Lah Nitroalkenes - Reverse Addition
Lah Nitroalkenes - Reverse Addition
Lah Nitroalkenes - Reverse Addition
The flask was placed under a hydrogen pressure of 60 p.s.i. and shaken for one hour at room temperature. The flask was then heated t o 70' and a t the end of the second hour an almost quantitative amount of hydrogen had been taken up' water was added to the reduction mixture, dissolving the precipitate in thereductionvessel. Thecatalyst w a s filtered off, and the filtrate of alcohol, water and spernlidirle trihydrochloride was evaporated to dryness on a steam-bath. The residue was triturated three times with 30-m1. portions of absolute ethanol, then recrystallized from a solution of 2 ml. of concentrated hydro-
chloric acid in 65 ml. of absolute ethanol, yielding 2.1 g. (28.7%) of spermidine trihydrochloride melting a t 256-
~ Calcd, for C,H&Ci3: ~ l . N, 16.5. Found: 9, 16.4, 16.3. The chloroaurate and the picrate derivatives were pr:pared and forlnd to melt at 219,5-220.5~ and 210-2,2 , respectively. Dudley, Rosel1hei1n Starling7 reported a melting point of 220-222' for spermidine chloroaurate and a melting point of 210-212' for spermidine picrate. CORAL GABLES, FLORIDA RECEIVED OCTOBER 1951 10,
258'. ~
THE
A. Reduction of 1-Phenyl-2-Nitropropene1 Introduction It has been shown213 that when compounds of the type ArCH=CRX, where X is a polar group such as CHO, COOH or NO2, are treated with lithium aluminum hydride Via normal addition] the double
bond, as well as the functional group, is reduzed. However experiments4 on the reverse addition of the above reagent to cinnamaldehyde demonstrated that the carbonyl function could be selectively reduced by such a method. I n view of the above findings and of the fact that nitroolefins prepared from aromatic aldehydes conform to the type, ArCH=CRX, where X is NOz, the reverse addition of lithium aluminum hydride to nitroolefins was studied, as a continuation of our earlier investigat i ~ n to correlate this method with those previ,~ ously reported5 for the reduction of nitroolefins. In these latter cases a wide variety of products has been obtained but a t no time has any uncondensed6 p-aryl nitroalkane been isolated. The nitroolefin chosen for study was l-phenyl-2nitropropene-1 .
Experimental
Reduction of 1-Phenyl-2-nitropropene-1 8-Phenylisoto propylamine and N-( p-Phenylisopropyl)-hydroxylamine.-
(1) Condensed from a part of the dissertation submitted to the Graduate Faculty of Fordham University in partial fulfillment for the degree of Doctor of Philosophy. (2) (a) R. F. Nystrom and W. G. Brown, THIS JOURNAL, 69, 1197 (1947); (b) 69, 2548 (1947); (c) 70,3738 (1948). (3) R . T.Gilsdorf and F. F. Nord, J . Org. Chem., 16, 807 (1950). (4) F. A. Hochstein and W. G. Brown, THIS JOURNAL, 70, 3484 (1948). (5) (a) G. Alles, ibid., 64, 271 (1942); (b) 0 Schales, Ber., 68, . 1579 (1935); (c) J. Kindler, E. Brandt and E. Geglhaar, Ann., 611, 209 (1934); (d) B. Reichert and W. Koch, Arch. der Phnrm., 173, 265 (1936); (e) L. Beauvault and A. Wahl; Compl. rend., 134,1145 (1902); (f) A. Sann and A. Schellenberg, Ber., 60,1513 (1917); (9) H.Cerf de Mauney, Bull. soc. chin., [5]7 , 133 (1940); (h) E.P.Kohler and N. L. (7) D . H. Dey, J . Chem. Soc., 18 (1930). Drake, THIS JOURNAL, 46, 1281 (1923). (8) R. L. Shriner and R . C. Fuson, "Identification of Organic Com(6) T h e reduction of w-nitmstyrene to 2,3-diphenyl-1,4-dinitro- pounds," 3rd edition, John Wiley and Sons, Inc., New York, N. V., butane has been 1948,p. 206.
