NEREQ‘CE‘EEMESZ'EEY: LAB‘ELE'W; A¥§ {NFRARED SFECTEAL $113355 05 CHLQRQEES (2.4:FENTANENGNATQ) E’ENTAHAF’TGCY‘ELG?EN‘EAE}i‘gNY’LZRCGNSEM Thesis 5m- é‘m ‘39qu a? M; 5. msmm SYATE {EHEEFEESEW E Dean Bufier 3%? THESIS LIBRARY Michigan State University i4 .Ifivw—yfi. .4...” ABSTRACT STEREOCHEMISTRY, LABILITY. AND INFRARED SPECTRAL STUDIES OF CHLOROBIS(2.4-PENTANEDIONATO)PENTA- HAPTOCYCLOPENTADIENYLZIRCONIUM BY E. Dean Butler The stereochemistry of chlorobis(2,4—pentanedionato)— pentahaptocyclopentadienyl zirconium, (h5-C5H5)Zr(acac)2C1, has been investigated in solution by nuclear magnetic resonance spectroscopy. The simplest configuration con— sistent with the results is based on an octahedron in which the C5H5 ring Occupies one stereochemical position and the chlorine atom is positioned gig to the ring. A superior description based on a dodecahedron is also discussed. Four geometric isomers have been observed for the benzoyl— acetonate analogue. This latter result supports the stereo— chemical assignment for (h5-C5H5)Zr(acac)2Cl. At elevated temperatures (h5—C5H5)Zr(acac)2C1 under- goes a rapid, first order, stereochemical rearrangement process. The kinetics of the rearrangement have been studied by nmr line broadening techniques. Also, infrared vibration frequencies in the region 1600—155 cm.1 have been assigned for (h5-C5H5)Zr(acac)2C1. vAssignments for (h5-C5H5)Zr(acac-d7)2C1 and (h5—C5H5)Hf(acac)2Cl are in— cluded for comparison. STEREOCHEMISTRY, LABILITY, AND INFRARED SPECTRAL STUDIES OF CHLOROBIS(2,4-PENTANEDIONATO)PENTA- HAPTOCYCLOPENTADIENYLZIRCONIUM BY E? Dean Butler A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry 1969 ,— \\ C\ M; 5%? '\_. 9‘K N To Brendii ii ACKNOWLEDGMENTS Graduate work has at least two results: it leaves an indelible mark on the psyche of the student, and it leaves him with a debt of gratitude to many people. I acknowledge three distinct types of assistance. First, the ideas that represent the nucleus of this thesis were provided by Dr. T. J. Pinnavaia, who is not only an academician and practi— tioner in the highest sense, but a patient, understanding mentor as well. Second, I thank Dr. J. B. Kinsinger for allowing me to hold a teaching assistantship in the Depart- ment of Chemistry while I pursued an MBA as well as the de— gree for which this work is partial fulfillment. Finally, for the inspiration that has made the agony of this graduate student more than worthwhile, I thank Brendii. iii TABLE OF CONTENTS Page I. INTRODUCTION . . . . . . . . . . . . . . . . . 1 II. EXPERIMENTAL . . . . . . . . . . . . . . . . . A. Preparation of Compounds . . . . . . ~Reagents and Compounds . . . . . . . . General Techniques . . . . . . . . . . Chlorobis(2,4-pentanedionato)penta- haptocyclopentadienylzirconium . . . . Chlorobis(2,4-pentanedionato-d7)penta- haptocyclopentadienylzirconium . . . . 10 Chlorobis(2,4—pentanedionato)penta— haptocyclopentadienylhafnium . . . . . 10 Chlorobis(l~phenyl-1,3-butanedionato)— pentahaptocyclopentadienylzirconium . 11 COWU'IUUI 03 0' lb OONH B. Attempted Preparation of Chlorobis(1,1,1—tri- fluoro—Z,4-pentanedionatopentahaptocyclo- pentdienylzirconium . . . . . . . . . . . . 12 C. Attempted Preparations of Bromobis(2,4-3 pentanedionato)pentahaptocyclopentadieny1— zirconium . . . . . . . . . . . . . . . . . 13 1. Reaction of Zirconocene Dibromide and Tetrakis(2,4-pentanedionato)zirconium. 13 2. Reaction of Zirconocene Dibromide and Acetylacetone . . . . . . . . . . . . 14 D. Physical Measurements . . . . . . . . . . . 15 1. Molecular Weight Determinations . . . 15 2. Conductance Measurements . . . . . . . 15 3. Melting Point Determinations . . . . . 16 4. Infrared Spectra . . . . . . . . . . . 16 5. Nuclear Magnetic Resonance Spectra . . 17 iv TABLE OF CONTENTS (Continued) Page III. RESULTS AND DISCUSSION . . . . . . . . . . . . 20 A. Preparative Chemistry . . . . . . . . . . . 20 B. Characterization of (h5—C5H5)Zr(acac)2Cl, (h5-C5H5)Hf(acac)2Cl,.and (h5-C5H5)Zr(bzac)2Cl . . . . . . . . . . . 21 C. Kinetics of Configurational Rearrangement for (h5-C5H5)Zr(acac)2Cl . . . . . . . . . 41 D. Assignment of Infrared Vibrational Fre— quencies For (h5-C5H5)Zr(acac)2Cl, (h5-C5H5)Zr(acac—d7)2Cl, and (hs—C5H5)Hf (acaC)2C1 o . o o . . . . . o o 57 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . 73 APPENDIX . . . . . . . . . . . . . . . . . . . 76 A. Least Squares Analysis, Computer Program . . . . . . . . . . . . . . . 76 B. Nmr Line Shape Analysis, Computer Program . . . . . . . . . . . . . . . 79 LIST OF TABLES TABLE Page I. Conductivity and molecular weight data for (h5—C5H5)M(dik)2Cl complexes . . . . . . . . 22 II. Possible stereoisomers for (h5-C5H5)Zr- (acac)2Cl based on a dodecahedron and a square antiprism . . . . . . . . . . . . . . 35 III. Observed line shape parameters for the —CH- proton resonance components of (h5-C5H5)Zr- (acac)2Cl . . . . . . . . . . . . . . . . . . 52 IV. Kinetic data for the interchange of -CH= group environments in (h5-C5H5)Zr(acac)2Cl . 53 V. Infrared vibrational frequencies for some (h5-C5H5)M(§cac)2Cl complexes in the region 1600-600 cm 1 . . . . . . . . . . . . . . . . 58 VI. Infrared vibrational frequencies for some (h5vc5H5)Mjacac)2Cl complexes in the region 600—150 cm . . . . . . . . . . . . . . . . 60 vi Figure 10. 11. 12. 13. LIST OF FIGURES Page Proton nmr spectrum of (h5—C5H5)Zr(acac)2Cl in benzene solution at 360 . . . . . . . . . 26 Possible cis and trans configurations for (h5-C5H5)Zr(acac)2Cl based on a simple octahedral model . . . . . . . . . . . . . . 29 The D2 dodecahedron and the D4d square anti prism2 . . . . . . . . . . . . . . . . . . . . 31 Two views of the C bicapped prism . . . . . 33 2h Methyl proton resonance lines of (h5-C8H5)Zr(bzac)2Cl in benzene solution at 36 . . . . . . . . . . . . . . . . . . . 38 Geometric isomers of (h5-C5H5)Zr(bzac)2Cl based on aD2d dodecahedron . . . . . . . . . 40 Temperature dependence of the relative chemical shifts for the methyl protons of (h5-C5H5)Zr(acac)2cl in chlorobenzene . . . . 43 Temperature dependence of the -CH= and CH3 proton resonance lines for (h5-C5H5)Zr(acac)2Cl in benzene . . . . . . . . . . . . . . . . . 45 Temperature dependence of the frequency separ— ation between the —CH- resonance components for (h5-C5H5)Zr(acac)2Cl in benzene . . . . . . . 48 Log k vs 1/T plot for exchange of nonequiva- lent -CH= group environments in (h5-C5H5)Zr- (acac)2Cl in benzene solution . . . . . . . . 55 Infrared spectrum of (h5-C5H5)Zr(acac)2Cl in the region 350- 155 cm . . . . . . . . . . 67 Infrared spectrum of (h5-C5H5)Zr(acac-d7)2Cl in the region 350-155 cm"1 . . . . . . . . . 69 Infrared spectrum of (h5—C5H5)Hf(acac)2Cl in the region 350-155 cm 1 . . . . . . . . . . . 71 vii I. INTRODUCTION Coordination compounds containing both cyclopentadi- enyl and B-diketonate ligands are known for only a few transition metal elements. The chromium compound bromo- (acetylacetonato)pentahaptocyclopentadienylchromium, (h5-C5H5)Cr(acac)Br,* was the first known compound of this class. The compound was synthesized by Thomas1 in 1956 by reaction of Cr(acac)3 and C5H5MgBr in benzene solution. Thomas suggested that a class of compounds of this type should exist. The discovery of a mixed B-diketonato- cyclopentadienyl complex of zirconium by Freidlina, Brainina, and Nesmeyanov2 followed in 1961. Recently, mixed complexes of titanium and vanadium have been reported by Doyle and Tobiasat4. The titanium and vanadium complexes are cations of the general type (h5-C5H5)2M(dik)+, where dik can be the conju— gate base of various B—diketones (233' acetylacetone, benzoyl— acetone, dibenzoylmethane, egg;)3t4. [The analogous cationic complexes with tropolonate and ethylacetoacetate5 and a *The nomenclature_used in this thesis to describe the con- nexity of the C5H5 moiety is after F. A. Cotton, J. Amer. ' Chem. Soc., 90, 6230(1968). Trivial names, however, will of- ten be used $3} convenience (33,, dichlorobis(pentahaptocyclo- pentadienyl)zirconium will be referred to as zirconcene di- chloride.) 1 2 neutral complex with squarate5, [C404]2-, have also been prepared. It is interesting to note that in the case of ethyl acetoacetate, coordination to vanadium is normal, whereas coordination to titanium occurs yi§_one keto oxygen and the "ether" oxygen. The complex cations can be pre- cipitated readily as salts with C104- by reaction of (h5-C5H5)3M(Clo4)2 with the conjugate base of the diketone in aqueous solutions. Salts with other large uninegative anions have been isolated from aqueous solution by reaction of (h5-C5H5)2MC12 and the diketonate in presence of the desired anion4I5. Zirconium compounds containing B-diketonate and cyclo- pentadienyl ligands are of special interest in the present work. The first reported compounds, Chlorobis(acetyl- acetonato)cyclopentadienylzirconium, (h5-C5H5)Zr(acac)2Cl, and the benzoylacetonate analogue, (h5-C5H5)Zr(bzac)2Cl, were prepared by reaction of (hEPC5H5)2ZrC12 and the dike- tonez. Other synthetic routes to (h5-C5H5)Zr(acac)2Cl included (a) reaction of Zr(acac)2Clz and C5H5Na in tetra- hydrofuranz, (b) metathesis reaction between (h5-C5H5)2ZrC12 and Zr(acac)46 in toluene and (c) reaction of [(h5-C5H5)2ZrCl]20 or (h5-C5H5)2Zr(OC2H5)Cl with acetyl- acetone7. »It was also reported6 that (h5-C5H5)Zr(acac)2Br could be obtained from metathesis reaction of (h5-C5H5)2ZrBr2 and Zr(acac)4, although a product of suitable purity could not be isolated from the reaction when it was repeated in this laboratory. Brainina, Freidlina and Nesmeyanov 3 suggested that the structure of (h5-C5H5)Zr(acac)2Cl may be based on an octahedron with the chloride atom positioned gig or trans to theIP-bonded cyclopentadienyl ring. However, evidence in support of either structure was lacking. A second type of mixed ligand zirconium complex re- ported by the Russian chemists is a zirconoxane of the type [(h5-C5H5)Zr(acac)2]20. The compound was prepared by hydrolysis of (h5—C5H5)Zr(acac)2Cl in presence of an alcohol or an amines. Zirconium compounds of the type (h5-C5H5)Zr(dik)3 (dik = benzoylacetonate or dibenzoylmethanate), wherein zirconium achieves an inert gas configuration, are also