SYNTHESIS OF THE WURTZITE FORM OF SILICON CARBIDE

SYNTHESIS OF THE WURTZITE FORM OF SILICON CARBIDE. Kenneth M. Merz, and Robert F. Adamsky. J. Am. Chem. Soc. , 1959, 81 (1), pp 250–251...
7 downloads 0 Views 276KB Size
COMMUNIC.~TIOI\TS TO THE EDITOR

2.50

CYTODEUTEROPORPHYRIN' Say:

Warburg degraded cytohemin, the prosthetic group of the saucrstoflubertragende Ferment and of cytochrome-a, to cytodeuteroporphyrin, I , a dicarboxylic acid. Although the yields of methylmaleinimide and hematinic acid obtained on oxidizing I suggested a deuteroporphyrin isomeric with 11, its tribromo-derivative revealed three free positions, not two.2 This and the low carbon in the hematinic acid could be rationalized by assuming that the latter contained its des-methyl derivative. I, if derived from protoporphyrin IX, might then be 11, H for Me' or Mes.

Mea

H Me'

H

Br P

P

Me

L-01. s1

methanol its photograph was identical with that of the third. We fused cytoheiniti concentrates from beef or horse heart with resorcinol and cleanly separated I from some I1 by countercurrent distribution. Its identity with 11, M for Mes, was shown by the m.p (108-209") mixed n1.p. and infrared mull spectruni of the ester, the m.p. (237-239') and mixed m.p. of its copper complex, and by X-ray photographs of these, of the free acid, and of the bromo-ester (1n.p. ca. 280-300 dec.). We also synthesized 11, H for h4ea, and H for NeY. Like TI, H for neither resembled I. (G) N.R C. Postdoctoral Fellow.

DIVISION OF PURECHEMISTRY NATIOSAI, RESEARCH COUXCIL 0 PTAIL'A, CAYADA

RECEIVED JULY

G. S. MAR& D. K. D O U C A L L ~ E. BULLOCK^ S I:. MCDONALD 25, 1958

HBr

e -

m

SYNTHESIS O F T H E WURTZITE FORM O F SILICON CARBIDE

Sir:

We have found that the product of the thermal decomposition a t 1500 =I= 50' of 0.006 mole fraction CH3SiC13in Hz, a t a flow rate of 1200 ml./min., contains crystals of the long-missing wurtzite form 2 - Methyl - 5 - carboxy - pyrrole - 4 - propionic of silicon carbide. The crystals were deposited on acid (from its ester3), m.p. 132" dec. (Found: C the graphite wall of the furnace tube, as clusters of 54.68; H, 5.81; N, 6.99), decarboxylated to 2- transparent blue, green, or colorless prismatic methyl-pyrrole-4-propionic acid, m.p. 80" (Found : needles. The crystals are often stepped, and often C, 62.50; HI 7.13). This with 2-formyl-4-methyl- have globular masses of cubic silicon carbide a t one 5-carboxypyrrole-3-propionic acid4 gave 4,.5'-di- or both ends, or a t the discontinuities in the stepped methyl - 5 - carboxypyrromethene - 3,s' - dipro- crystals. The longest crystals are 2-3 mm. in pionic acid hydrobromide, m.p. 160" dec. (Found: length and up to 0.2 mm. in diameter. Prism, C, 48.87; H, 4.94; N,6.35; Br, 17.98), brominated pyramid, and basal faces are observed on some of to 111 which contained perbromide. I n methyl- the larger crystals The needles show high biresuccinic acid a t 200°, I11 and 3,5,4'-trimethyl- fringence typical of a silicon carbide, and parallel 4,3',5'-tribromopyrromethene hydrobromidej extinction indicating the c axis to be along the gave as expected an alkyl porphyrin removed by needle axis. chloroform extraction, a tetra-propionic acid and X-Ray rotation, Laue, Weissenberg zero and lI3,5-trimethyl-porphin-6,7-dipropionic acid, IT, upper-level photographs have been taken with a H for Me6, isolated as a discrete symmetrical band number of the crystals, using Cu and Mo radiation. ( K = 1.6) by countercurrent distribution between The crystals have been rotated about both the a 0.28% HC1 and ether. Its methyl ester, m.p. con- and c axis, and all results are consistent with the sistently 198-202" (Found: C, 71.31; H, 6.16; N, wurtzite or (in the nomenclature of Ramsdell') the 10.44), formed a copper complex, m.p. 236-240') 2H structure.' The unit cell dimensions, as measand a tribromo-derivative, m.p. ca. 285' dec. ured from a Weissenberg photograph with a crystal (Found : Br, 32.11). X-Ray powder photographs rootating about the a axis, are a = 3.076, c = 5.048 showed that this ester separated as variable mix- A. The space group is PB3mc. The interplanar tures of two polymorphic forms from cold methanol spacings, measured from a powder photograph of but consistently as a single third form from hot. the material, may all be indexed on the basis of the Warburg's I ester, < 1 mg., melted a t 194-198' above cell; however, intensity measurements have (lit. 188",bromo-derivative 298"), undepressed with not as yet been made. The measured lines account the ester of 11, H for Me6. Its X-ray powder photo- for 23 of the 26 possible reflections for the 2H strucgraph represented one of the two first mentioned ture which are accessible with Cu radiation. synthetic forms but after it crystallized from hot The apparent reason for the failure to find 2H silicon carbide until now is probably that nearly (1) Issued as N.R.C. Document. (2) 0. Warburg, H. S . Gewitz and V. Viilker, Z. Nelurjorschung, all investigators have limited themselves to study lob, 541 (1955). We have continued this work at the suggestion of of crystals grown in the commercial silicon carbide Professor Warburg, who kindly provided a specimen of cytodeuterofurnaces, where, presumably, conditions are not porphyrin methyl ester from horse heart. favorable for growth of the 2H form. One fact (3) S. F. MacDonald and R. J. Stedman, C a n . J . Chem.. 33, 458 (1955). now apparent is that a - S i c can be grown a t rela(4) A related sequence is described by H. Fischer, W. Siedel, and tively low temperatures where only &Sic was preL. T. d'gnnequin, A n n . , 600, 156 (1933). (1) L. S. Ramsdell, A m . Mineral., 3 2 , 64 (1947). (5) H. Fischer and H. Scheyer, A m . , 434, 248 (1923).

