Neglected RDX Process

I have a large PDF Merck (11th ed.) and could not find any "sulfaminic" acid. But that doesn't mean it doen't exist of course. I just could not find any info....and the search engines continue to correct the name....

Dyslexia

With regards to the usefulness of NH4 salts in the nitrolysis reaction producing
RDX , the sulfate salts are even more interesting than is ammonium nitrate ,
since the yield is increased even beyond
what is gotten using the nitrate . The sulfate is more interesting also because there are a series of sulfate salts , not only the normal sulfate , but the acid sulfate , and the pyrosulfate , which have
untested potential .

It is not clear if the NH4 salts will have as much value in increasing yields in nitrolysis
mixtures where a lower excess of HNO3 is
being used , and HDN is the precursor instead of hexamine . Neither is it clear if
the stoichiometric ratio of NH4 salt which is optimum should be calculated on the basis of the amount of HNO3 , or on the basis of the hexamine or hexamine contained in HDN . So there are some unknowns with regards to adapting the patents ratios to a more economical nitrolysis using from 2 to 2.5 ml of 1.5 HNO3 per gram of HDN . Also it is known that urea nitrate in .5 - 1% added to the
hexamine by itself can give a 4 or 5% yield increase in the nitrolysis using low amount of HNO3 , like 2 ml HNO3 per gram of HDN . Dicyandiamide ( and probably even better its nitrate ) produces a similar increase when used the same way . The one thing in common about these two substances and ammonium salts is that they tend to act as anti-oxidants in nitric acid , reducing the oxidizing decomposition products of nitric acid to nitrogen and water , while themselves being decomposed . Another
variable is whether it is better to add the NH4 salt to the acid , or premix it with the HDN or hexamine , or apply both strategies .

It is this property of ammonium salts which accounts for a more pure form of nitric acid being obtained by distillation
of H2SO4 + NH4NO3 mixtures than is gotten from use of alkali nitrates .

Since nitrolysis mixtures are heavily polluted with the very sorts of unstable decomposition products which behave similarly as an accellerated decomposition
of HNO3 itself , this is probably why is
realized the usefulness of NH4 salts and
urea and dicyandiamide in the nitrolysis
mixtures , as they stabilize the deteriorating acid and oppose the autocatalytic decomposition from accumulating byproducts , maintaining a higher nitration quality for the acid while the nitrolysis is in progress by slowing its
deterioration to a more oxidizing and contaminated acid . This is even more important when the lessser amounts of acid are being used for economy .

That's my theory concerning the usefulness of NH4 salts under conditions
which are much different from the K-process parameters , and involve no
" recombinant formaldehyde " reactions
which seem unlikely to me . As I see it ,
no free fromaldehyde would long survive in a nitrolysis mixture , to be reacting with anything except the nitric acid itself and
quickly converted to formic acid as a stable
end product and byproduct in the mixture .
I just can't visualize any Knoffler style scheme for the utilization of the formaldehyde , maybe in reaction with nitroamide which could be present , but
not according to the classical textbook
comments on the Knoffler reaction ....IMHO those theories just have it wrong whatever is occuring to explain some anomalous utilization of byproduct formaldehyde . It may be that formaldehyde is somehow being returned to the reaction , just not by the mechanism they are suggesting , which is doubtlessly an oversimplification of what is actually occurring . A nice way of saying they have an idea of what has occurred but little clue as to the mechanism Anyway , there are
many complex reactions occurring in a nitrolysis , and there is probably a different explanation for the reactions which are most likely at different temperatures and concentrations of
acids and NH4 salts . Only in certain
reaction conditions which are specific to
the particular temperatures and ratios of reactants , will an increase be produced
beyond a plain HNO3 nitrolysis . If the
correct proportions are not used , then
a negative effect is the result , and that
is something of an enigma itself to explain .

Mean Dyslexia you? Happens me to never.