known9. A similar compound has been prepared‘with 8-hydroxyquinoline9. The general method of synthesis involved the interaction of Zr(C5H5)4 and the free ligand in benzene. The method may have some practical limitations, because the authors report that reaction of Zr(C5H5)4 and acetylacetone yielded a viscous oil which, according to analysis, was a binary compound of (h5-C5H5)Zr(acac)3 with benzene. The present research deals primarily with a characteri— zation of (h5—C5H5)Zr(acac)2Cl in solution. The main ob— jectives were (a) to deduce the stereochemistry of the molecule in solution by nuclear magnetic resonance spectros- .copy, (b) to define quantitatively the stereochemical labil— ity of the molecule by nmr line broadening techniques, and (c) to assign the infrared vibrational spectrum of the mole- cule, especially in the low energy region where Zr-Cl, Zr-O, 4 and Zr—(h5-C5H5) stretching modes should be infrared active. Previous attempts to determine the stereochemistry and lability of six-, seven-, and eight-coordinate zirconium acetylacetonate complexes of the type Zr(acac)2C12, Zr(acac)3Cl, and Zr(acac)4, respectively, have been unsuc- cessfulllllz. At temperatures even as low as -130°, these latter molecules undergo rapid configurational rearrangement processes which average the expected nonequivalent methyl proton and -CH= proton environments. Recent proton and fluorine magnetic resonance studies also failed to provide stereochemical information for tetrakis(1,1,1,5,5,5-hexa— fluoroacetylacetonato)zirconium(IV)13. II. EXPERIMENTAL A. Preparation of Compounds 1. Reagents and Solvents Zirconocene dichloride was purchased from Arapahoe Chemicals, Incorporated,and used without further purifica— tion. Zirconocene dibromide and hafnocene dichloride were prepared by reaction of stiochiometric amounts of cyclopenta— dienylsodium with the appropriate metal tetrahalide in tetrahydrofruan under a dry nitrogen atmosphere. The gen- eral procedure has been described by Wilkinson and Birm- ingham14. The crude Zirconocene dibromide was purified by treatment with ijt "A" alkaline decolorizing carbon in chloroform solution. A Fisher burner was used to dry the carbon which was kept under a stream of dry nitrogen. Both compounds were recrystallized from chloroform carbon-tetra- chloride solution and dried at 80° under reduced pressure for two hours. The melting points of the compounds were checked against literature valueserlo. The cyclopentadienyl— sodium used in these syntheses was prepared according to the method of King and Stone15. 5 6 Anhydrous hafnium(IV) chloride was purchased in high purity (spectrographic grade) from Wah Chang Corporation. Anhydrous zirconium(IV) bromide was prepared by reac— tion of the elements according to the general procedure of Young and Fletcher16 as their method of preparation from the oxide works equally well when zirconium metal chunks are substituted for zirconium dioxide and carbon. Acetylacetone was purchased from Matheson, Coleman and Bell. The ligand was freshly distilled before use, and the absence of organic impurities was checked by nmr spectroscopy. Acetykmetone deuterated in all eight positions was ob— tained in the following manner. Anydrous sodium carbonate (4 g) was added to a mixture of 100 ml of freshly distilled acetylacetone and 250 ml of deuterium oxide (99.5+ %). The mixture was placed in a 500 ml round bottom flask, to which was attached a 15 in. vigreaux column and a distillation head. The mixture was refluxed for eight hours, and the product which distilled below 101° was collected. The distillate separated into two layers on cooling to room temperature. The less dense layer was mostly acetylacetone, and the more dense layer was a solution of heavy water and acetylacetone. The heavy water - acetylacetone layer was redistilled, and again the product distilling below 1010 was collected. This second distillate also separated into two layers on cooling to room temperature. The process of re- distilling the heavy water-acetylacetone layer was repeated until the distillate yielded insufficient heavy water-acetyl- acetone to carry on the operation. The acetylacetone layers 7 obtained from each distillation were combined, and the entire process was repeated three additional times, each time starting with a fresh supply of deuterium oxide and anhydrous sodium carbonate. The deuterated acetylacetone was distilled three times from anhydrous magnesium sulfate to ensure dryness. The product was better than 99% deg- terated, as judged by nmr spectroscopy. Trifluoroacetylacetone was purchased from Columbia Organic Chemicals Company, and was freshly distilled before use. Benzoylacetone was obtained from Eastman Organic Chemi- cals Company and was sublimed at 600 in vacuo before use. All organic solvents used in the synthesis and purifi- cation of products were reagent grade and were dried over suitable desiccants. Toluene, carbon tetrachloride, benzene, and hexane were refluxed over calcium hydride. When benzene was used as an nmr solvent, it was further distilled from lithium aluminum hydride. Chlorobenzene was distilled from phosphorus pentoxide and tetrahydrofuran was distilled from lithium aluminum hydride. Deuterochloroform was prepared according to a slight modification of the procedure described by Paulsen and Cooke17. Freshly distilled hexachloroacetone (151 ml, 1.00 mole) and 99.5+ % deuterium oxide (10 g, 1.0 mole) were placed in a 500 ml round bottom flask. The mixture was cooled in an ice bath, and 10 ml of pyridine was added dropwise with stirring. The mixture darkened immediately 8 and became quite warm. After the reaction had subsided, the solution was refluxed for two hours. Then the deutero— chloroform was distilled and dried over anhydrous calcium sulfate in a glass-stoppered erlenmeyer flask in the freezer section of a refrigerator. -It was not necessary to add ethanol—d1 as a stabilizing agent when the deutero- chloroform was stored in this manner. When used as a sol— vent for ir and nmr studies, the solvent was freshly dis— tilled from phosphorus pentoxide. The chloroform was more than 99% deuterated, as judged by nmr spectroscopy. Nitrogen used as a purging agent was "Hi-Pure" nitrogen purchased from the Liquid Carbonic Division of General Dynamics and is referred to herein as "dry nitrogen". 2. General Techniques Any preparation involving the reaction of hydroscopic compounds was conducted in a side-arm erlenmeyer flask or a three-necked flask. The reactions were carried out under a stream of dry nitrogen or in vacuo.- Readily hydrolyzable metal tetrahalides were transferred to the reaction vessel in a glove bag, and the weight of material delivered to the flask was determined by difference. A magnetic stirrer was used to stir the solution. All glassware was washed with detergent and water, rinsed with distilled water and acetone, dried at 180°, and cooled in a calcium sulfate desiccator whenever. possible. Ground glass joints were usually greased with silicone grease. 9 Filtrations were carried out with a specially designed glass frit Buchner funnel similar to that described by Holah13. Recrystallizations were performed in a glass stop- pered erlenmeyer flask fitted with a sideearm and stopcock. While solutions were being heated during recrystallization, dry nitrogen was passed through the side—arm and over the solution to prevent contact with air. Compounds were stored in screw—cap vials fitted with Teflon liners and were kept in a calcium sulfate desiccator. Hafnium and zirconium tetrahalides are especially hydrolyz- able; therefore, they were stored in screw-cap vials sealed with paraffin wax. 3. Chlorobis(z,4—pentanedionato)pentahaptocyclopenta- dienylzirconium A modification of the method of Freidlina, Brainina, andNesmeyanov2 was used to synthesize this compound. Zir— conocene dichloride (5.0 g, 0.017 mole) was placed in a 250 side-arm erlenmeyer flask to which was attached a dropping funnel containing 50 ml of freshly distilled acetylacetone. The flask was evacuated through the side-arm and heated at 80° in an oil bath for thirty minutes to ensure dryness. Then the acetylacetone was added to the erlenmeyer flask, and the solution was stirred at 80° for 15 min. During the heating period the connection to the vacuum pump was opened every two minutes to remove volatile products from the very slightly yellow, clear solution. The excess acetylacetone ml 10 was then removed under reduced pressure to leave a white powder. The total heating time was approximately twenty- five minutes. It may be noted that longer heating times gave a less pure, yellow product. Also, a yellow product was obtained if the volatile products were removed in a stream of dry nitrogen rather than under reduced pressure. The white powder was recrystallized under anhydrous condi— tions from benzene (6.3 g, 95% yield). Melting point is 189-190° d; lit.2 188-190° d. m. Calcd for C5H52r(C5H702)2Cl: c, 46.20; H, 4.91. Found: C, 46.01; H, 5.00. 4. Chlorobis(2,4-pentanedionato-d7)pentahaptocyclo- pentadienylzirconium Zirconocene dichloride (2.0 g, 0.0068 mole) and acetyl— acetone-d8 (50 ml, 0.5 mole) were reacted as was described for preparation of the protonated analogue, except that the reaction time was 55 min. The pale yellow crude product was recrystallized twice from benzene-hexane. The yield was 0.30 g (11%). Melting point is 188—190° d. 5. Chlorobis(2,4—pentanedionato)pentahaptocyclopenta- dienylhafnium This compound was prepared by reaction of hafnocene di— chloride (5.79 g, 0.0152 mole) and acetylacetone (50 ml, 0.50 mole) according to the method described for the zir- conium analogue, except that the reaction time was 30 min. 12 Anal. Calcd for C5H5Zr(C10H902)2Cl: C, 58.40; H, 4.51. Found: C, 58.47; H, 4.60. B. Attempted Preparation of Chlorobis(J,1,1-trifluoro-2,4- pentanedinato)pentahaptocyclopentadienylzirconium Zirconocene dichloride (1.0 g, 0.0034 mole) and tri- fluoroacetylacetone (20 ml) were reacted at 90° for 45 min. according to the general procedure described for preparation of (hs-C3H5)Zr(acac)2Cl. The excess trifluoroacetylacetone was removed under reduced pressure. The resulting brown, viscous oil yielded yellow crystals on further pumping at room temperature. The crystals were dissolved in a minimum amount of benzene. Addition of a small amount of hot hexane to the hot benzene solution and cooling produced a mixture of a dark yellow, flocculant precipitate and colorless crys- tals. The precipitate and crystals were filtered under anhydrous conditions and discarded. Addition of more hot hexane to the hot filtrate and cooling yielded clear, color- less crystals. The crystals were dried at room temperature in vacuo; the yield was 1.0 g (48%). The melting point was 91.0—93.6°. The compound gave a negative test for chloride ion with acidic silver nitrate solution. The reaction has been repeated by J. J. Howe and the product purified by vacu— um sublimation. Chemical analysis indicated the product to be (C5H5)Zr(C5H402F3)319. 