A a: a! ‘ 7“ JV; 'Qs‘mwie ...v.,7~,,_.". .L';“. . ‘1: w‘.‘ V- y-uv.‘ an q u . .q,,_4 ”’1' a, .‘.L, ‘. :fx’.l rip}.\’{~v" $225 -.. ‘5 ’ I ‘ - x ‘. ' ' ’ . r 7. ‘ . ' ‘ ,. . ' n.4- w- 4,.» : J: -" ~ " I "aw "- ' - '. ”1; ' 5‘3!» . ., , ., . A I , \ " ‘ , 5-53 :,|J,., ' . .. w“ ., 2" . .- ‘ . glut] -.‘ .n }-' .'v a; A ,. . M v .3 g 2;, {4231‘ A»-o ,- ‘_ . . i: ' 9' J ‘ , r 4: X), , ' " ‘. , . .,.. 1v v a . '. . r .J; , (.1 , , Juan‘s fl . ‘. , fi-‘x’u- ,. . . my > ‘. ,2. I, z .. ,z . ..',w.b:,f_,'_._‘ ..,': 1'“ 1! a ,, n1! m ”I, 1 ’ u r.I. .1 ‘v.'~ Jiynn .u-v v" -'. .. my. , ”nu” .. _V H ...- - -, ..m-4 .n‘ at”. f! ” Md r" .1. u—aur' ” '2 , U...” . "f I:I...— - «c’. w ,. q , . »ro- . »-- ,4, , .pm.” ' ."V ‘ " .nunv'“ , r . I ,4 «a -“.’.."“I_ ‘ l'rr" ,,,..,, _, . v- ,1 a 3 MN,» ,_,,.,.. .4 “a. , .4 ‘!,,~. F‘ H. M. r u-.— ~ I t ,.mm"v'|.4v,,,.,..~ V ’" adv4"’ ' ' _ pm '3 .tm n» .m- . -.- , ,.,» "s I v I. .u w. a~ '1”: I ; ‘ . 4‘ r “ V .mw‘ / a4 n '4‘ l .... . g... ”" r'. .p- ”1» 7" '3 .. , .r J}! . _‘ a.- .. _ . . rr ‘ “4- .0," .r , , 7N"? ..- r: v, ',. nod ' , .u- - .1. "M54,“ .. -. ‘~’ V n ' “WT-3542:. {V’ f. .. “.m- ‘ yp'N-* ~ .a‘ ,. m. . ‘ ~ mun“, ' ‘ 'r‘ n ‘ ‘ -4: «>- . wr'rr‘; ' , .' . “4.. 1.“ .-- .1' ‘;4$hv , _ ‘ . . an“. o‘,‘ . .,, a .. " ’”‘ < - ’ aru’t' W! 51.3” .,.~ , r- ""“t y .3,“ n ,. I a. ’.’I J a» ,.— v a, -r- m a.» .~ “c .w.- - . rX'h—l' . ,... Abl‘qiv v . . , yr-I. 1J4 . . w!- 2...“? - . 9-" ’ u-.. "“" ..m-m .. N , ' q: I": :t,""'..".~1: >1 .. .. J _,‘ ' n- " an ’W‘ gnu: .— ’11:", .. .- ‘ a ”313." 1.3!}1', "!T~'-F'rr~";1' 7J2 - 371.. "V. r. ‘0‘ W - I‘M l r V'VVUWU‘ LEBMRY w Wm weUN.TEW}L{IBFAWE§- MW llllllllLlll‘llll University This is to certify that the thesis entitled "Hydride Transfer Reactions" presented by Michael Tornaritis has been accepted towards fulfillment of the requirements for Ph . D . degree in Chemistry 6 {iv/b Major professor Date July 14. 1989 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution PLACE It RETURN BOX to remove this checkout from your record. TO AVOID FINES mum on or before one duo. DATE DUE DATE DUE DATE DUE JL II P MSU Is An Mmotlvo ActiorVEqml Opportunity Institution HYDRIDE TRANSFER REACTIONS By Michael Tornaritis A A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOROF PHILCBOPHY Department of Chemistry 1989 ABSTRACT HYDRIDE TRANSFER REACTIONS By Michael Tornaritis The purpose of this research was to examine the structure of the transition states of intermolecular hydride transfer reactions. i.e.. whether they are linear or trigonal, by using as a probe hydrogen-deuterium kinetic isotope effects. Three sets of reactions were examined. The first set involved reactions of triarylmethyl cations with triarylmethanes and their deuterated (at central carbon) analogs. These reactions, which are considered to be one-step hydride transfer reactions, gave kH/kD isotope effects in the range 7-9. In view of the fact that the transition states are quite symmetrical with similar force constants t1 and f2, the large kinetic isotope effects exhibited by these reactions support a linear transition state (I). The data were not accurate enough to determine activation energy parameters and assess the extent of any mechanical tunneling. The second set of reactions involved the less sterically hindered xanthene system by reacting substituted xanthenes with the 9-xanthyl cation and xanthene with the diarylmethyl and triarylmethyl cations. The obtained kH/"D kinetic isotope effects, 3.5-7.5, are consistent with transition states (II, III, IV) where the angle between donor-hydrogen-acceptor is 120-150° on the assumption that the force constants on either side of the hydrogen are the same; it is also possible that the transition states are linear with somewhat unequal force constants on either side of the hydrogen 4. Q ‘ Hit»~ II III IV The third set involved the reactions of triphenylsllane with triarylmethyl cations. No assessment of the structure of the transition states (V) of these reactions could be made because the kinetic isotope effects (1.6-2.2), which are in the range assigned by 0me to cases where the angle between donor-hydrogen-acceptor is about 90°, may be due to the expected large differences in the force constants f1 and f2. - ‘ x .+ e I, . aging- ’znwuz‘ \x O £0 L . V Our investigations showed that the only reacting center of the triarylmethyl cations is the central carbon atom. No hydride attack on the aryl rings, followed by hydrogen rearrangement to the central methine carbon, was detected. Similar conclusions were drawn from our studies with the 9-xanthyl, 2,6-dimethoxy-9- xanthyl and dlanisylmethyl cations. The identity reaction between the triphenylmethyl cation and triphenylmethane was found to be appreciably slower than the corresponding reaction of xanthene with the 9-xanthyl cation. TO MY FAMILY To my father I own the life of to my teacher the enlightened life. Alexander the Great You understand who your . true friend is when you are in need and danger. Ancient Greek proverb Worship your Father and Mother Bible ACKNOWLEDGEMENTS I would like to express my deep appreciation to my advisor, Professor G.J. Karabatsos, for his guidance, encouragement and friendship. My thanks are also extended to the Department of Chemistry at Michigan State University including both faculty and staff. Financial support provided by the department is gratefully acknowledged. I will always remember my friends and co- workers including Asaad Salehi, A. Analyti, D. Voyiotzoylou. M. Kondylis and W. Park as well as the the Greek and Cypriot Student Association and the Greek American Community for their friendship. Finally, I dedicate this work to my family for their continuous love and support. TABLE OF CONTENTS LIST OF TABLES CHAPTER I INTRODUCTION INTRODUCTION CHAPTER II TRIARYLMETHYL CATIONS REACTIONS EXPERIMENTAL l. Preparations Triphenylmethane enriched by 50% 0-13 at the methine carbon Trianisylmethane Tri-p-tolylmethane Anisyldiphenylmethyl cation ll. Reactions studied Reaction of Trianisylmethane with Triphenylmethyl Tetrafluoroborate Reaction of Tritolylmethane with Triphenylmethyl Tetrafluoroborate Reaction of Trianisylmethane with Anisyldiphenylmethyl Tetrafluoroborate Reaction of Triphenylmethane with Triphenylmethyl Tetrafluoroborate Spectra RESULTS AND DISCUSSION Calculations of rate constants and kinetic isotope effects (RH/k0) Reaction of Trianisylmethane with Triphenylmethyl Tetrafluoroborate Reaction of Tritolylomethane with Triphenylmethyl Tetrafluoroborate Reaction of Trianisylmethane with Anisyldiphenylmethyl Tetrafluoroborate Tables Investigations of the magnitude of the isotope effect for reaction of Trianisylmethane with Triphenylmethyl Tetrafluoroborate Search for possible hydride attack on carbons other than 0-1 of the cations Reaction of Triphenylmethane with Triphenylmethyl cation Conclusions page .5 .5 .5 .5.5.5.5 .5.5.5 V0) 01 0| 0| 01 43-503” ”Nd —L unl-L 29 29 31 32 33 41 41 42 43 81 St St LA 91 VA 8L it (It L9 L9 79 179 IS 09 09 67 6’7 87 817 1.? 9V 9? 9V 9? VI? uoueo Muiewlfisgueul a- i-eueusMueudul pue eueusuiueudul suogieiedeid ‘l 'IVLNalNlHadXE SNOILSVBH BNV'IIS'IANal-Idlul Al Hahn/HO suogsmouoo UOItOGS Sill! U! SUOIIOBSJ out 10; stoeue edoiosg em to Mewwns uoneo Mutuex-s em to 6-0 ueut Jeuio suoqmo uo arcane epupliu elqgssod to; names eteloqomnuene J. Mqiuex-s un eueuiuexmeinep-e-Axouiewig-9‘z Io uogioeeu eieioqomnueuel Muteumiueudul uwvl eueqwexoremep-a-Iixoqiewig-9‘z io uogioeeu eteiolqoied Muiestguega IIIIM eueutuexmeineg-s to uouoeeu excioqomouenel Muiewlliueudgpuisguv llth OUBIIIUBX ID 0099998 eieioqomnuenel Mmewuiueudul uuM eueutuexmeinea-s to uonoeeu (Ox/Hy) mastic edema oneugx pue stuetsuco etc: to suonelnolea NOISSflOSIO CINV 81.10838 eitoeds eieioqomnuenei Mutuex-s un eueqtuexomnep-e-lixoutewgo-9‘z to uouoeea eieioqomnuenel Mutewlliueudul mm eueutuexmeinep-s-Axouiewig-9‘a to uouoeeu eieiolqmed Iliuiewllisguego un eueutuexmeinea-s to uogioeeu eieioqmonuenel Muteuilliueqdlpllisguv mm eueuiuex Io uogtoeea eteioqomnuenel Iliuieumiueudul un eueutuexmeineo—e to uouoeeu Pelpnis SUOItOBBH 'lI SiBJONOJSd Iliuteunisguego eueqtuexmetnep-e-Axoqtewgp-g‘z uoueo Muiuexlixoutewgp-g‘z uoueo Iliqtuex eueuiuexmetnepgp-s‘e pue eueuwexoxemep—s SUOIIBJBdSJd °| ‘lVLNEIINItISdXB SNOILSVBH SNSHLNVX III UBLdVHO II. Reactions studied Triphenylmethyl cation reaction Anisyldiphenylmethyl cation reaction Trianisylmethyl cation reaction Reaction of Triphenylmethyl Tetrafluoroborate with Silyl Deuterated Triphenylsilane Spectra RESULTS AND DISCUSSION Calculations of rate constants and kinetic isotope effects (RH/RD) Triphenylmethyl cation reaction Anisyldiphenylmethyl cation reaction Trianisylmethyl cation reaction Search for possible hydride attack on carbons other than C- 1 of the cations Summary of the isotope effects for the reactions in this section Conclusions GENERAL CONCLUSIONS APPENDIX REFERENCES vi 78 78 79 79 8O 81 90 93 143 Table 1. Table 2. Table 3. Table 4. Table 5. Table 6. Table 7. Table 8. CHAPTER III Table 1. Table 2. Table 3. Table 4. Table 5. Table 6. Table 7. UST OF TABLES CHAPTER II TRIARYLMETHYL CATION REACTIONS Rate constants and isotope effects of the reactions of trianisylmethane with triphenylmethyl . tetrafluoroborate Rate constants (kH) from aliquote solutions taken from the different runs of the reactions of trianisylmethanes with triiphenylmethyl tetraflurorborate and stabilized at 26°C. Rate constants (I‘D) from aliquote solutions taken from the different runs of the reaction of trianisylmethane-I-d with triphenylmethyl tetrafluoroborate and stabilized at 26°C. Isotope effects for the reaction of trianisylmethane with triphenylmethyl tetrafluoroborate. Rate constants and isotope effects of the reaction of tritolylmethane with triphenylmethyl tetrafluoroborate. Isotope effects for the reaction of tritolylmethane with triphenylmethyl tetraflurorborate. Rate constants and isotope effects of the reaction of trianisylmethane with the anisydiphenylmethyl tetrafluoroborate. Isotope effects for the reaction of trianisylmethane with anisyldiphenylmethyl tetrafluoroborate. XANTHENE REACTIONS The dependence of mm on the ratio of the concentrations of Meuteroxanthene and triphenylmethyl tetrafluoroborate at different temperatures. Integration values of the xanthyl cation protons in the reaction of 9-deuteroxanthene with dianisylmethyl perchlorate. Integration values of the xanthyl cation proton in the reaction of 9-deuteroxanthene with triphenylmethyl tetrafluoroborate (Reaction 10, Table 1). Integration data of the xanthyl cation protons (solution of 9-xanthenyl in deuteroacetonitrile). Isotope effects for the reactions of 2,6-dimethoxy-9- deuteroxanthene with triphenylmethyl cation. Integration of the 2,6»dimethoxyxanthyl tetrafluoroborate in CDacN solution. Isotope effects for the reaction of 2,6-dimethoxy-9- deuteroxanthene with 9-xanthyl cation. vii 33 34 35 36 37 38 39 40 65 68 68 69 7O 71 72 CHAPTER I INTRODIDTION Reactions involving hydride transfers are quite common. and include not only those of academic interest but also several carried out on an industrial scale 9.9., the catalytic cracking of hydrocarbons. as well as some biochemical reactions such as the reversible oxidation of alcohols to aldehydes and ketones by the use of alcohol dehydrogenases and NAD+(P)-NAD(P)H coenzymes. The purpose of the research described in this thesis was to probe the structure of the transition states that involve hydride transfer, i.e.. where they are linear (I) or trigonal ll. [R----I-I----R']+ . + t I II Before focusing on hydride transfers. however. a brief review of what is known about hydrogen transfers In general (proton, hydrogen atom, hydride) will be given. Before 1960 the hydrogen atom or ion which was transferred from a donor to an acceptor was treated as a tree species. For simplicity, this species was disassociated from the donor and associated with the acceptor In a two-center type reaction. In 1960 Westheimer1 and Melander2 separately introduced the three-center type transition state in a linear conformation for intermolecular transfers of a hydrogen atom or an ion. Westheimer's suggestion for this transition state resulted from the theoretical treatment of the reaction: AH + B 2 A----H---B ——-- A + H8 For the linear transition state A----I+---B of this reaction he took into account only the symmetric and antisymmetric stretching vibrations in his calculations of deuterium kinetic isotope effect. 