Actually, I wanted to avoid working with the older methods and try something new like an R-salt process or the titled method using sulfamic acid but I just couldn't afford the chems and I just hate distilling acid... But Roscoe posted that patent re: higher yields via ammounium nitrate....Hell, I had to try it.
But mind you I am not THAT wasteful of acids. So I used a 5:1 ratio with DNH and simple hex w/ the addition of NH4NO3. I need to be looking at the patent and I dont have it in front of me but I believe I used Example 2 so that is one part by weight ammonium salt to fifty parts acid (tiny bit) but my extrapolation from that varied from the example which used a 1:12 ratio of acid to hexamine. I used a 1:5. I received 23.5 grams totally dry product from 25 grams hexamine - That's close to 70 when I looked at the DNH yield which was better and averaged so I would say I was about at 70 via the process. I don't have my notes here so I can't show the math but it's close...(I just got taught a lesson about drying my product and made sure that stuff was BONE dry this time) My DNH nitration was much better but I don't want to post the figures as that was a smaller one and I don't have the figures in front of me (but all the other factors were the same, temp, time in each bath, etc).
The major issue was that when I used a simple nitration previously I has a "flash" when I added the hex (it would actaully fire!) even in amounts that were tiny. It was dangerous and time consuming. By the use of the patent proceedure I received no flash and I believe my yields improved by worthwhile percentage. Previously I believe that from a ratio of 25gr to 125 gr of HNO3 I would get about 19gr and it would take all damn day (using un-nitrated hex). I am extremely careful and wear the appropriate protective gear but that flash seemed unavoidable. I didn't enjoy nitarting hex as it seemed that it in-and-of-itself was wasteful of material. My purpose was not to make RDX per se' but to learn more about the niration process. Thus if I had x amount of starting material and received y amount via an intermediate nitration (DNH) in a sence I was receiving close to the same final product. If I could improve the final yield even if it was a few perrcent and not have to waste hex through a di-nitration, that was useful information, plus eliminating the flash....it seems like I got just what I had hoped for in that info. Example 2 claimed 85% yields as I remember but they used a lot of acid. 12 to 1 ratio I believe.

Jeez, I just re-read this post, it's early out here and I'm at work - I am babbeling.....forgive me; you get the idea of what I was trying to get accross.


[Edited on 9-11-2005 by quicksilver]

Your thoughts on accellerated decomposition are well taken. I wonder were these processes to take place in industrial settings if some decomposition could be avoided..... However from what little I worked with I would say that the quality of the HNO3 has an impact in a small-scale experiment. - It DID indeed work out the way the patent pointed to. But I would hasen to add that this was not preformed repeatedly and that on other occations I have gotten some variences for no explainable reason I could find. But not this time.
When variences have occured in times past I noted everything I did. But I can find no yardstick for them.
In this instance I took the leap that the stoichiometric ratio of NH4 salt which is optimum would be calculated on the basis of the amount of HNO3. But you say Dicyandiamide [& urea nitrate] may also be of value....and point to them acting as anti-oxidants in nitric acid. This rings a bell as acid quality has been a big thing with me. If it's REALLY clear white fuming; I get results.

I had thought about the sulfate and it's potential. Since for no other reason than it's unbelievable availablity I would enjoy testing it as well. - Could you please elaborate on the pyrosulfate?

This would be fantastic for only 100 ml of acid!
The whole issue of the theoretical amount of acid and that which is utilized in practice puzzels me. I am turned off by the demands of acid by many energetic material synths in patents ( twelve to one ratio is really too much for me....as in the patent we are talking about....) but at THAT level it makes the endevour practical.
When I read 70 grams of hex to 875 grams of acid I thought I read it incorrectly and it was really 375!


What do these guys think; that acid is free or something?
I hope you try it: I am very curious.

Yeah that was my post that got misdirected to the old server during the changeover and lost in limbo for awhile . Then it was imported and pasted back into the thread .

I am sure I once had the process worked out to quantitative yield , but thinking back on it now it seems like it may have been a two stage distillation which I was doing years ago . There was a routine where an excess of H2SO4 was used to first distill d. 1.5 nitric until a measured quantity had come over , and then the receiver was changed , an additional amount of NH4NO3 was added and the distillation continued to produce azeotropic nitric as a second yield utilizing the excess acid from the first distillation , and the combined " fractions " amounted to the quantitative yield , with part being d. 1.5 HNO3 , and part being d. 1.41 HNO3 , 97% and 68% respectively .

I'll probably find the notes from where I worked it out the first time , after I work it out again . Keeping track of lab notes is the only way not to be continually mystified by your own work

( Great minds think alike )

That 74% yield may hold at an even
lesser ratio of HNO3 to HDN , such as
1.7 ml HNO3 per each 1 gram HDN ,
but it is getting into the range where the
concentration of the acid is a more significant factor in whether you can
maintain yields at 74% or see diminishing yields because of the increasing dilution
of the acid from the relatively greater byproduct water from the nitrolysis .

