C . LT'. PLIXMFRIND
2720
N.L.DR.\KE
L'ul. iii
that 95-98'$& of the theoretical chloride ion had formed. distillations were steam distilled to give additional 2,3The white solitl v a s recrystallized from C P . acetone to dimethyl-2,3-tlinitrobutane (1.2 g.); the total yield of 2 3 1ieltl 3-n1ethyl-4-(2-nitro-2-propyl)-isoxazolit1e oxide, :j7.X tlimethyl-2,3-tlinitrobutaric (6.4 g . ) \vas I 1 %. g , 111.p.77-78". 3-Methyl-4-(2-isopropenyl)-isoxazole .--3-?iIet h + L ( 3oxide ( 2 5 g., U.14 mole) was 4 m l . Calcd. for C7111?N101: C, 44.67; 11, ti.4.j; s , 11itro-2-propyl)-isoxazolii1e stirred vigorously with 2 % aqueous sodium hydroxide (200 14.89. Found: C, 44.71; H, 6.24; S,14.2s. i d . ) a t 55-60', \Vithin several minutes, the isoxazoline The alcoholic filtrate I was evaporated a t reduced pressure t o a volume of approximately 150 ml. Ethyl ether (100 oside had dissolved to give a clear yellow solution. After 5 mirirites, cloudiness developed and an oil separated a t the 1111.) was added and the mixture was kept in a refrigerator bottom of the reaction mixture. The mixture was heated for 24 hours. The yellow solid which formed mas filtered; for one hour a t 55-60", cooled, and then extracted with the filtrate I1 was saved. The yellow solid was washed ethyl ether; the aqueous layer, upon acidification, evolved with water (100 ml.) and filtered to yield additional Rmethy1-4-(2-nitro-2-propyl)-isoxazolineoxide, 4.4 g. The osides of nitrogen. The ether extract was dried over sodium sulfate, concentrated and then distilled at fj0' (1 m m . ) total yield of isoxazoline oxide was 42.2 g., 36% of theoretito give a volatile product (condensed a t -80") and a high cal. (The reaction temperatures were varied from -20" to 30'; a t the higher temperatures the yields of isoxazoline boiling residue (12.4 g . , exteiisively decomposed). Rrdistillation of the volatile product gave 3-methyl-4-(Z-isooxide ranged from 10-15% and the amounts of tarry residues propenyl)-isoxazole (1.4 g., 8.1% yield), a colorless liquid were much greater). The aqueous filtrate, after acidification, gave a carbonyl group reaction with 2,4-dinitrophenyl- of powerful, fragrant odor, insoluble in water and soluble in usual organic solvents, b.p. 70-71' (17 mm.), n z 01.4863, ~ hydrazine; this reaction arises from conversion of sodium infrared spectra Fig. 1 E . The isoxazole reacted with satZpropanenitronate by the Nef2' reaction to acetone. urated mercuric chloride to give a white solid which soite~i.; Filtrate I1 was evaporated a t reduced pressure; the reand the11 dissociates at 124-126'. sulting thick sirup was cooled for 2 days in a refrigerator. -4solid precipitated and was filtered (5.2 g.); the filtrate 111 .lnnl. Calcd. for C7H9SO: C, 68.29; H, 7.41"; ?;, 11.38. was saved. The white solid, after recrystallization from Found: C , 68.27; H , 7.17; S,11.31. ether, was identified as 2,3-dimethyl-2,3-dinitrobutane, m.p. The high boiling residue was distilled in a Hickman-type 205-207", lit.24 209"; no depression by an authentic" still (0.08 mm., considerable decomposition) to give a vissample (m.p. 208-210"). cous yellow oil (3.2 g . ) ; redistillation of this product (0.02 The filtrate I11 was distilled to give a yellow liquid, b.p. mm., b.p. 50') gave a colorless oily liquid (2.8 g . ) , n% 50-75' at 2 m m . Ether (J0 ml.) was added and the solu1.5335 which turned dark-red quickly on storage. Contion was cooled t o -78 . A yellow solid precipitated sistent analyses for this material could not be obtaiiictl bewhich was filtered from the cold solution; the filtrate I\' cause of its rapid decomposition; the analyses fell within was saved. The yellow solid, 5-rnethyl-2-nitro-2,4-hexa- the ranges expected for mixtures of 3-rnethyl-4-(2-nitro-2diene ( 3 . 