Jan. 5, 1959

COMMUNICATIONS TO THE EDITOR

251

Treatment of chloromethyldimethylsilane (I, X = C1) with 0.0539 M KOH in 95% ethanol gave an extremely rapid evolution of hydrogen (reaction was complete in less than 15 seconds) which correRESEARCH AND DEVELOPMENT DIVISION sponded to 31y0of the theoretical (31y0solvolysis, KENSETII M. MERZ THECARBORUNDUM COMPANY ROBERTF. ADAMSKY reaction 2) and a precipitate of KCI, 67y0yield (in NIACARA FALLS,NEWYORK good accord with the expected 69% rearrangement, RECEIVED NOVEMBER 10, 1958 reaction 1). The expected disiloxanes, condensation products of the silanols from (1) and (a), also BASE-CATALYZED REARRANGEMENT OF HYDRIDE were obtained in good yields. A similar experiment with iodomethyldimethylsilane (I, X = I) ION FROM SILICON TO CARBON WITH DISPLACEMENT OF HALIDE ION gave 11% solvolysis (2) and 89% rearrangement AT THE MIGRATION TERMINUS (1). Sir: Base-catalyzed solvolysis of phenyldimethylsilane in the presence of bis-(chloromethy1)-tetraThe concept of a pentacovalent silicon intermediate, Si(5), in the base-catalyzed solvolyses of methyldisiloxane gave 100% hydrogen and no retriorganosilanes, whose geometry approximates a duction of the C-CI bonds in the latter. Furthertrigonal bipyramid in which the entering group more, treatment of I (X = C1) with 0.04 M KOH (hydroxide) and the group ultimately displaced in 70% dioxane-30% DzO gave 77% rearrangement (hydrogen) may form an angle near 90" with the product which contained no C-D bonds as shown central silicon,2 leads to the view that the original by an infrared spectrum. The relatively small change in the ratio (1)/(2) four bonds to silicon must undergo considerable excitation and loosening in the conversion of Si(4) to with change from -CHzCI to -CH2I, compared to Si(5) since this change would also involve the con- the large change in solvolysis rates with change versions: Si(4), sp3 bonding, neutral + Si(5), sp3d from RCl to R I (iodides are generally more reactive suggests a common Si(5) bonding, unit formal negative ~ h a r g e . ~ On the by a factor of basis of these considerations i t was of considerable intermediate for (1) and ( 2 ) in which the -CHzX interest to determine whether the hydrogen func- group and the hydrogen function form an angle tions in chloromethyldimethylsilane (I, X = C1) near 90' with the silicon atom (a geometry exceedand in iodomethyldimethylsilane (I, X = I) be- ingly favorable for back attack by the activated come sufficiently activated in the Si(5) complex hydrogen function on the -CH2X group). These formed with hydroxide ion to act as intramolecular interesting rearrangements are receiving further reducing groups for the carbon-halogen bonds, study. (4) A. Streitwieser, Jr., Chem. Revs.. 66, 602 (1956). which must, themselves, undergo considerable acCOLLEGE OF CHEMISTRY AND PHYSICS L. H. SOMMER tivation in the change from Si(4) to Si(5) H E 1>ENNSYLVANIA STATE UNIVERSITY \v. P BARIE,JR. We wish to report the synthesis of these silanes, TUSIVERSITY PARK,PA. D. R. ~VEYENBERG their novel and rapid rearrangement upon treatRECEIVED KOVEMBER 19. 1958 ment with KOH in 95% ethanol, and the proof that these rearrangements are intramolecular