3 t—A"++"‘B—. “k--H ##B symmetric antisymmetric and ignored the bending vibrations and the structures of the A and B groups. Westheimer concluded that this linear transition state was the one which imposed the lowest energy and the minimum of steric Interactions. Melander's suggestion of this same transition state resulted from the theoretical treatment of the shape of the landscape as a whole for the above reaction and is equivalent to that of Westheimer's. The first suggestion of a nonlinear transition state for hydride transfer processes was made by Hawthorne and Lewis3. They suggested that “the only reasonable point of attack of an electrophile on a bond to hydrogen must be the center of high electron density. that is, the bond. and a linear configuratlon A----H---B is not to be expected.“ They based their suggestion on kinetic studies of the hydrolysis of pyridine diphenylborane C536\O/H .. l3 Cm: \&’\\ U+ n.0—-n.+ (3")st +( C \x- . Hydrolysis of pyridine diphenylborane Hawthorne and Lewis proposed the following transition state for the BH bond hydrolysis aid A.“ suggested an electrophilic attack on the high electron density 8H3 bond from the electron poor side of the Ha, which is the opposite side of the oxygen. The possibility of all three atoms using the pair of electrons is shown by the dotted lines. Also, they suggested that the trigonal transition state may be general for hydrogen transfers. 4 The non-linear transition state in Intermolecular hydride transfer reactions was favored also by Olah, who studied“ the hydride abstraction from alkanes by using alkylfluoroantimonates. Working with superacid solvent systems, he was able to study stable carbocations and their reactions. He suggested for these reactions the three- center bonded pentacoordinated carbonium ion transition state, f! R,C-i{ + up -- [mcv'ihcm] —- rec-ca. + a‘ especially in tetriary-tetriary systems where the highly unsymmetric bridged carbonium ion minimizes the steric interactions. Olah based his conclusions on the relative composition of the product mixtures which were determined by gas chromatography, mass spectrometry and NMR spectroscopy. He emphasized, however. that due to the complexity of the systems, small changes in conditions may differentiate the composition of the product mixtures. A non-linear structure for C--H--C was proposed5 by Raghavachari, Whiteside, Pople and Schleyer for protonated ethane, C2H7+. According to their theoretical calculations, which were carried out an optimized HF/6-31 G' level, the lowest energy structure, CI. which is shown below, H" H \“H H ""'C/.\ V” H\ “H ””6 c \\“ H\\\ \H *c “c \\ H/ x». H \ X c, ‘H..II c. "is strongly distorted from a Dad structure with an angle CH1C of 121.7°. It is also twisted by about 30° from a staggered Cs configuration.” The transition states of intramolecular 1,2-hydride rearrangments are by necessity non-linear. lntrarnolecular-hydride rearrangements may occur from one 5 carbocatlon to a slightly less stable carbocation. whenever there Is no other rapid reaction for the former catlon to undergo. and the energy barrier to rearrangement is relatively low. Examples for these rearrangements are those occurring in the propyl cation. The low temperature NMR spectra of the propyl CaHT". In superacid medium show hydrogen and carbon scrambling.6 The superacid medium eliminates other reactions that would compete with the hydride rearrangements. The transition states for these rearrangements have been calculated from detailed molecular orbital calculations for the CaH7+ by Pople. Schleyer and co-workers.7 using STO-3G.4-31G and 6-316 sets. The structures examined were I-Vlll. , . a / . . _c*\° _g" ? my ,c—cue \F _-é< ,c-éwe (SJ- ’4 d .21 a Cs,- 1 II III N \cf g g I C’ii‘C C/ “5C?! C/';l\ Sec/(3%.? 8’ s; I? ' ‘s- a? “—Cx. I I V VI VII VIII The calculations predict VI to be a transition state, whereas VII. which lies at an energy ' minimum, is an Intermediate. Because of the special contraints imposed by the system the transition state VI is by necessity nonlinear. More examples of intramolecular hydride transformations are the hydride shifts which are facile In C3-Cn sized rings in superacid medium. The geometry of the ring forces the hydride nearby the empty p orbital to bridge with it. This is exemplified in the 1H and 130 NMR studies by Sorenson and Kirchens for the 1,6-dlmethyl-cyclodecyl cation which proved to exist in a symmetrical u-hydrido 1,6-hydrogen-bridged SITUCIUI‘B: [1 ‘CXIS CH CHILI T I H The strainless cis-decalin geometry is depicted, which shows no deviation from linearity of the C-H-C angle. Some bending. however. which would improve the cis-decalin type confonnatlon, would be expected. The deuterium kinetic isotope effect has been used extensively for the determination of the mechanisms of chemical reactions. The theory of the magnitude of this effect is very well established, For the reaction mentioned previously AH(D) + 8—,: A----H(D)-B ——> A + (D)HB Westheimer1 states that I'the major portion of the kinetic Isotope effect arises from contribution to the activation energy from changes In zero-point energy which occur when the reactants are converted to an activated complex.” Using the stretching vibration frequency for hydrogen and deuterium he calculated the activation energies for the above reaction and thus the difference in the zero-point energies of the hydrogen and deuterium species to be 1.15 kcal/mol at 300°K. This corresponds to a factor of 7 for the isotope effect. He assigned this magnitude of the Isotope effect for a linear transition state. For many reactions. especially those involving intramolecular hydrogen transfer. the transition state Is not linear. O'Ferral9 calculated the effect of such a geometric 7. change upon the magnitude of kHII‘D at 25°C. He considered for his studies a model which Involved a minimum of five atoms. This model ls shown below along with the force constants that have been used in the calculations: The maximum isotope effect is expected in a linear transition state (l.ll) when the force ‘- (I) x...H...?“: (II) {mufuir' n; constants on either side of the hydrogen are the same andthus there Is a balance of force on it (no motion of H in the symmetric stretch). In a‘nonlinear transition state (mm the balance does not exist and a smaller Isotope effect is expected. '(I) By assuming a reasonable I H dependence on transition state structure, as expressed by the order, x. of the bond to hydrogen being broken. and (ii) by supposing that in the limit X—) 0 or 1. force constants in the transition state reduce to their corresponding values In reactants and products,” O'Ferrall concludes: “For the linear transition state (IS-180°). the isotope effect showed the normal dependence on x, with kHII‘D approaching unit as x——-) 0 or 1 and at a maximum for x - 0.5. For the most extreme non-linear cases. with 6-60° or 90°, the dependence of kH/"D upon x was much less marked and for no value of x did the Isotope effect depart substantially from unity. For larger angles, B-IZO' or 150°. behavior intermediate between that for the linear and extreme non-linear transition states was found." These results can be represented by Figure 1. I1>I2 I1II2 f1 .7, X \ a - X I ,\x \' x - The second set involves the less sterically hindered xanthene system reacting with substituted xanthenes, and with dianisyl. and triaryl compounds. These systems would lie at the top (xanthene-xanthene) CH3. .60 4M:_> 6 Q CH30 .6 “ (DI xanthene-triaryl) :43?" ope 10 - a . ‘ —. END“ “”64? -°@i"‘. The kinetic isotope effect, - - except for the xanthene reaction - - would thus depend not only on 9. but also on the relative magnitudes of f1 and f2. The third set involves the reactions of triphenylsilane with the triarylmethyl cations. .~/" I © ©SI-HD)+ +0—u- o ‘ an I- "I (Drug... + Q. Iii/x i- _. Si + (D)H—QQ oip‘fios p This system most likely would lie at the left low side of the curve of the Figure 1. CHAPTER II TRIARYLMETHYL CATIONS REACTIO‘IS 11 EXPERIMENTAL I . Reparations Triphenylmethane enriched by 50% 0-13 at the methine carbon: The title compound was prepared according to the following procedure. Barium carbonate (3.2 gr), 50% enriched in C-13. was placed in a 50ml three-necked round- bottomed flask of the apparatu d ' t d In Figur (1). Figure 1. Apparatus used for the preparation of 13002 After the air of the system was evacuated, (concentrated) sulfuric acid was added dropwise to the barium carbonate. Ninety percent of the three-necked round-bottomed flask was filled with sulfuric acid, its contents were stirred, and finally the solution was warmed at 45°C for about 10 minutes. The stopcock of the 500 ml separator tunnel was closed and the produced carbon dioxide was trapped in the funnel. The separator funnel then was connected with a 500 ml round-bottomed flask containing phenylmagnesium bromide (2.94 g) in ether (60 ml). The carbon dioxide slowly and cautiously, because of the closed system and the excthermicity oi the reaction, was allowed to react with the Grignard reagent. The solution was shaken for 10 more minutes and then was hydrolyzed with saturated ammonium chloride. The ether layer was separated and dried over sodium sulfate and after distillation of the ether, benzoic acid was obtained. The overall yield of the benzoic acid formed was 76%. Following basically the procedure of Ray, Kurland and 13 Colter11 for the preparation of carbinols, the benzoic acid was refluxed overnight with ethanol in the presence of sulfuric acid to form the ethyl ester (95% yield). The ethyl benzoate (1.46 g) was reacted with phenylmagnesium bromide (4.4 g) to produce triphenylmethanol. Treatment of the carbinol with acetyl chloride according to the procedure of Bachmann12 gave triphenylmethylchloride, which upon reduction with excess lithium aluminum hydride led to triphenylmethane enriched by 50% C-13 at position 1. For the reduction reaction, the chloride (1 g) was dissolved In benzene or other (30 ml) in a 100 ml round-bottomed flask and the lithium aluminum hydride (about 0.1 g) was added slowly . The reaction mixture was stirred, refluxed overnight, and the excess lithium aluminum hydride was destroyed by addition of water. The organic layer was separated, dried over sodium sulfate, and after distillatlon of the solvent gave triphenylmethane. The NMR spectrum of the product in acetonitrile Is shown in spectrum 21. . Trianisylmethane: The trianisylmethanol for the preparation of the title compound was prepared following the procedure of Ray, Kurland and Colter.11 For this, anisyl acid was treated with methanol in the presence of sulfuric acid to form methylanisate. The ester was ' reacted with 4—methoxyphenylmagnesium bromide to produce trianisylmethanol.”-13 This carbinol also was produced by the reaction of excess 4-methoxyphenylmagnesium bromide (3.5 times) with diethylcarbonate. The crude product was purified by formation of the picrate. For this to the hot solution of trianisylmethanol (7 g) in ethanol (50 ml) was added with stirring plcric acid (7 g). The solution was cooled in an Ice bath and the red complex which precipitated was filtered and washed with methanol. The complex was dissolved in benzene and was extracted with 5% sodium hydroxide solution until the yellow petals which were formed with the addition of the sodium hydroxide were dissolved. Then the organic layer was separated and was dried over 14 sodium sulfate. After evaporation of the solvent trianisylmethanol was obtained as a red oil. This carbinol was treated with acetyl chloride to prepare of trianisylmethyl chloride, which was then reduced by lithium aluminum hydride or lithium aluminum deuteride to give trianisylmethane, or deuterated trianisylmethane. The reduction procedure was the same as that used to prepare triphenylmethane. The trianisylmethanes were recrystallized from ethanol. The acetonitrile solutions of the trianisylmethane gave spectrum 2. TrI-p-tolylmethane: The trI-p-tolylmethane was prepared according to the procedure described by Newman and Deno.14 Diethylcarbonate was reacted with p-tolylmagnesium bromide to give tri-p-tolylcarbinol, which when treated with acetyl chloride, gave tri-p- tolylmethyl chloride. Reduction of the chloride with lithium aluminum hydride or lithium aluminum deuteride gave tri-p-tolylmethane or deuterated triptolylmethane. Anisyldiphenylmethyl cation“: The carbinol needed for the preparation of the title cation was synthesized according to the general procedure of Ray, Kurland, and Colter11 for the preparation of carbinols from benzophenone and Grignard reagents. To a solution of benzophenone in ether, was added excess 4-methoxyphenylmagnesium to produce anisyldiphenylmethanol.13 This carbinol (4 g) was dissolved in propionic anhydride (37 ml) and the solution was cooled in Ice. When tetrafluoroboric acid (3.65 ml, 48% aqueous solution) was addded dropwlse, solid anisyldiphenylmethyl tetrafluoroborate was precipitated. The cation was filtered under argon and was washed with cooled petroleum ether. The acetonitrile solution of this cation gave spectrum 15. 