Preparation of polynitro compounds from amino acids

Tetrahedron, 1963, Vol. 19 Suppl.l, pp, 165 to 176. Pergamon Press Ltd. PREPARATION OF POLYNITRO COMPOUNDS FROM AMINO ACIDS H. FEUER, G. B. BACHMAN, C. R. KOLLER and W. A. SWARTS Department of Chemistry, Purdue University, Lafayette, Indiana Abstract- Amino acid esters, formaldehyde and trinitromethane react readily in a Mannich type condensation to form N-trinitroethylamino acid esters in good yields. These products decompose slowly on standing but are stabilized by further nitration to the N-nitro-N- trinitroethylamino esters. Hydrolysis of the esters to the free acids.is readily accomplished by warming with a mixture of hydrochloric and acetic acids. The above reactions are applied to both IX- and p-amino carboxylic acids and also to both a- and (3- amino dicarboxyJic acids. Azidcs of the substituted (3-amino acids decompose normally to give isocyanates, but azides of the substituted a-amino acids decompose abnormally and yield none of the expected iso- cyanates. THE object of this work was to prepare compounds suitable for conversion to polynitro polymers. Since polymers are readily prepared in various ways from isocyanates the synthesis of polynitro-isocyanates and -diisocyanates was under- taken by the well known Curtius rearrangement of acid azides which can be readily prepared from acid chlorides and metal azides. The Mannich type reaction of trinitroethanol (I) or trinitromethane (II) and formaldehyde with H.FEUER, G.B.BACHMANN, C.R.KoLLER and W.A.SWARTS Instead they were subjected to further nitration with fuming nitric acid. The various N-nitro-N-trinitroethylamino esters which were prepared are listed in Table 1. - Analyses R m.p.e Yield Found Calc. CH2C0 2Et 78° 72% C, 23-35; H,2-76; C, 23'16; H,2'92; N,22'42 N,22'51 (CH2)2C02Eta 410 - C, 25'67; H,3'42; C,25-84; H,3'38; N,21'24 N,21'53 CH(C02Et)CH2C02Et 73° 76% C, 30-38; H,3-86; C, 30'23; H, 3'78; N,11'89 N,17'63 CH(C02Et) (CH2)zC0 2Et 52° 73% C, 32'38; H,4'04; C, 32'12; H,4'13; N,17'18 N,17'03 CH(CH2C02Et)2 67-69 0 31'5% C, 32-16; H,4'07; C, 32'12; H,4'13; N,17-07 N,17'03 " Prepared from the corresponding acid chloride. b Decomposition point. The hydrolysis of ester V with concentrated hydrochloric acid gave a 79% yield of N-nitro-N-trinitroethylglycine (VI) after refluxing for 24 hours. When dilute hydrochloric acid (20%) was used, the hydrolysis was much faster, but lower yields of VI were obtained. Ammonium chloride was isolated which in- dicated that some decomposition had occurred. The hydrolysis of V and its homologs to the corresponding acids was found to proceed best on heating with a mixture of acetic and concentrated hydrochloric acids, for the reaction was completed in 2 hours or less and afforded the acids, which are enumerated in Table 2,' in high yields. Analyses R m.p." Yield Found Calc_ CH2C02H 149'5° 79% C, 17-17; H, 1'81; C, 16'96; H,1-76; N,24'45 N,24'73 (CH2)2C02H 151-152'5° 6% C, 20-48; H,2'56; C, 20'20; H,2'35; N,23-50 N,23-57 CH(C02H)CH2C02H 1730 60% C, 21'40; H, 2-02; C, 21'11; H,2'05; N,20'38 N,20'S2 CH(C02H)(CH2)2C02H 134'5° 62% C, 23'65; H,2'57; C, 23'66; H,2'S3; N,19'78 N,19'71 CH(CH2,C02,H)2 165-166° 56% C, 23'5'8; H,2'70; C, 23-66; H,2'53; N,19'62 N,19'71 CH(CH2C02H) 1520 15%b C, 26'34; H,3'18; C, 26'01; H, 2'97; . (CH2) 2C02H N,18-47 N,18'97 a Decomposition point. b Based on ~-aminoadipic acid. 166 Preparation of polynitro compounds from amino acids Acid VI was readily converted on refluxing with thionyl chloride to N-nitro~ N -trinitroethylaminoglycyl chloride (VII). Efforts to convert N-nitro-N-tri- nitroethylaspartic acid with thionyl chloride or oxalyl choride to the corre- sponding chloride were unsuccessful. However, treatment of the acid with phos- phorus pentachloride gave the desired N-nitro-N-trinitroethyl aspartyl chloride and in addition a small amount of N-nitro-N-trinitroethylaspartyl anhydride. The various N-nitro-N-trinitroethylamino acid chlorides which were prepared are listed in Table 3. The structures of these acid chlorides were checked by re- converting them to the original esters by reactions with ethanol. Compound VII was converted to N-nitro-N-trinitroethylaminoglycyl azide (VIII) by reac- tion with sodium azide in acetic acid or with silver azide in toluene. The latter reagent gives the better yields (90%vs. 60%) but is fare more hazardous to use because of its proneness to explosion. Compound VIII was isolated in a relativ- ely pure crystalline form, which gave fairly good