viously thought to be formed. Complete details of the synthesis, and the crystallographic measurements, will be reported later.

-

I

OH

REACTIVITY AT BRIDGEHEAD SILICON. 11. 1-SILABICYCLO [2.2.2;OCTANE

Sir: CH3SiCH2-X 1 cE13

I

1

1

OH

I

----f

CH8~iCHe-X+H+S- (2) I

J

CHI

Reduction of chloromethyldimethylchlorosilane with lithium aluminum hydride in ether solution gave chloromethyldimethylsilane (I, X = C1) in 59y0 yield having b.p. Sl.5' (732 mm.). Found: Si, 25.6; C1, 32.14. Treatment of I with sodium iodide in dry acetone gave iodotnethyldimethylsilane (I, X = I) in 35% yield having b.p. 128' (731 mm.), T Z ~ O D 1.5043, and d20 1.542. Found: Si, 14 1; MRD,38.5.

+

(1) RaSiH f O H S H -+ RaSiOH f Hz f S - , where SH repre= krlsilanel [OH-]. sents solvent (95% ethanol) and -d[silanel/dt (2) L. H. Sommer and 0. F. Bennett, THIS JOURNAL, 79, 1008 (1957); L. H. Sommer, 0. F. Bennett, P. G. Campbell and D. R. Weyenberg, rbid., 79, 3295 (1957). (3) The spad hybrid orbital, although present in PFs and PCls in the vapor state, is uncommon and certainly of much higher energy than spa. Formation of an Si(6) intermediate containing two hydroxide groups is untenable on kinetic grounds, and such an intermediate containing one hydroxide and one solvent molecule is unlikely on steric grounds dud contrary to other evidence nhich cannot be detailed here.

The special geometry associated with bridgehead atoms in medium-sized bicyclic systems has led to an important use of these compounds in the determination of the stereochemical requirements of S s l and SNZdisplacements on carbon.lV2 In previous reactivity a t the bridgehead silicon atom of 1-silabicyclo[2.2.l]heptane (I) was found to be greatly enhanced relative to (C2HJ3SiH. The objective of synthesizing a bridgehead silicon compound that would be almost free of strain (in order to obviate unusual reactivity a t (1) Bridgehead chlorides containing the bicyclo[2.2.llheptane system are inert toward nucleophilic reagents (back attack is not possible) and have greatly reduced reactivity toward halophilic reagents: P. D. Bartlett and L. H. Knox, THISJOURNAL, 61, 3184 (1939); P. D. Bartlett and E. S. Lewis, ibid., 7 9 , 1005 (1050); W. von E. Doering and E. F. Schoenewalt, ibid., 73, 2333 (1951). (2) The bridgehead bromide, l-bromobicyclo[2.2.2]octane shows greatly enhanced reactivity compared to the more strained l-bromobicyclo[2.2.llheptane in solvolysis reactions with or without added Ag+ as a catalyst. However, even the former is less reactive than 1-butyl bromide by a factor of five powers of ten. W. von E. Doering, M. Levitz, A. Sayigh, M. Sprecher and W. P. Whelan, J r . , THIS JOURNAL, 75, 1008 (1953). (3) L. H. Sommer and 0. F. Bennett, I b i d . , 79, 1008 (1957); L. H. Sommer, 0. F. Bennett, P G. Campbell and D. R. Weyenberg, ibid., 79, 3295 (1957).