15 ll. Baacticnsstudiad Reaction of Trianisylmethane with Triphenylmethyl Tetrafluoroborate (spectra 1 -1 2): (1) P-(CH3006H4ieCHlD) + (CeHs)SCBF4——) P-(CH3006H4)3CBF4 + (CsHsiacHlD) m For the title reaction the amount of reactants used for each run ranged between 0.267 x 10-3 and 0.3ssx10-3 moles (0.0894 and 0.1229 g). In each case, the reactants were dissolved in 4 ml 0030N. In some experiments 2 ml CDacN and two times the above amount of reactants were used. The reactions were performed in the dry box. Reaction of Tritolylmethane with Triphenylmethyl Tetrafluoroborate (spectra 1 3-1 4): (2) p-(CH306H4I3CHID) + (CSHSISCBF4 —-+ P-(CH306H4I303F4 + (CeHsiscHID) mm For the reaction of tritolylmethane with the triphenylmethyl tetrafluoroborate maximum concentrations of reactants were used amounting to about 0.6x10'3 moles (0.18 g) in 1 ml CDacN. The reactions were performed in the dry box. Reaction of Trianisylmethane with Anisyldiphenylmethyl Tetrafluoroborate (spectra '15-20): (3) p-(CH30C6H4)3CH(D) + P-(CH3006H4)(06H5)2CBF4——> p-(CH3006H4I308F4 + p-(CH30CsH4IICeHSI2CH(D) mm The reaction of trianisylmethane with anisyldiphenylmethyl tetrafluoroborate was canted out in the same way as the above reaction in terms of quantities and conditions. 16 Reaction of Triphenylmethane with Triphenylmethyl Tetrafluoroborate (spectra 21-24): (4) (CsHsla‘acH + (CeHsiecBF4—-> (06H5)3‘303F4 + (CsHsiacH am For the title reaction 0.210 g of triphenylmethyl tetrafluoroborate and 0.164 g of triphenylmethane having the central carbon labeled at the methine carbon with 50% C-13 were reacted In 2 ml CDacN. A Bruker 250 MHz NMR instrument was used to study all these reactions. In a typical reaction, the reactants, in 1 : 1 ratio were weighed In two separate vials. Acetonitrile-da, 1 or 2 ml, was added to the first vial containing the carbocation and the solution was transferred to the second vial which contained the hydrocarbon. Zero time was assigned at the onset of the mixing. After the solution was transferred into an NMR tube, the sample was placed In the probe that was kept at constant temperature and spectral measurements were taken. The Relaxation Delay value for the 1H NMR program which was used in this study was optimized to twenty for the reaction of trianisylmethane and triphenylmethyl tetrafluoroborate, at a time when the rate of product formation was practically zero. Each experiment was performed twice with freshly prepared starting materials. In the second run the acetonitrile was added first to the hydrocarbon and then transferred to the vial containing the carbocation. The following spectra, 1-24, are representative of the reactions studied: '1-7 a e U} ,‘D U! \. «ma... amen. Er:— III- P i 3.3.90... «80.33 . .o :5. e. 5e 5.333.333. ooze: 303320.»: .... 833.35 tan: in. =6 3.3:. cued .: 53 2:3. 2.. Icon}. ..n. I 9803:: m .e 52 o. Spaniaeaoae 83031.3: .. 3.2. II _ as u. i ‘L «mun... «who. ... L... avenge. a .a 5m. 0. e. 3.52:3 o. r 333133.53 2.... 31531305530. kflsqr e can. +8§9§5ta§o§fie no can : + can a ingenxfi + mucos- a. u. 2:. a £03 8.6: 3 303.836 So .3382. 0. 33:0: . voice: 3.5.30.2 530.58 0. 3.5.3.3052... end .3 3529.2. 8.5369: (.5 .fig 23 3:78.38 o. :6 3233.30.31 8:03. W. knew .5 h ... q 0 an: I D . 9.0 u e Po 0.. 18 30°C + \N ' no avatar) ‘L—vcm 095-» ' a C , i [a 1. Gr '1. 1,}... . 4 (how. + decal/J / ... . /2 5° ' ____ _________ 0.600 /s303 (58514} .‘L #1 ‘ l 0.664 I _ L. L..---;t_fi'._ ___-______'___ _ _ _ _ _ _'_‘l_____"_\____._ 4.047 //.428 5 hours 0326 0325 m L L 15974 Spectre BA-BC also represent reaction 1, which was repeated with an equimolar amount of trianisylmethane and its deuterated counterpart and which were reacted with excess of triphenylmethyl cation (4 times mmols) in CDaCN. The progress of the reaction was followed by 1H NMR at 30°C. These I8 hours spectra also show the integrals of the peaks due to the methoxy and methine hydrogene of the reectente end the products. at time lntervele of 1/2. 5 end 18 houre after the mixing of the reactants. (spec; I IIJ 1%” ‘1 L358 I I 3.95 .3 41°C ' + ICHpCHI'JCi-r' +(Cf-i0Q-3CDJ- (QCf-flCf-QQ'IAC +- (QC/140) . 31/9 / .' / . .’ 4 SpectreBDendeEarethoeeofenaliquotofthe above mixture Mitch was placed at 41°C. The integrals of the peaks due to the methoxy and the methine hydrogene are also shown. 29.32 r52 60) L. _JL_JL__ 6 how 7.093 8826 . r 32 s E) MU I I Milli L ... ._/‘J1KJL._JKj\-IU .JLU‘J-‘LU. IL. .. _:::_,_ _ _ __ ,l J... PPM Mk V. $000.33 4 .0 32 2 30 3033.323. 320: + 81035.5 III, E |> I git: «mum. «mm m. E ... Mhso. ' 30230030203033.8230 30331303030 03. 30 3033.32.33? 3303132330 3.350 402002.36. «Ozuchaostun I b I ...8: 02.3: are: .— 2 i -. 8109510 + as 23.? «00233 3. .0. 3.0 000233 2 30 3033.323. 02.0: (.3 3.3313033? Ms 30:3 . gr 2.2 3.x.I3 2 30 23 3 80833.0. 21 Spectra 11 and 12 represent the reaction of deuterated trianisylmethane with twice as many moles of the triphenylmethyl cation. after 35.4% and 95.6% of the hydrocarbon was reacted. TheeespectraehownopeakaintheareabetweenSppmandBppm. (CH 0 4- - 3 CIHaACD‘t' QC '—-O C H30GH4JJC+03CC v I: 2 ; ' 35.4 o/b I 1 I 1'1 " flit E 1 3 1 .v Ji__." I; 3-11:: ’JV \ k (SP. [I . a 95.5 o/o ‘ JL L . L L I 3:. 2.. 'J 0.. 0.! 5.0 4.. PPM 22 «000.23 6 3030030 30 300:0: .3 00.100: 00503.00 38.330330 0:0 30 3033.333. 83... M00233 .0 3030030 30 303.0: 0. 38.530303 (.3 30 .3033: 00:0? 31.03.00 D + 9. 0+ II. + R.1092. .00 + 9000 «Mn 5. 0 0 . 0 + A006 I. . R. I 0 I 00+. ... _ _r 0 0 . 0 Ann: Gm I: 0.0 ... 0.. 0.. ... 0.. 0.. ... 0.. 3a: .23. i A A r I + (CH3; Cal-.1”... .‘rsrzC Spectrum 15 is that of the anisyldiphenylmethyl cation. I ! t ' l . I (S 21,5 " (crepcmcritori-(CHQCfiJIQngC -' ; i. 9 4 a... tango cmc‘wc HFQH‘NQHSIIC Hi0) ' : 3 a : s v; "”35,“ I - U. ‘._/’L - (S R I 5} Spectrum 16 represents the reaction of trianisylmethane with the anisyldiphenylmethyl cation after 94% of reaction. " ....iMi L (SE! 7 I u 1. o 0.0 5.0 4.0 3.0 2 .0 PPM' Spectrum 17 is a representative spectrum for reaction oi the anisyldiphenylmethyi cation with deuterated trianisylmethane after 51% reaction completion. The peaks shown in spectrum 16 also appear in this spectrum, except for the peaks due to the methine hydrogens; no peaks are detected at 5.49 ppm or 5.36 ppm. 24' .UUJJ ,1 '__. i . -l J 35.9. 18 .' (c 9%." a + (51C9‘1""€"‘.’§ ‘-' ‘C”0€“,C*cuc~”.“z'r~’;“'co 3' “ °/° 'I ll '--- . j 1 Lol" 1 1' Lg}; ~ {SR/.9) Spectra 18-20 also represent the above reaction at the following reaction times: at mixing, at 3.4% reaction completion and 48.7% reaction completion. Again. no signals appear between 5 ppm and 8 ppm. 4 8. 7 0F PPM 25 Spectrum 21 is that of triphenylmethane enriched with 50% 13c at the methine carbon. In addition to the methine singlet at 5.58 ppm. the spectmm shows the methine doublet (J13c-H - 127.5 Hz) or the 13C~labeled compound. The mulltplet in the area7.0-7.4 ppmisduetothehydrogensofthephenyl rings ol the hydrocarbon. 0:21-14 1U IL Spectrum 22 represents the reaction oi high concentrations of the above labeled triphenylmethane with triphenylmethyl catlon. run for five months at 40°C. it should be noticed that the ratio or the signals due to the methine proton of the unlabeled over the labeled compound has increased substantially at the end of the reaction. + + ‘ cafe H + 03c —-. of? + 0??! H . l . L . . ' ‘ ‘ I I 1:. 3.. 1.. Ce. 5.. PF“ 0.. 3.. 2.. I3 . C 4. (SE 23) o, c ’ g LL Spectrum 23 Is the 13C spectrum taken from a solution oi triphenylmethyl cation. The peak at 212 [3 ppm represents the methyl carbon of the cation. C (I!) 1: 4. 4- (1 ($224) $3011 + 03c——p 2,04- 035+! I 1L._L"_L L l l l‘ . ‘ ' I L 2.... 1,... ”.‘.PPM ”it... no.0 5 0.. 2 0.. Spectrum 24 which is a ‘30 spectrum. was taken from the same sample the same day the spectrum 22 was taken. it should be noticed that the ratio of the signal due to the methyl carbon over the other signals in spectrum 23 is larger compared to that of spectrum 22. 26 Spectrum 1 is that of the triphenylmethyl cation (commercial) in acetonitrile. The small pentet peak at 1.93 ppm belongs to the acetonitrile (D. 99.6%) which was the solvent used for this study. The multiple peaks shown downfield between, 7.7-6.3 ppm, belong to the hydrogens of the phenyl rings of the cation. Spectrum 2 is that of trianisylmethane and shows singlet peaks at 3.66 ppm and 5.37 ppm representing the methoxy groups and the methine hydrogen respectively as well as two doublets at 7.6 ppm and 8.1 ppm representing the hydrogens of the phenyl rings. Spectrum 3 is that of a mixture of trianisylmethane and triphenylmethane. In this spectrum. in addition to the peaks shown in spectrum 2. are shown a singlet peak due to the methine hydrogen of the triphenylmethane (5.55 ppm) and multiple peaks due to the phenyl hydrogens of the triphenylmethane (6.1-6.4 ppm). Spectra 4, 5. and 6 were taken by monitoring the process of reaction 1 between equimolar amounts of trianisylmethane and its deuterated counterpart with two equivalents of the triphenylmethyl cation. in spectra 4 and 5 in addition to the peaks of the triphenylmethyl cation. trianisylmethane and triphenylmethane. appear the peaks due to the trianisylmethyl cation. These are a singlet at 4.05 ppm from the methoxy groups and two doublets at 6.3 ppm and 6.6 ppm representing the hydrogens of the phenyl rings. In spectrum 6 the peak at 5.37 ppm. which was due to the trianisylmethane methine hydrogen, has disappeared. Spectra 6A-60 also represent reaction 1, which was repeated with an equimolar amount of trianisylmethane and its deuterated counterpart and which were reacted with excess of triphenylmethyl cation (4 times mmols) in CDacN. The progress of the reaction was followed by 1H NMR at 30°C. These spectra also show the integrals of the peaks due to the methoxy and methine hydrogens of the reactants and the products, at time intervals of 1/2, 5 and 16 hours after the mixing of the reactants. 27 Spectra 60 and 6E are those of an aliquot of the above mixture which was placed at 41°C. The integrals of the peaks due to the methoxy and the methine hydrogens are also shown. Spectrum 7 is that of the trianisylmethyl cation and shows a singlet due to the methoxy group at 4.05 ppm and two doublets due to the hydrogens of the phenyl rings at 7.3 ppm and 7.5 ppm. Spectra 6 and 9 are the spectra of the trianisylmethane and the trianisylmethane-triphenylmethane mixture respectively. Spectrum 10 is the spectrum of the trianisylmethyl cation with triphenylmethane, 24 hours after mixing of the two in acetonitrile. Spectra 11 and 12 represent the reaction of deuterated trianisylmethane with twice as many moles of the triphenylmethyl cation, after 35.4% and 95.6% of the hydrocarbon was reacted. These spectra show no peaks in the area between 5 ppm and 6 PPM- Spectrum 13 represents the reaction (2) between deuterated tritolylmethane and the triphenylmethyl cation. In .this spectrum. the signal due to the methyl groups of the reactant is a singlet at 2.22 ppm and that due to the methyls of the product is a singlet at 2.61 ppm. The AA'BB' signal of the hydrogens of the phenyl rings of the tritolylmethane appears in the area 6.9-7.1 ppm. Spectrum 14 represents the reaction of tritolylmethane with the triphenylmethyl cation. The methine hydrogen of the reactant appears as a singlet at 5.40 ppm and that of the product at 5.55 ppm. Spectrum 15 is that of the anisyldiphenylmethyl cation. The methoxy group of the cation is a singlet at 4.19 ppm and the hydrogens of the phenyl rings appear as a multiplet in the area between 7.1 ppm and 7.9 ppm. Spectrum 16 represents the reaction of trianisylmethane with the anisyldiphenylmethyl cation after 94% of reaction. The methine hydrogen of the product 28 hydrocarbon appears as a singlet at 5.49 ppm and that of the reactant hydrocarbon at 5.36 ppm. Four singlets representing the methoxy groups of the reactants and the products appear in the area between 3.5 ppm and 4.5 ppm. The signal at 4.19 ppm