2 g., 3.5% yield) melted at 25-28' after several repropyl)-isoxazole ( V I I ) and 3-methyl-4-(2-isopropenyi)crystallizations from cold ether. Its infrared spectrum conisoxazole (VIII) (similar agreement was obtained from the tains well-defined bands at 6.15 (conjugated diene system), infrared and ultraviolet spectra of the product). The fi.6.5 (nitro group conjugated with diene system) and 7.65 p freshly-distilled liquid dissolves slowly in dilute aqueous (nitro group); absorption in the ultraviolet region (95% alkali to give a product having the characteristic odor anti ethanol) occurs at 230 mp ( E , ~ , ~ ,7,700) and 335 m p many of the properties of 3-1nethy1-4-(2-iso~~ropenyl)-isoxa1 I ,700), zole; it is thus believed t o consist mainly of unstal~lc:iAnal. Calcd. for C:H1,SOy: C, 58.57; H, 7.80; N, 9.93. methy1-4-(2-nitro-2-propyl)-isoxazole (L'II). Found: C, 58.11; H, 7.06; S,9.15. Ultraviolet Spectra of VI11 and of Trimethylisoxazole .The filtrate IV was distilled t o yield a yellow liquid9 (33 The ultraviolet spectra of I-III and of tri~nethylisosazolc iri g.): b.p. .50-58.5" (0.8 mm.), T Z ~ ~1.4812-1.5042. D Rc- .50% ethanol (1 X 10-4 df,resp.) are simple and almost itlciitical; each spectrum exhibited only very little fine structure. peated distillations' failed to result in effective purification The absorption spectrum for \'I11 is described as: estended of this misturc. The combined residuesg (20.3 g.) from all absorption a t 220 m p (log B 4.13), continuous decreased 127) J 17. h-ei, A n n . , 280, 263 (1851). absorption over the range 220-270 m p (log E range of 4.13(28) E . .A. Braude, E . R. H. Jones and G. G Rose, J . Chem. .Sor , 2.71), and weak absorption a t 270-400 mp. The spectruni 1104 (1517), report t h a t t h e ultraviolet spectra of conjugated nitrw of trimethylisoxazole indicates maximum absorption a t 223 olefins contain from one t o three hands of similar intensities ( 6 - l o p m p (log E 3.92) with continuous decreased absorptioti over apart) in t h e region o f 225-260 m p : enlax 3300-1)700. l-iYitro-l,3the range 223-250 m p (log E range of 3.92-2.78) nntl wcxk pentadiene absorbs a t 223 mp (erIIsx 5500) a n d 298 m p ( e r r l n , ahqorption over the range 2.5n-400 m p .
-
~
~~~~
~~~
~
~
12,000): 8-nitrostyrene al>+c,rtnat 227 ,,,:, lTi.Tr00).
rnp !cmnY
1).-)00) a n d 300 rnp
COLUMBUS. OHIO
(C
/CO'J?.RII3UTION FROM TIIE DEPARTMES.T O F CIIEZIISTRY O F THE
USIvERSITY O F
MARYLAND]
An Improved Procedure for Preparing Primary Nitroalkanes by the Victor Meyer Reaction BY C. IY.PLURIMER AXD N. L. DRAKE RECFIVED sovEnIBER
in,
1953
An improved procedure is tlescrihetl for prepnring six primary nitronlkniics of high purity formerly obtained.
iii
tnuch better yields than those
Of these, the first gives considerably better yields, although they are low. hiore recently, a few prihigher homologs of primary nitroalkanes. These mary nitroalkanes have been prepared by the l,-thave been prepared almost exclusively by two addition of an alkylmagnesium bromide to an a$methods: (1) the well-known reaction between an unsaturated nitr~olefin.~In general, the yields are alkyl halide and silver nitrite,' and ( 2 ) the reaction comparable to those obtained froin the Victor of an a-halocarboxylic acid with sodium nitrite.? Ifeyer reaction. The review and extension of the Victor 11cycr rc( I ) V Meyer and 0 Stuher, R P Y ,5 , 203 (1S72)
-1search of the literature has revealed the need
of a good laboratory procedure for preparing the
(2) IT Kollie .7 f i i n i t C h m i
6 , 427 (1872)
(3) G D Rurkley and R Fllrrv .7 rhPm Tor
i4'lt f l ' l 171
TABLE I CUYPARATIVE YIELDSA N D PHYSICAL CONSTANTS OF PRIMARY NITROALKANES Product
Yield,
%
B.p. 'C.