In a three-necked two-liter flask equipped with a mechanical stirrer, dropping funnel and a condenser through which was suspended a low temperature thermometer (openings protected with calcium chloride tubes), a solution of 12.2 g. (0.075 mole) of 1-phenyl-2-nitropropene-1in 300 ml. of absolute ether was cooled to below -30' with an acetoneDry Ice-bath. During rapid stirring, a solution of 4.25 g. (0.112 mole, the calculated amount for the reduction of the nitro group) of lithium aluminum hydride in 100 ml. of absolute ether, was added a t such a rate that the temperature of the reaction mixture was maintained between -30 and -40". Exothermic reaction progressed during the addition and the color of the nitroalefin was discharged. The temperature in the reaction flask was then allowed to fall below -40, the freezing bath was removed and the temperature allowed t o rise t o 15'. Hydrolysis was carried out with 400 ml. of 20% aqueous sodium potassium tartrate. The addition of the first few drops of this solution caused slight reaction indicating that the hydride had not been completely utilized. The aqueous layer was extracted with two additional 50-ml. portions of ether and the combined ethereal extracts, after drying over Drierite, yielded, on rectification, 4.4 g. (44%) of 8-phenylisopropylamine, b.p. 72-74' (4.0 mm.) and 2.6 g. (23%) of N-( p-phenylisopropy1)-hydroxylamine, b.p. 116-118' (4.0 mm.). The latter substance readily solidified. Recrystallization from light petroleum ether afforded a white crystalline solid, m.p. 63-64'. . . Anal. Calcd. for CsH13NO: C, 71.42; H , 8.66. Found: C,71.69; H , 8.42. The 8-phenylisopropylamine isolated was characterized as its hydrochloride and phenylthioureide: The hygroscopic hydrochloride, prepared by bubbling anhydrous hydrogen chloride through a solution of the amine in absolute ether, after recrystallization from an absolute alcohol-ether combination, had a m.p. of 147-148'. In the literature,' the m.p. is listed as 145-147'. Anal. Calcd. for GHt4ClN: N, 8.15. Found: N,8.30. The phenylthioureide, prepared in the usual way,8 had a m.p. of 131.5-132.5' after recrystallization from alcohol. Anal. Calcd. for C I ~ H I ~ N C,S71.07; H, 6.71. Found: ~ : C, 71.30; H , 6.38. The previously unreported N-( p-phenylisopropy1)-hydroxylamine readily reduced Tollens reagent at room temperature. It had the same melting point as, and did not dcpress the melting point of, N-( 8-phenylisopropyl)-hy-
1538
VOl. 74
Nef
6N
1
i-
LiAlHd
6N
6N
CoHaCH?CI-ICHs t SO?
SPT
Complex I I
Complex I1
-A -
__f
SPT
CaHsCH=CCHs
XHOH
' I
C~HSCH~CCH~
I1
SPT
> C6HsCHzCHCHt
"OH $LiAlH4
COHSCHIICHCHJ t
I
SPT
Complex VI
KHz
Fig 1.
