~-~-NAPHTHYLXYCLOPROPANECARBONITRILE AND DERIVATIVES
Aug., 1945
1249
zole ring, forming a methylol derivative with a of two imidazole rings. Inasmuch as the second melting point of 210-215' (uncor.). As shown compound is not stable, we are inclined to favor by the melting point, neither of Neuberger's com- the first alternative as the more probable. pounds is the 1 : l mole compound we have isoSummary lated. The conditions of the reaction between 1. The reaction between solutions of i(-)amino acids and aldehyde no doubt determine whether ring closure takes place resulting in pyri- histidine containing one equivalent of sodium dine compounds or whether simple methylol de- hydroxide and various amounts of formaldehyde has been followed by polariscopic and hydrogenrivatives are formed. In our wurk we expect that the 1: 1 mole com- ion measurements. 2. Under the experimentd conditions, histipound is the usual methylol type of derivative of the amino acid. The second mole of aldehyde dine forms two compounds, one of the AF type must link together two molecules of the 1: 1 mole and one of the A2F3 type as the aldehyde concencompound. This could be accomplished by reac- tration is increased. 3. The equilibrium constant of the latter retion with two methylol groups- 2R-NHCH20H HCHO R-NHCH+-CH~-OCHSH-R H ~ Oor action has been calculated. by reaction of the aldehyde with the imino groups GENEVA,N. Y. RECEIVED MAY14, 1945
-.
+
+
[CONTRIBUTION FROM THE WALKER LABORATORY OF THE RENSSELAER POLYTECHNIC INSTITUTE]
1-a-Naphthylcyclopropanecarbonitrileand Some of its Derivatives! BY JOHN B. CLOKEAND THOMAS S. LEARY 1-a-Naphthylcyclopropanecarbonitrile(I) and some of its derivatives have been made as a part of a general study of compounds containing the cyclopropane ring. Compound (I) was prepared from a-naphthylacetonitrile, ethylene dichloride and sodium amide in liquid ammonia
MCHZCN
+
ClCHzCHzCl C-CN
+ 2NaNIIz +
+ 2NaCl-k 2"s
\N/ (111) A'
The use of the cyclopropyl ketimine-pyrroline rearrangement as a test for the presence of the cyclopropane ring in such compounds is based upon considerable work, some of which has alby methods already reported for similar com- ready been r e p ~ r t e d . ~ *The ~ v ~ &ect of heating pound~.~*~ on the rearrangement of the ketimine base (11) The presence of a cyclopropane ring in (I) is may be seen in the top curve of Fig. 1, which was supported by its stability in the presence of po- obtained by the differential thermocouple method tassium permanganate, which has been used for already de~cribed.~It will be noted that the dethe differentiation of cyclopropane from ethylenic composition of the ketimine (11) is evident a t and also by the fact that it will give a about 1 4 4 O , becomes more rapid as the temperaketimine, 1-a-naphthylcyclopropyl phenyl keti- ture rises, and appears to be complete around mine (11), whose cyclopropane structure is dem- 196'. The effect of heating on the l-a-naphthylonstrated by its rearrangement to give the iso- cyclopropylphenylketimmonium chloride (IV), meric 1-phenyl-2-a-naphthylpyrroline (111). CH2CH2C(CloH7)-C(=NHaCI)CsHs, may be seen in Whether (111) has the conventional A2 pvrroline I structure or the A 1structure or whether ihese two the middle and lowest curves of the figure. A structures may be in tautomeric equilibrium has comparison of these Curves with those obtained not been ascertained. with the simpler phenyl cyclopropyl ketimine, CH$2H2C(H)-C(=NH)C!aH~,6 would indicate that (1) The data reported in this paper have been taken from a thesis presented by Thomas Samuel Leary t o the Rensselaer Polytechnic Institute in partial fulfillment of the requirements for the degree of Master of Science in 1939. (2) Kaowles and Clske, THIS JOURNAL, 64, 2028 (1932). (3) Murray and Cloke, ibid., 68, 2014 (1936). (4) Gotkis and Cloke. ibid.. 66. 2710 (1934).
the replacement of the hydrogen (H)in the cyclopropane ring by the a-naphthyl group increases the resistance of the compound to rearrangement. The (5) Cloke, ibid., 61, 1174 (1929).
