. 31’3””??? I; 2‘99. i l" HESIS iazffi‘i 3.9:. ."f a Kr ‘Rfi '."é.;x9»- .73: 5% 2.4.2511? state ,h .Usiveesity This is to certify that the thesis entitled THE SYNTHESIS AND CHARACTERIZATION OF A NOVEL MOLYBDENUM-IRON-SULFUR CLUSTER presented by Paul Ernest Lamberty has been accepted towards fulfillment of the requirements for Masters degree in Chemistry H. . Major professor Date_E'_”_'_ij_ 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution _4_.._h——-— a’... l MSU l RETURNING MATERIALS: Place in book drop to 1.18? ’s ‘ remove this checkout from Mr record. F'N_§_S_ wi‘.‘ 3. F . ,5 ’ .V ..~ . : f‘jt: 2?. ~ ‘ ' 3’3 1 0 .Il ; ‘l . V 'Cp'u ‘13-“. r.v I ‘ 11' 3? :"rm 2 ‘ «#31? THE SYNTHESIS AND CHARACTERIZATION OF A NOVEL MOLYBDENUM-IRON-SULFUR CLUSTER By Paul Ernest Lamberty A THESIS Submitted to Michigan State University in partial fulfillment of the requirement for the degree of MASTER OF SCIENCE Department of Chemistry 1983 g: _ //.a :;t_:f2:z:// ABSTRACT THE SYNTHESIS AND CHARACTERIZATION OF A NOVEL MOLYBDENUM-IRON—SULFUR CLUSTER By Paul Ernest Lamberty Reaction of three equivalents of Fe282(CO)2- in tetra— hydrofuran with one equivalent of molybdenum pentachloride in acetonitrile at -78°C affords a novel molybdenum-iron-sulfur cluster containing a 1:6 molybdenum to iron ratio. The com- plex has the formulation [MoFe6S6(CO)l8]2-, and has been isolated as the 1,2-ethylenebis(triphenylphosphonium) and benzyltriethylammonium salts. The optical spectrum shows absorptions at: Amax: 327, “59, 572 (sh), 668 (sh).. The infrared spectrum shows peaks at 2070, 2037, 1990 cm'1 in the carbonyl stretching region. Electrochemical measurements indicate one quasi—reversible oxidation and two reductions, one of which is reversible. Conductivity measurements support the formation of the complex as a dianion. Mossbauer spectra have been recorded at A.2 K in a 600 G magnetic field, and indicate that all iron atoms are equivalent, giving only a simple quadrupole doublet. Paul Ernest Lamberty Magnetic susceptibility measurements show that the com- pound exhibits an unusually large magnetic moment (“eff = H.85 BM) and slight temperature-independent-paramagnetism. Attempts to obtain crystals suitable for an X-ray struc- ture determination have met with little success; possible solutions for the problems encountered are discussed. To My Family: Robert, Margaret, John and Thomas And to Jan, a very special friend Thank you. ii ACKNOWLEDGMENTS I would like to gratefully thank Professor Bruce A. Averill for his advice and guidance throughout my graduate career. His knowledge of synthetic inorganic chemistry and insight on the interpretation of the results have been in- valuable. I would like to acknowledge Dr. E. Mfinck and Dr. T. A. Kent for obtaining the Mossbauer data and Walter E. Cleland, Jr. for his aid with running the electrochemistry experiments. I would like to thank Walter E. Cleland, Jr., Jim Davis, Mark Antonio, Vijay Kumar, and my other co-workers for their friendship and advice on matters pertaining to my research and graduate career. Special thanks are extended to the Department of Chemistry for providing a teaching assistantship and the University of Virginia for support during my last term. I would also like to thank Professors Brubaker, Eick, and Babcock for serving on my committee. And last, special thanks to Mrs. P. Warstler for the typing of this thesis, to Mrs. J. Moore for her much ap- preciated secretarial help, and to Jo Kotarski for her excellent work on the graphics. iii TABLE OF CONTENTS Chapter Page LIST OF TABLES. . . . . . . . . . . . . . . . . . . vi LIST OF FIGURES . . . . . . . . . . . . . . . . . . vii LIST OF ABBREVIATIONS . . . . . . . . . . . . . . . x I. INTRODUCTION . . . . . . . . . . . . . . . . . 1 II. EXPERIMENTAL . . . . . . . . . . . . . . . . . 12 A. Materials. . . . . . . . . . . . . . . . . 12 B. Methods. . . . . . . . . . . . . . . . . . 15 C. Syntheses. . . . . . . . . . . . . . . . . 16 1. Preparation of u-Dithiobis- tricarbonyliron) . . . . . . . . . . . l6 2. Preparation of (BzEt3N)2- [MoFe6S6(CO)18]. . . . . . . . . . . . l7 3. Preparation of [Ph3PCH2CH2PPh31- [MoFe6S6(CO)l8]. . . . . . . . . . . . 19 III. RESULTS AND DISCUSSION . . . . . . . . . . . . 20 A. Synthesis. . . . . . . . . . . . . . . . . 20 B. Solution Conductivity. . . . . . . . . . . 25 C. Infrared Spectroscopy. . . . . . . . . . . 28 D. Electronic Spectroscopy. . . . . . . . . . 33 E. Mossbauer Spectroscopy . . . . . . . . . . 38 F. Magnetic Susceptibility. . . . . . . . . . A7 G. Electrochemistry . . . . . . . . . . . . . 53 iv Chapter H. X-ray Structural Studies IV. CONCLUSIONS. REFERENCES. Page 62 65 66 Table II III IV VI VII VIII LIST OF TABLES Elemental Analysis Data for the 1,2—Ethylenebis(triphenylphosphonium) and Benzyltriethylammonium salts of [MoFe6S6(CO)18]2-. Solution Conductivity Data Obtained in Acetonitrile. Infrared Spectroscopic Data in Carbonyl Stretching Region. Electronic Spectroscopic Data for [More6s6(00)l8]2‘ and Fe282(CO)6 Absorption Spectra of Some Non-oxo, 6-Coordinate Molybdenum (IV) Complexes. Mdssbauer Data for [MoFe6S6(CO)18]2-, Fe2S2(CO)6, and Other Iron Compounds Electrochemical Data for [More6s6(c0)18]2‘. Electrochemical Data for Other Compounds. vi Page 21 27 32 36 37 A6 55 61 Figure II III IV VI LIST OF FIGURES Schematic representation of the proposed models for the iron- molybdenum cofactor. Schematic representation of the structurally characterized MoFe3Su cubane clusters prepared to date Schematic representation of the linear molybdenum—iron-sulfur clusters prepared to date. Reaction scheme for the formation of [MoFe6S6(CO)18]2-. Generation of Fe2s2(CO)§- was accomplished with LiEt3BH, after which one equivalent of molybdenum pentachloride was added to every three equivalents of Fe282(CO)g—. Infrared spectra of [MoFe6S6(CO)18]2-, (top), and Fe282(CO)6 in acetonitrile at 20°C. Calibrated with polystyrene film Optical spectra of [MoFe6S6(CO)18]2-, (solid line), and Fe282(CO)6 in aceto— nitrile solution at 20°C vii Page 10 23 3O 35 Figure VII VIII IX XI Approximate ranges of isomer shifts observed in iron compounds relative to metallic iron at room—temperature. S refers to the spin quantum number. Mdssbauer spectrum of [MoFe6S6(CO)18]2- as a polycrystalline sample diluted with boron nitride. with a 600 G magnetic field. Méssbauer spectrum of Fe2SZ(CO)6 as a polycrystalline sample diluted with boron nitride. “.2 K with a 600 G magnetic field. Magnetic susceptibility of (BZEtBN)2- [MoFe6S6(CO)18] as a function of temperature in the solid state, ob— tained in a 2.000 kG magnetic field. Units for Xm are EMU-M- Inverse of magnetic susceptibility of (BzEt3N)2[MoFe6S6(CO)18], (M°EMU-l), as a function of tem- perature. Slight downward slope is due to temperature—independent- paramagnetism. magnetic field Obtained in a 2.000 kG viii Recorded at “.2 K Recorded at Page Al “3 “5 51 53 Figure XII XIII XIV Differential pulse polarogram of first and second electrochemical 2- reductions of [MoFe6S6(CO)18] . Cyclic