NOTES
3418
tion from 700 ml. of 1: 2 acetonitrile-alcohol gave 22.5 g. of purified product, m.p. 161-163'. Three further recrystallizations from the same solvent mixture gave an analytical sample of m.p. 365 mp (e 18,500) and 262.5 (13,100). 167-169", Anal. Calcd. for ClIH10N40,: C, 42.59; H, 3.25; N, 18.06. Found: C, 42.52; H , 3.34; N, 18.00. For comparison, nitrofurantoin has AitxDMF 367.5 mp (e 17,900) and 265 (12,700).
VOL. 29
triphenylsilyllithium to give as the main product, so far isolated and identified, the dodecamethylcyclohexasilane. The mechanism for the formation of this higher cycle may be like that proposed for the conversion of octaphenylcyclotetrasilane to decaphenylcyclopentasilane. Experimental
The Preparation of Some a,o-Dichloro Permethylated Polysilanes HEKRYGILMANA N D SHOHEIINOUE Chemistry Department, Iowa State University, Ames, Iowa Received February 11, 1964
In connection with the synthesis of polymeric organosilicon compounds having special thermal characteristics we have developed a convenient procedure for the preparation of a ,w-dichloro permethylated polysilanes. The method involves essentially the cleavage of dodecaniethylcyclohexasilane by phosphorus pentachloride in sym-tetrachloroethane. This is an adaptation of a technique used for the preparation of a,w-
All reactions were carried out with oven-dried glassware in an atmosphere of oxygen-free, dry nitrogen. The tetrahydrofuran was dried and purified by refluxing for a t least 24 hr. over sodium, followed by distillation into lithium aluminum hydride and redistillation from this hydride immediately before use. The dodecamethylcyclohexasilane was prepared by the reaction of dichlorodimethylsilane with lithium.6 The temperatures reported are uncorrected. Preparation of 1,6-Dichlorododecamethylhexasilane.-A mixture of 23 g. (0.06 mole) of dodecamethylcyclohexasilane, 15.5 g. (0.073 mole) of phosphorus pentachloride, and 100 ml. of sym-tetrachloroethane was heated rapidly to reflux and kept there for 7 min. A t this time the yellow color disappeared completely. The solvent and low-boiling materials were distilled off under 1 mm. of pressure a t room temperature, leaving an oily residue. This was subjected to distillation under reduced pressure. After some low-boiling materials had distilled, a fraction considered to be mainly dodecamethylcyclohexasilane (boiling a t 115-130' a t 1.0 mm.) solidified in the side arm of the distillation flask, and was removed when the distillation was interrupted. Subsequent distillation gave 7.45 g. of a fraction (b.p. 160-162" a t 0.8 mm.) which solidified when cooled (m.p. 43.5"45").
Anal.
Calcd. for C12H36C12Si6: C1, 16.9. Found: C1, 16.4,
17.1.
dichloro perphenylated polysilanes from cyclic perphenylated polysilanes. It enjoys advantages over that described for the synthesis of some a,w-dihalo permethylated polysilanes. For example, 1,Zdihalo tetraniethyldisilane has beeri made by the controlled demethylation of hexamethyldisilane by concentrated sulfuric acid followed by reaction with the corresponding ammonium halide. The 1,3-dichlorohexaniethyltrisilane has been obtained by Kumada and I ~ h i k a w a . ~ The preparation of 1,6-diiodododecamethylhexasilane by the reaction of dodecaniethylcyclohexasilane with iodine has been reported, but no experimental details were given.4 In our studies we have observed that the kind and yield of the a,w-dichloro perniethylated polysilanes vary with the time of reaction and significantly with the nature of the solvent. The 1,6-dichlorododecaniethylhexasilanewas treated with niethylniagnesiuni iodide and phenylmagnesium bromide to give tetradecaniethylhexasilane and 1,6diphenyldodecaniethylhexasilane, respectively. The latter compound was prepared in order to obtain by n.1n.r. the ratio of aromatic to aliphatic protons. Corresponding derivatives were also made for the 1,4-dichlorooctaphenyltetrasilane. Additionally, hydrolysis of 1,6-dichlorododecaniethylhexasilane gave oxadodecamethylhexasilacycloheptane.b An interesting reaction was observed between 1,4-dichlorooctamethyltetrasilane and lithium and was catalyzed by (1) H. Gilman and G . L. Schaebke in "Advances in Organometallic Chemistry." Vol. I. F. G . A . Stone and R. West. Ed., Academic Press Inc., New York, N. Y.,1963, pp. 89-141. (2) M. Kumada, et ai..J . O T ~Chem., . 11, 1264 (1956). (3) RI. Iiurnada and M. Ishikawa, unpublished studies. (4) U. Graf L U Stolberg. Angew. Chem., 76,206 (1963). ( 5 ) Unpublished studies. T h e reaction of dodecamethylcyclohexasilane with phosphorus pentachloride in carbon tetrachloride was first examined preliminarily by R . A. Tomasi.