Mm.
1-Nitrobutane 2-Methyl-1-nitropropane 1-Nitropentane 3-Methyl-1-nitrobutane 1- Nitrohexane
63.8 44.0 66.6 61.0 69.4
58.0-58.3 60.3-60.7 75 -76 68.0-68 3 91 -92
23 44 23 24 24
I-Nitroheptane
60.6
707-108
25
n%
MD'~ 0 5 4
Obsd.
M D I ~ Yield, %
Calcd.
1.4080 0.9673 26.24 26.30 W8 .9602 26.32 26.30 168 1.4050 1.4161 .9493 30.97 30.92 . . . ,9458 30.97 30.92 28" 1.4143 ,9354 35.62 35.54 40g 1.4218 43 3 .9269 40.16 40.16 . . . 1.4263
Previously recorded nTD
1.4108 1.4050 1.4218 1.41806 1.4245
....
T
20' 258 2OI3 20.6'' 20'
d
d%o 0.975l' dZ6a 0.9625' d2"4 0.947513 d20.610.559912 dB1 0.948813 d174
0.9476"
action by Reynolds and Adkins4 suggests that yields troleum ether (45-55'), allowing the temperature of between 55-7070 of primary nitroalkanes may be of the stirred reaction mixture to rise to 40°, and obtained by treating the corresponding alkyl bro- maintaining this temperature ( 3") by occasional mides with an excess of silver nitrite with no sol- cooling, until all the alkyl halide has reacted ((5-8 vent, running the reactions cold for several hours, hours). No effort was made to shield the reaction and then heating under gentle reflux until all the mixture from light. Isolation of the .crude product bromide has reacted (5-6 hours). On the basis of is effected by conventional methods. Most of the analytical data, the products isolated were claimed principal impurity, alkyl nitrite, is removed a t to be mixtures consisting only of alkyl nitrites and reduced pressure through a short column. The residue is treated with cold, concentrated sulfuric nitroalkanes. Considerable doubt has been cast on the validity acid6containing a little urea (to remove any nitrous of these results by the later work of Kornblum5 acid), and the resulting solution is poured over ice. which clearly demonstrates that under the reaction Extraction with light petroleum ether gives a yelconditions used by Reynolds and Adkins, consid- low liquid which is almost pure product, as demonerable amounts of alkyl nitrate, as well as the de- strated by the final distillation. Little, if any, forecomposition products of alkyl nitrite (alcohol and run is obtained, and the entire contents of the pot aldehyde) are formed. (excepting a trace of brown residue and 1-3 ml. of In spite of the low yields obtained, the Victor column hold-up) distils within a temperature Meyer reaction is still the most convenient general range of one degree or less. I n all preparations, method of preparing homologs of primary nitroal- the several fractions taken within this temperature kanes higher than 1-nitropropane. Modifications range had the same refractive indices ( f0.0003). of the original procedure have not, up until now, Experimental included the use of lower reaction temperatures The silver nitrite used by the addition of a which, in the light of Kornblum's work, should be solution of sodium nitritewast o prepared a solution of silver nitrate. an obvious improvement. A number of experi- The silver nitrite was filtered off, washed thoroughly with ments carried out in this Laboratory have indi- ice-water, then with 95% ethanol, air dried in a dish procated that in some preparations, the solvent petro- tected from light for one day, and finally over calcium chloat reduced pressure for one day. It was stored in a leum ether is preferable to a dialkyl ether, the sol- ride brown bottle protected from the light, and was used up t o vent commonly employed. I n addition, there was 10 weeks after it was prepared. some evidence that a t low temperatures (
