THE REACTION OF l-METHYL-l, 2 - DICARBACLOSD - DODECABORANE (12) AND 1,2-BIS (DIPHENYLPHOSPHINO) CARBORANEUZ) WITH PLATINUM SALTS Thesis for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY RONALD MICHAEL ROGOWSKI 197 l "WL 5- ’3 This is to certify that the thesis entitled THE REACTION OF l—METHYL-l , 2-D ICARBACLOSO- DODECABORANE (12) AND 1 , Z-BIS (D IPHENYLPHOSPHINO) CARBORANE (12) WITH PLATINUM SALTS presented by Ronald Michael Rogowski has been accepted towards fulfillment of the requirements for Ph-D- degree in Chemistry Kc. QA. Major professor Date g‘q‘q’ 0-7689 amoma av T; “MAG & SMIS' BUUK BINDERY INCA , LIBRARY BINDERS ' I mummy}. menial] I THE REACTION OF l-METHYL-I,2-DICARBACLOSO-DODECABORANE(12) AND 1,Z-BIS(DIPHENYLPHOSPHINO)CARBORANE(12) WITH PLATINUM SALTS BY Ronald Michael Rogowski A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1971 ACKNOWLEDGMENTS The author is deeply indebted to Dr. Kim Cohn for his competent guidance and personal interest throughout the course of this research. The author is grateful to Mr. Eric Roach for obtaining 31F and 11B nmr spectral data. The author wishes to express his gratitude to the Department of Chemistry, Michigan State University, and especially to the Inorganic Division faculty members for their tutelage and for the financial aid and experience gained as a Graduate Teaching Assistant. The author also wishes to express his deepest appreci- ation to Dr. RObert N. Hammer and Dr. Wilma N. Bradley for their encouragement and friendship. ii TABLE OF CONTENTS Page INTRODUCTION . . . . . . . . . . . . . . . . . . . . 1 EXPERIMENTAL . . . . . . . . . . . . . . . . . . . . 6 A. Experimental Methods . . . . . . . . . . . 6 B. Materials . . . . . . . . . . . . . . . . . 7 C. Synthesis: General Reaction Procedures . 7 1. Synthesis of the cis-Platinum Phosphine Complexes . . . . . . . . . . . . . . 2. The Reaction of 1-Methyl-1,2-dicarbacloso- dodecaborane(12) with cis—Platinum Phosphine Complexes . . . . . . . . . 14 a. The Preparation of QfPt[P(C6H5)3]2[-C c-CH3]2 . . 14 B10H1o b. The Preparation of SrPt[P(CeH5)2(CH3)]2[-c C-CH3]2 16 BIOHIO c. The preparation of ngt[P(C5H5)(CH3)2]2[-C c—CH3]2 20 BIOHIO d. The Preparation of QfPt[P(CH3)3]2[-C C-CH3]2 . . . 24 BioHio 3. The Reaction of l-Methyl-l,2—dicarbacloso- dodecarborane(12) with gig and trans-Di- chlorobis(tri-gfbutylphosphine)platinum- (II) . . . . . . . . . . . . . . . . 26 4. The Preparation of 1,2-bis(diphenyl- phosphino)carborane(12) . . . . . . . 30 iii TABLE OF CONTENTS (Cont.) Page a. The Reaction of 1,2'bis(diphenyl- phosphino)carborane(12) with Potas- sium Tetrachloroplatinite(11) . . 30 D. Attempted Syntheses . . . . . . . . . . . . 39 1. The Reaction of 1,2—Dicarbacloso-dodeca- borane(12) with Titanium and Vanadium Dicyclopentadienyldichlorides . . . . 39 2. The Reaction of 1 CH3 "C C‘Li + n-C4H10 (C235)20 Vii '— B10H1o BioHio . 0° 2CH3—C C-Ll + PtClzL > PtL -c C-CH ) $07 2 (C2H5 >20 2( V 3 2 + 2LiCl BioHio B10H10 where L is a phosphine ligand. The slurry of lithiomethyl- carborane(12), CHa-CN67C-Li was freshly prepared by charging BlOHlo a nitrogen purged 100-ml three—necked flask with 1-methyl- 1,2-dicarbacloso-dodecarborane(12) (2 mmol) which contained 50 ml of dry diethyl ether. The flask was fitted with a magnetic stirrer, a nitrogen inlet, and an addition funnel with a nitrogen outlet. This flask was then cooled to 0° and a solution of nfbutyllithium (2 mmol) in 10 ml of dry diethyl ether was added over a period of 10 minutes yi§_the addition funnel. The solution was maintained at 0° and was stirred during this addition procedure. The reaction was allowed to proceed for 30 min at 0°. At the end of this time a 1 mmol sample of the phosphinoplatinumdichloro com- plex in a suitable solvent was added over a period of 15 min. (An insoluble platinum complex was added as a suspension in solvent.) The stirred solution was maintained at 0° for 45 min and then was allowed to warm to room temperature (25°) over a period of one hr. That a reaction took place was 9 indicated either by a color change or by the disappearance of the insoluble platinum slurry. Sometimes the mixture was allowed to reflux for 15 min. The solvent was then removed by distillation in vacuo. The products were re- crystallized from appropriate solvents and then dried in_ vacuo . 1. Synthesis of the cis-Platinum Phosphine Complexes The following platinum compounds, gigfdichlorobis- (trimethylphosphine)platinum(II) [QfPtC12[P(CH3)3]2]: gig- dichlorobis(dimethylphenylphosphine)platinum(II) [ngtClz[P(CH3)2(C6H5)]2]: gisfdichlorobis(methyldiphenyl- phosphine)platinum(II) [ngtClz[P(CH3)(C6H5)2]2]; and SEE- dichlorobis(triphenylphosphine)platinum(II) [ngtClz[P(C6H5)3]2] were prepared and characterized ac— cording to previously described methods (42-44). Character- ization data are found in Table I. A typical ir spectrum, as well as 1H and 31F nmr spectra for Sigfdichlorobis- (methyldiphenylphosphine)platinum(II) is shown in Figures 1-3. No nmr was obtained for the gisfdichlorobis(tri- phenylphosphine)platinum(II) complex because the compound was insoluble. aomimom Anew can as.m om.m om.sa ma.va aAamaova oaaisaa Amsv maaimaa ha.m ma.m as.aa sm.as aAamaoVAamova ammiaam ii- sa.m ao.s ma.am as.mm AsmaovaAamova mamimam . Aesv mamuvam ma.s em.v mo.SH am.Sa aflamova Accsomv A.uaqv .m.2 Awesomv Amuoonev Awesomv Amnomnev me me ow... 0e a 0&0: .mmxmamaou «AqvaAauvumlmHo osu mo when coflummflumuomnmzu .H magma 11 .Honszsou mop mcoflumnomnm mzu monocmp x .HHSE HOnsz m mm Gnu NHNAnmUVAnmoUvaNHUumlmHo mo Eduuommm UmumumcH .H musmflm AHIEUV mucosvmum one cow oooH ofimfi oowfi coma coma opbm comm h b p P P n p P p p h r p - 12 o .NHUNQU CH «HNAmmUVAmmwovmamaoumimflo mo Esuuommm HE: ma am .N musmflm .mocouwwmu Hmcuouxm am no poms mm3 mOvm .NHUNQU CH «RNAmmUVAnmoUvauHUumimHo mo Esuuoomm was man .m musmflm c 5.1 m o.ma 0 aces 13 All: mmo ommm 14 2. The Reaction of 1-Methyl-1,2-dicarbaclosododecaborane(12) with cis—Platinum Phosphine Complexes a. The Preparation of C -Pt[P(C6H5) 3]2[ -C C-CHsiz BioH The reaction was carried out as previously described. After the 1ithiomethylcarborane(12) (2 mmol) was prepared, a slurry of gisfdichlorobis(triphenylphosphine)platinum(II) in dry diethyl ether was added to the stirred 1ithiomethyl- carborane(12) solution maintained at 0°. This addition took 10 min. The reaction was then allowed to proceed at 0° for one hr after which time it was allowed to warm to 25° over a one hr period. The disappearance of the insoluble platinum slurry and the appearance of a clear yellow solution was evidence that a reaction took place. The solvent was then removed by distillation in vacuo. The yellow solid, which remained in the reaction flask, was recrystallized from a 70:30 mixture of carbon disulfide and methylene chloride. A small amount of nfpentane was employed to initiate the pre— cipitation of the product from solution. The amber solid was then filtered and dried in vacuo. The 60 MHz 1H nmr (CS2) spectrum exhibited two peaks at a 2.1 (broad singlet, relative intensity 1, due to C- methyl protons of methyl carborane(12)), and 6 7.3 (broad multiplet, relative intensity 5, due to the triphenylphos— phine protons). The 1H nmr spectrum is shown in Figure 4. We could not obtain 31P and 11B nmr spectra because of the low solubility of this product. 15 . amo as 32 amaovfi «A 38699:.“ OHmO Hm am imHo mo EsHuoomm HE: ma o .v ousmflm 16 The infrared spectrum obtained as a Nujol mull and shown in Figure 5, exhibits characteristic absorptions at 740 (m) and 2550 (s) cm—l. The absorption at 740 cm"1 is attributed to the B-H cage structure and the one at 2550 cm- is ascribed to the 8-H stretching mode. Other ab- sorptions appear at 510 (s), 545 (ms), 670 (s), 993 (w), 1020 (mw), 1090 (ms), 1115 (w), 1180 (w), 1305 (w), 1570 (w), and 2720 (w) cm-l. The bands tentatively assigned to Pt-Cl stretches present in the starting material (300, 320 cm-1) were absent from the product. The analytical data for this compound and the subsequent alkylphenylphosphine platinum complexes of methyl carborane(12) are represented in Table II. The compound was slightly soluble in carbon disulfide, methylene chloride, and benzene. It is stable in air. b. The Preparation of Q-Pt[P(C6H5)2(CH3)]2[-CVc-CH3]2 B10H1o The reaction procedure was identical to the one pre- viously described for the preparation of the triphenylphos- phine platinum complex. An amber solid was isolated. The 1H nmr spectrum (CS2) exhibited three peaks at o 1.9 (doublet J = 12 Hz, due to the methyl protons attached to phosphorus), 2.05 (singlet, due to C-methyl protons of the carborane), 7.3 multiplet, due to the phenyl protons of the phosphine ligand). The overlap of peaks at a 1.9 and 2.05 made the accurate determination of relative peak areas im- possible. The 1H nmr spectrum is shown in Figure 6. This 17 cmm F .Honsz on map mcoflumuomnm och monococ x OHmOHm .HHSE HOnsz m mm 5.5 «HmAnmmova«Ammoiuboivumimao mo Ednuoomm cmHmHMCH AHIEUV mocmskum ccv com com cccH chH cch p _ P . - F i1 coca cch comm p . L. 000m . .m musmflm comm - i1 18 A.asoomav coaumma .. How in mm.m u- .ws.am aAamova A.asouasv maaimma ans aw» us.a ma.m sw.mm mo.mm AamaovaAamuva A.asooaav soauoom Haw aoa As.m ms.m sa.os om.ms «AsmaUVAamova A.aeoomsv maaimaa moss macs so.m as.m sm.se ss.me aAamaoVa .m.2 Apcsomv Amuoonav Accsomv Amuoonev Accsomv Awuomsev a .3.2 .3.2 me me as us OHmOHm .mmonmEOU «AqvnflmmUiUbUivumimHo 93 mo moon coflumnaumuomumco .HH manna 19 . «mo 5 azamovaflamaoEaA£8690; m.H .mc.N o OHmOHm am I umimflo mo Esnuoomm HE: ma m.h o .c Tuscan 20 product was too insoluble to obtain 31P nmr Spectral data. The 118 signal consisted of a broad doublet at 629.4 up- field from B(OCH3)3. The infrared spectrum of this compound mulled in Nujol is shown in Figure 7. The characteristic absorptions due to the carborane were present, (740 and 2550 cm-1) while the Pt-Cl stretches were absent (320 cm—l). The solubility of this complex was similar to that described for the triphenylplatinum carborane complex. See Table II for the analytical data obtained from this complex. c. The Preparation of g-Pt[P(C6H5)(CH3)2]2[-CVc-CH3]2 BioHlo The reaction was carried out in an analogous manner to that previously described, however, the reaction mixture was allowed to reflux for 15 min at 36°. The 1H nmr spec- trum (C82) exhibited three peaks 5 2.0 (doublet J = 11 Hz, due to the methyl protons attached to phosphorus), 2.1 (singlet overlapped with a 2.0 doublet, due to the C-methyl protons of the carborane), 7.4 (multiplet, due to the phenyl protons of the phosphine ligand). The nmr spectra are shown in Figures 8 and 9. The 31P nmr (CH2C12) 5 120.4 (broad singlet). The 118 nmr (CH2C12) 5 21.4, 22.6, 23.8 (three broad absorptions, intensity 1), 27.7 (broad singlet, intensity 3), 29.4 (broad singlet, intensity 4), 31.0 (broad singlet, intensity 4), 33.4 (broad singlet, intensity 3). 