HMX synthesis

[Microtek]

I have obtained some acetic anhydride and paraformaldehyde in order to prepare some HMX. So far I have tried the synthesis on Mega's site which I have also seen several other places with very minor differences. My attempt was cut short because I had to leave, so both nitrating time and refluxing was reduced by more than half.
Even so, I got some white precipitate very similar to RDX, and I'm going to try the operation again today with proper timing. I did the test run with 1/30 the batch size compared to Mega's recipe, and so far a have a few comments on the reactions:
Temperature control is surprisingly easy, at least for this volume of reactants. None of the critical reactions are very exothermic and I very rarely needed the cooling bath I had prepared. I used the hot-plate set on low heat to heat the reaction vessel and contents to 45 C and then used a piece of cotton wool to maintain the temp between 43 and 45 C. This system could maintain temp with stirring for up to 10 min.
Speaking of temperature control, how critical is the 44 C mark and how long exactly do you need to maintain this temp ? Is it neccessary to keep it there right up to the point where you start refluxing ?
And when ice is added to cool the mix to 20 C, the ice is added directly into the reaction vessel right ? I didnt add much ice, but temp quicly dropped to -10 C.

[S. Toppholzer]

Don't you need that - er... was it boron triflouride?... as a catalyst for the reaction to work?

[DBSP]

You are thinking of the RDX process.

[Microtek]

I'm pretty certain that the BF3 catalyst is not used in this reaction. The method that I used was referenced as a commercial process ( though I don't know when it was used or if it still is ) and a process using BF3 was also referenced as a proposed method, so I don't think it was an omission.

[S. Toppholzer]

yes. Sorry. Just looked up the HMX process at Mega's lab site. Man, how much I'd like making that stuff... alas, it seems there is no way for me getting some acetic anhydride somewhere, hmpft...

[Microtek]

As far as I can see from the numerous sources that reference this method, the alpha modification of HMX is produced in quite high purity in this process. However it's the beta modification that is desired as it has superior density and VOD as well as better stability. Most sources indicate that the transition can be accomplished by "digesting" or "percolating" with 94 % aqueous acetone, or by dissolving in 94 % acetone and distilling the acetone off while replacing it with water. I suppose the digesting bit is simply an archaic term for refluxing but the question is for how long the refluxing needs to go on to ensure complete transition.

[Mr Cool]

The longer the better, I guess. It's not going to decompose at any significant rate while in the acetone, so if your reflux condenser is efficient and you have a stable heat source you could just set it running one evening, go to bed and get it out in the morning.

But, a few hours would probably do IMHO.

[spydamonkee]

would someone please be able to tell me what IIRC means?
just i see it used alot in this forum and its frustrating when i try to work out what it stands for :/ i know most others like IMHO, LMAO, etc etc
cheers

[zaibatsu]

If I Remember Correctly, IIRC <img border="0" title="" alt="[Wink]" src="wink.gif" />

[Richy]

is HMS less powerful than RDX because it has a lower density or what?
isnt the detonation velocity higher?

[Microtek]

Why do you get the impression that HMX is less powerful than RDX?
In some tests, RDX is just as good as HMX, but generally HMX beats "little brother".

[mongo blongo]

Remember that RDX contains some HMX after synth. Pure RDX is generally less powerful than pure HMX.
Also If you can't get Acetic Anhydride for HMX synth but you can make RDX then you could have a go at separating them. It can be done by agitating your crude RDX in Dimethyl Sulfoxide which will dissolve the HMX present and not the RDX. A small amount of water can be added as water is a cosolvent of RDX. Filter and ppt the HMX with water.

[zeocrash]

I was looking at a a patant for making RDX (3,937,703 its been uploaded to ftp already). according to this patant, the HMX produced with RDX Can exist in several polymorphic forms, possesing high impact sensitivity, and unless completely removed, would increase the impact sensitivity of RDX significantly. I Know that DMSO dissolves HMX but not RDX, and is available for rubbing horses muscles down. unfortunatly i cant find any horse superstores in my local area. does anyone know what other products contain DMSO, or What other substances disolve rdx or hmx, but not both.