droxylamine synthesized according to an earlier method9 : A mixture of 1.5 g. (0.01 mole) of phenylacetoxime dissolved in 25 ml. of 70% alcohol containing 0.365 g. of hydrogen chloride and 0.1 g. of platinum black was shaken under hydrogen until 0.01 mole was taken up. The catalyst was removed by filtration. The volume of the filtrate was increased t o 75 ml. with water and extracted with ether. The aqueous portion was then neutralized with aqueous sodium bicarbonate and the precipitate was collected. On recrystallization from light petroleum ether, there was obtained 0.64 g. (42%) of N-( @-phenylisopropyl)-hydroxylamine, m.p. 63-84', which was not depressed when mixed with the sample obtained from the reverse addition of lithium aluminum hydride t o 1-phenyl-2-nitropropene-l . N-( 8-Phenylisopropyl)-hydroxylamine Oxalate .--\Vhen 0.18 g. (0.02 mole) of oxalic acid dissolved in 10 ml. of absolute ether, was added, with stirring, to a solution of 0.6 g. (0.04 mole) of N-( 8-phenylisopropyl)-hydroxylamine in 10 ml. o absolute ether, there was obtained a white precipitate f of the neutral oxalate which, on recrystallization from an absolute methanol-ether mixture, melted at 175-176'. Anal. Calcd. for C Z O H Z ~ N ~C, ~ 61.21; H , 7.19. O : Found: C, 61.30; H, 6.99. N-( @Phenylisopropyl)-Fnitrophenylnitrone .-The procedure employed here was essentially reported previously. lo There were mixed together 0.3 g. (0.02 mole) of N-( 8-phenylisopropyl)-hydroxylamine dissolved in 6 ml. of absolute alcohol and 0.3 g. (0.02 mole) Fnitrobenzaldehyde in 10 ml. of absolute alcohol. After standing 24 hours, the alcohol was removed under reduced pressure and the residue was recrystallized from absolute ether (freezer). There was thus obtained a light yellow crystalline product, m.p. 119-120'. Anal. Calcd. for Cl&aN&h: C, 67.59; H , 5.67. Found: C, 67.45; H, 5.50. ~-Phcnylisopropylamhe.-A solution of 6.0 g. (0.04 mole) of N-( ,3-phenylisopropyl)-hydroxylamine in 80 ml. of absolute ether was treated with 1.9 g. (0.05 mole) of lithium aluminum hydride dissolved in 150 ml. of absolute ether in the usual manner and refluxed for an additional hour. After cooling, a sample of the reaction mixture gave a positive Gilman-Schulze color test* indicating an excess of the hydride. Hydrolysis was carried out with 400 ml. of 20% aqueous sodium potassium tartrate and after separation of' the layers, the aqueous portion was extracted with two additional 75-1111.volumes of ether. Rectification afforded 2.0 g. {37%) of @-phenylisopropylamine,b.p. 72-74' (4.0 mm.), ident&d by its phenylthioureide. N-( p-Phenylisopro yl)-hydroxylamine and Phenylacetoxime.-When 2.13 g. 6.056 mole, half the amount necessary
(9) G. Vavon and Crajcinovic, Bull. soc. chim., (41 13,231 (1928). (10) C. W.Watt and C . M. Knowles, J . Olg. Chcm., 8,540 (1943).
April 5, 1952
1839
an ethyl acetate-alcohol mixture afforded orange crystals of m.p. 152.5-153.5' which was not depressed when mixed with an authentic sample of the 2,4-dinitrophenylhydrazone of phenylacetone. This m.p. was originally reported as 155.5-156.5' .I4 Anal. Calcd. for CljHlrNaOc: N, 17.83. Found: N, 17.70. Phenylacetone (viaPhenylacetoxime).