The yield corresponds to 27.0% of the dimethylsilylene units in the starting material. Five other preparations carried out under similar conditions gave an average yield of 24.3YG. With a reflux period of 30 min. and a 1:2 ratio of dodecamethylcyclohexasilane to phosphorus pentachloride in sym-tetrachloroethane, the products isolated were 1,3-dichIorohexamethyItrisilane (16.7%), 1,4-dichlorooctamethyltetrasilane (24.3%), and 1,6-dichlorododecamethylhexasilane(13 .6Y0). From a prolonged refluxing reaction of a mixture of dodecarnethylcyclohexasilane in sym-tetrachloroethane, but with no phosphorus pentachloride present, the yield of 1,6-dichlorododecamethylhexasilane was 14%. I t is interesting to note that no 1,6-dichlorododecamethylhexasilane was isolated from a reaction with phosphorus pentachloride (1 :1 ratio) with carbon tetrachloride as the medium or other reactant.& Tetradecamethy1hexasilane.-To 3.3 g. (0.0079 mole) of 1,6dichlorododecamethylhexasilane in 60 ml. of tetrahydrofuran was added dropwise 0.0150 mole of methylmagnesium iodide in 25 ml. of ether. Immediately after the addition, Color Test I' was negative. An additional 0.0032 mole of the Grignard reagent was added and the mixture was stirred a t room temperature for 5 hr. After the customary work-up, the oily product was distilled under reduced pressure to give 2.98 g. (48%) of tetradecamethylhexasilane, b.p. 103-105' at 0.7 mm., n 3 0 ~1.5140 (lit.3 b.p. 117-119" a t 1 mm., 15G151" a t 7 mm., n 3 0 ~1.5139). 1,6-Diphenyldodecamethylhexasilane.-To 0.1 mole of phenylmagnesium bromide in 50 ml. of tetrahydrofuran was added dropwise 6.37 g. (0.015 mole) of 1,6-dichlorododecamethylhexasilane in 60 ml. of tetrahydrofuran. The reaction mixture was refluxed for 40 hr. after the addition. The colorless solid product obtained subsequent to the usual work-up was recrystallized a few times from methanol to give 6.89 g. (90%) of compound melting a t 86-87'. Anal. Calcd. for C24H46Si6: c , 57.29; H, 9.22. Found: c , 57.44, 57.64; H , 9.23, 9.29. The infrared absorption spectrum showed the phenyl-silicon, alkyl-silicon, and polysilane groups, but no Si-OH or Si-0-Si bonds. The ratio of aromatic-aliphatic protons by n.m.r. w~ts 1:3.58 and 1:3.62 (calculated, 1:3.60). Preparation of 1,3-Dichlorohexamethyltrisilaneand 1,4-Dichlorooctamethyltetrasilam-A mixture of 15.0 g. (0.043 mole) of (6) H . Gilman and R. A. Tomasi. J . Org. Chem.. 18, 1651 (1963). See also ref. 4 for a preparation using sodium-potassium alloy. (7) H.Gilman and F. Schulze. J . Am. Chem. Soc., 47, 2002 (1925).