[Microtek]

beta-HMX has a sensitivity between PETN and RDX IIRC. The other polymorphs of HMX can be converted to the beta polymorph by recrystallization from acetone. Or by dissolving in acetone, then distilling the acetone off while replacing it with water. I expect you could do this even with RDX present in the system, and it would raise the VOD of the product while only becoming slightly more sensitive.
BTW, RDX is not considered safe enough to use alone; at least a few percent of wax is needed to desensitize it ( for military application ).

[Microtek]

Recapitulating my current standpoint in HMX synthesis:
My first attempt was to use the process on Mega's site. It has the advantages of requiring only Ac2O in the way of inaccessible chemicals and does not require isolation of intermediate products. On the other hand it relies on very specific nitrating conditions to produce HMX rather than RDX. Thus, it is neccessary to maintain a temperature of 44 +- 1 C for relatively long periods of time.
Also, my yields using this method were not so good though that may have been due to the HNO3 being less than 100%.

Another method which is described in a ( slightly suspect ) patent is the nitration of SOLEX with an HNO3/P2O5 system. SOLEX is similar to HMX but has an acetyl group in place of one NO2 group. It can be produced by nitrating TAT with the same system as above. TAT is again similar to SOLEX, but has acetyl groups on all four ring-nitrogens. As can be seen from this, the basic ring structure remains constant throughout these steps and I assume that this would tend to preclude any RDX from appearing in the final product.
TAT can be made from DAPT which is not the same as DPT. It is a bicyclic compound with an outer ring that is similar to the tetraazacycloctan ring, but the 1 and 5 nitrogens are connected via a methylene group in the center of the ring, the other two nitrogens have acetyl substituents.
For making DAPT I used the method from the patent:
- 5 g water is added to 5 g hexamine which dissolves most of it at room temp. It is then cooled to 0 C and 2.5 g ice is added.
- To this slurry 7.2 g acetic anhydride is added over 30 seconds which causes temp to rise to ca 50 C.
- After about one minute the reaction should be complete and the acetic acid byproduct needs to be removed.
- This is done by evaporating it selectively by passing a stream of air saturated with water over the solution at 40-50 C for as long as it take to remove all acidity from the vapours.
- Soln is then heated to 130-140 C to remove formaldehyde polymers and water. This is also done in an air stream ( dry of course ).
- After 20-30 minutes of this, mix is cooled and DAPT precipitates or solidifies and can be collected.

[Hex]

Microtek,

Checked my notes this morning - found out I actually had to use a proper vac pump, rather than a water one, to get my DAPT to go solid on the rotary evap. I wa doing it at about 60C. It's kinda waxy, but re-crystallises into nice...er, shapes (my inorganic chemistry is letting me down again). DAPT is pretty stable; I don't think you'll have any probs with hydrolysis in the work-up unless you take the temperature very high.
For a synthesis avoiding anhydride, I'd be tempted to try nitrating the di-nitroso analogue of DAPT, which can be made in 76% yield using acetic acid, hexamine and sodium nitrite. There's only 3 records of the likely initial nitration product ("HMX" but with two nitro groups replaced by nitroso), and two of those have it as a product from bacterial degradation of HMX. I'll see if I can get hold of the other one. This material is likely to be a pretty quick explosive in it's own right, and it might be possible to oxidise/nitrate it to HMX.

<small>[ February 12, 2003, 03:49 AM: Message edited by: Hex ]</small>

[Mr Cool]

Now that I can try out. It really sucks having no anhydride :( .
Ah yes, I remember now reading about that (the dinitroso analogue of DAPT) in the Swedish Informania files (it sucks not having access to any scientific journals, too!).
Maybe the dinitrodinitroso analogue of HMX, which should be formed by nitrating the dinitroso-DAPT-like molecule could be oxidised with a percarboxylic acid? I know that's the prefered method to get an N-&gt;O dative bond with pyridine etc. It's worth a try at least.