-A solution of 2.13 g. (0.056 mole, half the amount necessary for the reduction of the nitro group) of lithium aluminum hydride in 100 ml. of absolute ether, was added to 12.2 g. (0.075 mole) of l-phenyl2-nitropropene-1 in 300 ml. of absolute ether, a t -30 to -40', as above. After the temperature of the reaction mixture had been allowed to rise to 15", 300 ml. of 6 N hydrochloric acid was introduced to hydrolyze the intermediate organometallic complex. The layers were separated and the ethereal layer was successively washed with two 100-ml. volumes of 6 N hydrochloric acid. The combined aqueous acidic portions were steam distilled. The distillate was extracted with the original ether layer and two additional 50ml. portions of ether. Rectification afforded 2.9 g. (29%) of phenylacetone, b.p. 73-74' (4.0 mm.). The 2,4-dinitrophenylhydrazone was prepared and recrystallized as previously indicated. Its m.p. was 152.5153.5'. The semicarbazone was prepared according to the usual procedure.16 Recrystallization from 75% alcohol afforded a white crystalline product of m.p. 186-187'. This m.p. was previously reported'c as 187'. Anal. Calcd. for C10H13N30: C, 62.80; H, 6.85. Found: C, 63.25; H, 6.61. l-Phenyl-2-nitropropane. A.-Lithium aluminum hydride (0.855 g., 0.0226 mole, a 20% excess of the amount required for the reduction of the double bond) in 100 ml. of absolute ether, was added to 12.2 g. (0.075 mole) of 1phenyl-2-nitropropene-1 in 300 ml. of absolute ether a t -40 to -50" as described above. The organometallic complex was hydrolyzed with 20% aqueous sodium potassium tartrate. On rectification, there was obtained 4.7 g. (38%) of l-phenyl-2-nitropropane, b.p. 103-104' (4.0 mm.). Anal. Calcd. for CgH1102: C, 65.43; H , 6.71. Found: C, 65.99; H, 6.69. B.-When the above experiment was repeated using as the hydrolytic agent the calculated amount of 1 N hydrochloric acid which was introduced dropwise over the course of three quarters of an hour, 1.5 g. (15%) of phenylacetone and 6.9 g. (56%) of l-phenyl-2-nitropropane were obtained. l-Phenyl-2-nitropropane gave a definite blue-green coloration characteristic of the pseudo-nitroles produced by secondary nitro compounds, when treated according to the conditions of the nitrous acid test.'? The coloration was chloroform extractable. A similar test performed omitting the sodium nitrite, gave only a very faint green color. &Phenylisopropylamine.-By the normal addition, 1.65 g. (0.01 mole) of l-phenyl-2-nitropropane in 50 ml. of absolute ether, was treated with 0.68 g. (0.018 mole, a 20y0 excess) of lithium aluminum hydride in 70 ml. of absolute ether. Hydrolysis was brought about by means of 150 ml. of 20% aqueous sodium potassium tartrate. Upon distillation of the ethereal layer, there was obtained a small amount of oil which gave a phenylthioureide of m.p. 131.5132.5' which was not depressed when mixed with an analyzed sample of the phenylthioureide of p-phenylisopropylamine. Nef Reaction18 on 1-Phenyl-2-nitropropane.-One-half of a gram of l-phenyl-2-nitropropane was dissolved in 10 ml. of aqueous solution containing 0.5 g. of sodium hydroxide. This solution was added dropwise to a solution of 2.5 ml. of concd. sulfuric acid and 16 ml. of water during rapid stirring and cooling with an ice-bath. An oil separated with the characteristic odor of phenylacetone. It was extracted with ether and after the removal of the ether, was treated with semicarbazide hydrochloride and sodium acetate in
(14) W. D. McPhee and E. Klingsberg, THIS JOURNAL, 66, 1132 (1914). (15) R. L. Shriner and R. C. Fuson, "Identification of Organic Compounds," 3rd edition, John Wiley and Sons, Inc., New York, N. Y., 1948, p. 170. (16) P. A. Levene and A. Walti, J . B i d . Chcm.. 90, 81 (1931). (17) V. Meyer, A n n . , 176, 120 (1875). (18) H. B. Hass and E. F. Riley, Chcm. Revs., 31, 399 (1943).
the usual manner for the preparation of a semicarbazone.16 I n this way, there was obtained a white crystalline product which, after recrystallization from 80% alcohol, had the same m.p. as, and did not depress the m.p. of, an authentic sample of the semicarbazone of phenylacetone (186-187'). Absorption Data.-The infrared curves (Figs. 2, 3, 4) were obtained on a Perkin-Elmer 21-Double beam spectrophotometer using Nujol mulls of the samples.