See the attached patent page 2
Example 1 . The acid is of 99.5% concentration , and the nitrolysis
described is for hexamine . Doing
the rough mathematical analysis ,
it looks like the ratio is for each gram of hexamine ~ 3.84 ml of the HNO3 ,
producing a yield of 74.3% .

Extrapolation for HDN instead of hexamine , that is 1.7 ml of HNO3
per each 1 gram of HDN .

Perhaps the inclusion of NH4 salts which are catalytic would be of benefit that may
offset the deterioration of the acid so that
the utilization of the acid is improved sufficiently at the lower ratios to improve the economy of the reaction even with lower concentrations of HNO3 like
95 - 96 % acid . If this occurs , the economy of the nitrolysis might improve
considerably in terms of yield of RDX per
each ml of HNO3 , even with decreasing
perentage yields based on theory . For example if a significant added amount of HDN was only 50% converted to RDX ,
it would be worthwhile to gain the additional end product from the less
efficient conversion of the final portions of HDN , even though it would lower the
overall percentage yields based on theory
for HDN . The economics of the acid utilization are important enough that it
influences the rationale about where the
" extra " amount of HDN is no further value for causing diminishing return or
instability of the reaction . For example
if the basic synthesis gives 80% yields
based on HDN , and adding half again the amount of HDN reduces the theoretical yield to 65% , the overall economics favor
the 65% reaction if acid utilization is the priority . It gives me a headache to analyze these sorts of curves to find the
" sweet spot " for the economics of such reactions , a task which is only done from observing many reactions studied in depth by the devoted bean counter

Attachment: GB658976 Cyclonite Improvements.pdf (410kB)
This file has been downloaded 1219 times

The experiment was done with addition of the HDN at 21 C over 30 minutes and after stirring for an additional 30 minutes the mixture was drowned . A very poor yield of RDX resulted .

It is my belief that some heating to a higher temperature is definitely
required for these reactions which use lower amounts of HNO3 .

In early patents as well as many of the later patents which describe much higher HNO3 content mixtures , nearly all of the procedures describe heating the mixture to some extent after all of the HDN is added and has reacted for a few minutes at a lower temperature . One patent describes heating to 40 C for a holding time of at least 1 hour , and several patents describe heating to 55 - 60 C for
at least 5 minutes , which seems to be about the minimum even for higher acid content muxtures .

Urea Nitrate is described as useful to be added to the mixture after the HDN addition and reaction at the lower temperature is completed , before the mixture is raised to the higher temperature . At the elevated temperature there is a decompositon of byproducts , a reaction controlled and stabilized by the urea nitrate , so that it proceeds smoothly and safely .

But I believe that it is more than just the
decomposition of byproducts which is caused by heating the mixture , particularly for the lower acid content mixtures . The heating is what actually drives to completion the reaction which
has started at the lower temperature during which the HDN was added , so heating the mixture is essential for much more than just decomposition of byproducts . Omitting the heating and holding time for a low acid content mixture
will therefore drastically reduce the yield ,
since the heating is necessary to drive the desired reaction to completion , regardless of the decomposition of byproducts . There is likely a correlation between byproducts and the desired product in a low acid content mixture , so you don't get one without the other , and evidently you get very little of either in the absence of the required heating

Actually I suspected this was true since I have experimented before with RDX many years ago and never had good yields from any experiments which did not involve heating . But I wished to give the
" cold process " described for higher acid content mixtures in a few patents another
try to see if the added ammonium sulfate would change the reaction profile with regards to temperature , since I hate heating a mixture like this which would seem to have so much potential for an
" event " .....wouldn't it be nice if the process could be completed at a lower temperature , reducing the fanny pucker factor by at least 50% overall

But alas .....no such luck !

I am going to have to repeat the experiment , with the heating that appears to be essential , to get any decent yields from the low acid content mixture . This is possibly a general rule ,
even for higher acid content mixtures , because years ago I even tried the
" cold process " with much higher acid content mixtures and had poor yields anyway even with the higher acid ratios ,
never getting any decent yields of RDX until I tried heating the mixture and discovered it was the heat which did the trick for the reaction . The optimum temperature and holding time may be quite specific to a particular acid content reaction mixture . Just guessing , about 55 C for 5 - 10 minutes seems like a reasonable starting point for this mixture , with a subsequent experiment
at 15 - 20 minutes for comparison .