(when I say "oxidised", I mean to HMX of course)

<small>[ February 12, 2003, 10:00 AM: Message edited by: Mr Cool ]</small>

[Mr Cool]

Well I've scanned through the SIF for useful stuff, so I thought I'd post what I found here.

The best method of obtaining DPT consists in introducing hexamine dinitrate to 90% sulfuric acid at 8-15C. After 45 min. all is poured on ice and the solution is filtrated. The filtrate is neutralized with 28% ammonia to pH = 5.5-6.5 DPT precipitated [165].
A method of making octogen from DPT (1 mole) consists in
acting with NH4NO3 (1.6 mole) and nitric acid (3.2 mole) at 60-65C for one hour. Water is then added and all is heated for 12 hours in a steam-bath. The yield was 75% of the theoretical [from DPT] (according to [5]).

[I guess the heating for 12h in water decomposes the less stable products, leaving safer HMX?]

And for making the dinitroso-DPT thing:

Dissolve 7g of hexamine in 50mL of water, and add 13mL of conc HCl. Mix this solution with 400g of crushed ice.
Dissolve 21g of sodium nitrite in 50mL of water and slowly add it to the first solution while stirring rapidly.
After the addition, filter out the crystals and dry them. Dissolve them in warm acetone and dump the solution into 100mL of 5% sodium bicarbonate solution. Filter and dry the ppte.
The yield will be around 7g.

I'm going to try this last procedure now to verify it, since that is a rough translation and may not be 100% accurate.

I hope that was useful to someone other than just me...

Progress update:

It looks to be quite a good procedure.
I used 100g of ice, I thought it might be a bit hard to stir it well with 400g, and I used 30g of 15% HCl because I don't have any conc, but apart from that I followed the procedure exactly (except for the minor point that I added the ice to the HMTA soln before the HCl).
Rate of addition of NaNO₂ solution was approx. 0.5mL/second, adequate to prevent any NOx forming in noticeable amounts.
I filtered the mixture immediately after I finished the addition, although I kept the solution and some ppte is till forming, slowly. It's in the fridge now. The ppte so far (not dried or recrystalised yet) looks to be about 5 or 6g, taking into account the fact that it's a bit foamy.

Update:

Recrystalisation required about 75mL of acetone, at just below its boiling point. Some solid remained, which I do not believe was dinitroso-DPT. After filtering, to the solution was added 50g of crushed ice, and it was chilled to 0*C. After stirring for five minutes at this temperature, the ppte, which had lost most of its yellow (dis?)colouration, was filtered and air dried. The yield was a fairly shocking 5.1g, 55% :( .

I'll have another go with a carboxylic acid (probably citric acid, since it's the only one I have that's pure or concentrated at the moment) before I try altering anything.
One thing I don't like about this method is the fact that R-Salt has an almost identical synthesis, so the product is likely to be contaminated with it, and I have no way of measuring the degree of contamination.

<small>[ February 12, 2003, 04:02 PM: Message edited by: Mr Cool ]</small>

[Microtek]

HEX: Concerning that direct synthesis of DADN from the DATP solution with HNO3/H2SO4, I found a patent ( US 3926953 ) which details method of avoiding the exotherm and decreasing HNO3 consumption by adding urea to the DAPT soln before addition to the nitrolysis mix:
Soln of DAPT in acetic acid:
0.25 mol Ac2O is added to slurry of 0.10 mol hexamine, 0.082 mol ammonium acetate and 0.139 mol H2O at 5-10 C ( as you said in your post in the TNPDU thread ).
0.02-0.10 mol urea is added.
DAPT --&gt; DADN:
The above soln is added over ca 6 minutes to a nitrolysis mixture consisting of 0.33 mol 90 % HNO3 and 2.25 mol conc H2SO4, keeping temp at 35-40 C.
After addition, mix is aged 10 minutes at 40 C, ice is added and mix is drowned in water ( quite a lot of ice and water is used here ).

