i [y “Av 'd-.r-pn|.‘l—~H- ‘1 D to! fO'A : .. SQ, «0;.4 I ‘ma.’ Prv'l a I. .1 a I o F a n ,_ . .._, ’r' I a 1'3: x’nflkfi‘ f-fi!.u bureau—v- - II o 01 an 3‘." Jr». rm- ,1 I a. ’v ‘5‘ an In. in I This is to certify that the dissertation entitled OXYGEN-LIGATED FE-S AND MO-FE-S CLUSTERS AS POSSIBLE MODELS OF NITROGENASE presented by Miriam Elaine Rogers has been accepted towards fulfillment of the requirements for Ph . D . degree in Chegis try / Major professor Date June 26. 1986 MS U i: an Affirmative Action/Equal Opportunity Institution 0-12771 MSU LIBRARIES W RETURNING MATERIALS: Piace in book drop to remove this checkout from your record. Elfi§§_wi11 be charged if book is returned after the date stamped beiow. n‘ 7 fl OXYGEN-LIGATED FE—S AND MO-FE-S CLUSTERS AS POSSIBLE MODELS FOR NITROGENASE By Miriam Elaine Rogers A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1986 ABSTRACT OXYGEN-LIGATED FE-S AND MO—FE-S CLUSTERS AS POSSIBLE MODELS OF NITROGENASE By Miriam Elaine Rogers A tridentate ligand, l,8,l3-tris[(N-4-hydroxyphenyl)carboxamido]- triptycene ((HO),-tripod), was designed and synthesized. Complexation of . the ligand with a tetrameric iron-sulfur cluster was examined in order to prepare an oxygen-ligated mixed-ligand Fe.S. cluster as a possible model for the P—clusters of nitrogenase. The basal portion of the ligand is a 1,8,13-trisubstituted triptycene, prepared from the Dials-Alder reaction of a 1,8-disubstituted anthracene with an ortho-substituted benzyne. In the process of synthesizing this macrocyclic ligand, an investigation of the factors governing the regioselectivity of the cycloaddition reaction was conducted. Twenty-one new trisubstituted triptycene derivatives and a new synthetically useful benzyne precursor, Z-amino-S-methoxycarbonyl benzoic acid, were prepared and characterized by ‘H NMR, ”0. NMR and IR spectroscopies, as well as mass spectrometry. Proton nuclear magnetic resonance studies of the reaction between [Pr.N],[Fe.S.(SEt).] and the (OH),-tripod ligand or between Miriam Elaine Rogers [Et.N],[Fe.S.Cl.] and the trisodium salt of the (H0),-tripod ligand were conducted. Analysis of the isotropically shifted peaks, by comparison with reported chemical shifts of related iron-sulfur clusters, indicates a mixture of tetrameric (Fe‘84) and hexameric (Fe‘Sd forms of the ligated cluster is generated in solution. An oxygen-ligated linear Mo-Fe-S cluster, [Et.N]¢[S,MoS,Fe(OAc),], was prepared as a possible model for the FeMo-cofactor of nitrogenase. Tetraethylammonium tetrathiomolybdate was reacted with anhydrous ferrous acetate in acetonitrile to yield the acetate-ligated dimer. The optical features are similiar to other dimers that have an FeS .Mo core. Variable temperature ‘H NMR and magnetic susceptibility measurements show paramagnetism that follows Curie Law behavior. An observed peff = 4.9 BM is consistent with an S = 2 ground state and indicates an Fe(II)-M0(VI) description of this complex. To my parents: Mr. and Mrs. R. Rogers ii ACKNOWLEDGMENTS I would like to acknowledge D. A. Holtman for assistance in obtaining temperature-dependent magnetic susceptibility data on [Et.N];[S;MoS,Fe(OAc);] and Dr. D. Herold for providing high resolution mass spectra on several trisubstituted triptycenes. I would like to thank Drs. Brian Ward, J. Michael Williams and Vijay Kumar for their helpful discussions and suggestions concerning synthetic problems. I am appreciative that Professors Pinnavaia, Eick, Chang, LeGoff, Hart and Averill have served on my Committee. Special thanks to Dr. Hart,lfor allowing me to work in his labs for a year; to Dr. Chang for his patience and understanding on many long-distance phone calls concerning problems of working in absentia; to Dr. P. for assuming Chairmanship of my Committee at the last minute; and to Dr. Rick for serving as second reader. . I am grateful to Michigan State University, the University of Virginia and Dr. Averill for financial support during my graduate studies. I thank the Department of Chemistry at M.S.U. for permitting me to complete my research at U.Va. I would like to thank my coworkers: Paul Lamberty, Gay Lilley, Elise Ponzetto, Susan Hefler, Charles Hulse, William Frazier (my running coach), Teresa Zirino (my fellow cohooter) and Douglas Holtman (Krossword King (oops, I spelled it wrong” for making the lab a more pleasant place to be. In addition, I would like to thank my "bosses" in the NMR lab: Dr. J. Scott Sawyer (fearless leader of the F-team) and iii Dr. Jeffrey Ellena (NMR Czar) for interesting times around the magnet. More than thanks are owed to Mr. Michael Hoard, whose guidance and enduring friendship throughout the (past eight years has meant a great deal to me (and may have changed my life); to Dr. Susan Kauzlarich for her constant encouragement, advice and friendship; and to my parents, for believing in me. iv TABLE OF CONTENTS Chapter LISTOFTABLES.............. LIST OF FIGURES. . . . . . . . . . . . . . LIST OF ABBREVIATIONS. . . . . . . . . . . . I. INTRODUCTION. . . . . . . A. Nitrogenase and its relevance B. Synthetic models of nitrogenase. . . . . C. Design of oxygen-ligated models of nitrogenase. II. RESULTS AND DISCUSSION. . . . . . . . . . A. (B0),-Tripod ligand . . . . . . . 1. Synthesis . . . . . 2. Physical properties . . . . . . . . B. Binding study of the (BO),-tripod ligand to an Fe.S. core. . . . . . . . . 1. Synthetic approach . . . . . . . . 2. Proton nuclear magnetic resonance studies. 0. [Et.N]3[S,MbS,Fe(OAc),] . . . . . . 1. Synthesis . . . . . . . . . . . 2. Physical properties . a. Optical . . . . . . . . b. Proton nuclear magnetic resonance. c. Magnetic susceptibility . . . III. EXPERIMENTAL . . . . . . . . A. Materials . . B. Physical methods. . Page vii . viii xi 18 24 24 24 46 67 67 68 71 74 78 78 79 Chapter Page 0. Preparation of 1,8-disubstituted anthraquinones . . . 80 D. Preparation of 1,8-disubstituted anthracenes . . . . 81 E. Preparation of o-methoxycarbonylbenzyne precursor. . . 83 F. Preparation of 1,8,13- and 1,8,16-trisubstituted triptycenes . . . . . . . . . . . . . . . 84 G. Preparation of 1,8,13- and 1,8,lG—tris[(N—substituted)- carboxamidOJtriptycenes . . . . . . . . . . . 95 He Remtion 0f [Pr‘N] 2 [Fe.S. (Sgt) g] With (BO),-tripod ligand. . . . . . . . . . . . . 98 I. Reaction of [Et.N],[Fe.S.Cl,] with Na,(0,-tripod) . . . . . . . . . . . . . . 98 1. Preparation of Na,(O,-tripod). . . . . . . . 98 2. Reaction of chloro tetra-er with the trisodium salt of (BO),—tripod. . . . . . . . . 99 J. Preparation of [Et.N],[S,MoS,Fe(OAc),] . . . . . . 