.7 ’1 @333 . M1,. 1 33331135133335.3333 $61. $5431.55. 3., gifiggifigis. , 31x.“ 5314:; i '3 11,35" >1 Wk '31 ‘1 l 3.1. , _ .42: ' 3* 33.11 . .41. 1.” “3121333“ ”313‘ . 333} it"! (123%? 3 .:i‘ 5,333: 3131‘ “ ‘3 hkrfikfi'u‘g , 3‘1 . In, 1 3:37.33: .- ‘33 ‘3 \1? a“ x: ‘ \‘l . .2. =53» 13,3. m} .1. ““1“ “(31% 1.. ~~ ~' “ '1. ”3&3 flx‘fl 3333333 ~..l 1.1.x: $31,337: ‘- fiié‘n‘n‘ 3‘ £133 . Y. ' 1.71 I" . 1? 1:313:3-13-‘3'5 I“: ‘ k u, putrid” ( f t f" .. .rfi-é‘gfifi'} ‘ 1171.931151115 c " “um-$- » I 1 3433:3141 1 \Lfi,‘ '- ‘1‘ ‘ l A “kh- 12153.5 L331 nun“ ’ , 1. Mg}. 3‘ 1,, .~ 341....915“ ‘f‘n- ‘4 . “as 1.5 . 313M 1:: - v19”; 7% ‘EL‘L kwjtlxl‘ls 53w. . ~ 33 ‘3 :23: , 2. ~13; ‘x‘r \- $5133 1 , ‘1 s '1»)- :55‘5. ,‘hgcugm LE . . .. .. ‘35?“ gtgggi‘l‘fiifi. .. ‘ w ‘1 t: .' . ww Kilhi‘ .1 1%}: nm at? 1 " 3:11. 3133f“ 3 , r2 . 3 ~ :1 ‘1‘. 3 «1‘3.» .. :‘3. ,. ‘ . -- ..:.2..»~s=~:1‘ 5-; .4 ll ‘5“ "3' 1-2:;2~§&u5,"a ' "3‘ mm! “W .22. ' 13,535,. ,,.V_ 5 5 '50., 3’ 3.. . 137:; $3“ 1.51%“..‘Ezf‘1‘ .‘Séfi , haw“: 1533553” “531.113.43.33“ 133-“ 13351331. 3 :33. :5“ “$3351. ‘01 .. V1? N {5th 1-4..“ ”13'. ‘ . ‘ 33% “5.3+ f. 3.31.} e: .‘1‘; ‘ of"; “5115,4855 3113‘ 1} n W3C": \‘ , 113144 MAE”; 5 3e , . ~1m"? ”3.21.3134- “353%“ ~31; 11.1». ‘5“ 5,5 . . .33» J M? W. .1‘ ‘7'»: ‘, W“ P’“ n .1 ‘11.; ,1 $35553; -1‘1k .33.; $1.»: JLEJ - 12-13 . .7 WMKQS; 33.?Ely31 3.3 I 31% 3333.3 3131:? H. 3 '1 33‘?“ 3 SS {15,5 ;'_ ' 11.1911}: . ”FL fi?"{\ M3" 3‘ “3:; J3 h 3‘ ”3313;312:533 ““3333" 3:1: «. 3 . .. 1 ‘ ,5 €13.35.» .: {5,13 “31.1% .311 3 :31.- .1 ,. 253:; m «.333. 333.3313 - MM '35. .. ‘13311 133.31 ., m l ‘5’ ME @5523.“ ‘5‘ i. 1\.:.\.'3 a. M kVVVrE: V‘s-0,; <31 3.»... . ~53 mi ‘xrm ‘5‘“;‘1‘ \. :V‘S: W»? “‘7 ‘EEE‘ “(HAE‘vVRHE-‘iifih 51 k ”3:333“ 33-“ §fit 3:331 - mi: 1:“ ,. 441%}:sz L “y’gmxxv .,.;1 -. . 3;." 31.1.1); ~de1 v CS. 534‘ ‘ 5:74:31 “\qui' «Eng-{fix r23; .muu..VV 1931 *gufird‘é? in), 3% {Exéflait .. ‘3“: Y1 ‘ L ‘ “1. ‘19:“ \‘EIVV ‘50:} 1. ”E‘VQK 2"!” Lu. J 1. ~ “3:71:63. 1... 1»..3~‘3.“c a‘mtt‘g, W 1.1 ~ ‘ ' 1%“ VV “Er-M 1 “iii-"1;“ Qua-.1... “lily“. fit‘ )1 , ~. " K911143542 u; ‘03" C‘Q‘l‘gt . ’31.“. ‘30. “$1115.th “.511“ h“‘7"“r5331,V‘Nfim'ci' v-u “831'?!“ “l 5 ma... ”xx-1.3m N V \‘ $1.}.uQ1RJX. 3 q. A \\~ xi «3.- ":xatct‘ 1 1361 -‘ .. 1.1.3:. ’2‘“ W V v33 :3 1,, £51 353.3% . {122.5% ~3v5§31313h .11" " I?" H'HESIS This is to certify that the dissertation entitled The Synthesis, Characterization and Reactivity of Iron—Sulfur and Mblybdenum-Iron—Sulfur Complexes with Phnoxide Terminal Ligands presented by Walter Edward Cleland, Jr. has been accepted towards fulfillment ofthe requirements for Ph . D . degree in Chemistgz Major professor E K Datew MSU is an Affirmative Action/Equal Opportunity Institution 0 12771 hi LIBRAny "ionic”. Sta te ufl‘Versity ———- MSU LIBRARIES \, RETURNING MATERIALS: Piece in book drop to remove this checkout from your record. FINES wiii be charged if book is returned after the date stamped beiow. THE SYNTHESIS, CHARACTERIZATION AND REACTIVITY OF IRON—SULFUR AND MOLYBDENUM—IRON—SULFUR COMPLEXES WITH PHENOXIDE'TERMINAL LIGANDS By Walter Edward Cleland, Jr. A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry l98u ‘3-“ --—=‘ is/chf§¢fi% ABSTRACT THE SYNTHESIS, CHARACTERIZATION AND REACTIVITY OF IRON—SULFUR AND MOLYBDENUM—IRON—SULFUR COMPLEXES WITH PHENOXIDE TERMINAL LIGANDS By Walter Edward Cleland, Jr. The phenoxide—ligated tetranuclear iron—sulfur clusters [FeuSu(OAr)u]2- (Ar=Ph, p—Tol) have been synthesized by re— action of [FeuSuCluj2— or [FeuSu(SR)u]2— with NaOAr or HOAr, respectively. The X—ray crystal structure of (EtuN)2[FeuSu— OPh)u] has been determined and shows a short Fe—O distance (mean 1.865(17)A). Optical spectral features of these com— plexes are blue shifted compared to corresponding bands in the arenethiolate complexes, while the magnetic properties remain essentially unchanged. Isotropically shifted phenyl proton resonances are observed in the l H NMR spectra. These shifts are approximately twice as large as corresponding shifts observed for the arenethiolate analogs. The 57Fe Mossbauer spectrum has been obtained and consists of a single quadrupole doublet. Electrochemical data show that substitu— tion of thiophenoxide by phenoxide results in negative shifts for the first and second reduction potentials of the [Feusu]2+ "" ' - =_-.. . .7, ” ‘*"'-\ IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIll-IIIIII__________———I , ~La, Walter Edward Cleland, Jr. core. The phenoxide complexes react with electrophiles such as acyl halides and thiophenol to yield the halide and thio— phenoxide substituted iron—sulfur tetramers, respectively. The binuclear iron—sulfur clusters [Fe2s2(OAr)u]2_ (Ar= Ph, p—Tol, p—C6H4Cl) have been prepared by direct synthesis and by reaction of [Fe2S2Clu]2_ with NaOAr. The crystal structure of (BuuN)2[Fe282(OPh)u] has been determined. The results of a variety of other physical measurements including optical, 1H NMR and 57Fe Mossbauer spectroscopy, electrochem— istry and magnetic susceptibility are reported. Ligand ex— change reactions with electrophiles are also discussed. The phenoxide—ligated "double cubane” complex [MO2F3688— (SEt)3(OPh)6]3_ has been synthesizedby:reaction of [Mo2Fe6S8— (SEt)9]3_ with PhOH. Electronic and 1H NMR spectroscopy, electrochemistry and magnetic susceptibility data are reported. Ligand exchange reactions with electrophiles are discussed. General effects of phenoxide ligation to iron and the biological implications for tyrosyl coordination to iron— sulfur and molybdenum—iron—sulfur centers are discussed. To My Family: Alice, Walter and Rebecca ii ACKNOWLEDGMENTS I would like to express my sincere appreciation to Professor Bruce A. Averill for his advice and guidance throughout the course of this work. His insights concerning synthetic inorganic chemistry and the interpretation of physical data have been invaluable. I would like to thank the following individuals for their contributions to this work: Dr. E. Munck and Dr. T. A. Kent for obtaining the 57Fe Mossbauer spectra; Pro— fessor G. C. DeFotis and D. A. Holtman for assistance in obtaining variable temperature magnetic susceptibility data on (EtuN)2[FeuSu(OPh)u]_ Professor James A. Ibers, Dr. M. 3 Sabat and Professor E. Sinn for the crystal structures of (EtuN)2[FeuSu(OPh)u] and (BuuN2[Fe2s2(0Ph)u]. I would like to thank Craig Silvis, Paul Lamberty, Jim Davis, Mark Antonio, Susan Kauzlarich and my other coworkers for their friendship and advice on various matters through— out my graduate career. I would like to acknowledge the Department of Chemistry for providing a teaching assistantship, and special thanks to Professors Brubaker, Eick, Chang and Allison for serving on my Committee. I would also like to acknowledge the University of Virginia for support during my last year. iii Special thanks to Mrs. P. Warstler for her expert typing of this dissertation and to J. Kotarski, B. Draper, and L. Ruiz for their work on the graphics. iv TABLE OF CONTENTS Chapter LIST OF TABLES. LIST OF FIGURES LIST OF ABBREVIATIONS I. INTRODUCTION. II. EXPERIMENTAL . A. Materials and Methods B. Preparation of (RAN)2[FeuSu(OAr)u] (Ar = Ph, p—Tol) Salts. l. (EtuN)2[FeuSu(OPh)u]. 2. (BuuN)2[FeuSu(O—p—Tol)u]. C. Preparation of (RuN)2[Fe282(OAr)u] (Ar = Ph, p—Tol, p—C6H4Cl) Salts. l. (BuuN)2[Fe2SZ(OPh)u]. 2. (EtuN)2[Fe282(OPh)u]. 3. (MeuN)2[Fe282(O—p—Tol)u]. A. (EtuN)2[Fe2S2(O—p—C6HuCl)u] D. Preparation of (EtuN)2[Fe2SEClu]. E. Preparation of (Et3NCH2Ph)3- [MO HFe6S8(U-SEt) (OPh)6]. F. Preparation of the Fe— S Long Wave— length Compounds. . . . . . V IIIIIIIIII-_______LLLLiLILLLLLLLLILA______LL_LI Page ix xi 22 22 27 28 28 29 31 31 31 31 32 32 33 33 Chapter 2. 3. MeuN/O—prol/LW EtuN/O-p-Tol/LW G. Ligand Substitution Reactions 1. Reaction of [FeuSu(OPh)u]2- with PhSH Reaction of [FeuSu(OPh)u]2- with PhCOCI Reaction of [Fe2S2(OPh)u]2- with PhSH Reaction of [Fe282(OPh)u]2- with PhCOCl Reaction of [M02Fe6S8(u-SEt)3- (OPh)6]3' with PhSH Reaction of [Mo Fe S (u—SEt) - 3_ 2 6 8 3 (OPh)6] with PhCOCI Reaction of Fe-S Long Wavelength Compound with PhSH. . . H. Physical Measurements III. RESULTS AND DISCUSSION. A. [FeuSu(OAr)u]2— (Ar = Ph, p—Tol). l. \IONUWJIUO Synthesis X-ray Structure Electronic Absorption Spectra Magnetic Susceptibility Proton Nuclear Magnetic Resonance 57Fe Mossbauer Spectra. Electrochemistry. vi Page 34 3A 35 35 36 36 36 37 37 37 38 A0 A0 A0 Al 5A 59 63 76 80 Chapter 8. Ligand Exchange Reactions 9. Fe—S Long Wavelength Compounds. 10. Summary 2- _ B. [Fe2s2(0Ar)ul (OAr - Ph, g-Tol, B—C6HuCl) 1. Synthesis 2. X-rayStructure. 3. Electronic Absorption Spectra A. Magnetic Susceptibility 5. Proton Nuclear Magnetic Resonance 6 57Fe Mossbauer Spectra. 7 Electrochemistry. 8. Ligand Exchange Reactions 9 Summary 0. [M02Fe688(sst)3(orh)6]3‘. 1. Synthesis 2. Electronic Absorption Spectra 3. Magnetic Susceptibility A. Proton Nuclear Magnetic Resonance 5. Electrochemistry. ON Ligand Exchange Reactions 7. Summary vii Page 86 98 121 12A 124 126 129 13“ 135 1A3 1A7 151 153 15A 15A 156 156 159 165 169 170 Chapter Page IV. CONCLUSIONS. . . . . . . . . . . . . . . . . . 172 REFERENCES... . . . . . . . . . . . . . . . . . . . 178 viii Table II III IV VI VII LIST OF TABLES Page Bond Distances (A) and Angles (deg) for (EtuN)2[FeuSu(OPh)u]. . . . . . . 48 Comparison of Structural Parameters for Compounds with the [FeASA]2+ Core. . . . . . . . . . . . . . . . . . . . 52 Electronic Spectral Features, Mag- netic Moments, and Isotropic Shifts of Phenoxide Protons of [FeuSu(OAr)u]2_ Complexes . . . . . . . . . . . . . . . . . 57 Comparison of Relative Isotropic Shifts for Various Metal-Sulfur Clusters in CDBCN Solution. . . . . . . . . 75 Electrochemical Data for [FeuSu— (OAr)u]2- Complexes . . . . . . . . . . . . 83 Chemical Shifts for Phenoxide and ThiOphenoxide Ligands in [Feusu- (OPh)u_n(SPh)n]2- Species (N = O,l,2,3,u) in CD3CN Solution at 22°C. . . . 95 Ratios of Equilibrium Constants for PhS—PhO Exchange in CD CN 3 Solution at 22°C. . . . . . . . . . . . . . 97 ix Table VIII XIX Electronic Spectral Features and Isotropic Shifts of Phenoxide Pro— tons of [Fe2SZ(OAr)u]2—, [M02Fe688- (SEt)3(OPh)6]3— and Fe-S Long Wave- length Complexes. Electrochemical Data for [Fe2s2— (OAr)u]2_, [Mo2Fe688(SEt)3(OPh)6]3-, and Fe—S Long Wavelength Complexes. Comparison of Average Bond Distances (A) and Angles (deg) for Compounds with the [Fe282]2+ Core Page 102 113 128 Figure 'J-j_4 LIST OF FIGURES Page Schematic of four types of Fe-S centers that occur in non—heme iron sulfur pro— teins . . . . . . . . . . . . . . . . . . A Hypothetical view of electron trans— fer through the nitrogenase system. Elec— tron transfer within the enzyme is from the AFe—HS cluster of the Fe protein (MgATP)2 complex to the p—clusters to FeMo—co to substrate. . . . . . . . . . . . 7 Possible models for the P cluster in— volving oxygen ligation at three vertices of a “Fe-AS core. . . . . . . . . . . . . . ll Schematic of structural models pro— posed for the FeMo—cofactor . . . . . . . . 15 Schematic of the structurally charac— terized clusters possessing the MoFe3Su cubane core including the "double cubane" and the ”single-cubane" complexes . l8 Schematic of the structurally charac— terized "linear" MoS2Fe clusters. . . . . . 20 xi Figure 10 11 12 Page A portion of the FeMSA(OPh>A core, showing 50% probability ellipsoids, the atom labelling scheme, and interatomic distances . . . . . . . . . . . . . . . . . A3 A stereoscopic View of the [FeASA(OPh)A]2— ion. Probability ellipsoids are drawn at the 50% level. The hydrogen atoms are not included. . . . . . . . . . . . . . A5 Stereodiagram of the unit cell of (EtuN)2[FeuSu(OPh)u] down the b axis. The 30% probability ellipsoids are shown. The hydrogen atoms are omitted for clarity . . . . . . . . . . . . . . . . . . A7 Electronic spectra of the [FeuSu(OPh)u]2_ and [FeASA(SPh>AJ2- ions in acetonitrile solution at 22°C. . . . . . . . . . . . . . 56 Magnetic susceptibility of solid (EtuN)2[FeuSu(OPh)u] (O) as a func- tion of temperature, compared to curves calculated for single J values of —160 (-—-), —175 (———), and —190 (...) cm‘l. . . 62 Proton magnetic resonance spectra (250 MHZ) of (EtuN)2[FeuSu(OPh)u] in d3-MeCN solution at various temperatures. Peaks from protons Figure 13 1A 15 16 of the cation are indicated by Q, sol- vent by S, and unidentified impurities by X. Chemical shifts are in ppm from internal MeuSi (TMS). Proton magnetic resonance spectra (250 MHz) of (BuuN)2[FeuSu(O-p—Tol)u] in d3—MeCN solution at various tem— peratures. Peaks from protons of the cation are indicated by Q, solvent by S, residual water by W, and un— identified impurities by X. Chemical shifts are in ppm from internal MeuSi (TMS) Temperature dependence of isotropic— ally shifted ligand proton reson— ances of (EtuN)2[FeuSu(OPh)u] (o) and (BuuN)2[FeuSu(O-p_—Tol)u] (O) in d —MeCN 3 57Mossbauer spectrum of polycrystal— line (EtuN)2[FeuSu(OPh)u] at 4.2 K in zero applied field Cyclic voltammetry and differential pulsed polarography scans for (EtuN)2- [FeuSu(OPh)u]. Solvents and scan rates are indicated xiii Page 66 68 7O 78 82 Figure 17 18 19 20 Page Optical spectra of a 3 mM solution of (EtuN)2[FeuSu(OPh)u] in MeCN treated sequentially with 0-5 equivalents of PhSH at 22°C. Optical pathlength: 0.2 mm. . . . . . . . . . . . . . . . . . . 90 Proton magnetic resonance spectra (250 MHz) of a 10 mM solution of (EtuN)2[FeuSu(OPh)u] treated sequen— tially with the indicated amounts of PhSH at 22°C. Peaks from protons of the cation are indicated by Q, solvent by S, residual water by W, and un- identified impurities by X. Chemical shifts are in ppm vs. MeuSi internal standard (TMS). . . . . . . . . . . . . . . 92 Electronic absorption spectra of EtuN/O-p-Tol/LW (———) and (BuuN)2— [FeuSu(O-p-Tol)u] (--—) in aceto- nitrile solution at 23°C. . . . . . . . . . 101 Proton magnetic resonance spectra (250 MHz) of EtuN/O-p-Tol/LW in d3-MeCN solution at various temperatures. Peaks from protons of the cation are indicated by Q, solvent by S, and un- identified impurities by X. Chemical xiv Figure 21 22 23 2A shifts are in ppm from internal MeuSi (TMS) Temperature dependence of isotrop— ically shifted ligand proton reson— ances of EtuN/O—p—Tol/LW in d —MeCN 3 solution. 57Fe Mossbauer spectrum of polycrystal— line EtuN/O—p—Tol/LW at “.2 K in zero applied field Cyclic voltammograms (above)and dif— ferential pulse polarograms (below) of EtuN/O—p—C6HUC1/LW and MeuN/O—p—Tol/LW in NMP. Cyclic voltammetry was per— formed at glassy carbon electrode at 100 mV/s. DPP was performed at DME at 5 mV/s Proton magnetic resonance spectra (250 MHZ) of a solution of EtuN/O—p-C6HuCl/LW treated sequentially with the indicated amounts of PhSH at 22°C. Peaks from protons of the cation are indicated by Q, solvent by S, and residual water by W. Chemical shifts are in ppm vs MeuSi internal standard (TMS) XV Page 105 107 111 115 118 Figure 25 26 27 28 Electronic absorption spectra of the [Fe2s2(OPh)u]27 (———) and [Fe2s2(SPh)u]2— (-—-) ions in acetonitrile solution at 22°C. Proton magnetic resonance spectra (250 MHz) of (BuuN)2[Fe282(OPh)u] in d3-MeCN solution at various tempera— tures. Peaks from protons of the cat— ions are indicated by Q, solvent by S, and unidentified impurities by X. Chemical shifts are in ppm from in- ternal TMS. Proton magnetic resonance spectra (250 MHZ) of (MeuN)2[Fe2S2(O—p—Tol)u] in d —MeCN solution at various tempera— 3 tures. Peaks from protons of the cations are indicated by Q, solvent by S, and unidentified impurities by X. Chemical shifts are in ppm from internal TMS. Temperature dependence of isotropi— cally shifted ligand proton resonances of @uuN)2[Fe2S2(OPh)u] (o) and (MeuN)2- [Fe2S2(O—p_—Tol)u] (C) in d —MeCN. 3 Page 131 137 1A1 Figure 29 3O 31 32 33 34 Page 57Fe Mossbauer spectrum of polycrystal- line (BuuN)2[Fe282(OPh)u] at 4.2 K in zero applied field. . . . . . . . . . . . . 145 Cyclic voltammograms for (BuuN)2— [Fe282(OPh)u], (MeuN)2[Fe2S2(O-p—Tol)u] and (EtuN)2[Fe282(O—p—C6HuCl)u] in NMP at glassy carbon electrode. Scan rates are 100 mV/s. . . . . . . . . . . . . . . . 149 Electronic spectra of the [Mo2Fe6S8(SEt)3- (orh)6]3' (———) and [Mo2Fe6s8(SEt)9]3' (--—) ions in MeCN at 22°C. . . . . . . . . 158 Proton magnetic resonance spectra (250 MHz) of (EtBNCH2Ph)3[Mo2Fe6S8— (SEt)3(OPh)6] in d3—MeCN. Peaks from protons of the cations are indicated by Q, solvent by S, and unidentified impurities by X. Chemical shifts are in ppm from internal TMS. . . . . . . . . . 162 Temperature dependence of isotropically shifted ligand proton resonances of (Et3NCH2Ph)3[Mo2Fe688(SEt)3(OPh)6] in d3—MeCN. . . . . . . . . . . . . . . . . 164 Cyclic voltammogram and differential pulsed polarogram for (Et3NCH2Ph)3— [MO2Fe6S8(SEt)3(OPh)6] in MeCN. Scan xvii Figure 35 Page rates are 100 mV/s (CV) and 5 mV/s (DPP) . . . . . . . . . . . . . . . . . . . 164 Schematic of known Fe—S and Mo—Fe—S clusters with phenoxide ligands to iron. . . . . . . . . . . . . . . . . . . . 