The actual yields from two experiments will not be known until the samples are completed drying . I will provide the dismal numbers when I have them ,
but from what I see it is probably on the order of a 35% yield using the ammonium sulfate cold process , and maybe 45% using a parallel warm process . All I have confirmed by the two experiments is that
warmer is better for a low acid content mixture , both during the addition of the
HDN and for the raising of the temperature even higher afterwards .
I also am finished with ammonium sulfate which seems to be a bust , and will try ammonium nitrate as the ammonium salt for future experiments with low acid content nitrolysis mixtures .

It appears to me that the addition of HDN is better at a rate of addition regulated 40 C than at 20 C , and that after a few minutes past the end of the addition ,
raising to 60 C for about 10 minutes is
about right . There are visual markers for the reaction when it is actively proceeding , effervescence during the additions with a temperature rise , which soon subsides and has to be maintained by subsequent additions . And after the temperature is ramped up beyond 50 C ,
a second stage of effervescence occurs
with the decomposition of byproducts and/or completion of the earlier reactions . This second stage of the reaction is mildly exothermic and its completion will be marked by a sudden dip in temperature by 4 - 5 C from the
stable temperature of 60 - 63 where it was actively effervescing . Whether it is
best to quench the mixture right then when the dip in temperature is noted ,
or whether it is best to drain the hot water bath and allow a slow cooldown before drowning the mixture , is a matter of which I am not sure .

For the warm process I mixed 1 gram of urea nitrate with the HDN , and I also mixed another 1 gram of UN with the acid
and the ammonium salt .

I did notice that the quenched reaction mixture for the warm process reeked of
formaldehyde , but the mixture from the cold process was nearly odorless .

Also I am thinking that microteks proportions are likely about the limit in terms of efficiency of acid utilization , and if that method using just HDN and HNO3
does give 74% yield reliably , then it is
an optimized procedure which will be difficult to improve . What I am seeing
is making me doubtful of the adaptation of the patents improved yield methods for higher acid content mixtires to have similar value in improvemnet of yields for lower acid content mixtures . But I will reserve conclusions in that matter until after tests using NH4NO3 as the ammonium salt .

I am also contemplating using even higher temperatures for finishing the reaction , 75 C perhaps , but I am unsure
if I will do this . I absolutely cringe at the prospect of heating such a mixture which seems to be heckling me and daring me .... as if it was quietly chanting

" Sprengel rules ! "

as the mercury is steadily rising

I know that industrially they use(d) efficient cooling during this stage (destruction of linear nitramines) to control the exotherm. Wonder what would happen without cooling, and at what temp. My gut feeling tells me that with most of the hex already reacted, and the non-reacted 20-30% partly decomposed, there would not be enough fuel for an *explosion* (ala Sprengel HE). I could imagine a runaway decomposition, but less violent than for NG. Reason: RDX is much more thermally stable, and the mix without SA will boil off earlier, limiting the temp.

Still nothing for my kitchen... Someone else wanting to try it deliberately?

You got it....I am going to go over the whole of the experiment and perhaps alter temp only (perhaps just 5-10 degrees). And in addition I am going to use the MicroTek concept. here. .
I would be really pissed if I wasted a 2-3hour distilltion. But it does seem too intriquing to not give it a try.

W Process

"Cold process produced 15.5 grams
Warm process produced 19.0 grams"

That equals 34% and 41% respectively, putting it in the range of my *worst* yields from 50g/100ml without ammonia salt addition.

This leaves the most important question open: What yield would have resulted from 55g/100ml *without* the salt addition? And secondly, what quality was the acid? For anything but home-brew, non-vacuum nitric it 's a pity. And a proof for your over-proportional yield-reduction caused by too much HDN.

Would you mind to try 50g and 45g with *the same* acid? And of course 55g without the ammonium sulphate, and everything at different temps please ... we keep you busy!