voltammograms of the first reduction process (top), and the oxidation process of [MoFe6S6(CO)18]2-. Scan rate: 100 mV/sec Schematic representation of the proposed structure for [MoFe686- (CO)18]2-. Compound has been isolated as (BzEtBN)+ and 2+ (Pb PCH CHZPPh3) salts 3 2 ix Page 57 59 63 LIST Solvents: MeCN EtCN MeOH EtOH Et2O i-ProH THF DME Reagents: KOH MoCl5 MoClu Fe(CO)5 Na2S Na2S5 Fe2S2(CO)6 LiEt3BH Miscellaneous: 2- [Ph3PCH CH B zEt3N BM SCE Ph +- PhuAs 2+ PPh3] 2 2 OF ABBREVIATIONS acetonitrile propionitrile methanol ethanol diethyl ether isopropanol tetrahydrofuran dimethoxethane potassium hydroxide molybdenum pentachloride a-molybdenum tetrachloride iron pentacarbonyl sodium sulfide sodium polysulfide u-dithiobis(tricarbonyliron) lithium triethylborohydride u-dithiobis(tricarbonyliron) dianion l,2-ethylenebis(triphenylphosphonium) benzyltriethylammonium Bohr magneton standard calomel electrode phenyl tetraphenylarsonium I. INTRODUCTION The limiting factor in agricultural production is the supply of fixed nitrogen. Every year over “0 million tons of anhydrous ammonia fertilizer are produced to supply the demands of our nation's farmers. Ammonia is manufactured industrially via the Haber-Bosch process,1 which utilizes high temperatures (ca. “50°C) and high pressures (ca. 350 atm). In order to manufacture ammonia on an industrial scale, very large amounts of energy are consumed, most of which is obtained from the use of fossil fuels. Because of the decreased supply and increased cost of fossil fuels, research has been focussed on developing potential alternative processes for ammonia production. Recently, much attention has been given to the metallo— enzyme, nitrogenase, which is responsible for biological nitrogen fixation and has the unique and valuable property of catalyzing the reduction of nitrogen gas to ammonia at ambient temperature and pressure. The exact mechanism of this reduction is not known, but knowledge is slowly being accumulated through many approaches, one of which is the construction of synthetic models for the metal centers of nitrogenase. The enzyme is known to consist of two oxygen—sensitive components, an iron protein and a molybdenum-iron pro- tein. The iron protein (m.w. 60,000) contains four iron and four sulfide atoms per molecule. Spectroscopic and cluster extrusion experiments indicate that the iron and sulfur are present in a FeuSu cluster arrangement.2’3 The molybdenum-iron protein (m.w. m220,000) contains approximately 32 iron atoms, 30:2 sulfides, and 2 molyb- 2 It has been found that four denum atoms per molecule. of the metal components of the molybdenum-iron protein consist of Feusu units. The rest of the iron and sulfur and all of the molybdenum is contained within a low- molecular weight cofactor.“ This cofactor, which can be removed from the molybdenum-iron protein, contains six to eight iron atoms and approximately ten sulfides per molyb- denum atom and constitutes a novel cluster. The low- temperature (6-20 K) EPR spectrum of the cofactor shows an axial signal5 with slight rhombic distortion (g-values 2.0, 3.65, and A.3). This has been shown to originate from one of the Kramer's doublets of an S = 3/2 system.5’6 Low temperature 57Fe Mossbauer spectra of the isolated 7 cofactor are similar to the spectra assigned to the co— factor centers of the molybdenum-iron protein, except for slight broadening of the lines and for the presence of a small quadrupole doublet (AEQ 2 0.9 mm/sec., 6 2 0.“ mm/sec.), which suggests that approximately twenty percent of the total iron is present in another environment. Analysis of the spectrum of the metal centers in the native protein has shown the presence of six iron atoms in each S = 3/2 metal center,8 in agreement with the presence of six irons per molybdenum in the cofactor. Mo K-edge extended X-ray absorption fine structure (EXAFS) spectroscopy indicates two or three iron atoms at approximately 2.71 A and four or five sulfur atoms at approximately 2.35 A as nearest neighbors to the molybdenum. Recently Fe-edge EXAFS studies9 indicate that the iron atoms in the cofactor have an average of 3.“ i 1.6 sulfur atoms at 2.25 A, 2.3 i 0.9 iron atoms at 2.66 A, 0.u i 0.1 molybdenum atoms at 2.76 A, and 1.2 i 1.0 oxygen atoms at l.8l A as nearest neighbors. Thus far, available evidence concerning the actual mechan— ism of reduction of dinitrogen to ammonia indicates that electrons are transferred from an external reductant (re- duced ferredoxin or dithionite) to the iron protein to the molybdenum-iron protein to substrate.2 Therefore, it is believed that the substrate coordinates to and is reduced at the iron-molybdenum cofactor within the molybdenum- iron protein. Due to the extreme oxygen sensitivity of and problems in purifying the cofactor, its structure has not yet been determined. However, a number of structural models have been proposed; see Figure I. Until recently, no synthetic molybdenum—iron-sulfur clusters were known, making it dif- ficult to assess the validity of the proposed structures. Figure I. Schematic representation of the proposed models for the iron-molybdenum cofactor. / 38 \l S \ s—Fe’ I / M: x \s/\ Mo '2_ I S F: /\ —S 8 Figure I By synthesizing and characterizing molybdenum—iron- sulfur clusters the inorganic chemist hopes to provide a body of data, which upon comparison with corresponding data obtained for the cofactor, will contribute to an eventual elucidation of the structure of the iron-molyb- denum cofactor and the mechanism of action of nitrogenase. Synthetic efforts have thus far led to two classes of molybdenum—iron—sulfur clusters: the "cubane" complexes and the "linear" type clusters. Examples of "cubane" clusters includezlo-27 [Fe6Mo2S3(SR)9]3-, [Fe6Mo2S9(SR)8]3-, [Fe7Mo2SB(SR)12]3-’u-, and [FeuMoSu(SR)3(C6H402)3]3-, presented in Figure II, all of which contain the Fe3MoSu core unit. EXAFS studies on these compounds show that the Mo-Fe and Mo-S (bridging) distances are virtually identical to that of the Mo site 10’28’29 and the observed 57Fe Mossbauer of nitrogenase, isomer shifts and quadrupole splittings are close to those reported for the iron-molybdenum cofactor.7’3O But these are not detailed models for the cofactor due to improper stoichiometry (3 Fe/Mo and too much S=) and none of these compounds exhibit EPR spectra that approximate that of the cofactor. The "linear" class include: [FeMoSuX2]2— (X = SAr, OAr, Cl), [Fe2Mos6x213‘ (x = SAr, x 2 2— 31-37 . . and [Fe2MoSuCl2] , shown in Flgure III. These also = 55), [Fe(MoSu)2]3— do not constitute accurate models for the iron-molybdenum Figure II. Schematic representation of the structurally characterized MoFeBSu cubane clusters pre- pared to date. Figure II P "3’ - q as SR - F! S g S F: RS\F¢\-5 5-7457 R5 — -; SR RS / §\ SR 9-(Fo\/)S>M{S>M-{s\>¥e¥s s-fiés—M-S-M—SE‘FP-‘Is RS sn RS SR L. _ L .1 F — 3'.4’ Ra}: g g fiffi _ I 58 éfii€>uég>i -3>M{S\ Fe>(5 ns/xfiés \s/ \z/ \séfi/ / R \ RS SR L. _ ._3- p _ 13 RS SR RS Q76 5 S—-F 2M0C1u + C1 5 2 2 C-CCl 3 3 A 250 ml, round-bottomed, side arm flask was charged with 8.0 g M0015. 