analytical values. Analyses R m.p." Yield --~'.' IFound Calc. CH2COCI 81-82° 80% C, 15'74; H, 1'21 ; C, 15'92; H, 1'32; N,23-38 N,23'21 (CH2)2COCI 75-76° 80% C, 19'51; H, 2'01; C,19'0l; H,I~90; N,21'81 N,22-18 CH(COCI)CH2COCI 62-63° 30% C, 19'20; H,I'35; C, 19'04; H, 1-32; N,18'43 N, 18'51 CH(COCl)(CH2)2COCI 71-72° 66% C,21'22; H,I'78; C,21'42; H,I'78; N,17'81 N,17'82 CH(CH2COCI)2 114-116° 60% C,21'46; H,I'88; C,21'42; H, 1'79; N,17'34 N,I7'85 CH(CH2COCI) 73-74° 92% C, 24'05; H,2'55; C, 23-65; H, 2'21; . (CH2)2COCI N,17'05 N,I7'24 a Decomposition point. When VIII was decomposed in inert solvents the only solid material which could be isolated was a polymeric material that varied in color from yellow to deep red and decomposed with charring over wide ranges of temperature (60-150°) Gases formed in the decomposition of VIII were analyzed. When the decomposition was carried out in toluene at 80-100° the gaseous mixture con- tained approximately 72% inert gas, 26 % carbon dioxide and 2 % carbon monoxide. When the decomposition was carried out at 55-60° the mixture was different and contained 94'5% inert gas, O' 5% carbon dioxide and 5 %carbon monoxide. No nitrogen dioxide, nitrous oxide or hydrogen azide were present. These data were checked by repeat runs using an Orsat apparatus and a mass spectrograph for analysis. Apparently, the gaseous mixture evolved, at the lower temperature especially, does not differ materially from what would be 167 H. FEUER, G.B. BACHMANN, C.R.KoLLER and W.A. SWARTS expected from the normal dissociation of an azide. Hence the formation of an abnormal product from the decomposition of VIII seems more likely to have occurred as the result of instability of isocyanate or more likely its tendency to polymerize. As in the case of VIII, rearrangement of other acid azides such as Nciitro-Nctrinitroethylaspartyl and N-nitro-N-trinitroethylglutamyl azides, in which the N-nitro-N-trinitroethylamino group is located on an a-carbon, led to tarry materials. On the other hand, ,B~amino acids, in which the Nvnitro-Nctrinitroethyl group is located on a If-carbon, behaved normally in the decomposition of their azides and gave isocyanates contaminated with polymeric material. ,B-(N-Nitro-N-trinitroethylamino)ethyl isocyanate (IX) was obtained from ,B-N-nitro-N-trinitroethylalanyl azide in 17% yield. The structure of IX was established by elemental analysis and by conversion to its ethyl carbamate which also analyzed correctly. CHCl,(02N)aCCH2N(N02)CH2-CH2CONa rs:: (02N)aCCH2N(N02)(CH2)2NCO IX ,B-(N-Nitro-N-trinitroethylamino)-1,3-propyl diisocyanate which was obtained from the rearrangement of ,B-(N-nitro-N-trinitroethylamino)glutaryl diazide was a gummy solid which could not be crystallized but which analyzed fairly well. It gave a crystalline diurethane with ethanol which analyzed correctly. ,B-(N-Nitro-N-trinitroethylamino)-I,4-butane diisocyanate was also a gummy solid. Its diethyl urethane was obtained as an oil. These isocyanates were extremely sensitive to atmospheric moisture; on short exposure they became crumbly solids that contained bubbles of entrapped gas. H is very likely that the isocyanate groups reacted with water to yield the un- stable carbonic acid group that is known to decompose with the evolution of carbon dioxide. EXPERIMENTAL Ethyl N-trinitroethylglycine Ethyl glycine hydrochloride was prepared from glycine according to the procedure of Harris.f To a mixture of 7'24g (0,04 mole) trinitroethanol and 5'6g (0'04 mole) ethyl glycine hydrochloride in 50ml water was slowly added 8ml 5 N sodium hydroxide while the temp was maintained at 20°. The solution was stirred for 15 min at 20° and an oil separated which solidified to a light yellow solid when the solution was allowed to stand at 5° for 30 min. This solid was filtered, washed with 50ml cold water and dried on a clay plate. There was obtained 6'7 g (63 %) of ethyl N-trinitroethylglycine, mop. 116-119° (dec). Ethyl N-trinitroethyl-N-nitroglycine To a nitrating mixture prepared by mixing 10ml fuming nitric acid (sp. gr. 105) with 10ml cone sulfuric acid at 0° was added 6'7g (0'025 mole) of ethyl N-trinitroethylglycine at 25° with stirring. This mixture was slowly heated to 60° and poured onto 500g ice after being 2 Harris and Weiss, Liebigs Ann. 327, 365 (1903). 