NOTES
NOVEMBER, 1964 dodecamethylcyclohexasilane, 19.7 g. (0.0946 mole) of phosphorus pentachloride, and 100 ml. of sym-tetrachloroethane was refluxed for 3 hr. The yellow color disappeared rather rapidly. The oil obtained after removal of the solvent and low-boiling materials was distilled under reduced pressure to give 5.4 g. (corresponding to 25.6% of the initial dimethylsilylene groups) of 1,3-dichlorooctamethyltetrasilane, b.p. 89-92' a t 15 mm., n% 1.4830 ( l k 8 b.p. 82-83" a t 11 mm., n% 1.4852), and 3.42 g. (17.5'%) of 1,4-dichlorooctamethyltetrasilane,b.p. 76-78' a t 1.O mm., n Z o D 1.5035. Anal. Calcd. for CsH24C12Si4:C1, 23.4; MR, 92.23. Found: C1, 23.7, 23.7; MR, 92.70. I n seven other preparations with shorter reflux times ranging from 15 min. to 2 hr., the average yield of the 1,4-dichloro compound was 22.77, based on the dimethylsilylene units in the starting compound, and the yield of 1,3-dichlorohexamethyltrisilane was 17.67,. I n an experiment using carbon tetrachloride in place of symtetrachloroethane, and refluxing overnight, the yields were 22.10/, of 1,2-dichlorotetramethyldisilane,23.870 of 1,3-dichlorohexamethyltrisilane, and 17.1% of 1,4-dichlorooctamethyltetrasilane (all on the basis of dimethylsilylene units in the starting material). Decamethy1tetrasilane.-From a reaction of 0.148 mole of methylmagnesium iodide in 60 ml. of ether and 7.45 g. (0.0246 mole) of 1,4-dichlorooctamethyltetrasilanein 60 ml. of tetrahydrofuran there was obtained, subsequent to distillation under reduced pressure, 3.98 g. of colorless liquid, n% 1.4863. This material was washed first with cold sulfuric acid, then with water, and after drying was redistilled to give 3.11 g. (4870) of decamethyltetrasilane, b.p. 108-109' a t 15 mm., n% 1.4878 .(lit.3,8 b.p. 109' a t 15 mm., n% 1.4872). 1,4-Diphenyloctamethyltetrasilane.-Frorn a reaction between 0.084 mole of phenylmagnesium bromide in 70 ml. of tetrahydrofuran and 4.24 g. (0.014 mole) of 1,4-dichlorooctamethyltetrasilane in 30 ml. of tetrahydrofuran was obtained an oil which upon treatment with methanol gave fine crystals. These on recrystallization from methanol gave 2.3 g. (42.5Y0) of 1,4-diphenyloctamethyltetrasilane, m.p. 59-61 AnaZ. Calcd. for C20Ha4Si4: C, 62.10; H , 8.85. Found: C, 61.93, 61.81; H, 8.87, 8.90. The infrared spectrum was quite similar to that of 1,6diphenyldodecamethylhexasilane. The n.m.r. aromatic-aliphatic proton ratios were 1:2.38, 1:2.36, and 1:2.33 (calculated, 1:2.4). Reaction of 1,4-Dichlorooctamethyltetrasilane. A. With Sodium.--A reaction mixture of 4.9 g. (0.0162 mole) of 1,4-dichlorooctamethyltetrasilane and 0.75 g. (0.0324 g.-atom) of sodium in 80 ml. of toluene wm refluxed overnight subsequent to the addition of a small quantity of sodium-potassium alloy as an initiator. From the reaction mixture wm obtained, subsequent to pouring into absolute ethanol containing a small amount of acetic acid, a solid. This wm washed with aqueous methanol, then with methanol, and dried to give 0.6 g. of powder. An additional 3 g. of material was obtained after removal of the solvents from the mother liquor. These a8 yet unidentified polymeric materials are insoluble in toluene, m evidenced by nonextraction of compounds by hot toluene after 5 days. What may be related polymers were also observed under corresponding conditions starting with 1,3-dichlorohexamethyltrisilane. B. Lithium Catalyzed by Triphenylsilyllithium .-To triphenylsilyllithium (prepared in tetrahydrofuran from 0.31 g. of hexaphenyldisilane and 1.0 g . of lithium) was added 9.47 g. (0.0312 mole) of 1,4-dichlorooctamethyltetrasilane. The addition was effected slowly a t room temperature to maintain the characteristic color of silyllithium; then the mixture was stirred a t room temperature for 2 hr. After filtering through glass wool to remove excess lithium, the filtrate was worked up as usual to give a crystalline solid. This was recrystallized from acetone to give 3.6 g. (49.670) of dodecamethylcyclohexasilane which was identified by mixture melting point and comparison of infrared spectrum with an authentic specimen. The only other product so far isolated is 0.1 g. of a polymeric material. From another experiment in which the brown color of silyllithium was intermediately lost, the only pure product isolated was 36.7% of dodecamethylcyclohexasilane. O.