176 xviii Solvents: MeCN EtCN MeOH i—PrOH Et2O THF DMF DMA DMSO NMP Reagents: PhCOCl PhOH PhSH p—TolOH p-ClC6HuOH Miscellaneous: EtuN+ + MeuN LIST OF ABBREVIATIONS acetonitrile propionitrile methanol isopropanol diethyl ether tetrahydrofuran N,N,—dimethylformamide N,N,—dimethylacetamide dimethylsulfoxide N—methylpyrrolidinone benzoyl chloride phenol thiophenol para—cresol para—chlorophenol tetraethylammonium tetramethylammonium BM SCE MCD EPR ATP TMS S2-g—xylyl tetra-n-butylammonium para—tolyl para-chlorophenyl phenyl ethyl methyl Bohr magneton saturated calomel electrode magnetic circular dichroism electron paramagnetic resonance adenosine triphosphate tetramethylsilane ortho—xylyldithiolate INTRODUCTION Metalloproteins and metalloenzymes are biomolecules that contain metal ions in integral stoichiometries. These metal ions are often the site of electron transfer or catalysis and are therefore termed 'active sites'l. These biomolecules are essentially elaborate coordination com— plexes, which often possess unusual structural and physical properties. For this reason, metallobiomolecules have received increased attention by inorganic and physical chemists, which has led to the emergence of the inter— disciplinary field of bioinorganic chemistry. One aspect of this field involves the "synthesis of relatively low molecular weight complexes, which, ideally, are obtainable in the crystalline form and approach or duplicate the biological unit in terms of composition, ligand types, structure and oxidation levels"2. Here the thesis is that synthetic metal complexes, termed 'synthetic analogues', can approximate or 'model' the properties of the related biomolecule and thus yield useful information concerning its biological structure and function. Possibly the most significant success of this approach has occurred in the area of the non—heme iron—sulfur proteins3_5, which contain as their active site iron and acid labile sulfur in FenSm units where n = l, m = O or n = m = 2, 3, or 4 as shown in Figure 1. In addition, evidence for an F8384 center in inactive aconitase has recently been reported3b. Synthetic analogues of the one, two and four iron sites have now been prepared and structurally and physically characterized“. These complexes usually contain alkyl or aryl thiolate terminal ligands in place of cysteinyl residues in the protein and have proven to be credible structural and electronic models of the protein active sites. The metalloenzyme nitrogenase, which performs the dif— ficult and useful function of reducing dinitrogen to am— monia for use in biosynthesis, was first isolated in 19606 but remained relatively obscure to chemists until the mid— l970's. Nitrogenase is found only in prokaryotic micro— organisms. These microorganisms can either form symbiotic aissociations with higher plant life or live independently; vaideties are known that grow either aerobically or anaero— bjxcally. Studies of the enzyme show that it possesses sever— al. properties that have stimulated a now keen interest by inLDrganic chemists. For example, nitrogenase fixes di— nitxrogen under extremely mild conditions (ambient temperature anti pressure) compared to the industrial process (450°C armi 200—300 atm). Therefore, a long-range goal of research irl this area is the development of a catalytic system for diJiitrogen reduction based on the metal centers of Figure 1. Schematic of four types of Fe—S centers that occur in non—heme iron sulfur proteins. SR RS SR l }: ,S\ F/ e e R3/£§5.e>” nitrogenase. In addition, the presence of novel metal— 7 sulfur clusters with peculiar physical properties offers a particularly challenging problem to the inorganic chemist. Nitrogenase consists of two oxygen-sensitive proteins: the iron protein (Fe protein), containing iron and acid labile sulfur; and the molybdenum—iron protein (MoFe protein), containing molybdenum, iron and acid labile sulfur. Both proteins are essential for activity in the presence of an external reductant and MgATP. The exact mechanism of dinitrogen reduction remains as yet unknown, 8,9 but available evidence suggests the pathway of electron flow shown schematically in Figure 2. Electron transfer is from external reductant (not shown), usually reduced ferredoxin or dithionite, to a (MgATP)2 complex of the Fe protein, to the MoFe protein to substrate. Thus, it is the MoFe protein which actually reduces the substrate. The iron protein has a molecular weight of m60,000 and contains four iron atoms and four sulfides per molecule. Cluster displacement and transfer techniques together with EPR results have shown that the iron and sulfide comprise 10,17 a single 4Fe—4S cluster Evidence that the Fe protein binds two moles of MgATPll and that hydrolysis of ATP is only observed during transfer of electrons to the MoFe proteinl2 seems to indicate that the function of the protein is to act as a transducer, coupling the energy released by ATP hydrolysis to electron transfer. Figure 2. Hypothetical View of electron transfer through the nitrogenase system. Electron transfer within the enzyme is from the 4Fe-4S cluster of the Fe protein (MgATP)2 complex to the P-clusters to FeMo—co to substrate. m ohsmflm £28m. one: 520$ on. N omocmoonzz The MoFe protein has a molecular weight of m220,000 and contains two Mo atoms, 32i2 Fe atoms and 32:3: per mole- cule. It contains two types of metal—sulfur clusters7’13—l6: a soluble, low molecular weight iron—molybdenum cofactor (FeMo—co) of unknown structure, and a variant of the normal 4Fe—4S clusters termed the P—clusters. Approximately half of the iron and acid-labile sulfur present in the MoFe protein constitute four 4Fe—4S clusters that can be removed from the protein in addition to FeMo-co. This has been shown in three ways: by cluster displace- ment experiments using fluorinated thiols and 19F NMR 17,18. spectroscopy for identification of Fe-S clusters by cluster transfer techniques using low molecular weight apoferrodoxins to accept displaced Fe—S clusters followed 19 by reduction and EPR quantitation ; and by quantitation of the EPR spectra of the MoFe protein in partially denatur- ing solvents20. The 57Fe Mossbauer spectra of the MoFe protein as iso— lated can be separated into a magnetic portion due to FeMo— co centers and a diamagnetic portion. The diamagnetic por- tion consists of two quadrupole doublets in a 3:1 ratiolu’ 21’22. Parameters of the less intense doublet (AEq = 3.02 mm/s, 6 = 0.69 mm/s) are typical of high spin Fe2+ in a tet— rahedral sulfur environment, and the species responsible has thus been called component Fe2+. The more intense doublet has 6: 0.64 mm/s, which is in the range for Fe2+, but has an unusually small quadrupole splitting, AEq = 0.81 mm/s; this iron has been termed component D. These two components behave as if they were diamagnetic in strong magnetic fields. It has been suggested that the only way to explain the apparent diamagnetism of the iron atoms of components Fe2+ and D is that they exist in spin coupled 14 13,14,— units, termed the P—clusters Further evidence 23-26, including combined EPR, Mossbauer and MCD measure— ments on dye—oxidized MoFe protein has led to the conclu— sion that the P—clusters are a variant of the normal 4Fe— 4S cluster in the all ferrous (F6484)O oxidation state. Several suggestions have been made as to the means by which the protein could differentiate three of the iron atoms from the fourth in a 4Fe-4S core. The first, and most obvious, is to exchange an oxygen or nitrogen donor terminal ligand for the cysteinyl mercaptide common to 'normal' clusters. Since ferrous iron has a low affinity 27 for saturated amine ligands , tyrosyl phenolate or gluta— mate or asparate carboxylate seem to be the most reason— able choices. Schematic representations of these models are depicted in Figure 3. The other possibilities include the addition of a fifth ligand to three of the iron atoms or protein imposed distortions of the 4Fe—4S cubane core from its normal geometry. At present, no evidence is available to determine which if any of these possibilities is correct. 10 Figure 3. Possible models for the P cluster involving oxygen ligation at three vertices of a 4Fe—4S COPG . 11 L\ “ oquEG m opswfim /on_ m l\ll|l..0u\ \ : mos» . n _ o I .S U 12 Because of the limited data available on the P—clusters, efforts to apply the synthetic analogue approach have been few. To date, no synthetic clusters with oxygen or nitrogen terminal ligands have been isolated. Johnson and Holm28, however, have generated in solution the [F9484- (OAc)u]2_ ion and reported briefly on some of its prOperties. - 0 2 3 and Recently, the complexes [Fe4s4(SC6H4‘9'OH)4J [F s SPh) c t ) 12' 29 h' ° en 4( 2(S2 NE 2 2 , w lCh contain one and two five-coordinate iron sites, respectively, have been pre— pared. The other half of the iron and sulfide and all of the molybdenum present in the MoFe protein comprise two iden- tical iron-molybdenum cofactor units. This cofactor3l_3u can be removed from the protein and exhibits spectroscopic properties very similar to those of the MoFe protein from 32b show the presence of which it came. Analytical data six to seven iron atoms and approximately 8 or 9 sulfides per molybdenum. The low temperature (6 - 20 K) EPR spec- 22335 shOWS an aXial Signal With g-Values 2.0, 3.8, trum and 4.3, which has been attributed to one of the two Kramer's doublets of a S = 3/2 spin system. Isolated FeMo—co ex- hibits a complex magnetic 57Fe Mossbauer spectrum nearly identical to the magnetic portion of the Mossbauer spectrum 20336 of this Of the MoFe protein. Detailed analysis SDectrum indicates the presence of spin coupled S = 3/2 units containing m6 iron atoms each. In addition, 57Fe 13 37a 0 ENDOR experiments n 57Fe enriched MoFe protein indicate the presence of no fewer than six nonequivalent iron sites within the S = 3/2 centers. 95Mo ENDOR experiments on 95Mo enriched MoFe-protein suggest that the molybdenum is a diamagnetic even—electron ion with even formal valency. Molybdenum and iron K-edge extended X—ray absorption fine structure spectroscopy (EXAFS)1uum2provided the only structural information available to date on FeMo-co. Molybdenum K-edge results indicate 4 S atoms at m2.35 A, 2—3 Fe atoms at 2.72 A and 1—2 additional 3 atoms at m2.47 A as nearest neighbors to M038-u2. Iron K-edge results indicate an average of l.3:l.0 O (or N) atoms at 1.8 A, 3.4:1.6 s atoms at 2.25 A, 2.3:o.9 Fe atoms at 2.66 A and 44. Combined EPR, 57Fe Moss- 0.41:0.1 Mo at 2.76 X from Fe bauer and EXAFS results clearly indicate a novel cluster type of structure for FeMo-co. Several speculative struc- tural models have been proposed for FeMo—co based on the data outlined above. These models are shown in Figure 4. It should be emphasized that these are speculative models, and no synthetic compounds with these Specific structures have been prepared. Because it is believed that FeMo-co is the site of 7,16,45—48 and because of dinitrogen binding and reduction the relative lack of known complexes containing Mo, Fe, and 8:, most synthetic work to date has been directed toward the preparation of new Mo-Fe—S clusters as models for 14 Figure 4. Schematic of structural models proposed for the FeMo-cofactor. l5 _s\ /S—/Fe/ S 8/8 /Mo;—S | \Fe/ \hlflo/ \e/ --S We 73 / x I I \ S S I III / \ S—Fe Fe—S s; 1‘s RFe “S 84 F” \ Fe F'e — — eI \ | /Fe\- -S\ |,3 —|-/Fe\ 'Fe Fe,— S —Fe lFe— S \Fe' 'stor5\,_—e/ [‘se‘Nk»{;‘ // ‘st'rl ‘~s,'r \\ S (I \s L III N SR RS ' ,el 1 z’ ,5 S\\Fe’\ |\ \ e—S S—Fe’ RS|\ '/5\R.’SR \ S S / \ \IFe l l'=,,SR / I \ I \ \s/‘\Fe§,s \ / S l s \ / \Mo’ Fe—S S—Fe~ SR L L/ |\L I‘ L I m Figure 4 l6 FeMo-co. Since 1978, two classes of Mo-Fe—S cluster have been prepared and characterized. The first consists of complexes containing the MoFe3Su "cubane" coreuo’u9-7O, including [Mo2Fe688(SEt)9]3—, [Mo2Fe7SB(SEt)l2]u- and [MoFe3Su(R3,6-cat)]3-; see Figure 5. These complexes are prepared from simple starting materials in "self assembly" reactions similar to the approach used for synthesis of the ferredoxin analogues. Although the stoichiometry, spec- trOSCOpic properties and EXAFS results confirm that these complexes are not synthetic FeMo-co analogues, they repre- sent, at present, the closest approach to such an entity. The second class of Mo—Fe—S clusters consists of a series of compounds based on the MoS Fe unit. These 2 clusters have an extended "linear" array of metal atoms compared to the cubane class. Examples of "linear" clus— 71-83 include: [MoFeSuX212— (X = SAP: OAr, Cl)’ ters [MoFeZS6X2]3_ (x = SAr; x2 = s5), [Fe(MoSu)2]3- and [MoFe2SuCluj2- (shown in Figure 6). These clusters also do not represent synthetic FeMo—co analogues due to in- appropriate stoichiometry and Mo oxidation state and to discrepancies in Mo—S distances obtained from EXAFS analysis. A variety of factors, then, have stimulated interest in oxygen ligation of metal sulfur clusters. These include the incorporation of oxygen ligands into the proposed struc- ture of the P-clusters as Shown in Figure 3 and recent re- 44,84—86 ports , including Fe edge EXAFS data, which suggest 17 Figure 5. Schematic of the structurally characterized clusters possessing the MoFeBSu cubane core including the "double cubane" and the "single- cubane" complexes. L 1 3- RS SR > s 2 s 4 RS /e\- —; \ SR 8: Fe/>S> M zS> M —S<\Fe;/S Fe—S/ \S/ SéFe / R \ RS SR —' RS \. ,. S g e_ we .32 p.43 5 \s / R R RS L '13- RS SR \F O. S :- >(SR NP-S Fb"$ \\ / S—'R! \ SR 5'? IQ‘ Fb-—S O O EtS’\/ X \ / S/ m2\m/\m2\m so2.\aemo\/\2 \oz/ \odx \oz/ d/ \ou. , \omz/ -m m\ m .m.. m - m 25.. m .m.. m .o m m a 2 \ / \ 2 \ x/ \ of \m \Ol 02 \mm mm .. < / \ .. .2 21 that oxygen ligands are bound to the iron atoms within the FeMo—cofactor. In addition, preliminary X—ray dif- fraction studies of the recently discovered 3 Fe ferr — doxin's87’88 indicate that one of the ligands is an oxygen donor ligand (ligand L in Figure 1). It should be noted that with two exceptions, none of the synthetic analogues of the iron sulfur sites nor any member of either class of Mo—Fe—S clusters has' been isolated with oxygen ligands to iron. Those exceptions are brief reports on the iso— lation of the [Fe(OAr)u]_ ion89 and the [MoFeSu(OAr)2]2_ ion8l, which appeared subsequent to the beginning of the research described herein. Thus, an investigation of oxygen ligation of metal sulfur clusters was in order. Phenolate was the oxygen ligand of choice as it is a ligand available to nature in the tyrosine moiety. Further, since the thiophenolate analogs of many of the metal sulfur clusters are known, the use of phenolate would facilitate a more direct comparison of the change in physical properties upon exchange of an oxygen atom for a sulfur atom. This dissertation describes the synthesis, characteriza— tion and reactivity of several oxygen ligated Fe—S and Mo—Fe— 8 clusters, including the first such cluster to be isolat— edgo, [Fe4S4(OAr)4]2_ (Ar = Ph, peTol). Also included are [Fe282(OAr)u]2_ (Ar = Ph, p-Tol, p—CéHuCI), [M02Fe6S8- (SEt)3(OPh)6]3_ and a compound of as yet unknown structure COntaining iron, sulfide and phenolate. II. EXPERIMENTAL A. Materials and Methods All operations were carried out under an atmosphere of pure, dry dinitrogen unless otherwise specified. Dinitro— gen was purified by passage over hot BASF R—3—ll catalyst and supported phosphorous pentoxide (Aquasorb). Solvents and reagents were either distilled under an inert atmos— phere or thoroughly degassed by repeated evacuation and flushingvfljtipure dinitrogen prior to use. Acetonitrile was purified by the following three step procedure. Reagent grade acetonitrile was refluxed for several hours and distilled from calcium hydride. The distillate was stirred for 12 h with 5 g/L each of an— hydrous sodium carbonate and potassium permanganate, fol— lowed by distillation at room temperature using a dry ice/icopropanol cold trap. Final distillation from phos— phorous pentoxide produced acetonitrile of acceptable purity for synthesis as well as for electrochemistry and other physical measurements. Propionitrile was purified by a similar procedure using 2 g/L potassium permanganate rather than 5 g/L. Tetrahydrofuran and diethylether were purified by distillation from sodium/benzophenone ketyl or lithium 22 23 aluminum hydride. Methanol and isopropanol were distilled from magnesium methoxide and aluminum isopropoxide, respec— tively. N-methylpyrrolidinone was distilled from calcium hydride and then from barium oxide. Spectroscopic grade DMA and DMF were stored over 4 A molecular sieves. Benzoyl chloride was distilled from barium oxide at room temperature using a liquid nitrogen trap. PhOH, p—TolOH, and p-ClC6HuOH were either sublimed twice or sublimed followed by vacuum distillation. Anhydrous sodium phenolates were prepared in one of two ways: by reaction of the approrpiate phenol with sodium methoxide in anhydrous MeOH, followed by re— moval of the solvent in vacuo, addition of MeCN and re— peated evaporation to dryness; or by reaction with metallic sodium in THF, followed by filtration and evaporation to dryness. Thiols, lithium sulfide and all other reagents were of commercial reagent grade and used without further purification. Tetra—n—butylammonium chloride was prepared from a 10% aqueous solution of the hydroxide by neutralization with 6N hydrochloric acid followed by removal of water in vacuo. Addition of MeCN to the oily mass and repeated evaporation to dryness resulted in a waxy material which was stored under an inert atmosphere. Tetraethylammonium phenoxide was prepared by the metathesis of sodium phenolate with tetraethylammonium chloride in MeCN. Filtration of the reaction mixture to remove NaCl followed by volume reduction 24 and cooling to -20°C caused separation of white crystals, which were collected by filtration, washed with diethyl ether and dried in vacuo. Tetra-n-butylammonium per- chlorate and tetraethylammonium perchlorate were prepared by addition of a Slight excess of perchloric acid to an aqueous solution of the corresponding tetraalkylammonium halide. The crude product was collected by filtration, washed with water, recrystallized twice from hot iePrOH and dried in vacuo. (EtuN)2[FeuSu(SPh)u] and (MeuN)2[FeuSu(S—t-Bu)u] were prepared by published proceduresgl’92. (EtuN)2[FeuSuClu] and (MeuN)2[FeuSuClu] were prepared from (EtuN)2[FeuSu- (SPh)u] and (MeuN)2[FeuSu(S-t-Bu)u], respectively, by reaction with PhCOCl as described93. Several new salts of previously reported iron—sulfur tetramer dianions were prepared. (EtuN)2[FeuSu(SEt)u] - This compound was prepared by a modification in the stoichiometry of the procedure reported by Christou and Garner for the preparation of [Fe4S4(SR)u]2— clusters92. Five, rather than four, equiva- lents of sodium alkyl thiolate were used in the preparation of the reaction mixture, which was stirred for 18 h. After filtration, the reaction mixture was treated with three equivalents of EtuN01 dissolved in a minimum of MeOH. The crude product was precipitated by removal of W70% of the solvent in vacuo, followed by addition of three volumes 25 of water. The crude crystalline product was collected by filtration and dried in a stream of dinitrogen, washed with THF and vacuum dried. Recrystallization was accomplished by dissolution of the crude product in a minimum of a 9:1 (vzv) mixture of THF/MeCN, filtration and slow cooling to -20°C. Typical yields of recrystallized product were 40-50%. UV-Vis and 1H NMR spectroscopic prOperties due to the di- anion were essentially the same as those reported for other 94,95. salts of this dianion This compound is extremely soluble in a variety of polar organic solvents, making it an attractive candidate for experiments requiring high tetramer concentrations. (BuuN)2[FeuSu(SEt)u] - This compound was prepared using the same stoichiometric modification described above. The filtered reaction mixture was treated with three equiva- lents of BuuNBr in MeOH, which caused immediate separation of the crude microcrystalline product. The material was collected by filtration, washed with MeOH and vacuum dried. The crude product was then dissolved in hot (m60°C) DMF, filtered and cooled slowly to —20°C, which caused separation of well formed crystals. The yield was 43%. UV-Vis and 1H NMR spectrOSCOpic properties were consistent with its formulationgu’95. (BuuN)2[FeuSu(S-i-Pr)u1 - This compound was prepared by the same modification of a literature procedure des- cribed above. Three equivalents of BuuNBr in MeOH was 26 added to the filtered reaction mixture followed by addi— tion of one volume of water which caused Separation of the product as microcrystals. The crystals were collected by filtration, washed with i—PrOH and dried in vacuo. The crude product was recrystallized by dissolution in hot (N60°C) DMA, filtration, and slow cooling to -20°C. The yield was 40%. UV—Vis and 1H NMR spectroscpic properties were the same as those previously reported for salts of this dianiongu’95. (EtuN)2[FeuSu(S-t-Bu)u] — This compound was prepared by the same stoichiometric modification of a literature procedure described above. Addition of three equivalents of EtuNCl dissolved in water to the filtered reaction mix— ture caused separation of the product as an amorphous solid which was collected by filtration, washed with i-PrOH and dried in vacuo. The crude product was crystallized by dis- solution in a hot (W60°C) mixture of 3:1 (vzv), MeCNzDMA, filtration and slow cooling to -20°C. The yield was 70%. UV-Vis and 1H NMR spectroscopic properties were the same as those previously reported for this tetrameric di— aniongu’gs. (EtuN)2[Fe2S2(SPh)u196’97 and (EtuN)2[Fe2S2Clu]92 were prepared as described. (EtuN)FeC1u98 was prepared as described and recrystallized from MeCN/i-PrOH. Ammonium99 and benzyltriethylammoniumu9 salts of tetrathiomolybdate and (Et NCH2Ph)3[Mo2Fe688(SEt)9149 were prepared by pub- 3 1ished procedures. 