21 camcam .Hohsz ou opp mCOHumHOQO ozu monocmc x .HHSE Hohsz m mm can «mmmUiUKWVUIH«MAmmuvuAmmmuvmmumimflo mo Esuuoomm commumaH .b Tuscan AaiEov mocosvoum cmm ccv com com cccH ccmH cch cccH cccm ccmm cccm ccmm . b F b ~ _ - _ p _ p P 22 OHmOHm .mmo CH m_mmUIUN©VUIHm_mAmmUVAnmmUVmHDmimHo mo mnuommm HEC me can mam .m musmflm am AAA v.cmH o 23 .mocmnmmmn Hmcnmuxo cm mm com: mmB «Ammuovm OHmOHm .mHUmmU CH mammUlUbulammmnmmUVAmmcUVmHumlmHO m0 EDHuomflsm HES mHH o.am e.ea e.aa am Tam _e.am_ mam _ v.3 a AIIIII . . _ _ . "Ammoovm .m onsmflm 24 The infrared spectrum of this compound mulled in Nujol, is shown in Figure 10. Again the two absorptions character- istic of the carborane cage are present (740 and 2550 cm-l). See Table II for the analytical data. This complex was slightly more soluble in organic solvents than the previous phenylphosphineplatinum carborane complex, but was less stable in air, presumably because of hydrolysis. d. The Preparation of ngt[P(CH3)3]2[-C C-CH312 B10H1o The general reaction procedure described previously was employed. After allowing the reaction vessel to warm to 250 for two hours, some of the insoluble trimethylphosphine platinum complex began to dissolve. After the solution was allowed to reflux for 15 min, it became cloudy and slightly yellow in color. After removal of the diethyl ether sol- vent by distillation in vacuo, a yellow residue remained in the flask. A buff colored solid was isolated after recrys— tallization. This solid decomposed when moist. Precise analytical data could not be obtained. The 1H nmr spectrum (C82) consisted of two resonances, a 2.0 (broad overlapping singlet and doublet, due to the methyl protons of the phosphine and the C-methyl protons of the carborane). After two hrs the upfield peak decreased and a new resonance 6 3.5 [broad singlet, due to the start- ing material, 1-methyl-1,2—dicarbacloso-dodecarborane(12)] appeared. 25 .Hoflsz ou osc mcoflumuomnm on» monocmp x OHmOHm . .HHSE Hoflsz m mm can «HmmUiukmvoiH«HmfimmUVAnmmovmuumimHo mo Esuuommm cmHMHMCH .cH musmflm AaiEov mocmsvmum cmm com com com cccH ccmfi cch cccH ccwH ccmm cccm ccmm . p p _ F F _ _ _ P _ p 3 26 The infrared spectrum of this compound exhibited the characteristic absorptions of the carborane cage. The ap- pearance of strong bands at 1630 and 3400 cm"1 (OH vibra- tional modes) suggests that hydrolysis and decomposition of the product had occurred. The infrared spectrum is shown in Figure 11. 3. The Reaction of l-Methyl-l,2-dicarbacloso-dodecarborane(12) with cis and trans—Dichlorobis(tri—Efbutylphosphine)- platinum(II) The gig and trans isomers of dichlorobis(tri—nfbutyl- phosphine)platinum(II) which were used for these reactions were prepared and characterized according to the method described by Kauffman and Teter (45). M.P. 140-1440 and 65—660 for the gig and t£§n§_complexes respectively (lit. (45) 142—144° and 65-66°). The reaction procedure described in the previous section was employed. Because the trans_ complex was soluble in benzene, benzene rather than diethyl ether was employed as a solvent. After addition of the yellow trans complex to the reaction flask a small amount of white precipitate appeared. The reaction was allowed to warm to 23°. The benzene was removed by distillation in vacuo, and a golden solid remained in the reaction vessel. This solid was dissolved in 70:30 carbon disulfide-methylene chloride mixture and then recrystallized from nfpentane. The 1H nmr spectrum (C82) exhibited three resonances, d 1.0 and 1.6 (broad multiplet, due to the nfbutyl groups 27 .Hoflsz on mac mcoflumnomnm mzu monococ X OHmOHm .HHSE Hohsz m mm con «HmmUIUKWVUIHNHmAmmuvmgumimflo mo Esnuoomm pmumumCH .HH ousmfim Eov mocmsomum .omm wow owo oww oowH oomH ooWH oowH oowH comm fiwom owmm 3 28 on the phosphine), and 2.1 (singlet, due to C-methyl pro- tons of the methyl carborane). The infrared Spectrum of the compound mullen in Nujol, is Shown in Figure 12. The characteristic absorptions (725 and 2580 cm-1) of the carborane cage were detected. The peaks at 1630 and 3400 cm.1 are attributed to OH stretching modes which arise from the hydrolysis of this complex. AQQI, Calcd. for C30H80B20P2Pt: C, 39.42; H, 8.76. Found: C, 39.75; H, 8.30. Molecular weight determined in benzene was 880 (theoretical 913). The melting point was 185 i 5° (decomposition). The compound was partially soluble in methylene chloride and carbon disulfide. It appears to be moisture sensitive and decomposition was evident upon exposure to air for 15 min. The gig isomer of dichlorobis(tri-nfbutylphosphine)plati- num(II) was prepared according to the method of Kauffman and Teter (45). The general reaction procedure described for the Ergg§_isomer was used. The white 213 isomer was insoluble in benzene and in most other organic solvents, thus it was added as a slurry in benzene. Upon warming to 250 the white slurry disappeared and the solution turned yellow. After distillation in vacuo, a yellow oily liquid remained in the flask. After recrystallization of the liquid from a 30:70 mixture of carbon disulfide and nfpentane a pale yellow solid remained. Attempts to obtain reproducible nuclear magnetic reso- nance data from this product were unsuccessful. The complex 29 oHOflsz 0U QDU mCOHUQHOWQM $5..“ mmuocmmu um OHmOHm .HHSE Honsz m no son «HmmUIUN®VUi_«HnAmmeovaumimcmnu mo Eduuommm commumcH .ma onsmflh AHIEUV mucosqoum cmm oov coo cow oooH coma coma coca oowH ooom comm ooom comm ooov - . p L p p b i. P p . . L - T N 30 proved to be insoluble in most organic solvents. The 1H nmr spectrum (C82) gave three broad peaks at a 1.0, 1.6, and 2.1. The Shifts were approximately the same as for the trans complex. The growth of an absorption peak at a 3.5, sug- gests the product undergoes decomposition to the starting methyl carborane(12). The infrared Spectrum for the £35 compound Showed the characteristic carborane cage absorp- tions as well as OH stretching modes which are attributed to the hydrolysis of this compound. Several attempts to get precise analyses and melting points on this gig compound were unsuccessful. 