100 ml degassed carbon tetrachloride and A0 ml degassed tetrachloroethylene were added via cannula. The flask was then equipped with a reflux con- denser and an external mercury bubbler (with trap). The mixture was heated to reflux for 2.5 days. After cooling the mixture was filtered, washed with carbon tetrachloride, and vacuum dried. The product was collected as dark black microcrystals. The X-ray powder pattern showed intense d spacingscfi‘6.18, 5.97, “.50, 4.07, 3.30, 2.88, 2.60, 2.56, 2.22, 2.11, 2.02. Literature38 (d,I) values: 9.092, 5; 6.099, 100; 5.888, 100; 5.383, 5; 5.27M, 5; 5.091, 10; H.866, 10; H.618, 10; H.4AU, 80; “.386, 5; H.0A2, 80; 3.266, 30; 2.936, 5; 2.856, 30; 2.750, 5; 2.705, 5; 2.632, 100; 2.602, 25; 2.522, 25; 2.491, 2; 2.365, 2; 2.320, 2; 2.270, 2; 2.221, 50; 2.18“, 20; 2.131, 20; 2.102, 50; 2.075, 1; 2.012, 50; 1.961, 2; 1.897, 5; 1.863, 5; 1.838, 30; 1.811, 2; 1.753, 20 1.721, A0; 1.679, 1; 1.633, 30; 1.586, 30; 1.572, 2; 1.506, 10; 1.A67, 10; l.u57, 2; 1.385, 5; 1.37u, 5; 1.3u8, 2, 1.336, 2. 15 B. Methods Optical spectra were recorded by means of a Cary Model 17 or Cary Model 219 spectrophotometer. Infrared spectra were obtained with a Perkin—Elmer Model 457 or Model 237B grating infrared spectrophotometer by using sodium chloride or potassium bromide solution cells. Variable temperature magnetic susceptibility measurements were performed on a SHE Corporation SQUID susceptometer opterating at 2 KG. Room temperature magnetic susceptibility measurements were also performed with an Alpha Faraday balance using Hg- [Co(SCN)u] as a calibrant. Electrochemical measurements were made with a PAR Model 17UA polarographic analyzer employing either do polarography (dropping mercury electrode) or cyclic voltammetry (glassy carbon electrode). Mdss- bauer spectra were measured by Dr. T. Kent and Prof. E. MUnck at the Grey Freshwater Biology Institute, University of Minnesota, Navarre, Minnesota. Isomer shifts are re- ported vs. metallic iron foil at room temperature.39 Conductivity measurements were obtained using a Markson "ElectroMark" Analyzer. Melting points were obtained by using sealed capillaries in 22222 and are uncorrected. Elemental analyses were performed by Galbraith Labora- tories, Inc., Knoxville, Tennessee under the supervision of Bill Longmeyer. 16 C. Syntheses 1. Preparation of u-Dithiobis(tricarbonyliron) Fe282(CO)6 was prepared by the method described by Hieber and GruberuO as modified by Seyferth and co-workers. H-l- 2Fe(CO)i' + S ' + Fe2s2(CO)6 + 3H28 + 2C0 + NaCl 2 5 Sodium polysulfide solution was prepared as follows: To a solution of A00 ml distilled H20 and 100 g Na S, 2 55 g sublimed sulfur and 10 ml 50% aqueous KOH (w/v) were added. After the solution was allowed to stir until most of the sulfur dissolved, the solution was filtered and transferred to a 1000 ml side arm flask, degassed, and cooled to 0°C. A 3000 m1, round-bottomed, three-necked flask equipped with a mechanical stirrer, gas—inlet tube, and a tight fitting serum stopper was charged with 200 m1 degassed methanol and 35 ml filtered iron pentacarbonyl. After the vessel was cooled to 0°C and degassed several times, 80 m1 of degassed 50% aqueous KOH (w/v) was added via can- nula. To this solution, the Na2S5 solution was added as quickly as possible through a funnel, against a counter flow of nitrogen. During the addition, the mixture was vigorously stirred. A mildly exothermic reaction resulted with formation of a dark red solution. The reaction mixture was stirred at 0°C for two hours. A1 l7 Cautious acidification of the reaction mixture was accom- plished by slow addition of NAOO m1 6 M HCl from an addition funnel. Evolution of hydrogen sulfide and carbon monoxide resulted, and a brown precipitate and colorless solution were formed. After filtration, the brown solid was air dried and divided into three portions. Each portion was extracted with hexanes until no color was present in the extraction solution. The combined extracts were evapor- ated on a rotary evaporator. A reddish semi-crystalline product containing a mixture of FeS2(CO)6 and Fe3S2(CO)9 resulted. Sublimation at 40°C (0.1 mm pressure) overnight yielded 5-10 g of pure, ruby-red, air-stable crystals (12-24%, based on Fe(CO)5), m.p. U6-A7°C (1it.uO A6.5°C), I.R. (MeCN solution): 2085 s, 20uu vs, and 2006 vs cm’l A2 1). (lit. 2081, 2038, 1996 cm- 2. Preparation of (BzEt3Nl2[MoFe6§6(CO)181 . 2- F82S2(CO)6 + 2L1Et3BH + Fe2S2(CO)6 + Et3B + H2 2- fl 2- . 3Fe2S2(CO)6+»MOC1 + [hoFe686(CO)18] + LlCl 5 Fe282(CO)6 (1.20A3 g; 3.50 mmole) was dissolved in 90 m1 degassed tetrahydrofuran in a 300 m1 Schlenck tube, and cooled to -78°C by using a Dry-Ice/isopropanol bath. Then 7.2 ml of 1.0 M LiEt3BH solution was added at a 18 rate of 0.5 ml per 5-10 min with use of a gas-tight syringe. In a 50 m1 round-bottomed, side arm flask, MoCl (0.3190 g; 5 1.168 mmole) was dissolved in 10 m1 degassed acetonitrile. The resulting room-temperature solution was then added dropwise to the dianion solution at a rate of two drops per second, using a gas-tight syringe. After complete addition, the reaction mixture was slowly allowed to warm to room temperature. To this, (BzEt3N)C1 (0.5845 g; 2.45 mmole) dissolved in 30 m1 MeCN was added. The resulting solution was stirred for one hour, filtered to remove LiCl, and the filtrate was evaporated in £3939 to approximately half the original volume and slowly cooled to -20°C in a freezer. Filtration at 0°C resulted in m50 mg crystalline purple by—product. The filtrate was then evaporated in vague at room-temperature to N20 ml, after which 20 m1 de— gassed isopropanol was added drOpwise via cannula. fter slowly cooling the solution to -20°C and allowing it to remain undisturbed for two days, the mixture was filtered at 0°C, resulting in 350 mg crystalline product. The product was recrystallized by dissolving in 20 m1 of 50% MeCN in THF, slowly adding 10 m1 isopropanol, and cooling to -20°C for three days. This resulted in collection of 200 mg pure crystalline product (11.3% yield). Anal. Calcd for (BzEt3N)2[MoFe6S6(CO)18]: C, 34.93; H, 2.91; N, 1.85; s, 12.73; Fe, 22.17; Mo, 6.35. Found: c, 34.95, H, 3.55; N, 2.12; S, 14.01; Fe, 22.04; Mo, 6.78. IR (MeCN 19 solution): 2070, 2037, 1990 cm-1. Optical spectrum: 327, 459, 572 (sh) and 668 (sh) nm. 3. Preparation of [PhBPCH2CH2PPh3][MoFe6§6(CO)181 The procedure was the same as described for the benzyltriethylammonium salt except for the work up, which was changed due to the different solubility characteristics of this salt. CH After the addition of [Ph3PCH PPh3JBr2 (0.71 g, 2 2 1.28 mmole), the mixture was evaporated in yacug to dry— ness and redissolved in a minimum amount of EtCN. Et2O was added drOpwise until no further precipitation of LiCl/LiBr was noticed. After filtration and addition of an additional 5-10 m1 Et20, the solution was cooled to -20°C, whereupon crystalline product separated in N20% yield. Recrystallization was accomplished from EtCN/ Et2O. The product was isolated