RMRRANGEMENTS 0F SILYL ENOL ET HERS 'Ehws for the Degree of Ph. D. ‘ MICHIGAN STATE {ENIVERSITY JUDETH A. MCCLARIN 1973 ,4 "1"" _~ 1‘3 LIBRARY I”; Higan Sta‘SC _ afiaivcrsity rS ‘wo. C’- ~v I. presented by fl/f/cC/W has been accepted towards fulfillment of the requirements for fl. 0 (1959‘. in [éfl/J [9/ Major professor ' lulu IV I am “on: t 1 am mm: m uaum nmozni m Ira-31w- Si] Luhere : ms ‘. tenfigu DI tran ABSTRACT REARRANGEMENTS OF SILYL ENOL ETHERS By Judith A. McClarin Silacycloalkane derivatives of acetylacetone, CHSSi(CH2)x(acac), (where x - 3, 4, and 5) have been prepared and characterized. The com— pounds possess an open chain enol ether structure which gives rise to configurations in which the uncoordinated oxygen is positioned cis (L) or trans (kl) to the siloxy group. (.353 0 ms ’ , ,Si-——o\ >—cn / ,Si—o H C" A c==c 3 CH2 A \ =c/ (CH2 x-2 CH3 (CH2 x-Z CH3 04/; 3 (p (w The cis isomer undergoes a rapid intramolecular stereochemical rearrange- ment which can be detected by nmr spectroscopy: a o o c/CH3 o .' T” ‘\ R Si—O \c-cn 3‘ --‘.—— R Si’ 1, ,.\c—H ‘=‘ R 51-0 \C-CH 1’- 3 / 3 3 ._‘ / 3 \ / 3 g H CH — / a \H 3 CH3— 7'2 new. .' 'v. 1' A 4 riative r aspective Trims silyldipi‘ throne pessess 0 he tr_an_s large vat in the l mcoordi fitter 1 Smith retent than 1 bearr: 1.1 x 'ffi Judith A. McClarin fi>"rhe Eggg§_isomer is stereochemically rigid. Ring strain causes the 333_ E7 to Eggn§_ratio to increase in the order CHSSi(CH2)S(acac) < CH3Si(CH2)4(acac) < CHSSi(CH2)3(acac), with values of 0.25, 0.45, and 2.23, respectively. The rate of the rearrangement increases in the same order with estimated relative rate constants at 25° of 2 x 103, 0.9 x 106 and 1 x 108 sec.1 , respectively. Trimethylsilylhexafluroacetylacetone (CH3)3Si(hfac) and trimethyl- silyldipivaloylmethane (CH3)3Si(dpm), where the substituent on the B- diketone is CF3 and t-C4H9, respectively, were prepared and feund to possess open chain enol ether structures. (CH3)38i(hfac) exhibits only the 53393 isomer and (CH3)38i(dpm) exhibits only the gi§_isomer. The large variation in the gi§_to tran§_ratio is explained by the differences in the long range electrostatic interaction between the silicon and the uncoordinated carbonyl oxygen. The rate of rearrangement at 25° for 5127(CH3)38i(dpm) is at least five orders of magnitude more facile than the analogous acetylacetone compound. The chiral derivative £i§;(C6H5CHz)(CH3)(C6H5)Si(dpm) was prepared in order to follow the stereochemical course of the rearrangement by nmr spectroscopy. The difference in activation free energy for the low energy retention processes and a higher energy inversion processes is greater than 17.8 Real/mole. The relative rate constants at 25° for the observed rearrangement processes and the inversion processes are estimated to be 6 7 1.1 x 10 and 1.5 x 10' sec-1, respectively. More than 1013 rearrange— ments occur at 25° without inversion of configuration. _.—-- The :eafl'm‘ gremd 5‘ cf the u tamed 5-diketo grocesse 5’34) C3 illih m2 manner The was obse of the 1 acetyla. Judith A. McClarin The above relationships between structure and lability show that the rearrangement involves the transformation of an incipient pentacoordinate ground state into a five coordinate intermediate by nuc1e0phi1ic attack of the uncoordinated carbonyl oxygen on the silicon at an adjacent tetrahedral face or edge. The trigonal bipyramid intermediate has the B-diketonate spanning an axial-equatorial position. These displacement processes that occur with retention configuration at the triorganosilyl group can indeed involve the fbrmation of a five-coordinate intermediate which may pseudorotate. Earlier views maintained that such processes are uncommon. The methylsilacyclobutane derivative of acetylacetone (CH3)Si(CH2)3(acac) was observed to undergo an irreversible rearrangement involving opening of the silacyclobutane ring fellowed by a Michael-type addition to the acetylacetone moiety to give structure (ill). F‘ REARRANGEMENTS 0F SILYL ENOL ETHERS By Judith AttMcClarin A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1973 ACKNOWLEDGEMENTS I would like to express my appreciation to Professor T. J. Pinnavaia. Through his guidance and patient efforts as a teacher I have acquired much. I wish to thank Professor C. H. Brubaker fer being my second reader and fer his interest during the past four years. I also wish to thank the Chemistry Department and Michigan State University for financial support and for the opportunity to acquire skills as a teacher. I am deeply grateful to my parents, Mr. and Mrs. L. McClarin for their constant encouragement and love. ii .é. - 1. M II. EXP] :ns-ncn BI! I. II. III. TABLE OF CONTENTS INTRODUCTION . . . . . . . EXPERIMENTAL . . . . . ..... . . . . . . A. Reagents and Solvents . . . . . . . . . . . . . . . . B. Synthesis. . . . . . . . . . . . . . . . . . . . . . . 1. General Synthetic Techniques . . . . . . . . . . . 2. Trimethylsilylhexafluroacetylacetone . . . . . . . 3. Trimethylsilyldipivaloylmethane. . . . . . . . . . . 4. Benzylmethylphenylchlorosilane . . . . . . . . . . . 5. Benzylmethylphenylsilyldipivaloylmethane . . . 6. Methylchlorosilacyclobutane. . . . . . . . . . . 7. MethylchlorosilacycIOpentane . . . . . . . . . . 8. MethylchlorosiIacyclohexane. . . . . . . . . . . 9. Cyclic Organosilyl Acetylacetone Derivatives . . 10. Rearrangement Product of Methyl- (acetylacetonato)silacyclobutane . . . . . . . . . ll. l,l,3,3,S,S-hexamethyl-I,3,S-trisilacyclohexane. . 12. l,3,5,7-tetrasi1aadamantanes . . . . . . . . . . . C. Analytical Data. . . . . . . . . . . . . . . . . . . . 0. Infrared Spectra . . . . . . . . . . . . . . . . . . . E. Nuclear Magnetic Resonance Spectra . . . . . . . . F. Preparation of Solutions for IR and NMR Studies. RESULTS AND DISCUSSION . . . . . . . . . . . . . . . . . Preparation and Characterization of Cyclic Organosilyl Acetylacetone Derivatives. . . . . . Preparation and Characterization of (CH3)3Si(hfac) and (CH3)3Si(dpm). . . . . . . . . . . . . . . . . . . Preparation and Characterization of (C6H5CH2) (CH3) (C 6H$)Si(dpm). o o . o o o o o . o o o Attempted Preparation of 1,3,5, 7-tetrasilaadamantane Derivative . . . . . . . . . . . . . . . . . . . . . . . Kinetic Study of the Intramolecular Rearrangement of cis-triorganosilyl-B-diketone Derivatives . . smary O O O O O O O O O O O I O O O O O O O O O BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . . . iii Page . 15 15 15 15 16 16 16 17 . 17 18 . 18 . 18 . 19 19 21 21 21 22 22 . 23 . 23 42 . 45 49 51 . 66 68 Elm—w— Table IX. Table II. III. IV. VI. VII. VIII. IX. LIST OF TABLES Page Summary of Stereochemistry of Selected R38i*x Displacement Reactions . . . . . . . . . . . . . . . . . . . . 2 Sommer's Mechanisms for Displacement Reactions at a Triorganosilyl Group . . . . . . . . . . . . . . . . . . . . S Equilibrium Ratio of Cis/Trans Enol Ether Isomers for Triorganosilylacetylacetonates . . . . . . . . . . . . . . . . 7 Activation Energies for Intramolecular Rearrangements of Compounds Containing R38i Groups. . . . . . . . . . . . . . 8 Synthesis and Analytical Data for Cyclic Organosilyl E1101 Ethers O O O O O O O O I O O O I O I O O O O O O O O O o 20 Proton Chemical Shift Data for Cis and Trans - Cyclic Organosilyl Acetylacetone Derivatives. . . . . . . . . . . . . 32 Equilibrium Ratio of Cis and Trans Enol Ether Isomers for CH3$i(CH2)x Derivatives ’of Acetylacetone . . . . . . . . . . . 36 Infrared Absorbtion Bands for Cyclic Organosilyl Acetylacetone Derivatives. . . . . . . . . . . . . . . . . . . 38 Equilibrium Ratios of Enol Ether Isomers for Trimethylsilyl- B‘diketonates o o o o o o o o o o o e o o o o o o o o o o o o o 44 Kinetic Data for the Intramolecular Rearrangement of Cis-Cyclic Organosilyl Acetylacetone Derivatives . . . . . . . 63 iv Figure LIST OF FIGURES Proton nmr spectra of A) CH3Si(CH2) (acac) in CCl , B) the thermal rearrangement produc of CH381(CH233(acac) in CC14, C) the methanolysis product of the rearranged form in CC14 . . . . . . . . . . . . . . . . . . . . . . . Infrared spectra of A) CH3$i(CH2)3(acac) in CHZCIZ, B) the thermal rearrangement product of CH3Si(CH2)3(acac). . . . Proton nmr spectra of A) CH3Si(CH2)4(acac) in CC14, B) CH3Si(CH2)5(acac) in CC14 . . . . . . . . . . . . . . . . . Infrared spectra for A) CH3$i(CH2)3(acac) in CHZCIZ, 8) CH3Si(CH2)4(acac) in CC14, C) CH3$i(CH2)5(acac) in CC14 . . Proton nmr Spectra of A) (C HSCH CH3)(C6H )Si(dpm) in CC14, B) benzyl methylene proton 28 Mechanisms which would account for the exchange of the non-equivalent R groups of a 8-diketone in cis-R381 (B’diketone) o o o o o o e o o o o e o o o o e o o o o o o o o 0 Temperature dependence of the proton nmr spectrum of (C6HSCHZ)(CH3)(C6H5)Si(dpm). . . . . . . . . . . . . ..... Temperature dependence of the proton nmr spectrum of (CH2)SSi(CH3)(acac) in CH2C12° . . . . . . . . . . . . . . . p8 ten in C 140 o o o o o o o o o Page . 26 29 . 34 . 