is that of the methoxy group oi the anisyldiphenylmethyl catlon; that at 4.05 ppm is the signal of the methoxy groups of the trianisylmethyl cation; and those at 3.69 ppm (3.692 and 3.695) are the signals due to the methoxy groups of the two hydrocarbons. Spectrum 17 is a representative spectrum for reaction of the anisyldiphenylmethyl cation with deuterated trianisylmethane after 51% reaction completion. The peaks shown in spectrum 16 also appear in this spectmm, except for the peaks due to the methine hydrogens; no peaks are detected at 5.49 ppm or 5.36 ppm. Spectra 16-20 also represent the above reaction at the following reaction times: at mixing, at 3.4% reaction completion and 46.7% reaction completion. Again. no signals appear between 5 ppm and 6 ppm. Spectrum 21 is that at triphenylmethane enriched with 50% 13c at the methine carbon. In addition to the methine singlet at 5.56 ppm. the spectrum shows the methine doublet (Jtac-H - 127.5 Hz) oi the 13C-labeled compound. The mulitplet in the area 7.0-7.4 ppm is due to the hydrogens oi the phenyl rings of the hydrocarbon. Spectrum 22 represents the reaction of high concentrations of the above labeled triphenylmethane with triphenylmethyl cation, run for five months at 40°C. It should be noticed that the ratio of the signals due to the methine proton of the unlabeled over the labeled compound has increased substantially at the end of the reaction. Spectrum 23 is the 13C spectrum taken from a solution of triphenylmethyl cation. The peak at 212 ppm represents the methyl carbon of the cation. Spectrum 24 which is a 130 spectrum. was taken from the same sample the same day the spectrum 22 was taken. It should be noticed that the ratio of the signal due to the methyl carbon over the other signals in spectrum 23 is larger compared to that of spectrum 22. RESULTS AND DISCUSSICN WWW): The rate constants and kinetic isotope effects of reactions (1), (2), and (3) were determined at several different temperatures. All the reactions were second order. 1:1, and were followed by NMR. For the second order reaction A + B ———) C + B the rate constants were calculated from the equation: * .1_ a(a-x) t k- a(a-x) / SlOpe :- k / where x is the product‘s concentration at time t and a is the initial concentration of the t-—-—-)S€C starting materials. Reaction of Trianisylmethane with Triphenylmethyl Tetrafluoroborate ( 1 ) : . Reaction (1) was carried out at three different temperatures; namely 22°. 30°, 40°C. From each sample run at a given temperature (22°, 30°, 40°C), an aliquot was taken and run at 26°C, to show the consistency of the measurements. The amount of reactants used for each run ranged between 0.000267 and 0.000366 moles (0.0894 29 3 O ' and 0.1229 grams). in each case the reactants were dissolved in 1 ml acetonitrile-d3. In some experiments 2 ml solvent and two times the above amounts of reactants were used. Each NMR tube containing the sample was placed in the NMR probe at constant temperature and measurements were taken up until 15-20% reaction completion. which usually took 34 hours. for the hydrogen compounds. The reaction of the deuterated trianisylmethane proceeded much more slowly. Consequently, the concentration of the deuterated reactants was approximately 2-3 times larger than that of the undeuterated ones. This higher concentration required 6-9 hours for 10% reaction completion. Rate constants were calculated from the integrated areas of the methoxy groups of the products and the reactants. The sum of the area under the peak of the methoxy group at 3.69 ppm (reactant) plus the area under the methoxy peak at 4.05 ppm (product). denoted as A4 and A3 respectively, equal the initial number of moles of trianisylmethane a So A3W+A4W-a a “I. A3 -r A4 X-AaW For solution volume' V: 1000X x- —v 1000a a . —v and the rate constant in time t can be written as k V 1— " 1000aA4 t ' The volume of the solutions was measured as follows: The same amount of reagents for each individual run was placed in a 1 ml or 2 ml volumetric flask. Either 1 ml or 2 ml of solvent was added to the flask. The volume of the solution above the line which indicates the 1 ml or 2 ml was measured with a 0.5 ml syringe. 31 A The k is the slope of 17%;? versus t. Using the least squares method the slope, the intercept, the slope deviation (standard) and the 95 percentage confidence slope interval, for each run in each temperature were calculated. The computerized data and results appear in the appendix (page 96-107, 136) as well as the computer program (pages 94-95) which was used for the calculations. Also in the appendix appear the computerized kHI"D (page 136). The 1: deviation of this interval was calculated according to the equation: :1:(Ab)2 - (£5)2(Aki-i)2 + [(EI‘D—ZWHFAKDZ k (Ab - deviation of the k—E, Akl-l - slope deviation of the kH; AkD - slope deviation of the kD). The results from these calculations are summarized in Table 1 and Table 4. Tables 2 and 3 summarize the rate constants for the reactions carried out at 26°C. Reaction of Tritolylmethane ‘ with Triphenylmethyl Tetrafluoroborate (2): The reaction of tritolylmethane with triphenylmethyl tetrafluoroborate is extremely slow. Consequently, maximum concentrations of reactants were used amounting to about 0.0006 moles (0.16 grams) in 1 ml CD30N. The protiated compounds were measured up to 10-15% reaction completion, which required 4-10 days, and the deuterated counterparts until 3-6°/e reaction completion, which required 10-15 days. An even was used for the 30°C measurements and a sund bath for 255°C and 395°C. The reactions with hydrogen and deuterium reactants were done simultaneously and were at the same temperature in close proximity. For reaction (2), the areas of the methyl group peaks of reactant and product, A4 and A3 respectively, were used. Calculations of the slope (rate constant), the intercept, the slope deviation (standard), the 95 percentage confidence slope interval and the kHI"D, for each run at each temperature were carried out as above (appendix pages 106-119, 139). Tables 5 and 6 summarize the results from these calculations. 32 Reaction of Trianisylmethane with Anisyldiphenylmethyl Tetrafluoroborate (3): Reaction (3) was carried out in the same way as reaction (2) in terms of quantities of compounds and temperature used. The two areas under the methoxy peaks. the one in the reactant carbocation at 4.19 ppm, which represents one methoxy group, and the other in the product carbocation at 4.05 ppm, which represents three methoxy groups, were donated as A3 and A4 respectively. So in this case A A3W+3—4W-a A anw and the rate constant for volume v at time t can be written as k . v-Afi . _1_ 3000-a-A3 t The computerized data and results for each run in each temperature appear in the appendix (pages 126-137, 140) as well as the computer program which was used for this reaction (pages 124-125). These results are summarized in Tables 7 and 6. 0:. 000000 a . 0:. .0... u Twp: .0 2...: .... n... ...... .4 .. D... to“: m.0.x.~0.0«~..~. 4.02.0933... 0.03.0 0.313.330. #3100030... 0.03.9. 0.03.0 0.05.00.32... 0.312.330. we“: 0.03.00.30.00. 0.02.00.30.00. 0.03.8 0.30.0 0-3120300. 0-318.039... 0.0.3.0 0.02.00.03E0. 0-02.00.02.06. 0.03.00 :03... :9. ...... .93.... .9. .3. 0.. .0.0.000.0:_.0..0. .E.0E...:0..0... 5.3 0000.05.00.00... .0 00.800. 0... .0 0.00.0 000.00. 000 0.00.0000 0.0... .F 0.00... 34 0-0.3.0 000... .00.. 0000. 0-0.30.0 000.00 00.0 00.0. 0-0.x0..0 000.0. 0.0.. 0.0.. 000.. 05000.0 ..00000 0.0.3.0.. 0-0.30.0 000.0 000.. 00.0. 0-0.50.0 00.0. {0.0 00v... 0.... 0000000 ..00000 0.03.00 0.02.3.0 000... 000.. 000... 0.0200... 000... 000.. 0000. 00... 0000000 0.00000 0.00000 0. o. .0 00. 0... .0. 7000.3. :0 .4 0< 000. 00.0.. 00.0.00 0.0.0.0080... 00.5.00 0.0.... 6.00 .0 00030.0 000 20.000.02.000. 30.00.30.000... ....3 0000.00.30.00... .0 000000. 00. .0 000. .0200... 00. 0.0.. 000.0. 00000.00 .00....0 0.0.. ...... 0.00.0000 0.0... .0 0030. 35 0.0...000 000... 000.0 00000 0.3.0.0 .00... 000.0 00000 0-0.500 000... 000.0 0.000 0.0.500 0..0. 000.0 00000 000.. 00.5000 0.00000 0.000-0.. 0.0...000 000.0. 00... 00.00 0.03.000 000.0. 00... 00000 0.02.000 000.0. 00... 00.00 0.0...000 0.0.0. 00... 00000 000.. 0000000 0.00000 0.00000 0.0.30.0 000.00 0....0 00000 0.0.00.0 000.00 000.. 0.000 000.. 0000000 0.03.00 0.03.00 0. o. .0 0:. 00. .0. 700075. 9. .4 2 000. 00.0 > 00.0.00 0.0.0.0008... 000200 0.0.... .900 .0 0000.080 000 20.000.02.000. 30.05.6000... 0...... ...-0000.08.00.00... .0 00:000. 0... .0 00:. .0200... 0... 0.0.. 0020. 0000200 .00....0 .00.. .0... 0.00.0000 0.0: .0 0.00.. 36 Table 4. Isotope effects for the reaction of trianisylmethane with triphenylmethyl tetrafluoroborate. t°C kH/ko' kH/ko" 22.002 3.1 00.7 6.61:0.5 30.0002 7.4002 7.3002 40.0002 7.0002 7.1 00.4 ' - first run " - second run E: 9.83 a . E: a... u C Tm72 .0 9.5.. c. 0.. .1; 37 23.0 «633.33.: 0.313631: 33.0 «638.33.: 0628933.: «63.2. 33.0 0.318.333 «632.356. «.3; 9318.30.58 «5:21.326. «.398 936.» $313.33.: «.Sxfitoamfie «3.0 0.319.391 «632.383 «63.8 :93: :9. ...: .92.. .9. .5. 0.. 6.2322598. 3:883:29... 5.3 28:35.22... .0 5:82 2: .0 «Soto 398. 93 «.5228 2a.... .m can... 38 Table 6. The isotope effects for the reaction of tritolylmethane with triphenylmethyl ' tetrafluoroborate. t°C kI-i/kp' "H/kD" 25.0002 6.5024 7.0005 ‘ 30.0002 0400.7 8.81:1.5 40.0002 ' 0200.5 9.5005 ' - first run " - second run .5. 9.89.. n : ...... .m... a . Twp: .0 9...... ... 0.. .7... 39 «.39. «31233.3 0.312.338 9390 «...—«3932.3 o-o.x.«93.m.«. «.393. 93«.«. 0.313.335 «.o.x.««.38.3 93.... «318.38.: «313.3092 «.398 «.39.: 0.313.336. «318.33.... .39.. 93:8.386. «6:34.386. «.398 :93... :9. ...... .910... .9. .... o... 6.8390258. _>;.oE.>co§E>m.=a o... 5.: 8285325... .0 5.88. a... .o 98.3 828. Ea $5.28 Ea: .«. can... 40 Table 8. Isotope effects for the reaction of trianisylmethane with anisyldiphenylmethyl tetrafluoroborate. t°C kH/ko' kH/ko“ 26.0102 8.61:1.1 10.3107 30.0102 9.11:0.6 12.7:10 40.0:02 8.9103 9010.2 ' - first run " - second run The large value of the isotope effect for reaction (1) was also confirmed by the results of the following experiment: An equimolar amount of trianisylmethane and its deuterated counterpart (0.147 mmol) was reacted with triphenylmethyl tetrafluoroborate (4x0.147 mmol), in acetonitrile (1 ml), and the progress of the reaction was followed by 1H NMR. Two reactions were run at 30°C and 41°C. spectra 6A-6E. At 30°C and at t-5 hrs, half of the hydride was transfered; however, the methoxy peak of the product was quite small compared with that of the starting material. This indicates that only a small amount of deuterium was transferred (sp. 68). At t-18 hrs, sp. SC, whereas hydride transfer was nearly complete, the methoxy peaks showed the presence of a large quantity of unreacted deuterated starting material. Likewise, at 41°C and at t-6 hrs, whereas hydride transfer was almost complete. again the methoxy peaks indicated that only a small fraction of deuterated starting material was reacted (sp. 65). The spectra 6A-GE and the integration data shown confirm the large magnitude of the kinetic isotope effect. .. .. . . . .... ...... .. ...... . .- d .. - . .. . .. - The possiblility of. hydride attacking the phenyl rings of the triphenylmethyl tetrafluoroborate followed by hydride rearrangments to the methine carbon was investigated as follows: Trianisylmethane-1-d (0.1263 g)and triphenylmethyl tetrafluoroborate (0.2489 g). in about 2 ml coacu, were allowed to react at 40°C until 95% completion. NMR showed no signal at 5.55 ppm where the methine of triphenylmethane absorbs (sp. 12). On high magnification a very small proton peak is detectable which can not be integrated because of its small magnitude. This very small peak is evident as early as when the reaction is only 1-2% complete and its intensity does not change after 95% reaction completion. It is also observed in all the reactions where deuterated compounds 42 were used. It is probably due: a) to the hydrogen species showing a very weak peak at 2.1 ppm in the 1H NMR spectrum of the CDacN, and which disappears when the reaction is carried out in this solvent or b) to unlabeled trianisylmethane present in the deuterated trianisylmethane which was prepared from «99.9%» LiAlD4 . Reaction of deuterated trianisylmethane and anisyldiphenylmethyl tetrafluoroborate was allowed to proceed to 51% completion. Upon mixing the components the 1H NMR (spectrum 18) shows no signal at 5.55 ppm, which denotes no anisyldiphenylmethane. After 3.4% reaction, (sp. 19), no signal is detected in this region. Even after three months, at which time 48.7% of reaction has occurred. no signal due to a methine is detectibie. If hydrogen rearrangement from the phenyl ring to the carbocation center had occurred, a methine proton at 5.55 ppm would have been detected. W The reaction of triphenylmethyl tetrafluoroborate (0.210 g) and triphenylmethane (0.164 g) having the central carbon labeled with 130 (50% excess), reaction (4) (CeHsia‘3CH + (CeHslscBF4 > (CeHsia‘308F4 + (06H5)3CH l ‘ . (4) at 40°C and at high concentration (about 2 ml solvent was used) was run for five months. The reaction was monitored by 1H NMR in the region of 5-6 ppm (5.32, 5.58, 5.83). As shown in spectrum 22, when compared with spectrum 21 of the starting material, reaction (4) has proceeded. Similarly, 13C NMR analysis of the same reaction corroborated this finding by showing an increase in the intensity of the methine carbon of the cation at 212.6 ppm (compare spectra 23 and 24). 