as dark needles in 15% yield. Anal. Calcd for [Ph3PCH2CH2PPh3][MOFe6S6(CO)18]: C, 40.03; H, 2.04; P, 3.69; S, 11.45; Fe, 19.94; Mo, 5.71. Found: C, 42.41; H, 2.78; P, 3.88; S, 12.83; Fe, 19.63; Mo, 6.90. IR and optical spectra are identical with those of the benzyltriethylammonium salt. III. RESULTS AND DISCUSSION A. Synthesis The compound was formed by reaction of three equiva- lents [Fe2S2(CO)6]2- with one equivalent molybdenum penta- chloride in tetrahydrofuran solution. After undergoing a metathesis reaction and upon concentration of the solution, the compound was isolated as either the BzEt3N+ or 2CH2 The stoichiometry of the compound was determined by (Ph3PCH PPn3)2+ salt. elemental analyses of the two isolated salts. The com- position percentages found agreed with those calculated for the proposed formulation: [MoFe686(CO)18], see Table 1. Furthermore, by comparing the %P or %N in the respec- tive salt, it was found that either one (Ph3PCH2CH2PPh3)2+ cation or two (BzEt N)+ cations were present. When taking 3 this into consideration, the compound must be a dianion with the postulated formulation of [MoFe6S6(CO)18]2-. The synthesis of [MoFeéS6(CO)18]2- from molybdenum pentachloride and dithiobis)tricarbonyliron) dianion, Figure IV, can be viewed as a combination oxidation-reduc- tion and ligand-exchange reaction. The molybdenum is reduced from Mo(V) to Mo(IV) with some of the dianion 20 21 Table I. Elemental Analyses Data for the 1,2-Ethylenebis- (triphenylphosphonium) and Benzyltriethylammonium 2- salts of [MoFe6S6(CO)18] . + (Ph3PCH2CH M13)? 750 % a % P % s 7. Fe % Mo 2 Calculated: 40.03 2.04 3.69 11.45 19.94 5.71 Sample 1 40.05 2.69 4.31 11.05 17.18 4.82 Sample 2 42.41 2.78 3.88 12.83 19.63 6.90 (BzEt3N)+ %c % H % N % s % Fe % Mo Calculated 34.93 2.91 1.85 12.73 22.17 6.35 Sample 1 34.95 3.43 2.25 12.88 21.94 6.44 Sample 2 34.95 3.55 2.12 12.88 22.04 6.78 22 .Ionoovmmmmm do mucofiw>fisvo omega zpm>o 0p poppm mm: oUHLonomucod Escopnzaoe do osoam>asoo oso soars sooan .zmmomaa seas oosnafiaeooon was uofioovmmmoa mo COHpmpocoo .ImmeAoovmmommozg mo cowmeLO% ecu Low oEozom COHpowom .>H opzw °r-l m um «I + mnmm + >H ossuaa Es: Es: Hoseomouué All uses. + O m O .J .2 7 G) 24 presumably being oxidized to its precursor, Fe282(CO)6, or to some other unidentified iron-sulfur compound. Three [Fe2S2(CO)6]2- units replace all the chlorides and co- ordinate to the molybdenum forming an octahedral sulfur environment around the molybdenum. The exact mechanism has not yet been established. Since [Fe2S2(C0)6]2- is thermally unstable, it is necessary to generate the dianion at -78°C. Tetrahydrofuran was found to be a suitable solvent, since it does not freeze at this low temperature and it is unreactive to the dianion. Unfortunately, molybdenum pentachloride is unstable in 43 tetrahydrofuran, producing the unwanted MoOCl3 complex; so it is therefore necessary to use acetonitrile as sol- vent. Since the molybdenum pentachloride will also react with acetonitrile after several hours to afford the MoC1u(MeCN)2 complex,uu it is necessary to minimize the time that the molybdenum pentachloride is left in solution. Also, since the dianion is decomposed by acetonitrile only a minimum amount can be used. Upon addition of the molybdenum pentachloride/aceto- nitrile solution to the [Fe2S2(C0)6]2_/tetrahydrofuran solution, the dissolved molybdenum pentachloride reacts rapidly with the dithiobis(tricarbonyliron) dianion at -78°C, thus limiting the amount of decomposition by tetra- hydrofuran. It is not unreasonable to suspect some de— composition which may be reflected in the low yield (m11%). 25 If the corresponding MoOCl3 is formed in some minute amount, the formation of an iron—sulfur-molybdenyl complex should result. The formation of the purple complex, as described in the experimental section, can be explained as a forma- tion of an iron-sulfur compound from the dithiobis(tri- carbonyliron) dianion. This purple complex has been re- 45 produced from other reactions not containing molybdenum. This compound has not yet been characterized, but single crystals have been obtained in an extremely low yield; a crystal structure determination is in progress. In the corresponding reaction using molybdenum tetra- chloride, no pure products were isolated. The infrared spectrum of the reaction mixture showed a large number of bands in the carbonyl region, which suggests the formation of a mixture of compounds that could not be separated by crystallization with the quaternary cation salts (EtuN+, + BuuN+, PhuAs+, PPN , (Ph3PCH CH PPh3)2+, BzEt N+) used. 2 2 3 The [MoFe686(CO)l8]2' cluster has been isolated as + N , (Ph3- various quaternary salts, including BuuN+, BzEt 3 PCH CH PPh3)2+, and PhuAs+. Single crystals of the 2 2 (Ph3PCH2CH PPh3)2+ and BzEt N+ salts have been obtained. 2 3 B. Solution Conductivity By comparison of solution conductance measurements, it 44,46 has been shown that the magnitude of conductivity falls into ranges depending on the number of ions in 26 solution. Thus, separate ranges are generally found for 1:1, 2:1, and 3:1 electrolytes. This neglects the forma— 47 tion of ion pairs, which frequently occurs in some low dielectric constant solvents. If this does occur, a sub- stantially low conductivity reading will be observed, usually well below the expected value. Solution conductivity measurements were performed on two different salts of the [MoFe6S6(CO)18]2- complex, and compared to several 1:1 and 2:1 electrolytes. These were performed on acetonitrile solutions with electrolyte con- centrations of 3.0 x 10-3 and 1.5 x 10-3 M. Relevant data are given in Table II. For 3.0 x 10.3 M solutions of known 1:1 and 2:1 electro- lytes,the ranges of 3.85-2.39 x 100 micromhos and 6.34- 4.57 x 100 micromhos were found, respectively. Conduc- tivity measurements for 3.0 x 10'3 M solutions of the 1,2-ethylenebis(triphenylphosphonium) and benzyltriethyl- ammonium salts of the compound gave readings of 2.34 x 100 micromhos and 4.34 x 100 micromhos, respectively. These values fall on or slightly below the low end of the 1:1 and 2:1 electrolyte ranges, as might be expected given the likely size and consequent relatively low mobility of the proposed dianion. Assuming that there is no ion pairing, this experiment suggests evidence that the complex is indeed a dianion in solution, thus supporting the formulation of the compound .pcomopa wCfiLHwQ COH Hmfipcmpmnsm * 27 m:.m z muoaxm.fi :m.: z mnoexm meAoovommoaozLNHZManmL H:.H z muoaxm.fi :m.m z muoexm flmHAoovomooaozLflmrddmzomzodrdu *om.H z mIOme.H *mm.m z mnoaxm mgmzsdvhmgsostL mua sm.m z muoaxm.fi em.: 2 micexm mtmfimsddmmomm0dmsdg NHH mm.m z mnoexm.a oo.: 2 muoaxm amozmfl236mL Hum m:.m z muoaxm.a :m.o z muoexm mgoadvhmrddmmommoamrdg Nae :m.e z m-oaxm.fi mm.m z muoexm mmardLm233mL mo.m z muoaxm.e mm.m z muoaxm HQZJom ms.a z muoexm.a mm.m z msoexm hzoaogfl235mL Hue AmocEOLOHz coaxv COHmeQCoocoo oEmz pawm maze zpfi>fipozpcoo oquOLpomHm .oeaspasooooa ca oosasooo mono soa>aposesoo soaosaom .HH dense 28 as being a monomeric unit with the oxidation state of the molybdenum being +4. C. Infrared Spectroscopy Infrared spectroscopy has been used primarily as a qualitative method for examining product purity. The pure product showed three well resolved bands in the carbonyl stretching region, while the crude semi-crystalline material isolated from the reaction mixture showed the presence of additional bands which were not as well defined. All Spectra were taken in solution, usually acetonitrile or propionitrile, with a NaCl IR solution cell adapted with tight fitting serum caps. The infrared spectrum of [MoFe6S6(CO)18]2- in the region of 2000 cm"1 represents C-O stretching modes. Three bands were observed at 2070, 2037, and 1990 cm’l, as shown in Figure V. In dithiobis(tricarbonyliron) three infrared active bands are observed at 2081, 2038, and 1996 cm-1, as reported by Hieber and Beck,“2 see Figure V. Assignment of all infrared and Raman active bands to their respective modes in dithiobis(tricarbonyliron) was accomplished by Scovell and Spiro.