168 Preparation of polynitro compounds from amino acids allowed to cool to 250. A white solid precipitated which was filtered off, washed with water and dried in a vacuum desiccator. There was obtained 5'68 g (72 %) of ethyl N-trinitroethyl- N-nitroglycine, m.p. 760. Recrystallization from carbon tetrachloride gave 5'6 g white needles, m.p. 780 (dec). N-Nitro- N-trinitroglycine (VI) Ethyl N-nitro-N-trinitroethylglycine (9'7g, 0'031 mole) was dissolved in a mixture of 50ml acetic acid and SOml cone hydrochloric acid and the solution heated on a steam bath for about 90min. Removing any volatile materials in vacuo and repeated washing of the residue with water left a white solid which was dried in vacuo at 250. Recrystallization from ethylene dichloride gave 7·0g (79 %) of N-nitro-N-trinitroethylglycine, m.p., 149~5° (dec). N-Trinitroethyl-N-nitrog/ycy/ chloride (VII) One gram (0'0035 mole) of N-trinitroethyl-N-nitroglycine was suspended in 15ml thionyl chloride and heated at reflux. All of the compound dissolved in 30 min and after 4 hr there was no more evolution of hydrogen chloride. The thionyl chloride was distilled off, and the residual oil crystallized upon addition of 35ml pet ether (60-700). The white solid was filtered off, washed with pet ether, and the excess solvent was removed in vacuoat 250. There was obtained 0'85 g (80 %) ofN-trinitroethyl-N-nitroglycyl chloride, m.p, 81-820.Recrystall- ization from carbon tetrachloride did not raise the m.p. Ethyl N-trinitroethyl-N-nitroglycine, m.p. 78° was obtained when VII was treated with an excess of absolute ethanol at 250 for I hr and the mixture poured into water. Methyl N- trinitroethyl-N-nitroglycine, m.p. 103° was prepared in the same manner. N- Nitro- N-trinitroethylg/ycy/ azide (VIII) Compound VII (17'8 g, 59'1 mmoles) was dissolved in 50ml glacial acetic acid and to the solution was added in small batches at 150 with stirring 4'1 g (63 mmoles) sodium azide. The resultant slurry was stirred for about 5 hr while the temp rose to 25°. Itwas then diluted with 50ml methylene chloride washed with 8 x 50ml portions of distilled water and the solution dried over magnesium sulfate. Pouring the dry solution into 150ml hexane and keeping the turbid solution in a freezer gave 11·5g (63,6 %) of N-nitro-N-trinitroglycyl azide (VIII) m.p, 76-780 (dec). Compound VIII could also be purified by dissolving it in chloroform at 25-300and placing the solution in the refrigerator overnight. (Found: C, 15-70; H, 1'78; N, 33'73; Calc. for C4H4Ng09 : C, 15'58; H, 1-31; N, 36'3%). Decomposition ofN-nitro-N-trinitroethy/g/ycy/ azide (a) In toluene at 80-1009 • A 500ml 3-necked round-bottomed flask was equipped with a Friedrichs condenser, a thermometer in a well, and a stopper. A lead from the condenser was connected to both a gas measuring buret and a jug of saturated sodium sulfate solution by means of a 3-way stopcock. The flask was charged with a solution of 10'7 g N-nitro-N-trinitro- ethylglycyl azide in 500rnl dry, analytical reagent grade toluene; the volume of the system above the solution was approximately 200m!. When the solution was heated, very slowevol- ution of gas was noted at about 350; decomposition became appreciable at 55°, and vigorous at about 70°. The temp was raised to 800 while the system was swept with twice its volume of decomposition gases .. The effluent gases were then admitted to the collecting jug. When decomposition slowed down at 800, the temp was raised to 1000, causing a further evolution of gas that quickly subsided. The lead to the collecting jug was kept open during this last ephemeral evolution of gas. 169 H.F I!UER, G.B.BACHMANN, C.R.KoLLER and W.A.SWARTS The gases were then passed through an Orsat apparatus for a quantitative determination. An average of 4 analyses gave the following approximate percentages of gases: carbon dioxide -- 26 %, carbon monoxide - 2 %, inert gas - 72 %. A red, gummy, polymeric mass was obtained when some of the toluene solution of the decomposition product was poured into hexane . (b) /11 chloroform at"58-59°. The apparatus used was identical with the one described in part a. Decomposition began at 350, and became noteworthy at 55°. The temp was main- tained at 58-590 for about 2!hr as the system was swept with twice its volume of de- composition gas before a sample was collected. Examination of this sample in the Orsat showed the presence of 0'5 % carbon dioxide, 5 % carbon monoxide, and 94'5 % inert gas. Removal of the solvent ill vacuo at 250 after the decomposition had been terminated afforded the red polymeric mass, and some unchanged azide . (c) /11 toluene at 70-720 and 72-95 0; gases analyzed ill a.mass spectrometer. The apparatus