13 C. Attempted Preparations of Bromobis(2,4-pentadionatol- pentahaptocyclopentadienylzirconium Brainina and Friedlina6 have reported that metathesis reaction of equimolar amounts of zirconocene dibromide and zirconium acetylacetonate in toluene yields (h5-C5H5)Zr(acac)2Br (yield 39%, melting point 203.5-205° d). Attempts by this author to duplicate their results were un- successful. Also, the reaction of zirconocene dibromide and acetylacetone, which is analogous to the reaction found successful for preparation of (h5-C5H5)Zr(acac)2Cl, did not yield the desired product. Both reactions are described here. 1. Reaction of Zirconocene Dibromide and Tetrakis— (gj4-pentanedionato)zirconium Zirconocene dibromide (1.14 9, 0.00299 mole) and Zr(acac)4 (1.46 9, 0.00299 mole) were placed in a 250 ml side-arm erlenmeyer flask to which was attached a dropping funnel containing 15 ml of toluene. The solid reactants were heated in vacuo at 100° C for 30 min. to ensure dryness. Then the toluene was added to the flask; the reactants dis- solved immediately to give a yellow—orange solution. After the solution had been heated at 1000 for 3 hours, the sol- vent was removed under reduced pressure. A brown, oily substance remained. The substance was dissolved in benzene,and the solution was filtered to remove a small amount of in— soluble brown material. Addition of hot hexane to the hot 14 filtrate and cooling gave brownish crystals. Recrystal- lization from benzene-hexane gave colorless crystals which melted at 190°. These crystals were regarded as being zirconium acetylacetonate; melting point was 194—196°. The reaction was repeated in tetrahydrofuran solution under a slight positive pressure of dry nitrogen. ~After a 2 hour reaction time at 50°, the solvent was removed under reduced pressure. A fine, deep yellow solid was ob- tained. This yellow solid turned green after it had aged at room temperaturefbr 0.5 hour in the evacuated reaction flask. After aging 2 hours the product was brown, and after aging 24 hours it was a brown oil. 2. Reaction of Zirconocene Dibromide and Acetylacetone Zirconocene dibromide (2.5 g, 0.0066 mole) and acetyl- acetone (25 ml, 0.25 mole) were reacted at 80° according to the procedure described for preparation of (h5-C5H5)Zr— (acac)2Cl. After reaction times of 5 min. and 30 min., white crystals, which had a melting point near the melting point of (h5-C5H5)ZZrBr2, were recovered from the reaction mixture. After a reaction time of 1 hour a brown powder was obtained. This brown powder was insoluble in benzene, hexane, acetone, and methylene chloride and did not melt below 300°. The reaction was repeated in carbon tetrachloride solu- tion. Zirconocene dibromide (1.0 g, 0.0026 mole), acetyl- acetone (5.4 ml, 0.0052 mole), and 50 ml of CCl4 which had 15 been dried over calcium hydride were placed in a 125-ml, side—arm erlenmeyer flask to which was attached a condenser and a P205 drying tube. The apparatus was swept with a slow stream of dry nitrogen while the solution was heated to just below reflux temperature for twelve hours. ~No reaction was observed as all the zirconocene dibromide was recovered unchanged from the reaction solution. D. Physical Measurements 1. «Molecular Weight Determinations The freezing point depression method was used to de— termine molecular weights. Benzene was chosen as the sol— vent, and its molal freezing point depression constant was found to be 5.410° per molal solution when benzil was used as the calibrating solute of known molecular weight. The benzene was dried over lithium aluminum hydride before use. Freezing point depressions were measured for solutions in the concentration range 0.0062—0.0131 molal with a Beckman differential thermometer graduated at intervals of 0.01°. Temperature readings, however, were estimated to i 0.001° with the aid of a magnifying lens. 2. Conductance Measurements The conductivity bridge used in this study has been described previously2°. It was operated at a frequency of 1000Hz. The conductivity cell was a Freas—type solution 16 cell with bright platinum electrodes. The temperature of the cell was maintained at 25.00 i 0.02° by a Sargent 5-84805 thermostatic bath assembly filled with light mineral oil. The cell constant at 25.00° was found to be 0.214 cm-l. An aqueous solution of potassium chloride was used as the cali— brating solution of known specific conductance21. The nitrobenzene used for the conductance measurements was washed successively with 1:1 H2SO4, water, and 1M_NaOH. Washing with 1MLNaOH was repeated until the wash solution was no longer colored. The nitrobenzene was then washed with water, allowed to stand over molecular sieves for one day, and finally vacuum distilled from a fresh supply of molecular sieves. 3. Melting Points Determinations A Thomas-Hoover 6406-H capillary melting point appara— tus containing G. E. Silicone Oil SF96(40) was used to determine melting points. Thermometers graduated in incre- ments of 0.2° were used, and they were calibrated using an appropriate standard such as vanillin (melting point 81-830), sulfapyridine (melting point 190—193°), or caffeine (melting point 235—237°). 4. Infrared Spectra Infrared spectra in the region 400—625 cm-1 were recorded with a Perkin-Elmer‘Model 237B grating infrared spectro— photometer. Polystyrene film was used to calibrate the 17 instrument. The accuracy of the reported frequencies in this range is estimated to be i 5 cm_1. Solution Spectra were obtained in 0.1 mm matched KBr cells with amalgam seals. The KBr pellet method was used to obtain spectra of solids. Spectra in the range 700-150 cm_1 were recorded on a Perkin-Elmer Model 301 grating spectrophotometer, which was calibrated with the pure rotational bands of water vapor. The estimated accuracy in this region is i 2 cm_1. The instrument was purged with dry nitrogen for recording spectra in the 310-150 cm—1 region. Solution spectra were obtained in 0.4 mm and 1.0 mm polyethylene molded cells; the Nujol mull technique was used to obtain solid spectra. Nujol mulls of each compound were carefully examined in the region 3800-3000 cm_1 to verify the absence of water and hydroxyl groups. All solutions were prepared in a glove bag purged with dry nitrogen. Deuterochloroform and benzene were dried as described in section II.A.1. 5. Nuclear Magnetic Resonance Spectra Variable temperature proton nuclear magnetic resonance spectra were obtained with a Varian A—60 high resolution spectrometer operated at 60 MHz. The probe temperature was controlled to within i 0.5° by a Varian temperature control— ler, Model V-6040. Temperatures were determined by measur- ing the chemical shift differences for methanol (low temper- atures) and ethylene glycol (elevated temperatures). The 18 audiofrequency side—band technique or a standard sample containing seven non—interacting compounds with chemical shifts uniformly distributed over a magnetic sweep width of 500 Hz was used to calibrate the magnetic sweep width. In the case of (h5-C5H5)Zr(acac)2Cl, relative signal intensities were determined by planimetric integration. The temperature dependence of the nmr line shapes were determined in benzene solution at a concentration of 15 g/100 ml of solvent. Small radiofrequency fields were employed to avoid saturation effects. Because of the small radiofrequency fields employed, the -CH= proton spectra had a low signal to noise ratio, especially at temperatures above 65°. It was necessary to resort to low filter band width settings and high spectrum amplitude settings. Although (h5-C5H5)Zr(acac)2Cl and (h5-C5H5)Zr(bzac)2Cl were found to be insensitive to atmospheric moisture when in the solid state, these compounds were extremely sensi— tive to moisture in solution. Therefore, prolonged exposure of the crystals to the atmosphere was avoided. ~Preparation of solutions was done in a dry-nitrogen-filled glove bag. The compound was weighed into a small glass vial, and the solvent added with a one milliliter syringe. The vial was heated gently to effect dissolution, and the solution was transferred to an nmr tube with the syringe. The nmr tube was stoppered and removed from the glove bag. Then the solu- tion was frozen in liquid nitrogen, and the tube was sealed with a flame. The benzene, chlorobenzene, and 19 deuterochloroform used to prepare the solutions were dried as described in section II.A.1. III. RESULTS AND DISCUSSION A. Preparative Chemistry Chlorobis(acetylacetonate)pentahaptocyclopentadienyl- zirconium,(h5—C5H5)Zr(acac)2Cl, and the benzoylacetonate derivative, (h5-C5H5)Zr(bzac)2Cl, are readily obtained by reaction of zirconocene dichloride and the free diketone as originally reported by Freidlina, Brainina, and Nesmey- anovz. An analogous reaction between (h5-C5H5)2HfC12 and acetylactone gives (h5—C5H5)Hf(acac)2C1. The volatile products of these reactions are presumably HCl and cyclo- pentadiene, although no attempt has yet been made to con— firm their identity. Zirconocene dibromide is considerably less reactive than zirconocene dichloride toward acetylacetone. For ex— ample, reaction of the dichloride and neat acetylacetone affords (h5-C5H5)Zr(acac)2Cl in essentially quantitative yield within 15 min at 80°. Under similar conditions re- action with the dibromide is incomplete after 30 min. 