99 IV. CONCLUSIONS . . . . . . . . . . . . . . . . 101 REFERENCES . . . . . . . . . . . . . . . . . . 104 Table II III IV VI LIST OF TABLES Page 3We MBssbauer isomer shifts and quadrupole splittings of various Fe-S clusters. . . . . . . . 19 Ratios of anti to syn isomers of trisubstituted triptycenes. . . . . . . . . . . . . . . . 36 Calculated and observed 1’0 NMR shifts for aromatic resonances of the (RO),-tripod ligand . . . . . . . 44 Isotropic shifts for the ligand protons in [Fe¢34(03-tril>°d) (51%)] ” and [host (Os‘tripod) a] "" species in Measo-d. solution . . . . . . . . . . 50 Electronic spectral features and molar absorptivities for [SgMOSaFO(OAC)3]2-, [SaflosaFeCI,]", and [SaflosaFe(OPh),]" complexes . . . . . . . . . . 70 Isotropic proton nuclear magnetic resonance shifts ((AH/B)iso) observed for the OAc group of [Eth]3[S,M083Fe(OAC)3]3- s s s s . s s s s s s 73 Figure 10 LIST OF FIGURES 4 Page Schematic drawing of the electron transport chain of nitrogenase. . . . . . . . . . . . . . 4 Schematic drawing of fear types of Fe-S centers that occur in proteins. . . . . . . . . . . . . 5 Schematic drawing of structural models proposed for the FeMo-cofactor. . . . . . . . . . . . . 7 ”Fe Massbauer spectrum of the MoFe protein, taken at 30 K (from ref. 16) . . . . . . . . . . . . 8 Schematic drawing of the P—clusters of nitrogenase based on the ”Fe Massbauer and magnetic properties of the MoFe protein. . . . . . . . . . . . 10 Schematic drawing of structurally characterized Mo-Fe-S clusters. . . . . . . . . . . . . 12 Schematic drawing showing the distortions of the [Fe.S.]+ core in: (left) [Et,MeN],[Fe.S,(SPh).]; (right) [Et.N],[Fe.S.(SCB,Ph).]. From ref. 40b . . 14 Possible models for the P—clusters, involving oxygen ligation at three vertices of a 4Fe-4S core . . . . . . . . . . . . . . 16 Schematic drawing of the transformation of ”hexamers" to ”tetramers”. . . . . . . . . . l7 Sch-atic drawing of [Fe.S.(L) (SR)]"; where L is a tridentate ligand. . . . . . . . . . . . 21 Figure 11 12 13 14 15 16 17 18 19 20 Page Schematic drawing of the (HO),-tripod ligand. . . . 22 Schematic drawing of a triptycene structure showing the numbering of atoms . . . . . . . . 25 Schematic of a synthetic approach toward the preparation of triptycene 1,8,13-tricarboxylic acid . . . . . . . . . . 26 Schematic of linear versus convergent synthetic approaches . . . . . . . . . . . . . . . 30 Schematic of convergent synthesis approach to triptycene 1,8,13-tricarboxylic acid . . . . . . 31 Schematic drawing of the synthesis of the (R0),-tripod ligand . . . . . . . . . . . . 34 "C NMR (360 MHz) of the (B0),-tripod ligand in Me,SO-d.. . . . . . . . . . . . . . . . 43 Schematic showing spin density stabilization at the ortho and para positions of an aromatic ring. . . . 49 ‘B NMR spectra (360 MHz) of the (BO),-tripod ligand and its reaction with [Pr.N],[Fe.S.(SEt).] in Measo-d. after a 4 h period and a 12 h period . . . 52 ‘B NMR spectra (360 MHz) of Na,(O,-tripod), reaction between Na,(0,-tripod) and [Et.N],[Fe.S.Cl.], and reaction of [Pr.N],[Fe‘S.(SEt).] with (B0),-tripod and five equivalents of PhSR in Measo—d. . . . . . . . . 57 Figure 21 22 23 24 25 26 27- Page ‘H NMR spectra (360 MHz) of the reaction between [Et.N],[Fe.S.Cl.] and Na,(O,-tripod) in Me;SO-d.. The ratio of cluster to ligand (C:L) used in the reactions is indicated next to each spectrum. . . . 61 Schematic of possible Fe-S clusters present fra- interconversions of an [Fe.S.(O,R)(Cl)]" cluster in solution. . . . . . . . . . . . . . . 63 Schematic drawing of two possible coordination modes of the (OH),-tripod ligand to an Fe.S. hexagonal core. . . . . . . . . . . . . . 65 Electronic spectrum of [Et.N],[S,MoS,Fe(OAc),] in CB,CN. . . . . . . . . . . . . . . . 69 ‘B NMR spectra (360 MHz) of [Et‘N],[S,MoS,Fe(OAc),] in CD,CN at various temperatures. . . . . . . . 72 Plot of isotropic shift ((AB/H)iso) versus l/T for [Et.N],[S,MoS;Fe(OAc),] . . . . . . . . . 75 Plot of x versus T for temperature-dependent magnetic susceptibility data on [Et.N],[S,MoS,Fe(0Ac),]. . . . . . . . . . . 76 ATP BM t-Bu Bu .N" cat CPE DEAE DMA DME dtc EPR Et,N Et.N'* EtOAc EtOH EtSR EXAFS FeMo—co glu (Hob-431M LIST OF ABBREVIATIONS adenosine 5’ -diphosphate arene adenosine 5 ' -triphosphate Bohr magneton tertiary-butyl tetrabutylammonium catechol Corey-Pauling—Koltun diethylaminoethyl N,N-dimethylacetamide dimethoxyethane dithiocarbamate electron paramagnetic resonance ethyl triethylamine tetraethylammonium ethylacetate ethanol ethanethiol extended X-ray absorption fine structure iron-molybdenum cofactor gas chromatography glutamate l ,8, 13-Tris[ (N-4-hydroxyphenyl)carboxamido]triptycene xi HPLC IR MCD Me MeOR Me,so MS Na, (0,-tripod) NBS NMR O,R ”xyls 3 Ph Pr,N"’ ref. TLC ms tyr high-pressure liquid chromatography infrared meta magnetic circular dichroism methyl methanol dimethylsulfoxide mass spectrometry trisodium salt of (HO),-tripod N—bromosuccinimide nuclear magnetic resonance -O,-tripod ortho ortho—xylyldithiolate para inorganic phosphate phenyl tetrapropylammonium reference thin-layer chromatography tetramethylsilane tyrosinate I. INTRODUCTION A. Nitrogenase and its relevance Nitrogenase is the metalloenzyme responsible for the catalytic reduction of dinitrogen to ammonia in biological systemsl. The protein is found in prokaryotic microorganisms ranging from strict anaerobes to strict aerobes. The enzyme reduces nitrogen (with concomitant reductive dephosphorylation of ATP to ADP) at room temperature and atmospheric pressure, as shown in equation 1. N, + 811* + 8s" + lZA’l‘P —-—» ZNH, + IZADP + 12Pi + H, (1) catalyst: nitrogenase temperature: ambient pressure: 1 atm. Industrially, the Haber-Bosch process is used to reduce dinitrogen by reacting three moles of hydrogen with one mole of nitrogen over a catalyst of alumina treated with iron and potassium oxide, as shown in equation 2. 3H, + N, ——. 