US2798870A - Method for preparing explosives - Google Patents

Description

July 9, 1957 PERCENT YIELD OF CYCLONITE o 5 l 8 BACHMANN 2,798,870

METHOD FOR PREPARING EXPLOSIVES Filed July 16, 1943 REACTION TEMPERATURE -(DE.GREES CENTIGRADE) Warner E E2 achmann United States METHOD FOR PREPARENG EXPLQSIVES WernerEnBachmann,Ann Arbor, Mich, assignor to the United Statesof America as'representad by the taryof War Application July 16, 1943, erial No. 495,078

11 Claims. (Cl. 260-448) O2NN N 1\|IO2 and therefore, on-thisassurnption, its systematic chemicalname is cyclotrimethylenetrinitramine. It has also been variously called Hexonite, Hexogen, T4, C6, andcyclonite. As a matter of convenience, however, theterm Cyclonite will hereinafter be used .to'designate this material.

The explosive in question appears to have been discovered by .Henning (D. R.P. 104,280,.June 14, 1899), who prepared it by converting hexametnylenetetramine (hereinafter called "hexamine), to the-corresponding dinitrate an'd then a'ddingthe dried dinitrate to cold, very concentrated nitricacid. 'For' a number of years little interestwas shown in the new explosive. However, when the *developmentdf syntheticmethanol opened thedoor to-explosives basedon'formaldehyde, interest in cyclonite was "renewed, and between 1919 and 1921, von Herz (DJR. P. 298,539 and 299,028; G. B. 145,791; Swiss 88,750; U. S. 1,402,693) disclosed a so-called direct nitration (nitrolysis) method of preparingcyClonitebY treating hexamine directly with cold, concentrated nitric acid, according "to the scheme indicated below and referred to herein as Scheme 1 or Equation 1:

characterized by a number of significant disadvanages.

Inithe firstplace, althoughonlyrthreemoles of nitric acid Cir are theoretically required by Equation 1, in actual practice a very large excess of nitric acid (from 20 to 25 molesper mole'of. hexamine) is required to obtain yields of 'to (See, for example, Hale, Jour. Amer. Chem. Soc., 47, 2754 (1925).) In the second place, the reaction is frequently accompanied by a violentfume-otf which renders large scale manufacture hazardous and diificult to control. Thirdly, that portionof the hexamine molecule which is represented in theprojection formula as in the middle of the ring (the so-called central part of the hexamine) represents one-half (3 moles) of the six moles of formaldehyde required for the preparation of the hexamine. This half-equivalen of formaldehyde represented by the central part appears to he completely wasted in the direct nitration process, througheventual conversion to carbon dioxideand other. reaction products. Moreover,the loss of the half-equivalent of formaldehyde is attended by a reduction of a large amount of nitric acid to N204. Accordingly, in large scale manufacture, a very large proportion of the equipment is representedbyapparatus for the recovery of nitric acid and the oxides of nitrogen, and even though this recovery apparatus may obviate serious nitrogen loss, it will be apparent that the formaldehyde which is oxidizedto'COz is irretrievably'lost. Accordingly, theloss of the halfequivalent of formaldehyde per mole of hexamine, and the enormous nitric acid recoveryproblem, constitute the main defects of the direct nitratioif process.

Despite the many disadvantages of the direct nitration processes, for over forty years the foregoing degradative procedures ofHenning and of von Herzappear to have been the only known methods available for cyclonite production. Within recent times, however, a synthetic method has been described by schiessler and Ross inULS. application Ser. No. 444,254, filed May 23, 1942. This method (which is known as the so-called M'cGill Process) involves the treatment of formaldehyde (e. g. in .the form of paraformaldehyde) with ammonium nitrate at 60 to 85 C. in the presence of a dehydrating agent (e. g., acetic anhydride) according to the overall Equation 2:

By the foregoing method, cyclonite may be obtained in yields as high as50 to 55% of the theoretical, based on the material balance indicated by the overall Equation 2.