Another patent describes a way of recovering HNO3 and H2SO4 from HMX production by keeping the waste acid flows from DAPT/DADN production separate from the DADN --&gt; HMX waste acid flow. Then H2SO4 can be recovered simply by boiling the DAPT/DADN waste flow as any traces of organic matter ( down to 1 ppm ) is destroyed and leaves as gas under the harsh conditions of sulfuric acid at 170 C ( they do it at 50 torr ).

BTW, I've only been able to find a single method for the nitration of DADN to HMX ( 3.6 mol N2O5, 26 mol HNO3 per mol DADN ). Do you know of any other ?

[Hex]

Microtek: Haven't been able to find a precise procedure for this, but have found some data on a Swedish Process which uses 1:2.5:9 DADN:Nitric:Polyphosphoric acid at 65C for 90 minutes, giving a 73% yield. I'd guess the DADN gets added to the cold mixed acid, then heated to the reaction temperature; the work up is probably just drowning in ice/water.
Very interested to hear how your work with the dinitroso compound goes - it may be possible to "oxidise" your product to HMX using good quality 100% nitric (I'm not sure if this is an oxidation reaction or nitrolysis at N-NO, but it works for a few compounds) I've seen similar ones which have 10% H2O2 (presumably anhydrous) in the nitric.

[Microtek]

Are you sure that is right ? 1:2.5:9 DADN:HNO3:P2O5 will surely be almost impossible to stir with that much solid. Did you mean 1:2.5:9 DADN:P2O5:HNO3 ?
Anyway, it is Mr Cool who might be doing the nitroso-derivative as I don't have access to nitrite, it has been labeled "toxic" be the ministry of health which means that a lab needs to have permission to work with poisons to be able to order it. The company I work for doesn't have those permits so I'm cut off from a lot of compounds. It's not all toxic substances that are labeled as toxic and I can order these with impunity. But then there are some substances which aren't very poisonous that ARE labeled...

I think I'll make a moderately large batch of DADN and then experiment with the nitrating conditions for the last step or steps.
As you know if you've read Lukasavages patents, he claim not only that method for making TAT which I can't get to work, but also conversion of 100 g TAT to 100 g SOLEX with 150 g HNO3 and ( I think ) 48 g P2O5. The SOLEX --&gt; HMX process is similar.
Of course, if the DAPT to TAT method is not as good as he says it is then the rest is suspect as well, but experimentation will have to be the deciding factor.

[Mr Cool]

I've now accumulated around 20g of the dinitroso compound to experiment with, but further work is going to have to wait because I'm going away this weekend. I would have had more, but I couldn't resist burning some. It's like HMTA, but more vigorous and with a nice red tint to the flame. Very pretty.
I think it might be possible to oxidise it (or whatever) to the dinitro compound and add the two further nitro groups on in a single step. I've done a bit of reading, I'll need to try and find out a lot more before I try anything out, but from what I've gathered so far it seems that the -NO --&gt; -NO₂ reaction procedes at low temperatures, and then further nitrolysis occurs under more vigorous conditions. So if I start off in an (salt/)ice bath and then move it to room temperature and then to a hot water bath it might hopefully work a bit (lol, I'm not terribly optimistic at this stage!).

The yield is still quite low for producing it though, the larger batch I did gave a yield of a fraction over 60%. This time I mixed the nitrite, ice and HMTA first, and then dripped in 15% HCl, just because I fancied some variation really. Precipitation of the product commenced quite suddenly after adding 62g of acid, which was done in 2g portions (90g in total).
Again, there is some stuff which seemed less soluble in acetone than the majority of the ppte. It's pure white in colour, in contrast with the pale yellow of the other compound. Intriguing...

I'm also going to have a crack at HMX from DPT, even if the yields are poor for making DPT all the chemicals are cheap and common.
And then there's RDX from sulphamic acid, the ammonium sulphamate is drying so that will be ready for me when I get back. It's annoying how I always seem to get bursts of motivation just before I go away, and then when I come back I'm too tired and can't be arsed to experiment!!