2MB, (2) catalyst: M303/K30/Fe temperature: 450 '0 pressure: 200-300 atm. 2 This process, which supplies the most widely used form of nitrogen fertilizer (NIL), requires a high operating temperature (450 ‘C) and pressure (200-300 atm.). Nonrenewable fossil fuels, such as natural gas and liquid hydrocarbons, are used as sources of hydrogen, as well as to supply energy to achieve the harsh operating conditions. Inorganic chemists are trying to duplicate the unique physical and chemical properties of nitrogenase in order to develop synthetic analogues that approximate the catalytic properties of the biomoleculez. The long-range goal of this type of research is to synthesize a model that could eventually be developed into an efficient commercial catalyst. Nitrogenase can be separated into two protein components, the Fe protein and the MoPe protein, by anion exchange (DEAR-cellulose) chromatography. The Fe protein has a molecular weight of 60,000 Daltons and has been shown to contain a single tetranuclear 4Fe-4S cluster3. The Fe protein binds two MgATP units to form an Fe protein-(MgATP) , complex4. The MoFe protein consists of two subcomponents which have a combined molecular weight of 220,000 Daltons. Assays have shown the presence of 32 irons, 32 acid-labile sulfurs and 2 molybdenums per MoFe protein3. An isolatable sub- component, from the extraction of the MoFe protein with N-methyl- formamide, contains 1 Mo, 6-8 Fe and 6-10 S= per mole. This subunit is essential for enzymatic activity and has‘ been termed the iron- molybdenum cofactor, or FeMo-co. A mutant strain (W45) of Azotobacter vinelandii produces an inactive MoFe protein, which lacks the cofactor unit, and consequently the ability to reduce dinitrogens. The other subcomponent of the MoFe protein consists of two tetrameric iron-sulfur clusters which were identified by either cluster 3 displacement experiments6 (1 ’F NMR detection of substituted fluorinated thiols) or by cluster transfer techniques5 (EPR detection of Fe-S clusters transferred into apoproteins). These clusters are a variant of the normal 4Fe-48 clusters and have been referred to as P-clusters7. The MoFe and Fe proteins function together as an electron transport chain. A schematic representation of the nitrogenase system is shown in Figure 18. Electrons are transferred from an initial electron donor such as a reduced ferredoxin to the Fe protein-(MgATP), complex9. The binding of MgATP to the iron protein may serve to make electron transfer more efficient by causing a conformational change in the protein structurelo. Electrons are then shuttled (via the P-clusters, which may act as electron reservoirs) to the MoFe cofactor. This is the site of the actual reduction process. For every two electrons transferred to substrate, four to five moles of ATP are hydrolysed to ADP4. Substrates that can be reduced (besides dinitrogen) are azide, alkynes and isonitriles, while carbon monoxide inhibits activity. A great deal of research has been aimed at deducing the exact structure of the metal centers in the MoPe proteinstn. Four different types of clusters, shown in Figure 2, have been found in various other iron-sulfur proteins by X—ray crystallography. The extreme oxygen sensitivity of nitrogenase and its high molecular weight have limited the amount of structural information obtainable by X-ray crystallography. The technique of EXAFS (extended X—ray absorption fine structure) spectroscopy has been employed to study the immediate environment of the Mo and Fe atoms in the protein. Mo K-edge scattering has been analyzed as arising from 4 S atoms at 2.35 A, 2-3 Nitrogenose /’I, \I N iii 7 '53 cu W“ ‘1‘ a: “l \---- x 1 9 as .2 * I.) a N I Fe Protein MoFe Protein Figure 1. Schematic drawing of the electron transport chain of nitrogenase. I \ /s\ SR [Fe-4S] [2 F's-ZS] RS Fe RS -— '\"" SR R 5% \Fe/ s\f-!e\/ S, \ s / \s/ | g_ -S L SR RS " I 7 F I 8/ 9 SR [ere-as] [4Fe-4S] ' Figure 2. Schematic drawing of four types of Fe-S centers that occur in proteins. 6 Fe atoms at 2.72 A and 1-2 additional 8 atoms at 2.5 A from M012o13. The Fe K-edge EXAFS data on the MoFe protein suggest 1.3 1; 1.0 O (or N) atoms at 1.8 A, 3.4 3; 1.6 S atoms at 2.25 A, 2.3 3; 0.9 Fe atoms at 2.66 A and 0.4 1; 0.1 M0 at 2.76 A from Fe“. Low temperature (8-25 K) EPR spectra of the MoFe protein display g values near 4.32, 3.65 and 2.0115. This is quite different from a normal tetranuclear cluster which has g values near or below g = 2.0. Several structures, shown in Figure 3, have been proposed for FeMo-co based on the physical data obtaineda. The ”Fe Mdssbauer spectrum of the MoFe protein, shown in Figure 4, consists of four quadrupole doublets, referred to as "M", ”S", "D” and "Fe""" components“. Least squares fitting of doublet "8" indicates that it accounts for only two iron atoms per holoenzyme, and it is uncertain whether this is due to a persistent impurity or if this iron is an integral part of the MoFe protein. Component "M" is assigned to the metal cluster of FeMo—co, while components "D" and "Fe""" are assigned to the P-clusters. M8ssbauer spectra taken in strong magnetic fields show that the P-clusters are diamagnetic (S = 0) in the resting enzyme. The ratio of intensities of the D:Fe"" doublets is 3:1. Analysis of isomer shifts (6 = 0.69 mm/s) and quadrupole splitting (AEq = 3.02 mm/s) indicates that the Fe" doublet represents high-spin iron(II) coordinated tetrahedrally to sulfur. The more intense doublet (D) has an isomer shift 6 = 0.64 mm/s, (which is in the range of Fe“ shifts), but has an unusually small quadrupole splitting (AEq = 0.81 mm/s). It is concluded that the Fe“ and D components make up a set of spin-coupled Fe,S. clusters (P-clusters), with each cluster containing one Fe" and three D ferrous atoms as illustrated in / \ 5/ s\?/S\ \ “as For. / \ / ' ‘ M0 F F! ‘M F9 "s’ \s/.\ / ‘s/ | ‘5’ \ L 7 \l‘e—S S—Fe / \a’s‘m’ 3‘ / \s \ / ‘s’ I ‘s \ / Fe—S L 8—H; I‘ t / x s - — Fe—S \ . z \ R I) / MO-Sl ‘Fe 3 Fai’ / I ._.s "Fa-l— FG- -.§ :45 {—ng / I_Tf.\ |’_g/ \ s V l \g L RS 2" ' a; .. Rsl\ y5\a’ fire I, ’ .158 "5 ('3;on s‘iII’ Isa L/\°\L Figure 3. Schematic drawing of structural models proposed for the FeMo-cofactor. 0.0 A 7_A__.' T‘ W Y 1 T r v _1, vy/ . M , x! *1 in EFFECT IN PER CENT --._..__.L_-._. L U r *flo-a-d‘. R a“... ..-.1 .. «.0 -3.0 -2.0 -l.0 0.0 1.0 2.0 3.0 4.0 VELOCITY IN (MM/SEC) Figure 4. "Fe MBssbauer spectrum of the MoFe protein, taken at 30 K (from ref. 16.). Figure 5. Additional studies with combined EPR, Mdssbauer and MCD spectroscopies” indicate that the oxidation state of the Fe,S. core in the native enzyme (PN) is probably zero (4 ferrous atoms), while magnetic suseptibility measurements 18 of the oxidized MoFe protein show an S = ’/a spin state for the oxidized P-clusters (Pox). B. . Synthetic models of nitrogenase Many experimentalists have tried to mimic the composition and spectroscopic properties of the structurally unique metal cluster of FeMo-co. Three classes of Mo-Fe-S clusters have been prepared and characterized. Structural examples are shown in Figure 6. The first class of models contains an MOS ,Fe unit, and are referred to as "linear" clusters. Various linear arrays of metal atoms include binuclear metallic clusters such as [S,MoS,FeL,]" (L 3 SM”, 0120, OAr21, OAczz, N023; L, = 8,24, o-xylszzs) (I), trinuclear clusters such as [c1,res,uos,rec1,]=- 26 (I_I), [s,uos,res,re(SAz-),P- 21%” (ILL). [s,uos,res,uos,]=- 28 (fl) and [(RS),MoS,FeS,Mo(SR),]" 29 (y), as well as a new type of hexanuclear cluster [MoOFe,S.