4 3 Conclusions All the reactions studied in this section represent one-step hydride transfer reactions. The transfer occurs from the donor to the acceptor molecule with no new carbon-carbon bonds or any other products having formed. Investigations of the triphenylmethyl and anisyldiphenylmethyl cation reactions confirm that the only cationic reacting center in the triarylmethyl cations under the conditions of this study is the central carbon. The possibility of hydride attack on the aryl rings of the triarylmethyl cations followed by rearrangement to the methine carbon does not occur to a detectable extent. The reaction of triphenylmethyl tetraflouroborate and triphenylmethane proceeds very slowly. The very large isotope effects exhibited by these reactions, slightly larger than the maximum kinetic isotope effect O'Ferral reports for linearity, support a linear transition state for the reaction of triarylmethyl cations with triarylmethanes. The data are not accurate enough to determine activation energy parameters and assess the extent of any quantum mechanical tunnelling. CHAPTER III XANTHENE REACTIONS 44 EXPERIMENTAL I. Reparations 9-Deuteroxanthene and 9,9-Dldeuteroxanthene: A 30 ml THF solution of 9-xanthone (4 g commercial) was reduced to the corresponding alcohol in a 100 ml .round-bottomed flask by gradual and careful addition of excess of LiAlD4 (about 0.5 g) while stirring overnight at room temperature. The excess LiAlD4 was destroyed by water and the THF was evaporated. Ether was added to the flask and the organic layer was separated, was dried over M2804. and the solvent was evaporated. The alcohol was then treated like the alcohols in Chapter II with acetyl chloride at room temperature in THF for 40 minutes to yield the chloro derivative. The solvent was then evaporated under vacuum, the chloride redissolved in THF and treated like the chlorides in Chapter II with LiAIH4 or LiAlD4 for 20 minutes at room temperature. The acetonitrile solution of the 9—deuteroxanthene gave spectrum 26 and that of 9.9—dideuteroxanthene spectrum 27. Xanthyl Cation: The xanthyl cation (sp. 46) was prepared by the reaction of excess xanthene (0.8 g commercial) with the triphenylmethyl tetrafluoroborate (1.3 g) in CHacN (3.5 ml) at room temperature followed by addition of ether (about 4 ml) and cooling that led to the precipitation of xanthyl tetrafluoroborate. The small peak at 3.93 ppm and the impurities shown near the benzene ring protons in the sp. 46 may have originated from the species that exhibited the peak at 2.1 ppm and which disappeared. 45 46 2,6-Dlmethoxyxanthyl Cation: The 2,6-dimethoxy-9-xanthone which was used for the preparation of the 2,6- dimethoxyxanthyl cation was prepared according to the procedure described by P. Grover, G. Shah and R. Shah.16 For this 1 g 012,2°-dlhydroxy-4,4'- dimethoxybenzophenone (commercial) was heated with 20 ml of water in a 60 ml volume closed tube at 210°C for 48 hours. The cycloketone formed was crystallized from ethanol. This 2,6-dimethoxy-9-xanthone was reacted with an excess of mum in THF to form the corresponding alcohol (mp. 165-170°C). An excess of this alcohol was then treated with triphenylmethyl tetrafluoroborate in Cl-lacN to form the substituted xanthyl cation (sp. 47) under conditions similar to those needed to prepare the xanthyl cation. The 2,6-dimethoxyxanthyl cation was precipitated as the tetrafluoroborate salt by addition of THF, the solid was filtered and washed with THF. This cation reacted with cycloheptatriene in C03CN to form the trophylium ion; however, it was unreactive toward triphenylsilane in CDacN. 2,6-Dlmethoxy-9-Deuteroxanthene: The title compound was prepared by careful addition of an excess of LiAlD4 (about 0.15 g) to a stirred suspension of solid 2,6-dimethoxyxanthyl tetrafluoroborate (1 g) in ether (30 ml) in a 100 ml round-bottomed flask at room temperature. The excess LiAlD4 was destroyed after 20 minutes by water and the organic layer was separated and dried over Na2804. The white solid deuteroxanthene (mp. 115-116°C) had an NMR in acetonitrile solution with a characteristic symmetrical triplet due to the C-9 proton at 3.88 ppm (sp. 48). 47 Dlanlsylmethyl Perchlorate: This cation was prepared by a method similar to that used for the formation of anisyldiphenylmethyl tetraflouroborate in Chapter II. When 4,4-dimethoxybenzhydrol (4 9 commercial) was treated with perchloric acid (0.022 moles) which was added dropwise to an ice-cooled propionic anhydride solution (70 ml) of the benzhydrol, red crystals of the cation perchlorate formed. The crystals were filtered under argon and washed with petroleum ether (B.P. 35-60°C). The solid cation, whose NMR is shown in spectrum 37, is stable for 1-2 hours at room temperature and for a few days if refrigerated. However. a solution of the cation in C030N decomposed within 30 minutes. 48 II. Beadicnsfludied Reaction of 9-Deuteroxanthene with Triphenylmethyl Tetrafluoroborate (spectra 25-36) ' PM 0.0 ”0%2°;_:'%¢3°H(°) +* (1)&(2)spoctra25-ae . . The title reaction was carried out by weighing the 9—deuteroxanthene (0.025 g) and the cation (one third to four times the moles of the 9-deuteroxanthene), individually, in two separate vials. Either the cation or the 9-deuteroxanthene or both were dissolved in CDacN and the reactants were mixed together. For the reaction at 0°C, the 9-deuteroxanthene was dissolved in acetonitrile and then introduced into one arm of the flask shown in Figure 1. The triphenylmethyl cation was introduced into the other side-arm. Figure 1. The apparatus which was used for the low temperature reactions. The arms were then submerged in ice-water for ~20 minutes. At his time the contents of the two arms were mixed together. The flask was kept at 0°C for ~15 minutes and then warmed to room temperature. Spectra were then taken. The above procedure was used for the reaction of 44°C with an oil bath used to warm the arms of the apparatus. 49 Reaction of Xanthene with Anisyldiphenylmethyl Tetrafluoroborate (spectra 44-45): (H) ' 00") +H(D) [g ’0 + HCHaOCsHo (Curlew-3:9— MCHaOCeH.) (CeHshCHw) + (3. (3) spectra 44-45 For the reaction of xanthene with anisyldiphenylmethyl tetrafluoroborate. equimolar amounts of xanthene and anisyldiphenylmethyl tetrafluoroborate (0.14 mmols) were dissolved individually in 1 ml CDacN and the cation solution was transferred to the vial which contained the xanthene solution. The progress of the reaction was monitored by 1H NMR at 25°C. The reaction of dideuteroxanthene was carried out in the same way. A small peak appears at 2.1 ppm in the 1H NMR spectrum of 003CN, which disappears when these reactions are carried out in this solvent. Another impurity appears at 3.7 ppm near the methoxy peaks in both experiments Reaction of 9-Deuteroxanthene with Dianlsylmethyl Perchlorate (spectra 37-39, 41-43): D *D(H) +9iCHaOCaH4lz 0°” 2—°.C> p-(CHaOCal'hiz CH(D)H+ (4) spectra 37-39 spectra 41-43 For the title reaction, the 9-deuteroxanthene (0.025 g) and the dianisylmethyl perchlorate (two and a half times the moles of the 9—deuteroxanthene) were weighed individually in separate vials. Trideuteroacetonitrile (3 ml) was used to dissolve the cation and the reactants were then mixed together. 50 Reaction of 2,6-Dimethoxy-9-deuteroxanthena with Triphenylmethyl Tetrafluoroborate (spectrum 49): (5) spectrum 49 For the title reaction an excess of the cation (1 .3-2.5 times) which was initially dissolved in CDacN (0.26-0.5 ml) was added to the solid 2,6-dimethoxy-9« deuteroxanthene (0.0034-0.0048- 9). Reaction of 2,6-Dimethoxy-Q-deuteroxanthene with 9-Xanthyl Tetrafluoroborate (spectrum 50): oak»; “1.0. ...; (a) spectrum so The title reaction was carried out in the presence of excess 9-xanthyl cation (1.5 to 3 times molar excess). Each of the reactants was dissolved separately, either in the minimum amount of CDacN before mixing, or the cation was dissolved in coacN and then added to the solid 2,6-dimethoxy-9-deuteroxanthene. The quantities of the hydrocarbon which were used for each experiment were 0.0023 g to 0.025 g and the CDacN from a few drops to 1.5 ml. A Bruker 25 MHz spectrometer was used to study all the reactions. Spectra 25- 50 are representative of the reactions studied. "‘ acetonitrile. I 1 ($825) [_¥ ' 1k; Spectrum 25 is that of xanthene in acetonitrile, which was the solvent used in this study. ($825) AF Spectrum 26 is that of 9—deuteroxanthene in ——.—_— ($227} . \ Spectrum 27 is that oi 9,9-dideuteroxanthene in acetonitrile solution ._ A g A l A ‘ .0. ’0‘ .0. 7'. .0. 3.. ‘.° PPM 52' Spectrum 28 represents the reaction of equimolar amounts of xanthene and triphenylmethyl tetrafluoroborate. (5828/ Ii J Spectrum 29 represents the reaction oi the triphenylmethyl cation with excess 9.9- dideuteroxanthene. " D D 1’ +0 V o maniac—903+ 960 J (SP. 29 2 , I A fl II II I III #lQ—ll ) I)"; , 6"). ll fl. -. . D D 1.- " 3" IIII A‘QS’DT‘“ “95’ IIIIIII"I I, 'i‘\ ...? .i. W7 .3. .T. .i. .f. Spectrum 30 represents the reaction of the mixture listed in spectrum 29 with triphenylsilane-t-d. No peak appears at 4.02 ppm. ‘53 Spectrum 31 represents the reaction oi the mixture listed in spectrum 29 with xanthene. ' A+Cm—+.+ ($23!) 1 Spectrum 32 represents reaction oi the mixture listed in spectrum 29 with excess triphenylsilane ' ' A + @§;H——o ($17.32) —--—P _JL CDS ' 0 + I F SLOW -X ADDITION . I I 4L J a 2L - L L L L L a l L NJ 9.! am 1.0 FPM 0.0 5.0 4,0 3.0 Spectrum 33 represents the reaction resulting from a slow and ' . periodic addition oi deuterated triphenylsilane with the reaction mixture at xanthene and triphenylmethyl cation. 54 dgCHro) ($334} g (SP. 35) .’SR 36) L_ 1 1:0 1 0:0 0: 0.0 PPM 1.0‘ 0:0 5:0 4.0 3.0 Spectra 34, 35 and 36 represent the reaction of 9odeuteroxanthene with one third, one and tour mole excess oi triphenylmethyl tetrafluoroborate respectively. 55 (c H.0 cm, c HC/‘O. L LL Spectrum 37 is that oi oi dianisylmethyl perchlorate in acetonitrile. l l 1 (SF? 37} j l 000 I (3.9, 38} If Spectrum 38 is the spectrum oi the xanthene in acetonitrile. L L k L 1 g 0p 03:) +(Cl-7;0QH,),CH _. +(CH.0Q’-(),Ch; ($19.39) - L tLLl't ,‘.‘.|'lli_ou.__ Spectrum 39 represents the reaction oi dianisylmethyl perchlorate with excess xanthene. L I l i 1.0 i 0.0 0.0 0.0 PPM 1.0 0.0 3.0 56 Spectrum 40 represents the reaction oi dianisylmethyl perchlorate with excess deuterated triphenylsilane. A characteristic symmetrical triplet appears at 3.8 ppm. u H‘f' H .+ C. 3OCSDE‘D + (235,-De-sCHpCDgDof- 63,5; ' EX. . OCH, OCl-v, KL/ / f$R4J) Ifi ! # i ..LULI ' 57‘ Spectra 41, 42 and 43 represent the reaction oi 9-deuteroxanthene with one third, one and two mole excess of the dianisylmethyl perchlorate respectively. a 1" i- (c H,0 QHACH —» w 35399:) (0.17400. I140 HfD) H III. M; I. ‘I L I / / (SP. 4 2 ) h ILILLJISS J0 / . 2 ( SP. 4 3 ) J'IJUJ IolgfloL~k _ a IIU‘I \___ 11.0 1 0; 0:0 0 0 1:0 PPM 0.0 5? 1:0 3.0 58' Spectra 44 and 45 represent the reactions at xanthene and 9,9—dideuteroxanthene with anisyldiphenylmethyl tetrailuoroborate respectively. 03:3 +(0-(0C7—(2CH,),C+ H abs + (CH, 0C,H.)(C,H,),C‘H (SF! 44) l L "IL—bd'IU' L4: J“) H L.’ ¥ ‘_ #fi-JLH; ' OD m + (CH, 0C.H.)(C. H,),C+—————» D as + (cmocmmc. 113,00 {$19.45) WU IIII LL 1 0.0 0,0 0. 0 1. 0 0.0 0. 