[Microtek]

I have now done some experimenting with the hexamine --&gt; DAPT --&gt; DADN synthesis and apart from some initial minor problems, I must say that this is one of the best synthesis I have ever done ( except for PETN, but then that is so much more simple ).
I'll present the entire synthesis along with my observations even though it is not much different from what is evident from the above discussion:

For all steps involving a weighing of reactants, a homemade scale was used which is accurate to 0.01 gram.
Hexamine used was a no-name brand which is very likely technical grade hexamine that is pressed into pellets without binder.
Acetic anhydride was J.T.Baker, 95 % purity
Other reagents were from pharmacies or drugstores.

Hexamine --&gt; DAPT

- 3.46 g ( 0.058 mol ) 100 % acetic acid and 4.9 g ( 0.058 mol ) 20 % NH3 in water were mixed dropwise with rapid stirring
- 10.0 grams of hexamine ( 0.070 mol ) was pulverized and added to the above soln forming a slurry, which was cooled to 5 C.
- 10.8 g ( 0.101 mol ) acetic anhydride was added with manual agitation over the course of an hour, keeping temp at 5-10 C ( my cooling bath could have been much better, in which case I could have completed the addition faster ). The mix starts out as a slurry and as more AC2O is added, it first takes on the appearance of a colloidal suspension and becomes rather viscous ( which is why I did the agitation by hand instead of with my newly made magnetic stirrer ). Then as the end of the addition approaches, it slowly clears up and becomes less viscous. After addition, mix was stirred
30 minutes at 10 C with a magnetic stirrer.

DAPT --&gt; DADN

This step was done with 11 mL of the produced soln ( 38 mL total ) In order to do other tests on the remaining volume.

- 0.82 g ( 0.020 mol ) urea was ground up and added to the mix. It took quite some time to get it all to dissolve ( ca. 10 min ).
- A nitrolysis bath was prepared by adding 3.0 mL ( 4.5 g, ca. 0.07 mol ) distilled HNO3 90+ % to 50 g H2SO4, 96 %. The bath was placed in a water bath at 15-20 C.
- 11 mL of the DAPT soln was added during vigorous stirring with a paddle stirrer made by attatching a stainless steel paddle from a cocktail mixer to a small electric motor. The addition was done at such a rate as to keep temp at 40 C. I found that even though there is no sign of an abrupt exotherm when one mol urea is used for each mol DAPT ( or hex ), rapid stirring is still neccessary or the yield will suffer a lot.
- The addition was done in just 5-10 minutes and afterwards, the mix was held at 40 C for 10-12 minutes by exchanging the cool water for progressively hotter baths.
- After this short aging period, the mix was poured over ca 200 g ice cubes in a 1-liter beaker and 300 mL cold water was added. A stirring magnet was added and the diluted soln was stirred for 3-4 hours to precipitate most of the DADN. It is odd that while DADN is extremely insoluble in water, it can take the better part of a day for any precipitate to show at all after drowning, if the soln is left undisturbed.
- The pure-white precipitate was filtered through standard filter paper over a coffee filter and once completely dry weighed. Weight was an incredible 5.45 grams from 11/38 of the original hexamine slurry corresponding to 18.83 grams for the entire batch or a yield of 92.2 % from hexamine. Considering that a little more precipitated from the mother liqour over the next day, and that some stuck to the filter paper, true yield must be almost quantitative.

[IPN]

Two questions:

1. You said: </font><blockquote><font size="1" face="Verdana, Arial, Helvetica">quote:</font><hr /><font size="2" face="Verdana, Arial, Helvetica"> 3.46 g ( 0.058 mol ) 100 % acetic acid </font><hr /></blockquote><font size="2" face="Verdana, Arial, Helvetica">Do you mean glacial acetic acid or anhydrous acetic acid?