“8 Since there are only three carbonyl stretching fre- quencies present in the infrared spectrum of the [MoFeESR- (CO)18]2- cluster, the same number that is present in the infrared spectrum of Fe282(CO)6, suggests evidence that there 29 Figure V. Infrared spectra of [MoFe6S6(CO)18]2-, (top), and Fe2S2(CO)6 in acetonitrile at 20°C. Cali- brated with polystyrene film. TRANSMTTANCE We) a 8 a E? IOOb i3 a TRANSMITTANCE (7.) m C) .h C) l I 0 1 2500 2000 l500 FREQUENCY(cm") G) C) I O 1 1 2500 2000 l500 FREQUENCY(cm") Figure V 31 are three individual dimer units present in an arrange- ment where their vibrations are not strongly coupled. Upon comparison of the three carbonyl stretching fre- quencies given for Fe2S2(CO)6, [Fe2S2(CO)6]2-, and [Mo- Fe6S6(CO)18]2', see Table III, an interesting trend was observed. When Fe2S2(CO)6 is reduced to [Fe2S2(CO)6]2-, the net negative charge on the sulfur is transmitted inductively to the iron, thus increasing the extent of back-donation from the iron to the carbonyls. This in turn, increases the Fe-C bond strength and decreases the (DEC) bond strength, resulting in a downward shift of the carbonyl stretching frequencies. After coordination of the [Fe282(CO)6]2_ units to the molybdenum, some of this negative charge is withdrawn from the sulfur to the molybdenum, thus reducing the amount of charge transmitted to the iron. This decreases some of the back-donation from the iron to the carbonyls, causing a slight decrease in Fe-C bond strength and a slight increase in (EEC) bond strength, resulting in an upward shift in the carbonyl stretching frequencies. Further detailed studies employing infrared and Raman spectroscopy would be necessary to examine the other vibra- tions in the molecule. 32 Table III. Infrared Spectroscopic Data in Carbonyl Stretch- ing Region. Compound Frequency (cm-l) Fe2S2(CO)6 2081 2038 1996 x 2- [Fe2S2(CO)6] 2025 1975 1945 2- [MOF8686(CO)18] 2070 2037 1990 * Since this compound is thermally unstable, these frequencies may be incorrect. 33 D. Electronic Spectroscopy The optical spectrum of [MOFe6S6(CO)18]2- is shown in Figure VI, and pertinent data are presented in Table IV. Due to the high molar extinction coefficients of all the observed peaks, they must be ligand to metal charge transfer in nature rather than d-d transitions. Unfortunately, the spectra of only a few non—oxo, six coordinate molyb- denum (IV) compounds have been reported; a survey of these is given in Table V. All of these molybdenum (IV) com- plexes exhibit very intense bands in the region of 250-550 nm with molar extinction coefficients of m5-l6 x 103 M-1 cm-l. These have been assigned as L + Mo charge transfer transitions. Therefore, the intense absorptions at 327 nm (e = l -1 cm-l) can be 37000 m‘ om‘l) and 459 nm (e = 9700 M assigned as S + M (M = To or Fe) charge-transfer transi- tions. Similar absorptions with high molar extinction co— efficients in the 300-500 nm region were observed in other 31-37 molybdenum-iron-sulfur clusters, in which the charge- transfer transitions were assigned to S + M (Mo or Fe). At this point, it is not possible to differentiate between S + Mo and S + Fe transitions, although the absence of the band at m450 nm in Fe2SZ(CO)6 and Fe2SZ(CO)§- suggests that the 459 nm peak is probably due to an S + Mo charge— transfer transition. Assignment of the two weak shoulders at 572 nm 34 .ooom pm 2 0H .pzaom wads . peso . poem ca a Aoovmmmoa new . Aosa a we .Hom V -mflwagoovo a m oaozL do mppom Qw awed .aao .H> as , sue .a 35 o!« 1OOofix1 (n-‘mI-W) ’ com H> ossrae gag com com com oov con 5 . [I]. q u o / /// / ,1 . III/ : .0000— / 11 1 11 1 1 _ 1 H ” 1oooo~ _ 1 1 1 11 1 1 1 10000m — , _ _ 1 1 _ 1 .nxxgov .1 1 11 11 : .oooon . 1 (..w1)'-W)3 36 Table IV. Electronic Spectroscopic Data. Extinction Coefficient Compound Peak (nm) (2 mole-1 cm-l) [More6s6(00)18]2’ 327 37,000 459 9,700 572 (sh) 4.900 668 (sh) 2,900 Fe2S2(CO)6 278 (sh) 8,800 333 14,000 450 (sh) 1,000 *Fe s (00)2‘ 330 -11 2 2 6 430 (sh) -—— 580 111 * Since this compound is thermally unstable, these absorptions may be incorrect. 37 Table V. Absorption Spectra of Some Non—oxo, Six Co— ordinate Molybdenum (IV) Complexes. Peak Extinction Compound Anm Coefficient Assignment Ref. MoClg- 242 CT 50 272 CT 294 CT 352 CT 387 d-d 450 d-d MoClu(dipy) 239 14000 CT n+n* 51 302 12000 CT w+w* 368 1200 CT 397 800 d-d 546 690 d-d MoClMMeCN)2 240 3000 CT 44 285 2400 CT 314 2900 CT 359 W900 CT? 400 m400 CT? 495 75 d—d Mo(sal-NEt)2C12* 352 15290 CT 52 430 4360 CT 550 5810 CT * Sal = N-substituted salicylalimiato. 38 1 1 (e z 4900 M- cm-l) and 668 nm (e = 2900 M— cm_l) can be only speculative. Their molar extinction coefficients are again too high to be d-d transitions, so they must be due to charge-transfer. The energies of these transi- tions are different from the usual high energy bands seen in molybdenum-iron—sulfur clusters. This energy difference may be interpreted by examination of the o and n symmetry of the bonding orbitals within the complex. This allows both So + M and St + M charge—transfer transitions with differing energies. Most pseudo-octahedral molybdenum (IV) complexes exhibit two peaks in the visible region above 400 nm, which have been tentatively assigned to the d-d transitions: 3T2g + 3 “9 These peaks usually have molar -1 cm-1 44,51 3 Tlg and lg' extinction coefficients well below 1000 M 3 Tlg(P> + T It is a reasonable assumption that these d-d transitions are obscured by the lower energy charge-transfer transitions at 668 nm and 572 nm. E. Mdssbauer Spectroscopy The 57 Fe Mdssbauer spectrum of [MoFe6S6(CO)l8]2- has given useful insight towards the structure of this complex. In the proposed structure, all of the iron atoms are equivalent with each having the formal oxidation state of +1. The expected 57Fe M8ssbauer spectrum would consist of a doublet with an isomer shift and quadrupole splitting 39 near the values obtained for Fe(I) in other related com— pounds. 