Feb. 20, 1960
COMMUNICATIONS TO THE EDITOR
1001
shifts were 1.75, 1.75 and 1.45 p.p.m., respectively, to higher fields than the resonance due to the solvent protons. Cooling the solutions to their freezing point, approximately O0, produced no line splitting. Therefore in diinethylaluminum chloride also, exchange between bridging and nonbridging groups probably occurs readily. In addition to methyl exchange within the dimer, halogens and methyl groups may exchange between molecules in the mixed methylaluminum halides. Trimethylaluminum, aluminum iodide and the intermediate iodomethyl compounds were shown to be in rapid e q ~ i l i b r i u i n . ~A. ~solution of methylaluminum chloride (25%) in dimethylaluminum chloride (75y0) shows only a single sharp proton resonance signal both a t room temperature and a t -60". Since the chemical shift between the two compounds is 12 cycles a t a frequency of 40 Mc. one can estimate3 that the maximum energy for intermolecular methyl exchange is 12 kcal. mole-l. This is similar to that reported, 6-14 kcal. mole-', for the methyl groups in trimethylaluminum.2 In this case, however, it is difficult to envision a mechanism for intermolecular methyl group exchange which does not involve dissociation to monomers. One possible mechanism for the exchange is shown.
pentammine complexes. Some question had been raised as to the existence of nitritoammine cobalt(111) complexes3 so that the isolation of such rhodiuin(I1T) and iridium(II1) coinpounds is further proof of this type of isoinerism, if additional proof is needed. The preparation of these compounds serves as a good example of the usefulness of reaction mechanism studies in the carrying out of preparative work. On the basis of kinetic studies4 it had been proposed that the formation of C O ( ~ H ~ ) ~ O N ~ ~ + takes place by an attack of NnOs or NO+ on the oxygen of Co(NH3)50H2+. That there is no Co-0 bond cleavage in this reaction was substantiated by oxygen-18 experiments.6 I t would therefore appear that a similar low energy reaction path should be available to analogous metal complexes. Then if rearrangement to the stable nitro compound is sufficiently slow, reaction a t mild conditions may yield the kinetic nitrito product. This communication reports just such an observation. At elevated temperatures the reactions of Rh( N H S ) ~ O H and ~ + Ir(h'H3)50H2+with NOz- are reported6 to produce salts of Rh(NH3)5N0~2f and 1r(NH3)5N0z2+. However, a t ice-bath temperatures similar reaction mixtures yield instead salts of Rh(NH3)50N02+and Ir(NH3)50N02+. For example, to an ice cold solution containing 0.67 g. of [Ir(nTH3)5H20]C13 in 20 cc. of water was added / / \ 0.96 g. of solid NaN02. Once the solid had disc1 CH3 C1 solved 1 cc. of 6111HCl was added and a crystalline CH3 C1 CH3 material separated immediately. This salt was \ / collected and washed with dilute ice cold HC1 and AI + acetone, then air dried (yield 0.54 g. or S3%.) /---\ /---\ C1 CHI C1 Anal. Calcd. for [Ir(NH3)50NO]Clz:N, 21.3; H, 3.8; C1, 18.0. Found: N, 21.2; H, 3.7; C1, 17.6. Under similar conditions a good yield of [Rh(NH3)60NO]Cln was obtained. Anal. Calcd. for [Rh(NH3)50NO]Cl2: N, 27.3; (4) A. V. Grosse and 1. M. Mavity, 1. Org. Chem., 6, 106 (1940). (5) IC. S. Pitzer and H. S. Gutowsky, THISJOURNAL,68, 220 H , 4.9; C1, 23.3. Found: N, 27.9; H , 4.7; C1, (1946). 22.7. BAKERCHEMICAL LABORATORY CORNELL UNIVERSITY S. BROWNSTEIN The presence of M-ON0 linkage was verified and ITHACA, XEWYORK,AYD B. C. SMITH its rate of isomerization to M-NOz followed by an DIVISIONOF APPLIED CHEMISTRY examination of the infrared spectra of the solid NATIONAL RESEARCH COUNCIL G. ERLICH compounds. Using KBr disks of the solids, abOTTAWA, CANADA A . W. LAUBENGAYER sorption peaks a t 1460 and 1060 cm.-l are found RECEIVED DECEMBER 16, 1959 which decrease in intensity and simultaneously two new bands appear a t 1420 and 830 ern.-'. These bands have been shown to be due to M-ON0 LINKAGE ISOMERISM : SYNTHESIS AND ISOMERIZATION OF NITRITOPENTAMMINE and M-NO2, respectively, for cobalt(II1) comCOMPLEXES OF RHODIUM(II1) AND IRIDIUM(II1)~ plexes.' The isomerizations of solid [M[NH3]sONO] Clz Sir: in KBr disks have half-lives of 6 hr. a t 48', 3.5 The only reported example of linkage isomerism hr. a t 22" and 2.5 hr. a t 57" for A I = Co(III), for metal complexes is that of nitro (Co-N02) and nitrito (Co-ONO) ammines of ~ o b a l t ( I I I ) . ~Rh(II1) and Ir(III), respectively. Work is now in We wish to report the successful synthesis of Rh- progress on the kinetics and mechanism of forma(NH3)60N02f and Ir(NH3)60N02+salts and their (3) R. Duval, C. Duval and 1. Lecomte, Bull. S O C . chim., [51 14, isomerization to the corresponding stable nitro- 1048 (1947). (1) This research was supported by the United States Air Force through t h e Air Force Office of Scientific Research of the Air Research and Development Command, under Contract No. AF 49(638)-315. Reproduction in whole or in p a r t is permitted for any purpose of the United States Government. (2) S. M. J@rgensen,2. anorg. Chem., 5 , 169 (1893).