(CO),,]" 30 (y_I_). A second class of synthetic clusters contains the MoFe,S. cubane core. Examples include linked double cubane structures such as [Mo.Fe.S.(SEt).1=- 31 (12.1.). [Mo.Fe.s.(sat).1°- 31:32 (my. [Mo.re.S.(SEt).(0Ph).P- 33. [Mo.Fe.s.(SEt>..J‘- 34 (1.3.0. [uo,p..s.(oue).(sa>.1=- 35 do and [Mo.Fe.s..(cae>.r- 36 (a); and single cubanes such as [MoFe.S.(SEt),(cat),]" 37 (E) and [MoFe,S.(SR),(cat)(L)]" 38 (L = solvent) (ELI). A new class of hexameric Fe-S clusters was reported in 1985. The first example of 10 S—F Fe 3— /Fe2+ 7%}?/i :-=e—:" o/f -—s /_.l e- -- o {—Fz .1, L? P°"(I+) _ p" (0) 825/2 8= 0 Figure 5. Schematic drawing of the P-clusters of nitrogenase based on the 5 ’Fe M8ssbauer and magnetic properties of the MoFe protein. 11 Figure 6. Schematic drawing of structurally characterized Mo-Fe-S clusters. 12 s s .31.) .. /\/’ \s/ \s/ x, ,A . )J" (‘3 s/ a/ \4/ \u . o « oJ. Hoax,“ ... Qavmv . /h\s . .... a J a . /H\.o m 'Kk/i 9 h.- x .._ ...i .. . _ 9 3 . \ t 1.4\ _ / . \ u A, .. M/ . a a l. a L «a we a... x L. e \x. e . _....... when n all... F _ F . OOl/I W \SR E . film... 6 94”.»... m M p fl\ ” _ .m . a s I. F . rm»... rill. a... .2 a... . . r - . V... 3.... .,...,....m 8 8 /.w\ m fl]fl/..n/»n\fl SHE'L. O / \ \. /.. skew... \ , ..\s/ _w_ SKY. /. .l m. n . 8 / / ./ (x F a... .. .NHV. X \ /. ..\. /. s/ Ask. .\\fl/. ”if“ In 13 this class of cluster to incorporate Me is [Fe.S.(OAr).(Mo(CO),),]"' 39 (X_I!). Although some of these clusters have similiar properties to FeMo-co, none serves as an appropriate analogue. Synthetic efforts towards modelling P-clusters have been less intense, owing to the limited data available. The ”Fe Mbssbauer parameters suggest that the P-clusters are a variant of the normal Fe‘S‘ tetrameric cluster. Hypothetical models have been proposed to account for the differentiation of "D" iron atoms from "Fe""" iron atoms, while maintaining a spin—coupled 4Fe-4S structure. One explanation is that the protein backbone enforces some sort of geometrical distortion on the Fe.S. unit such as to alter its physical properties from a normal cluster. Models of the oxidized state of the P-clusters ([Fe.S.]"' core) are analogous to synthetic structures that have an [Fe.S4(SR).]" formulation. A number of these reduced clusters have been reported (R = CR3Ph40, Ph‘u, t-Bu‘z, Et42, M942) and their X-ray structures show a lack of stereochemical regularity of the odd electron Fe.S. core. Two different core structures with distinct electronic properties are found. Figure 740b illustrates that [Fe.S.(SPh).]"’ exhibits a tetragonally elongated cubane core (elongated Dad structure with an idealized 4-fold axis), while [Fe.S.(SCR,Ph).]"' exhibits a tetragonally compressed core (imposed 2-fold axis). Magnetic studies and EPR spectra40b show that [Fe.S,(SPh).]"' has an S = V: ground state, while [Fe.S.(SCH,Ph),]" has an S = ’/2 ground state. This suggests that a protein-imposed distortion on an Fe.S. core could result in an S = 5/: ground state (as observed in the oxidized P—clusters). A threefold distortion could differentiate three of the iron sites from the fourth and may explain 14 PhS\ PhS’/ '4' 3- _ 3- 2 S I ' /$Ph /SCHZPh \FFI;S. 25\F: ——-S I __....—-Fe/ P}, :1]: —‘L’5 5/ \ CHZSS l SPh T .\ SCHzPh 3.. [F9454(5Ph)4] [Fe4s4 (SCH 2mg]; Figure 7. Schematic drawing showing the distortions of the [FO‘Sg]+ core in: (left) [ELQMON]3[FegSg(SPh)g]; (right) [Et.N],[Fe.S.(SCI-I,Ph).]. From ref. 40b. 15 the observed physical data. The second model, shown in Figure 8, has non-thiolate ligation on three of the four iron atoms. Ferrous iron has a low affinity for saturated amine ligands”; therefore, the most reasonable candidates for coordination would be oxygen-containing amino acids such as tyrosine, glutamic acid, or aspartic acid. This would lead to a mixed-ligand cluster with the formulation [Fe.S.(OR),(SR)]"'. The lability of the terminal ligands and the "tendency to yield a statistical distribution of all possible mixed-ligand complexes"8 has hindered synthesis in this area. Mixed-ligand clusters of the type [Fe,S.L,L',]"’ (L = Cl, L' :- 31%.“, 091145; L = SPh, L' = 0191.45, p-to1y145, iii-45) have been isolated in the solid state, but solution studies are difficult due to facile redistribution of terminal ligands. Recently, synthetic models of F648. cores with all oxygen ligands were prepared. The acetate cluster, [Fe.S.(OAc).]", has been generated and examined in solution, but not in the solid state46. The crystalline phenoxide-ligated cluster, [Fe,S.(0Ph).]", has been isolated and examined“. Phenoxide ligation seems to stabilize a new class of hexameric clusters. The crystal structure of [Fe.S.(OC.R.-p—CH,).]" 48 has been recently reported, whereas the thiolate hexamers are usually not observed due to the rapid transformation to tetramers, as shown in Figure 9. If an F9585 center were present in an iron-sulfur protein, extrusion by thiophenol would lead to the metastable [Fe.S.(SPh).]", followed by rapid conversion to tetrameric [Fe.S.(SPh).]". A third explanation of the physical data would be that each "D" atom is 5-coordinate. Diethyldithiocarbamate has been used as a bidentate ligand on an Fe-S cubane ([Fe,S.L,L',]" (L = 0149. SPhso; 16 Fe/SCys __ l / Ter Fe 3 S I Ter) I ’,Fe--- ---5 5’ Fe - \OTyr SCys Fe” Gin-Fe-j—S/ I 61:19: ’,Fe-- - - 5' Fe ‘06“: Figure 8. Possible models for the P-clusters, involving oxygen ligation at three vertices of a 4Fe—4S core. l7 { r 3‘ r w 2- x X , /X Ei—Fe Fe.’ x i—F' 2 x - : ‘ sI x '— 3 \fi/ge./.l 5. L ,‘Fe X’i” i / .— \x 5—5 / \ x x . J L .J ll 2 [F2686x613‘ 3£Fe4s4><412‘ Figure 9. Schematic drawing of the transformation of "hexamers" to "tetramers". 18 L' = Etadtc)), forming a mixed-ligand cluster with monodentate and bidentate ligands. [Fe.S,(SC.H.-o-OR).]" 51 is another example of a cluster that contains a 5-coordinate iron atom. The solid state structure shows three conventional tetrahedral FeS, sites and one unique Fes.o site. [Fe.s.(c1),(st,dien=- 49 has also been isolated and is the only mixed-ligand cluster with a 3:1 ratio of different ligands. It is interesting to note the effects of the various substitution patterns of the clusters on the "Fe M5ssbauer parameters. Table I shows the isomer shifts and quadrupole splitting values for selected samples citied above, and for the P-clusters. Oxygen ligation increases 6 by ~0.04 mm/s and AEq by 0.14 m/s over that of tetrahedral thiolate coordination. Five-coordination increases 6 ~0.18 mm/s and AEq by ~0.70 mm/s. Clusters in the reduced state, [Fe.S.