27 B. Preparation of (RMN)2[FeuS“(OAr)“] (Ar = Ph, p— Tol) Salts. These complexes were prepared by either of two methods. Typical examples of each are described in detail below. Method 1, from (RAN)2[FeQSH(SR')u] (R' = Et,4t:Bu). To 5.5 g (6.5 mmol) of (EtuN)2[FeuSu(SEt)u] dissolved in 100 mL of MeCN was added a solution of 26 g (260 mmol) of PhOH in 75 mL of MeCN. The reaction mixture was evap— orated in vacuo at 30°C to i 50 mL volume, diluted with 100 mL MeCN, and again evaporated to a final volume of i 50 mL. The color of the solution at this point was an orange-brown, rather than the green-brown of the starting material. Addition of 300 mL of i—PrOH to the filtered solution resulted in separation of the product as dark orange—brown microcrystals, which were collected by filtra— tion, washed twice with i—PrOH, and vacuum dried. Re— crystallization was accomplished by dissolution of the crude product in a minimum volume of MeCN at room tem— perature, filtration, addition of N3 volumes of i—PrOH, and slow cooling to —20°C. Typical yields are m80% of analytically pure product after one recrystallization. Concentration of the mother liquors and slow cooling to —20°C affords a second crop (~10%). Method 2, from (RHN)9[Fe“S“01“]. A mixture of 1.46 g (2.52 mmol) of (Et4N)2[Fe4S4Cl4] and 1.23 g (10.6 mmol) of 28 anhydrous NaOPh was taken up in 80 mL MeCN. An immediate color change from brown to orange-brown was observed, accompanied by formation of a white precipitate. The re- action mixture was stirred for 2 h, and the precipitate was removed by filtration and washed with a small portion of MeCN. The combined filtrate and wash were concentrated in vacuo until a dark crystalline precipitate began to form. Addition of several volumes of THF and cooling to -20°C caused complete precipitation of the microcrystalline product, which was collected by filtration, washed with THF or i—PrOH, and recrystallized as in Method 1 to give analytically pure product in 85% yield. The product was found to be identical in all respects with that obtained by Method 1. 1- fluflgtfim (0Ph>ul_~ This compound was prepared as described above by Method 1 or 2. Analytical data were obtained on a sample prepared by Method 1. Anal. Calc'd for C40H60N204S4Fe4: C, 48.80; H, 6.14; Fe, 22.69; N, 2.84; O, 6.50; S, 13.03. Found: C, 48.49; H, 6.26; Fe, 22.08; 0, 6.93; S, 13.14; m.p. 177°C (d). This compound was prepared by Method 1. Anal. Calc'd for C60H90N2O4S4Fe4: C, 56.96; H, 7.97; Fe, 17.66; N, 2.21; 29 S, 10.14. Found: C, 56.28; H, 7.79, Fe, 16.97; N, 1.92; S, 10.04; m.p. 132°C (d). 0. Preparation of (RHN12[Fe2S2(OAr)u] (Ar = Ph, p-Tol, p—C6E4Cl) Salts. These complexes were prepared by either of two methods. Typical examples of each are described in detail below. Method 1. Direct Synthesis. To 1.3 g (8.1 mmol) FeCl3 dissolved in 50 mL MeCN was added a solution of 2.2 g (8.1 mmol) of BuuNCl dissolved in 25 mL MeCN. The solution changed color at this point from orange-brown to light yellow. This solution was then added to a slurry of 3.75 g (32.3 mmol) of NaOPh in 50 mL of MeCN. The color of the supernatant solution changed immediately from light yellow to bright red. The reaction mixture was stirred for 30 min and the precipitate was removed by filtration. The filtrate was treated with 0.37 g (8.1 mmol) of Li S, and 2 the mixture was stirred for 18 h, during which time a slow color change to a much more intense red occurred. The reaction mixture was then filtered and evaporated in vacuo to a volume of <20 mL. Addition of 80 mL Et20 and cool— ing to —20°C for 8 h caused separation of the product as dark red microcrystals, which were collected by filtra— tion, washed with EtZO and vacuum dried. Recrystalliza- tion was accomplished by addition of 50 mL Et2O to the crude product followed by addition with stirring of several 30 1 mL aliquots of MeCN until all of the material was dis— solved. Filtration and Slow cooling to —20°C caused separation of large dark red prisms. Volume reduction of the mother liquors, addition of N1 volume of Et2O, and slow cooling to -20°C afforded a second crop (~10%). Typical yields are 40-50% of analytically pure product after one recrystallization. Method 2. From (RMN12LFE2S29141; A solution con- taining l g (1.73 mmol) of (EtuN)2[Fe2S2Clu] dissolved in 40 mL MeCN was added with stirring to a slurry of 0.84 g (7.26 mmol) of anhydrous NaOPh in 10 mL MeCN. An im- mediate color changefrom purple to deep red occurred. The reaction mixture was stirred for 0.5 h and filtered, and the filtrate was concentrated to half its volume. Slow addition of one volume of Et2O to initiate crystalliza- tion, followed by cooling to -20°C for 18 h, caused com— plete separation of the microcrystalline product, which was collected by filtration, washed twice with Et20 and vacuum dried. The crude product was recrystallized by dissolu— tion in 25 mL MeCN, slow addition of 20 mL Et2O, and slow cooling to —20°C. Concentration of the mother liquors and slow cooling to —20°C afforded a second crop. Total yield was 61%. 31 1. (BuuN)2[Fe2S2(OPh)41. This compound was prepared as described above by method 1. Anal. Calc'd for c 6H92N20u82Fe2: c, 65.09; H, 8.98; 5 N, 2.71; S, 6.22; Fe, 10.81. Found: C, 64.36; H, 8.71; N, 2.69; s, 6.30; Fe, 11.31. m.p. 182°C (d). 2. (EtuN)2[Fe2S2(OPh)u]. This compound was prepared as described above by method 2. Anal. Calc'd. for C4OH60N2O4S2F82: C, 59.40; H, 7.48; N, 3.46. Found: C, 59.68; H, 7.57; N, 3.79. 3. (M8uN)2[Fe2S2(O-B-T01)u]. This compound was prepared by method 1. Anal. Calc'd for 6H52N20u82Fe2: c, 57.45; H. 6.97; N, 3.72; S, 8.52; C3 Fe, 14.84. Found: C, 57.64; H, 6.67; N, 3.80; S, 8.12; Fe, 14.98. m.p. >225°C. 4. (FtuN)2[Fegs2(0-9-06Hu01)ul. This compound was prepared by method 1. Anal. Calc'd for CuoH56N20401uS2Fe2‘ C, 50.76; H, 5.96; N, 2.96; S, 6.78; Fe, 11.80. Found: C, 51.55; H, 6.09; N, 3.05; S, 7.99; Fe, 11.75. m.p. 138-1400C (d). 32 D. Preparation of (EtuN12EF92E2Qlul; This previously reported compound was prepared by a modification of method 1 above for the preparation of (RAN)2[Fe2S2(OAr)u] salts. This compound was also pre— pared using an alternative source of iron by a procedure analogous to method 1. A solution of 4 g (12.2 mmol) (EtuN)- [FeClu] in 100 mL MeCN was treated with 0.56 g (12.2 mmol) Li2S and stirred for 18 h. Subsequent workup of the re— action mixture proceeded as described in method 1. The yield was 51%. Anal. Calc'd for C16H40N2S2C14Fe2: C, 33.24; H, 6.97; N, 4.85. Found: C, 39.22; H, 8.10; N, 5.72. B. Preparation of (EtQNCHZPh)3[Mo2Fe6§8(SEt)3(0Ph)6l; To 3.16 g (1.65 mmol) (Et NCH2Ph)3[MO2Fe6S8(SEt)9] dis- 3 solved in 200 mL MeCN was added 14 g (148.8 mmol) PhOH dis- solved in 75 mL MeCN. The reaction mixture was evaporated in vacuo at 30°C to <50 mL, diluted with 150 mL MeCN and again evaporated to a volume of <50 mL. This dilution and evaporation step was repeated once, followed by addition of 300 mL THF. Slow cooling to -20°C caused separation of the product as black plates, which were collected by filtration, washed twice with THF and vacuum dried. Recrystallization was accomplished by dissolution of the crude product in %20 mL MeCN which was 25 mM in PhOH, filtration, addition 33 of m150 mL THF which was also 25 mM in PhOH, and slow cool— ing to -20°C. Typical yields were 50-60%. Anal. Calc'd for 081H111N3O6811FG6MO2: C, 46.30; H, 5.30; N, 2.00; S, 16.80; Fe, 15.90; Mo, 9.10. Found: C, 45.39; H, 5.553 N, 1.87; S, 16.72; Fe, 16.56; Mo, 8.10. m.p. 55°C (d). F. Preparation of the Fe-S Long Wavelength Compounds. l. EtuN/O-p-C6Hu01/LW. To 5.15 g (5.3 mmol) (EtuN)2[FeuSu(S—27Bu)u] dissolved in 250 mL MeCN was added 27.33 g (213 mmol) p—ClC6HuOH dis- solved in 100 mL MeCN. The resulting solution was evap- orated in vacuo to a volume of <75 mL, diluted with 200 mL MeCN, and again evaporated to a final volume of <75 mL. A color change from green-brown to orange-brown occurred dur- ing this step. Addition of 300 mL irPrOH and cooling to -20°C for 18 h caused separation of the product as microcrystals which were collected by filtration, washed with i—PrOH and vacuum dried. Recrystallization was accomplished by dissolution of the crude product in 100 mL MeCN, filtration, slow addition of 250 mL i-PrOH and slow cooling to —20°C. The yield was 0.91 g of dark brown crystals. Anal. Calc'd for (EtuN)2[F°uSu(O'E‘C6HuCl)u]’ C4OH56N2O4S4014F84: C, 42.80; H, 5.03; N, 2.50; S, 11.43; Cl, 12.64; Fe, 19.90. Found: C, 42.53, H, 5.02; N, 2.44; S, 12.30; C1, 13.49; Fe, 20.80. 34 2. MeuN/O-p—Tol/LW A solution containing 1 g (1.55 mmol) (MeuN)2[FeuSuClu] dissolved in 100 mL MeCN was treated with 0.85 g (6.54 mmol) NaO-p—Tol. An immediate color change from dull brown to orange—brown occurred. The mixture was stirred for 2 h, filtered, and the filtrate evaporated in vacuo until pre— cipitation of a crystalline solid began. Addition of ml50 mL i—PrOH and cooling to —20°C for 12 h resulted in separation of the microcrystalline product, which was col— lected by filtration, washed with i-PrOH and vacuum dried. Recrystallization was affected by slowly cooling a saturated solution of the product in warm (m35°C) EtCN to —20°C for 18 h. The yield was 0.2 g Anal. Calc'd for (MeuN)2— [FeuSu(O—p-Tol)u], C36H52N2O4S4Fe4: C, 46.56; H, 5.64; N, 3.02; S, 13.81; Fe, 24.06. Found: C, 45.32; H, 5.27; N, 3.14; S, 13.33; Fe, 23.30. 3. EtuN/O—p—Tol/LW. To 3.8 g (3.9 mmol) (EtuN)2[FeuSu(S-t—Bu)u] dissolved in 300 mL MeCN was added 16.96 g (157 mmol) p—TolOH in 50 mL MeCN. The volume was reduced in vacuo to <75 mL, diluted With 250 mL MeCN and again reduced to a final volume of <75 mL. Slow addition of toluene until a few droplets of dark oil formed, filtration, and slow cooling to —20°C Caused precipitation of a well formed crystalline solid, 35 which was collected by filtration, washed with THF and vacuum dried. UV-Vis and 1H NMR spectroscopic properties of this product were essentially identical, except for cation resonances in the NMR, with those of MeMN/O-p— Tol/LW. G. Ligand Substitution Reactions. 1. Reaction of [FeuSu(OPh)u]2- with PhSH. This reaction was monitored in separate experiments by 1H NMR and electronic spectral measurements described below. Two stock solutions, one 2.74 mM (EtuN)2[FeuSu(OPh)u] in MeCN, the other 1 M PhSH in MeCN, were prepared. An aliquot of tetramer solution was introduced into a quartz Optical cell with 0.2 mm path length, and a spectrum was recorded after each addition of each aliquot of thiol solution. Spectra were recorded within 5-10 min after each successive addition of thiol solution. Solution volume changes were negligible due to the high concentration of the thiol solution. A limiting spectrum was obtained on addition of >4 equivalents of thiol. Two stock solutions were made up in d3-MeCN: one 10 mM (EtuN)2[FeuSu(OPh)u] and the other 0.1 M PhSH. A 0.4 mL aliquot of tetramer solution was transferred to a 5 mm NMR tube fitted with a rubber septum. Spectra were taken 15—20 min after each addition of thiol solution. A 36 limiting spectrum was obtained on addition of >4 equivalents of thiol as shown by the appearance of resonances due to free PhSH. 2. Reaction of [FeuSu(OPh)u]2_ with PhCOCl. To 5.0 mL of a 7.42 mM solution of (EtuN)2[FeuSu(OPh)u] in MeCN was added 165 uL of 0.89 M PhCOCl in MeCN. The resulting solution was stirred for 0.5 h during which time a color change from orange-brown to a dull brown occurred. An optical spectrum of the solution showed 97% formation of [FeuSuClu12— 3. Reaction of [Fe2S2(OPh)u]2— with PhSH. To 5.0 mL of a 7.88 mM solution of (BuuN)2[Fe2SZ— (OPh)u] in MeCN was added 166 uL of 0.95 M PhSH in MeCN. The resulting solution changed color from red to maroon rapidly and was stirred for ml5 min. An optical spectrum of the solution was consistent with 98% formation of 2- [Fe2S2(SPh)u] . 4. Reaction of [Fe2S2(OPh)u]2— with PhCOCl. To 5.0 mL of 7.88 mM (BuuN)2[Fe2S2(OPh)u] in MeCN was added 0.7 mL of 0.89 M PhCOCl in MeCN. The mixture was stirred for 12 h. A gradual color change from red to purple took 37 place. An optical spectrum of the solution indicated 95% formation of [Fe282Clu]2_. 5. Reaction of [No2Fe6s8(SEt)3(0Fh)6]3‘ with PhSH. To 5.0 mL of 2.8 mM (Et3NCH2Ph)3[Mo2Fe688(SEt)3(OPh)6] in MeCN was added 94 uL of 0.95 M PhSH in MeCN. The solu— tion color changed rapidly from deep orange-brown to red- brown. An optical spectrum of the solution showed that 2- [Mo2Fe6S8(SEt)3(SPh)6] had been formed in 98% yield. 6. Reaction of [Mo2Fe6SB(SEt)3(0Fh)6]3' with PhCOCl. To 5.0 mL of 2.8 mM (Et3NCH2Ph)3[M02Fe6SB(SEt)3- (OPh)6] in MeCN was added 94 uL of 0.89 M PhCOCl in MeCN. The solution changed color from deep orange-brown to dark brown with a purple cast over the course of 0.5 h. An optical spectrum of the solution was consistent with 95% formation of [Mo2Fe688(SEt)3Cl6]3'. 7. Reaction of EtuN/O-p—C6H401/LW with PhSH. This reaction was monitored by 1H NMR spectrosc0py. Two stock solutions were prepared in d —MeCN: one was 3 5.0 mM EtuN/O-p-C6HuCl/LW assuming a formula weight of 1122.25, corresponding to (Et4N)2[Fe4S4(O‘B'C6H4Cl)4]’ and the other 0.1 M PhSH. A 0.4 mL aliquot of Etu/O- p—C6Hu01/LW solution was introduced into a degassed 5 mm 38 NMR tube fitted with a rubber septum. Aliquots of PhSH solution were added and spectra recorded at 15 min inter- vals. Complete reaction, as shown by the appearance of free PhSH signals, was reached at 6-8 equivalents of added PhSH. At >8 equivalents of added PhSH the spectra showed further reaction to form substantial amounts of [FeuSu(SPh)u]2—. H. Physical Measurements. All samples were handled under anaerobic conditions. Optical spectra were obtained on either a Cary 219 or a Cary 17 spectrophotometer. Proton NMR spectra were ob— tained on a Bruker WM-250 Fourier transform spectrometer equipped with a variable temperature unit. Room temperature magnetic susceptibility measurements were performed on an Alpha Faraday balance using Hg[Co(SCN)u] as calibrant. Var- iable temperature data were performed on an SHE Corpora— tion SQUID susceptometer operating at 2 k0 and on a modi— fied PAR 155 vibrating sample magnetometer operating at 10 kG. Electrochemical measurements were performed on a PAR 174 R polarographic analyzer equipped with an HP 4030 A XY recorder. Either DC polarography at a dr0pping mercury electrode or cyclic voltammetry at Pt flag or glassy carbon electrodes were performed. All solutions contained either 50 mM EtuN(C10u) or 50 mM BuuN(ClOu) as supporting electrolyte. Potentials were measured versus the 39 saturated calomel electrode and a Pt wire was employed as the counter electrode. Mossbauer spectra were measured by Dr. T. Kent and Professor E. Munck at the Grey Freshwater Biology Institute, University of Minnesota, Mavarre, Min— nesota. The Mossbauer spectrometer was of the constant acceleration type and has been described elsewhereloo. Isomer Shifts are reported versus metallic Fe foil at room temperature. Melting points were obtained in sealed capillaries in vacuo and are uncorrected. Elemental analyses were performed by Galbraith Laboratories, Inc., Knoxville, Tennessee. III. RESULTS AND DISCUSSION A. Lgeusu(0Ar),]2‘ (Ar = Ph, p—Tol) 1. Synthesis Addition of excess phenol to an acetonitrile solution of the alkyl thiolate cluster [Fe4S4(SR)4J2-’ R = alkyl, establishes an equilibrium whereby a small fraction of thiolates is displaced by phenolate (Reaction 1). [Fe4S4(SR)4]2— + xs ArOH Z [FeuSu(OAr)u]2- + 4 RSH (l) The original observation of this equilibrium101 suggested a route to the synthesis of phenolate substituted tetramers. By removal of solvent and volatile thiol (R = Et, E-Bu) in vacuo, the reaction can be driven to the right and complete substitution achieved. The resulting products, [FeuSu(OAr)u]2_, are isolated as their tetraalkylammonium salts in 80% yield after recrystallization. These complexes are stable in the solid state and in solution in the absence of dioxygen and water. In some cases, it is difficult to purify the phenolate tetramers by recrystallization because of their high solu- bility in polar organic solvents, especially the parent 40 41 phenol. This constitutes a drawback of the preparative method utilizing a large excess of phenol, which can make precipitation of the compound difficult. An alternative method of preparation employs a ligand exchange reaction between an appropriate salt of [Fe4S4C1412- and anhydrous sodium phenolate in acetonitrile (Reaction 2). [FeuSuClHJZ- + 4 NaOAr + [FeuSu(OAr)u]2— + 4 NaCl (2) This method makes use of the insolubility of NaCl in aceto- nitrile to drive the reaction to completion. Phenolate tetramers prepared by this method are of Similar purity and obtained in yields comparable to those found for the first method of preparation. 