2-7-10 by JohnWW]

Picric-A - 2-7-2010 at 02:02

JohnWW- I thought it was white P that was very soluble in CS2, not red P... could be both i suppose...

Try soaking the matchboxes in acetone, then scraping the red P off. that is as pure as you will get it uness you then sublime the red P, makign white P then convert that to pure red P.

woelen - 2-7-2010 at 02:11

Quote: Originally posted by JohnWW  
Red P is very soluble in CS2. If you can get that solvent, try it. It is also soluble in liquid NH3. (I am not sure about ".88" NH3 solution, though).
There is no solvent for red P at all which does not destroy the red P. If it does dissolve in some solvent (e.g. a mix of HNO3 and HCl does the job when heated), then it is oxidized to (usually) H3PO4. Indeed, it is white P which dissolves very well in CS2. The resulting liquid is extremely dangerous, it is self-inflaming in air, because the CS2 quickly evaporates and finely divided white P remains behind.

Random - 2-7-2010 at 03:53

I will try to soak matchboxes in acetone to remove glue at least. What are other things that are on matchboxes?

woelen - 2-7-2010 at 04:47

The brown material usually consists of a mix of red P, fine glass/silica particles to make it more abrasive, some glue and IIRC also some antimony sulfide (Sb2S3) in order to ease ignition. The amount of red P is only a few tens of percent, the rest is useless crap and it is not easy to separate from the red P. There is no solvent for red P, nor is there for the glass/silica and separation will be very difficult. You also need a LOT of match boxes in order to obtain decent amounts. One match box contains just a few mg of red P.

If you are in the EU, then just buy red P, it's not that expensive. If you are in the USA then try to find another area of chemistry which is interesting for you...

Anders Hoveland - 5-7-2010 at 23:33

Probably using boric oxide on a sodium pyrophospate would allow operation at a much lower temperature than silica and calcium phospate. All I want is the P4O10 for drying HClO4 and HNO3, so there is no need to reduce the oxide with charcoal.

mario840 - 27-7-2010 at 22:38

Hi guys !

You tell that PCl3 cannot be make from red phosphorus ?? So what about method this :

We make a little PCl3 first from say 10 g of WHITE phosphorus , so we have about 37 ml of pure PCl3 then we add this PCl3 to a red phoshporus in the flask and then GAS with chlorine ?? this may work ???

 Pages:  1  ..  3    5    7  ..  15