'Furthermore, the product obtained from the dibromide re— action is an intractable brown powder and not the desired (h5-C5H5)Zr(acac)2Br. Also, no reaction was observed on heating a 1:2 molar mixture of (h5—C5H5)ZrBr2 and acetyl- acetone in carbmitxfirachloride solution for 12 hours. 20 21 Brainina and Freidlina6 have reported that metathesis reaction of equimolar amounts of zirconocene dibromide and zirconium acetylacetonate in toluene yields crystalline (h5—C5H5)Zr(acac)2Br. When the reaction was repeated by this investigator, a brown oily substance was obtained as a crude product. The only crystalline compound that could be isolated from the crude product was unreacted zirconium acetylacetonate. A yellow solid was obtained when the re— action was conducted in tetrahydrofuran. However, this yellow solid was apparently thermally or photochemically unstable as it decomposed to a brown oil within 25 hours when stored in vacuo at room temperature. ,Attempts to prepare chlorobis(trifluoroacetylacetonato)- pentahaptocyclopentadienylzirconium, (h5-C5H5)Zr(tfac)2Cl, by reaction of zirconocene dichloride and trifluoroacetyl— acetone under conditions found successful for preparation of the acetylacetonate derivative did not yield the desired compound. The reaction afforded instead crystalline (h5—C5H5)- Zr(tfac)3. The identity of the compound was established by J. J. Howe.19 The only compound of this type reported pre- viously is a binary compound of (h5-C5H5)Zr(acac)3 with benzene9; however, this binary compound is a viscous oil. B. Characterization of (h5-C5H5)Zr(acac)2Cl, (h5-C5H5)Hf— jacac)2Cl, and (h5-C5H5)Zr(bzac)2Cl Conductivity and molecular weight data for (hs’C5H5 )Zr (acaC) 2C]. I (115 “C5H5 )Hf (aCaC) 2C]. , and (h5—C5H5)Zr(bzac)2Cl are collected in Table I. ‘The com— pounds are very weak electrolytes in nitrobenzene solution. 22 Table I. Conductivity and molecular weight data for (h5-C5H5)M(dik)2cl complexes. Conductivityé- Molecular Weight2 Molarity , ohm.1 {Molarity x 103 cm2 mole-1| x 103 Found Calcd (h5-C5H5)Zr(acac)2Cl 7.3 <0.014 : 22.1 407 390 (h5-C5H5)Hf(acac)2Cl 10.0 <0.014 : 6.19 517 514 (h5-C5H5)Zr(bzac)2Cl 10.0 <0.014 1 33.1 454 477 I gin nitrobenzene at 25°. EIn benzene. 23 A typical 1:1 electrolyte, such as [Ti(acac)3][SbC15], would exhibit a molar conductivity of SE: 25'ohm-J‘cmi'2mole-1 at a concentration of 1.0 x 102 M,12 In benzene solution the compounds are monomeric. Infrared spectra indicate that all four carbonyl groups are coordinated to the central metal in (h5-C5H5)Zr(acac)2Cl and (h5-C5H5)Hf(acac)2Cl. Both com- pounds exhibit a bond due to a symmetric carbonyl stretching vibration near 1595 cm-1. No bonds were found in the region 1626-1695 cm.1 where ketonic carbonyl modes have been re- ported for silicon (IV)22 and platinum (II)23.25.26 acetyl- acetonates containing uncoordinated carbonyl groups. Infra- red data also indicate h5-bonding of the C5H5 ring to the central metal. Huggins and Kaesz27 note that an h5-bonded C5H5 group can be differentiated from a h1-bonded ring by its simpler infrared spectrum. The most noteworthy differ- ences are the multiple C-H stretching frequencies and the presence of the free double bond absorptions near 1600 cm-1 in the spectrum of the h1-bonded C5H5 group. 1 (h5-C5H5)Zr(acac)2Cl and the hafnium analogue show only one C-H stretching vibration and no vibrations due to a free double bond in the C5H5 group. More cogent evidence for an hl-bonded C5H5 ring has been obtained by Stezowski and Eickzs. Their single crystal X-ray diffraction analysis shows that the C5H5 group is centrosymmetrically bonded to the zirconium atom. Since the infrared spectrum of (h5-C5H5)Zr(acac)2Cl in solution is essentially identical to the spectrum obtained 24 for the solid state, the C5H5 ring must be h5—bonded to zirconium in solution also. Infrared frequency assign- ments will be assigned and discussed in detail later. The proton nmr spectrum of (h5-C5H5)Zr(acac)2Cl in benzene solution is shown in Figure 1. The resonance line at 1 3.50 is assigned to protons on the cyclopentadienyl ring. The remaining six lines are due to -CH= and methyl protons on the acetylacetonate ligands. The two -CH= lines at T 4.75 and T 4.82 have relative intensities 1:1.03 i 0.03. Four methyl lines occur in the region T 8.31 to T 8.45. Relative to the methyl proton line at T 8.31, the two lines of equal intensity at T 8.37 and T 8.38 have a combined intensity of 2.03 i 0.05, and the line at T 8.45 has an intensity of 0.99 i 0.03. The intensities were de- termined by planimetric integration of five spectra; errors are estimated at the 95% confidence level. The nmr data connot be interpreted in terms of an equi- librium mixture of compounds of different stoichiometries arising from disproportionation of the (h5—C5H5)Zr(acac)2Cl complex. Equal molar mixtures of (h5-C5H5)Zr(acac)2Cl and Zr(acac)2C12, Zr(acac)3Cl, Zr(acac)4, Hacac, or (h5-C5H5)ZrClz gave spectra characteristic of the individual compounds initially mixed19. That is, the relative intensi- ties of resonance lines found in the spectrum of a solution containing only (h5—C5H5)Zr(acac)2Cl are not altered by varying the ligand composition. Therefore, it must be con- cluded that the observed methyl and —CH- proton resonances 25 .ucm>aom mo HE oofi\m m.b ma coflumuuamocou “Aux: omv 6mm pm COHuSHom mcwncmn CH HUNAUmumVHNAmanImnv mo Eduuommm MEG cououm .H wusmflm 27 result from nonequivalent environments for these groups in the (h5-C5H5)Zr(acac)2Cl molecule. The existence of a single, sharp cyclopentadienyl resonance is attributed to rapid rotation of the ring about the metal-ring axis. The simplest configuration which places each of the methyl and -CH= groups in nonequivalent environments is based on an octahedron with the C5H5 ring at one stereo- chemical position and the chlorine atom at a position gig_ to the ring(Figure 2). It may be noted that a trans con— figuration (Figure 2), which would have apparent C2V sym- metry in the presence of rapid rotation of the C5H5 ring, cannot be present in an appreciable amount, as judged from the relative intensity data. Configurations somewhat similar to the gi§_configuration may be derived from a D2d dodecahedron and a D4 square antiprism (Figure 3). In d these higher polyhedra it is assumed that the ring occupies a triangular face in forming three bonds to zirconium. For example, in the case of dodecahedron, the C5H5 ring may occupy an AAB face with the two ligands spanning a g-g pair of edges and the chlorine atom at a B position. An analo— gous anti—prismatic configuration may be readily visualized with the acetylacetonates spanning an s-s pair of edges. It is of interest to note that a configuration identical to that described in terms of the simplest octahedral model is possible based on aInCapped prism (3:3, Figure 4). The ad- vantage of thelficapped prism over the octahedral model is that the three bonds between zirconium and the C5H5 ring are more readily visualized.‘ 28 .Hmooe Hmucmzmuoo mameflm m :0 comma HONAomomVHNAmmmUImLV How macaumusmflmcoo mommy can mHo wanflmmom .N musmflm .N madmam 30 Z .Amomfiv mmm .N ..Emco rmwocH .aouuo>aflm .> .b can cumom..q .b Eouw ma AQV pom Amv How soapmgo: xouHm> paw omen ow on Q .AQV Emflumfluam mundqm Q wsu Cam .Amv Gonpmnmomvop one .m musmflm .31 .m musmfim 32 .wa 00mm Hmasmcmfluu one mmflmsooo mean ammo may ma Um>0a£om ma AN onswflm ..MMQ coupmzmuoo mHmEHm m mo mEHmu CH Umnfluommo umnu ou HMUHuchH coflumusmflmcoo 4 .Emflum UwQQMUHQ gnu msu mo m3wa> 039 .v musmflm _33 \’ \ \ z \ I \ \ ’2 \ z ’\ \\/ x x. z / \ I s I’ \ /\ .v musmam I . 34 Stezowski and Eick (28), in their single crystal thay diffraction structure determination, have shown that the molecule is more nearly dodecahedral than octahedral or anti- prismatic in the solid state. In view of this result one might have hoped to observe more than one stereoisomer in solution because at least five potentially detectable stereo- isomers are possible for the dodecahedral model. Seven additional isomers are possible if the C5H5 ring is cap- able of bonding through an ABB triangular face. Four stereoisomers are possible for the antiprism model. Table II ' lists the stereoisomers possible for dodecahedral and anti- prismatic configurations. However, no isomerization could be detected even after the compound was heated in benzene solution for 24 hours at 80°. The proton nmr spectrum of (h5-C5H5)Hf(acac)2Cl in benzene contains the same number of C5H5, -CH= , and CH3 resonance lines as the zirconium analogue. Presumably the two compounds have the same molecular structure in solution. The proton nmr spectrum of the benzoylacetonate deriva- tive, (h5—C5H5)Zr(bzac)2Cl, however, is considerably more complex than that described for the acetylacetonate deriva- tive. .In deuterochloroform solution at room temperature the compound exhibits a complex set of phenyl proton lines in the region 11.9 to T 2.8, two C5H5 lines near T 3.4, two —CH= lines near 1 3.7, and at least eight CH3 lines in the region I 7.7 to T 8.1. The methyl proton lines are better resolved in benzene solution and are shown in 35 Table II. Possible stereoisomers for (h5-C5H5)Zr(acac)2Cl based on a dodechahedron and a square antiprism Ligand Positions ”$22,223:“ .1 Id AABa g,bb AC IId AAB g .g B 111d AAB 9'm B IVd AAB b'a B Vd AAB g,b A VId ABB m.a B VIId ABB 9,3 B VIIId ABB b'a A de ABB m.9 A Xd ABB 9.9 A XId ABB m.m A XIId ABB 9,9 A Antiprismatic Isomers Ia — S , S - " 11a - . ' 111a - ' ’ Iva ' ' ' a . Face notation. bEdge Notation. cVertex notation. See Figure 3 for definition of edge and vertex notations. 