described in part (a) was modified in that the lead from the condenser was connected to a wet test meter and an evacuated gas collecting tube. A total of 6'16g (0'02 mole) of N-nitro- N-trinitroethylglycyl azide was dissolved in 500ml toluene, and the temp of the solution quickly raised to 70-72°. The system was swept with twice its volume of decomposition gases before a sample was collected for mass spectrometric analysis . The temp was then gradually raised to 95° and the system swept with its own volume of decomposition gas before the next sample was collected for analysis. A total of approximately 660ml gas at 260 and 742·8mm was evolved by this decomposition, corresponding to a total of 587ml at S.T.P.; this amount is 131 % of the calculated 448ml (0'02 mole) expected from a normal Curtius degradation. Diethyl N-Nitro-N-tl'inifroethylaspartate A mixture of 42g (0'222 mole) of diethyl aspartate, 33'4g (0'222 mole) of trinitromethane, and 18m! 37 % formalin solution (6'65g, 0'222 mole of formaldehyde) in 150ml distilled water was stirred at 0-5° for about 1hr and allowed to stand at that temp for another hr. Water was then decanted from the heavy, oily diethyl N-tdnitroethylaspartate which was dissolved in about 100ml ether and then dried over magnesium sulfate. The drying agent was then filtered off, the ether removed in vacl/o, and the residual oil dissolved in 75 ml acetic anhydride. This solution was placed in a 300ml 3-necked round-bottomed flask quipped with a mechanical stirrer, a thermometer, and an addition funnel, and was cooled to 50. The solution was stirred vigorously while 35ml fuming nitric acid was added over 1hr as the temp was maintained at 5-15 °. After the addition of the nitric acid was completed, the mixture was stirred for 1hr at 25°, and then poured over crushed ice. The crude diethyl N-nitro- N-trinitroethylaspartate was dissolved in fuming nitric acid at 5° and then poured over crushed ice for purification. A yield of 66'8 g (76'1 %) pure diethyl N-nitro-N-trinitroethyl aspartate, m.p, 72-73° (dec) was realized. N-Nitro-N-trinitroethylospartic acid A total of 40g diethyl N-nitro-N-trinitro.ethylaspartate was dissolved in a mixture of 400ml glacial acetic acid and 400ml cone hydrochloric acid, and the solution heated on the steam bath for 2hr. The solvents were removed ill vacuo, while being heated on a steam bath, and the residue dissolved in about 200ml ether. After the ethereal solution had been dried over magnesium sulfate, it was concentrated to about 100m1 and added to approximately 400m1 hexane. After this mixture had remained in the refrigerator overnight, a total of 15'1 g (44%) of N-nitro-N-trinitroethy1aspartic acid, rn.p. 1670 (dec) was filtered off. When the solvents were allowed to evaporate in the hood, a considerable residue was left. This was dissolved in about 25ml ether, 50ml hexane added and this mixture kept in the re- frigerator for I week, to give 5·4 g more product. The total of 20'5 g acid is 59·6 %of the theo- retical amount. A small amount of this acid decomposed at 173D after several purifications from ether and hexane . 170 Preparation of polynitro compounds from amino acids N-Nitro-N-trinitroethylaspartyl chloride and N-nitro-N-trinitroethylaspartic anhydride In a 50011 round-bottomed flask fitted with a reflux condenser and drying tube were placed 1'05 g (3'8 mmole) N-nitro-N-trinitroethylaspartic acid and l'92g (9'5 'mmole) phosphorus pentachloride and the mixture kept at 40° for 24hr. Phosphorus oxychloride was removed in vacuo and the residue extracted with 5 x 30ml portions of carbon tetrachloride. Evaporation of the solvent in vacuo left an oil which crystallized to give O'70g (49%) of crude N-nitro- N-trinitroethylaspartyl chloride, m.p. 53-56° (dec). Pure acid chloride (0,43g, 30%) m.p. 62-63° (dec) was obtained by dissolving it in 40ml carbon tetrachloride and precipitating with 140011 pet ether (60-70°). Cooling the residue, which remained after the carbon tetrachloride extraction, at - 10° for 2 days gave a yellow semi-solid which after 3 recrystallizations from carbon tetrachloride afforded a small amount of N-nitro-N-trinitroethylaspartic anhydride, rn.p., 113-114° (dec). (Found: C, 23'32; H, 2'88; N, 21'03; Calc. for C6HsNsOu : C, 22'29; H, 1'56; N, 21'67%). DiethyIN-nitro-N-trillitroethylglutamate .A solution of 71g (0'297 mole) diethyl glutamate hydrochloride, 45g (0·t98 mole) trinitro- methane, and 24'5ml 37 % formalin solution (9'05 g, O'302 mole, slight excess of