0 1.0 3.0 59 Spectrum 48 represents the xanthyl tetrafluoroborate in acetonitrile solution. ) ($946 . J" 74L ... Spectrum 47 is that oi 2,8odlmethoxyxanthyl tetrafluoroborate in acetonitrile solution. The singlet H 1 CR? Istigma», (SP. 4 7’ I _. L ___ _Jon I Spectrum 48 is that oi the 2,6-dlmethoxy-9- I I i deuteroxanthene in acetonitrile. D “.0 7 l (SP.43’ / /\,_. l 0.0 5.0 0,0 3.0 1 0.0 ’.I 60 Spectrum 49 represents the reaction oi 2.8- dimethoxy-9-deuteroxanthene with the triphenylmethyl tetrailuoroborate. cfllam + $3 C—fl mccfl“ 06”; C“? u 0 (SP. 491 .. UIII. i m. .l/ " L J 4‘ 1 J 0 1 r L 4 ; Spectrum 50 represents the reaction oi 2,8- dimethoxy-9-deuteroxanthene with the xanthyl tetrafluoroborate. Ex: o” ' ”’0’ .. I -' + .k.“___) «71°10 q“ CWN’Y‘QC’Q CW ergo ‘ ‘ oer!3 [3:3 I .S‘ F’. 5 O J I 1 1.0 1 0.0 0.0 0.0 1.0 0.0 3.0 4.0 3.0 PPM 61 Spectrum 25 is that of xanthene in acetonitrile. which was the solvent used in this study. A singlet due to the hydrogens on carbon 9 appear at 4.02 ppm and two multiplets due to the hydrogens oi the aromatic rings appear between 6.6 and 7.3 ppm. Spectrum 26 is that of 9.deuteroxanthene in acetonitrile. It looks like spectrum 25 but instead of the singlet at 4.02 ppm in spectrum 25, a symmetrical triplet appears at 4.02 ppm characteristic of the hydrogen on carbon-9 split by the deuterium. Spectrum 27 is that of 9.9-dideuteroxanthene in acetonitrile solution and the only peaks that appear are the two multiplets due to the hydrogens oi the aromatic rings between 6.6 and 7.3 ppm. No peak appears at 4.02 ppm. Spectrum 28 represents the reaction oi equimolar amounts of xanthene and triphenylmethyl tetrafluoroborate. The singlet at 5.55 ppm and the multiplet between 8.1-8.4 ppm are due to triphenylmethane. The xanthyl cation shows a singlet at 10.34 ppm. and a multiplet. a dupiet and a triplet between 8.0 and 8.8 ppm. Spectrum 29 represents the reaction of the triphenylmethyl cation with excess 9.9-dldeuteroxanthene. No peak appears at 10.34. 5.55 or 4.02 ppm. Spectrum 30 represents the reaction of the mixture listed in spectrum 29 with triphenylsilane-i-d. No peak appears at 4.02 ppm. Spectrum 31 represents the reaction of the mixture listed in spectrum 29 with xanthene. A peak at 10.34 ppm and an unsymmetrical triplet at 4.02 ppm appear in the spectrum. (The unsymmetrical triplet at 4.02 ppm is due to the superimposition of the singlet oi the xanthene and the symmetrical triplet oi the 9—deuteroxanthene). Spectrum 32 represents reaction of the mixture listed in spectrum 29 with excess triphenylsilane, whose silane hydrogen appears at 5.48 ppm. A symmetrical triplet appears at 4.02 ppm. Spectrum 33 represents the reaction resulting from a slow and periodic addition of deuterated triphenylsilane with the reaction mixture of xanthene and triphenylmethyl cation. An unsymmetrical triplet appears at 4.02 ppm. 62 Spectra 34. 35 and 36 represent the reaction of 9-deuteroxanthene with one third. one and tour mole excess of triphenylmethyl tetrafluoroborate respectively. Spectrum 37 is that oi oi dianisylmethyl perchlorate in acetonitrile. A singlet due to the methoxy groups appears at 4.14 ppm, a singlet due to the methyl hydrogen appears at 8.99 ppm and the AA'BB' hydrogens oi the phenyl rings appear In the region 7.3 - 8.3 ppm. Spectrum 38 is the spectrum of. the xanthene in acetonitrile. Spectrum 39 represents the reaction of dianisylmethyl perchlorate with excess xanthene. in addition to the peaks of the xanthyl cation and xanthene two duplets at 6.8 and 7.1 ppm (AA'BB' pattern) and two singlets at 3.72 (methoxy) and 3.81 (methylene) ppm due to the dianisylmethane are shown also in this spectmm. Spectrum 40 represents the reaction of dianisylmethyl perchlorate with excess deuterated triphenylsilane. A characteristic symmetrical triplet appears at 3.8 ppm. Spectra 41. 42 and 43 represent the reaction of 9-deuteroxanthene with one third. one and two mole excess of the dianisylmethyl perchlorate respectively. Spectra 44 and 45 represent the reactions of xanthene and 9.9- dideuteroxanthene with anisyldiphenylmethyl tetrafluoroborate respectively. Spectrum 46 represents the xanthyl tetrafluoroborate in acetonitrile solution. Spectrum 47 is that of 2.6-dimethoxyxanthyl tetrafluoroborate in acetonitrile solution. The singlet at 9.53 ppm is due to the hydrogen on carbon 9; the doublet at 8.2 ppm and the multiple at 7.4 ppm are due to the hydrogens oi the phenyl ring; and the singlet at 4.17 ppm is due to the methoxy groups. Spectrum 48 is that of the 2,6.dimethoxy-9-deuteroxanthene in acetonitrile. The doublet at 7.2 ppm and the multiple at 6.6 ppm are due to the hydrogens of the aromatic rings. The singlet at 3.77 ppm is due to the methoxy groups and the symmetrical triplet, which is characteristic of the hydrogen on carbon 9, appears at 3.9 ppm- 63 Spectrum 49 represents the reaction of 2.6-dlmethoxy-9-deuteroxanthene with the triphenylmethyl tetrafluoroborate. Spectrum 50 represents the reaction of 2.6-dimethoxy-9-deuteroxanthene with the xanthyl tetrafluoroborate. RESULTS AND DISCUSSICN WWWWD): The kinetic isotope effects for the reactions of xanthenes. except reaction 3. were calculated from the rates of competitive transfer of hydrogen versus deuterium from the same carbon. Carbon 9 of the xanthene and the substituted xanthenes was the one bearing the hydrogen and deuterium. in this case the isotope effect is given by the ratio of the concentration of hydrogen atoms transfered from carbon 9. versus the concentration of the hydrogen atoms remaining on this carbon. These concentrations correspond to the integrals of the areas the hydrogens signals in the proton NMR.. Because reaction (3) proceeded quite slowly. the isotope effect was calculated from the individual kH and kD determinations. which were carried out following the procedure described for the reactions in Section ll. ?:a¢;tl)on oi 9-Deuteroxanthene with Triphenylmethyl Tetrafluoroborate The average isotope effects were calculated to be 3.1:02, 3.4 t 0.4 and 3.7 :t 0.1 at 44°C. 24°C and 0°C respectively. They were determined from the ratio of the area under the peak at 5.56 ppm and the area under the peak at 10.34 ppm. Table 1 shows the kHI"D for this reaction, determined under different conditions. 64 65 Table 1. The dependence of the kH/ko on the ratio of the concentrationof 9-deuteroxanthene and triphenylmethyl tetrafluoroborate at different temperatures. Entry °C concentration ratios kH/"D (9-deuteroxanthene/ triphenylmethyl cation) 13 24.0:0.5 1:0.33 1.94 2‘)1 24.0:t0.5 1:0.50 2.30 3° 24.0105 1:0.66 2.37 4a 24010.5 1:1.00 2.97 5° 24.0:t0.5 1:1.00 2.65 6° 24.0:0.5 1:2.00 2.98 7° 24.0:0.5 1:3.00 3.35 8° 24.0:0.5 1:4.00 3.11 eaob 24010.5 1:4.00 3.88 10° 24.0:I:0.5 ~ 1:4.00 3.37 11d 24.0305 1.4.00 3.36 12° 24.0:i:0.5 1:4.00 3.12 13° 0.03:0.5 1:4.00 3.78 14° 0 03:05 1:2.00 3.65 15° 44.0305 1:4.00 3.38 15° 44.0;05 1:4.00 2.93 , 8* A solution of triphenylmethyl tetrafluoroborate in CDacN (1 ml) was added to the solid 9-deuteroxanthene (0.025 g). b Reaction 9 was also followed for 1.5 hr at intervals of 30 minutes. kH/kD values were 3.82. 3.39, and 3.93. 0 A solution of 9odeuteroxanthene in 0030N (1 ml) was added to solid triphenylmethyl tetrafluoroborate. d A solution of triphenylmethyl tetrafluoroborate in CDacN (1 ml) was added to the solution of xanthene in C03CN (1 ml). ° A solution of xanthene in CD30N (1 ml) was added to the solution of triphenylmethyl tetrafluoroborate in CD30N (1 ml). The kH/"D isotope effects were determined by dividing the area under the signal at 5.56 ppm due to the methine proton of triphenylmethane by the area under the signal at 10.34 ppm due to the methine proton of the xanthyl cation. All the reactions of xanthene. 9-deuteroxanthene and 9,9-dideuteroxanthene with triphenylmethyl tetrafluoroborate were studied in acetonitrile. All of them were extremely fast. 66 The following observations and comments can be made: a) b) d) 6) When the 9-deuteroxanthyl cation, generated from the triphenylmethyl cation and 9.9-dideuteroxanthene in situ. was reacted with triphenylsilane-1-d(¢3Si-D). the only product observed was 9.9- dideuteroxanthene. Therefore. the only carbocationic center reacting is the 0-9 (sp. 29-30). When the 9-deuteroxanthyl cation was reacted with xanthene. a proton peak appeared at 10.34 ppm (sp. 31) that is assigned to the 9-xanthyl cation. This finding establishes the xanthene-xanthyl cation equilibrium in solution. In the reaction of the xanthyl cation with triphenylsilane (slow and periodic addition of the silane), an unsymmetrical triplet appeared at 4.02 ppm (sp. 33). This signal was duplicated in the 1H NMR of a mixture of xanthene and 9-deuteroxanthene. indicating that both species are present in the carbocation reaction. When the 9-deuteroxanthyl cation was reacted with triphenylsilane (fast addition of the silane). only a symmetrical triplet was observed at 4.02 ppm (sp. 32). The above independent reactions show that the identity equilibrium between xanthene and the 9—xanthyl cation occurs quite easily. in order to eliminate this equilibrium which complicates the determination of kH/I‘D, the ratio of triphenylmethyl cation xanthene was varied. it was found that a 3-4 times excess of triphenylmethyl cation over xanthene gave reasonably constant ki-l/"D values (Table 1). 67 Reaction of Xanthene with Anisyldiphenylmethyl Tetrafluoroborate (3): Because the reaction between xanthene and anisyldiphenylmethyl tetrafluoroborate proceeded slowly. the rate constants kH and kD were calculated separately (by using the program which appears in the appendix. pages 94-95) and found to be (4.14:t0.10)x10'3 and (6.48:0.07)x10'4, respectively (appendix, pages 120-121). The progress of the reaction was followed by monitoring the methoxy peaks of the product and the reactant (sp. 44-45). The isotope effect for this reaction was calculated to be 64:02 at 25°C (appendix. pages 141). When equimolar amounts of xanthene and anisyldiphenylmethyl tetrafluoroborate (0.14 mmoles) were dissolved in C030N (2.0 ml) and the progress of the reaction was monitored by 1H NMR at 25°C, in 1/2 hour the reaction was 32% completed and an equal amount of each product was formed. as shown by the same integral for the peaks at 5.6 ppm and 10.5 ppm. with no indication of decomposition or by-products. The reaction of dideuteroxanthene which was carried out as above took 172 minutes to arrive at 32% completion. The percent reaction completion after 50 minutes was 44.3% for the xanthene and 12.7% for the dideuteroxanthene. Reaction of 9-Deuteroxanthene with Dianlsylmethyl Perchlorate (4): The isotope effects determined for the reactions of xanthene with dianisylmethyl perchlorate cover a wide range because of significant errors involved in the integration of various peak signals. Table 2 summarizes the integration values for the various xanthyl cation protons in the reaction of 9-deuteroxanthene with dianisylmethyl perchlorate. 68 Table 2. Integration values of the xanthyl cation protons in the reaction of 9-deuteroxanthene with dianisylmethyl perchlorate Entry Concentration Time 10.3 ppm 8.7 ppm 8.4 ppm 8.1 ppm ratio' sec (1H)° (4H)b (2H)b (2H)b 1 1:2.5 1.8x103 0.420 7.257 3.493 3.739 3.6x103 0.479 6.991 3.507 3.625 5.411103 0.457 6.759 3.392 3.509 ° Ratio of 9-deuteroxanthene : dianisylmethyl perchlorate. 3* Xanthyl cation methine hydrogen. b Xanthyl cation phenyl ring hydrogens. Table 3 below summarizes the integration values for the various xanthyl cation protons in the reaction of 9-deuteroxanthene with triphenylmethyl tetrafluoroborate (reaction 10. Table 1). Table 3. Integration values of the xanthyl cation protons in the reaction of 9-deuteroxanthene with triphenylmethyl tetrafluoroborate (Reaction 10. Table 1). Entry Concentration 10.3 ppm 8.7 ppm 8.4 ppm 8.1 ppm 5.4 ppm ratio" (1H)c (4H)d (2H)d (2H)d (1H)° 1 1:4 0.491 8.393 4.595 4.154 1.653 " Ratio of 9-deuteroxanthene : triphenylmethyl tetrafluoroborate. ‘3 Xanthyl cation methine hydrogen. d Xanthyl cation phenyl ring hydrogens. ° Triphenylmethane methine hydrogen. Comparison of the data summarized in Tables 2 and 3 reveals that the isotopic effects of the two reactions are quite similar. Unfortunately. precise measurements of the primary isotope effects was difficult due to secondary isotopic effects and significant errors involved in the integration of the xanthyl catlon multiple peaks. Table 4 presents the integration data of the protons of 9-xanthyl tetrafluoroborate in CDacN solution. 69 Table 4. Integration data of the xanthyl cation protons (solution of 9-xanthyl tetrafluoroborate in deuteroacetonitrile). 10.3 ppm 8.7 ppm 8.4 ppm 8.1 ppm (”0‘ (4H)b (2H)b (2H)b 10.573 53.288 26.679 10.764 25.311 a Xanthyl cation methine hydrogen. 