36 Figure 5. The existence of eight methyl lines is consistent with a single stereochemical coordination geometry or stereo- isomer, such as the dodecahedral stereoisomer IId in Table II. Because of the asymmetry of the diketonate ligand, a total of four geometric isomers are possible when the chlorine atom is positioned gig to the C5H5 ring and the benzoyl- acetonate ligands span a set of dodecahedral edges. The four geometric isomers are illustrated in Figure 6. .Each isomer is expected to give rise to two methyl proton lines, two -CH= lines, and a single C5H5 line in presence of rapid rotation of the C5H5 ring about the metal—ring axis. The existence of only two -CH= and two C5H5 lines in deutero— chloroform indicates that the chemical shifts of these pro— tons are less sensitive to differences in chemical environ- ment than the terminal methyl protons. In benzene solution one C5H5 line and four rather broad -CH= lines are observed. The chemical shift difference for the two non—equivalent protons in (h5-C5H5)Zr(acac)zcl is also dependent on the nature of the solvent. At a concentration of 7.5 g/100 ml of solvent, for example, the -CH= lines are separated by 22: 0.07 ppm and 33. 0.01 ppm in benzene and deuterochloro- form, respectively, whereas in chlorobenzene solution a . single, sharp line is observed over the temperature range -20° to 36°. Another interesting feature of the spectrum of (h5-C5H5)Zr(acac)2Cl is the temperature dependence of the chemical shifts for the methyl protons in benzene and chlorobenzene below 60°. Near room temperature only three 37 .ucm>HOm mo HE OOH\m n.» ma coaumuucmocou “Aumz oov com um :oHusHOm econcmn GA HUNAUMNQVHNAmmmUImsv mo mmcfla mocmcommu cououm Hmnuwz .m ousmflm .m musmam «I 0. 53$ ? Hg? .1} fix; 3?? '38 'Figure 6. 39 Geometric isomers of (h5-C5H5)Zr(bzac)2Cl based on a Dad dodecahedron in which the chlorine atom is positioned gig to the C5H5 ring and the diketone ligands span the same pair of edges. Only the four edges which define the equatorial plane in a hard sphere model of the dodecahedron are shown. .40 41 of the expected four methyl lines are observed, because of temperature dependent solvation effects which result in identical chemical shifts for two of the resonance lines. The temperature dependence of the relative chemical shifts of the methyl protons in chlorobenzene is shown in Figure 7. Above 60° the methyl lines broaden and then coalesce into a single broad line at 93, 80°. A similar line broadening phenomenon is observed for the —CH= resonance lines above 60°. The line broadening is due to a rapid configurational rearrangement process which averages the nonequivalent methyl or -CH= proton environments. The determination of the kin— etics of the exchange process from the line broadening be— havior is described below. C. .Kinetics of Configurational Rearrangement for (hs-C5H5 )Zr(acaC)2C1 . As described above, (h5-C5H5)Zr(acac)2Cl is suffici- ently stereochemically rigid at room temperature to permit observation of non—equivalent -CH= and CH3 proton environ- ments by nmr spectroscopy. As the temperature is increased above 22- 60° in benzene, however, the set of two -CH= lines and the set of four CH3 lines each collapse into a single broad line, which then sharpens as the temperature is increased above the coalescence temperature of 22: 80°. The temperature dependence for both types of proton reso- nance lines is illustrated in Figure 8. The line broaden- ing is attributed to a rapid configurational rearrangement 42 .Uamfiw umw3oa um mummmmm LUHLB mafia Hmcume mnu ou 0>Humawn @muuon mum mumanm one .ucm>aom mo HE ooH\m m.h ma coflumuucmosoo umcmncmnoungo CH HUqumomVHNflnmnolmnv mo mGODOHm Hmnume mzu How upmanm HMUHEOSU o>HumHmH 03p mo mocmwcmmmp musumnmmfiwa .h musmflm .43 .b musmam g . , 0. £25 . Om _ 00 03. .ON _ _ . _ m . J IN If X 1 _ _ ..l.l.._ _ i. L... zH Figure 8. 44 Temperature dependence of the —CH= and CH3 proton resonance lines for (h5—C5H5)Zr(acac)2Cl in benzene. Concentration is 15 g/100 ml of solvent. The spectrum amplitude used to record the spectra was larger for the -CH= lines. , M fimi CH3 , «LA,M l. t_ 46 process which averages the nonequivalent environments for the —CH= and CH3 groups on the acetylacetonate ligands. The exchange of -CH= group environments represents an exchange of nuclei between two nonequivalent sites of equal population. For processes of this type, Gutowsky and Holm29 have derived the dependence of the nmr line shape on (1) 6v, the frequency of separation between the resonance com- ponents assuming no exchange and no overlap of the compo- nents, (2) T2, the transverse relaxation time, and (3) the quantity T, which is given by TATE/(TA + TB), where T and A T are the mean lifetimes of protons at each site. When B the two sites are equally populated, T = T = 21. A B Normally, a limiting value of the experimentally ob— served frequency separation, ave, is reached as the tempera— ture is lowered, and this limiting value of éve is taken as 5v. As can be seen in Figure 9, five for the -CH= pro- tons of (h5-C5H5)Zr(acac)2Cl in benzene is temperature dependent over the entire temperature range investigated. The observed dependence of ave below 55° cannot be attributed to exchange effects. Within experimental error, the widths of the resonance components remain constant over the tem- perature range from 31-550. Above 55° the resonance com- ponents broaden markedly as the rate of —CH- group ex- change increases. Therefore, the exchange is slow below 55°, and the observed temperature dependence of éve must ibe due mainly to temperature dependent solvent effects \Mhich cause the chemical shift difference between the two 47 .mcmNch mo HE o0H\m ma ma Coaumuucmo Icou .wcmucmn Ca HUqumomqufimmmUImsv mo meHH mocmcommu cououm "mo: may cmmBqu coaumummmm mocwskum may mo mocwocwmmw musumummfime .m musmflm O. .a muamam 0o.m2wk .om om 0¢ 48 A _ q _ _ _ r0 Nonlv°éwas wad zH 49 resonance components to decrease with increasing temperature. Thus 6v will take on a different value for every temperature. The appropriate values of 5v in region of fast exchange were found by extrapolating the linear portion of the five 'gg T curve to the region of fast exchange (gf,, dashed line in Figure 9). The extrapolated line was assumed to represent the temperature dependence of the frequency of separation that would be observed in the absence of ex- change. Since the ratio of the line width at half-maximum amplitude to the frequency of separation is less than 0.3, no correction for 6v due to the slight overlap of the two resonance components was necessary29. The resonance components for the -CH= groups in the region of slow exchange do not have equal widths; therefore TZA f T23. .In benzene solution at a concentration of 15 g/100 ml of solvent, the average values for the line width at half-maximum amplitude were 0.62 Hz and 0.55 Hz, respectively, for the low field and high field line in the region of slow exchange (36°). Corresponding values of the transverse relaxation times, Z/Zw x half width, are 0.513 sec and 0.579 sec, respectively. These values of T2 were assumed to be independent of temperature over the entire region of exchange. *It is interesting to note, however, that the line widths are apparently dependent on concen- tration. Values of T2 at a concentration of 7.5 g/100 ml of solvent were 0.624 sec for the low field -CH= line and 0.965 sec for the high field -CH= line at 36°. 50 With the appropriate values of 5v and T2, values of the mean lifetimes for the -CH= groups in (h5-C5H5)Zr— (acac)2Cl in benzene solution were obtained by comparison of the observed nmr line shapes in the region of exchange with the line shapes calculated from the full Gutpwsky— Holm equation. The complexity of the Gutowsky-Holm equa— tion necessitated use of the Control Data Corporation 3600 computer of the Michigan State University Computer Labora- tory. The program used is presented in the Appendix. .Six line shape parameters were used for comparing the observed nmr spectra with the computer—calculated spectra. Below the coalescence temperature the spectra were compared with regard to five, the frequency separation between the two C-H resonance lines and the quantities r and r', which are defined as the ratio of the maximum amplitude of the low and high field resonance lines, respectively, to the between-peak minimum amplitude at (VA + VB)/2. At coales— cence and above the line widths at one—fourth, one-half, and three—fourths maximum amplitude were compared. -At each temperature three copies of the spectrum were recorded and the shape parameters averaged. The calculated spectra con— sisted of 100 coordinate pairs spaced at intervals of 0.36 radians/sec about the mean frequency of the spectrum. ‘The reliability of the shape parameter method was checked by comparison of 24 coordinate pairs for spectra obtained at two temperatures below coalescence (81.20, 94.7°). The observed shape parameters for the —CH- resonance components 51 are given in Table III. Values of T for the interchange of -CH= group environ- ments are given in Table IV along with values of log k, where k, the first order rate constant, is given by 1/2T. -Within experimental error the coalescence temperature for the -CH= proton lines at a concentration of 7.5 g/100 ml of benzene was the same as that observed at a concentration of 15 g/100 ml (79.20). The mean lifetime at the coales— cence temperature for the -CH= protons at the former concen- tration was determined from a line shape analysis in which appropriate values of T2 (0.965 sec for the low field line, 0.628 sec for the high field line) and 5v (2.90 Hz) were used. The value of T for the 7.5% solution (0.118 sec) was in good agreement with the value of T obtained at a concentration of 15 g/100 ml of solvent (0.124). Therefore, the exchange is indeed a first order process. The Arrhenius activation energy, Ea’ the frequency factor, A, and the activation entropy, AS*, are also in— cluded in Table III. The activation energy and the fre- quency factor were determined from the least squares straight line of a log k Xfin 1/T plot (see Figure 10). The activa— tion entropy was calculated at 25° from the expression AS*=RlnA—R[1+ln£hl where R, T, k, and h have their usual meanings. 