Now, as indicated in connection with Scheme 1, in the degradative nitration process, thethree moles of formaldehyde (represented by the central part) are destroyed per mole of cyclonite produced. On the other hand, in the more recent Schiessler'and Ross synthetic method, three moles of formaldehyde are required per mole of-cyclonite produced (Scheme2).. It was accordingly of interest to determine whether the formaldehyde lost in the degradative nitration" process, could be usefully employed in the synthetic process. This proposal neglected important considerations of mechanism, since the wasted formaldehyde in the direct nitration process appears as certain formaldehyde-nitration products, and ultimately as CO2, rather than asformaldehy'de. Nevertheless, prior investigators carried out the direct nitration process in the presence of ammonia donors (e. g,, ammonium nitrate) in the hope that the wasted formaldehyde might combine with the ammonia donor to reform 'hexamine and thus .be returned to the main reaction. The results, howexer, gave little encoura'ge ment thatsuch a process would effect an appreciable improvement in yield, and indicated that the reaction prod uctsof the lost half-equivalent of formaldehydewould 3 not combine with ammonium nitrate in the presence of strong nitric acid to yield cyclonite.

The object of the present invention is to provide a new and improved method of preparing cyclonite from hexamine, whereby two moles of cyclonite are theoretically obtainable from one mole of hexamine, and which actually produces in excess of one mole of cyclonite per mole of hexamine.

A further object is to provide What may loosely be termed a combination process wherein both the degradative and synthetic processes of the prior art may be considered as taking place concurrently.

Another object is to provide a process which is capable of producing higher actual yields of cyclonite per mole of hexamine-assuming optimum conditions and yields in each case-than the aggregate yield obtainable by first carrying out the direct nitration process on one mole of hexamine, and then separately carrying out the McGill process, starting with paraformaldehyde in an amount equivalent to all the formaldehyde lost in said direct nitration process.

A further object is the provision of a process for preparing cyclonite which obviates the enormous nitric acid requirements of the direct nitration process and which requires much less acetic anhydride than the McGill process.

Still another object is a method of preparing cyclonite which requires less ammonia and formaldehyde equivalents per pound of cyclonite produced, than any previously known method.

Other objects and advantages will be apparent as the invention is hereinafter more fully described.

As a result of an extended investigation, it has been found that the foregoing objects and advantages may be attained by the process of the present invention, which involves the treatment of a hexamine compound with an ammonia donor and nitric acid under conditions effecting the elimination of the elements of water, the reaction being carried out under certain critical temperature con- 'ditions, as hereinafter explained.

The hexamine compound mentioned in the preceding paragraph may consist of hexamine itself, or a hexamine derivative such as a hexamine salt, for example, hexamine acetate, hexamine mononitrate, hexamine dinitrate and the like. The ammonia donor preferably consists of ammonium nitrate, although other ammonium salts such as the acetate and similar equivalent materials may be employed if desired. The elimination of the elements of water is elfected in the presence of a volatile fatty acid anhydride, such as a lower fatty acid anhydride, for ex ample, acetic anhydride, propionic anhydride, butyric anhydride and the like, which combines with the eliminated water.

The conditions prevailing when the foregoing reactants are brought together, exert an important influence on the course and extent of the reaction. As a matter of convenience, these conditions may be classified as (1) preferred proportions of reactants and (2) critical ranges of temperature at which the reaction may be carried out.

Considering firstly the preferred proportions of reactants, it has been found that the optimum conditions for the reaction approach those indicated by the theoretical material balance of the overall equations, which (in the case of hexamine and hexamine dinitrate) may be written as follows:

(3) Hexamine+4HNO3 2NH4NO2 6(RCO)2O=2 cyclonite+ IZRCOOH (4) Hexamine dinitrate+2HNOs+2NH4NOs+ 6 (RCO)2O=2 cyclonite-l-IZRCOOH In most cases, an excess of nitric acid, ammonia donor and dehydrating agent over the theoretical amounts required by the overall Equation 3 or 4, are usually desir- 'able. Thus when hexamine itself is employed, excellent yields of cyclonite (over 75% of the theoretical, based 4 on Equation 3, supra) may be obtained using approxi mately 2.3 to 3.2 moles of ammonium nitrate, 4.5 to 5 moles of 98% nitric acid and 6.8 to 8.0 moles of acetic anhydride per mole of hexamine. Likewise with hexamine dinitrate, excellent yields (over 80% of the theoretical, based on Equation 4, supra) may be obtained using approximately 2.3 to 3.2 moles of ammonium nitrate, 2.4 to 2.8 moles of 98% nitric acid, and 6.8 to 8 moles of acetic anhydride per mole of hexamine dinitrate.

While the foregoing proportions are preferred, satisfactory yields of cyclonite may be obtained over a wide range of operating conditions, particularly if the reaction temperature and time are varied to suit the particular proportions of reactants selected for carrying out the reaction. Thus for example, Table I gives illustrative ranges for each of the reactants, which may be used to give satisfactory yields of cyclonite when the reaction is carried out at about 6075 C. For optimum yields, particularly where it is desired to lower the consumption of reactants, the ammonium nitrate and nitric acid are preferably reduced roughly in the same molecular proportions.

TABLE I. MOLAR PROPORTIONS OF REACTANTS Hexamine Nitric Acid Ammonium Acetic Anhy- (98%) Nitrate dride 1 4 to 6 2 to 4 (or more) 6 to 8 Hexamine Nitric Acid Ammonium Acetic Anhydinitrate (98%) Nitrate dride 1 2 to 4 2 to 4 (or more) 6 t0 8 It will thus be evident that a Wide choice in the proportions of reactants is available to the operator in carrying out the process of the present invention. Moreover, nitric acid less concentrated than 98% (e. g., 70%) may be employed, provided additional anhydride is used in an amount equivalent to the Water present in the less concentrated nitric acid.

Before leaving the discussion of reactant proportions, it is deserving of emphasis that all four reactants (i. e., hexamine compound, ammonia donor, nitric acid and fatty acid anhydride) appear to be essential to the reaction. If any one of these four is omitted, substantially no cyclonite is obtained. Thus, in one experiment where ammonium nitrate was the only reactant omitted, substantially no cyclonite was produced at C., although it might be expected that at this temperature the direct nitration reaction might take place between the hexamine and the nitric acid. Such however, does not appear to be the case, except to a very limited extent (e. g., about 2%). Accordingly, all four reactants are considered necessary to the production of cyclonite by the process of the present invention.

Referring now to the critical temperature for the prep aration of cyclonite, there is, in general, an unusually narrow minimum temperature range for each specific set of reaction conditions (e. g., proportions of reactants, time of reaction and details of procedure). If the proportions of reactants or the time of treatment or the details of procedure, are changed, this narrow minimum temperature for cyclonite production may shift somewhat. Generally speaking, however, the minimum temperature range for appreciable cyclonite yields will be found to lie somewhere above approximately 35 C. to 40 C.

The foregoing general principles may best be illustrated by reference to the single figure of the appended drawing which is a graphical representation of the effect of time and temperature changes on cyclonite yiel d when the reaction is carried out in all cases with the same proportions of reactants and the same general procedure. Curve S (solid line) shows the effect of temperature on yields of cyclonite when the reaction time is relatively short (12 /2 minutes). Curve L (broken line) shows the temperature etfectzwhen the reaction time-is-relatively long} (45 minutes). Ineach case thereactantswerei used. in the following proportions: hexamine, 33.6 g.; nitric. acid (98%), 75 g.; ammoniumnitrate, 55 g.; and acetic anhydride, 180 cc. Except for the variations. indicated, the procedure in all cases wassubstantially otherwise identicalv with that described hereinafter in. connection with Example IV.

By inspection of the two curves shown inthedrawing. it will beevident .thatthe effect of temperature on the yield isiunusually pronounced inbothminstances. Moreover, in the case of the relatively short reaction time (curve. S) the minimum temperature appears to lie above 40?C.,. whereas forthe relativelylong reaction time, thenriuimum lies above about 35 C.