4O 48 SS . S7 5T0”? exampj stere< ceord: reten1 quasi- £10“? pra’J neces: react: solver involw LiAlH‘ In so: to the r[91m I. INTRODUCTION Bimolecular nucleophilic displacement reactions at a triorganosilyl group may occur with retention or inversion of configuration. Several examples are summarized in Table I. Gielen1 has recently interpreted the stereochemistry of these reactions in terms of the fbrmation of five- coordinate RSSiXY intermediates. Displacement reactions which occur with retention of configuration are believed to occur by the fbrmation of a quasi-cyclic structure between the entering nucleophile and the leaving group. The reactions occur when the leaving group is a weak nucleophile (pKa>10) and electrophilic assistance from the attacking species is necessary to separate the nucleophile from the silicon. Retention reactions are known for the displacement of hydride ions in a variety of solvents and alkoxy groups in non-polar solvents. A typical case involving retention of configuration is the reaction of R Si*0Me and 3 LiAlH4 (Equation 1). .' .‘ (1) R351*0Me+ 1.1th4 + R331: 0 :AIH , * R351 H In some cases, displacement of a fluoride ion occurs with retention, due to the coordinating ability of fluorine with electrophilic centers (Equation 2). . ,~,» . had-aw . Tabl R351 R 51 R 51 R 51 R.$l 3 R 51 R 51 11.51 3 11.31 J R 51 R 51 R Si R 51 R 51 2 Table 1. Summary of Stereochemistry of Selected RSSi*X Displacement Reactions. Reactant Reagent Product Solvent Stereochemistry R35i*H LiAiH4 R35i*H Bu20 retention R381:H KOH R381:0H xylene retention Rssi H KO-t-C4H9 RSS1 0- -t-C4H9 t- -C4H90H retention R35i H Cl2 RSSi:Cl CCI4 retention RSSi:H HSCOSH R38i*0H CCl4 retention R38i 0R t-CuMgCl R351 H ether retention RSSi*0R LiAlH4 RSSi:H ether retention R35i*on non R38i ox xylene retention R3Si*F R35i*ox R35i OSiR3 xylene retention RSSi:F EtLi R3Si:Et ether retention Rssi OMe MeOH Rssi OMe MeOH inversion R381 ocoa LiAiH4 R35i*H ether inversion RSSi 000R KOH R381:0H xylene inversion RSSi OCOR MeOH R381 OMe pentane inversion R381 CI(Br) H20 Rssi OH ether inversion R35i Cl(Br) KOH R331 0K xylene inversion R3Si*Cl ROH R381 0R pentane inversion Rssi:Cl(Br) NaBH(0Me)3 R381 OMe ether inversion R35i Cl CHSCOZK R38i*OCOMe C6H6 inversion Rssi:Cl(Br) LiAlH4 Rssi:H ether inversion R38i F LiAlH4 RSSi H ether inversion R381 Cl EtLi R381 Et ether inversion The 5 force :rigc SIBI'E Occur M11 The p- gener: biPYr: 105m R ,n (2) R381* F + EtLi + R—Si‘fie 193—» R381 *Et + LiF CH2 LiO CH3 The strained, quasi-cyclic structure characteristic of retention reactions forces one of the electronegative groups in an equatorial position of the trigonal bipyramid. Pseudorotation would be rapid and the resulting stereochemistry would be retention (Equation 3). R R R R - 2 1R 3 2 (3) 1\Si./ LY» X— Si51( xd ~R3 x R2 Y “Rz Pseudorotation would be slow, since both electronegative groups are in electronically favorable positions.4 The solvent may effect the stereochemistry of the reaction in some cases. Methoxy exchange reactions in methanol occurs with inversion. In contrast, alkoxy displacements occur with retention in non-polar solvents.2 Non-polar solvents favor intramolecular electrophilic assistance, whereas, polar solvents favor intermolecular assistance by solvent molecules.l Gielen's views on the stereochemistry of nucleophilic displacement at a triorganosilyl group are related in part to the earlier mechanistic interpretations of Sommer. Sommer's classification of bimolecular dis- placement processes are summarized in Table II. According to Gielen, nucleophilic displacement reactions involving R3Six compounds may be explained by one mechanistic approach where a stable five coordinate intermediate is formed which is sufficiently long lived to pseudorotate. The structure of this intermediate determines the stereochemistry of the reaction. Sommer believes that stable five coordinate intermediates are uncommon and that most of the nucleophilic reactions occur by concerted processes. He also believes that although the silicon 3d orbitals may, in certain cases, participate in the reaction, this does not mean that an intermediate is formed. The stereochemical data can be adequately explained by either approach. There is no evidence Sommer's Mechanisms2 Triorganosilyl Group. for Displacement Reactions at a Characteristics Stereochemistry Table II. Mechanism 8N2 - Si 5N1 - Si * . 8N2 - 81 *i' . 5N2 - Si Common mechanism fbr polar reactions involving trigonal bipyramid transi- tion state. Si may become electron rich compared to ground state. 3d orbital participation only if Ea is lowered: does not imply an inter- mediate. Common mechanism involving frontside attack; may involve attack by strong nucleophiles. Formation of quasi- cyclic structure, no ion pair formation. Rare mechanism involving formation of pentacoordinate intermediate in fast v equilibrium step. Unable to dis- tinguish kinetically from 5N2 - Si. [i.e. R Si*F + MeOH + RSSiF] 3 Rare mechanism involving fbrmation of pentacoordinate intermediate in slow step, unable to distinguish kinetically from 5N2 - Si, no example. Inversion for acyclic cases and retention fOr bridgehead Si Retention Racemization No information 6 to support the existence of a five coordinate intermediate, nor is there any conclusive evidence to support the mechanisms pr0posed by Sommer. The reversible intramolecular rearrangements of R Six compounds, where 3 one atom on X is displaced by another atom on the same ligand, represent systems in which structural and electronic parameters may be systematically varied without altering the basic displacement reaction. A typical rearrangements is illustrated in equation 6 for a cis-triorganosilyl- acetylacetone, R38i(acac). b. o o_C/CH3 o 6 RS'—0 gwcu11 .—-=R3i’/'+'"-\c-— —"“———RS 0 c'---cn"l ( ) 3 1 l’ 3 3 \\ "')/ 3 1 ‘\ __ /’ 3 =c o—c c—c\ \ _ / H a H CH3 CH 1’— 3 3 As illustrated by the data in Tables III and IV, the equilibrium gig to £322§_ratio as well as the rate of rearrangement, fbr the £i§_isomers of these compounds increases as the electron withdrawing ability of the silyl substituents increases, suggesting that a long range electrostatic inter- action exists between the uncoordinated carbonyl oxygen and the silicon atom in the ground state of the gi§_isomer. The gi§_to Egag§_ratios for (c6H6)(CH3)2 Si(acac)6 and (C6H5)2CH3$i(acac) are lower than expected based on Hammett-Taft 0* values fbr C H group and indicate a weakening 6 S of the long range interaction due to steric factors.6 Table III. Equilibrium Ratio of Cis/Trans Enol Ether Isomers for Triorganosilylacetylacetonates.6 Compound [CiSJ/[Trans] (n-C4H9)(CH3)ZSi(acac) ‘ 0.28 (C2H5)(CH3)ZSi(acac) 0.29 (CH3)SSi(acac) 0.34 (CFSCHZCHZ)(CH3)ZSi(acac) 0.39 (CH2 - CH)(CH3)ZSi(acac) 0.38 (C6H5)(CH3)ZSi(acac) 0.31 (C6H5)2(CH3)Si(acac) 0.25 Table IV. Activation Energies for Intramolecular Rearrangements of Compounds Containing R38i Groups. Compound Ea Kcal/mole Reference R33i (acetylacetone) 6 R3Si = (n-C4H9)(CH3)ZSi 14.1 (C2H5)(CH3)ZSi 14.0 (CH3)SSi 13.8 (CH2=CH)(CH3)ZSi 13.0 (C6HS)(CH3)ZSi <13.2>12.6 (CF3CH2CH2)(CH3)ZSi 12.2 (C6H5)2(CH3)Si 12.2 (CH3)38i (triacetylmethane) (I) 15.4 7 (2-C4H9)(CH3)25i (tropolane) (LLL) 8.29- 9 (CH3)3Si (pyrazole) (IX) 10 R3 = R4 = R5 = H ~32 R3 = R5 = cus, R4 = H 28 R3 = R4 = R5 = CH3 24 R3 : R5 = CF3, R4 = H <32 Rssi (1,3 dimethyltriazene) (X) 11 Rssi = (CH3)3Si 16.12' (CH3)(C2H5)ZSi 16.3 (CH30)38i 14.8 (CH3)(CH3O)28i 14.9 (CH3)2(CH30)Si 15.1 (CH3)(C1)ZSi 13.6 (CH3)2(C1)Si 13.8 Table IV (cont'd) Compound Ea Kcal/mole Reference (CH3)3SiN(CbHS)COR (W) 14 R = CHClz 19-2 CHZCI 17.5 CH3 11.2 CHZCH3 11.6 CH(CH3)2 9.5 (CH3)SSi (o-methylallyl) 47.7 22 (CH3)2(C6H5)Si (ormethylallyl) 47.2 (CH3)2(C6HS)Si (mB-dimethylallyl) 47.7 (CH3)SSi (o-phenylallyl) 42.5 13 a. Ea determined by temperature dependence of C nmr b. A61 values 10 Analogous processes for R3Si derivatives where the silyl group migrates between two oxygen atoms have been reported for triacetyl- methaneQ),7 substituted malonate(££)8 and tropolone(LLI)9 derivatives. The rearrangements are facile, with activation energies of 15.4 kcal/mole for (I) and 8.2 kcal/mole for (III) and 13.8 kcal/mole fer (CH3)33i(acac). The low activation energy of (III) supports the suggestion by Pinnavaia of a pentacoordinate intermediate fer these reactions, since here the dipolar species is highly stabilized.9 Substituent effects on the cis/trans ratio were observed for 11. When R a H or Me, only the cis isomer is observed, but when R = Ph, 10% trans isomer is observed,8 most likely due to steric effects. 