4. The Preparation of 1,2- bis(diphenylphosphino)carborane(12) This bidentate carborane ligand was prepared according to the method of Alexander and Schroeder (46). The air- stable white solid was recrystallized from petroleum ether m.p. 217-218° (lit. (46) 2190). Infrared, 1H, 319, and 113 nuclear magnetic resonance, and mass spectral data, not pre- viously reported for this compound, are shown in Figures 13- 15 and Table III. flat—l. Calcd. for C26H30B10P2: C, 60.93; H, 5.86; P, 12.11. Found: C, 60.82; H, 5.82; P, 11.10. a. The Reaction of 1,2-bis(diphenylphosphino)carbor- ane(12) with Potassium Tetrachloroplatinite(II) Two complexes were obtained when 1,2-bis(diphenylphos- phino)carborane(12) was allowed to interact with potassium 31 omm cow - b .Hohsz on 05c mcoflumHOQO map monocmc N OHmme .HHDE Homsz m mm com «Ammoovmiokmvoimmfinmoov mo Esuuowmm UmumumCH .MH musmflm ooo oom . OOOH L A Eov mocwsvoum HI oomH _ oovH p oooH omwfi ooom Cmmm ooom - — - 32 .mocwuowou Hmchuxo cm mm cow: mmB Ammoovm 0.3%on .mmo CH «mammovmiumwwoimmnmmmov mo Eznuoomm ch maH .VH ousmflm “Lem H.om nAri m m . moo m 38— Sea . 3:0 A V iiiiillllll'lllll‘l 33 mEB OHNOHm .mmo CH «AmmmovaUNWVUimmAmmouv Mo mnuommm HE: ma can mam .mH musmflm am .AI mm.» .owé e O m. we a A m so; .34 *- Table III. Mass Spectrum of (C6H5)2P-CV§7C-P(C6H5)2 . . BlOHlo (Ionization Voltage 70 eV) m/e Relative Intensity Assignment 512 100 [(C6H5)2P-C ' c-P(C6H5)2]+ BioH1o 435 37.5 [(C6H5)P'C C'P(C6H5)2]+ BioHio 404 15.5 [(CSH5)-CVC-P(C6H5)2]+ BioHio + 327 36.5 [(CsHs )2P-CVC 1 BioHio 262 14.5 [P(C6Hs)3]+ 250 15.0 [(CaH5)P=CVC 1+ BioHio 216 12.5 ? + 185 45.0 [P(C6H5)2] ~X- Boron cage compounds appear as a broad distribution due to boron isotopic effects. The center of each distribution is the m/e value reported. 35 tetrachloroplatinite(II). Both a 2:1 and a 1:1 complex of ligand to metal salt were prepared. To prepare the 2:1 compound, a 0.415 g (1 mmol) sample of K2PtC14 was dis— solved in a 60:40 methanol—water mixture and to this mag— netically stirred solution a 1.03 g (2 mmol) sample of the 1,2-bis(diphenylphosphino)carborane(12) was added. No immediate reaction was evident so the solution was stirred for 12 hrs at about 30°. After 12 hrs the solution was darkened by the formation of a gray-violet precipitate. This solution was cooled, filtered, and then washed with three 101mlponionscfi cold 50:50 methanol-water mixture. A white residue and a violet precipitate were observed. The white substance was removed by washing the filtrate with benzene. The violet solid was insoluble in carbon di- sulfide, benzene, methylene chloride, diethyl ether, chloro— form, and nfpentane, as well as HCl. A sample of this solid was sublimed in vacuo at 250°. The small amount of white solid sublimate which was collected was shown by means of its 1H nmr Spectrum to be the starting bisphosphino- carborane(12) ligand. The remaining violet solid did not appear to be altered by the sublimation process. A nuclear magnetic resonance Spectrum could not be ob— tained due to the extreme insolubility of the violet solid. An infrared spectrum of the solid mulled in Nujol is Shown in Figure 16. The absorptions at 750 and 2540 cm"1 are characteristic of the bisphosphinocarborane(12) ligand. There was no Pt-Cl stretch observed in the appropriate III. I I i 'Vi.i 36 .HOnsz ou moo mcoaumuomnm och monococ x OHmOHm 1:58 Hohsz m mm con NHUDQNHNAnmoovmiuwoimfinmmo: mo Enuuommm commumcH .oH wusmflm omm com com _i P _ oom p oooa P AHIEUV mocmsomnm ccmH covH coca coma oflcm ommm oflom » p b - 37 region of the infrared Spectrum (280 to 310 cm-l). This fact together with the chemical analyses suggests that the 2:1 complex of ligand to platinum salt was formed. Anal, Calcd. for C52H60B20012P4Pt: C, 48.37; H, 4.65; CI, 5.50. Found: C, 47.66; H, 4.48; Cl, 6.57. No molecular weight data were obtained due to the insolubility of this compound. The 1:1 ligand to platinum complex was first prepared by accident in an attempt to prepare the 2:1 complex. A 1 mmol sample of KthCl4 was dissolved in a 60:40 mixture of ethyl acetate and water. To this stirred solution, a 1 mmol sample of the 1,2-bis(diphenylphosphino)carborane(12) was added. No immediate reaction was observed. The reac— tion flask was cooled to 20°. After Six hours a white pre- cipitate was observed in the reaction vessel. The mixture was filtered; the precipitate was washed with water and then dried in vacuo. This gray~white solid was insoluble in carbon disulfide, benzene, methylene chloride, diethyl ether, chloroform, and nfpentane. I was unable to recrys- tallize the product and sublimation in vacuo at 250° left the product unchanged. No nmr Spectral data were obtained due to the insolu- bility of the complex. The infrared Spectrum of this com— plex mulled in Nujol is shown in Figure 17. The bands characteristic of the bis(phosphino)carborane(12) at 750 1 and 2540 cm- are again present. There are also bands at 280, 290, and 310 cm‘1 which we tenatively assign to Pt-Cl 38 was oos 000 L oom — ofioH .Honsz on can mcoflumuomflm emu monococ X OHEOHm AaIEoV mucoswmum ome sovfi cwofi omwfi owom oomm 2:58 Homsz m mm 23H «HUDQNAmmoovmiOWUimmflmmmov mo Eduuoomm UmHmHMCH .bH ousmflm ooom 39 stretches. This evidence together with the analytical data support the suggested 1:1 formula for this complex. Anal. Calcd. for C26H30810C12P2Pt: C, 40.10; H, 3.89: Cl, 9.11. Found: C, 39.83; H, 3.85; CI, 8.89. No molecu- lar weight data was obtained due to the insolubility of this product. D. Attempted Syntheses 1. The Reaction of 1,2—Dicarbacloso-dodecaborane(12) with Dicyclopentadienyldichlorides of Titanium and Vanadium The Similarity of carborane to benzene suggested that (C5H5)2Ti(-C c-) or the dimer, (C5H5)2Ti(-C C)2Ti(C5H5)2, B10H10 B10310 might be prepared through a reaction analogous to the pre- viously reported (47) reaction. THF (C5H5)2TiC12 + 2Li-C6H5 > (C6H5)2Ti(C5H5)2 + 2LiC1 Specifically, a 1 mmol sample of 1,2-dicarbacloso-dodeca- borane(12) was dissolved in 50 ml of dried tetrahydrofuran solvent. A 2 mmol sample of nfbutyl lithium in pf hexane was Slowly added,over a period of 5 min, to the stirred reaction vessel maintained at 0°. After 30 min the white slurry of the dilithium