formal- dehyde) in 300011 distilled water was stirred at 5°. Approximately 60ml5 N sodium hydroxide (0'3 mole) was added to the solution in 20 min. The mixture was stirred at 5_10° for another 30 min and then allowed to stand at that temp for another hour. The water was decanted and the oily diethyl N-trinitroethylglutarnate dissolved in about 150011 ether. The solution was dried with magnesium sulfate, the drying agent filtered off, the solvent removed in vacuo, and the remaining oil dissolved in 70ml acetic anhydride. This solution was placed in a 300ml 3-necked round-bottomed flask equipped with a thermometer in a well, a mechanical stirrer, and an addition funnel. The solution was cooled to 5° in an ice bath, and 35m! fuming nitric acid added over ll/2hr while the temp was maint- ained at 5-15°. After the addition had been completed, the nitration mixture was allowed to stir at 25° for another hr before being poured over crushed ice. The oil that precipitated out solidified when the mixture was allowed to remain in the refrigerator for a day or two. Puri- fication of the solid by dissolution in about 25ml fuming nitric acid at 5_10° followed by precipitation over crushed ice afforded 88·8g (72'6 %) diethyl N-nitro-N-trinitroethyl- glutamate, m.p, 51-52° (dec.). N-Nitro-N-trinitroethylglutamic acid Diethyl N-nitro-N-trinitroethylglutamic acid (30g) was dissolved in a mixture of 300ml glacial acetic acid and 300ml cone hydrochloric acid. The solution was heated on the steam- bath for 2hr and the excess acids removed in vacuo. The last traces of acid were removed by adding l5ml distilled water and evaporating in vacuo. This operation was repeated 3 times. Drying the residue in vacuo and recrystallizing from ethylene chloride gave 16'9g (62,2%) acid, m.p. 132-134° (dec.). N-Tl'initroethyl-N-nitroglutamyl chloride In a 50ml flask fitted with a reflux condenser and a drying tube were placed 1'7g (5mmole) N-trinitroethyl-N-nitroglutamic acid and 2'1g (l Ommole) phosphorus pentachloride. A vigorous evolution of hydrogen chloride began within 5 min and the flask was cooled in an ice-bath to moderate the reaction. After this initial reaction the mixture was allowed to stand at 25° until a homogeneous liquid resulted in 72hr. Thirty-five ml carbon tetrachloride was added and a small amount of insoluble material was filtered off. After removing the carbon 12 Times N. R. 171 H.FEUER, G.B.BACHMANN, C.R.KoLLER and W.A.SWARTS tetrachloride by distillation, the residual oil was heated for 1hr at 30° and 1mm in order to remove the phosphorus oxychloride. To the residue was added 40ml pet ether (60-70°) and the oil solidified to a white solid which was filtered, washed with pet ether and dried in vacuo at 25°. There was obtained 1'3 g (66 %) N-trinitroethyl-N-nitroglutamyl chloride, rn.p. 71-72° (dec). Recrystallization from a mixture of carbon tetrachloride and pet ether did not raise the m.p, Ethyl N-trinitroethyl-N-nitroglutamate, m.p. 51-52° was obtained by treating this acid chloride with an excess of absolute ethanol at 25° for 30 min and then precipitating the ester with water. Ethyl N-nitl'o-N-tl'initl'oethyl-{j-alanille A solution of 30·8g (0'2 mole) ethyl !J-alanine hydrochloride, 30'2g (0'2 mole) trinitro- methane and 17ml 37 % formalin solution (6'3g, 0'22 mole, slight excess of formaldehyde) in lOOml distilled water was stirred at 5° while 40ml 5 N sodium hydroxide (0'2 mole) was added. At times it was necessary to use a slight excess of sodium hydroxide before all the oily ethyl Nstrinitroethyl-fi-alanine precipitated out. The mixture was stirred for 30min at 5° after addition of base had been completed, and then allowed to stand at that temp for another hr. Water was then decanted, the remaining oil dissolved in about 75ml ether, and the ethereal solution dried with magnesium sulfate. After the drying agent had been filtered off and the ether removed in vacuo, the residual oil was dissolved in 50ml acetic anhydride, and this solution placed in a 300ml 3-necked round-bottomed flask equipped with a thermo- meter in a well, a mechanical stirrer, and an addition funnel. The solution was cooled to 5_10°, and the temp was maintained at 5-15° as 50ml fuming nitric acid was added over II / 2hr . The mixture was allowed to stir at 25° for another hr, and then poured over crushed ice to afford the oily ethyl Nsnitro-Nctrinitroethyl-fi-alanine. The fuming nitric acid-crushed ice purification technique did not afford the ester as a solid. The ester was obtained as