(SR).]", show AEq values that are much higher than those observed for [Fe.S,(SR).]" clusters. It appears that a combination of all three explanations would form a more accurate model of P-clusters. C. Desi n of ex en- ated models of nitro enase Interest in further investigating oxygen ligation to metal clusters has led to the design of two models to be discussed in this dissertation. ' The purpose of the first model was to mimic the properties of P-clusters. It was desirable to make a mixed-ligand Fe.S. cluster that would have three Fe-O bonds and one Fe—S bond and which could be studied in solution as well as in the solid state. The synthetic approach toward an [Fe.S.(O,R)(SR)]"' cluster was to synthesize a tridentate ligand that would bind to three corners of an 19 Table I. "Fe Mdssbauer isomer shifts and quadrupole splittingsa of various Fe-S clusters. Complex 6 LE; [re.s.(srn).]=- b 0.46 1.07 [re.s.01.]=- b 0.52 1.09 [re.s.01,(oph),]=- b 0.51 1.01 0.52 1.28 [Fe.s.(orh),(sph),]=- b 0.47 0.96 0.46 1.24 [re.s.(opn).]=- c 0.50 1.21 [re.s.(sc.ii.-e-on).] 1- d 0.43 0.75 0.48 1.22 0.63 1.84 [re.s.(sph),(8t,dtc),]=- e 0.39 1.34 0.65 1.67 [re.s.01,(st,dte)]=- f 0.51 1.07 0.64 2.13 [Fe.S.(SCR,Ph).]" 8 0.60 1.41 0.60 0.93 [re.s.(sph).]=* g 0.64 2.04 0.57 1.13 P-clusters d 0.64 0.81 A. Vinelandjj 0. 69 3. 02 aSpectra taken at 4.2 K and referenced to Fe metal at room temperature. bReference 45. 6Reference 47. dReference 51. 6Reference 50. 1'Reference 49. Spectum recorded at 77 K. 8Reference 40a. 20 F648. cubane, while allowing the fourth corner to have tetrahedral sulfur coordination as shown in Figure 10. This design would prevent ligands from rapidly exchanging in solution as observed with other mixed-ligand clusters”. Previous attempts to synthesize clusters constrained by large macrocycles used OP(NRC.R.-o-SH),, OP(NHCH,C.H.-o—SH), and CH,C(CR,C0;CH;SH)352 as ligands. Reactions of these compounds with Ra‘s. cubanes were followed by optical and ‘R NMR spectroscopies. Observed shifts in absorption showed that binding had occurred, but the correct coordination was not achieved. It appeared that these ligands were too floppy to tightly bind an Fe.S. cluster, and that polymers were probably formed. A new type of rigid ligand, shown in Figure 11, was designed. The foundation of the ligand is a symmetrically trisubstituted triptycene, with three arms that can attach to the Fe-S cluster. This ligand will be referred to as the (HO),-tripod ligand throughout this text. A CPE model indicated that there is less than 1.0 A distance between iron atoms on the Fe.S. cubane and oxygen atoms on the (HO),-tripod ligand. The structural rigidity of the triptycene and the ligand bite formed by the phenolates are such that polymer formation is unlikely to occur. Once one phenolate ligand binds to one vertex of an Fe-S cluster, the chelate effect should force coordination of only one Fe.S. cubane per ligand. The synthesis of the (RO),-tripod ligand and 1H NMR studies of its binding to an iron-sulfur center are discussed in this dissertation. A second investigation of oxygen-ligated clusters involved the synthesis and characterization of a new Mo—Fe—S linear cluster, [Et.N]3[S,MoS,Fe(OAc),]. This complex was synthesized before a note on the preparation, by Zhuang, et al.22, appeared in the literature. Only 21 Figure 10., Schematic drawing of [Fe.S.(L)(SR)]"’; where L is a tridentate ligand. 22 Figure 11. Schematic drawing of the (HO),-tripod ligand. 23 the electronic properties were reported on this compound at that time. The purpose of this model was to mimic the FeMo-cofactor, which is known to be a unique molybdenum-iron-sulfur center. The only oxygen-ligated linear Mo-Fe-S cluster previously reported was [SaMosaFe(OPh),]" 21, which was synthesized by a ligand substitution reaction of [S ,MosaFeClzP" and sodium phenoxide. Previous attempts in our laboratory to synthesize the iron-molybdenum-acetate dimer by ligand substitution reaction of [SaMosaFeCl,]" with sodium acetate or from [S,MoS,Fe(SAr)3]"' (Ar = Ph, p-tolyl) and acetic anhydride failed. A black decomposition product was isolated but not characterized”. This dissertation describes the direct synthesis of the acetate dimer from ferrous acetate (Fe(O,C,H,),) and thiomolybdate ([MoSd"). The optical properties, variable temperature ‘3 NMR data and magnetic susceptibility measurements are discussed. II. RESULTS AND DISCUSSION A. HO ,-trimd Ligand 1. Synthesis The (HO),-tripod ligand that was synthesized is shown in Figure 11 of chapter 1. The numbering scheme used for the triptycene structures is consistent with the 9,lO-dihydro-o-benzenoanthracene nomenclature used by Chemical Abstracts Service and is illustrated in Figure 12. The 1,8,13-trisubstituted structures are referred to as syn and the 1,8,16-trisubstituted structures are referred to as anti. A preliminary goal was to synthesize a symmetrically trisubstituted triptycene containing carboxyl groups at the l-, 8- and 13- positions. Friedman and Logu11054 initially reported a convenient synthesis of triptycene from the reaction of anthracene with benzyne. The first synthetic approach taken to produce the foundation of the (HO),-tripod ligand is shown in Figure 13. The preparation of 1,8-dichloro- anthracene (2) from 1,8-dichloroanthraquinone (1) was followed as reported in the literature“; but its conversion to 1,8-dicyano- anthracene (3)56 according to the literature procedure yielded little or no product. The procedure described56 for 3 involves reaction of 2 with cuprous cyanide in quinoline for 24 h, followed by digestion of the crude product with nitromethane, and finally recrystallization from acetic acid. This procedure was improved by treating crude 3 with 24 25 Figure 12. Schematic drawing of a triptycene structure showing the numbering of atoms. 26 000 ——:-:,—~ 21*“ + éijii” o , 2 coon coon coon coon cn3 93,3: .