salt of carborane appeared. The red titanium complex (1 mmol) was added over a 10 min period as a suSpension in THF. Immediate evidence of reac- tion was noted by a color change from red to dark green. After allowing this stirred mixture to react at 0° for 40 30 min, the system was allowed to warm to 25°. The solvent was removed by distillation in vacuo and a dark green oily residue remained in the flask. Attempts to recrystallize this oily product from benzene, methylene chloride, and £7 pentane proved futile. The 1H nuclear magnetic resonance spectrum of the resi- due taken in THF revealed peaks attributed to the cyclo— pentadienyl protons (o 6.5) as well as the characteristic peak (5 3.52) of 1,2-dicarbacloso—dodecaborane(12). Several other minor peaks attributed to the unreacted titanium complex and decomposition products were observed. After several hours in solution the 1H nmr peak attributed to the starting carborane (5 3.52) increased in intensity. Pre- cise analyses were never obtained. This reaction was repeated several times with variations in the reaction time (30 min to 3 hrs) and temperature (0° to 36°). All attempts to isolate a stable fraction from the oily residues proved futile. The vanadium analogue to titanium dicyclopentadienyl- dichloride was then investigated. The reaction procedures were identical to those employed in attempting to prepare the titanium complex, however, vanadium dicyclopentadienyl- dichloride was used as atreactant and THE as a solvent. Again a dark oily residue was the final product and all at- tempts at recrystallization proved unsuccessful. The ability of titanium to polymerize may be responsible for the difficulty in isolating a characterizable compound. 41 The two reactive sites of 1,2-dicarbacloso-dodecaborane(12) may enhance the chance of polymerization. Thus polymers of the type (C5H5)\\ (CsHs) \;.:’,(C5H5) T1‘\‘ ’//, 1 ,/// <3‘§7C ‘\‘~Cxé7c - BIOHIO BioHio .41“ may form upon reaction. To avoid this problem I used 1—methyl-1,2-dicarbacloso—dodecaborane(12) (40) instead of 1,2-dicarbaclosos—dodecaborane(12). In this moiety, one of the active C-H Sites has been blocked by the addition of a methyl group. This would eliminate the possibility of a polymer involving the carborane linkage. Reaction procedures identical to those previously described (pp 7,8) were employed. After addition of titan— ium or vanadium dicyclopentadienyldichloride the solution immediately turned from the original color of the starting material to a dark green. Removal of the solvent (THF) by distillation in vacuo left an oily residue. All atte ts at recrystallization were unsuccessful. The nuclear mag— netic resonance spectrum exhibited the characteristic peaks ascribed to the starting carborane material. The use of diethyl ether or benzene as the solvent resulted in no noticeable difference in the final product. No acceptable analyses were obtained on the system. 42 2. The Reaction of l-Methyl-l,2—dicarbacloso—dodecaborane(12) with Palladium and Nickel Chlorides Following the successful preparation of platinum- carborane compounds, reactions with palladium analogs of platinum were attempted. The starting materials for these reactions, gisfdichlorobis(triphenylphosphine)palladium(II) and transfdichlorobis(tri—dimethylaminophosphine)palladium(II) were prepared and characterized according to previously re— ported methods (48, 49), m.p. 260—2650 and 1200 (decomp.) (lit. (48,49) 260—270° and 119—120° respectively). The pro- cedures used to make the palladium compounds were identical to those described previously (pp 7—8). In a typical reac- tion, a 2 mmol sample of the lithium salt of methyl carbor- ane(12) was prepared under nitrogen. This was followed by the addition of 1 mmol of the appropriate palladium complex in diethyl ether. When Sigfdichlorobis(triphenylphosphine)- palladium(II) and transfdichlorobis(tri-dimethylphosphine)- palladium(II) were employed, no immediate reaction was ob- served but dissolution of the added palladium complex occur- red after the reaction flask was allowed to warm to 25° for one hr. The solution underwent a gradual color change from yellow to brown to black over this time period. An oily black residue remained after removal of diethyl ether by distillation in vacuo. These residues were partially sol- uble in carbon disulfide. A 1H nuclear magnetic resonance Spectrum revealed peaks which can be attributed to the starting material (5 2.0 and 3.5) and other minor peaks, 43 (5—10% relative intensity) probably a result of the decompo- sition of the unreacted palladium complexes, which increased in intensity upon standing. Attempts to recrystallize these products were unsuccessful. A change of the solvent from diethyl ether to benzene had no observable effect on the reaction. The analogous nickel(II) system was investigated X13 the preparation and characterization of the dichlorobis- (triphenylphosphine)nicke].CLI) (50,51) (m.p., 245-250°; lit. m.p. (51), 247—250°) and the subsequent interaction of this product with a 2 mmol sample of 1—methyl—1,2-di— carbacloso-dodecaborane(12). The reaction procedure used was identical to that described previously (pp 7—8). There was no evidence of immediate reaction, but upon warming to 25° the solution gradually darkened. The solvent (benzene) was removed by distillation in vacuo and a black residue remained in the flask. A 1H nmr spectrum of the black residue in carbon disulfide exhibited peaks which can be ascribed to the starting methyl carborane and a very broad resonance (6 7.3) due to the paramagnetic nickel complex. No further attempts at synthesis of nickel(II) complexes were carried out. DISCUSSION Sigma Bonded Methyl Carborane(12) Complexes Although no o-bonded transition metal complexes of methyl carborane(12) were reported at the start of this study, during the time these investigations were underway, three reports of Similar complexes appeared (32-34). In these reports,however, detailed spectroscopic data were not presented. In this study the results of a systematic investigation of o—bonded complexes will be reported, together with spectroscopic data. I was unable to obtain sufficient data, because of experimental difficulties, to fully characterize the influence of the carborane cage on the metal-carbon bond. The possibility of preparing o-bonded complexes by the interaction of gig-platinum phosphine compounds with methyl carborane(12) was established. . . 