a solid, m.p, 41° by reacting the acid chloride with excess ethanol at 25° for 3hr, pouring the solution over crushed ice, and subjecting the material so obtained to the fuming nitric acid purification. N-Nitro-N-trinitroethyl-!J-alanine The oily ester was dissolved in ether and thoroughly washed with distilled water to removed all traces of nitric acid. This was found to be essential in this reaction. After the ether had been removed in vacuo, the residual oil was dissolved in a mixture of 50ml glacial acetic acid and 50ml cone hydrochloric acid. The solution was heated on the steam bath for an hr and the solvents were then removed in vacuo, After two or three 1a-15ml portions of distilled water had been added and removed in vacuoto dispose of the last traces of hydrogen chloride and acetic, the solid residue was dried in vacuo. Recrystallization of the dry solid from ethy- lene dichloride gave 1'6g N-nitro-N-trinitroethyl-~-alanine,m.p. 151-152,5° (dec). N-Nitro-N-trinitroethyl-!J-alanyl chloride Nsnitro-Nctrinitroethyl-ji-alanine (0'45g) was added to l Ornl thionyl chloride, the mixture refluxed for 5 hr, and the resultant solution poured into 75ml hexane. After the turbid mix- ture had remained in the refrigerator overnight, a total of 0'25 g (53'5 %) N-nitro-N-trinitro- ethyl-ji-analyl chloride, rn.p. 75-76° (dec) precipitated. Placing the filtrate in the refrigerator for another week gave an additional 0'13g product for a total of 0·38g which is 79'5 % of the calculated amount. 172 Preparation of polynitro compounds from amino acids 2-(N-Nitro-N-tl'initl'oethylamino)ethyl isocyanate NvNitro-Nctrinitroethyl-ji-alanyl chloride, 0'25g (0'795 mmole) was dissolved with stirring in 5m1glacial acetic and the solution cooled to 15°.A total of O'5g (2.5mmoles) of sodium azide was added and the resultant slurry stirred at 15° for 90min. It was then diluted with 30ml chloroform, washed with 8 X 25ml portions of distilled water and the clear chloroform solution dried over magnesium sulfate. The azide was not isolated, but the dry chloroform solution was refluxed for about 3hr. Decomposition began at about 40°, and became rapid at 50°. Removal of the chloroform ill vacuo afforded an oil that was taken up in 5ml toluene. Addition of about 25ml hexane precipitated a little red, gummy, oily semi-solid. The supernatant solution was decanted from this mass and another 75m1 hexane added, producing a turbid mixture. After this mixture had remained in the refrigerator for a day, a total of 40mg (17'1 %) 2-(N-nitro-N- trinitroethylamino)ethy1 isocyanate, m.p, 48-49° (dec), precipitated out. (Found: C,20'50; H, 2'40; N, 28'60; Calc. for C6HaNo0 9 : C, 20'40; H, 2'04; N, 28'57%). The ethylurethane of this isocyanate was prepared by adding excess ethanol to the chloro- form solution of the azide. After decomposition had been effected, the solution was concen- trated in vacuo to about 5ml, and poured into about 50ml hexane. The oil that precipitated became solid after the mixture had been in the refrigerator for several days. A similar preci- pitation from ethylene dichloride by means of hexane afforded slightly golden yellow crystals, m.p. 74-75° (dec), (Found: C, 24'37; H, 2'95; N, 25'04; Calc. for C7H12NoOlO : C, 24'70; H, 3'52; N, 24·70 %). Diethyl{3-(N-llitro-N-trinitl'oethylamino) glutarate Ethyl ,B-aminoglutarate hydrochlorlde.f 12'Og (0'05mole) was dissolved in 20ml distilled water, the solution cooled to about 50° in an ice-bath, and mechanically stirred. A solution of 7'5g (0'05 mole) trinitromethane in 50ml distilled water and 5m137% formalin (approxim- ately 1'8 g, 0·06 mole, slight excess of formaldehyde) were added to the cooled solution of the amino ester hydrochloride. Ten ml 5 N sodium hydroxide (0'05 mole) was slowly added to the stirred solution; an oil precipitated as the base was added. The solution was stirred at about 5° for 1/2hr after addition of sodium hydroxide was completed, and the mixture allowed to stand at the same temp for another 1/2hr. The water was then decanted and extracted with about 75ml ether. This ether extract was used to dissolve the precipitated oil, and the solution dried over magnesium sulfate. The magnesium sulfate was :filtered off, the ether removed ill vacuo, the last traces with the aid of a vacuum pump, and the remaining dark red-brown oil dissolved in 45ml acetic anhydride. This solution was placed in an ice cooled 200ml 3-necked round-bottomed flask equipped with a dropping funnel, a "Tru-Bore" stirrer, and a