52 Table III. Observed line shape parameters for the -CH= proton resonance components of a (h5-C5H5)Zr(acac)2Cl __ b [fc Average Values ” _ I Temp. d e ,f g h h 1 °C éve 5v r r W1/g W1/2, W3/4 31.2 3.48 3.48 6.00 6.40 —-— ——— -—— 33.3 3.30 3.30 4.38 5.15 --- -—— ——— 45.3 3.15 3.15 4.71 5.25 ——- —-- ——— 50.3 3.07 3.07 3.22 3.88 —-— —-- -—— 54.3 ' 3.00 3.00 2.90 3.34 --- --- -—— 61.0 2.80 2.88 1.88 2.33 ——- -—— -—- 70.9 2.35 2.72 1.33 1.75 —-- —-— ——— 75.4 1.60 2.64 1.14 1.33 --— --— ——— 77.3 1.10 2.62 1.07 1.21 --- ——- _-_ 79.2 - 2.58 - — 4.80 3.67 2.45 81.1 ——- 2.55 --- —-- 4.48 3.05 1.96 86.7 --- 2.45 —-— --- 3.78 2.40 1.32 89.2 —-— 2.41 —-- --— 3.30 2.10 1.18 94.7 --- 2.32 —-- --- 2.60 1.50 0.88 103.5 —-— --— --- —-— 1.98 1.18 0.68 116.0 --- -—- --- --- 1.98 1.16 0.60 a15 g/ml of benzene. pAverage shape parameters obtained from three spectral measure— ments. cObserved frequency of separation between the two resonance components, in Hz. dFrequency of separation between the two resonance components in absence of exchange (Hz). 6Ratio of the low-field—peak maximum to the between—peak—min— imum. fRatio of the high-field—peak maximum to the between-peak minimum. gLine width at one-quarter maximum amplitude, Hz. hLine width at one—half maximum amplitude, Hz. lLine width at three-quarters maXimum amplitude, Hz. '53 Table IV. .Kinetic data for the interchange of nonequivalent -CH= group environments in (h5-C5H5)Zr(acac)2Cl. Temp, °C 1. sec log k 61.0 0.217 0.3625 70.9 0.176 0.4535 75.4 0.142 0.5467 77.3 0.130 0.5850 79.2 0.124 0.6055 81.1 0.101 0.6946 86.7 ' 0.0791 0.8008 89.2 0.0730 0.8356 _94.7 0.0550 0.9587 Ea = 10.4 t 0.8 kcal/mole log A = 7.126 i 0.497 AS* = —27.9 i 2.3 an k250= 0.293 Sec-1 aIn benzene; concentration is 15 g/100 ml solvent. 54 Figure 10. Log k gs, 1/T plot for the exchange of non— equivalent -CH= group environments in (h5-C5H5)Zr(acac)2Cl in benzene solution. 55 . . ' . J l . I . . . . - 2.7 I 2.8 32.9 3.0 l/TXIO Figure 10. 56 The activation energy of 10.4 kcal/mole falls in the range of values reported for first order configurational rearrangement processes of a variety of six coordinated metal B-diketonates30‘32. ‘The value obtained for the acti- vation entropy (-27.9 eu) is the most negative value among the reliable determinations of AS* for first order rear- rangement processes of metal fi—diketonates. .It should be noted, however, that the signal-to-noise ratio for the -CH= proton lines, on which the values of Ea and AS* are based, was limited by the solubility of the compound. .In order to obtain useable spectra, it was necessary to record the spectra at progressively lower filter band width levels as the temperature of the Sample was increased. It is well known that very low filter band width levels can lead to apparent broadening of resonance lines33. The lowest filter band width setting of 0.1 cps was used in the region above coalescence. Consequently, the values of log k above co- alescence would tend to be too low relative to the values obtained below coalescence. A systematic error of this type would cause the observed activation energy to be too low and the activation entropy to be too negative. Therefore, the reported values of Ba and AS* should be regarded as lower limits for these activation parameters. 57 D. Assignment of Infrared Vibrational Frequencies for (h5-_5H5)Zr(acac) Cl,*(h5-C5H5)Zr(acacld72gpl, and (h5 "C5H5 )Hfracac 2C1 The infrared spectrum of (h5’C5H5)Zr(acac)2Cl was re— corded in the region 4000-155 cm-l. Spectra of the acetyl- acetonate-d7 analogue and of (h5-C5H5)Hf(acac)2Cl were re- corded for comparison. The frequencies are listed in Tables V and VI. Except for a sharp weak band due to a C-H stretch- ing vibration of the C5H5 group near 3100 cm.1 and broad, weak bands in the region 3000-2900 cm-1 due to C-H stretch— ing modes of acetylactonate, no bands were observed above 1600 cm-1. The C-H stretching vibrations are not of inter- est in the present work and will not be discussed further. In general, bands observed in solid spectra were only slightly shifted from bands observed in solution spectra. 'However, several bands observed in solution were split in the solid; the splittings will be discussed later. A Nujol mull of each compound was carefully investigated in the 3500-3000 cm- region to verify the absence of water or hydroxyl groups in the samples. Bands appearing in the 1600-480 cm—1 region are ex- pected to be due mainly to vibrational modes localized in the acetylacetonate and cyclopentadienyl ligands. The vibrational modes of acetylacetonate assigned in Table V are based mainly on the normal coordinate calculations of Benke and Nakamoto26 for Pt(acac)Clz— and its deuterium analogues. The fundamental vibrations of the C5H5 group are assigned on thetbasis of the normal coordinate treatment for ferrocene by Lippincott and Nelson34 and on the extensive 58 mom mom x000 mm0 Aomomv m sue” a cum a mum m mmofi m mmofl m Anmn0umav 3> muofi 3> woofi 3> moofi 3> fieofi 3> mooH w . n 3> oHHH 3> NHHH . u 000 0:00 Anemonmnv 3> HmHH 3> fimHH 3> HmHH 3> HmHH 3> HmHH pawn mamamlcfl mlo Aomomv 3 wwHH I I 3 meH 3 meH 000 n5.0 + 3 memfi 3 mwmfi 000 saw 04440 Aomomv m 00m” 2 000H 5 mama m owmfi m mumfi 000 2mm «00 Aomomv m oomfi :0 owofi gm 000d 0 momfi m ommfi m Homfl Hum Emma 04440 Aomumv :0 «>02 mumfi 0 ohm” rm sumfi am mumfi 000 000 mm0 Aomomv e 000H 3> Hmofi 3> umofi a Hmwfi a umwfi 000 04440 Ammnonnnv .m.z gm omvfi 3 Hmwfi .m.z 0 .m.z . :0 0mmfi m 000H 0 000H gm mmmfi Hum swam 04440 Aumomv m 0mmH m mmmwz m ommfi m mmmfi m 0mmH m wmnfi m mmmfi m obmfi . m mwmfi m «00H m Homfl 0 000H 0 puma .000 2mm 04440 Aomomv m mmmfi m ummfi m uwmfi m 000H 0m ummfi 00Hom coausaom 0HHom coausaom \ moo: ucmcflfiovmum @HHOm Hoaflomoavmmflnmmoumnv Hoafls0u0000vnuflnmnounnv HoaAumomvHNAnmnonmgv mmmemEoo HUNAommeSAmmmUImnv meow How mmaocmsqmnm Hmcoflumnnfl> omumnmcH 0 Hugo 000-000H 000000 0:0 00 .> magma s 9 5 .uc0>aom an omusomno 0H03 scammn was» CH mocmm .nm “mum> .> “#003 .3 “Edflowfi .E “moonum .m H .00>H000H uo: ..m.z .Hmoasonm .Uonqu umaamm me 030 mn Uwcamuno 0H03 moflaom mo mnuommm .uc0>aom HE oon\m n ma coammnucmocoo "0:0NQ0Q ca Umcflmuno 0H03 £0a£3 IE0 com 3oamn mUGmQ may now ummoxm .EH0moHoanooumuswo CH Umcflmuno 0H03 muuommm QOHUSHomm . 0 0mm 2 H00 0 0mm 0 0mm 0 000 00H“ Aomomv a ham 0 com 0 a son 0 Hum 0:: + pawn 0GMHQIGH mno.0 00 000 00H“ Aomomv a 000 s 0H0 0 a mom a 000 pawn mamam -mouuso mu0 Aomomv a com a how 0 a can 0 m 0H0 m mom 0 cam 0:09 mamam -00-»:0 muo Ammu0umnv 0 Ham 0 Ham 0 0H0 m 0mm 0 4H0 pawn mamam .00.050 0-0 Anmm0nnav 3 «mm 2 «mm 0 5 H00 0 m Anmnuungv 3> 000 am «H0 0 3> 000 0 Hum nm0.0 Emma Aomomv a 0mm 5 000 a one a mam a 000 000 «m0T0 5mm Aomoav am >00 3 com 0 am 000 T 0:03 E maoH mamamucfl 0-0 Anmm0ungv m uHoH a Hmofi a Hmofi omoH m mmofi 0flaom 0HHom coflusaom 0flaom couusaom 0WD: HGMCHEOUGHQ H0mxomomvmmxmmmo-msv H0NA00-0momvHNAmmnounnv Honxomomvuaxumn0unnv Aomscflucouv .> magma 60 .00005 Honsz :0 00000090 0003 mnuummm oflaom .0G0>Hom HE ooH\m m.mm 00 coflumuucmocoo .mcmucmn CH omcflmuno 0H03 mnuommm Goausaomm 3 000 00 000 3 000 3 000 3.000 3 000 3 00m 0 000 0 000 0 000 0 000 0 000 0 000 00 000 00 000 00 000: 00 000 3 000 3 00m 3 o0me 00 000 00 000 .000 on: 0 000 0 000 0 000 0 000 0 000 0 000 .000 00-2 00 000 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 000 0 + 000 0-2 0 000 0 000 00 >00 0 000 0 000 00000 00000000 00000 00000 00000000 0002 ucmcflfioooum awnAbmomvmmAnmmolnnv HUNApplomomvuwfinmmolmnv HUunomomvHNfinmmolnav m HIEU mmfiloww GOHmwu may :0 mmxmamEoo avaAomomvanmmulnsv 0800 00m mwflocmswmum HMGOHumHQH> UmnmumcH .H> wanna 61 data compiled by Fritz35 for various cyclopentadienyl metal compounds. The spectrum of solid (h5—C5H5)Hf(acac)2Cl in the 1600-480 cm-1 region is virtually identical to the spectrum of the zirconium analogue (32,, Table V). Little mass de- pendence is consistent with assignment of the frequencies in this region to modes which involve primarily motion of the ligand atoms. The assignments for (h5-C5H5)Hf(acac)2Cl follow directly the assignments made for the zirconium analogue. Several features of the frequencies and assignments given in Table V for (h5-C5H5)Zr(acac)2Cl and the corre- sponding acetylacetonate-d7 derivative merit special comment. A considerable number of metal acetylacetonates exhibit a single, strong band in the region 1600-1550 cm.1.36 Based on Nakamoto's most recent calculations for a C2v one-ring modelA, the band in this region is due to a nearly pure symmetric carbonyl stretching vibration. -In deuterochloro- form solution (h5-C5H5)Zr(acac)2Cl exhibits two strong bands at 1597 and 1577 cm-1. Furthermore, the 1577.001.1 band 1 in the solid. split into two bands at 1586 and 1570 cm- More than one carbonyl vibration could arise from coupling of the carbonyl vibrations of the two different chelate rings. It is of interest to note that Shimarouchi and co— workers37 have assigned an extra band in the carbonyl region of Cu(acac)2 and Pd(acac)2 to a combination band between an infrared active c—H out—of-plane vibration and an infrared 62 inactive C-H out-of-plane vibration. A similar assignment is not possible for the 1577 cm"1 band in (h5—C5H5)Zr(acac)2Cl, because this band is shifted only slightly to lower energy in the acetylacetonate-d7 derivative. The band at 1524 cm.1 in the solution spectrum of (h5-C5H5)Zr(acac)2Cl is assigned to the asymmetric C:::C stretching mode of acetylacetonate coupled slightly with the C-H in-plane bending mode. In the deuterated compound these vibrations are decoupled, and the band splits into two bands at 1526 and 1496 cm-1. The origin of the split— ting, however, is not clear. Acetylacetonate is expected to give rise to three bands 1 due to degenerate in the region between 1500 and 1300 cm“ and symmetric methyl deformation modes and an asymmetric carbonyl stretching vibration. Since the rather broad band of medium intensity at 1427 cm-1 is absent in the acetyl— acetonate—d7 derivative, it is reasonable to assign this band to the degenerate methyl deformation. The shoulder that appears at 1450 cm_1 in the deuterated compound is as— signed to the C:::C stretching mode of C5H5. This shoulder 1 could not be resolved from the 1427 cm- band in the pro- tonated compound in solution or in the solid. The shoulder at 1373 cm_1 and the strong band at 1360 cm-1 in (h5—C5H5)Zr(acac)2Cl are assigned to the asymmetric carbonyl stretching and symmetric methyl deformation modes, respec- tively. These latter assignments are supported by the spec- trum of the deuterated compound. in which the 