1250
JOHN
B. CLOKEAND THOMAS S. LEARY
Vol. 67
ported by Coles and Dodds," which gave yields up t o
40
fiO7& These various methods have recently been dis-
cussed by Cambron,lZalthough we did not see his paper riritil after our work was done. a-Naphthylncetontrile.la-The CJoH,CH,CN was prepared from a-chloromethylnaphthalene and sodium cy0 aiiide by the same general procedure as that described in "Organic Syntheses" for benzyl cyanide." However, the -20 product was distilled from an anti-creep flask.16 The compound used by us bhiled from 182-186' a t 12 mm., and was obtained in yields up to 87Yob. 40 1-a-Na hthylcyclopropanecarbonitrile(I) .-A two-liter, 0 t hree-necfed, round-bottomed flask was fitted with a dropping funnel, a reflux condenser provided with a lime tower, x 20 and ap efficient mechanical stirrer which operated through Y v1 a mercury seal. One and a half liters of liquid ammonia .-P O was run into the flask followed by the addition of 2 g. of bright sodium. The stirrer was set in motion and one gram of hydrated ferric chloride was added; then 46 g. 60 (2 moles) of sodium, in the form of small chips, was added over the course of thirty minutes. The reaction mixture was then stirred until the initial deep blue color of the 40 sodium solution had disappeared, denoting the complete conversion of the sodium into sodium amide.a 20 At this stage, 167 g. (1 mole) of a-naphthylacetonitrile was added cautiously to the sodium amide solution over a period of an hour, when 700 cc. of anhydrous isopropyl 0 ether was added a t approximately the same rate as that at which the liquid ammonia evaporated. The reaction mixture was then allowed to attain room temperature, - 20 when it was refluxed gently on a steam pot, giving a dark 50 90 130 170 210 brown viscous liquid. At this point, the reaction mixture was cooled to - 10" Temperature, "C. i n a salt and ice-bath, when 99 g. (1 mole) of ethylene Fig. 1.-Heating curves. chloride was added during the course of an hour to the stirred mixture. The stirring was continued overnight curves also show an aging effect, which we have while the temperature of the bath rose to that of the room, likewise noted in related compounds. a t which time the mixture was refluxed gently for two On account of the slight solubility of the naph- hours. Here water was added, the solution was acidified with acetic acid and the ether layer separated and dried thylketimmonium chloride (IV) in water a t O", over anhydrous sodium sulfate. The distillation of t h e its rate of reaction with water to give the corre- extract in an anti-creep flask gave 118 g. of a fraction of sponding ketone and ammonium chloride was b. p. 153-159" at 3 mm. and a residue. The distillate ascertained in 50% aqueous acetone, giving a value ca,me over as a light ycilow oil which solidified in the reThe crystallization of the compound from 95Y0 for k min.-' of 0.01 in the equation for a uni- ceiver. alcohol gave 100 g. of a pure white crystalline solid of m. p. molecular reaction. Since this value is lower than W-85". Further recrystallizations failed to raise the that ascertained for CHzCHzC(H)-C(=NHzCl)CsHjmelting point of the product, which was obtained in a 52% L--J yield. The sublimation of a small sample of compound also (V) in water a t 0' and in 0.05 M solution, viz., 0.0249, and as Pundt7 found that diphenylketim- gave crystals of m. p. 85'. Likewise, when some of the was ground up with potassium permanganate in monium chloride is more readily saponified by product water, followed by the removal of the permanganate with 50% aqueous acetone than by water at Oo, it bisulfite and the recrystallizoation of the residual nitrile appears that the replacement of the hydrogen from alcohol, it melted at 84 . Anal. Calcd. for CldHIIN: C, 87.01; H , 5.74; X , (H) in the cyclopropane ring of (V) by the anaphthyl group decreases the rate with which the 7 . 2 5 . Fourid: C, 86.6; H, 5.63; X, 7.31. (VI).-When 2 coinpound is saponified by water. The phenyl g. I-a-Naphthylcyclopropanecarbonamide of the nitrile (I) was treated with hydrogen peroxide group, however, has little hy the modified Radziszewiski methodlB a small amount The nitrile (I) may be hydrolyzed to give the of amide was obtained. When this was recrystallized amide, CH2CHzC(CloH7)CONHz(VI), and the acid, from dilute alcohol, white crystals of m. p. 144' were obL A tained. Anal. Calcd. for Cj4HI3SO: N, 6.63. Found: X, ~HzCHzC(CioH7)COzH(VII). (j.32. A poor yield of lower nieltitig coinpoutid was obtained
30
s
-
3
Experimental
dXoromethylnapUhalene.-The
CloH7.CH2C1 which was first employed in our work was made in 35'3, yields by the BlancQmethod as modified by Darzens and L6vy.l0 Later we were more successful with the modification re(6) Cloke, Tlns JOURNAL, 6S, 117 (1940). (7) Pundt, "Thesis," Rensselaer Polytechnic Institute, 1913. (8) Knowles, "Thesis," Rensselaer Polytechnic Institute, 1931 (9) Rlnnr H a l / w c . chiin., W, 313 (1923). 110) T):trzcn. and 1 . 6 ~ C~o t.r i p ! rrnrl , 109,73 (19RG)
(11) Coles and Dodds, THISJOURNAL, 60, 853 (1938). (12) Cambron, Con. J . Research, 17, 10 (1939); Fuson and McKeever, "Organic Reactions," Vol. I, John Wiley and Sons, Inc., 1942, p . 70. (13) Wislicenus, Ber., 88, 507 (1905). (14) Adams and Thal, "Organic Syntheses," Coll. Vol. I , John Wiley and Sons, Inc.. p , 101. (1.5) "Synthetic Orranic Chemicals." Vol. VI. No. 4 , Eastman Kodsk Co..1933. (10) L1iirr.p xn