Discussion
The experiments carried out on 1 phenyl-2-nitropropene-1 and its reductive derivatives are outlined in Fig. 1. These reactions demonstrated that the reduction was stepwise and capable of regulation so as to afford the amine, hydroxylamine derivative, oxime or nitroparaffin in various mixtures. While the formation of Complex I was undoubtedly the favored reaction a t -40 to - 50' since the nitroparaffin, l-phenyl-2-nitropropane, was obtained in 56% yields, and while some of Complex I1 must have been formed a t -30 to -40" to account for the phenylacetoxime isolated, neither reaction has been proven as being mutually exclusive of each other a t the temperatures involved. The preparation of its hydrochloride and pKenylthioureide served to characterize the p-phenylisopropylamine formed. The identity of the N-(P phenylisopropyl)-hydroxylamine was proven by its ready reduction of Tollens reagent, the preparation of its oxalate, the nitrone formation and by the fact that it did not depress the melting point of an authentic sample prepared according to the method of Vavon and C r a j c i n o ~ i c . ~ The phenylacetoxime was characterized by virtue of its melting point, hydrolysis to phenylacetone and conversion to the 2,4-dinitrophenylhydrazone. The blue-green coloration, manifested by pseudonitroles which are formed by secondary nitro compounds under the conditions of the V. Meyer nitrous acid test," its reduction to 0-phenylisopropylamine with lithium aluminum hydride and conversion to phenylacetone when treated according to the conditions of the Nef reaction, established the structure of l-phenyl2-nitropropane. The reduction of N-(P-phenylisopropyl)-hydroxylamine to P-phenylisopropylamine with lithium aluminum hydride is novel since we have been able to find no report in the literature on the reduction of a hydroxylamine derivative with this reagent. It is noteworthy that the phenylacetoxime and N-( /3-phenylisopropy1)-hydroxylamine distilled together in the fraction obtained a t 116-11s' (4.0 mm.). The fact that the distillate failed to crystallize readily, led to the assumption that i t might be a mixture. This contention was verified by the chemical separation of the components and supported by the infrared absorption data. Curve H was obtained for N-( P-phenylisopropyl)-hydroxylamine. I t is noticed that there is a. strong absorption band a t 1108 cm.-l which is absent in curve 0. Curve 0, that of phenylacetoxime, demonstrates absorption a t 1666 cm. -l assignable to the C=N bond of the oxime group. Curve D, that of the reaction distillate, shows absorption at both the above frequencies and, in general, has all the peaks present in curves H and 0. The major portion of the phenylacetone formed
April 5, 1952
REVERSE ADDITION OF
LITHIUM ALUMINUM
TABLE I
HYDRIDE TO NITROOLEFINS
1541
Starting material
1-o-Chloro1-(2-Thienyl)-2phenyl-2nitropropene-1 nitropropene-1 l-Phenyl-2l-Phenyl-2o-Chlorophenyl- 2-ThienylKetone butanone pentanone acetone acetone 83-85 75-77 103-104 96-97 5.5 5.0 6.0 5.0 k? Yield, % 62 43 62 52 e 8 Hydroxylamine N-( l-Phenyl-2- N- 1-(Phenyl-2N-( p-o-Chloro3 phenylisobutyl)amyl)2 Propyl)77-78 58.5-60 88-89 72.69 73.70 58.22 * 4 Carbon 72.90 73,37 58.35 w Calcd. 6.52 9.15 9.56 0 Hydrogen 8.67 9.06 6.25 m Found P 148-14gb M.p., "C. 174-176 179-180' 97-98* .* Y Calcd. 64.37 66.18 53.22 48.71 2 Semicarbazones 'C Found 64.15 66.20 53.46 48.53 Calcd. 7.37 7.82 5.04 5.72 Hydrogen Found 7.10 7.66 4.92 5.56 140-141' 129.5-130.5 116-117 92-93 2,4-Dinitrop henyl17.49 Calcd. 17.04 16.34 16.06 hydrazones Found 17.05 16.20 17.48 16.40 I 0 All m.ps. and b.ps. are uncorrected. M. Montagne and M. Roch, Comfit. rend., 218, 679 (1944), list the m.p. as 148T. Thomson and T . S. Stevens, Chenz. Soc., 2607 (1932), report the same m.p. 150". D. Ivanov, Bull. soc. ckim., [5] 4 , 682 (1937), records the m.p. as 82 . Decomposition occurred during the steam distillation and no hydroxylamine derivative was isolated. P. Cagniant, Bull. soc. chim., [5] 16, 847 (1949), lists the m.p. as 194-195'.