0 (CH3) 381—0 \ >C—CH3 (CH3) SSi— o /\c -—-0 (:H3 /c=='—'c /CH3 /C =C\ CH3 > (:3 R o 3 l (Ref.7) g; (Ref.8) (2-C4H9) (CH3) 25i—0\ ;@ III (Ref.9) 11 Rearrangements where the silyl group migrates between two nitrogens have been reported for RSSi derivatives of pyrazoleq’V),10 triazine(¥),ll benzamideneqp,12 and for silyl hydrazine anions(¥L,I’).l3 Enhanced rearrangement rates for substituted pyrazoles are observed as the 3, 4, and 5 substituents become more electron releasing (25,, Table IV). The increase in electron density at the uncoordinated nitrogen would increase the extracoordinate silicon-nitrogen interaction. If the increase in electron density had the effect of strengthening the silicon- nitrogen bond the opposite effect would be observed. Steric repulsions between the 3, 5 substituents and the trimethylsilyl group would be relieved in a pentacoordinate activated complex leading to a decrease in activation energy when H is replaced by methyl in the 3,5 positions.10 The activation energy of rearrangement fer (V) is lowered as the silyl substituents become more electron withdrawing.11 The magnitude of this substituent effect is comparable to that observed fer R38i(acac) (Table IV). N 3 3 R c R 5 I, 3 x (Ref.11) (”133331- T" N R Si—N-—-—N-—=N-—CH 3 XL; (Ref.13) VI (Ref.12) 12 Migration between nitrogen and oxygen is observed in trimethylsilyl anilides(¥{£1) and amides(£§), as well as in the more complex disiloxa- diazines(§). The activation energy for rearrangement of the anilides decreases as electron releasing substituents are placed on the carbonyl group14 (g£., Table IV). This result is similar to the effect of sub- stitution on the pyrazole ring. It has also been observed that the equilibrium concentration of O-silylated isomer increases with increasing electron withdrawing ability of the phenyl substituents.l7 An inter- action between the uncoordinated oxygen and silicon referred to as in- cipient pentacoordination, exists in disiloxadiazines (29.16’17 The interaction was established in the solid state by x-ray diffraction studies.18 The distance between the exocyclic oxygen and silicon is 0.74 X less than the sum of the Van der Waals radii, but still longer than a silicon-oxygen covalent bond. The geometry of the incipient penta- coordinate silicon is distorted toward a trigonal bipyramid. o I! X . R—C-N [(CHS)381]2N—fi—R (CH3)SS1 0 I (Ref.15) Xiii (Ref.14,17) mi CQQSi//CH3 0 /\———H I I0 H3 :\O—z H§o\/SEH é (Ref.16,17) 13 The superior silylating ability of triorganosilylamides compared to triorganosilylamines can be explained by the incipient pentacoordinate 17 The reaction with a geometry of the silylamides in the ground state. proton donor may be described as the transfbrmation of a trigonal bipyramid or tetragonal pyramid ground state to an octahedral transition state. (Equation 7) R . 2 R o—c/ 2 R o-—-~c’R l l 1 R3 \ I '_ 1 (7) Si----N -—R -—9-*—'—-+ Si/N R -—-—» R308iR3 R R Rér’ \§g, + RZCONHRI g°x R H The lower activation energy for the reaction of silylamides compared to silylamines is due to the already expanded valence shell of silicon in the silylamide. The fluxional character of trimethylsilyl groups is also fbund in the intramolecular rearrangement of trimethylsilylcyclopentadiene,19’20 and the rearrangements of trimethylsilylindine21 and allylic silanes.22 In contrast to the rearrangements of R Si groups between electronegative 3 atoms, the rearrangement of allylic silanes (Equation 8) is uneffected by substitution at the silicon. The only substituent influence on rate CH3 (8) (CH3) 381 --‘C --H ——-=- (CH3) 3.81 --— (‘IH2 P— CH£==TCH Hfi====CH CH3 14 occurs when one product is clearly stabilized by the substituent.22 This effect is expected in a concerted rearrangement process.23 The activation energy for this process is considerably greater than those observed for the degenerate processes (g£,, Table IV). Triorganosilyl B-diketonate and triorganosilyltropolone are known to rearrange with retention of configuration at silicon,9’24 as do silyl 25,26 carbinols and B-ketosilanes. The presence of a cyclic structure in the activated complex, such as the one shown in Equation 6 for Rssi acac, indicates a similarity to the retention case of bimolecular nucleophilic displacements. As a general case, the facile rearrangement of R SiX———¥ may be 3 explained as an incipient pentacoordinate ground state transfbrmed into a pentacoordinate intermediate.l’6’9'l7 Evidence available at present tends to support this mechanism. The suggestion has been made that triorganosilyl compounds may undergo pseudorotation, similar to the phosphoranes.3’6 The purpose of this research is to elucidate further the kinetics and mechanism of the rearrangement of R Si B-diketonates. 3 We have studied the effect of ring strain on the reactivity and measured the difference in activation free energy between the retention and inversion process. II. EXPERIMENTAL A. Reagents and Solvents Methyl(3-chloropropyl)dichlorosilane was obtained from the Chemical Research Division of the Dow-Corning Corporation and was used without further purification. All other silicon reagents were obtained from Aldrich Chemical Company and were fractionally distilled in an N2 atmosphere before use. The purity of reagents was checked by nmr spectroscopy. Diethylether, benzene, hexane and tetrahydrofuran were dried for at least 24 hours over lithium aluminum hydride and freshly distilled before use. Methylene chloride, carbon tetrachloride and chlorobenzene were dried over calcium hydride fer at least 24 hours and freshly dis- tilled before use. a-Chloronapthalene was dried over molecular sieves, and carbon disulfide was dried over phosphorus pentaoxide. Pyridine was distilled from and stored over Drierite. Magnesium turnings and powdered magnesium was degassed under vacuum overnight before use. B. Syntheses General Synthetic Techniques All glassware was dried at 175°C overnight and cooled in a disiccator whenever possible. All manipulations of hydrosc0pic reagents and products were carried out in a nitrogen atmosphere. Liquid products were purified by fractional distillation through a 12-cm jacketed Vigreaux column. 15 16 Trimethylsilylhexafluroacetylacetone27 A solution of 11.7g (0.056 mol) of hexafluroacetylacetone and 20ml (0.157 mol) of trimethylchlorosilane was heated at reflux temperature fer eleven days. The reaction mixture was distilled, at atmospheric pressure to give trimethylsilylhexafluroacetylacetone in 43% yield (b.p. 128-129°). Anal, Calculated fer C H F O Si: C, 36.0; H, 3.57; 8 10 6 2 F, 40.7; Si, 10.0. Found: C, 34.9; H, 4.01; F, 40.6; Si, 10.9. TrimethylsilyldipivalOylmethane The sodium salt of dipivaloylmethane 10.4g (0.05 mol) dissolved in 50ml ether was added slowly to a solution of 5.9g (0.054 mol) trimethyl- chlorosilane in 50ml ether. The reaction mixture was stirred for two hours, and the resulting sodium chloride was removed by filtration. Distillation of the mixture under reduced pressure gave a colorless liquid product (b.p. 44-45°; 0.04 torr). The yield was 51%. The sodium salt used in the reaction was prepared by the reaction of dipivaloyl- methane and sodamide or sodium metal. Anal, Calculated for C H O Si: 14 28 2 C, 65.6; H, 11.0; Si, 11.0. Found: C, 66.2; H, 11.2; Si, 11.6. Bengylmethylphenylchlorosilane Benzylmagnesium chloride was formed by the slow addition of 90ml (0.78 mol) benzylchloride to 19g (0.78 mol) magnesium turnings in 700ml ether. After the mixture had been allowed to reflux for one hour, it was cooled to -78° in a dry-ice acetone slush bath and 100ml (0.78 mol) methylphenyldichlorosilane and 1000ml ether were added rapidly. The reaction mixture was stirred and allowed to warm slowly 17 to room temperature overnight. The solid was filtered from the slurry, and the ether was removed under reduced pressure. The remaining liquid was distilled under reduced pressure to give a colorless liquid product (b.p. loo-105°; 0.1 torr). The yield was 51%. The purity of the product was checked by integration of CH -CH2, and -C H nmr lines. 3’ 6 5 Benzylmethylphenylsilyldipivaloylmethane The preparation of this compound is analogous to the pre- paration of trimethylsi1yldipivaloylmethane. Distillation of the pro- duct under reduced pressure gave a viscous pale yellow product (b.p. 183-185°; 0.1 torr). The yield was 51%. Anal: Calculated fer C H 0 Si: C, 76.09; H, 8.68; Si, 7.12. Found: C, 76.34; H, 8.60; 25 34 2 Si, 7.05. Methylchlorosilacyclobutane28 Powdered magnesium 18.3g (0.75 mol) was activated by grinding with a mortar and pestle in a nitrogen atmoSphere and then heating it for 15 minutes at reflux temperature in 200ml ether containing 2ml of 1,2-dibromoethane. The magnesium slurry was cooled to room temperature and 39.9ml (0.25 mol) methyl(3-chlor0propyl)dichlorosilane, in 200ml ether was added slowly over several hours. The mixture was refluxed overnight, cooled to room temperature, and then stirred for 24 more hours. The product was separated from the filtrate by fractional 28 distillation at atmospheric pressure (b.p., 101-102°; Lit. 103.5-104° 731 torr). The yield was 60%. 18 Methylchlorosilacyclopentane29 To a mixture of 22g (0.90 mol) magnesium in 500ml ether, was added slowly 50.8g (0.400 mol) 1,4-dichlorobutane. The mixture was stirred fer 4-5 hours until the ether stopped refluxing. The resulting difunctional Grignard reagent was transferred to a large drOpping funnel and added slowly to 38.5g (0.360 mol) methyltrichlorosilane in 1 liter of ether. The mixture was heated at reflux temperature for 4 hours and then stirred at room temperature overnight. The slurry was filtered and the ether removed by distillation at atmospheric pressure. The remaining mixture was filtered and distilled at atmospheric pressure to give a colorless liquid product (b.p. 128-129°; Lit.29 132). The yield was 31%. Methylchlorosilacyclohexane29 The preparation of this compound is analogous to the prepara- tion of methylchlorosilacyclopentane. Purification of the product by distillation, at atmospheric pressure gave a colorless liquid (b.p. 160- 29 188°; Lit. 167°). The yield was 34%. chlic Organosilyl Acetylacetone Derivatives The general method was modified from the preparation of tri- methylsilylacetylacetone.6 A solution of pyridine (22,, 0.1 mol) in hexane ($2,, 20ml) was added slowly to equimolar amounts (£2, 0.1 mol) of acetylacetone and the silane in hexane (9a,, 80ml). The mixture was allowed to stir at room temperature overnight, except in the reaction with the silacyclobutane, in which case the mixture was stirred for one hour. The pyridinium chloride was filtered, and the hexane was 19 removed under vacuum. The remaining liquid residue was filtered again and distilled under reduced