2. Does the ACO2 mean something like actinium oxide or what?
If so where can you get something like that?
[It's Ac2O which is an abbrevation for acetic anhydride]
I´m just asking because this looks like a very easy thing to do.

<small>[ February 17, 2003, 01:59 PM: Message edited by: Machiavelli ]</small>

[Microtek]

Ac2O is indeed acetic anhydride ( I mistyped it as AC2O ), 100% acetic acid is the same as glacial acetic acid which is infact anhydrous ( waterfree ) but is NOT the same as acetic anhydride.
And as to your last comment, it does indeed appear that this route to HMX would make HMX at least as convenient to prepare as RDX ( assuming you can get the Ac2O and the P2O5 which will be needed for the last two nitro groups ).

HEX: In that Swedish method for DADN --&gt; HMX was the ratio you gave by weight or by mol ? I have done a test nitration using it as a weight ratio and assuming that the P2O5/HNO3 ratio should be reversed and have gotten a fluffy white solid when drowning in ice water.
I suppose it could be alpha-HMX, so I will try a slow recrystallisation from acetone to see if I can convert it to the beta polymorph. I will post yield when the material is dry.

[Hex]

Microtek,

The figures are weight ratios, - note that it's actually polyphosphoric acid (PPA) they use, rather than P2O5. It's a rather smudgy photocopy that I've got, and the tables (they tried many different ratios) are very small, but I'm fairly sure the figures are correct - and in the text they quote "at approx 70% weight polyphosphoric acid, pure HMX could be prepared in yields of 81-89 (99? - photocopy unclear)%" They suggest that free phosphoric acid inhibits the reaction and this is why such a large excess is needed. No mention of the quality of PPA - the standard stuff is equivalent to 115% H3PO4, but it's oligomeric, so higher percentages may be available.
They also say that little work has been done using P2O5 due to "impracticality" - presumably they mean on an industrial scale. Your reaction sounds quite promising, if it's not starting material it can only really be RDX or HMX, or most likely a mixture of both.

[Microtek]

I have just found out what polyphosphoric acid is; I thought it would be a crystalline material composed of around 30 % H3PO4 and 70 % P2O5 so some kind of low grade technical P2O5. I see now that it is in fact made by dissolving P2O5 in ca 85 % aqueous H3PO4 and is liquid. In that case I can understand what they mean by saying that P2O5 is impractical. I have quite serious problems with measurement and addition of the solid P2O5 to HNO3 because it is so hygroscopic; it clumps together on the spoon and sticks to any tool I use to handle it. Also, all operations with it must be done quickly to avoid it absorbing too much water. So, a liquid with a known density would be much easier to handle.
Regarding your comment that my product is probably a RDX/HMX mixture, I must say that I thought having the basic tetraazacyclooctane ring in the starting material ( DADN, TAT or SOLEX ) would preclude the possibility of RDX forming. Also, wouldn't SOLEX be another possible product of the nitration I attempted ?

[IPN]

Getting the acetic anhydride is quite simple (I need to make it my self) but I don’t know about the P2O5. I may get it from my chemical supplier if they will ship it. I guess it isn’t so hazardous.

But this procedure in whole is the next thing i will probably try, after my nitrosugar experiments.

[Hex]

Microtek,

You're right - RDX isn't a possible product from DADN or TAT. Not sure what I was thinking there! It is possible from DAPT however, but unlikely. Although the cyclohexane structure is still there in DAPT, it's still more likely to give a cyclooctane. Assuming the first step is nitration at one of the bridgehead nitrogens to give a quaternary amminium compound R3N-NO2(+), there's 3 methylene-nitrogen bonds that can break; 2 identical N(Ac)CH2-quaternary N and one N(other bridgehead)CH2-quaternary N. On the face of it, it would appear statistically twice as likely for the first of these to break, which would break up the cyclooctane ring, and leave only the cyclohexane ring intact. But when the N(bridgehead)-CH2-quaternary breaks, the positive charge is stabilised through R2N-CH2(+) &lt;-&gt; R2N(+)=CH2. This stabilisation isn't available through the N-Ac nitrogens as they are, as amides, already partially pi-bonded [R2N(+)=C(CH3)-O(-)].
That sounds as clear as mud, it's hard to explain without drawings. Draw DAPT with an NO2 on one of the bridgehead N's and you'll see what I mean.
Anyway, the important thing is, you're absolutely correct - it's unlikely to contain RDX!