(4) R. G. Peareon, P. ?if. Henry, J. G. Bergmann and F. Basolo, THISJOURNAL, 7 6 . 5920 (1954). (R) R . K. Murmann and H. Tauhe, THISJOURNAL, 78. 4886 (1956). (6) S. M . J@rgensen,J . p r a k f . C h e m . , [21 34. 414 (1886); A. Werner and 0. de'r'ries, Ann., 364. 103 (1909). ( 7 ) R. B. Penland, T . J . Lane and J. V. Quagliano, THISJ O U R N A L , 78, 887 (19561.
1002
COMMUNICATIONS TO THE EDITOR
VOl. 82
tion and rearrangement of these nitritoammine complexes. Other metal systems are also being investigated to see whether additional examples of linkage isomerism can be obtained.
It seems very likely that B4Si completely free of BGSi can be prepared by the reaction of the elements in inert atmosphere for periods longer than 2-3 hours a t temperatures below 1 3 i O o and above .Ibout 1200". Such studies are now in progress. DEPARTMENT OF CHEMISTRY BASi is highly oxidation resistant due to the NORTHWESTERN UNIVERSITY FRED BASOLO EVANSTOS, ILLINOIS G. S. HAMMAKERformation of a protective boron-silicon-oxygen 28, 1959 RECEIVED DECEMBER coating which first forms upon exposure of the niatcrial to air a t elevated temperatures. Shapes fabricated from BISi by powder metallurgical PREPARATION OF TETRABORON SILICIDE, BdSi techniques have withstood oxidation in air for over Sir: 100 hours a t 1370" and showed excellent thermal Much effort is being expended today in the search shock resistance, no cracks being observed when for high temperature, oxidation resistant materials. pieces were remorcd from 1x70" to room temperaIn this respect, we wish to report for the first time ture numerous times. The ceramic properties of the preparation in large quantities of a new silicide B&i are now being investigated. of boron, tetraboron silicide, B4Si.l RESEARCH DITISIOY The question of compound formation in the ALI,IS-CIfALVERS 1 l I - G CO EKVINCULTON system boron-silicon has been reviewed recently bT11~x.iA u 1 a - r 1, R'IS by Cline2 in his paper on investigations of BGSi. RFCLIVFCDECEMBER 30, 1959 We have not observed any evidence to date for the existence of BaSi as first reported by Moissan and REARRANGEMENT AND FRAGMENTATION Stock3 and more recently by Samsonov and LatyCARBENOID DECOMPOSITION OF ~ h e v a . Only ~ the presently reported B4Si5 and REACTIONS IN DIAZO HYDROCARBONS the known BGSi have been found. B4Si is prepared by heating the elements in an Sir: Tosylhydrazones of aliphatic aldehydes and keinert atmosphere a t temperatures not exceeding 1370'. -4bove this temperature, BiSi is thermally tones react with bases in aprotic solvents to give unstable ; and the interesting phase transforma- diazo compounds.lS2 The diazo compounds undergo thermal carbenoid decomposition's2 with loss tion B4Si + &Si is now being studied. Amorphous boron of 83 to 95% purity has been of nitrogen to yield olefins derived from hydrogenused, each material giving the identical product. migration and cyclopropanes by intramolecularCrystalline silicon of 98% purity was used. An insertion.' Carbon-skeleton rearrangement did intimate mixture of 500 g. of 86.6% boron6 (40 not occur appreciably even in systems involving moles) and 574 g. of 98% silicon,' +200 mesh, tcrt-butylcarbenes. Rearrangement in carbenoid (20 moles) was placed in a large metallurgical fire decomposition of diazo ketones is quite common, clay crucible of the sillimanite type8 and heated however. It is now reported that cyclopropanecarboxaldein an electrical resistance-type furnace in argon reacts with sodium methatmosphere to 1370' over 4-5 