1-Phenyl-2nitrobutene-1
l-Phenyl-2nitropentene-1
6.
sponding nitroparaffin, 1-phenyl-2-nitropropane1 is novel since no similar reduction, L e . , the selective reduction of the double bond, has been reported in the aryl series without condensation. The fact that either 1-phenyl-2-nitropropane or phenylacetone can be isolated from the same reaction mixture depending on the type of hydrolysis, indicates that the mechanism of the ketone formation when the lower concentration of hydride was used, is a modified Nef reaction. The transitory blue color which is characteristic of the Nef reaction, and which was observed during the hydrolysis, supports this contention. In view of the findings by Hochstein and Brown4 on the reduction of cinnamyl alcohol to hydrocinnamyl alcohol, i t is possible that the reduction of 1phenyl-2-nitropropene-1 to 1-phenyl-2-nitropropane involves only the double bondlg and is effected by 0.25 mole of lithium aluminum hydride per mole of nitroolefin.
which the steam had been passed, alkaline and extracting with ether. 1-o-Chlorophenyl-2-nitropropene-1.A. Modified Knoevenagel-Walter Synthesis.zO-There were refluxed together for 8 hours, 28.1 g. (0.2 mole) of o-chlorobenzaldehyde, 15.0 g. (0.2 mole) of nitroethane, 1.74 g. (0.02 mole) of n-amylamine and 30 ml. of absolute alcohol. On cooling, the mixture was stored in the refrigerator until the precipitation was complete. The mixture was filtered and the precipitate washed with a few ml. of absolute alcohol. There were thus obtained 20.0 g. of 1-o-chlorophenyl-2-nitropropene-l . An additional amount of 2.0 g. was obtained by concentration of the filtrate and washings, bringing the total yield t o 56%. B. Knoevenagel-Walter Synthesis.z'-When the same quantities of the above reagents were mixed and alldtved t o stand a t room temperature for 2 weeks, and the product was isolated in the same manner, there was obtained a total yield of 24.1 g. (61%) of I-o-chlorophenyl-2-nitropropene-1. Recrystallization from oabsolute ethanol produced bright yellow crystals, m.p. 40 Anal. Calcd. for CgHsClNO: C, 55.00; H, 4.08. Found: C, 54.98; H , 3.88. 2-Thienylacetoxime .-When 2-thienylacetone was treated in the usual mLnnerZ2 the preparation of an oxime there for was obtained, on recrystallization from petroleum ether (b.p. 60-75"), a white crystalline product, m.p. 91-92', which was not depressed when mixed with a sample prepared according to the method of Bouveault and Wahlze by the action of aluminum amalgam on 1-(2-thieny1)-2-nitropropene-1.
Discussion The preparation of the ketones listed in Table I demonstrated the general applicability of this method of synthesizing carbonyl compounds from p-arylnitroolefins. In general, the yields were good, the lowest (43y0) being obtained from the
(20) H. B. Hass, A. G. Susie and R. L. Heider, J . Org. Chcm., 16, 8 (1950). (21) E. Knoevenagel and L.Walter, Bcr., 37, 4502 (1904). (22) R,L. Shriner and R . C. Fuson,"Identification of Organic Compounds," 3rd edition, John Wiley and Sons, Inc., New York. N. Y., 1948, p. 202.