pressure. Boiling points, yields, and analytical data is presented in Table V. Rearrangement Product of Methyl(acetylacetonato)silacyclobutane The silacyclobutane derivative was found to rearrange at room temperature, and was stored in dry-ice. The rate of rearrangement was dependent on the purity of the sample. A highly purified sample had undergone approximately 50% rearrangement in 2 weeks. The rearrangement product was purified by distillation at reduced pressure (b.p. 36-37°; 0.5 torr). Angl, Calculated for C H 0 Si; C, 58.65; H, 8.75; Si, 15.24; 9 16 2 M.W., 184. Found: C, 58.66; H, 8.77; Si, 15.46; M.W., (in CC14) 193. l,l,3,3,5,5-hexamethyl—l,3,5-trisi1acyclohexane30 A small portion of 284g (1.73 mol) chloromethyldimethylchloro- silane and a few drops of iodomethane to initiate the reaction was added to a mixture of powdered magnesium (33,, 2g) in tetrahydrofuran. The remaining silane and enough tetrahydrofuran to bring the volume to 500ml was then added. The remaining magnesium (70.2g, 2.89 mol) was added over a period of 2 1/2 hours. The temperature was kept between 30° and 50° during the addition, and then the mixture was heated to 50° for an additional 2 hours. The reaction mixture was washed successively with 500ml, 300ml, and 100ml water and then dried over magnesium sulfate. The product was purified by distillation under reduced pressure to give a colorless liquid (b.p. 108-112°; 50 torr, 31 Lit. 112°; 50 torr). The identity of the product was further verified by infrared spectroscopy. The yield was 14%. 20 2.3» 78 omen/A \l- wm.ma Hm.m wa.ao om.ma m¢.m mo.mo mN we-ae n:ux\.m, V UGO“ me.e~ «a.» om.mm ea.ea ma.m em.oe em fiance N.ov I/za om-mn m \\\. =0 060d fleece m.ov x/xa on.ma Nw.w mo.wm v~.ma ma.m mo.wm NN mm-em azuxxx. mm m 6 am _ z . u ».eaoa> u. .uemoa ecsoasoo eeeoa copaasuHau . mafiaaom .mnogum Hocm fixuwmoemuuo owfiozu Mom memo Hmowpxfimc< ecu manoeueam .> «Heap 21 1,3,5,7-tetrasilaadamantanes31 A mixture of 10.4g (0.078 mol) aluminum chloride, and 17g (0.078 mol) l,l,3,3,5,5-hexamethyl-l,3,5-trisilacyclohexane, was heated to 100° for 2 hours. The tetramethylsilane which was formed was con- tinuously distilled from the reaction mixture. The reaction mixture was cooled, diluted with 200ml benzene and washed twice with 50ml water. The organic layer was dried over magnesium sulfate, and the benzene was removed under vacuum. White crystals were vacuum sublimed at room temperature from the resulting oil. The infrared spectrum showed a Si-Cl stretch at 527 cm'l, and silver chloride was precipitated from a nitric acid solution of the product. The nmr spectrum of the product indicated a mixture of silaadamantanes which were identified as l,3,5,7-tetramethyl-1,3,5,7-tetrasilaadamantane and l-chloro-3,S,7- trimethyl-l,3,5,7-tetrasilaadamantane by comparison of chemical shift values to reported value532(b.p. 114-127°). C. Analytical Data Microanalysis of compounds was performed by Galbraith Laboratories, Inc., Knoxville, Tennessee. D. Infrared Spectra Infrared spectra of silyl enol ethers were determined as 5% by volume solutions in carbon tetrachloride or methylene chloride by use of Perkin Elmer 2378 and Perkin Elmer 457 spectrometers. Matched liquid cells with KBr salt windows and a 0.1mm path length were used 1 for difference spectra of the solutions. The 2851 and 1603 cm' bands of polystyrene were used as reference frequencies. 22 E. Nuclear Magnetic Resonance Spectra Proton magnetic resonance spectra were obtained by use of a Varian A56/6OD analytical spectrometer operated at 60MHz. The probe temperature was controlled to :_0.5° with a Varian Model V-6040 tem- perature control. Temperatures were determined to :_0.5°, by use of a capper-constantan thermocouple inserted just below the sample tube and above the heat sensor. Magnetic sweep widths were calibrated by the audiofrequency side band technique. Tetramethylsilane was used as an internal standard, and chemical shift values were determined by positioning the side band of TMS over the resonance of interest. Integration of resonance lines was performed by planimetry. All spectra were recorded at a radiofrequency field strength well below the value necessary to observe the onset of saturation. F. Preparation of Solutions for IR and NMR Studies All solutions used in the infrared studies were prepared in a nitrogen filled glove bag and transferred to liquid cells while in an N2 atmosphere. Sample tubes for the nmr studies were dried as described in sec- tion II B. When possible, solutions were prepared in a nitrogen filled glove bag, and the tubes were sealed with a flame. Solutions of silane and chlorodifluromethane were prepared by distilling the solvent to the nmr tube on the vacuum line. All solvents were dried as described in section II A. Samples were used immediately after preparation since decomposition due to hydrolysis occurred upon ageing, even in a sealed tube. III. RESULTS AND DISCUSSION A. Preparation and Characterization of Cyclic Organosilyl Acetylacetone Derivatives The chlorosilanes used as starting materials in the pre— paration of the cyclicorganosilyl acetylacetone derivatives were synthe- sized by means of grignard reactions. Methylchlorosilacyclobutane was prepared by the ring closure reaction of methyl (3-chloropropy1)dichloro- silane with magnesium (Equation 9). Earlier work has shown that in (9) CICHZCHZCHZSi(CH3)(C1)2 + Mg + ClMgCH CH CH Si(CH3)(C1)2 4 2 2 2 <:::Si(CH3)Cl + Mgc12 order to obtain reproducibility good yields, increasing the surface area of the magnesium, as well as chemical activation is essential.28’33 Methylchlorosilacyclopentane and methylchlorosilacyclohexane were both prepared by the reaction of the appropriate difUnctional grignard reagent with methyltrichlorosilane (Equation 10). The presence of (10) Bng(CH2)ngBr + (CH3)Si(C1)3 + (CH Si(CH3)(C1) chlorine atoms on the silicon facilitates the ring formation.29 The reaction proceeds in higher yields with silicon tetrachloride as a substrate, but subsequent replacement of a chlorine with a methyl grignard produced only small amounts of the desired product. In all three cases the ether solvent was removed by slow distillation through a Vigreaux column. A substance co-distilled with the ether, 23 24 if the distillation was not executed slowly. This substance hydrolyzed readily on contact with air to give an acidic gas. The cyclic organosilyl acetylacetone derivatives were prepared by the reaction of the cyclic chlorosilane, acetylacetone, and pyridine (Equation 11). An analogous reaction was originally described by West34 (11) CH3$i(CH2)xCl + H(acac) + py + CH381(CH2)x(acac) + py - HCl for the preparation of (CH3)SSi(acac). To avoid ring-opening side reac- tions, the procedure was modified by keeping the reaction mixture at room temperature instead of heating it to reflux temperature. The compounds were pale yellow liquids which hydrolyzed on con- tact with moist air. The silacyclohexane and dislacyclopentane deriva- tives, after ageing at room temperature for 2 months in a vial sealed under nitrogen and stored in a desiccator, showed only a small amount of decomposition, as judged by nmr spectrosc0py. The silacyclobutane derivative was appreciably less stable even when highly purified. Inspection of the nmr spectrum after ageing at room temperature under the above conditions, indicated that more than one decomposition reaction occurs. Silacyclobutanes are known to be much less thermally stable than the 5 and 6 member ring compounds and will undergo ring opening reactions to form polymers.35 The decomposition products are high boiling and may be separated from the compound by distillation. In addition to a probable polymerization reaction, the silacyclobutane derivative under- goes a rearrangement reaction which is catalyzed by unknown impurities, 25 but is observed to occur at a lower rate even in highly purified samples. Analysis of the rearrangement product, which was purified by fractional distillation, has shown it is a monomer with the same stoiciometry as CHSSi(CH2)3(acac). The ambient temperature nmr spectrum of the product in CCl4 solution is shown in Figure 1 along with the spectrum of CHSSi(CH2)3(acac). The spectrum contains singlets at 19.78 and 18.79, a doublet centered at 18.23 (1:.95 Hz), a quartet centered at 14.80 (1=.9S Hz) and a complex series of lines between 19.4 and 18.2. There is no temperature dependence of the nmr spectrum between 120° and -80°C. In the infrared spectrum of the rearrangement product (Figure 2) there is 1 a strong band at 1657 cm' , which is within the proper region fer an -1 36 isolated double bond, usually found between 1680-1620 cm There is a weaker band at 1586 cm.1 which may be assigned to the coordinated alkoxide, usually fbund between 1500 cm.1 and 1600 cm'l.34 A Si-O frequency at 1050 cm.1 and a Si-C stretch at 840 cm"1 are also present. No change was observed in the nmr spectrum of the compound after it had aged for 2 months at room temperature. A structure consistent with the above data is shown below. The nmr lines may be assigned as follows: the Si-CH3 singlet at 19.78, Figure 1. 26 Proton nmr spectra (60 Mhz) of A) CH38i(CH2)3(acac) in CCl4 (25 m1/100 m1 solvent), B) the thermal rearrangement product of CHSSi(CH2)3(acac) in CCl4 (25 m1/100 m1 solvent), and C) the methanolysis product of the rearranged form in CCl4 (25 m1/100 m1 solvent). 27 o._o~ Figure 2. 