+ no... <4—~ CH3 6A 65 0002"! “NI [mm /’ ‘4'. l \ nooc coon coon 75 CM CM 5A 55 Figure 13. Schematic of a synthetic approach toward the preparation of triptycene 1,8,13-tricarboxylic acid. 27 aqueous ammonia. This resulted in the decomposition of an organocopper intermediate to give crude 3 and a blue solution of [Cu(NH,).]"". Purification by column chromatography to separate out l—cyano-S-chloroanthracene and some black tar yielded 3 in 41% yield. Although numerous substituted triptycenes have been prepared”, only one symmetrically trisubstituted compound was known prior to this work. Mori, et a1.58 had found that the Dials-Alder reaction of 2 with chlorobenzyne led to a mixture of 1,8,13- and 1,8,lS—trichlorotriptycenes in 16% combined yield. This type of Diels-Alder addition seemed like the best approach to prepare the basal portion of the (RO),-tripod ligand, triptycene-1,8,13-tricarboxylic acid (7s). 2-Amino-6- methylbenzoic acid (4) was chosen as a convenient benzyne precursor because of its commercial availability. Reaction of 3 with 4 in the presence of isoamyl nitrite in dimethoxyethane (DME)/triethylene glycol dimethyl ether yielded 57% of 1,8-dicyano-l3-methyltriptycene (5s) and l,8-dicyano-16—methyltriptycenes (5a). The nitrile groups on the triptycene caused the mixture of 5a and 5s to be only sparingly soluble in most organic solvents, except extremely polar solvents such as ' dimethylsulfoxide (Me,SO) and dimethylacetamide (DMA). The isomers were not separated at this point because the small polarity difference between anti and syn structures made isolation difficult. Conversion of the nitrile groups to carboxylic acids was attempted by reaction with concentrated sulfuric acid”, 85% phosphoric acid60 and 10% potassium hydroxide/Z—methoxyethanol56. All of these procedures resulted in recovery of starting material. A clean conversion to l3-methyl- triptycene-1,8-dicarboxylic acid (6s) and lS-methyl—triptycene- 1,8-dicarboxylic acid (6a) was accomplished by reacting the mixture of 28 5a and 5a with ROI! in ethylene glycol61 at 100 'C for four days. The isomeric products 6a and 6s were more insoluble than 5a and 5s, and consequently caused problems in attempts to oxidize the methyl substituent. Oxidation of the methyl group of 6a and 68 using mild reagents such as cerium ammonium nitrate (Ce(NH.),(NO,)‘) in glacial acetic acid“, chromium oxide (CrO,) in pyridine53 and tetrabutyl- ammonium permanganate (Bu.NMnO.) in pyridine“, was unsuccessful. When reaction conditions were made severe enough (6a and 6a in Bu.NMnO./pyridine for 48 h at 100 'C), evidence of oxidation of the triptycene structure to methyl-substituted anthraquinone was observed by ‘H NMR spectroscopy. Oxidants have been previously reported to cleave the triptycene system65. Attempts to oxidize the methyl group of the isomeric pair of 5a and 5s (which had greater solubility than 6) also met with failure. At this stage, it appeared as if the methyl group could not be converted to a carboxylic acid in one step, and therefore a multiple step approach was tried. Bromination of a methyl substituent has been known to yield a bromomethyl group“, which can subsequently be converted to a primary alcohol, and then oxidized to a carboxylic acid. Due to the low solubility of 6a and 6a, the three additional steps in the synthetic route would have to be clean conversions since purification techniques would be limited. Bromination reactions were conducted with the isomers 1,8-dichloro-lS-methyltriptycene (8s) and 1,8-dichloro-l6- methyltriptycene (8a) since 6a and 6s were difficult to isolate. Reaction of 8a and 8s with one equivalent N-bromosuccinimide (NBS) and dibenzoyl peroxide led to what appeared to be a bromomethyl substituent on the triptycene, as well as a dibromomethyl substituent. 29 When 2 equivalents of NBS were reacted with the triptycenes, 1,8-dichloro-13- and -16-dibromomethyltriptycenes (9a and 9s) were isolated. This product, in principle, could be converted to an aldehyde-substituted triptycene and then subsequently oxidized to yield the desired carboxylic acid substituent. At this point, the synthetic problems provoked a reinvestigation into the approach to 7s as outlined in Figure 13. The efficiency of the synthetic plan was reanalyzed, and it was realized that in order to isolate and purify large quantities of 7s, another synthetic route had to be designed. A linear and a convergent approach to synthesizing a hypothetical molecule A-B-C-D are contrasted in Figure 1467. The yields are calculated assuming 90% yield at each step. The linear approach to synthesizing the trisubstituted triptycene (7a) was to perform the Dials-Alder reaction first, and then try to convert the substituents to carboxylic acids. A convergent approach, to the same compound, would convert the substituents on the anthracene and benzyne to carboxylic acids prior to the cycloaddition reaction. This new approach is shown in Figure 15. The advantages of a convergent approach are increased overall yields of products and simplified purification. The linear synthesis complicates separation and purification problems because the change in physical properties of the compounds diminishes as the synthesis progresses. The only problem envisaged with the convergent approach was that the carboxyl substitutents on the benzyne and anthracene might prefer to orientate to give the anti isomer, due to steric hinderance during the cycloaddition reaction. We were interested in optimizing the yield of syn isomer since the target ligand is 30 LINEAR: A -—B—i A-B —£—. A-B-C -—D——i A-B-C-D 732 couvsncsm': A A-B —' B — A-B-C-D 81% c C-D _. D Figure 14. Schematic of linear versus convergent synthetic approaches. 31 Cl C) Cl cu 0 CN CuCN ’5“% O l COOCH3 I. 5'0“ 603 CO °°°°”3 COOCH33C 3coocn3 coocn 17A1‘7s3 MN on" Shanty ."0 COOH COOH 75 N02 00°” coon m3on coocn3 L2 ’5“% cases NH coocn3 coocn3 dCOOH + mCOOCH l€5 K3 ”2 99% cnaon COOH ‘5 14 Figure 15. Schematic of convergent synthesis approach to triptycene 1,8,13-tricarboxylic acid. 32 substituted in this fashion. This dilemma led to an investigation of various factors that are important in controlling the stereochemistry of the Dials-Alder reaction between various benzynes derived from 6-substituted anthranilic acids and 1,8-disubstituted anthracenes. The results are discussed at the end of this section. The convergent synthesis outlined in Figure 15 shows that l was converted to l,8-dicyanoanthraquinone (10) by reaction with cuprous cyanide in DMA. This is a modification of the original procedure“, which uses benzyl cyanide as the solvent. The organocopper intermediate was decomposed with dilute nitric acid to give 10 in 88% yield. Hydrolysis of 10 with 11.80. gave anthraquinone-1,8—dicarboxylic acid (11), which upon reduction with zinc dust in NILOH yielded anthracene-1,8-dicarboxylic acid (12). l,8—Bis(methoxycarbonyl)- anthracene (13) was synthesized as described69 by reaction of 12 with