1s42
VOl. 74
relatively uiistable 1-(2-tl~ienyl)-2-nitropropene-l.After 24 hours the precipitate was filtered and rccrystallized The ketones are assumed to arise via the same mech- from absolute alcohol. In this way there was obtained a anism as did phenylacetone in the previous experi- yellow crystalline solid, m.p. 157-158'. Anal. Calcd. C15H1,N~Oa: C, 66.65; H, 5.22. ment, ;.e., by a modified Nef reaction after the se- Found: C, 66.75; for 5.29. H, lective reduction of the double bond of the nitroalN-(6-Phenylethy1)-hydroxylamine Oxalate.-The neutral kene. oxalate was prepared as previously described and recrystalThe substituted hydroxylamines listed in Table lized from a methanol-ether combination. Its m.p. was I all readily reduced Tollens reagent a t room tem- 165.5-167' with decomDosition. Anal. Calcd. for Cl8H2&!~O,j: C, 59.30; H, 6.59. perature. Found: C, 59.85; H, 6.49. The experiments on the synthesis of the previPheny1acetaldehyde.-When 0.855 g. ously unreported 1-o-chlorophenyl-2-nitropropene- 1 20% excess for the reduction of the double (0.0225ofmole., a bond) lithium showed a slightly higher yield for the Knoevenagel- aluminum hydride was treated with 11.2 g. (0.075 mole) Walter synthesis (61% vs. 56%) as compared t o the of w-nitrostyrene via reverse addition a t -40 to -50' arid much shorter reaction time of the modified Knoe- hydrolysis was brought about by 400 ml. of 6 N hydrochloric acid introduced rapidly, there was obtained after venagel-Walter synthesis (8 hours vs. 14 days). steam distillation of the acidic hydrolytic mixture and recThe authenticity of the oxime prepared from 2- tification, 0 to 0.46 g. (0-57,) of phenylacetaldehyde, b.p. thienylacetone was proven by means of the mixed 63-69' (5.0 mni.). When hydrolysis of the intermediate organometallic coniitielting point determination with the sample obplex was effected with the calculated amount of 1 N hydrotained by the reduction of 1-(2-thienyl) %nitropro- chloric acid, added slowly, the ether layer afforded a crude lietie-1 with aluminum anialgain. yellow oil which, when treated according t o the nitrous acid
test, gave an orange-red coloration comparable in shade and intensity to that given by 1-nitrobutane when similarly treated. The crude oil was then triturated with a solution of 4.0 g. of sodium hydroxide in 75 ml. of water to yield a dispersion which, when added to an ice cold solution of 12.5 ml. of sulfuric acid in 80 ml. of water during rapid stirring (Nef reaction), allowed the separation of 0.92 g. (10%) of phenylacetaldehyde, b.p. 63-64' (5.0 mm.). The methone derivative of phenylacetaldehyde was prepared by the usual method.26 Its m.p. was 164-165', the same as reported p r e v i o ~ s l y . ~ ~ The 2,4-dinitrophenylhydrazone of phenylacetaldehydc was prepared as before. Its m.p. was 121-121.5' which was not depressed when mixed with an authentic sample.
Discussion In general, the experiments on the reduction oi o-nitrostyrene Via the reverse addition of lithium aluminum hydride indicated that i t behaved similarly to 1-phenyl-2-nitropropene-1.In most cases, the same derivatives were prepared to characterize the various products. However, a dissimilarity showed itself in the synthesis of the carbonyl derivative insofar as the yield of phenylacetaldehyde was very low. It is believed that here the mechanism of the reduction is still comparable while the ease of hydrolysis differs. The facts that a positive nitrous acid test was obtained for a primary nitro compound and that the yield of phenylacetaldehyde was improved when the conditions of the Nef reaction were more strictly imposed, support this contention. The preparation of phenylacetaldehyde marks the first reliable synthesis of the next higher homolog of benzaldehyde via w-nitrostyrene, since the reported reduction of o-nitrostyrene to the aldoxime followed by hydrolysis to the aldehyde,28has been described as being incapable of r e p e t i t i ~ n . ~ ~ Acknowledgments.-This study was aided ill part by a grant from the Office of Naval Research. The authors wish to express their gratitude to Miss Cecelia Vitiello of the Schering Corp., Bloomfield, N. J., for the infrared absorption data and their
( 2 6 ) E. (1946).