5 Xanthyl cation phenyl ring hydrogens. Table 2 demonstrates that the integration ratio of the xanthyl hydrogen at C-9 with respect to the other xanthyl protons changes with time. The above limitations notwithstanding. one may make the following comments regarding the reactions of dianisylmethyl perchlorate with xanthene (sp. 37-39) and 9-deuteroxanthene (sp. 41-43). a) b) C) As observed in reaction (1). also in this reaction, varying the amount of carbocation with respect to that of 9-deuteroxanthene resulted in different ratios of the xanthyl cations formed (sp. 41-43). Spectrum 42 represents the reaction of equimolar quantities of both xanthene and dianisylmethyl perchlorate. It shows that the only reaction occurring is the isotope exchange between the methine carbon of the cation and the xanthene C-9 hydrogens. and that no new carbon-carbon bond or any other by-products are formed. Reaction of the cation with deuterated triphenylsilane gives a symmetrical triplet at 3.8 ppm (sp. 40). The solid cation is stable for 1-2 hours at room temperature and for a few days if it is refrigerated. However, a solution of the cation in CD30N decomposes in a matter of 30 minutes (sp. 43). 70 d) Reaction of dianisylmethane (commercial) with its cation counterpart results in a black solution. The NMR spectrum of this solution indicates that decomposition has occurred. Reaction of 2,6-Dimethoxy-9-deuteroxanthene with Triphenylmethyl Tetrafluoroborate (5): Table 5 summarizes the isotope effects calculated from integration of the various protons (sp. 49). The average isotope effect for this reaction is kH/kD - 5.4 3: 0.5. Table 5. isotope effects for the reaction of 2,6-dimethoxy-9-deuteroxanthene with triphenylmethyl tetrafluoroborate. kl-l/kc kH/kc kH/kb kH/kc kl-l/kb Entry' 8.2943 vs 6.2569 vs 6.301-6.247=1 5562*) vs 4.167° vs 9.530d ppm 9.530d ppm vs 9.530d ppm 9.530d ppm 9.5306 ppm 1a 4.91 5.28 1b 4.91 1c 6.22 4.99 1d 5.59 2a 5.58 5.28 5.74 5.35 2b 5.25 4.74 5.38 20 5.56 5.11 5.35 20 5.67 3a 5.20 6.06 5.70 5.69 3b . 5.91 ’ Three reactions were run 1. 2 and 3. 1a, 1b, 1c. 1d give the isotope effects which were calculated from different integrations on the same sample. The same is true for 2a. 2b, 2c, and for 3a, 3b. 3 Integration of the 2,6-dimethoxyxanthyl cation phenyl hydrogens. b Integration of the triphenylmethane methine hydrogen. 0 integration of the 2.6-dimethoxyxanthyl cation methoxy hydrogens. d Integration of the 2.6-dimethoxyxanthyl cation methine ring hydrogens. Since this reaction proceeded at a fast rate, an excess of the cation (1.3-2.5 times) was used. Entry 1 represents the data for the reaction in which the cation was initially dissolved in CDacN. in order to consume the species represented by the peak at 2.1 ppm in the solvent, followed by the addition of this solution to solid 2.2-dimethoxy- 9-deuteroxanthene. Computation of the kH/"D for this reaction was troublesome if the 71 C-9 proton of the xanthyl cation is integrated with respect to the remaining aromatic protons of the xanthyl cation. because of the proximity of the latter with those of the triphenylmethyl cation. However, the former proton may be compared accurately with either the triphenylmethane proton (5.58 ppm) or the CCH3 protons of xanthene at 4.17 ppm. Entries 2 and 3 represent the same reaction condition as that on entry 1, without an excess of the cation present. Thus, in this case. the xanthyl cation aromatic protons may also be used for the computation of kH/"D. The values shown in Table 5 have been corrected for errors in integration based on the data in Table 6, which are the integration values of the protons of pure 2,6-dimethoxyxanthyl tetrafluoroborate. Table 6. Integration of the 2.6-dimethoxyxanthyl tetrafluoroborate protons in CDacN solution. 9.530 6.294 6.256 8.301-8.241 4.167 (1H)° ppm liH)b ppm 11H)” ppm (2H)b ppm (6Hi° ppm 1.885 1.902 1.967 4.274 13.453 ° 2.6-dimethoxyxanthyl cation methine hydrogen. 9 2.6-dimethoxyxanthyl cation phenyl ring hydrogen. c 2.6-dimethoxyxanthyl cation methoxy hydrogens. Reaction of 2,6-Dlmethoxy-9-deuteroxanthene with Xanthyl Tetrafluoroborate ( 6) : The kinetic isotope effects of the reaction of 2.6-dimethoxy-9-deuteroxanthene with xanthyl tetrafluoroborate are summarized in Table 7. Because of substantial errors involved in the integration of the small C-9 protons of the 2,6- dimethoxyxanthyl cation and the multiplicity of the other aromatic hydrogens, the computed isotope effects Vary greatly. 72 Table 7. Isotope effects' for the reaction of 2,6-dimethoxy-9-deuteroxanthene with 9-xanthyl cation. JkH/kc . kl-l/kD kH/kD kH/kc Entry“ Temperatur 8.2943 vs 6.256al vs 8.301-8.214° 4.167b vs °C 9.530° ppm 9.5306 ppm vs 9.5306 ppm 9.530° ppm 13 24.0:l:0.5 4.40 4.75 4.74 4.60 23 24.0:0.5 7.57 8.90 2b 24.0:t0.5 5.24 6.86 20 24.0:t0.5 7.57 8.85 33 24.0i0.5 8.42 9.45 3b 24.0:l:0.5 8.42 8.53 43 24.0:05 9.03 10.04 10.49 11.37 4b 24.0i0.5 7.85 9.16 7.85 8.57 53 24.0:t0.5 7.41 8.11 8.31 9.08 SD 24.0105 4.63 5.96 5.16 5.72 63 24.0:0.5 4.26 _73 0.0;05 3.65 4.04 3.74 3.64 ' The isotope effects shown in this Table have been corrected for errors in integration based on the data in Table 6, which are the integration values of the protons of pure 2,6- dimethoxyxanthyl tetrafluoroborate. ” Seven reactions were run 1, 2. 3. 4, ,5. 6, and 7. The isotope effects calculated from different integrations on the same sample are denoted by a, b. or c. 3 integration of the 2.6-dimethoxyxanthyi cation phenyl hydrogens. ‘3 integration of the 2,6-dimethoxyxanthyi cation methoxy hydrogens. 0 Integration of the 2.6-dimethoxyxanthyl cation methine hydrogen. This reaction was carried out in the presence of excess of the 9-xanthyi cation (1.5 to 3 times molar excess). However. as the cation concentration increased, so did the errors associated with integration. because of the interference of the nearby peaks of the cation. All entries in Table 7 involve reactions where each of the two reactants was separately dissolved in the minimum amount of solvent before mixing. except entries 4 and 5 where the cation was first dissolved in C03CN. and then added to the solid 2.6- dimethoxy-9-deuteroxanthene. 'ueborpliu out pus sbuu answers om eul buuseq uoqrso out 01 eusus eut urou leisuslt uinuetnep Jo; petoedxe euo out 51 peuuoi 100de out ‘SWJ. °uidd an» is teldirt [soutetuuiiis out Aiuo smous (00 'ds) eususMueqdm petsletnep out 1.1th etsioluoied Muteiuuisluslp to uouosel out uteri uinltoeds out ‘ulsbv 'uoqrso 5-0 out 01 tueurbususel epupliu tuenbesqns tlilM (s-o usut .leuto) etsloqoronusltet piutusxlixoutelulp-g‘z out to suoqlso out to euo uo epupliq to nostis sepnioxe Duipuu siut ‘uisliv WM 99's is teidui lsouteuiuliis out Aluo pemous (91: 'ds) punoduioo out to ulnitoeds eul 'BIBJOQOJOi'luBJlBi liiqtusxlixouteuiip -9‘z puos M01164 pepuedsns out at pepps ssM appetnep lunululnls tunlutu to sseoxe us ‘19016 111 eueutusxoretnep-e-lixouteullp-g‘z to uoltsisdeld out Dunno 'UOQJBO 5-9 out 01 tueulebususel epupliu iuenbesqns 11th (60 Hall: 19010) uoueo liulusx-s all: :0 suoqm all: to euo uo mane 90119411 to Mlllqgssod out epnioxe stueuliledxe seen; '(09 'ds) uldd 30'718 petoetep SBM used cu “(g-13841) D-L-eususuiueudm iouop epiletnep 6141 1mm petosei ssM etsioqoronusitet liiuiusxoretnep-s ueuM 'erouueuung 'uldd gay is lBldiJi isollteuiuiiis s Aluo pemoqs (39 'ds) uoltosel out to srtoeds BWN out ‘eususlliueudut rouop epglpiiu out 0th petseit ssM etsioqoronusliet Iliqiusxorelnecs ueuM ‘pessnoslp lipsells sv . . . .. . l0 ' ll: - :l 0 - I“! :I0 l00l- 10 -0:i: :0 0 l :0 00 0| 6L 74 + 03C fl. 3.614(6) ° (1) kH/kD - 3.421: 0.4 0 +01”) 0.0 + 030 —o—9> MCI-RD) + (23) e e kHlkD - 3.7:t0.1 I +D(H) “. .36 J1". ..ch) .* (2b) 0 kH/kD . 3.1 i 0.2 H(D) (H) - Q . O + 1011.06.11.) (0.115120 -z‘—°-» (011.com lowechlD) +* C kH/kD :- 6.4i0.2 A 3) D +Di”) ©‘O + (CH3006H4)2 C’H 34.2. (CH3065H4)2 CH(D) H + g Q (4) O Isotope effect is nearly the same as reaction (1) but not identified precisely. 7 5 O . * ' (H) 24°C “3&3: ”a W) + CHAwa (5) kH’kD I 5.41:0.5 isotope effect is large but not precisely identified. 7 6 Conclusions All the reactions studied in this section are consonant with a one-step hydride transfer from the donor hydrocarbon to the acceptor carbocation. These reactions proceed with large kinetic isotope effects, which place them in the range assigned by O'Ferral to cases where the angle between donor-hydrogen-acceptor is 120°-150°, on the assumption that the force constant on either side of the hydrogen is the same. It is also possible that the transition state is linear with unequal force constants on either side of the hydrogen. The isotope effect of the triphenylmethyl cation with xanthene (3.43:0.4) is smaller than the one of the same cation with the dimethoxy substituted xanthene (5.4 10.5). A possible explanation for this finding may be a more symmetrical transition state of the latter reaction that has more similar force constants. The same explanation may apply for the isotope effect of the former reaction which is smaller than the isotope effect of the same hydrocarbon and the anisyldiphenylmethyl tetrafluoroborate (6.4:t0.2). The identity reaction between xanthene and the 9-xanthyl cation is quite fast compared with the corresponding reaction of triphenylmethane and the triphenylmethyl cation (Chapter Ii). Investigations of the xanthyl and 2.6-dimethoxyxanthyi cation reactions confirm that the only cationic reacting center present in these cations, under the conditions of this study. is carbon-9. The possibility fo hydride attack on one of the carbons of the xanthyl or the 2,6-dimethoxyxanthyl cation (other than C-9) followed by rearrangement of hydride to the C-9 does not occur. Again the reactions of the dianisylmethyl perchlorate confirm that the only cationic reacting center in this cation is the carbon bearing the two phenyl rings and the hydrogen. CHAPTER lV TRIPHENYLSILANE REACTIONS 77 EXPERIMENTAL I animations Triphenylsilane and Triphenylsilane-1 -D: The title compounds were prepared‘7o18 by reduction of the corresponding chloride (commercial). like in Chapter II. with lithium aluminum hydride and lithium aluminum deuteride respectively. and purified by vacuum distillation. Trianisylmethyl Catlon15: When trianisylmethanol (6g) was treated with tetraflouroboric acid (4.6 ml. 48% aqueous solution) in propionic anhydride (45 mi). crystals of trianisylmethyl tetrafluoroborate formed immediately. The salt was filtered under argon and washed with petroleum ether (B.P. 35-60°C). The crystals, because of the fast crystallization, contained a very small amount of propionic anhydride. 11mm Triphenylmethyl catlon reaction (spectrum 51): 038%! + 03810 + 030+ ——’ '0aSi+ + ¢3CH(D) (1) 90.9mm The reactions of triphenylmethyl tetrafluoroborate with the triphenylsilanes were carried out twice. In the first run an equimolar amount of triphenylsilane (0.568 mmol) and its deuterated derivative dissolved inCDacN (1 ml) was added to an equimolar amount of solid triphenylmethyl tetrafluoroborate (0.568 mmol). In the second run an equimolar amount of the silane compounds (0.571 mmol) dissolved in CD30N (1 ml) ' ¢331+ decomposed. 78 79 was added to an equimolar amount of the cation dissolved in CDacN (1 ml). The reaction was also performed at 20°C by using the apparatus depicted in Figure 1 (page 48). For this reaction 0.585 mmol of each silane compound in 003CN (1 ml) and an equimolar solution of the cation (0.585 mmol) in c030N (1 ml) was used. Anisyldiphenylmethyl cation reaction (spectrum 52): ¢3SiH + ¢3$iD + P-(CH3006H4)(06H5)20*_" 'essi” + p-(CH30C6H4)(06H5)20H(D) (2) 309611310142 The reactions of anisyldiphenylmethyl tetrafluoroborate with triphenylsilanes were performed twice. In the first experiment. equimolar amounts of each component (0.693 mmol) dissolved In CD3CN (2 ml) were added to solid cation (0.693 mmol). In the second experiment. anisyldiphenylmethyl tetrafluoroborate (0.675 mmol) dissolved in CDacN (1 ml) was added to a solution of equimolar amounts of each silane compound (0.675 mmol) dissolved in CDaCN (1 ml). Trianisylmethyl catlon reaction (spectrum 53): ¢3SiH(D) + P-(CH3006H4)30*'—" "has“ + p-(CH30C6H4)3CH(D) (3) Win—5.3 For the reaction of trianisylmethyl tetrafluoroborate with triphenylsilanes. the cation (0.581 mmol) in 2 ml CDacN was added to equimolar amounts of triphenylsilane and its silyl deuterated derivative (0.581 mmol) dissolved in CD3CN (2 ml). in a different experiment, trianisylmethyl tetrafluoroborate (0.389 mmol) in 2 ml CDacN was added to an equimolar amount of triphenylsilane in 4 ml CDacN at 25°C. The reaction of deuterated triphenylsilane with the cation was similar. ' 038i+ decomposed. 