2. X-Ray Structure The crystal structure of (EtuN)2[FeuSu(OPh)u] consists of discrete cations and anions in an 8:4 ratio, respec— tively, per unit cell. The structure of the core Showing the numbering scheme and selected interatomic distances is presented in Figure 7. A stereoscopic View of the entire dianion is depicted in Figure 8, and stereosc0pic View of the unit cell along the b axis is shown in Figure 9. Selected interatomic distances and angles for the anion are given in Table 1. The tetraethylammonium cations are crystallographically ordered, possess normal angles, 42 Figure 7. A portion of the Feusu(OPh)u core, showing 50% probability ellipsoids, the atom labelling scheme, and interatomic distances. A w 43 Figure 7 44 Figure 8. A stereoscopic View of the [FeuSu(OPh)u]2- ion. Probability ellipsoids are drawn at the 50% level. The hydrogen atoms are not in- cluded. 45 Figure 8 46 Figure 9. Stereodiagram of the unit cell of (EtuN)2— [FeuSu(OPh)u] down the b axis. The 30% prob- ability ellipsoids are shown. The hydrogen atoms are omitted for clarity. 47 Figure 9 48 AmHVH.SOH coo: Amvam.ooa AmvmnfievohuAHVm MHVH.MOH Amvmlflswomlmmwm va.mOH Amvm-Ae on- H m . Afivm.ooa Amvmuxmvomuflfivm memmm.m Amvmuflcwwm AHVN.MOH AevmuflmvomuAmvm Asvmmm.fi coo: Amvwmm.m Afivmtflmvom Anew.mea Aevm-lmvos-xsvm A Smmm.s Ammvc-1 V Amammm.m imam-nmvoe lacs.mes Asvm-lmcoe-lmcn lacmmm.a Aamvo-lmco Ame om.m Aacm-lmcoe Ascm.mos Ascm-lmvoe-lscm lava m.a Amado-l cw Amvmem.m A cm-xmcoe lacs.mos Amvm-1mcos-xavm Amcamm.a Asaco-lmco Amvsam. imam-lmcoe lacs.mos Assn-lscos-lmcm Amcoam.m A cm-xscoe MMWWHMWfi MNWWHMMWMMHMMWW ...... locate o-o ...... Amcoam.m Amen-lscoa Asvmmm.m coo: ............ m-oe-m------------ Assamew.a coca “mammm.m Amen-xevoe onww.fi Aevonflevom Amvmom.m Aevmlxmvoa “Havoo.oo ANH nod coo: nevoom.a Amvouflmvom Amvmmm.m Afivmsfimvom Amavom.oo coo: onmow.H Amvouflmvoe Amvoem.m AmvmlAHVos Amvsm.oo Amvomnfizvomunavom onwem.fi flavouflavom Amvmm.oo Amvomnszomufimvom msom Amvos.om AevomuAmvomuAHVom uuuuuuuuu onom uuuuuuuuu Amvsm.ow szomnflmvomlmmvom Amvwm.oo AHVoEIAmVomnAmvom Amfivmms.m Am mow coo: “mamm.eo AHVoR-AmVoRIAeVom Aaavowm.m secs Amvoos.m coo: Amvmo.oe AmVoSIAHVoSIAmvom Amvmmm.m Aevmufimvm Amvmms.m AevomuAmvom Amvmo.oo Amvomnflavomuflevom Amvmsm.m AmvmnAme Amvoms.m AmvomuAmvom Ammvme.mm coo: Amvmom.m Aevmufifivm Amvmes.m ASVoEIAHVom Amvoo.mm Amvomnflevomuflavom Amvwwm.m Amvmufiflvm Amvmms.m Amvomuflavom Amvos.mm Advosuxmvomlfiavom Academo.m coo: mms.m coo: Amvmm.mm AmvomuAmvomnnsvom Amvzwo.m Aevmufimvm Amvome.m AsvomnAmvom Amvmm.mm Amvoa-xfivom-levoa Amvsos.m AmeIAOHVm Amvsss.m Amvoa-fiavoa IIIIIIIIIII omlomtomlllllllllll llilllllm....wllllllll Illlllllom....omllllllll 20Hz< .flefimmmoovemeomLNMZSSmL soc Amoov mon2< one Awm one 66—mco so moucoomwo 039 ohms: nmmms6>o mmsfio> do mamnE=c 6gp 6L6 mommgucwsca cw mocsmwdm .xnoz megpu NOF wucmgmwmmu Pop mocmdmtwmn .Pm mocmawmmmo vas6.6m ”Aeve_.N6 Amvom.am ”Aevae.le vase.mm ”A6v66.16 vame.6m MASVF_._6 rm-rm-rm 6.6p_ 6.e_1 1.m__ e.6__ x-6a-rm ..eo_ m.mop m.eol _.eep .m-6l-.m s.ms 6.6a m.ma w.mm 6a-rm-6a AmVe~.ee 116166.mm vaa_.ee ”Aevme.6m Amv__.oe sieves.6m levee.6m MASVF_._6 66-61-61 levemm.m 11N1666.m Aevwem.m “Amvsme.m Aevmmm.m ”Amvom6.m 1616mm.m ”Amvm66.m .m-tm m66.P cam.m mom.~ FmN.N x-6l Amvm_m.m ”Aevmmm.~ Amvm6N.N ”Aev66N.N va66N.N ”ASVA6N.~ Amvo_m.m ”Aevamm.m rm-6l levees.m ”Amvams.m Aev66A.N ”ANmeA.N Aevams.~ ”Amvems.m Aevwms.m M6AN16AA.N 66-61 eflexeaoveme6lgmfizeomg 6H6_66m66luwmzecmL amexeamvem66lgmmze6zu aheleamxcmveme6aummzeo6L 616e< \6ocopmwo .mcou +Nmememdu 62p sue: mocsoaeoo com msmumeosod Fessposcpm do cemwcoasou .HH o_hee 53 described above with regard to the structure of the [Fe4S4JZ+ core of [FeuSu(OPh)u]2— are also applicable to the other members of this family of complexes possessing the same core but different terminal ligands. Thus, the phenoxide tetramer provides additional evidence for the invariance of the structure of the [F8484]2+ core with various terminal ligands. The key structural feature of the [Fe4S4(OPh)4J2- ion is the presence of the Fe—O bond. The mean Fe-O bond dis- tance is 1.865 (17) A. It is difficult, however, to analyze the chemical significance of this bond length due to the lack of structurally characterized complexes with tetrahedral iron oxygen bonds. This distance is inter- 2‘ 81 (1.897) (13) A) and that in [Fe(OAr)]l— 89 (1.847 (13) A for Ar = mediate between that found in [MoFeSu(OPh)2] 2,3,5,6—MeuC6H, 1.866 (6) A for Ar = 2,4,6—Cl3C6H2). This trend is consistent with an increasing average Fe oxidation state in these complexes: 2+ for the MoFe dimer, 2.5+ for the phenoxide tetramer, and 3+ for the iron phenolate mono— mers. The range of these Fe-O distances is somewhat small, and less than expected based on trends in ionic radius. This suggests the presence of substantial covalent character in these bonds. Typical five and Six coordinate complexes with oxygen ligands show longer Fe-O distances than those observed for the monodentate phenoxide ligands in [FeuSu(OPh)u]2-. 54 103 04 Examples include [Fe(salen)Cl] and [Fe(salen)]20 l which have Fe-O distances of 1.88—1.92 A and [Fe(cate— 105 which has Fe-o distances of 2.02 A. Some cholate)3] of the difference between these Fe-O distances and that observed in [Feusu(OPh)u]2_ may be due to the expected decrease in bond distances between octahedral and tetra- hedral geometrle6. This effect,however, does not seem to fully explain the short Fe-O distances in [FeuSu(OPh)u]2- or [MoFeSu(OPh)2]2-. Recently it has been Shown that axial monodentate phenoxide coordination in certain five- coordinate Schiff base complexes, such as [Fe(saloph(CatH))] (CatH = catecholate) and [Fe(sa1en)HQ], (HQ = pehydro- quinone dianion), results in quite short Fe—O distances107 (1.828 (4) A and 1.861 (2) A, respectively). One might View coordination of phenolate to the trigonal FeS3 corner of a cubane as a situation somewhat analogous to the axial phenolate coordination in these five-coordinate Schiff base complexes. 3. Optical Spectra The Optical spectra of [FeuSu(OPh)u]2- and [F8484- (SPh)u12- are shown in Figure 10. Comparative spectral data are given in Table 3. The first observation to be made is that the spectra are qualitatively quite similar. Both exhibit a strong absorption maximum in the 400—500 nm range, with less well 55 Figure 10. Electronic spectra of the [Fe4s4(C)Ph)14]2 and [FeuSu(SPh)u ]2 ions in acetonitrile solu- tion at 22°C. 56 4'0 I l I I l I 3.8 — 3.5 _ (Et4N)2[Fe4$4(OPh)4] (—) 3.4 - (Et4N)2[Fe4S.(SRr-).] (---) Figure 10 .6ssoomos 66 c6R6o one was no oaoaccsoo mocanm Hooaeono .Ammo-mv Hm.m- .mw.6- “mofioe-m mmo.s- ”moss .mfiocond esoocmosoao .6> moeacn .oomm o6 occaosHom zommo sHo .oomm o6 6o6o6 peach 6hr cHo .Am-OH x Hugo H-zv so ca Amv Rosa .oomm o6 soapsHon 6HHtoacooooo cHo Ammonmvmw.ml Amlmvmw.m+ Eda .meOQORo ooflxocozm Amuavmm.mu .Am-mvmm.m+ .Am-avmm.m- .Am-mvfim.m+ do monasm cadohoomH mm.H mm.H m: «onmav «COLH pom aeon-6:52 Sfipocwmz Am.mevoem .Asavosm Ao.eevmmm .Anhvosm .mmosshooa Am.eaveme .Asmvem .As.mfivofie .Asmvomo Honooodm oscosoo6fim -mfleAHoeLm-oveme6eu -mfleflsdovemeosL .moonoEoo ImmzAh3 >05 Q’HAIULLI -9.39 -4.5I -3.96 S TMS 293 K o o m-H p-H i -9.'25 -4.7I'-'4|6 S TMS 233K Q o m—H _ P-H ___I 9’“ JLL__ -9.'09 49614.38 Figure 12 Figure 13. 67 Proton magnetic resonance spectra (250 MHz) of (BuuN)2[FeuSu(O-p-Tol)u] in d3-MeCN solu— tion at various temperatures. Peaks from protons of the cation are indicated by Q, solvent by S, residual water by W, and un— identified impurities by X. Chemical shifts are in ppm from internal MeuSi (TMS). 68 (n-Bu4 N )2[Fe4 84(0C6H4-p-CH3)4] S Q 363 K m-H LZ‘K’E’S A -x ~9525 -5.I4 $1.29 m-H I} Cg -9.08 '498 S 233K Q -CH TMS 9 3 Q 0 m-H w X L J x . -8.93 -4.98 Figure 13 69 Figure 14. Temperature dependence of isotropically shifted ligand proton resonances of (EtuN)2- [FeuSu(OPh)u] (o) and (BuuN)2[FeuSu(O—p—Tol)u] (o) in d3-MeCN. 70 :H ohswflm 86:. cm om 0v ON 0 ON.- 0v.- on- om- o..- o Ede-mid. 0.. 71 (A§)contact = _ Aye BS(S+1) H YH 3KT ’ where A is the proton-electron coupling constant, Ye and YH are the gyromagnetic ratios of the electron and proton, respectively; B is the Bohr magneton and g is the nuclear g factor. Thus, the contact shift is proportional to the magnetic susceptibility. The dipolar term118 is given by: (%§)dipolar = -(X|l _ Xi) |:3cos:0-l] r where XII and XI are the axial and equatorial components of the magnetic susceptibility; e is the angle between the mag— netic axis and the nucleus being observed; and r is the distance between the nucleus and the paramagnetic center. Single crystal magnetic anisotropy measurements are re— quired to evaluate the magnitude of the dipolar term. In ll8 can be observed practice, however, the r—3 dependence in the NMR spectra. There are two pathways for de- localization of spin onto a ligand, 0 and w delocaliza- tionll8. Characteristic features of contact shifted resonance arising from o delocalization are rapid attenua- tion of the shifts as the protons become further from the paramagnetic center and non—alternation of the signs of the shifts. Also, replacement of a proton by a methyl 72 group usually results in a methyl resonance Shifted in the same direction but smaller in magnitude than the proton shift. Features of contact shifts resulting from it spin delocalization include: both upfield and downfield shifts which alternate in direction between protons attached to adj acent carbon atoms; insignificant attenuation of the magnitudes of the shifts as the number of bonds between the metal center and the proton is increased; and sub— stitution of a methyl group for a proton affords a shift in the opposite direction and comparable in magnitude to that of the replaced proton. Thus, contact shifts in ali- phatic ligands usually arise from o delocalization while in aromatic ligands contact shifts usually arise mainly from 1r delocalization. Dominant contact interactions have been established for the series of complexesgb' [Fe484(SR)4]2- where R = alkyl, aryl. All alkyl tetramers Show rapid attenuation of the shifts as the number of carbon atoms in the chain are increased. In addition, all shifts are downfield, a situation consistent with antiparallel ligand to metal spin transfer as o delocalization of the parallel (posi— tive) Spin on sulfur would result in negative contact shifts. It should be noted that ligand to metal spin transfer in this system must necessarily be antiparallel, regardless of whether one considers the tetrahedral metal sites as Fe(II) (e3t23) or Fe(III) (821323). When R = Ph, high 73 field shifts are observed for the 23222 and paga protons while the aaaa protons are displaced to low field. When R = p—Tol, the 23222 and mafia protons remain unchanged in Sign and magnitude, while the methyl shift at the 2232 position is reversed in Sign and comparable in magnitude to that of the replaced proton. Antiparallel ligand to metal spin transfer would leave positive spin on sulfur, which could be delocalized through the N system of the phenyl ring as shown below: 65............ 1 Thus, positive spin would be placed on the 23222 and paaa positions and negative Spin at the @232 positions would result from spin correlation effects. The resultsgu for the mercaptide tetramers, therefore, are consistent with a o delocalization mechanism when R = alkyl and a n delocalization mechanism when R = aryl. In the phenoxide tetramer [Fe4s4(OAr)4]2- Ar = Ph the room temperature spectrum shows resonances due to the phenyl protons at —4.16, -4.71 and -9.25 ppm. Assignments are based on relative linewidths and the results of sub- stitution of the p-H by CH3. The ortho proton resonance eat approximately —4.3 to -4.9 ppm is clearly identifiable 74 by its relatively large linewidth, as dipolar broadening is expected to have an r'6 dependence118 Upon substitu- tion of the p—H for CH3, the resonance at -4.16 ppm dis— appears and is replaced by a signal at -4.9 ppm, with inte— grated intensity corresponding to three protons. Thus, the resonance at -4.16 ppm is assigned to the p—H, and the resonance at -9.25 ppm must therefore be due to the T‘H- The alternation of the signs of the shifts as one proceeds from a—H to M-H to p-H, the lack of attenuation of the shifts with increasing distance from the metal center, and the Sign reversal of the CH3 shift upon replacement of the p-H all suggest that dominant contact interactions with a N delocalization mechanism as described above are responsible for the observed isotropic shifts. These re— sults are qualitatively similar to those observed for the arenethiolate analogsgu described above. The magnitudes of the shifts relative to those of the T233 protons are listed in Table IV. The relative shifts for the paga protons are slightly less than the corresponding values for [FeuSu(SPh)u]2— due to a somewhat larger increase in the mafia vs the paga shifts, but the general trend 2232 > 22322 > T232 is obeyed. Since the relative shifts should 9“ provided the shifts are reflect relative spin densities contact in origin, the data support the conclusion that dipolar interactions are negligible in these systems. The only significant difference between the proton NMR 75 Table IV. Comparison of Relative Isotropic Shiftsa for Various Metal—Sulfur Centers in CD3CN Solu- tions. Complex ortho meta para [FeuSu(SPh)u]2— b 1.34 1.00 1.96 [FeuSu(OPh)u]2_ 1.04 1.00 1.29 [FeuSu(SPh)u]3- C h 1.00 1.68 [Fe282(SPh)u]2_ d 1.09 1.00 1.81 [Fe282(OPh)u]2_ h 1.00 1.27 [MoFeSu(SPh)2]2— e 1.25 1.00 1.49 [MoFeSu(OPh)2]2_ f 1.03 1.00 1.09 [Mo2re6s8(sst)3(sph)6]3‘ g’h h 1.00 1.65 [No2re6s8(sst)3(oph)6]3' h 1.00 1.30 aRelative to meta shift; 22-2500. bReference 94. CReference 137. dReference 96. eReference 72. fReference 81. gReference 136. hObscured. 76 spectra of the phenolate and arenethiolate tetramers is the magnitude of the observed shifts, which at any tempera— ture are approximately twice as large for the phenolate derivative as for the corresponding arenethiolate complex. Since the magnetic properties of the two clusters are virtually identical, the difference must be due to a larger hyperfine interaction in the phenolate complexes. This is consistent with the structural data, which suggest a relatively covalent Fe—O interaction. The isotropic shifts of the phenyl protons of the phenoxide tetramers are temperature dependent. The data plotted in Figure 14 show that the magnitude of the shifts increases with increasing temperature throughout the tem— perature range. This behavior parallels the magnetic prOperties of the clusters, as expected, and is similar to the temperature dependence of the shifts of the correspond— 94 ing arenethiolate tetramers These results are again consistent with intramolecular antiferromagnetic coupling. 6. Mossbauer Spectra The 57Fe Mossbauer spectrum of a solid sample of (EtuN)2[FeuSu(OPh)u] diluted with boron nitride was obtained at 4.2 K. It consists of a single quadrupole doublet with 6 = 0.50 mm/s, AEQ = 1.21 mm/s and r = 0.32 mm/s. The isomer shift is measured relative to Fe metal at room temperature. The spectrum is shown in Figure 15. Upon 77 57 (EtuN)2[FeuSu(OPh)u] at 4.2 K in zero applied field. Figure 15. Fe Mossbauer spectrum of polycrystalline 78 ma 6ssmaa meE v m m o _- N- m- e- . d u q u u u - - .om _ _ L o. . . - .o. . . Q62 .. ._ x we a 2.6m - om .......... 726966623269 ._. .§.ala.h-- :0” .. - LIL-1.x Issf$l=iittii.¥!itifi.f!l{: 1 0.0 P p b — p — - 79 application of a small (N600 G) magnetic field, the spec— trum remains essentially unchanged. This result is in— dicative of a diamagnetic ground state (S = 0) and is consistent with the magnetic susceptibility data. Similar spectra are obtained for both solid and frozen solution samples. These results are comparable to the arenethiolate clus— ters119 under similar conditions, except for the magnitude of the isomer shift, 6. Because the isomer Shift of the arenethiolate tetramers have previously been reported relative to that of Fe metal at the same temperature as the sample, the spectrum of (EtuN)2[FeuSu(SPh)u] was remeasured versus Fe metal at room temperature. The following parameters were obtained: 0 = 0.46 mm/s and AEQ = 1.20 mm/s. The increased isomer shift for the phenolate tetra— mer is consistent with increased ferrous character of the iron. This suggests increased electron donation of phen— oxide versus thiophenoxide to the FeuSu core and is con— sistent with increased covalency in the Fe—O bond relative to the Fe—S bond. This result is somewhat surprising since phenoxide is an inherently poorer electron donor than thiophenoxide, considering the relative oxidizing power of PhOOPh versus PhSSPh. This is consistent, however, with the structural results presented above and electrochemical data discussed below. 