thermometer. The solution was stirred vigorously while 15ml fuming nitric acid was added over a period of 45 min and the temp was maintained at5-10°. Stirring was continued for another 2hr atO-I2°, and the solution was then poured over crushed ice. The diethyl ,B-(N-nitro-N-trinitroethylamino). glutarate crystallized overnight, and was purified by dissolution in fuming nitric acid and subsequently pouring the acid solution over crushed ice. The yield of 6'5 g, m.p, 67-69° (dec), so obtained represented a yield of 31'4%. ,B-(N-Nitro-N-trillitroethylamillo) glutaric acid Ethyl ~-N(-nitro-N-trinitroethylamino) glutarate, 6'2g (15 mmoles) was dissolved in a mixture of 30ml acetic acid and 15m137 % hydrochloric acid, and the solution heated on the steam bath for about 21/ 2hr. The solvents were removed in vacuo, 20ml distilled water added, 3 H.Feuer and W.A.Swarts, J. Amer. Chern. Soc. 77,5427 (1955). 173 H.FEUER, G.B.BACHMANN, C.R.KoLLER and W.A.SWARTS and the resultant solution evaporated to dryness. The slightly brown residue was dissolved in 75ml ether, and the solution dried over magnesium sulfate. After the drying agent had been filtered off, 400ml hexane was added to the ethereal solution, and the resultant turbid solution placed in the refrigerator overnight. A yield of H g (55'7 %) ~-(N-nitro-N-trinitro­ ethylamino)glutaric acid, m.p, 165-166° (dec) was obtained. This acid may also be re- crystallized from ethylene dichloride using approximately 100ml solvent per gram of acid. The overall yield (based on diethyl ~-aminoglutarate) hydrochloride), was raised from 17'5% to 25 %in one small scale experiment in which the acid was recrystallized from ethylene dichloride and hexane subsequently added to the filtrate to obtain more product. The analyti- cal sample was recrystallized from ethylene dichloride. ~-(N-Nitro-N-trinitroethylamino) glutaryl chloride P-(N-Nitro-N-trinitroethylamino)glutaric acid, 4'77 g (13'4mmoles) and 6·59g phosphorus pentachloride (31'6mmoles, slight excess)were thoroughly mixed in a 100mi round-bottomed flask equipped with a condenser protected by a drying tube. The mixture was heated at 40-50° for 50hr, and phosphorus oxychloride was then removed in vacuo. Recrystallization of the acid chloride from carbon tetrachloride gave a total of 3'14g ~-(N-nitro-N-trinitro­ ethylamino)glutaryl chloride, m.p. 114-116° (dec). 2-(N-Nitl'o-N-trinitroethylamino)-1,3-pr,opane diisocyanate A solution of 0'4g (1'04mmoles) of ~-(N-nitro-N-trinitroethylamino)glutaryl chloride in lOml acetic acid was cooled to 15° and stirred mechanically. Approximately 0'35g (4'62 mmoles of sodium azide was added to the solution, and the resultant slurry stirred at 15°for 30min. The slurry was then diluted with 50ml chloroform, the solution washed with 8 x 50m portions of distilled water, and then dried with magnesium sulfate. After the drying agent had been removed, the chloroform solution of the azide was heated in a flask which had been thoroughly flamed out and flushed with dry nitrogen. Decomposition began at about 40°, and became vigorous at about 50°. The chloroform solution was heated under reflux for about 30 min, and the solvent then removed in vacuo. The yellow, gummy semi-solid material that was left after all the chloroform had been evaporated was protected from at- mospheric moisture and submitted for analysis. (Found: C, 24'15; H, 3'06; N, 27'15; Calc. for C7H7N7010: C, 24'07; H, 2-00; N, 28·08 %). Diethyl urethane of 2-(N-Nitro-N-trinitroethyl-amino)-1,3-propane diisocyanate The diethyl urethane of 2-(N-nitro-N-trinitroethylamino)-1,3-propane diisocyanate was prepared by adding a vast excess of pure dry ethanol to the dry chloroform solution of the diazide, and by refluxing the solution for about 30 min. The solution was concentrated to about 7ml and poured into 50ml hexane. The solid that precipitated overnight was re- crystallized thrice from ethylene dichloride to yield a solid m.p, 119-121° (dec). (Found: C, 29'03; H, 4'30; N, 22'14; Calc. for CUH19N7012: C, 29'93; H, 4'30; N, 22'22%). Diethyl p-aminoadipate hydrochloride ~.Aminoadipic acid", 12'05g (0,055 mole) was added to approximately 150ml absolute ethanol that had been previously saturated with hydrogen chloride at 0°. This mixture was refluxed for 16hr, and alcohol and hydrogen chloride were then removed in vacuo leaving an oil. Attempted purification by adding ether to a chloroform solution of the oil failed to yield a solid product. 