It should be noted that the. temperature-yield relationship shown in theJdraWing. applies, strictly speaking, only to the-particular proportions of reactants (mentioned above) which were maintained in obtaining the data for thegraph.v If the proportions of reactants, or the time of treatment, or both are modified, the curve maybe somewhat displaced to the leftor right on the horizontal 'axis, and theshape of the curve may also be alteredi Thus an increaseinthe reaction time tends to raise thewyields at the lower reaction temperature. However, the -graph* gives .a. general picture of the pronounced eifect of-temperature on the yield for any specific reaction condition. Moreover, ittappears generally to be true that the mini-- mum critical temperature for cyclonite production lies" somewhere :above about 35 to 40 C.

If the reaction is carried out below this minimum tem-- perature range (forexample, at -35 C.) theprincipal reaction'productisusually a substance melting at about 154 C. which appears to have at leastlone acyl" group inthe molecule. At temperatures somewhat above the minimunrrange given above, cyclonite will'be obtained in varyingyields depending'on the time of reaction and proportions of reactants used. Thecyclonite-so obtained may contain the above low-melting impurity to avarying but usually small degree, depending on the precise conditions obtained during the reaction. This impurity, however, is readily removed by digesting the unneutralized reaction mixture with water at 90100 C. for 1 /2 to 12 hours, which converts the low melting compound to'water soluble and volatile products without, however, detrimentally-altectingthe cyclonite. If the preparative reaetion is carried out at temperatures 15 1030" above tov C. the amount of low-melting impurity is consider ably decreased, while the yield of cyclonite improves. Thus the optimum temperature range for cyclonite production appears to be between to 90 C., and for commercial' operations to C. willusually betound 'to' be most satisfactory.

"in regard toa possible ceiling. reaction temperature forcyclonite preparation, as the temperature increases, theyield and quality of the crude cyclonite may-fall. offsomewhatat temperatures above about C. Cyclonite has, however, been obtainedabove this temperature (e. g., C.). The actual maximum-temperature at which cyclonite is to be produced should be determined withreference to safety considerations, and from this point of view onehesitates to run the reaction above C.

In: addition to the low-melting, readily removable byproduct mentioned above, the cyclonite produced by the process of the present invention may also contain. a highmelting compound (M. P. about279-280 C.) in varying amounts (up to 10%, forexample), depending'on the proportions and conditions employed. This compound, however, appears to be an explosive substance quite'similarto cyclonite in explosive power and brisance, and is be-' lievedto be ahomologue of cyclonite (i. e., cyclotetra methylenetetranitramine) fOt'lWhlCh reason it may conveniently be. designatedflby thev trivial term, homo-' cyclonite. Regardless of its structure, the presence-ofi homocyclonite. in the. cyclonite. produced. byrthe process: ofthe present invention. is not objectionable fromithepoint: of :view of ballistic power and .brisanceofthe productand: therefore homocyclonite is not to be regarded as 11111-3 desirable; impurity? in the cyclonite produced.=-by- 'thepresent process. For this reason, .conditionswhidhfavorthe. formation of relatively high proportions ofhomocyclonite. (e. g., up to. 10%) as a by-produc in cyclonite production .are not objectionable in respect to the? explosive power of the product. These homocyclonitefavoring conditions appear to include highanhydrideand. low nitric acid proportions.-