00 sis-PtclziP(CH3)nPh3_n12 + 2CH3-C C-Li (C2H5)20 > 310310 cis-Pt[P(CH3)nPh n]2[-c C-CH3]2 + 2LiCl — 3- BioHio (where Ph is a phenyl group). 44 45 Elemental analyses together with molecular weight data suggest that all the complexes are monomeric and non—ionic when dissolved in C82. The 1H nmr Spectral results indicate that bis(alkyl- phenyl)phosphinoplatinum methyl carborane(12) complexes have been isolated. The 1H nmr spectral data Show the presence of the methyl group on the carborane cage along with the appropriate phenyl and or methyl groups on the phosphine. The absence of the C-H resonance, characteristic of the starting methyl carborane(12), suggests this linkage has been destroyed on complexation, as expected. The most soluble compound (C82, CH2C12, and benzene) was cis-Pt[P(CH3)2Ph]2[-C C-CH312. The 319 nmr of this BioHio complex appears as a broad Singlet at 5 +120.4. Although the 195Pt-31P coupling constant could be obtained on cis— PtC12[(CH3)nPh ]2, they were not observed in cis- 3-n Pt[(CH3)nPh3_ [—C -CH3]2 because of the lower solu— n12 BioHio bility of the latter complexes. The 113 nmr of cis-Pt[P(CH3)2Ph]2[-C C-CH3]2 is BioHio Shown in Figure 9. The apparently simple low field triplet and high field quartet may be due to overlap of the absorp- tions or second order spectral effects. For cis-Pt[P(CH3)3]2- [-c C-CH3]2, cis—Pt[P(CH3)Ph2]2[-C C-CH312, and BioHio BioHio 46 cis-Pt[P(Ph)3]2[-C C-CH3]2 only very broad 118 resonance BioHio absorptions were obtained. In order to understand what effects the o-bonded methyl carborane(12) exerts in the platinum complexes, an attempt was made to obtain the 195Pt-31P coupling constants. It has been suggested (52) that as the value of the coupling constant increases there is a corresponding increase in the w-acceptor character of the phosphine ligand. Although I was able to obtain these data for the complexes of the type cis-PtC12[P(CH3) Ph ] extensive efforts to obtain n 3-n 2’ Similar data for the complexes in which the chloride ligand is replaced by methyl carborane(12) were unsuccessful be- cause the complexes which contained the carborane ligand were much less soluble than those which contained the chloride. The yields for the gig—pt[P(CH3)3Ph3_n] [—c C-CH3]2 BioHio products were about 50%, based upon the amount of methyl .carborane(12) recovered. As the phenyl groups on phos— phorus were replaced by methyl groups the yields decreased. The complexes which contained greater numbers of methyl groups were more easily hydrolyzed, which may account for the lower yields. It can be suggested on the basis of these observations that methyl carborane(12) is a poorer W-acceptor than a chloride anion when either of these are trans to phOSphines in platinum(II) complexes. Increased stability toward hydrolysis is observed for dichlorobis(methylphenylphosphine) 47 platinum(II) compounds as the methyl groups on the phos- phine are replaced by phenyl groups. This stability has been attributed to the increased r—acceptor character on phosphine because of electron delocalization within the phenyl groups (52,53). The phenyl groups on phosphorus enhance the w-accep- tor character of the phosphine. This results in greater overall bond strength between platinum and phosphorus (53). Methyl groups, however, act as electron donors; this weakens the w~acceptor character of phOSphorus and the resultant bond strength between platinum and phosphorus is decreased. The substitution of methyl carborane(12) for the chloride ligand results in phosphinoplatinum(II) methyl carborane(12) complexes which are more susceptible to hydrolysis than the corresponding dichlorobis(phosphino)- platinum(II) complexes. Since the phosphine groups remain unchanged, hydrolysis apparently is made easier through the replacement of the chloride ligand by methyl car- borane(12). This suggests that methyl carborane(12) is a weaker v-acceptor than a chloride ligand. Such a result is not unreasonable. The chloride anion, which lies in the middle of the triggfeffect series, can function as a v-acceptor (53). On the other hand, it has been reported that the gfcarborane(12) cage can function as an electron donor or acceptor (54,55). Since the methyl group attached to the boron cage acts as an electron donor 48 to the cage, the methyl carborane(12) ligand should be a better electron donor and a poorer n-acceptor (56). No kinetic or thermodynamic studies have been attempted on this system. The gi§_and trans complexes of dichlorobis(tri-nf butylphosphine)platinum(II) formed upon interaction with methyl carborane(12) were much more susceptible to hydrol- ysis than the corresponding phenyl phosphine complexes of methyl carborane(12). We suggest this because of the ap- pearance of bands associated with CH vibrations in the infrared spectrum (1600 and 3400 cm—1) and the appearance of an 1H nmr resonance which can be ascribed to the start- ing material, methyl carborane(12) (C-H resonance at 6 3.5). The expected gig complex from the interaction of di- chlorobis(tri-nfbutylphosphine)platinum(II) and methyl carborane(12) was not isolated. We suggest that a reaction similar to that described below can be used to rationalize the difficulty we encountered in the preparation of the gigfdichlorobis(tri-nfbutylphosphine)platinum(II) methyl carborane(12) complex. (Where C4H8 is a butyl group which is bonded to both phosphorus and platinum.) 