acidic methanol (149011). In order for the convergent synthesis approach to work, a new methoxycarbonyl—substituted aryne had to be developed. The two-step synthesis of 2—amino-6-methoxycarbonylbenzoic acid (16) involved selective esterification of 3-nitrophthalic acid (14) to yield l-methyl—2-hydrogen-3-nitrophthalate (15), which was reduced to the amino benzoic acid 16 (as shown in Figure 15). The literature preparation70 for 15 reported a 56% yield with a reaction time of 16 h. This procedure was modified to give an increased yield (77%) in a much shorter time (3 h). The hydrogenation of 15 for 8 h in 146011, using a catalyst of 5% palladium (Pd) on charcoal yielded a gummy yellow solid of 16. Upon standing at room temperature for more than 24 h, this compound undergoes intramolecular condensation reactions to form 33 amide-linked polymers and MeOE. Compound 16 was reacted with 13 to yield a 3:1 mixture of 1,8,l6-tris(methoxycarbonyl)triptycene (17a) and 1,8,13-tris- (methoxycarbonyl)triptycene (17s). These colorless isomers were separable by column chromatography (silica gel), although radial thin-layer chromatography was used as an efficient way to perform the separation. The radial chromatograph was equipped with a quartz cover which allowed the evolution of bands to be followed using a UV light source. The ester substituents on 17a and 17s provided increased solubility and circumvented many reaction condition problems seen with 5s and 5a or 6s and 6a. Hydrolysis to the triptycene tricarboxylic acids (7a and 7s) in XOR/M6011 was. easily performed. The convergent approach shown in Figure 15 provided a practical synthetic route to produce large quantities of the .(HO),-tripod foundation. This was mainly because of the increased solubility of 13 compared to 3 and consequently easier separation and purifications of the subsequent products. Conversion of 7s to the (HO),—tripod ligand is shown schematically in Figure 16. Compound 7s was reacted with thionyl chloride to yield triptycene-1,8,13-tricarboxylchloride (17s). This product was moisture sensitive and therefore immediately reacted with p-aminophenol in the presence of triethylamine, forming the (HO),-tripod ligand (19s) in high yield. Four other triptycene amides, 19a, 20a, 21a and 22a, were prepared by reaction of triptycene-l,8,16-tricarboxylchloride (17a) with p-aminophenol, propylamine, p-toluidine and p—aminobenzene p—xylene sulfide, respectively. An alternative preparation of these carboxamido-triptycenes involved reaction of 18s or 18a with seven 34 Figure 16. Schematic drawing of the synthesis of the (HO) ,-tripod ligand. 35 equivalents of amine (instead of using Et,N to scavenge hydrochloric acid). The reaction of 18s with p-aminothiophenol yielded a mixture of products, some of which appeared to be thioester-linked oligomers. Apparently the difference in pita of phenol (9.9)",1 versus thiophenol (6.4)72 causes a difference in reactivity. Solvents and reaction conditions were varied to try to change the amount of product formed, but side-products were always present. Attempts at separation of the product mixture proved to be futile due to solubility problems. Removal of the S-p—xylene protecting group from 22a, to yield the sulfur analog of the (RO),-tripod ligand, was also unsuccesful due to the insoluble nature of this compound. In the process of preparing the (HO),-tripod ligand, a series of new trisubstituted triptycenes was synthesized which gave some insight into the selectivity of the Dials-Alder cycloadditon between disubstituted anthracenes and monosubstituted benzynes. The synthesis and properties of the isomers were examined. The various l,8-disubstituted anthracenes used were dichloro-, dicyano-, and his- (methoxycarbonyl)anthracenes, of which the improved preparations were discussed earlier in this section. The three substituted benzyne precursors used included 2-amino-6-methylbenzoic acid, 2-amino- 6-chlorobenzoic acid",3 and 2-amino-6-methoxycarbonylbenzoic acid. The three monosubstituted anthranilic acids were reacted with the three disubstituted anthracenes to yield eight pairs of isomeric I substituted triptycenes, shown in Table II. The ninth possible pair of isomers (X = CN, Y = Cl) could not be prepared. Monosubstituted benzynes were generated in situ, by slow addition of a DME solution of Table II. 36 Ratios of anti to syn isomers of trisubstituted triptycenes. X X Y NH N we we s x = v -.-. new“ 8 Cl cs, 25% 75% 74: 5 ON on, 28% 72% 577: 23 000011, on, 312 69% 58% 24 c1 c1 77% 232 277: 25 coocn, Cl 73: 277: 207: 26 Cl coocn, 44x 56% 4 477: 27 on 00008, 992 17: 38% 17 coocs, cocoa, 76% 247: 62% ‘5 Yield is crude yield of both isomers prior to chromatography, based on anthracene starting material. Isomer ratios were obtained'by integration of the ‘3 NMR spectra of the crude triptycene mixture. 37 monosubstituted anthranilic acid into a solution of disubstituted anthracene and isoamyl nitrite in the same solvent. A twofold excess of monosubstituted anthranilic acid and isoamyl nitrite was used to insure complete reaction. Treatment of the product with aqueous NaOH neutralized the excess acid in solution and precipitated the product. In a few cases, the parent disubstituted anthracenes were present in the product mixtures. It has been reported that maleic anhydride can be used to scavenge anthracene58, but it was found that an easier way to separate the anthracene from the triptycene was by sublimation. The yellow starting materials, which have lower melting points and higher volatility compared to the triptycenes, were sublimed off to yield the white mixture of triptycene isomers. The lowest yields of triptycenes were obtained using chlorobenzyne as the dienophile. This suggests that yields of triptycenes depend on the ability of the substituted anthranilic acid to produce benzyne. It has been reported58 that 3-chlorobenzyne is not easily produced via aprotic diazotization of 3-chloroanthranilic acid, or from the isolated chlorobenzenediazonium-Z-carboxylate. Attempts to synthesize l,8-dicyano-13- or -l6-chlorotriptycene failed, presumably due to the difficulties in generating 3-chlorobenzyne combined with the very low solubility of 3. In comparison, the new aryne made from 3-methoxycar bonylbenzoic acid via aprotic diazotization gave triptycene products in 50-60% yields. Separation of six pairs of structural isomers was carried out I by semi-preparative RPLC using a silica gel column. The pairs of isomers formed using o—methoxycarbonylbenzyne were the easiest to separate. Two sets of isomers could not be separated by HPLC; trichloro— 38 triptycene and dichloro(methyl)triptycene. 