28 Infrared spectra of A) CH351(CH2)3(acac) in CHZCIZ, (5 m1/100 m1 solvent), B) the thermal rearrangement product of CH38i(CH2)3(acac) in CCl4 (S ml/100 ml solvent), in the region 1800 cm'1 to 1400 cm'l. 30 the =CCH3 singlet at 18.23, the -CH3 doublet at 18.79 and the =CH quartet at 14.80. In the silyl enol ethers previously studied, the coupling between the =CH and allylic CH3 protons was ~0.4Hz or less and the coupling between =CH and COCH3 protons was W0.6Hz.37 The first step in the rearrangement of CH3Si(CH2)3(acac) must be a ring opening reaction. These ring-openings are catalyzed by trace amounts of HCl, H20, and a variety of other nucleophiles, and they may also be thermally induced.38 Therefore, in successive preparations of CH Si(CH2)3(acac) 3 there was a large variation in the amount of rearrangement product observed, depending on the purity of the sample. Once the ring opens, it could attack the acac moiety by a Michael type addition illustrated below. To further verify the structure, the compound was allowed to react for 24 hours at room temperature with dry methanol, £011owed by removal of the methanol under reduced pressure. The nmr spectrum 0 '1: CH CH | 3 / 2\ / 3 C\Hz/31\R' 0\ ——————> C E\ CH \ 0 \_ 2 0 2 \ é-f/C CH3 CH/ ('3 \c—cu "' 3 01(3) \H {CHz/(IKC/ CH H M 31 illustrated in Figure 1 has a Si-OCH line at 23,, 16.6 of intensity 3 equal to the Si-CH line at 22,, 19.95. The =CH line and the acac CH 3 lines are replaced by a line at 17.6 and a complex multiplet at 18.0. 3 The solvolysis is illustrated in reaction 12. In general the reactivity of the Si-OR bond increases as the pKa of the conjugate acid of -OR decreases.39’4O (12) /Ei\o + CHSOH ——) /sAi\ I / 3 CH l H 2 \ C " CH -—— u— \ /C \ H \CH2/ \ 2 C CH3 CH CH CH3 Nuclear magnetic resonance spectroscopy in conjunction with infrared spectroscopy confirms the structural similarity of the cyclic derivatives to acyclic R Si(acac).6 The lines observed in the nmr spectrum at 3 ambient temperature are listed in Table VI. The acetylacetone moiety give rise to a =CH-proton multiplet near 14.6 and a =CH- singlet near 14.8, two methyl doublets near 17.8 and 18.0, and a methyl singlet near 18.0. Si-CH3 lines are present near 19.6. The ring methylene protons exhibit a complex set of lines between 18.5 and 19.5. The presence of two =CH lines indicates the presence of cis and trans isomers. The acetylacetone lines near 19.6, 8.0, 7.8 and 4.6 are to the cis isomers 32 .u:o>Hom HE ooH\m oH ma :ofipmeueoocou no mononomoo scum :oxmp uqsomeoo many new upon .0 .oocmeomon Hanoos vomauo>a mafia .a .:0Hu:Hom «How cw mauoomm muco>aom HaooH\HEmm ma :oflumaueoocoo “mosflm> e we counomoe mpmwnm HH< .m Hm.e no.m va.m mm.4 N¢.A oo.m ea.m afloauavammfinmuu ~m.e 00.x ma.m we.4 ow.a oo.m ma.m fiuauavmnmzuvumnmu oa.e ao.w mo.m mm.e Hw.a mm.a mo.m floaoavefiwzovammmu ~m.e mo.w no.a oa.e ma.a mm.a mm.m noaoavmnmxuvammxu _.:uu ammo mzu-um :uu mzooo mzu- mmu-um eesoaaou Kill .8503 a llll\ / noEomH asap. 1! 1\ mo>mue>mnoo occuoomaxuoo< HxHflmocmmno owfloxu . mcmhh mam mwu how mama wwwgm Hmowsonu :ouona .H> oHan 33 Figure 3. Proton nmr spectra of A) CH35i(CH2)4(acac) in CCl4 (25 m1/100 m1 solvent), B) CHSSi(CH (acac) 2)S in CCl4 (25 m1/100 m1 solvent). SE“. . O n QFOP CW CW 0* O. 0_ CW . . é. ..; if if C . 34 35 in accordance with the assignments made for (CH3)SSi(acac).37 The single acac methyl line in the gl§_isomer arises from rapid rearrangement, which will be discussed in section III E. The Si-CH3 lines of the glg and Ergp§_isomers are accidentally coencident at ambient temperature, but can be resolved in each case as the temperature is lowered. Separation of gl§_and £3223 - (CH3)3Si(acac) by gas chromatography and distillation through a spinning band column was unsuccessful, suggesting a rapid equilibrium between the isomers.37 Facile isomerization is supported by the fact that the same gls to ££§g§_ratio's were observed for samples of acyclic R35i(acac) derivatives immediately after distillation, and after ageing for 6 months at room temperature.27 CHSSi(CH2)4(acac) and CH35i(CH2)S(acac) were allowed to age for 2 days before determining the g§§_to Eggpg_ratio, but the gl§_to ££§§§_ratio of CH38i(CH2)3(acac) was determined immediately after distillation. The Ei§.t° l£§2§_ratios shown in Table VII were determined by plani- metric integration of the =CH lines. After the samples had aged 2 months at room temperature, the gl§_to ££§g§_ratios of the compounds remained con- stant. The amount of gl§_isomer observed increases as the ring size decreases, with a large increase in cl§_isomer observed for the four member ring compound. The magnitude of the change in the amount of gl§_isomer due to ring size is much greater than that observed due to inductive effects of silyl substituents (g£,, Table III). The data suggest a relationship between ring strain and the strength of the incipient silicon-oxygen interaction. 36 Equilibrium Ratio of Cis and Trans Enol Ether Isomers Table VII. for CHSSi(CH2)x Derivatives of Acetylacetone? Compound [cis]/[trans] CH381(CH2)3(acac) 2.23 CH3$1(CH2)4(acac) 0.45 CH351(CH2)5(acac) 0.25 . b (CH3)381(acac) 0.34 In carbon tetrachloride at ambient temperature; concentration is 25 ml/100 m1 of solvent. Value for this compound was taken from reference 6; the solvent is chlorobenzene; concentration is 0.6 m, 37 The structure of silacycloalkanes have been studied by electron diffraction.4l'43 The ring in silacyclobutane is highly puckered to allow adjacent CH2 groups to be staggered. The relative importance of angle strain in the carbon skeleton is lessened due to the deformation of the C-Si-C angle. Silacyclobutane and dichlorosilacyclobutane both were studied and found to have C-Si-C angles of 80°. The exocyclic angle was 109° and 105°, respectively.41 The deformation of the C-Si-C angle causes an alteration of the hybrid character, as determined by the exocyclic Si-H infrared stretching frequencies. There is an increased amount of 5 character to the exocyclic bonds.38 The amount of ring strain in silacyc10pentane is not as great as the strain pre- sent in the four-member ring.35 The ring is bent and the C-Si-C angle is 96°.43 The six member ring in 1,3,5-trisilacyclohexane has C-Si-C angles of 109.5° and adapts the chair conformation.43 Relief of ring strain by incipient pentacoordination and distortion towards a trigonal bipyramid ground state would strengthen the silicon-oxygen interaction in the feur and five member ring compounds. This effect is greatest for the four member ring. The reluctance of the six member ring com- pound to distort its tetrahedral angles is illustrated by the smaller amount of cis isomer relative to the acyclic derivative. Selective vibrational frequencies fer the cyclic organosilyl acety- lacetone derivatives are reported in Table VIII. The frequencies observed are in good agreement with the previously reported values for acyclic R38i(acac) compounds.6 The compounds all have a strong band 38 Table VIII. Infrared Absorbtion Bands for Cyclic Organosilyl . . a Acetylacetone Derivatives. Compound y c=o y C=C Si-O b c -1 -1 -1 (CH3)SSi(acac) 1678(1659) cm (1625)1588 cm 1032 cm d -1 -1 -1 CH38i(CH2)3(acac) (1677)1662 cm (1622)1589 cm 1044 cm CH38i(CH2)4(acac) 1680 cm‘1 (1625)1589 cm'1 1044 cm'1 . -1 -1 -1 CH381(CH2)S(acac) 1678 cm (1620)1587 cm 1042 cm a. In carbon tetrachloride unless otherwise stated; concentration is 5m1/100ml solvent. b. Data for this compound were taken from reference 6. c. Weak bands or shoulders are given in parentheses. d. Frequencies for this compound were recorded in dichloromethane. C 1.. II )- 1 Figure 4. 39 Infrared spectra in the region 1800—1400 cm"1 for A) CH38i(CH (acac) in CHZCI2 (S ml/100 ml 2)3 solvent), B) CH35i(CH2)4(acac) in CCl4 (5 ml/100 ml solvent) C) CH3Si(CH2)5(acac) in CCl4 (S ml/lOO ml solvent). 40 P 1 ‘ I 1800 1 I 1600 1400 CM" 41 between 1600-1700 cm.1 corresponding to the uncoordinated carbonyl stretching frequency. For (CH3)SSi(acac) the band at 1678 cm.1 is assigned to the Egggg_isomer with the shoulder at 1659 cm'1 assigned to the gl_s_isomer.44 In the silacyclobutane derivative, the higher energy C=0 stretch of the Eggn§_isomer appears as a shoulder on the strong 1662 cm.1 band of the gl§_isomer. The C=0 stretch of the gl§_ isomers of CH3$i(CH2)4(acac) and CH38i(CH2)S(acac) is not resolved from the C=0 vibration of the ££§n§_isomer. The presence of a Si-O group in each compound is indicated by.a strong band in the region 1000-1050 cm-1. The strong band at 23,, 1589 cm-1 is attributed to the C=C vibration. The intensity of this band is enhanced by conjugation with the carbonyl .34‘36 It has been suggested that the incipient silicon-oxygen group interaction should cause a shift in the carbonyl vibrational frequency to lower frequencies. The C=0 band in c»ketosi1anes is shifted 68-70 cm"1 to longer wavelengths relative to their carbon analogues, and in B-ketosilanes the band is shifted l9 cm-l. The C20 stretch of 6-ketosilanes is not shifted relative to their carbon analogues.45’46 It has been suggested that the 18 cm.1 low energy shift in the C=0 stretch of (CH3)SSi(troPolone), relative to free tropolone may be due to incipient pentacoordination in the silyl derivative.9 An infrared study of several trimethylsilyl derivatives of ligands where incipient pentacoordination may be anticipated44 led Sommer2 to the conclusion that no incipient silicon-oxygen interaction was