__~.~___
(8) G. K . Wilson and A . G. Smith, .I Org. . Chem.. 16, 557 (1961).
3419
Acknowledgment.-This research was supported by the United States Air Force under Contract AF 33(616)-6463 monitored by the Materials Laboratory, Directorate of Laboratories, Wright Air Development Center, Wright-Patterson Air Force Base, Ohio. The authors are grateful to Dr. Roy King for the n.m.r. and infrared spectral determinat,ions.
Ozonolysis of Certain Acetylenic Alcohols'
JOSEPH G. CANNON AND LASZLO L. DARKO Laboratory of Medicinal Chemistry, College of Pharmacy, State University of Iowa, Iowa City, Zuwa Beceived J u n e 4 , 1964
An earlier communication2 reported the preparation of some phenylcycloalkylglycolic acids by treatment of the corresponding phenylcycloalkylethynyl carbinols with cold, neutral permanganate. The products were difficult to separate from the reaction mixtures, and yields were poor. A possible alternate oxidation procedure involved ozonolysis of the triple bond. Hurd and Christ3 reported conversion of l-hydroxy-l-ethynylcyclohexane to 1-hydroxycyclohexane-1-carboxylic acid in 52y0 yield, by use of a considerable excess of ozone. When phenylcycloalkylethynyl carbinols were treated with a large excess of ozone, quantitative yields of phenylcycloalkyl ketones were isolated, which suggested that the desired glycolic acids were initially formed, but were rapidly decarboxylated by the powerful oxidizing environment. This interpretation was supported by the observation that benzilic acid is rapidly converted to benzophenone by ozone. When ozonolysis experiments were performed on the acetylenic carbinols I, using carefully controlled and measured amounts of ozone, the glycolic acids were obtained in moderate to good yields. 9-Ethynyl-9fluorenol was also subjected to ozonolysis, forming 9hydroxyfluorene-9-carboxylic acid. Ozonolysis appears to be useful in the preparation of otherwise difficultly obtainable, disubstituted glycolic acids.
R
I When a variation of the method of ozonolysis and of the isolation procedure was attempted, using l-phenyll-cyclopropyl-2-propyn-l-ol,an anomalous product (which was not a carboxylic acid) was isolated. A molecular weight determination of this product gave a value of 179. Elemental analysis indicated a niolecular formula of C12H1,,0,, which is consistent with the molecular weight determination. The compound decolorized permanganate instantaneously, and niicrohydrogenation data indicated the presence of one carbon-carbon double bond. Vigorous ozonolysis of the material gave rise to benzoic acid, and a partial (1) This investigation was supported by Grant M H 07775, National Institute of Mental Health. (2) S. B. Kadin and J. G . Cannon, J . Ow. Chem., 2T, 240 (1962). (3) C. D. Hurd and R. E. Christ, ibid., 1, 141 (1936).