80 7. Electrochemistry The electrochemical behavior of the phenoxide clusters [FeuSu(OAr)u]2_ (Ar = Ph, p—Tol) were examined by three methods including direct current polarography (DCP), differential pulse polarography (DDP), and cyclic voltam- metry (CV). Over the potential range +1.0 to -2.0 V, approximately 1 mM solutions of the complexes in acetonitrile and N—methylpyrrolidinone were used. Both compounds ex- hibit two well-defined quasireversible one—electron reduc— tions corresponding to sequential generation of the tri— and tetraanions. No evidence for a discrete oxidation process, corresponding to formation of the monoanion, [FeuSu(OAr)u]—, containing the as yet unisolated [FeuSu]3+ core, was found. Examples of a typical cyclic voltammogram and differential pulse polarogram are displayed in Figure 16. Data for both complexes as well as for the thio— phenolate tetramer in NMP are tabulated in Table V. Comparison of the data for the well-characterized complexes [FeuSu(SAr)u]2- (Ar = Ph, p—Tol) with that for the corresponding phenolate tetramers shows that both the 2-/3- and the 3-/4- reduction processes are approximately electrochemically reversible by any method used to examine them. Plots of log[i/(id-i)] vs. potential in all cases yielded slopes acceptably close to the theoretical value of 59 mV for a reversible one—electron process. In DPP, the widths at half-height, Wl/2, were somewhat less 81 Figure 16. Cyclic voltammetry and differential pulsed polarography scans for (EtuN)2[FeuSu(OPh)u]. Solvents and scan rates are indicated. 82 'l.l2 (Et4N)2 [Pamper-)4] redn. IOO mV/s MeCN ox. 'I.O4 L I I I I I I '08 'IO 'l.2 ”L4 V vs. SCE - I83 I -I.I7 I 5 mV/s NMP *1 I 1 J I J ’09 -I.I -I.3 -l.5 -l.7 'l.9 V vs. SCE Figure 16 83 .m\>E oo— mmom m> comm .m\>E m mmum m> Domwm a__ a6._- ma.e Nm- we._- -6\-m azz oe._ _m_- aa.o- mm_ 66._- mo._. Nm- mo.a- -M\-N azz flefiaamveme6aummze6mu m6._ mm- ew._- -6\-m 2662 aa.o ma - N_.F- A6._ he- m_._- -M\-N 266: 66_ mm._- Na.e mm- mw._- -e\-m azz m6.e e__- m_._- em_ ON.F- 66.6 66- a_._- -M\-N azz A_oenm-oveme6aamflzesma s6.o a6 - we._- 6.N mm- wo._- -M\-N 2662 mm_ mw._- ma.o 6m- ow._- -6\-m aZZ No._ NMP- e_._- ___ s_._- _6.6 mm- m_._- -M\-N azz mefleaoveme6lummzeomL do an a . a\ A>er A>vm A>ev A>V 6 Aze\v 6666666 >lom easoae66 a>o saaa sacs .moxm_asou -mfleAc 6_noh 84 satisfactory, as the theoretical value for a reversible one-electron reduction is 90 mV. They are acceptable, however, when compared to the arenethiolate analogs measured under identical conditions. In the CV experi- ments, ipa/ipC values were very close to the theoretical value of 1.0 in all cases. The E E values were in ac- pc pa cord with those obtained for the corresponding arenethiolate complexes examined under similar conditions and approached the theoretical value of 59 mV at slower scan rates. As with an extensive series of thiolate tetramers, several solvent and substituent effects are observed. NMP was chosen as it has been suggested that it is capable of stabilizing highly reduced Specieslzo. Indeed, data for the 3-/4- reduction of the phenolate and thiophenolate tetramers indicate more reversible processes than in MeCN. In addition, the phenolate tetramers are stable in NMP where as they are solvolyzed in DMF, a solvent commonly used for electrochemical studies. As with the thiolate tetramers, there are small (<0.07 V) negative shifts in the reduction potentials in NMP vs. MeCN for the phenolate tetramers. The polarographic diffusion currents are de- creased by a factor of two in NMP vs. MeCN, which may be due to the higher viscosity of NMP relative to MeCN. This behavior is also observed in the thiolate tetra— mers95 in NMP and in DMF. In fact, data obtained in DMF for the thiophenolate complex, when compared to corresponding 85 data in NMP, Show that EDMF z ENMP as do the slopes and diffusion currents. Substitution of the p-H by CH3 re— sults in a small (30-50 mV) negative shift in the reduc— tion potentials. This can be explained as a simple inductive effect of the electron releasing methyl group on the phenyl ring and is also observed in the correspond— 95 ing arenethiolate tetramers . Most importantly, the phenolate tetramers exhibit the same pattern of two one—electron reductions separated by approximately 600 mV that is typical of all [FeuSuLHJZ- clustersg3’95. The potentials of these reductions are, however, 100 to 150 mV more negative than those of the arenethiolate complexes. Simple electronegativity argu- ments would predict the phenolate tetramer to be more susceptible to reduction as the more electronegative oxygen atoms should decrease the electron density at the [FeuSu]n+ core relative to the thiophenoxide analog, as is observed :for the [FeuSuClMJZ- 93 and [FeuSu(OAc)u]2- 121 tetramers. IPhenoxide must therefore be capable of transferring more eelectron density to the [Feusu]2+ core than thiophenoxide. Uihis is consistent with a more covalent Fe—O interaction Eis suggested by the structural results. Similar behavior kuas been observed for the [S2MoS2FeL2]2_ system where L — SAr72, 0Ar81. 86 8. Ligand Exchange Reactions The phenoxide cluster complexes are extraordinarily sensitive to water or other acidic solvent impurities, necessitating rigorous purification procedures for sol— vents used in synthesis and for Spectroscopic characteriza- tion. These compounds are not stable for long periods of time in solvent mixtures containing protic solvents such as methanol and ethanol, highly coordinating solvents such as N,N—dimethylacetamide, N,N—dimethylformamide, and di- methylsulfoxide, and to a lesser extent ether solvents such as tetrahydrofuran. They are relatively stable in NMP arfl.MeCN. Whether the observed decomposition is due to (direct solvolysis or to reaction with trace acidic impuri— ‘ties is not clear, but the reactivity is comparable to that (of the halide ligated clusters [FenSuX4]2-,93 where X = (31, Br, I. Decomposition is accompanied by gradual loss <>f the longest wavelength feature in the optical absorb- tzion spectrum and eventual formation of insoluble black gxrecipitates. The phenolate tetramers also react with electrophiles SLlCh as acyl halides, as shown by quantitative formation of [IPeuSuCIMJZ- (demonstrated by optical spectra) and pre- SLunably phenyl benzoate upon treatment with four equiva- lients of benzoyl chloride (reaction 3). This reactivity ifs completely analogous to the reaction of acyl halides with thiolate ligated tetramerlel, although somewhat surprising 87 when one considers the lack of known examples of nucleo- philic behavior of coordinated oxygen ligands. 12‘ + 4PhCOOPh (3) 2.. [FeuSu(OPh)u] + “PhCOCl + [FeuSuClu Reaction of the phenolate tetramers with thiols such as thiophenol results in immediate and quantitative for- mation of the thiolate derivative (Reaction 4), as ex- pected from the evident lability of terminal phenolate ligands and the relative acidity of phenols and thiollel. Previous investigation of the thiolate tetramer series has demonstrated facile ligand exchange reactions in which coordinated thiolates are sequentially substituted. In this series, substitution tendencies roughly parallel aqueous aciditieslOl and arylthiols effect full substitu- tion of alkylthiolate ligated clusters. Further, kinetics data122 suggest a Simple mechanism in which the rate limit- ing step is protonation of the coordinated ligand followed 13y rapid separation of alkylthiol and coordination of errylthiolate. The phenolate tetramers fit into this re— eactivity pattern, as phenol is slightly more acidic than aleylthiols and substantially less acidic than arylthiols. Iieaction of phenolate tetramers with arylthiols have been 1 Clemonstrated by monitoring optical and H NMR spectra of tile reaction mixtures. 88 [FeuSu(OAr)u12- + 4Ar'SH + [FeuSu(SAr')u]2- + 4ArOH (4) Optical Spectra of a solution of (EtuN)2[FeuSu(OPh)u] titrated with 0-5 equivalents of PhSH are shown in Figure 17. The most notable effects of addition of PhSH are a progressive shift in the position of the lowest energy peak to longer wavelength, together with appearance of a peak at approximately 260 nm with the concomitant dis— appearance of the original peak at 240 nm. Apparant isos bestic points are observed at 408, 398, 297, 290, and 248 nm. The final spectrum obtained upon addition of > 4 equivalents of PhSH is identical to that of the [F8484- (SPh)u]2- ion95 measured separately. Examination of solutions containing (EtuN)2[FeuSu- 1H NMR spectroscopy provides more de— (OPh)u] are PhSH by tailed information on the nature of the species present. Selected spectra are displayed in Figure 18. Addition of n equivalents of PhSH (n i 4) results in the release of n equivalents of free PhOH as estimated by integration of the aromatic resonances of the free PhOH versus the cation peaks. This indicates that, as expectelel, the Inore acidic thiophenol quantitatively diSplaces coordinated pflrenolate. In addition, the isotropically shifted proton :resonances of coordinated phenolate decrease in intensity aJKI exhibit multiple peaks. At the same time, new sets of :resonances appear at approximately —8.3 (m-H), -5.9 (a-H), 89 Figure 17. Optical spectra of a 3 mM solution of (EtuN)2[FeuSu(OPh)u] in MeCN treated sequen— tially with 0-5 equivalents of PhSH at 22°C. Optical pathlength: 0.2 mm. 90 (Et4 N)2[Fe4S4(OPh)4] + nPhSH I J 1 1 I 300 400 500 600 700 Mnm) Figure 17 Figure 18. 91 Proton magnetic resonance spectra (250 MHz) of a 10 mM solution of (EtuN)2[FeuSu(OPh)u] treated sequentially with the indicated amounts of PhSH at 22°C. Peaks from protons of the cation are indicated by Q, solvent by S, residual water by W, and unidentified impuri- ties by X. Chemical shifts are in ppm vs. MeuSi internal standard (TMS). .92 Q'H B'H Pho Ph0 '925 '4'7' "415 S L442]: Q Q qu PhSH TMS PhQH ,A‘ ~ 93 2eq Ph SH PhOH p-H p—H PhS PhO 3e PhSH PhOH q r-H g—H Q—H PhS PhO p-H _ p_H PhS PhO 4eq PhSH PhOH PhSH Q A m-H PhS 9- PhS I PIhStI -8.ll8 -5'.9 -5.29 Figure 18 TMS Q TMS Q TMS 94 and —5.2 (p—H) ppm, which correspond to coordinated thio— phenolategu. The intensity of these sets of multiple peaks vary as n is changed. This variation in intensity of the peaks with PhS/PhO ratio allows the individual peaks to be assigned to the meta and para protons of the ligands in the mixed—ligand species [FeuSu(OPh)u_n(SPh)n]2— where n = 0—4. Chemical shifts of the mixed ligand tetramers are given in Table VI. The linewidth of the ortho proton signals is greater than their chemical shift differences in the mixed ligand species. No evidence for disruption of the FeuSu core was observed. In View of the evidence for relatively covalent Fe—O interactions in (EtuN)2[FeuSu(OPh)u] and the extreme difference in acidities of phenol and thiophenol, one might anticipate the equilibria to favor certain mixed— ligand species at the expense of others. That is, one Inight expect a non—statistical distribution of ligands among the various species present. The relatively well separated g-H and p—H signals of coordinated phenoxide arfl.thiophenoxide provide a direct method of examining tins. Concentrations of each mixed-ligand species were Eistimated from peak heights of the partially resolved m—H FJeaks of coordinated phenolate and thiophenolate, assuming tlhat the linewidths for all mixed—ligand species are the Samne. Their relative areas were determined and compared VVith the total integrated intensity of each set of peaks 95 Table VI. Chemical Shiftsa for Phenoxide and Thiophenoxide Ligands in [FeuSu(OPh)4_n(SPh)n]2- Species (n= 0,1,2,3,4) in CD3CN Solution at 22°C. Proton n = 0 1 2 3 4 a—H(Ph0) -9 25 —9 17 —9.10 —8 99 ————— p—H(Ph0) —4 15 -4.25 -4 34 —4 45 ————— m—H(PhS) ————— —8 38 —8 31 —8.25 —8 18 p—H(PhS) ————— —5 21 —5 24 —5.27 -5 29 avs. TMS (ppm) .- I measured versus the cation proton resonances as internal standard. Seven individual equilibrium constants, three 101, are required to describe of which are independent the equilibria of this system. The sequential substitu— tion reactions are described by the set of equilibrium constants Kn (n = 0-3). = [FeuSu(OPh)u_(n+l>(SPh)n+l][PhOH] n [Feusu(OPh)u_n(SPh)n][PhSH] Since the exact concentration of added thiophenol could not be measured, all of the individual equilibrium con- stants could not be obtained. The ratios K“ = KO/Kl’ K5 = Kl/K2 and K6 = K2/K3 can be determined independently. These ratios are also the equilibrium constants that represent ligand exchange reactions between tetramers. Experimental values of K4’ K5, and K6 obtained at various PhSH/tetramer ratios are given in Table VII. The average values do not differ significantly from those calculated for a statistical distribution of ligands among the various species present. Similar statistical distributions of ligands have been observed for exchange of thiolate ligands with other thiolatelel, and approximately statis- tical distributions are reported for mixed thiolate—acetate clusters as welll2l. 97 Table VII. Ratios of Equilibrium Constants for PhSZPhO Ex- change in CD3CN Solution at 22°C. [PhSH] K K K W 4 5 6 M-H(Ph0) 0 57 2.0 1 5 ___ 0.89 2.5 2 6 --- 1.30‘ 2.5 3.2 --- 1.67 2.6 2.8 --- 2.17 2.3 2.5 --- 2 77 --- 2.7 --- m—H(PhS) 1.30 —-- 2.5 3.7 1.67 -—- 2.5 2.8 2.17 —-- 2.7 2.5 2.77 —-- 3.4 2.7 Average 2.4(2) 2.7(4) 2.9(5) Statistical 2.66 2.25 2.66 98 9. Fe—S Long Wavelengph Compounda Preparation of an extensive series of substituted phenolate tetramers has been difficult because of a com— bination of high solubility of the clusters in polar organic solvents, their sensitivity to protic and strongly coordinating solvents, and a tendency to form oils. In an effort to obtain crystalline samples of certain pheno— late tetramers, the size of the quaternary cation was varied. In certain of these preparations, a crystal— line product with physical properties distinctly different from samples of authentic tetramer salts with the same phenolate were obtained. These materials were recrystallized from MeCN/a—PrOH or MeCN/THF and gave essentially identi— cal spectroscopic properties before and after recrystalliza— tion. Specifically, two compounds were obtained both start— ing with (RAN)2[FeuSu(SR')u](R' = Et, p—Bu)and either p— CH3C6HHOH or p—ClC6HuOH according to Reaction 1 in Sec— tion III.A.1. As will be discussed in more detail below, the main absorption band in the optical spectrum of these new compounds is shifted to longer wavelength relative to the corresponding phenolate tetramer; they have therefore been termed 'long wavelength' compounds. The abbreviations MeuN/O—p-Tol/LW, EtuN/O—p-Tol/LW and EtuN/O—p—ClC6Hu/LW are used to indicate the cation and phenol. Two different salts, MeuN and EtuN, of O—p—Tol/LW were obtained in 99 separate preparations. Materials possessing identical spectroscopic prOperties have been obtained using the apprOpriate starting materials and following Reaction 2 (Section III.A.1). By either procedure the yields of these compounds vary between 20 and 50%. Analytical re— sults Show N/Fe/S mole ratios of 0.5/1/1 for MeuN/O—p—Tol/LW and N/Fe/S/Cl ratios of 0.5/1/1/1 for EtuN/O-p-C6HMC1/LW. These data indicate one S= and one phenolate per iron atom, consistent with a cluster formulation. Despite ex- tensive efforts, crystals of sufficient quality for X- ray structural analysis have not been obtained. The optical spectra of (BuuN)2[FeuSu(O-p-Tol)u] and EtuN/O—p-Tol/LW are shown in Figure 19. Optical data for the two compounds are given in Table VIII. As can be seen in Figure 19, the absorption spectra of EtuN/O-p-Tol/ LW and p—cresolate tetramer are very similar. The major differences are a shift to longer wavelength of the main absorption band by N20 nm, while the band at 650 nm in the ‘tetramer spectrum is unresolved in the LW spectrum and tflie shoulder at N282 nm in the tetramer is more fully re- SC£1ved in the LW complex. Corresponding bands at 424 nm (tetramer) and 435 nm (LW) are tentatively assigned to Jfiigand—to-metal charge transfer. The magnetic susceptibility of the MeuN/O—p-Tol/LW nonasn meomm 66 coho ISHOm zoozlmt CH toLSwwozm .Amloa x HIEOHIEV Es «Adv K MQomm pm 2062 CH oopzmmozm Am-mv 66.m- Am-ov mm.m wm6.6wm.s:m 34\H6em66-d-o Ammo-ma 6H.6- Arlee mH.mI I Am-ov 6m.m ems.mmm.m6m 3q\aoeuouo 616666 6m.6a- Am-mv mm.mH Amumv ow.mu Am.HmVoos m m m m Am-ov 6 .Am.6sv6sm .Am.swvmmm -mfl Anaov Aommvmm 6m can Am-mv m6.m- A6.m.s6V66me .A6.HHV6H: : 6 I m m Am-ov 6 Aa.mmvmwm .A6.6mvmsm ImfleAao m Old-6V m 6gL Ammo-ma 66.:- Am-mv sm.mI A6.w.nnvooma .AR.MHVmHe e I m m Amuov em.6 Am.mmvewm .Am.wmvm6m Imfl lace-d-ov m 6gL Am-mv ew.s Am-mv em.mn AH.6.an66me.A6.mvaoe m m Am-ov 6 .A6.6mvomm .As.mmvoem -mfleflsdov m 6mg QEQQ .OmHAm\m63 moon muom one ImfleAsdovaommvmmeogmozL .Imflsnsaovmmmoah mo wCOpopm ooflxococm mo mpmflgm oflooppomH pom moRSpmom Hmppooom oesopoooam .HHH> oHQwB 103 (uncorrected) was obtained. Assuming a formulation as (MeuN)2[FeuSu(O—p—Tol)u] (molecular weight 928.46) and a ligand correction of —555.7 x 1076, a neff/Fe of 1.15 BM was calculated. This result is approximately the same as that obtained for authentic p—cresolate tetramer (1.32 BM), suggesting antiferromagnetic coupling and the possibility that the Fens“ core may be intact. Proton NMR spectra have been recorded at room tempera— ture for O-p—C6H401/LW and as a function of temperature for O—p—Tol/LW complexes in d3—MeCN solution. In addition to resonances of the cation and residual undeuterated sol— vent, isotropically shifted resonances due to the pheno- late ligands were observed. Room temperature isotropic shift data for the complexes are given in Table VIII. Representative 250 MHz spectra of EtuN/O—p—Tol/LW at various temperatures are displayed in Figure 20; a plot of isotropic shift vs. temperature for this complex is shown in Figure 21. The room temperature spectrum for EtuN/O-p—Tol/LW shows resonances at -l2.00, —8.31, and -1.29 ppm. Qualitatively, this spectrum is very similar to the room temperature spec— trum of [Feusu(O—p-Tol)u]2~ (Figure 13). Assignments are based on this similarity in addition to relative line widths and results of substitution of p—CH3 by p-Cl. The ortho proton resonance at approximately —1.29 ppm is clearly identifiable by its relatively large 1inewidth,as discussed Figure 20. 104 Proton magnetic resonance spectra (250 MHZ) of EtuN/O-p—Tol/LW in d3—MeCN solution at various temperatures. Peaks from protons of the cation are indicated by Q, solvent by S, and unidentified impurities by X. Chemical shifts are in ppm from internal MeuSi (TMS). 