occurring. Included in this study was (CH3)SSi(acac). The gl§_and Ergn§_carbonyl bands for (CH3)SSi(acac) and for the analogous methyl enol ether are reported equal.44 Variations in the extent of incipient pentacoordination in 42 silyl enol ethers as reflected by cis to trans ratios does not correlate with their relative C80 stretching frequencies. For example, the C=O 1 1 band 19.212? (CH3)381(acac) is at 1659 cm- and at 1662 cm' for CH38i(CH (acac). The shift is small and in a direction Opposite to 2)3 that which would be predicted on the basis of the gl§_to trans ratios. B. Preparation and Characterization of (CH3)SSi(hfac) and (CH3)3Si(dpm) Trimethylsilylhexafluroacetylacetone (CH3)SSi(hfac), prepared by the reaction of neat trimethylchlorosilane and hexafluroacetylacetone (Equation 13). Trimethylsilyldipivaloylmethane, (CH3)SSi(dpm), was (13) (CH3)SSiC1 + H(hfac) + (CH3)38i(hfac) + HCl prepared by the reaction of trimethylchlorosilane and sodium salt of dipivaloylmethane in hexane (Equation 14). A related compound, (14) (CH3)SSiCl + Na(dpm) e (CH3)38i(dpm) + NaCl trimethylsilyldiisobutylmethane was prepared by Alan Schwartz by the same method used to prepare (CH3)35i(dpm). The compounds were color- less to pale yellow liquids. They are hydrolyzed on contact with moist air. The nmr spectrum of (CH3)3Si(hfac) exhibits a single Si-CH3 line 19 at 19.65 and a single sCH- line at 13.70. The F nmr spectrum showed two sharp lines of equal intensity assigned to the two non-equivalent CF3 groups in the trans isomer. The chemical shift difference between the 43 fluorine lines is preportional to the magnetic field strength (442.5 Hz at 94.6 MHz and 260.0 Hz at 56.4 MHz), which indicates that the two lines are singlets and not an anomalous doublet.27 The nmr spectrum of (CH3)3Si(dpm) exhibits a singlet Si-CH line at 19.65 and a single =CH- 3 line at 13.70. The lfbutyl groups on dipivaloylmethane gave rise to a single resonance at 18.91. The presence of only one £7butyl line is (cu ) Si 0 0:: . 0 3 3"" ~\\\ ’,/C--CF3 (c113)3 si—-o \» C:='G \ C-C4H9 CE; \\‘H j__— I, . t-C4H9 (CH3)351(hfac) (CH3)SSi(dpm) - CH .___ 0\' 3 (”19351 0\ >c—c —«H C==: \\ C/ C\H CH3 / I CH3 H\\CH3 (CH3)3Si(dibm) attributed to the presence of only the gl§_isomer, which is undergoing a rapid stereochemical rearrangement (g£,, section IIIE). The nmr spectrum of (CH3)3Si (dibm) exhibits lines characteristic of both £i_s_ and £332§_isomers.47 0n comparison of the gl§_to £I§2§_ratios to tri- methylsilylacetylacetone, we find that as the methyl;hsreplaced by the more electron releasing isopropyl group the glg to £3§g§_ratio increases from 0.34 to 0.55 (Table IX). A further increase in electron releasing ability by replacement with Efbutyl groups results in only the cis isomer 44 Table IX. EQuilibrium Ratios of Enol Ether Isomers for Trimethylsilyl- B-diketonates.a Compound [cis]/[trans] (CH3)38i(hfac) R = CF3 < 0.02 (cu ) Si (acac)b R = CH 0 34 3 3 3 ' (CH3)3Si(dibm) R = CH(CH3)2 0.55 (CH3)SSi(dpm) R = C(CH3)3 >50.0 a. In carbon tetrachloride at ambient temperature, concentration is 25 ml/100 ml solvent. Data fer this compound were taken from reference 37; solvent is chlorobenzene; concentration is 0.60 m, 45 being observed by nmr spectroscopy. Steric effects, especially in the case of lfbutyl groups, should favor the gl§_isomer. Steric effects, however, do not influence the gl§_to £5325 ratio of the hexafluroace- tylacetone derivative where only the ££§p§_isomer is observed. It is probable that the electrostatic interaction between silicon and the uncoordinated oxygen is important in determing these gl§_to Eggn§_ ratios. The infrared spectrum of (CH3)SSi(hfac) exhibits a single C=C 1 1 stretching frequency at 1733 cm- , a C=C stretching frequency at 1625 cm- , and a 31-0 stretch at 945 en’l. (CH3)33i(dpn) exhibits a single c=0 stretch at 1676 cm'l, a C=C stretch at 1625 cm-1, and a 81-0 stretch at 1100 en'l. The presence of only one C=C frequency in each case is attributed to the (CH3)3Si(hfac) existing exclusively as the 23223. isomer and to (CH3)3Si(dpm) as the gl§_isomer. It has been suggested that the variation in the Si-O vibrational frequency from the normal range of 1000-1050 cm-l, is due to the inductive effects of the CF3 and £7butyl groups on the Si-O bond strength.27 C. Preparation and Characterization of (C6HSCH2)(CH3)(C6H5)Si(dpm) In an effbrt to synthesize a compound in which diastereotopic environments could be observed in the proton nmr spectrum, several different chiral triorganochlorosilanes were prepared. The chlorosilanes were prepared by the reaction of methylphenyldichlorosilane and the appropriate Grignard reagent. The chlorine was replaced with acac or dpm as described previously. Compounds of the type R(CH (C6H5)Si(acac) 3) 46 were prepared where R was equal to 2-propanol, 2-d-2-propanol,48 2,6-dimethylphenyl; and benzyl groups. When R was a benzyl group, it appeared that the diastereotopic methylene protons were present in the nmr spectrum, but they overlapped with the acac methyl lines. The deuterated acac derivative,49 (C6H5CH2)(C6H5)8i(acac-d7), provided a better resolved methylene pattern, but the spectrum was still complex due to the presence of Eli and £3323_isomers. The nmr spectrum of (C6H5CH2)(CH3)(C6HS)Si(dpm) is simplified by the presence of only the £2§_isomer. The compound was prepared as illustrated in equations 15 and 16. (15) CGHSCHz-Mg-Cl + (CH3)(C6HS)SiC12 + MgCl2 + (CGHSCH2)(CH3)(C6H5)SiCl (16) (C6HSCH2)(CH3)(C6HS)SiCl + Na(dpm) + (C6H5CH2)(CH3)(C6HS)Si(dpm) + NaCl The compound was a pale yellow viscous liquid that hydrolyzed in con- tact with moist air. After ageing for over a year at room temperature, the compound remained unchanged, as judged from nmr spectroscopy. The nmr spectrum of gl§:(C6H5CH2)(CH3)(C6HS)Si(dpm) contains a Si-CH3 line at 19.56, a £;C4H9 line at 19.00 and a =CH-line at 14.30. The C6Hs multiplets are centered near 13.0. The benzyl methylene protons are diastereotopic and exhibit an AB pattern centered at 17.23 (Av = 11.0 Hz, J a 13.5 Hz). The spectrum is illustrated in Figure 5. The intensity ratio of the central to the outer bands of the AB pattern is calculated to be 7.8:1, which agrees with the experimentally determined ratio. 47 ‘16.. .. -3 I ! Figure 5. Proton nmr spectra of A) (C6HSCH2)(CH3)(C6HS)Si(dpm) in CCl4 (25 m1/100 ml solvent), B) Benzyl methylene proton AB pattern in CCl4 (50 ml/100 ml solvent). 48 '- 0.0 0d 32.... cs 06 49 The infrared spectrum exhibits a carbonyl stretching band at 1675 cm- a C=C stretch at 1600 cm.1 and a Si-O stretch at 1100 cm'l. The spectrum is similar to that obtained for (CH3)3Si(dpm). D. Attempted Preparation of l,3,5,7-tetrasi1aadamantane Derivative The tetrasilaadamantane cage is prepared from 1,1,3,3,5,5- hexamethylsilacyclohexane by a ligand redistribution reaction catalyzed by AlCl3 (Equation 17). The AlCl3 also acts as a reactant, giving rise X I . Si (17) 2[CH Si(CH) ] “C13 2(CH ) Si + i/ 31—43” 2 3 2 3 -————-> 3 4 . 3 CH-—-Si 3 . 8.1- CH3 to chlorine in place of a methyl group. Massive amounts of AlCl between 3! 20—200% of the [CHZSi(CH3)2]3 weight, is used to produce facile redis- tribution at a moderate temperature.31 Frye purified the silaadaman- tanes by preparative glc. However, in the present work vacuum sublima- tion at room temperatures provided an easy purification route. Although fractional sublimation of the tetramethyl and trimethyl-chloro adamantanes appeared possible, the mixture was used without further purification. The 1,1,3,3,5,5-hexamethyl 1,3,5—trisilacyclohexane required as a starting reagent was prepared from chloromethyldimethylchlorosilane by a Grignard reaction (Equation 18). (18) (C1CH2)(CH3)251C1 + Mg + [CHZSi(CH3)2]3 + MgCl2 1 50 The tetramethyl and trimethylchloroadamantes were identified by nmr spectrosc0py and infrared spectroscoPy.32 The nmr spectrum of the sub- limed crystals exhibited lines at 19.74, 19.83 and 110.22 that are character- tistic of the silaadamantane with one Si-Cl functional group. The presence of the tetramethyladamantane was indicated by lines at 19.90 and 110.28. The infrared spectrum contained characteristic bands at 527 cm-l, 1027 cm-1, and 1248 cm-1, corresponding to Si-Cl, Si-CHZ-Si, and Si-CH3 frequencies, respectively. As an additional check for the Si-Cl functional group, AgCl was precipitated from a nitric acid solution of the crystals. The purpose of preparing the silaadamantane was to determine whether a reaction between the Si-Cl functional group and a B-diketone was possible and whether the product would undergo an intramolecular rearrange- ment. The mixture of silaadamantanes was allowed to react with the sodium salt of dpm in cm4 and also with Hacac and pyridine in CC14. The reactions were followed by nmr spectrosc0py. After a reaction time of one week at room temperature, the adamantane-Na(dpm) mixture gave rise to a new nmr line at 19.93, in addition to the original sila- adamantane lines. The reaction with Hacac and pyridine also gave rise to a product exhibiting in 19.93 line, but no lines characteristic of the acac moiety were present. Thus neither reaction mixture provided evidence of the desired product. The reaction of B-diketones and chloro- silanes proceeds with inversion of configuration,24 probably involving backside attack. The silaadamantane cage structure blocks a backside approach. 