1360 cm-lband is 63 absent. -It should be noted that Doyle and Tobias have as- signed a strong band at 1321 cm"1 in [(h5-C5H5)2Ti(acac)]C1043 and a shoulder at 1360 cm-1 in [(h5-C5H5)2V(acac)]ClO44 to a C5H5 ring stretching vibration. A similar band may be super- imposed on the band at 1360 cm-1 in (h5-C5H5)Zr(acac)2Cl, because the 1376 cm.1 band observed in the solution spec- trum of the acetylacetonate-d7 derivative is resolved into two bands at 1373 and 1361 cm_1 in the solid spectrum. Based on the compilations of Fritz, however, a band near 1360 cm—1 would not be expected to represent a fundamental vibration of an h5—bonded C5H5 group. The next band at 1280 cm"1 is due to the C:::C sym— metric stretches of acetylacetonate coupled slightly with the C-CH3 stretching mode. As was observed by Benke and Nakamoto for Pt(acac)c12. , this band splits into two bands at 1304 cm.1 and 1242 cm.1 upon deuteration of acetylaceton- ate. The band at 1185 cm_1 is assigned to the C-H in—plane bend of acetylacetone. This latter band is expected to shift below 900 cm—1 upon deuteration of the -CH= group. However, no band in either the solution or solid spectrum of the acetylacetonate-d7 compound could be unambiguously assigned to the C-D in-plane bend. Perhaps the band cor- responding to this mode is superimposed on the bands in the C5H5 bands in the 834-800 cm-1 region. It is also possible that the band may be too weak and broad to observe at room temperature. For example, Benke and Nakamotoze, observed 64 the C—D in-plane bending mode in [Pt(acac—d1)C12]- and the corresponding acetylacetonate-d7 complex only at liquid nitrogen temperature. Two very weak bands near 1130 and 1070 cm—1 are pres- ent in both the protonated and deuterated acetylacetonate compounds. The 1130 cm.1 band is in the region where the antisymmetric ring breathing mode of C5H5 should be observed, and the band is assigned accordingly. However, we are un- 1 band to an infrared active able to assign the 1070 cm- fundamental vibration of the C5H5 group. The band at 1023 cm.1 in the solution spectrum of (h5-C5H5)Zr(acac)2cl consists of two overlapping bands, one due to a methyl rocking mode of acetylacetonate and the other due to an in-plane C-H bending vibration of the C5H5 group. The assignment is supported by the fact that the band is approximately one half as intense in the deuterated compound. The two overlapping bands in (h5-C5H5)Zr(acac)2Cl are apparently resolved in the solid. The C—D in-plane bending frequency of the C5H5 group, however, is split in the solid, which accounts for the qualitative similarities between the spectra of the solid samples. With the exception of the very weak band near 906 cm-1 band assignments in the region 1000-600 cm.1 are quite straight forward. It is interesting to note, however, that the two types of C-CH3 stretching modes of acetyl- acetonate are resolved partially in the solid sample but not in solution. Also, the lower energy C—H out-of-plane 65 bending mode of the C5H5 group appears as a single band in solution, but the band is split in the solid phase. The two bands at 557 cm.1 and 536 cm-1 in the spectrum of (h5-C5H5)Zr(acac)2Cl are sensitive to deuteration of acetylacetonate. These bands are tentatively assigned to out—of—plane deformation vibrations of the acac ring. Bands due to metal-oxygen, metal-halogen and metal- h5—C5H5 stretching modes are expected to occur in the region 480-155 cm-l. Bands in this region are given in Table VI. In solution, the band at 318 cm_1 in @F0C5H5)Zr(acac)2Cl which is assigned to both Zr-O and Zr—Cl stretching vibrations, and the corresponding band at 294 cm-1 in the hafnium analogue are further resolved in the solid state. Spectra in the region 350 cm-1 to 155 cm—1 are shown in Figures 11-13. Fay and Pinnavaia38 have assigned zirconium-oxygen stretching modes in both seven coordinate Zr(acac)3Cl and eight coordinate Zr(acac)4 to strong absorption in the regions 450-421 00" and 314-301 00“. Zirconium-chlorine vibration in Zr(acac)3Cl38 and ZrCl4-2Diarsine39 have been assigned to bands at 314 cm_1 and near 300 cm-1, re— spectively. Similar assignments have been made for the analogous hafnium compounds at somewhat lower absorption frequencies. »No assignments have yet been made for zir- conium- or hafnium-h5—C5H5 stretching vibrations. The assignments in Table VI are based on the above data. 66 .0c0>aom HE ooH\m w xmw mo coaumuucmocoo m 00 coausaom 0G0Nc0£ s0 .mcfla @0000 “005E Hohsz C0 00000 .0:00 wwsmmo .0180 mmH Iomm coflmmn 030 CH HUqumomVHNA0m0010£V Mo Ednuommm UwumumcH .fiH musmfim 67 .an 00500h . . 70.0 00200000... 000 000 00m l 0 . — a _ 0 — T 0 N. . . . N m m as o. ss ’— . ~ o . a _0 08051 I 0.05 .. .. , . . . . ” ~ .30. .4 .. a . _. . .... .. .. g o . a s 3 . s on . N x u 3 0 as. .. 3r .0: I 0.0 0.... .. x ‘0’ .K s \‘~.k _ _ _ _ _ _ _ (%) BONVlllWSNVHl O 00. 68 .0058 Honsz 0 :0 00000090 003 Ednuo0m0 0:0 “0:80 000 10mm 200000 000 :0 000Ae010000v0NA0m0010£v mo E50000m0 00umumcH .00 000000 .NH musmflm Tao xozm 8mm“. 0mm 0mm 00m . - — - — - O _om}-88§mzmo-s . m \ssfiao N a. s v . . .s oo N. .- . & s); .. f .- V s v ~ - N .. x. N .— S g . N. g m a. a. u a . ll.— . . . . .. i . .. . . V s . . ’ ~ . s... ..\a \(x .o .s /\ _oa W s‘ 00 \\\ ’I\ a’\\\ II 3 s r \I s|l\ _ _ _ _ _ _ _ oo. 7O .ucm>H0m mo HE ooH\m m 2mm mo coflumnucwocoo m um coausHom wawucmn SH .mcfla wwaom "Hans Hondz.afl vflaom .mcfla Umnmmnv .nIEU mma Iomm coammn 0:» :H HUqumumvmmAmmmUImsv mo Ednuummm UmnmumcH .mH Gunman 71 .mH ouumflm TEo $02M DOMWE _ emu. _ 0mm own \J— . “.4” omeoogimzmows . .. .. a s \II (%) BONVLLIWSNVBL _to O O. 72 Unfortunately, it is not possible to assign unambiguously the metal-h5C5H5 stretching mode in these compounds. Such assignments might be possible if the h5-C5D5 analogues Were also investigated. BIBLIOGRAPHY 0:014:60 10. 11. 12. 13. BIBLIOGRAPHY J. C. Thomas, Chem. Ind., 1388 (1956). R. Kh. Freidlina, E. M. Brainina, and A. N. Nesmeyanov, Dokl. Akad.-Nauk SSSR, 138, 1369 1961;; Proc. Acad. Sci. USSR, Chem. Sect., 138, 628 1961 G. Doyle and R. S. Tobias, Inorg. Chem., , 1111 (1967). 9. G. Doyle and R. S. Tobias, Inorg, Chem., Z” 2479 (1968). Z. G. Doyle and R. S. Tobias, Inorg. Chem., , 2484 (1968). E. M. Brainina and R. Kh. Freidlina, Izv. Akad. Nauk SSSR, Ser. Khim., g, 1421 (1964); J. Acad. Sci. USSR, Div. Chem. Sci., g” 1331 (1964). E. M. Brainina, R. Kh. Freidlina, and A. N. Nesmeyanov, Dokl. Akad. Nauk SSSR, 154, 1113 1964); Proc. Acad. Sci. USSR, Chem. Sect.,“134, 143 1964). E. M. Brainina, G. G. Dvoryansteva, R. Kh. Freidlina, Dokl. Akad. Nauk SSSR, 156, 1375 1964); Proc. Acad. Sci. USSR, Chem. Sect.,“136, 633 1964). E. M. Brainina, E. I. Mortikova, L. A. Petrashkevich, and R. Kh. Freidlina, Dokl. Akad. Nauk SSSR, 169, 335 (1966;; Proc. Acad. Sci. USSR, Chem. Sect., 1667'681 1966 M. Kh. Minacheva, E. M. Brainina, and R. Kh. Freidlina, Dokl. Akad. Nauk SSSR, 173, 581 (1966); Proc. Akad. Sci. US§R, Chem. Sect., 173, 282 (1966). T. J. Pinnavaia and R. C. Fay, Inorg. Chem., 5“ 233 (1966). T. J. Pinnavais and R. C. Fay, Inorg. Chem., Z” 502 (1968). S. C. Chattoraj, C. T. Lynch and K. S. Mazdiyasni, Inorg. Chem., 1, 2501 (1968). 73 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 30a. 74 G. Wilkinsen and J. M. Birmingham, J. Amer. Chem. Soc., ZQ, 4281 (1954). R. B. Kin and F. G. A. Stone, Inorg.;Synthesis, 7, 101 (1963 . "’ R. C. Young and H. G. Fletcher, Inorg. Synthesis, 1, 49 (1939). ~ P. J. Paulsen and W. D. Cooke, Anal. Chem., 35, 1560 (1963). "’" D. G. Holah, J. Chem. Educ., 42“ 561 (1965). J. J. Howe, Michigan State University, personal com- munication, 1968. H. B. Thompson and M. T. Rogers, Rev. Sci. Instr., 21, 1079 (1956). J. E. Lind, J. J. Zwolenik, R. M. Fuoss, J. Amer. Chem. Soc., él/ 1557 (1959). R. West, J. Amer. Chem. Soc., 82“ 3246 (1958). B. N. Figgis, J. Lewis, R. F. Long, R. Mason, R. S. Nyholm, R. J. Pauling, and G. B. Robertson, Nature, 195, 1278 (1962). R. A. D. Wentworth and C. H. Brubaker, Inorg. Chem., 2, 1472 (1964). J. Lewis, R. F. Long, and C. Oldham, J. Chem. Soc., 6740 (1965). G. T. Behnke and K. Nakamoto, Inorg, Chem., 6, 440 (1967). D. K. Huggins and H. D. Kaesz, Progr. Solid State Chem., 1“ 470 (1964). J. J. Stezowski and H. A. Eick, Michigan State University personal communication, 1968. H. S. Gutowsky and C. H. Holm, J. Chem. Phys,, 22/ 1228 (1956). ' n -R. C. Fay and T. S. Piper, Inorg. Chem., Ex 348 (1964). D. C. Bradley and C. E. Halloway, Proceedings of the 9th International Conference on Coordination Chemistry, St. Moritz-Bad, Switzerland, 1966, p. 483. 31. 32. 33. 34. 35. 36. 37. 38. 39. 75 R. C. Fay and R. N. Lowry, Inorg. Chem., 6” 1512 (1967). T. J. Pinnavaia, J. M. Sebeson, II, and D. A. Case, Inorg. Chem., in press. A. Allerhand, H. S. Gutowsky, J. Jonas, and R. A. Meinzer, J. Amer. Chem. Soc., §§/ 3185 (1966). E. R. Lippincott and R. D. Nelson, Spectrochim. Acta, 12, 307 (1958). H. P. Fritz, Adva. Organometal. Chem., 1“ 280 (1964). -K. Nakamoto, "Infrared Spectra of Inorganic and Co- ordination Compounds," New York: John Wiley and Sons, Inc., 1963, p. 216. M. Mikami, I. Nakagawa, and T. Shimanouchi, Spectro— chimica Acta, 23A, 1037 (1967). R. C. Fay and T. J. Pinnavaia, Inorg. Chem., Z” 508 (1968). R. J. H. Clark, Spectrochim. Acta, 21, 955 (1965). APPENDIX APPENDIX A. .LeasteSquares Analysis, Computer Program To determine kinetic data from a log k.y§ l/T plot it is necessary to know the equation for the straight line that best fits the experimental points. This is conveni— ently done with the following least-squares Fortran computer program, written by Charles Sokol, which gives the slope and intercept along with their standard deviations. The data cards are prepared as explained by comment cards in the program itself. 76 .1117. /.\ UUOODDZ F~Zm>Drw O~3mzm~OZ xamoovr<3moov.~0mzaawo. o AI~m UDCOD>3 EHFF o>rocr>4m 41m rm>m4.mcc>nmm mrcnm.