00 3]2 C6H6 > CH3-C C-Li + cis-PtC12[P(C4H9) BioHio (CH3-c C-)Pt[P(C4H9)3][P(C4H9)2(C4H8)] + LiCl BloHlo 49 Turco reported similar results for the interaction of dichlorobis(tri-ethylphosphine)platinum(II) with methyl carborane(12) (33). He tentatively formulated the product of this reaction as [P(C2H5)3]Pt(-C C-CH3)[P(C2H5)2(C2H4)]. B10310 (Where the C2H4 group is bonded to both phosphorus and platinum.) The molecular weight data, which was low for the expected methyl carborane(12) complex of dichlorobis- (tri—gfbutylphosphine)platinum(II), is near the theoretical molecular weight for the product shown in the reaction (687). Precise analytical results were never obtained. No further investigations were attempted on this system. Magnetic susceptibility measurements of all the phos- phino methyl carborane(12) platinum complexes isolated were made on a Gouy balance. In every case, the compounds were found to be diamagnetic. Platinum Complexes with 1,2-bis(diphenylphosphino)carborane(12) The bidentate ligand 1,2-bis(diphenylphosphino)car- borane(12) was first isolated by Schroeder (46). The pre— viously unreported 1H, 31P, and 118 nmr Spectra are shown in Figures 12-14. The proton spectrum (Figure 12) is unusual in that two distinct phenyl resonance absorptions are observed. These are located at 6 7.35 and 7.80 respectively from TMS. This behavior generally is observed when phenyl groups are bonded to systems which allow electron delocalization (57). 50 The 31F nmr (Figure 13) of 1,2-bis(diphenylphosphino)- carborane(12) is a complex multiplet centered at 6 +78.8 from P406. The 118 nmr spectrum of 1,2-bis(diphenylphosphino)car- borane(12) (Figure 14) may be rationalized in the following manner. In the icosahedral structure for the QfBioczHio unit one readily recognizes the existence of four different types of boron atoms, namely, 3(6), 8(10), 9(12), and 4(5,7,11). This would give rise to four singlets of area 2:2:2:4. These absorptions are further split by Spin-Spin coupling due to the protons attached to the borons to give a predicted spectrum of four doublets of relative integral intensities, 2:2:2:4. Such a spectrum has been reported for the 11B nmr of gfcarborane(12) at 60 MHz (58) and has been rationalized in terms of selective overlapping of the expected doublets (59). The predicted spectrum of gfcarborane(12) which appears to consist of four doublets is Shown schematically in Figure 18a. It has been suggested that the high field doublet (d) is due to the 3(6) boron atoms located nearest to the 1 carbon atoms of the icosahedral cage of gfcarborane(12) (58). When 1,2 bis(diphenylphosphino)carborane(12) is prepared, the 3(6) boron atoms would probably feel the electronic effect of the carbon-phosphorus attachment more than the other boron atoms. A schematic reproduction of the 11B nmr spectrum obtained at 65 MHz of 1,2 bis(diphenylphosphino)- carborane(12) is shown in Figure 18b. It can be suggested, 51 "i ”—1 L i4 L___J L___.| ' u I a b I C E Figure 18a. Schematic representation of predicted Spectrum for QfBioczHio unit. Iu—i L. 2 3 I. N )— Ho Figure 18b. Schematic representation of spectrum resulting from selective overlapping in (C6H5)2P'C C‘P(C6H5)2 - B10310 52 therefore, that the resonances ascribed to the 3(6) boron atoms, (d in Figure 18a), are Shifted so as to overlap the high field member of d with the low field member of the resonance due to the 4(5,7,11) boron atoms, (c in Figure 18a). The low field member of d, the resonances ascribed to the 3(6) boron atoms, then overlaps with another low field doublet (b in Figure 18a). It is not possible to determine from the experimental data which boron atoms correSpond to the resonances ascribed to the 8(10) and 9(12) atoms (a and b in Figure 18a). The resultant Spectrum consists ofia low field 1:1 doublet and an apparent 2:3:2 triplet. This is close to the observed 11B Spectrum for the 1,2-bis(diphenyl- phosphino)carborane(12) ligand (Figure 14). It has been reported that the attachment of AS(CH3)2 groups to the of carborane(12) cage has resulted in a change of electronic environment about the carbon atoms so as to cause such Spectral results (55). The mass spectral assignments for 1,2obis(diphenyl- phosphino)carborane(12) are shown in Table III. The main peaks occur at m/e ratios of 512, 435, 404, and 327. These result from the loss of phenyl groups or the phenyl phos- phine itself from the boron cage. A strong parent ion peak is observed which suggests that the ligand is thermo- dynamically stable. The interaction of 1,2-bis(diphenylphosphino)carborane(12) with potassium tetrachloroplatinite(II) (K2PtCl4) gave a monosubstituted, PtLClZ, and a disubstituted, [PtL2]C12 53 complex (where L is 1,2—bis(diphenylphosphino)carborane(12)). Both of these products were insoluble in carbon disulfide, benzene, acetone, ethyl acetate, methanol, ethanol, methylene chloride, tetrahydrofuran, and hydrochloric acid. The identity of these products is suggested by the analytical and infrared spectral data. Specifically, the C, H, and P analyses agree with the proposed formulations. In addition, the presence of platinum-chlorine stretches is observed in the infrared spectrum of the monosubstituted product (Figure 17), but is absent in the infrared of the disubsti- tuted product (Figure 16). Attempts to obtain molecular weight data by means of mass Spectral techniques were un- successful. It is generally difficult to volatilize platinum compounds for mass Spectral analyses (60). I attempted to prepare o-bonded methyl carborane(12) complexes containing titanium, vanadium, palladium, and nickel by the use of similar techniques. The equations for the reactions are: 00 (C235)2O 1- (W-C5H5)2Mcl2 + 2CH3-C C-Li —> B10H10 (w-C5H5)2M(o-c C-CH3)2 + 2LiCl BlOH10 (where C5H5 is the cyclopentadienyl anion and M is titanium or vanadium). 