57 A correlation diagram for Fe Mdssbauer is shown in Figure VII, in which isomer shift ranges are drawn as a function of oxidation state.53 The isomer shift ranges for the different oxidation states in ionic iron compounds seldom overlap one another, so it is therefore relatively easy to determine the oxidation state of the iron in un- known iron compounds. The 57Fe Mbssbauer spectrum of [MoFe6S6(CO)18]2_ at 4.2 K is presented in Figure VIII. A doublet was observed with EQ = 0.80 mm/sec and 0 = 0.06 mm/sec. With comparison to the spectra of Fe282(CO)6 shown in Figure IX, and other compounds, Table VI, the isomer shift and quadrupole splitting values are in good agreement with those of similar Fe(1) compounds. Since no magnetic hyperfine interactions were observed, probably due to rapid electronic relaxation in the solid plus an integral spin ground state, nothing else can be determined except that all the iron atoms have the formal oxidation state of +1 and they are arranged so that they are all equivalent. Figure VII. 40 Approximate ranges of isomer shifts observed in iron compounds relative to metallic iron at room temperature. S refers to the spin quantum number. 41 DFe(l)S=§2 D Fe(I)S= ’12 L j Fe(ll)S=2 DFe ossmaa 43 m\EE c1 3_uo_m> v m N _ o _. N- m- 19.. . 11 I _1 1 _ .. ____ _ . l _ _ _ 1 _ _ _ 1 1 1 1 1 u 1 11 1 _ 1 .11 111.1. __ .. 11 1111 .1 1.1111 .1 1111.1»... 1.1.1. .. ;1_.:1=.11111§1n._._1.1.1...1111. 11111 11111111111511.111121621111...1.1.1.1..£31.11... .1 - p — P — _ _ _ _ waxed u! uoudlosqv 44 .Uamflm oaumcmmE 0 com m prz x m.: pm U®©LOo®m .mUflLpHC COLOQ Cpflz popzaflp oHQEwm ocflHHmBmmgozHOQ m mm mfloovmmmom mo ESLpoQO Losmpmmmz .xH ogsmflb 45 xH ogztflb m\EE c1 b_uo_m> o 1.. N- m- - _ _ . q _ q T. L I. .. .. I1 .I . . . . I1 1| . . 1 r . . .1 T. . I1 . 1| . . 1 1... . 1 1| . . l . T _ . .1 I . . _ I. 1 T . . . 1| .. . . i . . 1 ... .. . .1 .I ’33... .. I1 ... I .‘ I. l 3 t f T 3.1.21.1...3...c....i....._.................. .x. .I.......:1......t....t.................(.....l. 4 _ _ _ _ _ . _ rue-mad u! uoudiosqv 46 mm no.0 -mfiagsdmvoaL em mm.o -mflqAnamvzmaoaL om ea.o umflmAstuoummvmmmoaL mm x as am.m mwo.e mgoovoa am g on mmw.o mm.o onovoaonzmvoaonovunao 2m x om awo.a Hao.o mgoovoamgsdmvoamgoov I- u can .x m.: HH.H mo.o afloovmmmoa u- 0 com .x m.: ow.o oo.o mwagoovomooaoszAZmomumv .mom meMEom Am\EEV Am\EEvc pCSOQEoo qua .mUCSOQEoo COLH Locpo pew .ofloovmmmom .INmmHAoovmmwmm021 Low mpmm poswnmmmz .H> manms 47 F. Magnetic Susceptibility The room temperature magnetic susceptibility of [MoFe6S6(CO)18]2- has been determined by the Faraday method. The data gave rise to an effective magnetic moment of 4.85 BM, after corrections for diamagnetic con- tributions to the susceptibility from the ligands and cations by use of Pascal's constants. This effective magnetic moment is significantly larger than the spin- only value (“eff = 2.83 BM) expected for two unpaired electrons in an S = 1 ground state. If we assume that the large magnetic moment is due to a paramagnetic impurity, such as high-spin Fe(III), a con- tribution to the molar susceptibility of 1.72 x 10-3 EMU-M—l would be necessary to achieve the observed value of 4.85 BM. This would correspond to m15% of the total iron being present as high-spin Fe(III). If this were the 57Fe Mdss- case, an additional quadrupole doublet in the bauer spectrum would be expected, but was not observed. Octahedral molybdenum (IV) has a 3T1g ground state; thus an orbital contribution is expected due to the orbit- ally degenerate triplet ground state. Since the environ- ment around the molybdenum atom is highly symmetrical and an electron can occupy these degenerate orbitals, a circula- tion of the electron about the molybdenum is permitted, thus causing a large orbital contribution. Molybdenum (IV) complexes have a d2 electronic 48 configuration in which the shell is less than half full, thus the AL-S factor is positive, thereby causing a negative contribution to the effective magnetic moment. Most pseudo- octahedral molybdenum (IV) compounds have moments ranging from 1.9 - 2.8 BM. Typical values reported in the litera— 52 9 “eff = 3.02. If the [MoFe686(CO)18]2- complex is con- ture include: K2MoCl6, = 2.28; K2Mo(NCS)6,5 ueff sidered as a simple pseudo-octahedral molybdenum (IV) com- pound, orbital contribution cannot explain the large mag- netic moment. A third possibility for an increase in the magnetic moment is that the compound may disproportionate to Mo(III) and Mo(IV) in the solid state. This has been observed in (PYH)2MO(NCS)6:6O (“eff = 2.45), in which after storing for a few months in sealed tubes gave rise to a magnetic moment of N3.5 BM. The other physical measurements, how- ever, do not agree with disproportionation in the [MoFe6S6- (CO)18]2- complex. Another possible contribution for the large moment may be due to temperature-independent—paramagnetism; which has been observed in one molybdenum (IV) complex: MoO[py(anil)2]- 60 . . _ 2, (diamagnetic, Heff - 0.8 BM). C1 In order to examine the possibility that the large moment observed for [MoFe6S6(CO)l8]2- was due to an unusually large temperature-independent-paramagnetism, variable temperature 49 magnetic susceptibility measurements were taken. The tem- perature—dependent susceptibility data were found to con- form to Curie-Law behavior, X = C/(T + 0), c = 3.54 EMU- K°M-l and 0 = -1.47 K. The resulting plots of Xm XE T and l/Xm _s T are shown in Figures X and XI. In the plot of 1/Xm _s T, a slight downward trend is observed at higher temperatures. This trend is probably caused by a small contribution of temperature-independent-paramagnetism.61 It seems likely that this contribution is not very large since the deviation from Curie-Law behavior is very small. G. Electrochemistry The electrochemical behavior of [MoFe686(CO)l8]2— was examined by using polarographic and cyclic voltammetric techniques. All measurements were made in acetonitrile solutions containing 1.5 mM [MoFe6S6(CO)18]2- and 0.1 M [BuuNJECIOu] as supporting electrolyte. They were re- corded relative to a standard calomel reference electrode. The polarography techniques, employing a dropping mercury electrode, proved to be the most useful methods for studying the electrochemical behavior of this complex. Well—formed polarographic waves were obtained, which al- lowed reasonable assignments for the electrochemical data. In cyclic voltammetry, a glassy carbon electrode was used instead of the more conventional platinum electrode. With platinum, the electrode surface was discolored after 50 .Huz.:zm use Ex soc noes: .saoaa ofipocmms ox ooo.N m CH pocflmppo septum Uflaom on» an oLSmeoQEop mo soaoocsu a we mmagoovammoaoz1mA2momamv do aeaaaoaoooomsm oaoocmez .x ossmaa 51 x banana CE m1a§anm .1. 00 0Nm . 00m 0¢N .00N .0w. . 0m. l 0.0 _.0 #0 00 0.0 NO 52 .Uaowm owuocmwe ox ooo.N m CH pocHMppo .EmapocmmsmpwdlpeopcwdopcfllopzpwngEop Op 036 ma macaw Ugmzczop pzwfiam .opzpmgmgsop mo coHpocsm m mm .Aalzzm.zv H .m@ Aoovmmmomoz1NAZmpmmmv mo zpflfiflpfipdoomsm vapocmms mo empo>CH .Hx osswaa 53 Hx assuaa 05 222368. 0mm .ON.m ON ON OON ow. . o.N_ d 1 00. 