105 80°C p'CHs EH 0 w , {I ,n I . -|2.I7 -e.50 p-CH3 20°C m-H Q . I - . -I2.oo -8.3I '40°C P'CHs m-H Q -u.ee -e.Ie Figure 20 W TMS 106 Figure 21. Temperature dependence of isotropically shifted ligand proton resonances of EtuN/O-p-Tol/Lw in d3-MeCN solution. 107 6.0 ,,,Tf,,,,.,1j 5.0- ‘ 4.0- 3.0— 2.0— LO- Afl (ppm) o~ H -|.o— -20- -3.0— -4.o ~ ,50_. E‘H’_*—*—‘-4~4—+—+_*_+—+—+—« -60 >— P'CH3 W l I l l J J J J l l J J I _ 40 -2o 0 20 40 so 80 T(°C) Figure 21 108 in Section III.A.S. Upon substitution of the p—CH3 group by E—Cl, the resonance at —8.3l ppm disappears and no new resonances are observed. The remaining resonance at -l2.00 ppm must therefore be due to the meta protons. This pattern is also observed for the p—cresolate tetramer, which, together with the alternation of the Sign of the shift as one proceeds from g—H to m—H and reversal of sign at the p—CH3 and the lack of attenuation of the shifts with increasing distance from the metal, suggests that dominant contact interactions are responsible for the iso— tropic shifts of the LW complexes as with the phenolate tetramers. The only significant difference between O—p—Tol/LW and O—p—Tol tetramer is in the magnitude of the isotropic shifts, which at any temperature are approximately twice (2x) as large for the LW complexes as for the correspond— ing phenolate tetramer. The isotropic shifts of the O—p— Tol/LW complexes vary with temperature. The data plotted in Figure 21 show that the magnitude of the shifts in- creases with increasing temperature throughout the tem- perature range. This behavior should parallel the mag— netic behavior of the complex,assuming dominant contact interactions are responsible for the isotropic shifts. These results are similar to those described above for the phenolate tetramers and suggest that the phenolate ligands in the LW complexes are bound to Fe atoms that are antiferromagnetically coupled. 109 Since the molecular structure of the LW complex is not yet known, these NMR results cannot fully be inter— preted. They are, however, qualitatively similar to those obtained for the phenolate and arenethiolate tetra- mers and suggest that the above interpretation is con- sistent with the data obtained. The 57Fe Mossbauer spectrum of polycrystalline EtuN/O—peTol/Lw has been measured at u.2 K in a boron nitride matrix. It consists of a single quadrupole doub- let with 6 = 0.5“ mm/s and AEq = 1.12 mm/s. The spectrum is shown in Figure 22. Application of a small (m6OO G) magnetic field results in essentially no change in the spectrum suggesting a diamagnetic ground state. In addi— tion, the isomer shift and quadrupole splitting are not significantly different from those obtained for (EtuN)2- [Feusu(OPh)u] under the same conditions. These data indicate that the LW compound possesses an intact (FeuSu)2+ core or at the least a single iron environment with very similar ligand set and oxidation level. These results are consistent with the proton NMR data described above. The electrochemical properties of the LW compounds have been measured by DC polarography (DCP) and differential jpulse polarography (DPP) and also by cyclic voltammetry (CV) at glassy carbon and Pt flag electrodes. Approxi— Inately 1-2 mm concentrations were used, based on their ftnflnulation as tetrameric species. Measurements were 110 Figure 22. 57Fe Mossbauer spectrum of polycrystalline EtuN/O-p—Tol/LW at “.2 K in zero applied field. 111 mm mpsmflm m\EE c. m N _ O _n Nu m: ¢u . _ may _ q _ _ _ . l .5 : naugw_ I _ _ ___ -00. 1 _ _ _ - 0.: - _ _ _ - om. - _ _ - o.o_ _ numw adayax I , __._____ _.__ 1 O.N .l _.=__ z. ___ 5551-11—55 -—_=_—==___:==_ __—__._=. I -===_£=_;._.'=__r____!____ -.;=::_==2:.=_£_:___€:-£___£;=-__= I. 0.0 _ _ _ _ _ _ _ _ 112 performed in NMP solution over the potential range +1.0 V to —2.5 V. Both O—p—Tol/LW and O—p—C6HuCl/LW exhibit three reductions. No evidence for a discrete oxidation pro— cess was found. Electrochemical data are collected in Table XIX. Representative differential pulse polarograms and cyclic voltammograms of both complexes are displayed in Figure 23. This electrochemical behavior of the Lw compounds is in contrast to the corresponding phenOlate tetramers that exhibit two reductions over the same potential range. Ex— amination of the DCP and DPP results show that the first two reductions of the LW complexes have about the same potential separation (N600 mV) and approximately the same potentials as the corresponding processes for the pheno— late tetramers. Specifically, the potential of the first reduction of O—p—Tol/Lw is within 10 mV of the 2—/3- reduc— tion of O—p—Tol tetramer and the potential of the second reduction of O—p—Tol/LW is approximately 120—130 mV more negative than the 3—/4— potential of O—p-Tol tetramer. In addition, the slope and Wl/2 values for both reductions are reasonably close to the values obtained for the O-p—Tol tet- ramer, indicating quasireversible one electron processes. It should also be noted that id/C values for the first re- duction at least are in good agreement with corresponding values of the phenolate tetramers, suggesting that the formulation of LW compounds as FeuSu species is not un— reasonable. .A>V 0mm .Qoapdaom 20oz CHI .oongpwHU msflxwzm .LoEwppop 113 c zmzom mm mm meQEoo map %o soapmHSEpom co comma mGOHQSHom SE NIHm .m\>E ooam .oom \mooao m mm\>E mm. .mcmom m mo modam> ommmo>w ohm Haw mmom m> comm pm COHPSHOm mzz CHM oom wma.mu 33H :o.mn emfi.m ms: Ho.mu mm.an msfi ms.Hu cam.a Hm: ms.au : o I 0mm mom.au m NH.H: cam.o mm- OH.H- Ho m Dunno mH.m: m mm.Hn m m m ms.Hu mmfi me.au wom.H mm: os.Hn 1 omH mHN.H- mma mH.Hu emo.H om- mH.H- 3Q\Hoeuano woa mam.au omH mm.Hn um\-: mmH.Hu HNH :H.Hn -:\um Immoflgmovmflpmmvmmmmmmozg mm.au mmH mm.an :m.H Hm- m©.H- .:\Im a z o m m H:.H- 03H mm.H- :m.H mm- om.au -mxum -mm Aao m oumuov m mag :m.au mmH ms.Hu no.0 ms- ms.an .:\:m s I m m mm.H- oma mz.H- mm.o 3m- mm.H- -m\nm -mfi Aaoenanov m meg mm.au mmm om.Hu 3H.H mm- mm.Hu .:\.m s m m mm.Hu 0mm mm.a- :m.o ms- mm.H- -m\:m -mfl Agmov m mm A>EV A>V A>EV A>vm AZE\V mmooopm onQEoo mgmloam com m\a3 Q\©H o oam m\am o.m>o p.mmmm n.wmoo .moonQEoo apwzoam>m3 wcoq mlom new .-mfimgaaovmgpmmvwmommmoza .-mflzgp p—H > p—CH3), indicating that other effects may be dominant. Since halide substituents in the para position are capable of participating in resonance struc— tures that increase the negative charge at the donor atom, and since such effects are expected to be much greater for phenols than thiophenols (greater overlap of oxygen p orbitals with w system of ring), the observed order may be due in part to dominant electron donating n effects over— coming the normal 0 inductive electron—withdrawing effects, 134 resulting in a net red shift for the p—Cl versus the parent phenolate analog. Resonance effects would be especially important for the deprotonated phenolate anion, which could form quinoid type structures more easily. The trend found here for the phenolate dimers has also been 127 in the absorbtion observed by Ackermann and Hesse maxima due to ligand—to—metal charge transfer of 1:1 iron (III)-phenolate complexes in alcohol solution (p—H (580 nm) < 2-01 (585 nm) < p—CH3 (610 nm)). 4. Magnetic Susceptibility The magnetic susceptibility of the complex (BuuN)2— [Fe282(OPh)u] has been measured at room temperature by the Faraday method. The data were corrected for the diamagnetic contributions due to ligands and cations using 112 Pascals constants The resulting value for neff/Fe of 1.64 BM was obtained. This result is comparable to Inagnetic moments per iron obtained for [Fe282(82-o-xylyl)2]2— 128 (1.43 BM) and [FeZS Clu12- 93 (1.38 BM) and suggests 2 intramolecular antiferromagnetic spin coupling. This re— sult is also consistent with the structural results indi- cating dimensional invariance of the Fe2S2 core as des- cribed above. 135 5. Proton Nuclear Magnetic Resonance Proton magnetic resonance spectra have been measured for the series of phenoxide complexes [Fe2S2(OAr)u]2_, where Ar = Ph, p—Tol, pr6HuC1 in acetonitrile solution at room temperature and as a function of temperature for com- plexes with Ar = Ph and prol. In addition to resonances due to the cation and residual undeuterated solvent, isotropically shifted resonances due to the phenoxide protons at room temperature are observed; the room tem- perature shifts are given in Table VIII. Representative 1H NMR spectra at various temperatures of the 250 MHZ dimers with Ar = Ph and p—Tol are shown in Figures 26 and 27, respectively. The temperature dependence of the iso— tropic shifts is plotted in Figure 28. At room temperature, the phenoxide protons of [Fe2s2- (OPh)u]2- appear at -10.86 ppm and approximately —2.15 ppm. At higher temperatures, a broad signal appears slightly upfield of the cation resonance at approximately —3.0 ppm which is assigned to the 23322 proton. The par- tially obscured resonance at approximately —2.15 ppm, which moves downfield at lower temperatures, disappears upon substitution of the p—H for a CH3 group and is re— placed by a resonance at —6.81 ppm, whose intensity cor— responds to three protons. It is therefore assigned to the 27H, and the remaining resonance at —10.86 ppm is assigned to the m-H. Figure 26. 136 Proton magnetic resonance spectra (250 MHz) of (BuuN)2[FeZS2(OPh)u] in d3—MeCN solution at various temperatures. Peaks from protons of the cations are indicated by Q, solvent by S, and unidentified impurities by X. Chemical shifts are in ppm from internal TMS. 137 I -l062 Figure 5 90° o 00 _m-H x Q-H J1 J ._. l l -|I.I2 '259 20°C 5 Q ‘~ 00 m-H °-H W" .4 ' -IO.86 -2.8 '4Cf S _rD-H TMS ‘HMS lTflS Figure 27. 138 Proton magnetic resonance spectra (250 MHz) of (MeuN)2[Fe2S2(O-p—Tol)u] in d -MeCN solu— 3 tion at various temperatures. Peaks from protons of the cations are indicated by Q, solvent by S, and unidentified impurities by X. Chemical shifts are in ppm from internal TMS. [=3 I 4Lb4 80°C —im 20°C 9_CH3 m-H 4029 -6£l me 4038 -63| Figure 27 TMS TMS TMS 140 Figure 28. Temperature dependence of isotropically shifted ligand proton resonances of (BuuN)2[Fe2S2(OPh)u] (o) and (MeuN)2[Fe2S2(O—p—Tol)u] (O) in d3—MeCN. v6 95 ‘4 +3 141 J J l 1 J J J J 1 -40 -IO ‘20 ‘50 080 T(°C) Figure 28 142 Several features of these spectra suggest that, as with the thiophenolate dimers96, dominant contact inter— actions are responsible for the observed isotropic shifts: (i) the alternation of the signs of the shifts as one pro— ceeds around the aromatic ring; (ii) the lack of attenua— tion of the magnitude of the shifts with increasing distance from the metal center; (iii) the sign reversal of the CH3 group upon replacement of the p—H. The magni— tudes of the shifts relative to those of the maaa protons are listed in Table X. The relative shifts for the para protons are slightly less than the corresponding values for [Fe2s2(SPh)u]2_ due to a somewhat larger increase in the maaa vs the para shifts, but the general trend BEBE > 23222 > maaa is obeyed. Since the relative shifts should reflect relative spin densities94 provided the shifts are contact in origin, the data support the conclusion that dipolar interactions are negligible in these systems. The only significant difference between the NMR spec— tra of the phenolate dimers and the thiophenolate dimers is that the magnitude of the isotropic shifts is larger for phenolate than for the corresponding thiophenolate analogs. These shift increases are somewhat smaller than those for the tetramer system, where the phenolate shifts are approxi— mately twice as large as the corresponding thiophenolate shifts. These data again indicate more delocalization of unpaired spin density when phenolate replaces thiophenolate. 143 This behavior is consistent with the structural data indicating a relatively short Fe—O interaction. The isotropic shifts of the phenyl protons of the phen— oxide dimers are temperature dependent. The data plotted in Figure 28 show that the magnitude of the shifts in— creases with increasing temperature over the entire tem— perature range. This result provides further evidence for intramolecular antiferromagnetic coupling and parallels the magnetic behavior of [Fe2S2(82—a—xylyl)2]2— 128 and [Fe2S2Cl4]2_ 93, two structurally characterized dimers for which variable temperature magnetic data is available. 6. 57Fe Mossbauer Spectra The 57Fe Mossbauer spectrum of a solid sample of (BuuN)2[Fe2SZ(OPh)u] diluted with boron nitride was ob— tained at 4.2 K and zero field. The spectrum is shown in Figure 29. It consists of a single quadrupole doublet with isomer shift y = 0.37 mm/s, quadrupole splitting AEQ = 0.32 mm/s and linewidth F = 0.26 mm/s. The isomer shift is measured relative to Fe metal at room temperature. The spectrum was also measured in a small (N600 G) applied magnetic field. The spectrum remained essentially un— changed, indicating that a diamagnetic ground state (S = 0) is present, which is consistent with the magnetic data Suggesting antiferromagnetic coupling. These results are 144 Figure 29. 57Fe Mossbauer spectrum of polycrystalline (BuuN)2[Fe2S2(OPh)u] at 4.2 K in zero applied field. 145 mm enemas 0.0. 0.0 0.0 0.¢ 0.N 0.0 «<85 V m N _ 0 T N- n. Y. . . u d _ _ _ 4 _ T J _. _ _ _ _ _ I. _ J _ _ _ _ _ ___ _ _ _ .I _ _ I. . _ . _ _ I _ _ .. l__ __ _.__. ___... .. __ . 1 .5..._..__.._=:_=_.__._____..____...._.....-__1....__....__.._.==_.=___2...... _ 1..__.._._._=__=____==_.__._________¢___zf-E.‘a........=..£......§ J _ b _ _ b . P _ 3°04 146 comparable to those obtained for the thiolate dimers, except for the magnitude of the isomer shift, 6. Mossbauer spectra of the complexes [Fe2S2Xu]2_ 97 where X = SPh and X2 = S2—a—xyly1 have been obtained, but the isomer shifts were measured relative to Fe metal at 4.2 K. Holm and co-workers129 have reported a correction factor of +0.12 mm/s in the isomer shift for various reduced tetra- mers, and a difference of +0.11 mm/s was obtained for the isomer shift of [FeuSu(SPh)u]2_, Section III.A.6. Using a correction of +0.11 mm/s, the thiolate parameters are as follows: 6 = 0.28 mm/s, AEQ = 0.36 mm/s (X2 = 82—9— xylyl)97; 6 = 0.28 mm/s, AB = 0.32 mm/s (X = SPh)97. Q The increased isomer shift for the phenolate dimer is consistent with increased ferrous character of the iron. This suggests, as before, that phenolate is capable of donating significantly more electron density to iron than thiophenolate, which may be due to a substantial degree of covalency in the Fe-O bonds. This increase in isomer shift for the phenolate vs thiophenolate dimers is ap— proximately twice that found for the corresponding tetra- mer system and may reflect the presence of two coordinated phenolates per iron versus one per iron in the tetramer. The results obtained for the phenolate dimer are consistent With the structural and magnetic data described above and the electrochemical data described below. 147 7. Electrochemistry Electrochemical measurements on the phenoxide complexes [Fe2SZ(OAr)u]2— where Ar = Ph, p—Tol, and p—C6H4Cl have been performed using three methods: do polarography (DCP), differential pulse polarography (DPP), and cyclic voltammetry (CV). Measurements were made on 1-2 mM solu— tions in NMP over the potential range +1.0 to —2.0 V. Each compound exhibits two well—defined cathodic processes corresponding to formation of the tri— and tetraanions. Typical cyclic voltammograms for each compound are pre- sented in Figure 30. All electrochemical data are listed in Table XIX. Comparison of the data with corresponding data for the [Fe2S2(SAr)u12_ (Ar = Ph, p—Tol, p—C6HuCl)97 complexes reveals several similarities. First, the thiophenolate analogs also show two cathodic processes over the same potential range. In dc polarography, the diffusion cur— rents are in reasonable agreement with those obtained for the thiophenolate dimers and the phenolate and thiopheno- late tetramers measured under similar conditions for a one electron process. Slopes of log [i/(id—i)] vs E are usually somewhat less than the theoretical value of 59 mV for a reversible one electron reduction. Peak halfwidths obtained by DPP are in adequate agreement with correspond— ing values obtained for phenolate tetramers (Table V) for such a process. The slopes (DCP) and peak half—widths (DPP) 148 Figure 30. Cyclic voltammograms for (BuuN)2[Fe2S2(OPh)u], (MeuN)2[Fe282(0—p-Tol)u1 and (EtuN)2(O—p- C6H4Cl)4] in NMP at glassy carbon electrode. Scan rates are 100 mV/s. —lO -| 57 -|95 —|.4| 149 (Bu‘N)2 [FezSZ(OPh )4] . I 53 4.83 (E 1, N)Z[Fezs,(ocsH,—p-cn,] Figure 30 195 (Mew2 [Fe2 52(0- p-Tol)‘] 150 obtained for [Fe2S2(OPh)u]2' are unusually large com— pared to the p—Tol and p—CéHuCl analogs, suggesting that the reduction process for this compound are less revers— ible by these criteria. These data show that, as for the thiolate dimers, the electron transfer series [Fe2S2— 2—’3_’4_ is realized and that the electron trans- (OAr)u1 fers are quasireversible by polarographic techniques. In the CV experiments, two well—developed cathodic peaks with essentially no reverse (anodic) peaks are observed for each phenolate dimer. This result indicates that no 2—/3— or 3-/4— dimer process satisfies the diagnos— tic criteria for reversible charge transfer. This behavior is very similar to results obtained for the [Fe282(SAr)u]2- 97 and [Fe2SZClu]2_ 93 complexes. The reduction potentials for the phenolate dimers exhibit variation with substitution of the p—H for CH3 and Cl. By any method used, the order observed is p—Cl < p—H < p—CH3 with 20—100 mV negative shifts between members of the series. This trend is also observed for the cor— responding arenethiolate dimers, and can be explained by simple inductive effects of the electron releasing CH3 and electron withdrawing C1 groups. It is, however, some— what surprising considering the order observed for the shifts in the electronic absorbtion maxima, Section III.B.1. The most important results obtained from the 151 electrochemical data are that the same pattern of two one electron reductions typical of all [Fe2S2Lu12- clus— 97,93 ters is observed, but that the potentials of these reductions are approximately 200 mV more negative than those of the arenethiolate analogs. For reasons explained in Section III.A.7, simple electronegativity arguments would predict the phenolate dimers to be easier to reduce than the thiolate analogs. Again, phenoxide is apparently capable of transferring more electron density to the [Fe2s2]2+ core than thiophenoxide. This is consistent with a more covalent Fe—O interaction and is reinforced by the Mossbauer and 1H NMR results. Similar behavior 2— 72,81 has been observed for the [MoFeSuXZJ and the [FeuSuXu]2_ 95 systems where X = SAr and OAr. 8. Ligand Exchange Reactions The phenoxide dimer complexes are extremely sensitive to water, acidic solvent impurities and strongly donating oxygen solvents, as are the phenoxide tetramers. This behavior necessitates rigorous purification of solvents for use in recrystallizations and solution studies. The phenolate dimers react with electrophiles such as acyl halides. Treatment with four equivalents of benzoyl chloride in acetonitrile results in rapid quantitative conversion to the corresponding dimeric chloro complex 152 (Reaction 9), as demonstrated by optical spectra. This reactivity has also been documented for the thiophenolate 93 101 dimers as well as the phenolate and thiophenolate tetramers and proceeds, presumably, by the same mechan— ism93. 