54 00 2. cis—MC12(PPh3)2 + 2CH3-C C-Li (c H ) o > 252 BIOHIO glng(PPh3)2(—C C-CH3)2 + 2LiCl BIOHio (where M is palladium or nickel). However, no complexes could be isolated. Studies on some of these systems are still in progress by other investigators. REFERENCES 1. 10. 11. 12. 13. REFERENCES A. Stock, "Hydrides of Boron and Silicon,“ Cornell University Press, Ithaca, N.Y. 1933. W. N. Lipscomb, "Boron Hydrides," W. A. Benjamin, New York, 1963. E. L. Muetterties and W. H. Knoth, "Polyhedral Boranes," Dekker Publishing Co., New York, N.Y., 1968. E. L. Muetterties and F. Klanberg, Inorg. Chem., 5, 1955 (1966). W. H. Sweet, A. N. Soloway, and R. L. Wright, J; Pharmcol. Expil. Therap., 137, 263 (1962). F. A. Cotton and G. Wilkinson, "Advanced Inorganic Chemistry", 2nd ed. rev., Interscience, New York, N.Y., 1966, pp. 272—275. M. F. Hawthorne, Endeavor, 22/ 146 (1966). W. N. Lipscomb, Science, 153, 373 (1966). M. F. Hawthorne, "The Chemistry of Boron and Its Compounds," E. L. Muetterties, Ed., John Wiley and Sons, New York, 1967, pp. 286-323. R. L. Hughes, I. C. Smith, and E. W. Lawless, "Produc- tion of the Boranes and Related Research," R. T. Holzmann, Ed., Academic Press, New York, 1967, pp. 163-184. T. Onak, "Advances in Organometallic Chemistry," F. G. A. Stone and R. West, Ed., Academic Press, New York, 1965, pp. 263—363. E. L. Muetterties and W. H. Knoth, Chem. Eng. News, 44/ 88 (1966). T. Heying, J. W. Ayer, Jr., 8. L. Clark, D. J. Mangold, H. L. Goldstein, M. Hillman, R. J. Pollak, and J. W. Syzmanski,.1norg. Chem., 2, 1089 (1963). 55 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 56 Russell N. Grimes, "Carboranes," F. G. A. Stone, P. M. Maitlis, and R. West, Ed., Academic Press, New York, 1970. M. F. Hawthorne, D. C. Young, T. D. Andrews, D. V. Howe, R. L. Pilling, A. D. Pitts, M. Reintjes, L. F. Warren, Jr., and P. A. Wegner, J. Amer. Chem. Soc., 90, 879 (1968). "W M. F. Hawthorne, D. C. Young, and P. A. Wegner, J;_ Amer. Chem. Soc., 81/ 1818 (1965). M. F. Hawthorne and T. D. Andrews, Chem. Commun., 443 (1965). W. H. Knoth, J. Amer. Chem. Soc., 82/ 3343 (1967). D. E. Hyatt, J. L. Little, J. T. Moran, F. R. Scholer, and L. J. Todd, J. Amer. Chem. Soc., E2/ 3342 (1967). W. R. Hertler, F. Klanberg, and E. L. Muetterties, Inorg. Chem., Ex 1696 (1967). M. F. Hawthorne and T. A. George, J. Amer. Chem. Soc., 82” 7114 (1967). M. F. Hawthorne and A. D. Pitts, J. Amer. Chem. Soc., 82“ 7115 (1967). L. F. Warren, Jr., and M. F. Hawthorne, J. Amer. Chem. Soc., 82/ 470 (1967). M. F. Hawthorne and R. L. Pilling, J. Amer. Chem. Soc., 81“ 3987 (1965). L. F. Warren, Jr., and M. F. Hawthorne, J. Amer. Chem. Soc., 22/ 4823 (1968). M. F. Hawthorne, Accounts of Chem. Res., Vol I, 2x 281 (1968). M. F. Hawthorne and T. D. Andrews, J. Amer. Chem. Soc., 82” 2496 (1965). T. S. Piper and G. Wilkinson, J. Inorg. Nucl. Chem., 3“ 104 (1956). E. O. Fidcher, W. Hafner, and H. O. Stahl, Z. Anorg. Allgem. Chem., 282, 47 (1955). A. Zalkin, D. H. Templeton, and T. E. Hopkins, J. Amer. Chem. Soc., 81, 8988 (1965). 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 57 A. Zalkin, T. E. HOpkins, and D. H. Templeton, Inorg. Chem., 5, 1189 (1966). J. C. Smart, P. M. Garrett, and M. F. Hawthorne, J;_ Amer. Chem. Soc., 91” 1031 (1969). S. Brisadadola, P. Rigo, and A. Turco, Chem. Commun., 22“ 1205 (1968). c. M. Mitchell and F. G. A. Stone, Chem. Commun., 22“ 1263 (1970). T. L. Heying, J. W. Ager, S. L. Clark, R. P. Alexander, S. P. Papetti, J. A. Reid, and S. I. Trotz, Inorg. Chem., 2, 1097 (1963). D. A. Owen and M. F. Hawthorne, J. Amer. Chem. Soc., 92” 3194 (1970). H. D. Smith, J. Amer. chem. Soc., S1, 1817 (1965). R. Zaborowski and K. Cohn, Inorg. Chem., §x 678 (1969). B. Rosenberg and L. Van Camp, Cancer Research, 32” 1799 (1970). M. F. Hawthorne, T. D. Andrews, P. M. Garrett, F. P. Olsen, M. Reintjes, F. N. Tebbe, L. F. Warren, P. A. Wegner, and D. C. Young, Inorg. Syn., 10, 91 (1967). S. Papetti and T. L. Heying, J. Amer. chem. Soc., 86, 2295 (1964). K. A. Jensen, Anorg. Allgem. Chem., 229, 225 (1936). S. O. Grim, R. L. Keiter, and W. McFarlane, Inorg. Chem., 6, 1133 (1967). M. A. A. Beg and H. C. Clark, Can. J. Chem., 38” 119 (1960). G. B. Kauffman and C. A. Teter, Inorg. Synthesis, 1, 245 (1963). Roy P. Alexander and Hansjuergen Schroeder, Inorg. Chem.,fizx 1107 (1963). L. Summers, R. H. Uloth, and A. Holmes, J. Amer. Chem. Soc., zzfij3604 (1955). J. Chatt and F. G. Mann, J. Chem. Soc., 1622 (1939). J. M. Jenkins and J. G. Verkade, Inorg. Syn., Vol 6“ 2250 (1967). 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 60. 58 K. Yamamoto, Bull. Chem. Soc. Japan, 21/ 501 (1954). L. M. Venanzi, J. Chem. Soc., 719 (1958). S. O. Grim, R. L. Keiter, and W. McFarlane, Inorg. Chem., 6” 1133 (1967). F. Basolo and R. G. Pearson, "Mechanisms of Inorganic Reactions," John Wiley and Sons, Inc., New York, N.Y., J. A. Potenza and W. N. Lipscomb, J. Amer. Chem. Soc., 86, 1874 (1964). R; B. Zaborowski, M.S. Thesis, Michigan State University, 1969. D. E. Young, G. E. McAchran, and S. G. Shore, J. Amer. Chem. Soc., §§/ 4390 (1966). Personal communication with Dr. G. J. Karabatsos, Michigan State University, East Lansing, Michigan. G. R. Eaton and W. N. Lipscomb, "NMR Studies of Boron Hydrides and Related Compounds," W. A. Benjamin Inc., New York, N.Y., 1969, pp. 375-391. R. L. PiLUhg, F. N. Tebbe, M. F. Hawthorne, and E. A. Pier, Proc. Chem. Soc., 402 (1964). Personal communication from Mrs. R. Guile, Mass Spectral Technician, Michigan State University, East Lansing, Michigan.