~24mxom34 >20 41m m4>20>zo .IIZIIIdéom<~>4“ozm:Cfle41m:mfionm >20 Hzamnnmna now > mAI>_OIA FHZm 0024>~Z_zo moo or n fimsmz mXCmD~3m24>F UO~24modl~m UXCOD>3 n>Z mm CmmO 4C nbrncrbam 3CF4_UFF ECZm. 0241m nNqu O>4> o>DOImIOCFO I>r BCHZam HZ ncrczzm p n 40 wedlm mmnOZU C>4> O>DO wICCFO ~Om24~fl< AIW DCZ Cm~ZO norczzm ~ 40 mooAImZ OI4D ODDU ROD m>nI NXUHD~3mZA>F 03—24 cmfizm OOFCEZw _ 40 O HO n03 41m <>ch on X >20 OOFCZZm ”a 40 mo ROD 41m <>FCm Ofi Z .IIEIIInII>DOm ROD 4Im zmxa DCZ.IM40. >fiamnralm F>m4 O>4> 0>DO OR 0 41m F>m4 DCZ. AImBm flIOCFO :m OZm mF>ZK C>4> O>DU. - IOOIDnbo 333.3v 2., ; C WODE>4 Amuv ImC3Xuo 9-0C3d> Uofizam OZ 41m mAEDHOIfi FMZmo>ZamEOUH ~m 41m ”Zamnnquo Iszi? 2.:I. Hfl AZV HOO.HOO. N _ . .. _. . : .2 . .. .. m Dm>0 AOO.~HVA~0m24AHV.~nH.~O.HV H 3033>4 .~o>m. . : 30 b Munoz.» Dm>0 “@Q.Uv XAHV.<.~V u WODEDA Amnuoeov IIIIiIIIIIIImC?:x mC3x+xA~v,$.9CZXmOu0C3XmO+X.Ho**m w mczx24mDnU4nAmC3me*mC3mv SD~4W .0H.4. Z .4-WODE>4A*I*.*HIH ZCZUWU OW 3>4> UOCZHm 02 film 04D>~OIA F~Zm u*.~u. :egDHAIWAOqu -MOV :CKX. CZ/xomIeCZXu DowCZF N3 .WOD. .._>._IA* *o*mw CZXH *omlu I600. -VX.*UIC_<_r -l.I---; --IIIIIII.IINN WODEDAIAIA. -Ko*amc..._ ICC?..<9\ZH* onwmoo monk.“mC..<.XvU.C\ZH*onumomomwx.*amcmsr 3cm C\zI *.m_m .o.mx. x*mczxr 1-- ;.%;-Im<.\2vwoc>nmoI *.nwm. o.mx..¥mczxmol.mcsxvmosz *.hpm.o .mx.*.2I Imcm moc>onaecz>r . armoI *..n_m.o.mx. \.* m mcc>zmI *.m_m. o.mx.*3m czxr _r .s‘:. emeIlmczxcqoxz..Im.him. 93 034307» Efluflfl. Ami—.39 D WOD3>4 3*I*.*UOHZ4 ZCZQWD*.CHX.*X*.HmX.*<*V zofiqm 303.0.A~.x3~..<.~C.~u~.z._v ;.I o .non3>43* I.~U.HHx.mAn~m.q.vxvv snfiam Am“.yoCmronm.maomzamnnna.maom<324 IICOInonz>43*n*r*mronmu*.nHm.q+*nrcm on 332cm*.num.mm\.*.zqmnnmnqu*.npm 3.4.*orcm on 332cm*.npm.qv ,00 «o 00:. I III..: IIII.:, was nOZaHzcm . ,mzo;:II-IIIII:II-;I;:I:1 JI‘II-1.YI I' Doll .1... I .FODO IIIIIIIIqflr;LflOobmflmIOOO IIIIEIEI::. Z I iI I?:... IIiéII:.1.::II ODDS fiOCZfi u m# 20 F~4AWD~ZO I 13 OM Dbmmm IIIIIitiITILIIIII i,I I: I IOD FOHHND~ZO :iII,1:5.I23T53i.iIIIIIIIEIIIiIII.I11 II :.2 . .. ,I :. IIIIIHHI:IIIIIwI.II -I I, I IIIIIHWIIIIIHIWII III. IWIIIII III - I I I I,III.I I- -I I IIIHWHIIMII. MI I III.IIIIII:IIII.I I II I I I II II I I III I- I,I IIIiIII.IIIIIIIIIIIIEIIIIIWEIIEIIMMIIIIM-I.I I I I II I III II I I I. 'III I I I iIIIlIII l IlIlIIIl IIIIIII: IIIIIIIIIsI I. I -I.- I I I I I 79 B. Line Shape Analysis, Computer Program This program was written by John M. Sebeson, II. It is designed to calculate nmr line shapes by use of the Gutowski-Holm equation.29 The output consists of nmr spectra which appear as coordinate pairs (OMEGA, AMPLITUDE) where OMEGA is the chemical shift in radians per second relative to a central reference line, and AMPLITUDE is the height of the spectrum at point OMEGA. The program requires two data cards for each tempera— ture or for each set of T2 values for which theoretical spectra are desired. Any number of T values may be used for calculating spectra at a given temperature or set of T2 values. Also, any number of data card pairs may be used. The data should be entered as follows: Data Card # 1 Columns 1-10 Value of peak separation at the slow exchange limit (radians per second) divided by 2. Columns 11-20 T2 (= Z/half width in radians per seéond) Columns 21-30 T2B Columns 31-40 Proton fraction for site A Columns 41-50 Proton fraction for site B. Data Card # 2 Columns 1—10 The number of T values for which spectra are to be calculated Columns 11-20 The number of coordinate pairs desired in the output for each spectrum calcu- lated Columns Columns Columns Columns 21—30 31-40 41-50 51-60 80 The amount by which.T should be in- cremented for each spectrum calculated The incrementation constant for OMEGA values in radians per second The lower limit of T (the value of T for the first spectrum calculated) The lower limit of OMEGA in radians per second (this will be a negative number if the entire spectrum is to be included in the output or 0.0 if only the high field half of the spectrum is desired) . All data points must include a decimal point except the first two data points on Card #2 which must not contain a decimal point. The very last card after all the data card pairs must be exactly like the first card except that T2 - 0. A- This is the only way to terminate the program. IIKI I. 4.." .I.I.I--A4radnnwbr. p..>r;- I - - - - I - - -- - -- - - -- - I I DAIHEI Z. _> 4)... 10.2.43,”.U>.UA....C..._.-.- 312...»... .33.... ...A\A .. ).A.-AI -. III-I1.-- -I-IAIC.A.._.....C7A- Z». ssh-4.4.33. .UAnAFA AA.3-).AI..- - .. .. ..: .I-...I- -I: II; I- III-II -I.I III - I--. - - - - - JA::Z AC7. IWI4IAAJ.A.EA)))A.FA)33A.SA)))A 7.4%...)th I-.. .5. II I I- --IIIII 7.. A::uI I). . I..- -.IIIII.--..-.II..+.;.A.4.O:A- I -.J.o .......A....-4. ..- A.->. A. 3.4.4.. J-.--.-... .. -.- - I - - I - :1... A ..IAuI ~I7..u\....AI.4 . A O. --.I..-IIII-I.I.m-.II.A-u-30.A> . -A- -A-r.-.-IA-I-- I.-I...-.--..: - --.I .- : . . .. . .- I. -I - - - H- I- - I - -I :- --- - - . - £2.41 AZEEC. Q. 0 TC... A... AAIAIA-min-AA.4-AI.._IIAII..I.-mI4I-AI-..-Ar.II>-.I,.I-.-Am-,..:- III. .--C...-.-:.4AH._....-...A......._...CA...A.,...A.-.? IA...- OAI ......-.I:.- ..I..A....->.A), .A.-.A 5.... A..-I---II.I---- - If?! I I- I A4 A A...- AuflflCZA/Zfifi. _I A Z... HIE-1.... III.-- I- III. .-..&A 4.....4II.-.-AZ.-AA-.-......-h . AIM-IA. --I-.I ..II..I..-I.--.--. - - -- II - - - .I. --II-I.- I - -- - - - - - .3 WC.):....>.A.-._.AA..IAHA-A 07. 4:4...” A.-.-I>Z 1A...1A.3AI..A.,A.\..-II 4A....A4. )4 4.1.4 :3...- ..A/AhrIA-Ivi. 1.4%.... II.-.I-.I-..V-JD...-A.)..-...-.>.A.i A. 5) 4.”...Ai4-DAIII ZA- _.:-....A..-u:-...I...-JC .7. LA. ..n. . . - .. . . -. - I - VDD WOU$>4A3WADo I. IIIIIIIIIIlI-IlI-.IIII...—I..rr-«IA-sums)...SIM-v.) VI o-.M...J...Iwo..IoI--M “IA-4.x...IIIII5I. I- I. III I-I- -.lI- ...... I I - - -I- - I- I I --II -I. I... III- III .- - IIIII-.-I-III-.I.I--. I I - - - .I - EII v.3) UNA/.4 A 7. E: . .31.. A 7.. . Eu . _.\ . 1...... . 4AIAI . A-AI . -- - -....-M.-44.3 A..-.,.C.._.-_....>4 -A .u A .A.-.D....A...A.-..A.....4.7... - . . .. . . . . . 1 - - - - M3. ..fA4rI AZ..._.....r-.A ..,A_.>.A.-..->I-.4..:...U> A43 - IIIIIIIII- -I- HIM-I.«.II-O..A._.....->..A -.AInxAAA-474.142.!Micah-UTA!) oI..- o o . 3. 2.1.4.” >.. I MAI” ) o .o AVA/K . Zia-(.31.. .III 1.-..) o H . o H)?» 5: - - I .> H.34of. A3<.\.T.A.. UTA-34...... I-II-IIII-I-IIIJA... A «at. war“... .....AIA-AA-AII..-A-A-I.IM.A--..I4-_.--.III-IIIIIII-I . I -. .. I - - - - ..-- -.I .. 51-.---.- I-I. III-zI-I. I-- II-II-II - -I- - II: -- Ab WCJED4A-ACTD4IAI 7i? $.41_245,.u<.A .J<.A .42: {>7 AZ.I<.W4.2.J< IIII-IEIIA...4:AA. .14W344W4AAA; >..,-. W 7. A4 C..-€A.fl<.>. .. -.. - :5... 4. AA... ....._._... A... 131.40 . .Ir III- -.--I-- A .n...- .T- (7.4.24 A-AJ. ._.I44._IA A .....-_ ..A./.--A...IA..-_ AT 4. ...-...24..- I -.-..I.....<.. .I. J--. .I...< . I. ..A..._..-..-.;..;... - .,.:2_I_. .fl-qi- A.-... I. . I..< . hump-n. . - I--.-I I - A4<.Jo:A.1A.A.ATZII 3.4: s 771?: FATAI Cw.-4<. fil. _A IIIIIIII.IIIII-I-..-.... ....n.-.4.......-I._I..H_.\IA.. .. . 5.- - - -- - III . ----..I.- III-III..-- III-I-I.III- -I- . I-- I- II- . ----.I J). u A ”A .Z. . -III-II.-I-;A>Lm. EA«-I%IIIIII- -IIIiIIII-I.I III -II -I ..- . ;.I ;.-; i I.1.-:-j :1-5II I :L- -- -.- - . .1.-. A 4... AZ.....:.A,.4 . A: 4.)... 5.--..-..-..b-.rI.c4.,._...>.A..- ANA-$.42; -w-.AwIAI-l.,\-..AI II: -- - - - - - II -. - - - - 34ifi2XnC.- Aw>x.fl3 - II -I --..---..-A..,.......mD.>. Aw... ".._-..- ...-......,vA. -. - . - .. j - - - - - - - - - - 1 - I - - . CA...A.u.A-A,,_.:AAAA.\A4-.A..-.4JZAAIC.:. :2A.... ;xV+AA:-..>.)A+J.\4..AA-UAA\4UA.A - - -. I . .. I. n... A .2... A ”.4. >.....*-.A-A-. :7 32 A ... A I.,._.. .2 .. AL. ..,IT ..A A i . . - - 3A,...u..:.....-Ia..:A_;_.- A. o+4>..\-A.J>-. 4>..\ 2. Ar....A,...A.AAIo>IU3.. I ..-I-II .--.II.-.I-- A .AA+:A -vA/AASAZ. u JDZ4A -33 .OD. .IATA:A.4A_ 4.xZAZAA >:AAA.\4U>IA.\4u3.A. 7.. An... CCu4.2x \A:A?.x -...I\..-. I1-.\u-.)I-+-.-.--u )\JIJIIv:4u>.A_.I~ 0 w +.wAA..._.JIu..A. U+7Aa2£vL>UL A.¢....2...XJI)-._u».fl d O .\n1\\-I.)ld .\.1\.-J..u.. V“*U+.\JO< \AIJJ .25... AA,_ . _ ..A.. A.-.- ..A..>-_-..A.AH-AI.A.... .. -.\..4..J- . I. .o. I: f‘a a I -wo.::.-4 IACC.I3;. VS”). 30 33v. H-HAoZ ..pACC.\4z:JA m AA.- A2 ...A: A A ..1..4...-.A._-.. A..- ...:- .n..CA.....Aw-A-A;A-t COM. 4-3 37....- A 7: .7- >913»... ~ 17.-MIT.“ -I-II-III-I-..-II-II..I-III ...-III.- ,I I... -.I. II III- - II. III-..I.,III III-I.I-I-..-I.II-I:IIIIIII.II ...-II II-III.-III....II..I.-; ..I- I - --. I 4.4 3:. AI .wzm ---;IIII--IAIAQAn4CAAAA.-myquIdazflmA1m733442I,-- - - ------ - DDJ -.WDZIAZ:1 -.--.-.--.- .- .. AAHA> <_ ...C. A 7.. ...... .C A 4-.A...-IIMI.H,AI...-_._..4& . CD? A CE -.I-II I-.. -. -.I.-.-I-.-.-.-.I.-II..-.- .I.-II... IIII.II.I..-. .I--..-: I; I: - 034... T-CZ.....>4 AbDA ADA-AD4 A C CT. AI...- 77 ....C... ...... ...... .A..r.. .2 ...-...... ...-.I....AI ITAISAA-u ...-.._ A 7... ...... ...... u . AH. ) o J... ..--.-- - AA-IA. .zcifiACmc-I... .-.-.-...Am...7....-A.I-...-Inc _ .._-c . 031. . .. ... . . - . - .I. - .. . --I.-..I.. .I..-.- - 04b {UA4U AZSiC. 344. --I-IIIIJJ-4. 1:12:24.A?..33rA---d._-..fl.c.;.ANA-Am-.....-..I- ......r..Z-I..IC...--A.,_.-..... . 1+3 .nC-..AI1-_.rb4.Ao.z? ..c. 4.. 24.-«..I. ... A... - - .4742. A IIIIIII-I-mT-mI--w...+,....u-VAAAVM. . ----I ..III-- I- -I. E. I-.- - - -- -I ..--III--- II I- --I- I. ......uAA-AwA7A.....C....CD.1.)D. «62A . (I... . I--.I. -I--on. 1.9.2.2345: nil-......A I.-- ...»...x. A .... . .. -n ...-.._..4.n.....-.<. uA_1A.AC.r.-.I_......A,_..C-.n..._n- A+-.-..J.-.-.. --.w-..AI.-A-..-,I-.-.II-:. --I---.---- I..- - - -- --I|--I---I.ofl. I-II--II:---II- .I.... I- - I - - - - -- -- I- ---I - - - I I- - ... . A ...-n...- IA7_<...:.A-4I. ..wa 3 WCZ§>4 AAED.ADK.HICEWfi>. UK.DI>AUFA 4.:77. BK WLZWT44. AA.O--DIII....I-..I.--A.I.I 7.... ..I--III-.-III..:----. .- I --I -. .. - . .-I-II.-II-I-- .I. III-III-II- III-IIIIIIIIIIII-III- .I . . I. II-.-.. ....) A 4"... A7242...)- . I: C.... GA A . A . >. ..C....-A :-.A.._ .A .. WWW-4.4 IA A ... I -I- -- 5-4.-.AII-AAACA.C._.A4-- AA-L-I-.A)X.-JAI.4o-.A - .- - - -- ---- I .-- - b 1.3.4.4 A A... ...“.- - - N- 12.2.7...- A-7-A....T-.. -- I- - - - - - - - - -- - - 33 43 M33 .II-IIIII-M......»I...:.-.....I.AI...--A-... A.Z.C.n.I.H-.---I-: ...- -. II-I .. ...-I..I-II.I.-I.-IIIII . .I.- MD.» .I.-... 2...).-AAJIDAI--..7A14Am.-ZI14,1).In4AIC. I -I I -3 II-AIIA/IJI -- I.I-III-I-II -.I--II-I-IIII --I-.I- -- III II -I--I IIIIII -II - I I-I - 15n7.14:34 . D 1. II. III ‘ J.I' Al M I TIT‘IIWITWIWW lul WIT W!“ IHIWIFHIW Es 3 1193 03012 3896