Where preferred; additional materials may be employed; for example, as reaction mediaor as solvents: for:one:or: more of the reactants. Thus, the mother liquor'from; a previous run may be used as a reaction media; or the -hexamine. may be: dissolved in a suitable inert solvent, for example, .a substantially; anhydrous lower fatty acidlsuch' as.glaci'al acetie. acid. Moreover, .bytheuseof a hexamine': SOllltiOHiHJCOnjll'IlCtlOH with a suitable combinationof" theother-reactants; a procedure may be'employed wherein tall .the reactants are. charged totheareaction vessel. in: liquid/form.. Thus the hexamine may be dissolved in: acetic acidaand'thei ammonium nitrate in the-'nitric-acid; withvacetie'anhydride. as the third in liquid, to provide: ans allliquidafeedf modification within the ambit of 1 the: present-invention Thewdetails'of this all-liquid-feed? modification are .moreparticularly described and. claimed in the copendingapplieati'on of George F. Wright, Douglas: C. Downing and Henry-H. Richmond, Serial No'.495,082,-. filed July 16,- 1943, issued as;Patent No. 2,434,879; Ian-1' uary; 20,1948;

In order more clearly to disclose: thevnature ofrthen present invention,, several illustrative embodiments: :will

hereinafterbe described in considerabledetail.

clearly be understood, however, that this is done solely byway ofexample, and-is-notto be construed as inany' manner limiting the spirit or scope of the appended claims;

Example I (SOLID AND LIQUID FEEDS, USING HEXAMLNE- DINITRATE) efficient mechanical stirrer, one for athermometentwm for burettes, and a large. flanged one for the addition of solid). The flask was placedin a water bath not much larger irr diam'eter than itself. Thisbath was equipped with a thermometer andcoul'd be heated by steam and cooled =by running water.

Three operators are desirable to regulate the" addition. and control the" temperature for the: reaction; Sixty-fives grams-ofhexamine .dinitratewas divided into 26 substarrtially equal portions (each portion was weighed into atesti tube forconvenience. in handling); cc. of aceticanhydride wasplaced in a burette calibrated; in 26. sub stantially equal portions; and 26 cc. of 98% nitric.acicl was placed in another burette calibrated in 28 substan-: tially equal portions.

At the beginning of the run, 50 g. of dry'ammonium nitrate-(60420 mesh) and 30 cc. of glacial'aceticacid.

wereplaced inthe flask. The temperature was raised to 75 C. and 30 cc. of acetic anhydride was addedflhis is The- 30 co. in addition to the 145 cc. in the burette). resulting mixture was quite fluid andeasily stirred. Eflicient agitation is essential for optimum yields.

A-quarter portion of hexarnine dinitrate-was added'lto the mixture in the flask. This serves to inhibittheviolent'l reaction between the acetic anhydride and thenitrc acid..

Two portions of nitric acid were then slowly added. The

third portion of nitric acid was then added simultaneously.

It should t er-portion for quarter-portion) over a period of about twelve-minutes. The temperature withinthe flask was maintained at 74-76 C. throughout the addition. The first operator added the nitric acid and watched carefully the inside and outside temperatures, regulating the steam and cooling water. The second operator added the acetic anhydride to correspond to the addition of the nitric acid and, watching the nitric acid burette, called off the addition in quarter-portions (i. e., one-quarter, one-half, threequarters, all added). The third operator added the solid (hexaminedinitrate) to correspond to the addition of the nitric acid; he added a quarter of the contents of a test tube at a time. The addition of the liquids was continuous, and that of the solids, effectively so (i. e., in 100 quarterportions).

After the addition was complete, the mixture was stirred for minutes at the same temperature. The water bath was then removed and the mixture was allowed to cool to 60 C. (about 10 min. were required). I It was then filtered through a warm, coarse Jena funnel. The first-crop product was washed with 30 cc. of cold glacial acetic acid, and then with two 50 cc. portions of hot water. The air-dried product was granular and of good appearance; M. P. l203 C. (corn) with previous softening; yield, 81.1 g. or 75% of the theoretical, based on Equation 4, supra, representing an actual yield of about 1.5 moles of firstcrop, high grade cyclonite per mole of hexamine dinitrate. A second crop of less pure cyclonite may be obtained by drowning the filtrate in an equal volume of water, then digesting the mixture for one hour or more at about 95 C. and cooling. In this manner, from 8 to 10 grams of less gpure cyclonite, melting at about 190-200 C., may be recovered from the filtrate.

desired to reduce the number of liquid feeds, the nitric acid may be mixed with the acetic anhydride. In this case however, the nitric-acid-acetic anhydride mixture should be prepared and used at low temperatures (e. g., 5-15 C.) as soon as prepared.