54 a single scan due to problems associated with the adsorp- tion of sulfur-rich complexes onto the platinum surface. Due to this adsorption problem, usually the cathodic wave was of reasonable shape and magnitude, but the anodic wave was featureless. In using the glassy carbon electrode, several runs were necessary to condition the electrode surface, after which the waves became of a more reasonable shape. It was also necessary to clean the electrode surface prior to each scan. It is not unreasonable to assume that the only reason the dropping mercury electrode system gave better results is due to the continual generation of a fresh electrode surface with each new drop. Significant electrochemical data appear in Table VII. An example of a differential pulse polarogram for the first and second reductions is shown in Figure XII, and examples of cyclic voltammograms are shown for both the oxidation and first reduction processes in Figure XIII. Examination of the electrochemistry of this complex over the potential range of 0.0 to -2.0 volts revealed two electrochemical reductions, the first of which is reversible. Plots of log [i/(iD—i)] 1s voltage for the complex gave slopes near the theoretical value (59 mV) for a single, one-electron reversible reduction. Since the value of the diffusion current (iD) is 55 IH\IN oHnHwLm>9H > smH.o+ COprpon oHprLo>mLLH > ow.HI COHuozpop .UCN Im\IN mHnHmngo.H > mmH.HI :oHpUSBoL .pmH mmooopm mm>.o MNH+ mNzH.o+ III lllll IIII IIII IIIII +zmpmmm . . III lllll IIII IIII lllll m N N m new 0 mm + oomH o+ A can mo mom adv :oHpmpon ..... II- II--- emu- mae.a- ea.e a.am- eme.HI + ameaaa IIIII III IIIII III owe.HI IIII IIII IIIII AmcmmNmoNzommzmv .pog .ch mmw.o mOHI NmH.HI NHHI mNN.HI mw.N o.mmI mmH.HI +zmpmmm m . I . I I . I . . I . I m N N m tam o NNH me H mHH NHN H Nm N : Hm :mH H A sum mo mom Lav .pog .pmH do a Em w a .9%>fiw A>Vna A>sv A>V ”a <10 .wmmw :3 m sea mu m x3 o\o H .pHo> .ozo zcawLNOLmHom zcdmpmommHom mom ”.Nom mesm .me0 zoo: HpC®>Hom .INHwHHoovomooaoz1 sou some HaoHEozooseoon .HH> oHoae 56 . m: INH 8on a m 0 @021 00 mcoa .pospon OOH l“ m Q PH 0 HHX mhijm 57 ON. HHx ossmaa ”50> m..- u 58 Figure XIII. Cyclic voltammograms of the first reduc- tion process (tOp), and the oxidation 2. process of [MoFe6S6(CO)18] . Scan rate: 100 mV/sec. 59 “0.5 '|.O '-|.5 VOLTS l l l ‘03 0.0 + O. 5 VOLTS Figure XIII 60 dependent upon concentration of the complex ion, a rela- tionship is observed when comparing the 1D values for other compounds, see Table II. The diffusion current value of the compound is near 2.5 pA/mM, which correlates well to that of the iron-sulfur tetramers (1D 2 2.97 uA/mM for SPh and id = 2.45 uA/mM for OPh). Since the tetramers have a 2-charge, this then also supports the formulation of [MoFe s (00) 12' for the com ound 66 18 p ' The reduction of the compound at -1.l96 V is a re— versible 2-/3- process. The reduction at -l.80 V was found to be due to a multi-electron process on the grounds of an extremely large i value (7.97 uA/mM). D In comparison to the reduction potentials for Fe282(CO)6, see Table VIII, the potentials for the compound are very different as expected. The compound contains the Fe282(CO)6 unit already reduced by two electrons as compared to the Fe2S2(CO)6 itself. With comparison to [CH3SFe(CO) see Table VIII, a 312, compound where the unit is reduced, it can be concluded that the 2nd reduction of the compound, (-1.80 V), may correlate to the first reduction of the Fe282(CO)2_ unit (—l.9 V). The second reduction of the Fe282(CO)§- unit (-2.5 V) is well be- low any of the observed reduction potentials for the compound. A quasi-reversible electrochemical oxidation was found at +0.137 V, which correlated to the 2-/l— process. These results suggest that it may be possible to isolate the 3- and possibly the 1- species. 61 Table VIII. Electrochemical Data for Other Compounds. i Slope d Compound Solvent Process (mV) (pA/mM) Ref. [FeuSu(SPh)u]2- MeCN lst red -62 2.97 63 2nd red -50 2.24 [FeuSu(OPh)u]2- MeCN lst red -52 2.45 63 2nd red -50 2.75 Com ound E1 (V) E2 (V) Ref 9 1/2 1/2 ° Fe2S2(CO)6 -0.49 —1.87 (ill defined) [CH3SFe(CO)3]2 -1.9 -2.5 (ill defined) 64 62 H. X-ray Structural Studies Excluding magnetic susceptibility measurements, all other evidence suggests the proposed structure shown schematically in Figure XIV. It has approximate D3d sym- metry with an octahedral sulfur environment around the molybdenum. Attempts to isolate X—ray quality single crystals have met with little success. Different mono— and dication salts were used to try to achieve the best crystal size and shape. Only two different salts enabled the growth of reasonably sized single crystals. l,2-Ethylenebis(triphenylphos- phonium) gave needle like crystals, but contained severe surface defects and gave poor X-ray diffraction. Benzyltriethylammonium yielded hexagonally shaped flat crystals which diffracted, but not well enough to allow collection of a complete data set. It is possible that due to the almost spherical sym- metry of the compound, the MoFe6S6 octahedra are disordered and randomly oriented within the crystal lattice. A pos— sible solution to this problem is to remove the spherical symmetry by substitution of one or more of the carbonyls with another ligand, thus overcoming the disorder. Phosphine ligands are known to react with Fe2S2(CO)665 and undergo ligand exchange to afford the complexes Fe2S2(CO)5L and Fe2S2(CO)uL2 (L = P(C6H5)3). Initial experiments in which the [MOFe6S6(CO)l8]2- 63 O r 9‘ III -2- 3 c c’ \‘fflt 'I ‘ ‘§D~kg5”sk. )“C25c) Os 8’ \ \c I I 5 w che- —S 1: CEO 0" fly I c C FeVC§ S7 \\\ /// [\ ~O o O 47 VA _. C) J Figure XIV. Schematic representation of the proposed struc- ture for [HoFe6S6(CO)18]2_. Compound has been isolated as (523311)+ and (Ph3PCH2CH2— PPh3)d+ salts. 64 cluster was treated with triphenylphosphine yielded no substitution products. From the X-ray data that has been obtained, it has been determined that iron is bonded to molybdenum through sulfur linkages, but little else can be hypothesized. Further work is necessary to obtain better crystals in order to complete the crystal structure determination. VI. CONCLUSIONS AND PLANS FOR FUTURE WORK A novel molybdenum-iron—sulfur cluster has been prepared and partially characterized. The available data suggest the formulation of [MoFe686(CO)l8]2_ for the complex. As such, it apparently represents the first example of a new structural class of molybdenum-iron-sulfur cluster (the others are the linear and cubane clusters discussed earlier). To eliminate any speculation, a crystal structure de- termination is absolutely necessary. Future work will be focussed on growing better crystals in order to reach this goal. When the structure has been determined, the next step will be to study the reactivity of the cluster. Its use as a possible starting material to prepare new clusters more relevant to the iron-molybdenum cofactor should prove to be very interesting. Another student in the re- search group (G. Lilley) has recently developed a mild method for oxidative decarbonylation of M-S-CO clusters under reducing conditions. Application of this method, which involves treatment with thiolate-disulfide mixtures to this compound,shou1d lead to a new branch of molybdenum— iron-sulfur cluster synthesis. 