51 E. Kinetic Study of the Intramolecular Rearrangement of Eié‘ triorganosilyl-B-diketone Derivatives The temperature dependence of the nmr Spectrum of the com- pounds prepared for this study was attributed to an intramolecular rearrangement involving migration of the silyl group between the two oxygens in the B-diketone (see Eq. 6). The rate of the rearrangement was calculated by the simplified Gutowsky—Holm equation for the . coalescence of singlets (Equation 19).50 The simplified equation is , based on the assumption that 1A = 1B and T2Av is large. The general 1 (19) :l- a kc = ( fl ) Av where TA is the mean lifetime (sec) 1 A JP2' of a prbton at site A k is the rate constant (sec-1) cal- culated at the coalescence temperature Av is the frequency separation (Hz) in the absence of exchange validity of the approximate equations have been tested by a comparison to complete line shape analysis.51’52 Values for kc obtained by com- plete line shape analysis for equally intense coalescing singlets or doublets are within 6% of the rate obtained using the approximate equation, when Av is 20 Hz.51 An error of 6% for kc will result in an error for AGt of less than 0.1 kcal/mole at the coalescence temperature. The value of Av for the silyl dipivaloylmethane derivative was taken to be equal to the limiting frequency separation in the region of slow exchange. The frequency separations at the coalescence temperatures for the acac derivatives were extrapolated from the temperature dependences of Av in the region of slow exchange. The activation free energies were 52 calculated from Equation 20. 1 RT -AG (20) k = fi- exP (117-) The interchange of non-equivalent methyl groups in cis-RSSi(acac) derivatives (g£,, Table IV), is sufficiently slow to permit observation of two well resolved methyl proton resonance lines below -40°. As the temperature is raised the two lines broaden and merge into a very broad line, which then sharpens above the coalescence temperature. The activation parameters for the rearrangement of gl§7(CH3)38i(acac) have been previously determined by line broadening analysis.6 A first order 1 rate constant of 851 sec- , an activation energy of 13.8 I 0.5 kcal/mole and an activation entrepy of -0.8 :_2.5 e.u. were feund at 25°. (CH3)3Si(dpm) rearranged at a much greater rate than the corresponding acac derivative. In ClF CH solution the Efbutyl line remained sharp 2 even at —137°. A lower limit to the first order rate constant was estimated based on the Av observed in the region of slow exchange for . o t . . (CGHSCHZ)(CH3)(C6H5)Si(dpm). At -l37 , AG for (CH3)381(dpm) is 1 <6.7 kcal/mole. If one assumes AS so is >7.76 107 sec'l. is zero, as in the case of (CH3)3Si(acac), then k At the estimated coalescence temperature 2 (~95°) for (CH3)SSi(dpm), AG* is less than 9.4 kcal/mole. (CH3)3Si(hfac) exhibits only the £5§E§_isomer which is stereochemically ridged at tem- peratures up to 40°C. These results show that when the methyl group on the B-diketone is replaced by an isopropyl or a £:butyl group, the activation free energy for the rearrangement decreases, by at least 4 and 7 kcallmole, respectively. This decrease is most likely due to an 53 increase in electron density at the uncoordinated carbonyl oxygen. The magnitude of the effect is comparable to the effect of substitution on pyrazole in the rearrangement of (CH3)SSi(pyrazole)10 (g£,, Table IV), and supports the importance of the long range electrostatic interaction between the silicon and the uncoordinated carbonyl oxygen. Figure 6 illustrates the possible reaction pathways and the stereo- fr chemical result of each path. Attack by the uncoordinated oxygen at an 1 opposite tetrahedral edge will result in an intermediate in which the d diketonate spans two equatorial positions. This structure is unfavorable because two electronegative atoms are in equatorial positions4 and is because of the O-Si-O bond angle of 120°. Lingafelter and Braun have shown that acac normally assumes an O-M-O bond angle of 90°.53 Attack at an opposite edge lead to inversion of configuration at silicon. Attack of the uncoordinated oxygen at an adjacent tetrahedral edge or face will result in an intermediate with the diketonate spanning axial equatorial position, and will result in retention of configuration at silicon. Pseudorotation using an R group as a pivot will not alter the stereochemical result of the latter intermediate, however use of an oxygen as a pivot will interchange the two intermediates. The stereochemical consequences of the rearrangement process was determined by an nmr investigation of the chiral compound glg; (C6H5CH2)(CH3)(C6HS)Si(dpm). The temperature dependence of the nmr spec- trum is illustrated in Figure 7. The Erbutyl groups are resolved into two lines of equal intensity below -100°. The lines coalesce at -88.4 and sharpen to a single peak at -60°. The AB pattern of the benzyl- methylene protons is retained in the region of fast exchange indicating S4 Figure 6. Mechanisms which would account for the exchange of the non-equivalent R groups of a B-diketone in cis-R38i (B-diketone). nge of 1‘62 (S is-R O 09 O 1“.“ .lil: Figure 7. 56 Temperature dependence of the proton nmr spectrum of (C6HSCH2)(CH3)(C6H5)Sl(dpm) low temperature spectra were recorded in CIFZCH (12.5 m1/100 ml solvent) high temperature spectra were recorded in chloronaphthalene (12.5 m1/100 ml solvent). 57 t— 175 Hz -) -48.5 °c :88.4 °c 406.4 “C 58 189.0" 0 .f 'e 134.4" 0 643° c 59 retention of configuration. From these data the first order rate constant fer the retention rearrangement process at 25° is 1.1 x 106 sec-1 and the activation free energy is 9.2 kcal/mole. The AB pattern of the benzyl methyl protons persisted to a temperature of 211°C, although the separation between the central bands decreases, causing the two outer bands of the multiplet to dissappear. At 211° the inversion process is not yet observed. A lower limit to the first order rate constant for the inversion process was estimated from the simplified Gutowsky-Holm equation. Equation 21 for coupled AB spin systems is only accurate when (21) k = (41(sz + 6.1211” f. c 2 Av > J. Equation 19 is more accurate for small Av values. The value of Av at 211° was estimated to be 3.6 Hz. The estimate was based on the known coupling constant at ambient temperatures and the observed frequency separation at 211°C, along with the fact that the intensity ratio of the central to the outer bands must be greater than 10:1. The relationship between the intensity ratio and observed frequency separation between the bands (Ic/Io = (v1 - v4)/(v2 - v3) where v1 and v4 are outer bands) may be used to cal- culate Av. The first order rate constant was estimated to be 27 kcal/mole. The difference in AG tion and inversion process is >18 kcal/mole. Similar data have recently been reported for the R Si(acac) rearrangement and R3$i(tropolone) 3 rearrangements studied by C13 nmr spectroscOpy,9 where differences between the retention and inversion processes of >9.2 kcal/mole and >9.9 kcal/ mole, respectively, were found. The lower limit for the difference in 60 AGI values fer inversion and retention processes of (C6HSCH2)(CH3)(C6H5)Si(dpm) is >17.8 kcal/mole. At 25°, the ratio of rate constants for rearrange- ment and inversion is at least 1:0.79 x 1013. Fewer than one rearrange- ment out of every 1013 results in inversion of configuration. Space filling models of (C6HSCH2)(CH3)(C6HS)Si(dpm) indicate a great deal of steric hinderance between the bulky phenyl groups and the dipivaloyl- methane. This factor is probably important in the large difference in observed rates of rearrangement between (CH3)SSi(dpm) and (C6HSCH2)- (CH3)(C6H5)Si(dpm). The glgfcyclic organosilyl acetylacetone derivatives exhibited a single methyl resonance at ambient temperatures (see Figures 1 and 3). As the temperature was lowered the methyl line broadened and was re- solved into two lines of equal intensity below -40° and -100° for the six member and the five member ring compounds, respectively. The tem- perature dependence of the nmr spectrum of the silacyclohexane deriva- tive is illustrated in Figure 8. The methyl resonance in the feur- member ring compound was still sharp at ~142°, at which point the com- pound was no longer soluble in ClFZCH. The compound redissolved on warming to give a normal spectrum. The estimated first order rate constants and the activation free energies are listed in Table X. The rearrangement rate at 25° increased by three orders of magnitude when the six member ring was replaced by a five membered ring. A further increase in k25° of at least two orders of magnitude was ob- served for the four-membered ring compound. The effect of increasing the ring strain is to increase the reactivity of the compound towards 61 Figure 8. Temperature dependence of the proton nmr spectrum of (CH2)SSi(CH3)(acac) in CH Cl2 (12.5 ml/100 m1 solvent). 2 CCCCC s iiii 63 Table X. Kinetic Data for the Intramolecular Rearrangement of cis- Cyclic Organosilyl Acetylacetone Derivativesa d . -1 -1 , 4 -18 Compound Tc( C) AvO,Hz ch,sec k25°,sec AG ,Kcal/mole . . b 8 CH381(CH2)3(acac) <—142 17.15 38.1 >l.05x10 <6.5 CHSSi(CH2)4(acac) -88.8° 17.16 38.1 9.82x10S 9.3 CH38i(CH2)S(acac)c -20.0° 17.14 38.1 2.16x103 12.9 (CH3)SSi(acac) ----' ---- ---- 851 13.8 a. In ClF CH unless otherwise stated, concentration is 12.5 ml/lOO ml solvent. b. Assumed value. c. In CHZCIZ, concentration is 12.5 m1/100 ml solvent. d. Calculated from Eyring equation. e. Extrapolated value at 25°. 