hours, soaked a t hyde t~sylhydrazone~ oxide in diethyl Carbitol or N-niethylpyrrolidone 1370" for 2-3 hours, and furnace cooled. The product consisted of a glassy, slag-like top a t 1SOVja to yield (Equations 1-2) cyclobutene and an inner, black friable core. The latter was (60, 6 i % ) 5 b by riizg-expansion, ethylene (13, 10%) powdered readily and sieved through a 325 mesh and acetylene (13, 10%) by carbon-skeleton fragscreen, the fine portion being shown by X-ray rnedation, and I,&butadiene (4.5, 7%) by rirLgpowder diffraction patterns to be B4Si plus traces
[email protected]*b Methylenecyclopropane was not deof BsSi. Tetraboron silicide is readily distinguish- tected.6h Formation of cyclobutene is thus a able from hexaboron silicide by its X-ray powder prime example of extensive carbon-skeleton rearpattern,6the six strongest lines (d and respective rangement in I: simple carbenoid system.' Deintensities being: 4.1.5, 30; 2.75, 90; 2.67, 100; (1) l,, Friedman and €1. Shechtcr, THISJ O U I G ~ A L , 81, 3512 (19.59); 1.606, 50; 1.582, 30; 1.510, 30. Chemical analysis L. Friedman, Ph.D. Dissertation, T h e Ohio State University, 1839. (2) Similar concliisions have been reached by J. W.Powell and R I . C . of the fine fraction showed B, 60.6%; Si, 36.S%, and hlg, 0.65%. The B/Si molar ratio is 4-28, Whiting, Tiluahedioi?, 7, 305 (1'359) (3) W. E. Bachmann and W. S. Struve, "Organic Reactions," Vu]. I , slightly larger due to the presence of traces of B6Si Chapter 2, John U'iley and Sons, Inc., R e w York, N . Y . , 1942, p. 38. than the expected value of 4.00 required for B S i . (4) All tosylhydrazones p;rre satisfactory analyses. Yields of B4Siof 50-80%, based on boron, have been ( 5 ) (a) T h e tosylhydrazonrs and sodium tnethoxide were added t o achieved by proper choice of reaction conditions. the solvent a t 2 5 " , allowed t o stand for a few minutes and then heated
a.)
(1) T h e writer prefers the nomenclature "tetraboron silicide" in contrast t o "silicon tetrahoride" in consistency with t h e boron-carbon system in which BIC is called "boron carbide." ( 2 ) C . 17. Cline, J . Electrochem. Soc., 106, 3 2 2 (1959). (3) H. Moissan and A. Stock, CornPt. vend., 131, 139 (1900). (4) G. V. Samsonov and V. P. Latysheva, Doklady A k a d . N a u k S.S.S.R., 105, 499 (1955), Chem. Abstr., SO, 7639h (1956). ( 5 ) V. I. Matkovich, A c t a . C ~ y s f (1959). . T h e complete crystal structure of BlSi has been submitted f o r publication. (6) From F. W. Berk and Co., Wood-Ridge, N. J., and Metolsdts Corp., Hawthorne, N. J. (7) Union Carbide Metals Co. ( 8 ) Denver v l r e Slay 60,. I m t v w , &lV.
a t 180O. Decomposition is complete in less t h a n 5 min. (b) T h e tosylhydrazones decornpoie t o hydrocarbons in excellent yields; the percentages reported herein are actual yroduct compositions. (6) (a) T h e hydrocarbons were analyzed, separated, and identified by gas-chromatographic and subsequent infrared methods; comparisons were made with authentic samples. (b) Minor amounts of other hydrocarbons also were obtained. (7) (a) L. Hellerman and R . I,. Garner, THISJ O U R N A L , 57, 139 (1935), report t h a t thermal decomposition of l-diaro-2,2,2-triphenylethane gives triphenylethylene. I t is thus apparent t h a t phenyl migration occurs in triphenylmethylcarhene systems. (b) W.von E. Doering and P. X I . LaFlamme, Teirahedro?z, 2 , 75 (1958), raise the question t h a t reaction of 1,1-dihromocyclopropanes with magnesium QP audiirrn t r i K ~ Y Fallonas roily involve carbenoid procefines,