2- + PhCOZAr (9) 2_ [Fe282(OAr)u] + 4PhCOCl + [Fe28201u] Reaction of the phenolate dimers with four equivalents of thiophenol results in immediate formation of the thio— phenolate derivative (Reaction 10). This reactivity again is as expected in view of the lability of phenolate ligands and the much greater acidity of thiophenol vs phenollOl. [Fe282(OAr)u]2— + 4PhSH + [Fe2SZ(SPh)u]2_ + 4ArOH (10) Previous investigation of the thiolate dimers show that they too undergo facile ligand exchange processes. For example97, [Fe282(S2—a-xylyl)2]2— complex is smoothly con- verted to [Fe2s2(SAr)u]2' by treatment with the appropriate arenethiol. In addition, analogous reactivity has been demonstrated for the phenolate tetramers, already dis— cussed. It is apparent that the phenolate dimers undergo facile ligand exchange with retention of the Fe2S2 core. 153 9. Summary Phenolate substituted binuclear Fe—S clusters can be prepared and isolated in crystalline form as their tetraalkylammonium salts. The optical spectra exhibit the expected blue shift upon oxygen substitution. Struc- tural results show a relatively short Fe-O bond. Magnetic properties remain virtually unchanged upon phenolate sub- stitution while the isotropically shifted phenyl protons in the 1H NMR reflect a substantial increase in unpaired spin density in the aromatic ring. An increased isomer shift in the 57Fe Mossbauer spectra and a substantial negative shift in the reduction potentials suggest that phenolate is transferring more electron density to the [Fe2S2J2+ core, consistent with the structural and NMR results. Comparison of corresponding shifts in properties between phenolate and thiophenolate ligated Fe2S2 centers with analogous shifts observed for the F9454 centers show an interesting trend. The blue shifts in the optical absorption maxima, the isotropic NMR shifts of the phenyl protons, and the negative shifts of the reduction poten— tials are in each case approximately REESE as large for the Fe282 complexes compared to the FeuSu complexes. This may reflect coordination of two phenolates to each iron in the dimers compared to one phenolate per iron in the tetramers. Results obtained for phenolate coordinated Fe282 clusters 154 suggest that substitution of a cysteinyl thiol ligand to an F8282 core in a protein by tyrosyl phenolate should have only a minor effect on cluster redox properties. This ob— servation is entirely consistent with results obtained for the phenoxide tetramers, and indicates that tyrosyl liga— tion could provide a reactive coordination site with minimal change in cluster reduction potentials. 3+ has been implicated in a 130 Phenolate coordination to Fe including trans— 133 and uteroferrin , 130 variety of metalloproteins and enzymes 131, purple phosphatases132 134 ferrins and certain heme proteins 3+ catechol dioxygenases Tetrahedral phenolate coordination of Fe has not been reported for any of the biological systems cited above. In addition, only one other synthetic four coordinate Fe3+ complex with mondentate phenolate coordination has been 135 reported89. However, a 2Fe—2S center has been identified in a protein containing no cysteine amino acid residues, but only tyrosine, serine, glutamic and aspartic acid moieties, strong evidence for non—cysteine coordination has been re- ported for the Rieske center140 from Thermus thermophilus. c. £M92F_e6_s_8(SEt)3(OPh)6i 1. Synthesis The phenolate double cubane complex [Mo2Fe6S8(SEt)3— (OPh)6]3- can be prepared by reaction of [Mo2Fe688(SEt)9]3- 155 with a large excess of PhOH (Reaction 11). This re— action is completely analogous to the reaction of [Fe4s4' (SR)u]2_ with excess phenol (Reaction 1, Section III.A.l) and represents a ligand exchange equilibria which is driven to completion by removal of the volatile EtSH in vacuo. The resulting product is isolated as its tetraalkylammonium salt in 50-60% yields after recrystallization. 3" + [MO2Fe6S8(SEt)9] + XS PhOH [Mo2Fe688(SEt)3(OPh)6]3' + 6EtSH (11) Isolation and recrystallization are made difficult be— cause of the extremely high solubility in polar organic solvents. Once prepared and isolated, however, the complex is stable in the absence of oxygen and water. A schematic View of the proposed structure is shown below. I '1 3' PhO 0Ph~ F? gy>/\/\/ 0% \2\ .__ Mo._.s _\S; :M/\§/\S\/::/ /F F41C) ' ()F41 156 2. Electronic Absorption Spectra The electronic absorption spectra for [Mo2Fe6S8- (SEt)9]3— and [Mo2Fe688(SEt)3(OPh)6]3_ are presented in Figure 31. Spectral data are presented in Table VIII. Comparison of the spectra show the elimination of the original bands at 281 nm and 391 nm and replacement with bands at 233 nm, 270 nm, and 400 nm. Similar spectral changes are observed for the reaction of greater than six equivalents of PhSH with [MOZFe6S8(SEt)9]3~ to yield the [Mo2Fe6s8(sst)3(SPh)6]3‘ ionl36, which exhibits prominent shoulders at N346 nm and N418 nm, and has recently been isolated as its tetraethylammonium salt. Apparently cor— responding bands atN400 nm (OPh) and N418 nm (SPh) are tentatively assigned to terminal ligand—to—metal charge transfer transitions. The observed blue shift in this band is again expected based on electronegativity argu- ments already discussed in Section III.A.3. The magnitude of the shift (N20 nm) is somewhat less than expected based on results of the F8484 system (N40 nm), which possesses a structurally similar core and one terminal ligand to each iron. 3. Magnetic Susceptibility The room temperature magnetic susceptibility of (Et3NCH2Ph)3[Mo2Fe6S8(SEt)3(OPh)6] has been measured by Figure 31. Electronic spectra of the [Mo2Fe6S8(SEt)3— (oph)6]3‘ (——) and [Mo2Fe6S8(SEt)9]3— (———) ions in MeCN at 22°C. 158 0 Hm opzmflm ACE: oom _ own _ 0000N 0000? OOQow (Fun IM; )9 the Faraday method. After correction for the diamagnetic contribution of the ligands and cations using Pascal's constantsll2, an effective magnetic moment, peff, of 5.74 BM for the cluster was obtained, suggesting the presence of intramolecular antiferromagnetic coupling. This result is virtually identical to that measured for (Et3NCH2Ph)3[Mo2Fe6S8(SEt)9]51 (5.73 BM) and (BuuN)3_ 8 [MozFe6SB(SPh)9]5 (5.70 BM) and corresponds to a spin only value of S = 5/2 at room temperature. A recent 66 which has satisfactorally simulated mathematical model the variable temperature susceptibility data for represen— tative double cubane complexes shows the ground magnetic state to be S = l, with thermal population of higher ex- cited states resulting in the observed magnetic moments at higher temperatures. These results suggest that the mag- netic properties of the double cubane clusters, as with the Fe—S tetramers and dimers, are not sensitive to the nature of the terminal ligand. 4. Proton Nuclear Magnetic Resonance Proton magnetic resonance spectra of (Et3NCH2Ph)3- [Mo2Fe6S8(SEt)3(OPh)6] have been measured as a function Of temperature in acetonitrile solution. In addition to the resonances due to the ctaions and residual undeuterated solvent, isotropically shifted resonances due to the phenyl protons and bridging ethanethiolate groups are observed. 160 Representative 250 MHz spectra at two temperatures are shown in Figure 32, while spectral data are collected in Table VIII. Plots of isotropic shifts versus temperature are displayed in Figure 33. As in the usual convention, resonances downfield of TMS are negative and isotropic shifts are measured relative to the diamagnetic ligands. Assignments were made based on the well demonstrated spectra of [Mo2Fe6s8(SPh)9]3‘,5l’58 and [Mo2Fe6S8(SEt)3_ (SPh)6]3— 136 and are shown in Figure 32. No resonances attributable to u—OPh groups were detected. The data suggest that, for reasons already discussed in Section III.A.5, contact interactions are responsible for the isotropic shifts of the phenyl protons. Analysis of the isotropic shifts of bridging groups for a series of double cubane complexesSl’58 indicate that there is an appreciable dipolar contribution. In [Mo2Fe6S8(SEt)3(OPh)6]3_, reson- ances for the bridging methylene protons have approximately the same shift (—l6.4 ppm) as those for [Mo2Fe6S8X6]3_ where x = 01 (—15.49 ppm)l36, SPh (—14.49)l36, SEt (-15.0)51 and it is assumed that dipolar interactions are also in effect here. Comparison of the magnitudes of the isotropic shifts of the phenyl protons of the phenolate complex (Table VII) with those of the thiophenolate analog (—6.6 ppm (m—H), 136 +10.9 ppm (p—H) shows that the general trend of sub— stantially larger shifts for phenolate vs thiophenolate Figure 32. 161 Proton magnetic resonance spectra (250 MHz) of (EtBNCH2Ph)3[M02Fe6S8(SEt)3(OPh)6] 1n d3— MeCN. Peaks from protons of the cations are indicated by Q, solvent by S, and unidentified impurities by X. Chemical shifts are in ppm from internal TMS. ”my." Kfifx‘ik 162 S TMS 20°C EtsNCHzph CH2 l 5 M0/ \M0 p'H I l I 48.87 46.68 554 .400C 5 TMS EtsNCHzPH CH3 ' PhOH M0/ \M0 A m-H 7" 1'5 W '2281 48.76 8.45 Figure 32 163 Figure 33. Temperature dependence of isotropically shifted ligand proton resonances of (Et NCH2Ph)3— 3 [MO2Fe6S8(SEt)3(0Ph)6] in d3-MeCN. 20 I5 '20 164 [D -H // b-CH2 / J J J J J J '40 '25 'IO 5 20 35 Figure 33 165 analogs established for the phenolate tetramers and di- mers is followed for the double cubanes. These results once again suggest that phenolate is capable of more electron density to the metal center than thiophenolate, and are consistent with the electrochemical results discussed below. The temperature dependence of the observed isotropic shifts for the phenolate protons plotted in Figure 33 shows that the magnitudes of the shifts increase with decreasing temperature. This behavior parallels that obtained for 8 and the terminal thiophenolates of [Mo2Fe6S8(SPh)9]3-,5 is also consistent with the temperature dependence of the magnetic susceptibility of several double cubane com— 51,66 plexes of the same type. These results are again consistent with intramolecular antiferromagnetic coupling. 5. Electrochemistry The electrochemical behavior of [Mo2Fe688(SEt)3— (OPh)6]3— has been examined by differential pulse polaro— graphy and cyclic voltammetry. Measurements were performed on 1 mM solutions in acetonitrile over the potential range +1.0 to -2.0 V. The complex exhibited two well defined quasireversable one electron reductions corresponding to sequential formation of the 4— and 5— ions. A typical cyclic voltammogram and differential pulse polarogram are 166 shown in Figure 34. Electrochemical data are provided in Table XIX. Examination of the cyclic voltammetric results suggest that while reduction of the complex is not strictly electro- chemically reversible (Shown by [Ep —Ep |> 59 mV), it does approximate chemical reversibility, as shown by the value of ipc/ipa N 1. Peak half~widths obtained by DPP measure- ments, although somewhat larger than the theoretical value of 90 mV for a reversible one electron process, are in reasonable agreement with values obtained for the quasi— reversible one electron reductions of the phenolate and thiophenolate tetramers under similar conditions. Comparison of the corresponding reduction potentials for [Mo2Fe688(SEt)3X6]3— where X = SPh (E1 = —l.01 V, E2 = —1.19 V)l36; OPh (Table VIII); C1 (E1 = —o.83 V, E2 = 136 —l.01 V) show the same order of decreasing ease of reduction, C1 < SPh < OPh established for the Fe-S tetramer and dimer series. The shift of approximately -l20 mV for each of the two reductions on substitution of pheno— late for thiophenolate represents further evidence of the ability of phenolates to donate more electron density to the metal sulfur center than thiophenolates, therefore render— ing them more difficult to reduce. Similar results have also been obtained for the [MoFeSuX2J2— system where X = SAr72, OAr8l. 167 Figure 34. Cyclic voltammogram and differential pulsed polarogram for (Et3NCH2Ph)3[Mo2Fe688(SEt)3— (OPh)6] in MeCN. Scan rates are 100 mV/s (CV) and 5 mV/s (DPP). 2m opswflm mo..- 168 mm..- mm..- 169 6. Ligand Exchange Reactions The phenolate double cubane cluster is extraordinarily sensitive to water, acidic solvent impurities and strongly donating solvents, as are the phenolate Fe—S clusters. They also undergo reactions with electrophiles. Reaction with benzoyl chloride smoothly converts the phenolate complex to the corresponding chloro clusterl36’65 in quantitative yield as shown by optical spectra (Reaction 12). [Mo2Fe6S8(SEt)3(OPh)6]3_ + 6 PhCOCl + [Mo2Fe638(SEt)3016]3‘ + 6 PhCOZPh (12) Reaction with thiophenol quantitatively converts the phenolate complex to the thiophenolate analog136 (Reaction 13). [Mo2Fe6s8(SEt)3(oph)6]3‘ + 6 PhSH + [Mo2Fe688(SEt)3(SPh)6]3_ + 6 PhOH (13) Of the two atoms potentially susceptible to attack by electrophiles, the sulfur atoms of the bridging ethane— thiolates and the oxygen atoms of the terminal phenolate groups, only the terminal phenolates are reactive under 170 the conditions examined. This parallels the reactivity of the entirely thiolate ligated double cubanesl36’65 con- taining both terminal and bridging thiolate ligands, where bridge integrity was preserved in all cases. The ob— served ligand exchange reactivity is entirely expected based on the evident lability of phenolate ligands and the analogous reactivity of the phenolate Fe-S clusters. 7. Summary Reaction of phenol with an alkyl thiolate ligated double cubane complex effects phenolate substitution at the terminal positions; the product can be isolated as its benzyltriethylammonium salt. The phenolate cluster possesses unaltered magnetic properties, blue shifted op- tical spectra, and undergoes analogous ligand exchange re- actions with electrophiles as expected for terminal pheno— late substitution. Increased isotropic proton NMR shifts, and large negative shifts in corresponding first and second reduction potentials are consistent with increased donation of electron density to the [MoFe3Su]3+ cores for phenolate vs thiophenolate terminal ligands to iron. Similar behavior has been observed for the Fe—S tetramers and dimers and for the molybdenum—iron dimer systems. Several chemical and physiochemical studies of FeMo— couu’84—86, including iron K—edge EXAFS, have suggested that oxygen donor ligands are bound to the iron atoms 171 within the FeMo—cofactor. Also, FeMo—co can be extracted from the MoFe—protein without addition of thiols which are necessary for cluster displacement of typical Fe—S cen~ ters. The results obtained for the [Mo2Fe6S8(SEt)3(OPh)6]3_ ion indicate that phenoxide is a more labile ligand than ethanethiolate. In addition, electrochemical results in— dicate that [Mo2Fe6s8(SEt)3(oph)6]3‘ exhibits reduction potentials in the same range as [Mo2Fe688(SEt)9]3_. These results suggest that substitution of a cysteinyl thiolate ligand by a tyrosyl phenolate ligand to iron in FeMo—co should provide a more labile coordination site without significantly changing cluster reduction potentials. To date, the phenolate double cubane, [M02Fe6S8(SEt)3— (OPh)6]3—, and the phenolate dimer [MoFeSu(OPh)2]2— are the only examples of synthetic MoFeS clusters possessing oxygen ligation to iron. 172 IV. CONCLUSIONS The tetrameric iron-sulfur clusters [FeuSu(OAr)u]2_ (Ar = Ph, p—Tol) have been prepared by two types of ligand exchange reactions and isolated as their tetraalkyl- ammonium salts. The X—ray crystal structure of (EtuN)2— [FeuSu(OPh)u] has been determined and shows an Fe—O bond length somewhat shorter than expected. A variety of physi— cal measurements including optical and 57Fe Mossbauer spectros— copy and proton NMR spectrometry have been performed on these complexes and compared to corresponding results ob— tained for the [FeuSM(SAr)412- complexes. The optical spectra of the phenolate tetramers exhibit blue shifts of 40 nm vs corresponding features in the arenethiolate analogs. Proton magnetic resonance spectra indicate that the isotropic shifts of the phenoxide protons are contact in origin and that their magnitudes are approximately twice as large as those of the thiophenolate tetramers. The Mossbauer spectrum of [Fe4s4(OPh)4]2- has been ob— tained and shows increased ferrous character of the iron atoms upon oxygen substitution. Electrochemical results show approximately 100 mV negative shifts in reduction potentials upon oxygen substitution. Taken together, these results suggest that phenoxide ligands are capable 173 of donating substantial electron density to the [FeuSu]2+ core, presumably through a relatively covalent Fe-O bond. These complexes are also reactive toward electrophiles such as PhCOCl and PhSH and can be smoothly converted to the chloro and thiophenolate tetramers by these reagents, respectively. In addition, certain preparations of phenolate tetramers yield a complex of as yet unknown structure possessing related but distinctly different physical properties; it is probably a higher nuclearity cluster, possibly a hexane. The dinuclear iron—sulfur clusters [Fe2S2(OAr)u]2_ (Ar = Ph, p—Tol, B_C6H4Cl) have been prepared by two methods: direct synthesis, and by ligand exchange reac— tions involving [Fe2S2Clu]2—. They are isolated in pure, crystalline form as their tetraalkylammonium salts. Pre— liminary X—ray crystallographic results indicate a rela- tively short Fe—O bond. Optical spectra of these com— plexes exhibit the expected blue shifts upon oxygen sub— stitution and unaltered magnetic properties. Increased 57Fe Mossbauer isomer shifts isotopic proton NMR shifts, and negative shifts in the reduction potentials compared to results obtained for the corresponding arenethiolate complexes suggest that phenoxide ligands donate sub— stantial electron density to the [Fe2s2]2+ core. The phenoxide dimers undergo ligand exchange reactions by treatment with electrophiles such as PhCOCl and PhSH, which . 