80 Reaction of TriphenylmethylTetrafluoroborate with Silyl Deuterated Triphenylsilane (spectrum 54): 16810 + 036* —-’ 3331+ + 0300 (4) W The reaction of triphenylmethyl tetrafluoroborate with silyl deuterated triphenylsilane was carried out with 0.247 mmol of the cation and excess triphenylsilane-i—d (0.280 mmol) in 1 ml CDacN. A Bruker 250 MHz NMR instrument was used to study all the reactions. The following spectra 51-54 are representative of the reactions studied: ' 03$i+ decomposed. rigs-71+ c135 10+ czgc’io c0394 cos-0- cm I : : J J / / <05? (0) Spectrum 51 represents the reaction of equimolar amounts of triphenylsilane and its deuterated analog with an equimolar amount of triphenylmethyl tetrafluoroborate. The peak of the silane hydrogen of the triphenylsilane appears at 5.48 ppm and that of the methine of triphenylmethane at 5.59 ppm. II- L 1* A_ A A l L A a a 0.0 0.0 7.0 PPM 0.0 5.0 0.0 . CD @JSiHor- g3; 04: ¢c"—+ dgsi‘Hvi- @5707“ C) 53MB) <5 ' <35? ‘79 OCH ’ I . OCH. - ' II . 1 f P. D 2 . S 1 tall“ I 11.; . “JV,” PPM Spectrum 52 represents the reaction of equimolar amounts of triphenylsilane and its deuterated analog with an equimolar amount of anisyldiphenylmethyl tetrafluoroborate. The peak of the methine hydrogen of the anisyldiphenylmethane appears at 5.53 ppm. _ _ it; I 8 . 9. Mill. Gozwbarsmulecmw. I. \o 0.4.1 09413. 638 is ... _s 1 4mm mus g GmuAV . c a . ... . Pd...» $0355. «a 30:82... 51.31323. 855530033. so 2:2 .836 mouse. .: ...... 8355.. 9883:. me 30302.8 30 38.3: o‘ 3:522 2:258 2 33035.53 2:. .3 3503.8 25.3 (.5 a: 35322 2:25. o. 35.3.3051 33.823220. .25 use: 2 30 30550 333cc: 2 .30 59:31:53»? moves; 2 mbu 33. 2006\6 I; F . #1,- ; ... + 60.93.683.111 5.9..P 9.66 50 33:0: 2 35038.. 52.03.9530 15 mxeeu. ‘2 30 35:10 ueexe ace 8 3e p.53. 396cc? 83 Spectrum 51 represents the reaction of equimolar amounts of triphenylsilane and its deuterated analog with an equimolar amount of triphenylmethyl tetrafluoroborate. The peak of the silane hydrogen of the triphenylsilane appears at 5.48 ppm and that of the methine of triphenylmethane at 5.59 ppm. Spectrum 52 represents the reaction of equimolar amounts of triphenylsilane and its deuterated analog with an equimolar amount of anisyldiphenylmethyl tetrafluoroborate. The peak of the methine hydrogen of the anisyldiphenylmethane appears at 5.53 ppm. Spectrum 53 represents the reaction of equimolar amounts of triphenylsilane and its deuterated analog with an equimolar amount of trianisylmethyl tetrafluoroborate. The peak of the methine hydrogen _of the trianisylmethane appears at 5.37 ppm. Spectrum 54 represents the reaction of deuterated triphenylsilane with triphenylmethyl tetrafluoroborate. Except for the multiple peaks due to the phenyl hydrogens no other peaks appear in this spectrum. W Warren“ kHike). Typically, these reactions involved the addition of equimolar amounts of triphenylsilane and its silyldeuterated counterpart to an equimolar amount of the cation, in deuteroacetonitrile solution. The reactions were generally exothermic and proceeded instantaneously, except for the reaction with the trianisylmethyl tetrafluoroborate (3), which proceeded at a much slower rate. When two isotopic molecules, A1 and A2. undergo analogous irreversible reactions in intermolecular competition k1 A1 + B ———-) (productsh k2 A2 + B ——) (prodUCts)2 the isotope effect is given by: where a1° and a2° are the initial concentrations and a1 and a2 are the remaining fractions of the two isotopic species. From the 1H NMR spectra of the reaction mixture. the isotope effect was calculated by comparing the concentration of the triphenylsilane with that of the triphenylmethane (or anisyldiphenylmethane), which is equal to the concentration of unreacted deuterated triphenylsilane. JAL !Sl'i log [Ac] kD ' |A° - Al log [A0] 84 85 [A] - the area at 5.48 ppm of the triphenylsilane proton [Ac] - the area of the triphenylsilane proton at 5.48 ppm plus the area of triphenylmethane methine proton at 5.59 ppm (or anisyldiphenylmethane methine proton at 5.53 ppm, or trianisylmethane methine proton at 5.37 ppm). The isotope effect of reaction 3 was calculated from determination of the individual kl—l and k0, by following the procedure described for the reactions in Chapter II. Triphenylmethyl cation reactions (1)“: The reactions of triphenylmethyl tetrafluoroborate (sp. 51) with the triphenylsilanes were carried out twice. In the first run, an equimolar amount of triphenylsilane and its deuterated derivative dissolved in CDacN was added to an equimolar amount of solid triphenylmethyl tetrafluoroborate. The calculated kH/kD was found to be 1 .8i0.4. In the second run, an equimolar amount of the silane compounds dissolved in dideuteroacetonitrile, was added to an equimolar amount of triphenylmethyl tetrafluoroborate dissolved in CDacN; the isotope effect was calculated to bet 310.1. So, for this reaction kH/"D - 1.81:0.4. The reaction was also performed at -20°C. The isotope effect was found to be 2 .1 :l:03 . Anisyldiphenylmethyl I cation reactions (2): The reactions of anisyldiphenylmethyl tetrafluoroborate with the triphenylsilanes (sp. 52) were performed twice. In the first experiment equimolar amounts of each component dissolved in 0030N were added to solid anisyldiphenylmethyl tetrafluoroborate. giving an isotope effect of 2.03. 86 In the second experiment equimolar amounts of each silane compound dissolved in acetonitrile were added to an equimolar amount of anisyldiphenylmethyl tetrafluoroborate dissolved in acetonitrile, giving an isotope effect of 2.41. So, the averaged kH/"D - 2.2 :1: 0.2 for this reaction. Trianisylmethyl catlon reactions (3): The reaction of trianisylmethyl tetrafluoroborate with triphenylsilanes at 24°C is quite slow. For this reaction equimolar amounts of triphenylsilane and its silyl deuterated derivative dissolved in cogcu were reacted with an equimolar amount of the trianisylmethyl tetrafluoroborate. Upon completion of the reaction, as monitored by the disappearance of the methoxy proton peaks of the cation, 54-57% of the triphenylsilane (sp. 53) had reacted resulting in an isotope effect ki-l/kb - 1.38 s 0.09. In a different experiment equimolar amounts of triphenylsilane and trianisylmethyl tetrafluoroborate were mixed together in acetonitrile at 25°C and the progress of the reaction was followed by NMR. After 77 minutes, 36.3% of the cation had reacted. From the integrated areas under the methoxy peaks for the reactant and product, by using the program which appers in the appendix, pages 94-95, the kH for the reaction was found to be (2.00 :I: 0.03)x10‘3 (appendix, page 122). Similarly, for the deuterated triphenylsilane kD was found to be (1.21 : O.01)x10'3 (appendix, page 123). in this reaction 26% of the cation had reacted in 79 minutes. So, from this experiment kH/"D - 1.66 t 0.03 (appendix, page 142). A source of error in the reaction rates is due to the propionic anhydride trapped in the cation crystals during the cation preparation. The NMR of the reaction of the triphenylmethyl cation with deuterated triphenylsilane (sp. 54) shows no proton peak in the region 5-6 ppm indicating that only deuterium has been transferred to the C-1 of the triphenylmethyl cation. 87 When the reaction of deuterated triphenylsilane reacted with 38.7% of the trianisylmethyl cation (3) the NMR spectrum showed no proton peak in the region of 5- 6 ppm. This finding clearly demonstrates that the deuterium has been transferred from triphenylsilane to the central carbon of the trianisylmethyl cation. 88 S Hill “It" I. .H. I. 24°C egsiH + ¢3SID + e3c+ ——+ «1331+ + ¢30H(D) kH :6 - 1.8:t0.4 -2o°C' ¢3SiH + ¢3$iD + 03C+ ——) ¢3$i+ + ¢3CH(D) — - 2.1 21:0.3 24°C 1933‘“ + ¢3Si0 + (Cl-l3006H4)(Ce|'is)20+ —-9 ¢38i+ + (CH3OC6H4)(C6H5)ZCH(D) kit '6 I 2.2 It 0.2 25°C ¢3$iH(D) + (CH3006H4)3C+ —-) ¢38i+ + (CH30C3H4)3CH(D) kH E . 1.66 1; 0.03 8 9 Conclusions All the reactions studied in this section are one-step hydride transfer from the donor hydrocarbon to the acceptor carbocation. Investigations of the triphenylmethyl and trianisylmethyl tetrafluoroborate reactions with the triphenylsilanes confirm that the only cationic reacting center In the triarylmethyl tetrafluoroborates under the conditions of this study is the central carbon. Hydride attack on the aryl rings of the triarylmethyl tetrafluoroborates followed by rearrangement to the methine carbon does not occur. The isotope effects of the triphenylsilane reactions are very small. Their magnitude lay in the range which was assigned by O'Ferral to a very small angle (around 90°) between donor-hydrogen-acceptor, on the assumption that the force constants on either side of the hydrogen are the same. Since in the triphenylsilane reactions the force constants on either side of the hydrogen are not expected to be the same no conclusions can be made for the transition state in these reactions. 90 All the reactions studied in this study which involve triarylmethanes with triarylmethyl cations; xanthenes with xanthyl, triarylmethyl and dianisylmethyl cations; and triphenylsilanes with triarylmethyl cations represent one-step hydride transfer reactions. The transfer occurs from the donor to the acceptor molecule with no new carbon-carbon bonds or any other products being formed. Investigations of the triphenylmethyl, anisyldiphenylmethyl and trianisylmethyl cation reactions confirm that the only cationic reacting center in these cations under the conditions of this study is the methine carbon. Hydride attack on the aryl rings of the above cations followed by hydrogen rearrangement to the methine carbon does not occur. Also investigations of the 9-xanthyl, 2.6—dimethoxy-9-xanthyl and dianisylmethyl cation reactions lead to similar conclusions. The very large isotope effects exhibited by the reaction of triarylmethanes with triarylmethyl cations, which are slightly larger that the maximum kinetic isotope effect O'Ferral calculated for linearity, support a linear transition state for these reactions. The reactions of xanthenes with xanthyl, triarylmethyl and dianisylmethyl cations also proceed with large kinetic isotope effects, which place them in the range assigned by O'Ferral to cases where the angle between donor-hydrogen-acceptor is 120°-150°, on the assumption that the force constant on either side of the hydrogen is the same. It is also possible that the transition state is linear with unequal force constants on either side of the hydrogen. No assessment of the structure of the transition states of the reactions of triphenylsilanes with triarylmethyl cations can be made because the small kinetic isotope effects, in the range assigned by O'Ferral to cases where the angle between donor-hydrogen-acceptor is about 90°, may be due to large differences in the force constants on either side of the hydrogen bonds in the transition state. 91 92 The identity reaction between the triphenylmethyl cation and triphenylmethane is quite slow compared with the corresponding reaction of xanthene and the 9-xanthyl catlon. APPENDIX 93 94 Computer program for the calculations of the rate constants of reactions 1 and 2 from Chapter ll. 3 from Chapter III and 3 from Chapter IV. Oinclude Oinclude idetine LINE 80 mainl) I FILE *pr: char data_£ile[20]: float t[30]: float a.V,A3[30],A4l30]; /* data */ int n: /* data number */ double yt30]: double SumT-O,SumT2-0,Sum¥-O.SumY2-0,SumYT-0,SSt,SSy,SSty.SSE.SlopeDev: char titlelBO]: int 1: double store: double k,b: /* estimated slope and intersept '/ double sqrtli: /* Read data '/ scan£("§s',data_£ile): pr-fopenldata_£ile."r"1: qutsltitle.LINE.pr): fscanttpr."§£",&al: fscanf(pr,"§£",&V): i-O: while(!feo£(prl) { £scan£tpr,"%£",&t[i]); tscan£(pr,"%£",&A3[i]): fscanf einclude Odotin. LINE 80 main() { FILE 1'pr: char data_£ile[20]: float tl30]: float a.V.A3l30],A4[30]: /* data */ int n: /* data number */ double y[30]: double SumT-O,Sum12-O,Sum!-O.Sum!2-O.Sum¥T-O.SSt.SSy,SSty.SSE.SlcpeDev: char titlel80]; int 1: double store: double k.b: /' estimated slope and intersept */ double sqrtl): /* Read data */ scanti'ts'.data_£ile): pr-topenldata_£ile."r'): fgetsltitle.LINE,pr): £scan£(pr.”\£',aa): £scantlpr,'t£'.&v1: i-O: whileflteotlprli ( £scan£lpr,'§t'.£tfi)): tscant(pr,'§£',&A3[i]); tscanttpr,"$t',&a4(i)): ylil-A4fiJ'V/(3000*a'A3Iill: i++r } n-i-l: for (i-O:i