Example [I (SOLID AND LIQUID FEEDS, USING HEXAMINE INSTEAD OF THE DINITRATE) The apparatus was the same as that described above. 33.6 g. of hexamine, 43 cc. of 98% nitric acid and 160 cc. of acetic anhydride were used. Addition of the reagents required about 15 minutes. The temperature was maintained at 73-77 C. during the reaction. Before the other reagents were added, 50 g. ammonium nitrate was placed in the flask together with cc. of the cool, clear mother liquor from a previous run. The procedure was similar to that used in Example I. After portion number 5 of the other reactants was added, 25 g. of additional ammonium nitrate was added to the reaction mixture. In a typical run, a yield of 73% of the theoretical (based on Equation 3, supra) was obtained; M. P. (corn) 197-202 C. This represents an actual yield of approximately 1.46 moles of first-crop cyclonite per mole of hexamine employed.v A second-crop may be obtained from the filtrate by the procedure described in Example I. As in Example I, the second-crop cyclonite is less pure than the first-crop and may contain a relatively high proportion (e. g., 10 to 20%) of homocyclonite.

Example III (ALL-LIQUID FEED TO SUSPENSION OF AMMONIUM NITRATE IN A020, USING HEXAMINE INSTEAD OF THE DINITRATE In a 1-liter five-necked flask equipped with a mechanical stirrer, a thermometer and three burettes, are placed g.) of 98% 65 g. of ammonium nitrate (granular or finely divided) dried at 50 C. for 18 hours, and 60 cc. of acetic anhydride. The mixture is stirred and'heated in a water bath 75 C. (internal temperature). To the hot mixture are added separately and concurrently, from the three burettes (Note 1, infra), cc. of 95 acetic anhydride, 43 cc. (64.5 g.) of 98% nitric acid, and a solution of 33.6 g. of hexamine in 53 cc. of glacial acetic acid (Note 2, infra), during a period of about fifteen minutes.

The temperature is held at 7476 C. during additions. The reaction mixture is heated at 74-76 C. with stirring for five to fifteen minutes after the reactants have been introduced. 'i

Without cooling, the mixture is diluted to a concentration of about 30% acetic acid by the addition of 675 cc. of warm water (60 C.) and then heated on the steam bath (90-100 C.) with stirring for one to twenty-four hours. This digestion step destroys the low-melting impurities, as more particularlydescribed and claimed in my copending application, Serial No. 495,079, filed of even date herewith, now Patent No. 2,680,651.

The mixture is cooled to 20 C., with constant stirring, and filtered immediately by suction. The white, glistening, sandy single-crop crystals are washed with four 50 cc. portions of warm water (60 C.) and oven dried at 50 C. The yield is 76-79 g, or 71-74% of the theoretical based on Equation 3, supra, representing about 1.4 moles of cyclonite per mole of hexamine; M. P. 196- 197 C. (corn) with previous softening at 191 C. The main filtrate (but not the washings) is reserved for recovery operations.

In some instances better yields may be obtained with the above procedure if slight variations are made; for example, by adding some acetic acid and a small amount of the nitric acid to the acetic anhydride in the flask at the start of the reaction. Another alternative is to maintain the above described reaction mixture at 54-58 C., rather than at 75 C., with no other changes. The yields obtained with either of these modifications run as high as 82% of the theoretical, based on Equation 3, supra.

Note 1.-In order to secure a uniform introduction of of the reactants in equivalent amounts, the three burettes were marked with calibrations corresponding to 25 cc. portions of each reactant.

Note 2.This solution has a volume of about 75 cc. Since the solution is quite viscous and does not drain readily in the buette, it is convenient to pepare a small excess of the solution and determine the amount actually introduced by dilference in weight.