[Microtek]

Excellent! You had me quite worried there for a while...
Anyway I have weighed the dried precipitate and unfortunately, the yield was very poor; 0.475 grams or about 35 % so I'm going to try some of the other methods.
I have found patents which claim 87 % yield from 19.8 mol HNO3 and 3.6 mol N2O5 to each mol DADN, and of course I have to try Lukasavage's method though it sounds somewhat suspect; he's using 150 g HNO3, 48 g P2O5 and 100 g TAT to make 100 g SOLEX. That's a 1:1 molar ratio of TAT:P2O5 to attach three nitro groups, though I suppose that if the two first NO2's could be attached with HNO3 alone it might work... Anyway, I was thinking of trying this with DADN in place of TAT which must be at least as likely to work.
I have some problems with characterization of the product as I can't measure melting points with sufficient accuracy at those temps. So far I have done a little flame test, and it ( the product of my low-yield nitration attempt ) definately wasn't DADN. It burned with much more vigour than DADN, akin to NC/NG mixes emitting a hissing noise and producing no smoke.
Regarding the troublesome nature of P2O5, I was wondering if I couldn't dissolve it in some sort of inert solvent. Then it might be possible to quickly weigh off a given amount, dissolve it in the solvent to give a certain volume of solution and then use a volumetric method to add a certain amount of P2O5. For solvents I was thinking either glacial acetic acid ( though I guess that wouldn't be inert but exist in equilibrium with Ac2O ), acetic anhydride, nitromethane ( suggested as alternative liquid diluent for N2O5 in one of those patents ) or chlorinated hydrocarbons such as dichloromethane, chloroform or trichloroethane ( which is the only one I can buy pure without ordering at the chem supply ).

[Mr Cool]

Just to let you know the results of the few experiments I've been doing. There's not much, but I thought I'd let you know anyway.

I've given up on dinitrosopentamethylenetetramine. Even with freshly distilled (only once at 1 atm, not really good re-distilled in vaccuo etc., but should still be 90%) nitric acid at -12*C, it instantly disappeared with a poof of vapour, leaving black bits in the acid. I'm sure it could be made to work, but it's likely to be more effort than it's worth.
I've been getting really bad yields of DPT, around 15%. I've seen it stated as 25-30%, which is still bad, but about double what I got! Anyway, I'll have a go at converting that later tonight (I have 4.5g to do a little experiment with).
I have also seen a method for making DPT using nitrourea instead of HDN and H2SO4. Originally posted by ALEN:

"Materials:
21g nitrourea
120 ml 37% formaldahyde solution
200ml 5% sodium hydroxide solution
250 ml 28-30% ammonia solution

Procedure:
add to the nitrourea to the formaldahyde solution and heat it to 45 celsius and stirr untill all of the nitrourea is dissolved. When it has alldissolved, take it off the heat and let it cool to room temperature. While stirring, add 200 ml of 5% sodium hydroxide solution and continue stirring for 1 hour. While, stirring heat the mixture to 65 degrees celsius for 1 hour while stirring. Remove the mixture from the heat source and let it cool to room temperature. Slowly add 200 ml of 28-30% ammonia (could probally use more less concentrated ammonia). Filter off the precipitate and set it aside. Add 100 ml of 28-30% ammonia to the filtrate and filter off the precipitate. Combine the precipitate and wash it with 1L of water.

total yeild is around 11 or 12 grams."

This looks like a much better method in terms of yield. I'll try it when my ammonia soln. stocks are replenished...