174 Preparation of polynitro compounds from amino acids Diethyl P-(N-Nitro-N-trinitroethylamino)-adipate The oil obtained in the previous experiment was freed of hydrogen chloride by adding and subsequently removing in vacuo several portions of ethanol. The amino ester hydrochloride was dissolved in 100mI distilled water and the solution cooled to 5° while being stirred me- chanically. A solution of 8·4g (0'055 mole) of trinitromethane in distilled water and 5ml 37 % formalin solution (1'8g or 0'06 mole, slight excess of formaldehyde) were added to the stirred, cooled solution. A fairly dark oil precipitated, but went into solution slowly when sodium hydroxide was added. Addition of 11ml 5 N sodium hydroxide (0'055 mole) caused a yellow oil, different from the original oil, to precipitate from the solution. After addition of the base was completed, the mixture was stirred for 1/2hr, allowed to stand at 0° for another 30 min, the supernatant water decanted off and extracted with one 100mI portion ether. This ether was used to dissolve the oily diethyl ~-(N-trinitroethylamino) adipate, and the ethereal solution was dried with magnesium sulfate. The drying agent was filtered off, the ether re- moved in vaCI/O, and the residual brown oil dissolved in 30ml acetic anhydride. This sol- ution was placed in an ice cooled 100ml 3-necked round-bottomed flask equipped with a thermometer in a well, an addition funnel and a "Tru-Bore" stirrer. Nitration was per- formed at 5_10° by adding 16ml fuming nitric acid over 1hr while the solution was vigorously stirred. After the addition was completed, the mixture was stirred for 21/ 2hr at 4-17°, and then poured over crushed ice. The oily diethyl ~-(N-nitro-N-trinitroethylamino)adipate failed to crystallize, even after a purification treatment with fuming nitric acid. P-(N-Nitro- N-tl'initl'oethylamino)adipic acid The oily ester obtained in the previous experiment was dissolved in a mixture of 50ml glacial acetic acid and 50ml cone hydrochloric acid, and the solution heated on the steam bath for 2hr. The solvents were then removed in vacuo to dispose of all traces of acetic acid and hydrogen chloride. The residue was dissolved in 50ml ether, and after the etherealsolution had been dried with magnesium sulfate, about 400ml hexane were added, and the turbid mixture placed in the refrigerator overnight. A total of 3'01 g ~-(N-nitro-N-trinitroethyl­ amino)adipic acid, m.p. 152°(dec)was isolated. The yield was 3·01g (14'8 %) of the calculated amount based on ~-aminoadipic acid. P-(N-Nitro-N-trillitroethylamino)adipyl chloride ~-(N-Nitro-N-trinitroethylamino)adipic acid, 3'56 g was suspended in 50 ml thionyl chloride, and the mixture heated at 55-60° for 3 days, during which time the solid gradually went into solution. The solution was then concentrated to 10ml in vaCI/O, poured into 75ml hexane, and the turbid mixture placed in the refrigerator overnight. A total of 3'38g (85'8 %) fJ-(N-nitro-N-trinitroethylamino) adipyl chloride, m.p. 73-74° (dec) was obtained. Addition of another 50ml hexane to the filtrate afforded 0·19g more acid chloride after 6 weeks in the refrigerator. The total of 3·57g represents a yield of 91'5 %. In experiments in which the thionyl chloride was heated under reflux, the yields of solid acid chloride ranged from 0 % to 21-8%. 2-(N-Nitro-N-tl'initroethylamino)-1A-butane dlisocyanate A solution of 1'03 g (2'46 mmole) of fJ-(N-nitro-N-trinitroethylamino)adipyl chloride in 10ml acetic acid was cooled to 15° and stirred mechanicaIly. Addition of 0'8g (12'3 mmole) of sodium azide produced a slurry which was stirred at 15° for another 30 min. This slurry was diluted with 50ml chloroform, and the resultant solution washed with 6 x 50ml portions 175 H.FEUER, G.B.BACHMANN, C.R.KoLLER and W.A.SWARTS of distilled water and dried over magnesium sulfate. The chloroform solution of p-(N-nitro- N-trinitroethylamino)adipyl azide was placed in a flask which had been thoroughly flamed out while being swept with dry nitrogen. Decomposition of the azide began at about 45°, and became vigorous at 50°. The chloroform solution was heated under reflux for about 2hr, and the solvent subsequently removed in vacuo. The residue consisted of a yellow, gummy, semi-solid in this case also. This material was protected from atmospheric moisture and sub- mitted for analysis. (Found: C, 27-()0; H, 3'30; N, 25'35; Calc. for CaHgN7010: C, 26'44; H, 2'48; N, 26'99%). Acknowledgment - We are indebted to the Office of Naval Research for the financial support of this work. 176