When next weekend I distill my next nitric, I will do my share of tests. What I need is at least your acid density (and colour) for my results to be of any value. I might also try the pyrosulphate, it's one of the things we have in multi-kilo bags in the company, unlike nitric

I finally distilled some nitric, so here come the first results:

1. In 60ml freshly distilled nitric (d=1.53 at 0C) were dissolved 5% AN and 2.5% urea nitrate by weight. 25g HDN was added in 5 portions at 15-25C
(ice bath when needed). Mix was let stand at RT for 15 min, then heated to 55C within 3min, held there for 12.5 min and poured in 800ml ice water.
Neutralised in the filter with some 1M soda sln. then more water until neutral to litmus. Dried yield was 17.1g = 82%. HURAY

2. As above with 50ml nitric. Gave 15.4g or 74%.

3. With 50ml but without AN + UN. Nearly dry this morning, looks like at least 15g again. (Little time and temp variations unfortunately, I'll tell you tomorrow). ???

More coming....

[Edited on 13-12-2005 by Boomer]

Quote:

Originally posted by Boomer I finally distilled some nitric, so here come the first results: 1. In 60ml freshly distilled nitric (d=1.53 at 0C) were dissolved 5% AN and 2.5% urea nitrate by weight. 25g HDN was added in 5 portions at 15-25C (ice bath when needed). Mix was let stand at RT for 15 min, then heated to 55C within 3min, held there for 12.5 min and poured in 800ml ice water. Neutralised in the filter with some 1M soda sln. then more water until neutral to litmus. Dried yield was 17.1g = 82%. HURAY 2. As above with 50ml nitric. Gave 15.4g or 74%.

Good experiment , breaking 80% with an economic
HDN to HNO3 ratio is exactly the idea . Looks like
a good step towards a general method if it is
consistent .

Those results are tracking nicely with what I was guessing about the " sweet spot " ratio for HDN to HNO3 being something in the area of 40 to 42 grams of HDN per 100 ml of ~ 97% HNO3 . Your experiment #1 represents 41.66 grams HDN per 100 ml HNO3 . To allow a slight excess of
HNO3 and thereby provide a bit of insurance against the occasional poor yield from a reaction that is sensitive to acid quality , or other variables , I will probably settle for 40 grams
of HDN per 100 ml of HNO3 . To use microteks 50 grams of HDN per 100 ml HNO3 instead , would result in almost the same amount of RDX , but the quality requirement for the acid is more stringent and it uses more HDN than Boomers proportions in example #1 . Boomers proportions would likely give more predictable results in the presence of variables and even better predictability would be likely settling on 40 g. HDN per 100 ml HNO3 as a rule of thumb proportion providing a bit of headroom for reactant and reaction condition variables .

The raw number yield figures for RDX should be weighed as being only part of the story concerning the efficiency of the reaction where acid utilization is really the more significant
consideration for a lab scale synthesis , and the other factor
more important is reproducibility , getting consistent and predictable results . Increasing the excess of HNO3 favors
predictability for the reaction .

Experimenting with the effects of NH4NO3 and urea nitrate
makes sense for the likely increase in yields and the safety of the reaction . Another possibile candidate salt which may
have usefulness added to the acid is anhydrous magnesium sulfate , from baking dry epsom salt . This does react with
NH4NO3 to form a deliquescent double salt or mixture of ammonium sulfate and magnesium nitrate , which has dehydrating properties . It could possibly function as a dehydrating agent in the nitrolysis mixture . This might benefit the yields if it did not have some other negative effect on the reaction .

I could kick my own balls for letting the reaction conditions go astray
Yield of test No3 was 16.1g btw, which is 77%. Now does that mean that the addition of AN/UN did *decrease* yield? Or was it my 'fault' to let it react 4 min longer, and let it go up to 61C max instead of 55C, which turned out to be beneficial? Were those 10ml more nitric wasted in test No1 (only 1g more RDX for 15g more acid)? I am really angry at me! All I can do now is give more details:

- The reaction time after the last addition was chosen as 15 minutes for a reason in all three tests. This was the time an exotherm could be seen (rising from 15C to 26C and staying there in a 16C room), before it cooled down.

- In the third test, the heating time was 17 min instead of 12.5. Also the temp rose higher (61C instead of 55 like planned. The first time I had the thermometer in the water bath (55C), the second time I swapped it between the bath and the mix to find the mix approx 3C above the bath. It is a safe guess the bath was also at 58C in the first test. The last time the bath went to 58C, hence the high reaction temp. These deviations were caused by stupidly doing something else in parallel….