(27) R.H. Lin and R. Robinson, J . C h o n . Soc., 2003 (1938). ( 2 8 ) I,. Bouveault and X. \Val& BirLI. POC. cliim.. [ 3 j 29, 518
(1903). (29) IT. 13. l i a s s and E. I:. Riley, Clrcm. Revs., 39, 399 ( l o i 3 ) .
April 5, 1952
1S43
nterpretation. The analyses were performed by A. A. Sirotenko of this Department. The platinum black catalyst was placed a t our disposal by
Baker and Co., Inc., Newark 5 , N. J., and the nitroparaffins by Commercial Solvents Corp.
NEWYORK58, N. Y.
RECEIVED OCTOBER 1951 17,
DEPARTMENTCHEMISTRY, UNIVERSITY OF
OF KEW HAMPSHIRE]
I n a previous article2 a study of the system ammonium fluoride-potassium fluoride-water was reported. Similar measurements have been made on the corresponding system involving sodium fluoride in place of potassium fluoride. The system is a simple one; no double salts are formed, the solid phases being only ammonium fluoride and sodium fluoride. No evidence was found for the existence of hydrated sodium fluoride in the equilibrium mixture.3 TABLE I
ANALYSES SOLUTIONS AND WET RESIDUES OF
Solutions, weight % NH4F NaF Ha0 Wet residues, weight % NHaF NaF HzO
Experimental
As in the previous work, the Schreinemakers wet residue method was used. The apparatus was the same, with polyethylene cups fitted with motor-driven lucite stirrers. All chemicals were C.P. grade. The solutions and wet residues were analyzed for ammonium ion and sodium ion. Ammonia was determined by distillation from alkaline solution into boric acid and titration with standard acid. Sodium was weighed as sulfate after ignition with sulfuric acid and heating with ammonium carbonate. The low concentration of sodium fluoride in the equilibrium mixture made satisfactory analysis more difficult than with the potassium fluoride system. The solutions were slow in attaining equilibrium. In several cases the identity of the solid phase was checked by X-ray diffraction. using copper radiation with a nickel filter and a Philips 114.59 mm. powder camera. HzO
0.90 4.43 94.67 0.60 45.66 2.45 4.01 93.54 1.33 40.16 4 . 8 0 2.99 92.21 3.22 43.61 5.02 3.24 91.74 3.22 43.61 6.90 3.25 89.85 4.75 37.98 7.26 2.60 90.14 4.97 37.98 10.18 2.33 87.49 5.55 46.32 12.08 2.24 85.68 7.49 42.91 15.10 3.07 81.83 8.06 44.07 19.26 2.39 78.35 ... ... 20.31 2.57 77.12 12.12 38.29 ... 24.45 2.48 73.07 29.45 3.25 67.30 ... ... 33.26 3.15 63.59 12.76 61.99 34.37 3.49 62.14 20.28 41.05 35.97 4.22 59.81 21.76 43.61 41.57 2.88 55.55 14.87 62.83 44.32 0.08 55.60 ...b ... 44.49 .16 55.35 88.40 0.00 44.85 .23 54.92 ...b ... 45.50 .OO 54.50 ... ... a X-Ray analysis showed NaF as solid phase. aiialysis showed NH4F as solid phase.
...
49.59
...
...
...
...
11.60
... ...
X-Ray
NH4F Fig. 1.-The
(1) This work is part of a program of research on inorganic fluorides supported by the Research Corporation and the Atomic Energy Commission. (2) H . M. Haendler and A. W. Jache. THISJOURNAL, 72, 4137
(18.50). (3) Compare N. V. Sidgwick, The Chemical Elements and their Compounds, Vol. I, The Clarendon Press, Oxford, 1950, p. 94. and H. J. Emeleus in J. H. Simons. ed.. Fluorine Chemistry, Vol. I, Academic Press, Inc., N e w York, N. Y., 1950, p. 27.
Results
The results of the analyses of the solutions and
wet residues are given in Table I and representative points have been plotted in Fig. 1.
DURHAM, NEW HAMPSHIRE RECEIVED NOVEMBER 30,
1951