Example IV (ALL-LIQUID-F'EEDS, USING HEXAMINE INSTEAD OF THE DINITRATE) A yield of 70-90% cyclonite may be obtained by a modification of the procedure described in Example III, which permits the use of alHiquid-feeds. In this modification, the first liquid feed consists of the ammonium nitrate dissolved in the nitric acid with the formation of a clear, water-white solution (after the evolution of oxides of nitrogen). The second liquid feed consists of the hexamine dissolved in glacial acetic acid. The acetic anhydride constitutes the last liquid feed.

Thus, a solution of 55 g. ammonium nitrate in 50 cc. nitric acid was prepared and placed in one burette. In another burette was placed cc. of acetic anhydride while in a third burette was placed a solution of 33.6 g. of hexamine dissolved in 55 g. of acetic acid, 6 cc. of the ammonium nitrate-nitric acid solution in the one burette was added to 30 cc. of acetic acid and 30 cc. acetic anhydride (this amount is in addition to the 150 cc.) in the reaction flask. Thereafter the remaining three liquids in the three burettes were all added concurrently and equivalently. The reaction conditions and procedure were otherwise similar to those described in Example, IIL' 9 Certain aspects of the all-liquid-feed modifications of the present invention are more particularly described and claimed in the aforesaid application of George P. Wright, Douglas C. Downing and Henry H. Richmond, Serial No. 495,082, filed July 16, 1943, issued as Patent No. 2,434,879, January 20, 1948.

PURIFICATION The cyclonite produced in accordance with the present invention may be purified, if desired, either by recrystallization from any convenient cyclonite-solvent, (for example, 70% nitric acid, 50% acetic acid, dioxane, monobutyl ether of ethylene glycol, nitromethane, acetone, and the like); or by digestion with dilute nitric acid in an amount insufficient completely to dissolve the crude material. A particularly advantageous method of etfecting the first-mentioned method of purification (i. e., recrystallization) is more particularly described and claimed in the copending application of Johnson, Blomquist and McCrone, Serial No. 495,081, both of which are filed on even date herewith.

The second-mentioned method of purification (i. e., digestion with dilute nitric acid) is illustrated in the following example.

Example V (PURIFICATION) The first-crop material of Example I (melting point 20l203 C. corr.) was stirred with 55% nitric acid at 70 C. for one-half hour, 25 cc. of nitric acid being used for every g. of crude cyclonite. During the digestion, some gas was evolved as the nitric acid attacked and destroyed oxidizable impurities. The digested mixture was then cooled and the solid filtered. A 96% recovery of high grade cyclonite, melting at 205-206 C., with no softening below 200 C., was obtained. The second-crop material from Example I and II (which usually contains considerably more homocyclonite than first-crop material) may also be purified by digestion in a similar manner but with a somewhat longer period of digestion. Likewise the single-crop product obtained in Examples III and IV (which is intermediate in purity between the first and second-crop materials of Examples I and II) may be purified by the same general procedure. If the digestion is carried out for several hours (e. g., 6-7 hours) the crystals increase in particle size.

If desired, the filtrate from the digestion procedure may be used to digest further quantities of crude product. In some instances it may also be desirable to dissolve the crude cyclonite in 70% nitric acid and then to dilute the solution to 55% nitric acid strength (thereby partly precipitating the material), before carrying out the diges tion. Regardless of the details employed, however, the digestion procedure may be used to efiect a substantial improvement in the size and quality of the cyclonite crystals (where the starting material is inferior in this respect), and also to destroy many of the oxidizable impurities which may be present in the material prior to treatment. If it is desired to lower the acidity of the digested material, the product may be ground and di gested with a large volume of boiling water, or the aciddigested product may be purified by recrystallization, for example, in accordance with the procedure set forth in the above-identified copending applications.

Another method of purification which may be used, if desired, is to digest the crude cyclonite with hot water containing either a small amount (e. g., 25%) of acetone, or a small amount (e. g., 1%) of a buffering agent such as sodium tetraborate or sodium perborate. Thus the crude cyclonite may be poached at 75 C. for 3 to 6 hours with water containing 1% perborate. The thus purified material generally exhibits an increased frothing point and melting point, and is neutral to methyl red.