- The acid used must have been of better quality than usual. I recently went to distilling in three increments: From 400ml 96% SA and 400g AN the first 10ml are distilled off, receiver changed for a yield of 160ml of ‘the good stuff’, then 150g more An added and the old receiver put on again to use up all nitrate. Only the middle fraction was used for the RDX tests. No vacuum btw and 1-2 drops per second.

- Plus, in my older experiments I assume the HDN was not too good. In fact it had been stored at RT for some month and smelled of formaline. This, together with the better acid, must have been the reason for the good yields.

Unfortunately I am out of nitric again (60+50+50ml = 160ml duh). While I am happy with the results, they leave many questions open: Why no effect of the AN/UN? Would the first yield have been even better without? Why did 10ml more acid give only 1g more product? …

[Edited on 14-12-2005 by Boomer]

[Edited on 14-12-2005 by Boomer]

Quote:

Originally posted by Boomer I could kick my own balls for letting the reaction conditions go astray Yield of test No3 was 16.1g btw, which is 77%. Now does that mean that the addition of AN/UN did *decrease* yield?

Not really wasted in my opinion anyway . You are looking at absolute acid utilization efficiency as the goal with maximum yield of RDX in grams per ml of HNO3 , regardless of the conversion efficiency for the HDN and the reproducibility
which are also important factors . This sort of nitpicking
over the absolute acid utiliztion efficiency will likely never lead to a good efficiency good general method which is forgiving of small variables , and always gives acceptably good yields but at the slight expense of falling slightly short of maximum yields . It is something analogous to fine tuning a racing car to get an extra 10 miles an hour top speed above the 200 miles per hour it can do reliably all day without any problem , but the expense for that extra 10 is
that the engine might disintegrate on the first lap at the increased speed if the oil temperature is 5 degrees warmer .
The logic which applies to the analysis will determine which
implementation is superior for accomplishing a hundred laps reliably at 200 , or having bragging rights for one lap once in awhile at 210

Even better , don't be so picky about the acid quality , and
try an even more conservative 2.5:1 ratio for 3 experiments modeled on example #1 to see what is the reproducibility
and if yields will go even higher than 82% to maybe 88% based on HDN . If such an outcome occurs , then you will
pretty well have tagged the conditions for a general method .
I'll buy generalization and reproducibility with acceptably good yields at the expense of very little in terms of absolute
acid utilization efficiency , as a good bargain any day

You see that generalization will give you a " general recipe " that is guaranteed to work well every time , whereas nitpicking the conditions for absolute acid utilization efficiency will only identify an exceptional reaction condition where a maximum yield is obtained for very precise conditions that will likely be different by some variable that will have bearing when you try to repeat your own " maxed out " and " too refined " procedure . The next batch of distilled acid being different by even a fraction of a per cent ,
may be sufficient variable by itself to cause very different
results from what was carefully worked out before . But
a good general method is able to accommodate such variables without being adversely affected .

another possible nitrolysis catalyst / acid dehydration reagent

The wheels have been turning in old Rosco's head ,
wondering what value that bottle of Magnesium Oxide
sitting on the shelf in the health food store might be for
an off label use as a nitrolysis or nitration " enhancer "

And by golly it just could be a damn useful component
for reducing the water content in various processes
due to the high affinity for water of the nitrate of magnesium
which would form , at the expense of some of the acid
of course for forming that nitrate ......and I haven't run
the numbers on it yet , but it sure would seem to have promise of a net gain in that regard .

And it should be completely compatable with other additives
like urea nitrate or ammonium nitrate in the nitrolysis mixture
for conversion of HDN to RDX . This use of MgO added to
the acid mixture could also have value in many other nitration
mixtures where more radical dehydrating reagents have been found to be useful .

MgO + 2HNO3 ------> [ Mg(NO3)2 - 1 H2O ] mol.wt. 166.05

Added MgO will easily absorb 90% of its own weight of additional water beyond the monohydrate formation to form the trihydrate , and 180% of its weight to the pentahydrate .

MgO mol.wt. = 40.03
HNO3 mol.wt. = 63.01 ( X2 )

See the attached patent . In the patent process is
described the use of separately made Magnesium Nitrate
Trihydrate as a dehydrating agent ......and I am suggesting
even going beyond that and enhancing the dessicant effect
further by forming the nitrate in situ from the oxide .