65 REFERENCES 10. 11. 12. 13. 14. Schrauzer, G. REFERENCES N. Chemistry 29, 13 (1977). Orme-Johnson, W. H.; Davis, L. C.; Henzl, M. T.; Averill, B. A.; Orme-J ohnson, R. R.; Mfinck, E.; Zimmerman, R. Components and pathways in biological nitrogen fixation, in; Recent Developments in Nitrogen Fixation, (ed.) Newton, W. E. ; Postgate, Jr .; Rodriguez-Barrucco, C.; 131, New York-London, Academic Press (1977). Gillum, W. O. ; Mortenson, L . E.; Chen, J.-S.; Holm, R. H. J. Amer. Chem. Soc. 99, 584 (1977). Shah, v. K.; USA 75, 3249 Palmer, 0.; M Mortenson, L. Smith, B. E.; Rawlings, J.; Zimmermann, R. Chem. €53, 10 Mdnck, E.; Rh Brill, W. J.; 32 (1975). Antonio, M. R M. J.; Groh, Brill, W. J. (1977). ultani, J. S.; E. Arch. Bioc Lang, C. Bioc Proc. Natl. Acad. Sci., Cretney, W. C.; Zumft, W. C.; hem. Biophys. 153, 325 (1972). hem. J. 137, 169 (1974). Shaw, V. K.; ; Mfinck, E.; 01 (1978). odes, H.; Orme Shaw, V. K. B Chisnell, J. R.; Brill, W. J.; Orme-Johnson, W. H. J. Biol. —Johnson, W. H.; Davis, . C.; .; Teo, B.-K.; S. E.; Lindahl L iochim. Biophys. Acta. 490, Orme-Johnson, W. H.; Nelson, , P. A.; Kauzlarich, S. M.; Averill, B. A. J. Amer. Chem. Soc. 104, 4703 (1982). Wolff, T. E.; Holm, R. H. J Berg, J. M.; . Amer. Chem. Warrick, C.; Hodgson, K. 0.; Soc. 100, 4630 (1978). Teo, B.—K.; A $9. 1454 (197 Wolff, T. E.; Holm, R. H.; Wolff, T. E.; Holnl, R. li.; 5454 (1979). Wolff, T. E.; 174 (1981). verill, B. A. 9). Berg, J. M.; J. Amer. Chem. Biochem. Biophys. Res. Commun. Hodgson, K. 0.; Frankel, R. B.; Soc. 101, 4140 (1979). Berg, J. M.; Frankel, R. B. Berg, J. M.; 66 Power, P. P.; Hodgson, K. O.; J. Amer. Chem. Soc. 101, Holm, R. H. Inorg. Chem. 20, 15. 16. l7. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 67 Christou, G.; Garner, C. D.; Mabbs, F. E.; King, T. J.; J. Chem. Soc., Chem. Commun. 740 (1978). Holm, R. H. Chem. Soc. Rev. 00, 455 (1981). Christou, G.; Garner, C. D.; Mabbs, F. E.; Drew, M. G. B. J. Chem. Soc., Chem. Commun. 91 (1979). Acott, S. R.; Christou, G.; Garner, 0. D.; King, T. J.; Mabbs, F. E.; Miller, R. M. Inorg. Chim. Acta, 00, L337 (1979). Christou, G.; Mascharak, P. K.; Armstrong, W. H.; Papaefthymiou, G. C.; Frankel, R. B.; Holm, R. H. J. Amer. Chem. Soc. 000, 2820 (1982). Garner, C. D.; Acott, S. R.; Christou, G.; Collison, D.; Mabbs, F. E.; Miller, R. M.; in "Current Perspec- tives in Nitrogen Fixation", Gibson, A. H.; Newton, W. E., eds., Australian Academy of Science: Canberra, 40, (1981). Christou, G.; Collison, D.; Garner, C. D.; Mabbs, F.E.; Petrouleas, V. Inorg. Nucl. Chem. Lett. 07, 137 (1981). Wolff, T. E.; Berg, J. M.; Power, P. P.; Hodgson, K. 0.; Holm, R. H.; Frankel, R. B. J. Amer. Chem. Soc. 00%, 5454 (1979). Wolff, T. E.; Berg, J. M.; Power, P. P.; Hodgson, K. 0.; Holm, R. H.; Inorg. Chem. 00, 430 (1980). Armstrong, W. H.; Holm, R. H. J. Amer. Chem. Soc. 000, 6246 (1981). Armstrong, W. H.; Mascharak, P. K.; Holm, R. H. J. Amer. Chem. Soc. 000, 4373 (1982). Wolff, T. E.; Berg, J. M.; Holm, R. H. Inorg. Chem. 00, 174 (1981). Armstrong, W. H.; Mascharak, P. K.; Holm, R. H. Inorg. Chem. 0% 1699 (1982). V Cramer, S. P.; Hodgson, K. 0.; Gillum, W. 0.; Mortenson, L. E. J. Amer. Chem. Soc. 000, 3398 (1978). Cramer, S. P.; Gillum, W. 0.; Hodgson, K. 0.; Mortenson, L. E.; Stiefel, E. I.; Chisnell, Jr.; Brill, W. J.; Shaw, V, K. J. Amer. Chem. Soc. $00, 3814 (1978). Huynh, B. H.; Mfinck, E.; Orme-Johnson, W. H. Biochim. Biophys. Acta 000, 192 (1979). 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 68 Tieckelmann, R. H.; Silvis, H. C.; Kent, T. A.; Huynh, B. H.; Waszczak, J. V.; Teo, B-K.; Averill, B. A.; J. Amer. Chem. Soc. 000, 5550 (1980). Silvis, H. C.; Averill, B. A. Inorg. Chim. Acta. :5, 657 (198i). Coucouvanis, D.; Simhon, E. D.; Baenziger, N. C.; Stremple, P.; Swenson, D.; Kostikas, A.; Simopoulos, A.; Petrouleas, V.; Papaefthymiou, V. J. Amer. Chem. §99. 192. 1730 (1980). Coucouvanis, D.; Simhon, E. D.; Swenson, D.; Baenziger, N. C. J. Chem. Soct, Chem. Commun. 361 (1979). Coucouvanis, D.; Simhon, E. D.; Baenziger, N. C. J. Amer. Chem. Soc., 19%: 6644 (1980). Coucouvanis, D.; Baenziger, N. C.; Simhon, E. D.; Stremple, P.; Swenson, D.; Simopoulos, A.; Kostikas, A.; Petrouleas, V.; Papaefthymiou, V. J. Amer. Chem. S90. 00%, 1732 (1980). Tieckelmann, R. H.; Averill, B. A. Inorg. Chim. Acta. Ag, L35 (1980). Kepert, D. L.; Mandyczewsky, R. Inorg. Chem. Vol. 7, No. 10, 2091 (1968). Emptage, M. H.; Zimmerman, R.; Que, Jr., L.; Mdnck, E.; Hamilton, W. D.; Orme-Johnson, W. H. Biochim. Biophys. Acta. 12, 495 (1977). Hieber, W.; Gruber, J. Z. Anorg. Allg. Chem. 0%, 296 (1958). Seyferth, D.; Henderson, R. 8.; Song, L. Organometallics g, 125 (1982). Hieber, H.; Beck, W. Z. Anorg. Allg. Chem., 000, 265 (1965). Stiefel, E. I. The Coordination and Bioinorganic Chem- istry of Molybdenum in Progress in Organic Chemistry, Volume 22, 57, New York (1977). Allen, E. A.; Brisdon, B. J.; Fowles, G. W. A. J. Chem. Soc. 4531 (1964). Lilley, G.; Averill, B. A. unpublished results. Allen, E. A.; Feenan, K.; Fowles, G. W. A. J. Chem. Soc. 4531 (1964). 47. 48. 49. 50. 51. 52. 53. 54. 56. 57. 58. 59. 60. 61. 62. 63. 69 Davies, C. W. The Conductivity of Solutions, John Wiley and Sons, 11. New York (1933). Scovell, W. M. and Spiro, T. G. Inorg. Chem. 00, 304 (1974). Lever, A. B. P. Inorganic Electronic Spectroscopy, Elsevier, 273 (1968). Horner, S. M.; Tyree, S. Y., Jr. Inorg. Chem. 0, 568 (1963). Carmichael, W. M.; Edwards, D. A.; Walton, R. A. J. Chem. Soc. (A) 97 (1966). van den Bergen, A.; Murray, K. S.; West, B. O. Austral. J. Chem. 00, 705 (1972). Gfitlich, P.; Link, R.; Trautwein, A. Mdssbauer Spec— troscopy and Transition Metal Chemistry, Springer- Verlag Berlin Heidelberg New York 19 (1978). Gibb, T. C.; Greatrex, R.; Greenwood, N. N. J. Chem. Soc. (A) 1663 (1967). Herber, R. H.; Kingston, W. R.; Wertheim, G. K. Inorg. Chem. 0, 153 (1963). Mayerle, J. J.; Frankel, R. B.; Holm, R. H.; Ibers, J. A.; Phillips, W. D.; Weiher, F. Proc. Natl. Acad. Sci. USA 70, 2429 (1973). Dickson, D. P. E.; Cammack, R. Biochem. J. 000, 763 (1974)- Kostikas, A.; Petrouleas, V.; Simopoulos, A.; Cou- couvanis, D.; Holah, D. G. Chem. Phys. Lett. 00, 582 (1976). Horn, C. J.; Brown, T. M. Inorg. Chem. 0%, 1970 (1972). Mitchell, P. C. H.; Williams, R. J. P. J. Chem. Soc. 4570 (1962). Midollini, S.; Bacci, M. J. Chem. Soc. (A) 2964 (1970). Earnshaw, A. Introduction to Magnetochemistry Academic Press, London, New York, 101, (1968). Cleland, W. E.; Holtman, D.; Sabat, M.; Ibers, J. A.; Averill, B. A. J. Amer. Chem. Soc., in press. 64. 65. 70 Dessy, R. E.; Stary, F. E J. Am. Chem. Soc., 00(30, Rossetti, R.; Gervasio, G. -3 King, R. B.; 471 (1966). Waldrop, M. and Stanghellini, P. L. Inorganica Chim. Acta. 30, 73 (1979). "14111111111111111111111111117 3 1293 03015 5765