It will be apparent to those skilled in the art that the present invention has been described in great detail in the foregoing specification and also that many variations may readily be made without departing from the spirit and scope of the invention. I therefore intend to be limited only in accordance with the following patent claims.

I claim:

1. A process of producing an explosive, which comprises reacting a hexamine compound selected from the group consisting of hexamine itself and henamine salts, with an ammonia-yielding ammonium salt, concentrated nitric acid and a lower fatty acid anhydride, the reaction being carried out at a temperature above about 35 to 45 C.

2. A process of preparing cyclonite which comprises reacting a hexamine compound selected from the group consisting of hexamine itself and hexamine dinitrate, with ammonium nitrate and concentrated nitric acid in the presence of a fatty acid anhydride, said reactants being used approximately in the proportions of two to four moles of ammonium nitrate, four to six moles of con centrated nitric acid and six to eight equivalents of anhydride per mole of hexamine compound in the case of hexamine itself, and approximately in the proportions of two to four moles of ammonium nitrate, two to four moles of concentrated nitric acid and six to eight equivalents of anhydride per mole of hexamine compound in the case of hexamine dinitrate.

3. The process of claim 2 wherein said anhydride comprises acetic anhydride.

4. The process of claim 2 wherein said reaction is carried out at a temperature above about 3545 C. and below about 100 C.

5. A process of preparing cyclonite which comprises treating hexamine with ammonium nitrate and concentrated nitric acid in the presence of acetic anhydride, approximately in the proportion of 2.3 to 3.2 moles of ammonium nitrate, 4.5 to 5 moles of concentrated nitric acid and 6.8 to 8 moles of acetic anhydride per mole of hexamine, said reaction being carried out at a temperature suflicient to effect the formation of more than one mole of cyclonite per mole of hexamine.

6. The process of claim 5 wherein said reaction is carried out at a temperature above about 35 to 45 C. and below about 100 C.

7. The process of claim 5 wherein said reaction is carried out at 6075 C.

8. The process of preparing cyclonite which comprises reacting hexarnine dinitrate with ammonium nitrate and concentnated nitric acid in the presence of a lower fatty acid anhydride, approximately in the proportions of 2.3 to 3.2 moles of ammonium nitrate, 2.4 to 2.8 moles of concentrated nitric acid and 6.8 to 8 moles of anhydride per mole of hexamine dinitrate, said reaction being carried out at a temperature sufficient to effect the formation of more than one mole of cyclonite per mole of hexamine dinitrate.

9. The process of claim 8 wherein said reaction is carried out at a temperature above about 45 C. and below about 100 C.

10. The process of claim 8 wherein said reaction is carried out at 60-75 C.

11. A process of preparing cyclonite which comprises reacting hexamine with ammonium nitrate and concentrated nitric acid in the presence of acetic anhydride, approximately in the proportions of 2.3 to 3.2 moles of ammonium nitrate, 4.5 to 5 moles of concentrated nitric acid, and 6.8 to 8 moles of acetic anhydride per mole of hexamine, the reaction being carried out at 60 to C.

References Cited in the file of this patent Bachmann et al.: Journal Amer. Chem. Soc., vol. 71 (1949), pp. 1842-1845.

Claims (1)

1. A PROCESS OF PRODUCING AN EXPLOSIVE, WHICH COMPRISES REACTING A HEXAMINE COMPOUND SELECTED FROM THE GROUP CONSISTING OF HEXAMINE ITSELF AND HEXAMINE SALTS, WITH AN AMMONIA-YIELDING AMMONIUM SALT, CONCENTRATED NITRIC ACID AND A LOWER FATTY ACID ANHYDRIDE, THE REACTION BEING CARRIED OUT AT A TEMPERATURE ABOVE ABOUT 35* TO 45*C.

Images