If it is wished to make the magnesium monohydrate separately , so as to avoid any consumption of acid in its formation , simply long boiling of MgO with ammonium nitrate
to a cessation of evolution of ammonia should result in
a solution of Mg(NO3)2 . Heating this to dryness and then
to 330C produces the monohydrate .

Alternately , MgCO3 may be substituted in that boiling for
MgO . And if no source of MgO or MgCO3 is convenient ,
the MgCO3 can be precipitated from mixed hot solutions
of epsom salt and washing soda . So there are different
ways of obtaining / utilizing the Mg(NO3)2 monohydrate .

Another possible simple route to Mg(NO3)2 would be to
boil a mixture of Ca(OH)2 with mixed solutions of epsom salt and ammonium nitrate , until evolution of ammonia ceased ,
filter out the precipitated CaSO4 and concentrate the residual
solution of Mg(NO3)2 followed by dehydration by baking out the solids at 330C . But this strategy could be complicated by formation of double salts , so I am not certain this would work , but I think it likely would work .

[Edited on 15-1-2007 by Rosco Bodine]

Attachment: US5012019 Magnesium Oxide as Dehydrating Agent for HNO3.pdf (234kB)
This file has been downloaded 1514 times

I'm there - ....just as has been expressed before; my head (also) turns when seeing a flash of flesh from a promising looking girl and esoteric routes always intregue me as well.
The links were down to the posted material dealing w/ Ketone N so I couldn't read them but what I wonder is what is being done today: if the most efficient methods are not common knowlege re: RDX on lab scale (not industrial). I just can't see modern researchers still limited to mid-scale nitrations, etc. but I've been dead wrong before. Perhaps RDX is only produced on such a large scale that there has never been any need to move away from the simplier / direct methods of RDX production(?)...

This idea of using Mg(NO3)2 monohydrate as a nitrolysis and/or nitration acid dehydrant ,
is very intriguing to me , more than any idea I have ever considered as having possible value in nitrations actually . It would be a totally OTC available route to providing a dehydrating reagent which just might be an acceptable substitute for much more exotic and expensive reagents
used for the same purpose ....like phosphorous pentoxide , or acetic anhydride , for dehydration of nitrolysis mixtures ....or in the place of oleum in nitration
mixtures . And of course there are also possibilities there for usefulness in the preparation of nitric acid
of high concentration .....prior to use in any nitration
where additional Mg(NO3)2 monohydrate could have further usefulness .

Having HNO3 of extremely high concentration at the beginning and maintaining the acid at a near to anhydrous condition during the nitrolysis or nitration is
a very important aspect of creating ideal conditions
with regards to the efficiency of the synthesis particularly
for RDX ....where even a 2 or 3 percent change in the
water content of the nitric acid can affect the yield by
twenty times that amount . So anything that gives even
a little improvement to the quality of the acid can have
a drastic influence on the efficiency of the reaction .

This could also apply as well to other nitrations which are
very sensitive to the water content of the acid or mixed acids ....where having anything present which can efficiently tie up indigenous or reaction byproduct water
will be likely to improve the yield considerably .

This Mg(NO3)2 idea is moving to the top of my
" must do experiments list " because of its potential
value in virtually any nitrolysis or nitration reactions .

There is not good information easy to find about the
different temperatures and hydration states for
Mg(NO3)2 , melting points for the various hydrates ,
and what is the decomposition temperature for the
monohydrate ....which I suppose is the lowest stable
hydrate . PATR references conversion of the hexahydrate
to the monohydrate at 330C , but I think
that is wrong . They are probably referencing the
trihydrate or possibly a dihydrate , as I would expect
a stepwise sort of dehydration passing through a series of distinct hydrates lower than the hexahydrate . So anyone having any extensive data on the hydration states for Mg(NO3)2 , please share a link or post the data here .

I found one reference in a patent that the hexahydrate of
magnesium nitrate melts in its own water of hydration at 90C .....which is a fairly high melting point for a hexahydrate and would account for a very absorptive and dehydrating activity towards formation of the hexahydrate at temperatures much lower than that 90C transition point ,
where a much lower than six hydrate is in the reaction system .

This is much higher activity than for example the tetrahydrate
of calcium nitrate which melts at ~42C .

If anybody has the book on Mg(NO3)2 ....I want it


[Edited on 16-1-2007 by Rosco Bodine]

linkage with more recent thread

E-Process failures