64 nucleophilic attack by the uncoordinated oxygen. The decrease in the activation free energy may be attributed to the release of ring strain by formation of a five coordinate intermediate. The rate of hydro- lysis of cyclic organosilanes have been reported to increase from a relative rate of 0.1 for the six member ring compound to 10 for the five member ring compound and to 10,000-100,000 for the four member ring compound.54-56 The effects of ring strain on the reactivity of silicon compounds is analogous to ring strain effects on tetra- valent phosphorous.3’57’58 The most probable intermediate in the rearrangement of Eli? CHSSi(CH2)3(acac) and glngHSSi(CH2)4(acac) is illustrated below. Electronegative atoms normally occupy axial positions and four and TS ‘ C-—H 1* 9 e 0 -'d~——CH /. 0 ,./ 3 ,0 .hv'. 1H2 TIVCH 3 (”WY—CH2 where x = 1 or 2 4’57 The acac is five member rings span axial-equatorial positions. shown spanning an axial-equatorial position. The six member ring compound exhibits a rate of rearrangement comparable to the acyclic silyl derivatives. This result is expected because there is no 65 significant release of angle strain in the six member ring upon for- mation of the intermediate shown below. The affect of ring strain on the rate of rearrangement (i.e. nucleophilic displacement) correlates with the available data on the affect of ring strain on the reactivity of tetravalent phosphorous to nucleophilic attack. 1“: z’C if *9 C.” ICH-CH\ \O-Si’V 2 20H ' CH—CH/ 2 CH3 2 2 Ring strain also affects the stereochemistry of nucleophilic displacement reactions on phosphorous.3 Acyclic systems usually result in inversion, whereas a cyclic system, in which phosphorous is incorporated in a small ring usually results in retention of configuration. Displacement of a chlorine from silicon always re- sults in inversion of configuration, except when the silicon is incorporated in a five member ring and retention of configuration is observed.59 Incorporation of silicon into a six-member ring does not alter the stereochemical consequence of the reaction.60’61 Strictly analogous rearrangements are not known in carbon chem- istry, however a similar degenerate migration of acetyl groups 62-64 between two electronegative atoms has been reported. An example is illustrated in reaction 21.59 The reported activation 57,58 66 CH CH CH | 3 3 _ 1 3 “—0 ""0 C=C) /// ,,/C //’ (21) H"'"C/C \COCH3:H-C '1' .0>C\ "\‘—-’H——C< COCH \Ezo \C'3— CH3 \(E—O/ H3 CH3 CH3 energy is 15 kcal/mole. F. Summary The organosilyl B-diketonates studied all exhibit an open chain enol ether structure. In the ground state of these compounds the silicon atom is incipient pentacoordinate. Two factors are important in determining the degree of incipient pentacoordination: (l) The electron withdrawing or releasing abilities of the silyl and diketo- nate substituents and (2) The amount of angle strain at silicon. The kinetic data for rearrangement of the £33 isomer support a mechanism in which the land making is an important step (llg,, the process is a nucleophilic displacement). The data also indicate the formation of a five coordinate intermediate. The increase in reactivity observed when the angle strain becomes large is due to a lowering of the energy of the intermediate relative to the ground state. The relative AG° values of the ground state of the gl§_isomer, calculated from the El§_to ££§A§_ratios, differ by 1.3 kcal/mole when the ring size is decreased from six members to four members, whereas the AGI values for rearrangement, differ by more than 6.4 kcal/mole. Pseudorotation about an equatorial oxygen atom as a pivot does not occur at any appreciable rate, but there is no evidence to exclude 67 pseudorotation about an equatorial alkyl group as a pivot. In the absence of pseudorotation the displacement would require axial attack of the uncoordinated oxygen atom and equatorial departure of the leaving group. Although this does not violate the principle of microscopic reversibility,6s it is chemically improbable since apical attack and departure should be the perferential modes of bond making and bond breaking. Apical bonds are longer and weaker than the equatorial bonds.3 If one assumes apical attack and departure, then pseudorotation in the intermediate is required. BIBLIOGRAPHY 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. BIBLIOGRAPHY M. Gielen, C. Dehouch, H. Mokhtar-Jamai, and J. Topart, Reviews on Silicon, Germanium, Tin and Lead Compounds, l, 9 (1972). L. H. Sommer, "Stereochemistry Mechanism and Silicon," McGraw-Hill Inc., New York, (1965). K. Mislow, Accounts Chem. Res., g, 321 (1970). , ’1 E. L. Muetterties, W. Mahler, and R. Schmutzler, Inor . Chem., :_ g, 613 (1963). ”.421 _-‘:- . m2 (- J. J. Howe and T. J. Pinnavaia, J. Amer. Chem. Soc., gl, 5378 (1969). T. J. Pinnavaia, W. T. Collins, and J. J. Howe, J. Amer. Chem. Soc., 22, 4544 (1970). ‘xame H. Shanan-Atidi and Y. Shvo, Tetrahedron Lett., 2, 603 (1971). Y. N. Kuo, F. Chen, and C. Ainsworth, J. Chem. Soc. D, 137 (1971). H. J. Reich and D. A. Murcia, J. Amer. Chem. Soc., BE, 3418 (1973). D. A. O'Brien, and C. P. Hrung, J. Organomet. Chem., 32, 185 (1971). N. Wiberg and H. J. Pracht, Chem. Ber., 105, 1388 (1972). O. J. Scherer and P. Hernig, Chem. Ber., 101, 2533 (1968). R. West and B. Bechlmeir, J. Amer. Chem. Soc., 2A, 1649 (1972); and references therein. M. Fukui, K. Itoh and Y. Ishii, J. Chem. Soc. PII, 1043 (1972). J. Pump and E. G. Rochow, Chem. Ber., 21, 627 (1964). J. F. Klebe, J. Amer. Chem. Soc., 99, 5246 (1968). J. F. Klebe, Accounts Chem. Res., g, 299 (1970). F. P. Boer and F. P. Van Remoorten, J. Amer. Chem. Soc., 22, 801 (1970). H. P. Fritz and C. G. Kreiter, J. Organomet. Chem., A, 313 (1965). 68 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 69 . Davison and P. E. Rakita, Inorg. Chem., 2, 289 (1970). . J. Ashe, Tetrahedron Lett., 2A, 2105 (1970). . Kwart and ;L Slutsky, J. Amer. Chem. Soc., 2A, 2515 (1972). SEE>> . M. Frey and R. Walsh, Chem. Rev., 22, 103 (1969). T. K. Kusnezowa, K. Ruhlmann, and E. Grundemann, J. Qggpnomet. Chem. , 47,53 (1973). A. G. Brook, D. M. MacRae, W. W. Limburg, J. Amer. Chem. Soc., 22, 5494 (1967). A. G. Brook and J. D. Pascoe, J. Amer. Chem. Soc., 22, 6224 (1971). Ward T. Collins, M.S. Thesis, Michigan State University (1970). R. Damrauer, R. A. Davis, M. T. Burke, R. A. Karn, and G. T. Goodman, J. Organomet. Chem., A2, 121 (1972). R. West, J. Amer. Chem. Soc., 22, 6012 (1954). W. A. Krines, J. Organomet. Chem., 22, 1601 (1964). C. L. Frye, J. M. Klosowski, and D. R. Weyenberg, J. Amer. Chem. Soc. , 92, 6379 (1970). C. L. Frye, J. M. Klosowski, J. Amer. Chem. Soc., 2A, 7186 (1970). V. M. Vdovin, N. S. Nametkin, and P. L. Grinberg, Dokl. Akad. Nauk SSSR., 150, 799 (1963). R. West, J. Amer. Chem. 529., 22, 3246 (1958). K. A. Andrianov and L. M. Khananashvili, Organomet Chem. Rev. ,.2, 141 (1967). R. T. Conley, "Infrared Spectroscopy, Allyn and Bacon Inc., Boston (1972). J. J. Howe, Ph.D. Thesis, Michigan State University, East Lansing, Michigan, 1971. R. Damrauer, Organomet. Chem. Rev. A, 2, 67 (1972). E. Akerman, Acta. Chem. Scand., ll, 373 (1957). R. C. Mehrotra, International Symposium on Organosilicon Chemistry, Prague, 1965. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 70 L. V. Vilkov, V. S. Mastryukov, Y. V. Vaurova, V. M. Udoven, and P. L. Grinberg, Dokl. Akad. Nauk SSSR, 177, 1508 (1967). J. Laane and R. C. Lord, J. Chem. Phys., A2, 1508 (1968). K. G. Dzhaparidze, Canadidates Thesis, Moscow, 1955, see ref. 35. W. H. Knoth, Ph.D. Thesis, the Pennsylvania State University, University Park, Pennsylvania 1954. A. G. Brook, M. A. Zuigley, G. J. D. Peddle, N. V. Schwartz and 2 C. M. Warner, J. Amer. Chem. Soc., 22, 5102 (1960). « A. G. Brook and J. B. Pierce, Can. J. Chem., 52, 298 (1964). 1 Private communication, Alan Schwartz, Michigan State University, East Lansing, Michigan. W1.nu~_.a-..1 nut.‘ J. Frye, Ph.D. Thesis, Michigan State University, East Lansing, Michigan. E. D. Bailer, M.S. Thesis, Michigan State University, East Lansing, Michigan, 1969. H. S. Gutowsky and C. H. Holm, J. Chem. Phys., 22, 1228 (1956). D. Kost, E. A. Carlsen, and M. Raban, Chem. Commun., 656 (1971). T. J. Pinnavaia, J. M. Sebeson, II, and D. Case, Inorg. Chem., 2, 644 (1969). F. C. Lingafelter and R. L. Braun, J. Amer. Chem. Soc., 22, 2951 (1966). L. A. Sommer, O. F. Bennett, P. G. Campbell, and D. R. Weyenberg, J. Amer. Chem. Soc., Z2, 3295 (1957). L. H. Sommer and (1 F. Bennett, J. Amer. Chem. Soc., 22, 1008 (1957). L. H. Sommer, W. P. Baree, Jr., and D. R. Weyenberg, J. Amer. Chem. Soc., 2l, 251 (1959). F. H. Westheimer, Accounts, Chem. Res., l, 70 (1968). R. F. Hudson and C. Brown, Accounts Chem. Res., 2, 204 (1972). D. N. Roark, L. H. Sommer, J. Amer. Chem. Soc., 22, 969 (1973). 60. 61. 62. 63. 64. 65. 71 R. Corriee and J. Masse, Bull. Soc. Chim. Fr., 3491 (1969). H. Sakurai and Murakami, J. Amer. Chem. Soc., 2A, 5081 (1972). V. I. Minkin, L. P. Olekhnovich, Yu. A. Zhdanov, V. V. Kislev, M. A. Voronev, L. E. Nivorozhkin and Z. N. Budarina, Dokl, Akad. Nauk SSSR, 204, 1363 (1972). I. C. Calder, D. W. Cameron, and M. D. Sidell, Chem. Commun., 360 (1971). J. Castells, M. A. Merino, and M. Morino-Manas, Chem. Commun., 3: 709 (1972). 1 R. L. Burwell, Jr., and R. G. Pearson, J. Phys. Chem., 22, 300 (1966). I 4'- ‘WA 1, S ”'TITJ'ITIQHJLEMIMTIIMilli)111111171?!”