2— 2- . yield [Fe2s201u] and [Fe2s2(SPh)u] , respectively. 174 The phenoxide substituted "double cubane” complex [Mo2Fe6S8(SEt)3(OPh)6]3" has been prepared by reaction of [Mo2Fe6SB(SEt)9]2_ with PhOH. The phenoxide cluster ex— hibits the expected blue shifts in the optical spectra and unaltered magnetic properties. Increased isotropic shifts of the phenyl protons in the NMR spectra and nega— tive shifts in the reduction potentials are consistent with increased donation of electron density to the MoFe3Su cores for phenoxide vs thiophenoxide terminal ligands to iron. This complex undergoes ligand exchange reactions with electrophiles as expected for terminal phenolate substitu- tion. To date, terminal phenoxide substitution to iron has been achieved for the one-, two—, and four—iron Fe—S synthetic analogs and for the molybdenum-iron—sulfur dimer and "double cubane" complexes. A schematic representation of these complexes is shown in Figure 35. Results ob- tained for the complexes described in this work, together with results obtained for [Fe(OAr)u]l- and [MoFeSu(OAr)2]2—, reveal several apparently general trends for terminal phen- oxide substitution to iron when compared to results obtraind for the corresponding arenethiolate complexes. These include: i) unaltered magnetic properties, ii) approxi— mately a two fold increase in the isotropic shifts of the phenyl protons; iii) blue shifts of 20—80 nm in correspond— ing Optical absorption bands; iv) an approximately 0.04 175 Figure 35. Schematic of known Fe-S and Mo-Fe—S clusters with phenoxide ligands to iron. 176 OPh " PhO OPh 2' | ,s\ / /Fe\ Fe\S/F< Ph P \OPhO PhO 0Ph 2' Is\ Fe,OPh 2' PhD 5. 5 PhD . \.—./ u. / [5:3 \s 0 ” e- / /// PhO PhO \ s / \ s ‘oph — T 3- PhD OPh \ R / PhO / ys\ /s\ /s\/Fe\ OPh SEFe ;S—M—g—M—S< Fe—/‘s \/Fe>_§s/ \g/ \s/Ffi/ L PhO OPh Figure 35 177 to 0.1 mm/s increase in the 57Fe Mossbauer isomer shift, 6; v) approximately 100—200 mV negative shifts in correspond- ing electrochemical reduction potentials. Finally, these trends suggest that substitution of a cysteinyl thiolate ligand to an Fe—S center by a tyrosyl phenoxide in a protein, or to iron atoms within FeMo—co in the MoFe pro— tein of nitrogenase,would provide a means to a reactive coordination site without significantly altering the redox potentials of the cluster. REFERENCES 10. 11. 12. REFERENCES Ibers, J. A. and Holm, R. H. Science 1980, 209, 223. Holm, R. H. and Ibers, J. A. in Iron—Sulfur Proteins, Vol. 3, W. Lovenberg, Ed., Academic Press, New York, NY, 1977. Ch. 7. a) Averill, B. A. and Orme—Johnson, W. H., in Metal Ions in Biological Systems, Vol. 7, "Iron in Model and Natural Compounds," H. Sigel, Ed., Marcel Decker, Inc., New York, NY, 1978, Ch. 4. b) Beinert, H; Emptage, M. H.; Dryer, J.-L.; Scott, R. A.; Hahn, J. E.; Hodgson, K. 0.; Thomson, A. J. Proc. Natl. Acad. Sci. U.S.A. 1983, 80, 393. Holm, R. H. Acc. Chem. Res. 1977, 10, 427. Berg, J. M. and Holm, R. H. in Iron—Sulfur Proteins, T. G. Spiro, Ed., Wiley—Interscience, New York, NY, 1982, Ch. 1. Carnahan, J. E.; Mortenson, L. E.; Mower, H. F.; Castle, J. E. Biochim. Biophys. Acta 1960, 44, 520. Averill, B. A. Structure and Bonding 1983, 53, 59. Orme—Johnson, W. H.; Davis, L. C. in Iron-Sulfur Proteins, Vol. 3, W. Lovenberg, ed., Academic Press, New York, NY, 1977. Ch. Hageman, R. V.; Burris, R. H. Proc. Natl. Acad. Sci. USA 1978, 75, 2699- Orme-Johnson, W. H.; Davis, L. 0.; Henzl, M. T.; Averill, B. A.; Orme—Johnson, N. R.; Munck. E.; Zimmerman, R. in Recent Developments in Nitrogen Fixation Newton, W. P.; Postgate, J. R.; Rodrigue, N.; eds.; Academic Press, New York, NY, 1977, pp. 131—178. Tso, M.—Y.W.; Burris, R. H. Biochim. Biophys. Acta. 1973, 309, 263. Hageman, R. V.; Orme—Johnson, W. H.; Burris, R. H. Biochemistry 1980, 19, 2333. 178 13. 14. 8.0 15. l6. l7. 18. 19. 20. 21. 22. 23. 24. 25. 179 Zimmerman, R.; Munck, E.; Brill, W. J.; Shah, V. K.; Henzl, M. T.; Rawlings, J.; Orme—Johnson, W. H. Biochim. Biophys. Acta 1978, 537, 185. Huynh, B. H.; Henzl, M. T.; Christner, J. A.; Zimmer— man, R.; Orme—Johnson, W. H., Munck, E. Biochim. Biophys. Acta 1980, 623, 124. Orme-Johnson, W. H.; Lindahl, P.; Meade, J.; Warren, W.; Nelson, M.; Groh, 8.; Orme-Johnson, N. R.; Munck, E.; Huynh, B. H.; Emptage, M.; Rawlings, J.; Smith, J.; Roberts, J.; Hoffmann, B.; Mims, W. B. In Current Perspectives in Nitrogen Fixation, Gibson, A. H.; Newton, W. E., Eds.; Australian Academy of Science, Canberra: 1981, pp. 79—83. Nelson, M. J.; Lindahl, P. A.; Orme-Johnson, W. H. In Advances in Inorganic Biochemistry, Eichorn, G.; Marzilli, L., Ed.; Elsevier: New York, 1982, in press. Wong, G. B.; Kurtz, D. M., Jr.; Holm, R. H., Morten— son, L. E.; Upchurch, R. G. J. Am. Chem. Soc. 1979, 101, 3078. Kurtz, D. M., Jr.; McMillan, R. S.; Burgess, B. K.; Mortenson, L. E.; Holm, R. H. Proc. Natl. Acad. Sci. USA 1979, 76, 4986. Orme-Johnson, W. H.; Davis, L. C.; Henzl, M. T.; Averill, B. A.; Orme—Johnson, N. R.; Munck, E.; Zim— merman, R. In Recent Developments in Nitrogen Fixa- tion, Newton, W.; Postgate, J. R.; Rodriguez-Barrueco, C., Eds.; Academic Press; New York, 1977, pp. 131-178. Rawlings, J.; Shah, V. K.; Chisnell, J. R.; Brill, W. J.; Zimmerman, R.; Munck, E.; Orme-Johnson, W. H. J. Biol. Chem. 1978, 253, 1001. Munck, E.; Rhodes, H.; Orme-Johnson, W. H.; Davis, L. C.; Brill, W. J.; Shah, V. K. Biochim. Biophys. Acta 1975, 400, 32. Smith, B. E.; Lang, G. Biochem. J. 1974, 137, 169. Smith, B. B.; O'Donnel, M. J.; Lang, G.; Spartalian, K. Biochem. J. 1980, 191, 449. Johnson, M. K.; Thomson, A. J.; Robinson, A. E., Smith, B. E. Biochim. BiOphys. Acta 1981, 671, 61. Smith, J. P.; Emptage, M. H.; Orme—Johnson, W. H. J. Biol. Chem. 1982, 257, 2310. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39- 180 Stephens, P. J.; McKenna, C. E.; Smith, B. E.; Nguyen, H. T.; McKenna, M.-C.; Thomson, A. J.; Devlin, F.; Jones, J. B. Proc. Natl. Acad. Sci. U.S.A. 1979, 76, 2585. Cotton, F. A.; Wilkinson, G. in Advanced Inorganic Chemistry: A Comprehensive Text, John Wiley and Sons, New York, 1980, pp. 759. Johnson, R. W.; Holm, R. H. J. Am. Chem. Soc. 1978, 100, 5338. Kanatzidis, M. G.; Ryan, M.; Coucouvanis, D.; Simpoulos, A.; Kostikas, A. Inorg. Chem. 1983, 22, 1979. Johnson, R. E.; Papaefthymiou, G. C.; Frankel, R. B.; Holm, R. H. J. Am. Chem. Soc. 1983, 105, 7280. Shah, V. K.; Brill, w. J. Proc. Natl. Acad. Sci. USA 1977, 74, 3249. a) Burgess, B. K.; Jacobs, D. R.; Stiefel, E. 1. Biochim. Biophys. Acta 1980, 614, 196. b) Nelson, M. J.; Levy, M. A.; Orme-Johnson, W. H. Proc. Natl. Acad. Sci. USA 1983, 80, 147. Burgess, B. K. In Nitrogen Fixation: Chemical/Bio- Chemical/Genetics Interface, Muller, A.; Newton, W. E., Ed.; Plenum: New York, 1982. Shah, V. K.; Brill, W. J. Proc. Natl. Acad. Sci. USA 1981, 78, 3438. Palmer, G.; Multani, J. S.; Cretney, W. C.; Zumft, W. G.; Mortenson, L. E. Arch. Biochem. Bigphys. 1972, 153, 325- Huynh, B. H.; Munck. E.; Orme-Johnson, W. H. Biochim. Biophys. Acta 1979, 527, 192. a) Hofmann, B. M.; Venters, R. A.; Roberts, J. R.; Nelson, M.; Orme-Johnson, W. H. J. Am. Chem. Soc. 1982, 104, 4711. b) Hoffman, B. M.; Roberts, J. B.; Orme-Johnson, W. H. J. Am. Chem. Soc. 1982, 104, 861. Cramer, S. P.; Hodgson, K. 0.; Gillum, W. 0.; Morten- son, L. E. J. Am. Chem. Soc. 1978, 100, 3398. Cramer, S. P. Gillum, W. 0.; Hodgson, K. 0.; Mortenson, L. E.; Stiefel, E. I.; Chisnell, J. R.; Brill, W. J.; Shah, V. K. J. Am. Chem. Soc. 1978, 100, 3814. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 181 Wolff, T. E.; Berg, J. M.; Warrick, C.; Hodgson, K. 0.; Holm, R. H. J. Am. Chem. Soc. 1978, 100, 4630. Teo, B.-K.; Antonio, M. R.; Averill, B. A. J. Am. Chem. Soc. 1983, 105, 3751. Antonio, M. R.; Teo, B. K.; Cleland, W. E.; Averill, B. A. J. Am. Chem. Soc. 1983, 105, 3477. Teo, B.-K.; Antonio, M. R.; Tiekelmann, R. H.; Silvis, H. C.; Averill, B. A. J. Am. Chem. Soc. 1982, 104, 6126. Antonio, M. R.; Teo, B.—K.; Orme-Johnson, W. H.; Nelson, M. J.; Groh, S. E.; Lindahl, P. A.; Kauzlarich, S. M.; Averill, B. A. J. Am. Chem. Soc. 1982, 104, 4703. Shah, V. K.; Chisnell, J. R.; Brill, W. J. Biochim. BiOphys. Res. Commun. 1978, 81, 232. Newton, W. E. in Kirk-Othmer: Encyclopedia of Chem— ical Technology, Vol. 15, 3rd Edn.; John Wiley and Sons: New York, 1981, pp. 942-968. Mortenson, L. E.; Thornley, R. N. F. Ann. Rev. Bio— chem. 1979, 48, 387. Holm, R. H. Chem. Soc. Rev. 1981, 10, 455. Wolff, T. E.; Berg, J. M.; Hodgson, K. 0.; Frankel, R. B.; Holm, R. H. J. Am. Chem. Soc. 1979, 101, 4140. Wolff, T. E.; Berg, J. M.; Power, P. P.; Hodgson, K. 0.; Holm, R. H.; Frankel, R. B. J. Am. Chem. Soc. 1979, 101, 5454. Wolff, T. E.; Power, P. P.; Frankel, R. B.; Holm, R. H. J. Am. Chem. Soc. 1980, 102, 4694. Wolff, T. E.; Berg, J. M.; Power, P. P.; Hodgson, K. 0.; Holm, R. H. Inorg. Chem. 1980, 19, 430. Christou, G.; Garner, C. D.; Mabbs, F. E. Inorg. Chim. Acta 1978, 28, L189. Christou, G.; Garner, C. D.; Mabbs, F. E.; King, T. J. J. Chem. Soc. Chem. Commun. 1978, 740. Acott, S. R.; Christou, G.; Garner, C. D.; King, T.J.; Mabbs, F. E.; Miller, R. M. Inorg. Chim. Acta 1979, 35, L337. 56. 57. 58. 59. 60. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 182 Christou, G.; Garner, C. D.; Miller, R. M. J. Inorg. Biochem. 1979, 11, 349. Christou, G.; Garner, C. D.; Mabbs, F. E.; Drew, M. G. B. J. Chem. Soc., Chem. Commun. 1979, 91. Christou, G.; Garner, C. D. J. Chem. Soc., Dalton Trans. 1980, 2354. Christou, G.; Garner, C. D.; Miller, R. M.; Johnson, C. E.; Rush, J. D. J. Chem. Soc., Dalton Trans. 1980, 2363. Christou, G.; Mascharak, P. K.; Armstrong, W. H.; Papaefthymiou, G. C.; Frankel, R. B.; Holm, R. H. J. Am. Chem. Soc. 1982, 104, 2820. Wolff, T. E.; Berg, J. M.; Holm, R. H. Inorg. Chem. 1981, 20, 174. Armstrong, W. H.; Holm, R. H. J. Am. Chem. Soc. 1981, 103, 6246. Palermo, R. E.; Power, P. P.; Holm, R. H. Inorg. Chem. 1981, 20, 173. Christou, G.; Garner, C. D. J. Chem. Soc. Chem. Commun. 1983, 613. Christou, G.; Collison, D.; Garner, C. D.; Mabbs, F. E.; Petrouleas, V. Inorg. Nucl. Chem. Lett. 1981, 17, 137. Armstrong, W. H.; Mascharak, P. K.; Holm, R. H. J. Am. Chem. Soc. 1982, 104, 4373. Armstrong, W. H.; Mascharak, P. K.; Holm, R. H. Inorg. Chem. 1982, 21, 1699. Mascharak, P. K.; Armstrong, W. H.; Mizobe, Y.; Holm, R. H. J. Am. Chem. Soc. 1983, 105, 475- Palermo, R. E.; Holm, R. H. J. Am. Chem. Soc. 1983, 105, 4310. Coucouvanis, D.; Simhon, E. D.; Swenson, D.; Baenziger, N. C. J. Chem. Soc., Chem. Commun. 1979, 361. Tieckelmann, R. H.; Silvis, H. C.; Kent, T. A.; Huynh, B. H.; Waszczak, J. V.; Teo, B.—K.; Averill, B. A. J. Am. Chem. Soc. 1980, 102, 5550. 73. 74. 75- 76. 77. 78. 79. 80. 81. 82. 83. 84. 85. 86. 87. 183 Coucouvanis, D.; Baenziger, N. C.; Simhon, E. D.; Stremple, P.; Swenson, D.; Simopoulos, A.; Kostikas, A.; Petrouleas, V.; Papaefthymiou, V. J. Am. Chem. Soc. 1980, 102, 1732. Muller, A.; Tolle, H.—G.; Bogge, H. Z. Anorg. Allg. Chem. 1980, 471, 115. Tieckelmann, R. H.; Averill, B. A. Inorg. Chim. Acta 1980, 46, L35. Muller, A.; Sarkar, S.; Dommrose, A.-M.; Filgueira, R. Z. Naturforsch., Tei1.y B. 1980, 35, 1592. McDonald, J. W.; Friesen, G. D.; Newton, W. E. Inorg. Chim. Acta 1980, 46, L79. Coucouvanis, D.; Simhon, E. D.; Baenziger, N. C. 1; Am. Chem. Soc. 1980, 102, 6644. Dahlstrom, P. L.; Kumar, S.; Zubieta, J. J. Chem. Soc. Chem. Commun. 1981, 411. Coucouvanis, D.; Simhon, E. D.; Stremple, P.; Baen- ziger, N. C. Inorg. Chim. Acta 1981, 53, L135. Silvis, H. C.; Averill, B. A. Inorg. Chim. Acta 1981, 54, L57- Coucouvanis, D.; Stremple, P.; Simhon, E. D.; Swenson, D.; Baenziger, N. C.; Draganjac, M.; Chan, L. T.; Simopoulos, A.; Papaefthymiou, V. Inorg. Chem. 1983, 22, 293. Coucouvanis, D. Acc. Chem. Res. 1981, 14, 201. Burgess, B. K.; Stiefel, E. 1.; Newton, W. E. 1; Biol. Chem. 1980, 255, 353. Burgess, B. K.; Yang, S.—S.; You, C.-B.; Li, J.-G.; Friesen, G. D.; Pan, W.-H.; Stiefel, E. 1.; Newton, W. E. in Current Perspectives in Nitrogen Fixation, Gibson, A. H.; Newton, W. E., Ed.; Australian Academy of Science: Canberra, 1981; pp. 71—74. Newton, W. E.; Burgess, B. K.; Stiefel, E. I. in Molybdenum Chemistry of Biological Significance, Newton, W. E.; Otsuka, 8., Ed.; Plenum Press: New York, 1980; pp. 191. Ghosh, D.; Furey, W., Jr.; O'Donnell, 8.; Stout, C. D. J. Biol. Chem. 1981, 256, 4185. , 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 184 Antonio, M. R.; Averill, B. A.; Moura, 1.; Moura, J. J. G.; Orme-Johnson, W. H.; Teo, B.-K.; Xavier, A. v. J. Biol. Chem. 1982, 257, 6646. Koch, S. A.; Millar, M. J. Am. Chem. Soc. 1982, 104, 5255- a) Cleland, W. E.; Averill, B. A. Inorg. Chim. Acta 1981, 56, L9 b) Cleland, w. E.; Holtman, D. A.; Sabat, M.; Ibers, J. A.; DeFotis, G. C.; Averill, B. A. J. Am. Chem. Soc. 1983, 105, 6021. Averill, B. A.; Herskovitz, T.; Holm, R. H.; Ibers, J. A. J. Am. Chem. Soc. 1973, 95, 3523. Christou, G.; Garner, C. D. J. Chem. Soc., Dalton Trans. 1979, 1093. Wong, G. B.; Bobrick, M. A.; Holm, R. H. Inorg. Chem. 1978, 17, 1093. Holm, R. H.; Phillips, W. D.; Averill, B. A.; Mayerle, J. J.; Herskovitz, T. J. Am. Chem. Soc. 1974, 96, 2109. DePamphilis, B. V.; Averill, B. A.; Herskovitz, T.; Que, L., Jr.; Holm, R. H. J. Am. Chem. Soc. 1974, 96, 4159. Reynolds, J. G.; Holm, R. H. Inorg. Chem. 1980, 19, 3257. Mayerle, J. J.; Denmark, S. E.; DePamphilis, B. V.; Ibers, J. A.; Holm, R. H. J. Am. Chem. Soc. 1975, 97. 1032. Ryan, J. L. Inorg. Synth., Vol. XV, pp 231. Kruss, G. Justus Liebigs Ann. Chem. 1884, 225, 6. Emptage, M. H.; Zimmerman, R.; Que, Jr., L.; Munck, E. Hamilton, W. D.; Orme—Johnson, W. H. Biochim. Biophys. Acta 1977, 495, 12. Que, L., Jr.; Bobrik, M. A.; Ibers, J. A.; Holm, R. H. J. Am. Chem. Soc. 1974, 96, 4168. Bobrik, M. A.; Hodgson, K. O.; Holm, R. H. Inorg. Chem. 1977, 16, 1851. Gerloch, M.; Mabbs, F. E. J. Chem. Soc. A. 1967, 1598. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 185 a) Coggon, P.; McPhail, A. T.; Mabbs, F. E.; Mc- Lachlan, V. N. J. Chem. Soc. A. 1971, 1014. b) Gerloch, M.; McKenzie, E. D.; Towl, A. D. C. J. Chem. Soc. A. 1969, 2580. a) Anderson, B. F.; Buckingham, D. A.; Robertson, G. R.; Webb, J. Nature (London) 1976, 262, 7222. b) Raymond, K. N.; Isied, S. S.; Brown, L. D.; Fronczek, F.6F.; Nibert, J. H. J. Am. Chem. Soc. 1976, 98, 17 7. Beattie, J. K.; Moore, C. J. Inorg. Chem. 1982, 21, 1292. Heistand, R. H.; Roe, A. L.; Que, L., Jr. Inorg. Chem. 1982, 21, 676. Geurtz, P. J. M.; Gosselink, J. W.; Van Der Avoird, A.; Baerends, E. J.; Snijders, J. G. Chem. Phys. 1980, 46, 133. Aizman, A.; Case, D. A. J. Am. Chem. Soc. 104, 3269. Yang, C. Y.; Johnson, K. H.; Holm, R. H.; Norman, J. G. J. Am. Chem. Soc. 1975, 97, 6596. Thomson, A. J. J. Chem. Soc. Dalton Trans. 1981, 1180. Figis, B. N.; Lewis, J.; In Technique of Inorganic Chemistry, Vol. IV, Jonassen, H. B. and Weissberger, Ed., Interscience, New York, NY, 1965, pp. 137-248. Laskowski, E. J.; Frankel, R. B.; Gillum, W. 0.; Papaefthymiou, G. C.; Renaud, J.; Ibers, James A.; Holm, R. H. J. Am. Chem. Soc. 1978, 100, 5322. Herskowitz, T.; Averill, B. A.; Holm, R. H.; Ibers, J. A.; Phillips, W. D.; Weiher, J. F.; Proc Natl. Acad. Sci. USA 1972, 69, 2437. Hatfield, W. E.; In "Theory and Applications of Molecular Paramagnetism", Boudreux, E. A. and Mulay, L. N., Eds.; Wiley-Interscience: New York, NY, 1976, Chapter 7. Antanaitis, B. C.; Moss, T. H. Biochim. Biophys. Acta 1975, 405, 262. Papaefthymiou, G. C.; Laskowski, E. J.; Frosta— Pessoa, S.; Frankel, R. B.; Holm, R. H. Inorg. Chem. 1982, 21, 1723. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 128. 129. 130. 131. 132. 133. 134. 186 Horrocks, Jr., W. D.; In "NMR of Paramagnetic Mole- cules: Principles and Applications", LaMar, G. N.; Horrocks, Jr., W. D.; Holm, R. H., Ed.; Academic Press: New York, 1973; Chapter 4. Frankel, R. B.; Averill, B. A.; Holm, F. H. J. Phys. (Orsay, Fr.) 1974, 35, C6-107. Van Tamelen, E. E.; Gladysz, J. A.; Brulet, C. R.; J. Am. Chem. Soc. 1974, 96, 3020. Johnson, R. W.; Holm, R. H. J. Am. Chem. Soc. 1978, 100, 5338. Dukes, G. R.; Holm, R. H. J. Am. Chem. Soc. 1975, 97. 528. Que, L., Jr.; Anglin, J. R.; Bobrick, M. A.; Davison, A.; Holm, R. H. J. Am. Chem. Soc. 1974, 96, 6042. Hill, C. L.; Renaud, J.; Holm, R. H.; Mortenson, L. E. J. Am. Chem. Soc. 1977, 99, 2549. Averill, B. A.; Bale, J. R.; Orme-Johnson, W. H. J. Am. Chem. Soc. 1978, 100, 3034. Coucouvanis, D.; Swenson, D.; Stremple, P.; Baenziger, N. C. J. Am. Chem. Soc. 1979, 101, 3392. Ackermann, G.; Hesse, D. Z. Anorg. Allg. Chem. 1970, 375, 77. Gillum, W. 0.; Frankel, R. B.; Foner, S.; Holm, R. H. Inorg. Chem. 1976, 15, 1095. Laskowski, E. J.; Reynolds, J. G.; Frankel, R. B.; Foner, 8.; Papaefthymiou, G. C.; Holm, R. H. 1; Am. Chem. Soc. 1979, 101, 6562. Que, L., Jr.; Coord. Chem. Revs. 1983, 50, 73. Aisen, P.; Listowsky, 1. Ann. Rev. Biochem. 1980, 49, 357. Davis, J. C.; Averill, B. A. Proc. Natl. Acad. Sci. USA, 1982, 79, 4623. Antanaitis, B. C.; Strekas, T.; Aisen, P. J. Biol. Chem., 1982, 257, 3766. Que, L., Jr.; Epstein, R. M. Biochemistry 1981, 20, 2545. 187 135. Hawkins, C. J.; Parry, D. L.; Wood, B. J. Inorg. Chim. Acta 1983, 78, L29. 136. Palermo, R. E.; Power, P. P.; Holm, R. H. Inorg. Chem. 1982, 21, 173. 137. Reynolds, J. G.; Laskowski, E. J.; Holm, R. H. J; Am. Chem. Soc. 1978, 100, 5315. 138. Kanatzidis, M. G.; Dunham, W. R.; Hagen, w. R.; Coucouvanis, D.; J. Chem. Soc. Chem. Commum. 1984, 356. 139. Saak, W.; Henkel, G.; Pohl, S. Angew. Chem. Int. Ed. Eng. 1984, 23. 140. Fee, J. A.; Findling, K. L.; Yoshida, T.; Hille, R.; Tarr, G. E.; Hearshen, D. 0.; Dunham, W. R.; Day, E. P.; Kent, T. A.; Mfinck, E. J. Biol. Chem. 1984, 259, 124.