1,8,13- and 1,8,16-Trichloro- triptycene have been reported55 to be separable on an alumina column using benzene as the eluent. TLC on alumina plates showed no separation in benzene or other solvents. No separation was observed using HPLC conditions similiar to those used with other pairs of isomers. 1,8-Dichloro-l3-methyltriptycene was, however, separated from the anti isomer by slow evaporation of an ethyl acetate (EtOAc) solution of the mixture of compounds, whereupon the syn isomer selectively crystallized. Recrystallization from EtOAc gave the syn isomer in 95% isomeric purity. The ratio of syn to anti isomers was determined by integration of the ‘R NMR spectra of the crude triptycene product (from at least two separate reactions). Table II shows the ratios obtained along with combined yields for compounds 5, 8, 17, 23-27. It is clear from Table II that the substituent on the benzyne (Y) is more important in dictating the observed regiochemistry than the substitutents on the anthracene (X). When Y = 011,, the syn isomer is formed in a 2 or 3 to 1 ratio relative to the anti, regardless of the anthracene substituent X. Conversely, when Y = Cl, the anti isomer is preferred over the syn structure. Only when Y = 00,011,, does the anti to syn ratio depend noticeably on X. These ratios are interpreted as resulting from electronic effects of the substituents on the polarity of the orbitals of the benzyne with respect to the anthracene“. All three anthracenes examined have electron withdrawing substituents at the 1- and 8-positions, which are expected to stabilize a partial negative charge at the 9-position and a partial positive charge at the lO-position, with the stabilization being 39 in the order 0N > 00,011, > 01. This prediction of stabilization is based on the relative electron withdrawing effect of the substituents (Hammett op parameters“). Because the bonding orbital of the benzyne is the n-bond orthogonal to the aromatic system of the benzyne ring, the more important effect (of substituents is an inductive one via the a network, rather than a resonance effect. Qualitatively, an electron-releasing substituent will generate a partial negative charge at the substituted carbon (0-3) of the benzyne, thus stabilizing a polarization of the benzyne as 6+ at 0-2 and 6- at 0-1. An electron-withdrawing substituent will cause exactly the opposite polarization of the benzyne. Simple electrostatic matching of the polarized benzyne and anthracene satisfactorily accounts for the preferred regiochemistry in 5, 8, 23-25. For 17, 26, 27 the observed regiochemistry depends on the anthracene substituents X. This is rationalized by the recognition that the substituent-induced polarization (01) of the benzyne 17 electrons by the methoxycarbonyl group (01 = 0.20) is intermediate between those of the methyl and chloro groups (01 = -0.04 and 0.46, respectively)“. Consequently, an increased sensitivity to subtle changes in the elcetronic structure of the anthracene would be expected. For Y = 00303,, the anti preference correlates with the ability of the anthracene to stabilize negative charge at 0-9 and positive charge at 0-10. For X = 01, the resonance effect (on) is opposite in sign to op. It appears as if the resonance effect, which places a partial negative charge at 0-10, dominates, giving a slight excess of the syn isomer. The almost exclusive formation of the anti isomer for X = 0N is readily explained by the stronger combined 0p and 03 effects of 0N versus 40 00,011,. Alternatively, the favorable alignment of opposed dipoles in the transition state may be significant in determining the final orientation of substituents. For example, the lowest energy arrangement of dipoles?7 of the benzyne and anthracene to produce 27 yields the anti isomer, and may account for the virtually exclusive formation of the isomer. None of the results suggests that steric effects are significant with the substituents examined. 2. Physical Properties The melting points of the trisubstituted triptycenes are high, usually above 300 '0. The (HO),-tripod ligand melts at 420-424 ‘0 with decomposition. Melting points of syn structures are higher than their anti analogs. The extraordinarily low solubility of many of these triptycenes in common organic solvents made it difficult to obtain analytically pure samples; preparative HPLC runs typically produced ‘ 1 mg/run. Several elemental analyses were somewhat low in carbon, apparently due to occlusion of methylene chloride in the triptycene lattice. The presence of 08,01, was verfied by ‘H NMR; integration gave stoichiometries consistent with the elemental analyses. Heating the compounds at 125 '0 under a vacuum of 0.07 torr for 64 h did not remove all the chlorinated solvent present, as evidenced by a positive chloride test (sodium fusion method). Electron impact (direct exposure probe) mass spectra showed the molecular ion (M‘1‘) peak and usually, peaks due to the loss of substituents on the triptycene structure. High resolution mass spectra were obtained on triptycenes that analyzed low in carbon to confirm the exact mass. 41 The most effective way to differentiate the isomeric pairs of triptycenes was by their ‘H NMR spectra. The triptycene structure shows a characteristic pattern composed of four doublets, two doublet of doublets, and two singlets. The singlets are due to the bridgehead protons, and their shifts vary depending upon the substituents on the aromatic rings. The ‘H NMR of triptycene was studied by Kidd, et al.73, with the conclusion that chemical shifts are due to ring currents, bond anisotropies, and electron density contributions on the rigid structure. The data obtained confirm these results; the triptycenes with polar substituents show the largest Ad for the bridgehead protons. In each case, the syn isomer bridgehead proton signals show larger M than those for the anti isomer. The ABC spin system on each aromatic ring, shows J AB = 3809 and J A0 3 0. In some instances, the overlap of peaks made it necessary to decouple protons selectively in order to assign coupling constants. For 8a, 8a, 24a, 24s and 21a, 2-dimensional homonuclear chemical shift correlation (COSY) spectra were recorded in addition to 1-dimensional ‘H NMR to determine coupling constants. Characterization of 17s, 7s and 19s was simplified, owing to the three-fold symmetry and the resulting magnetic equivalence of the identically substituted aromatic rings. The ‘1! NMR of the (HO),-tripod ligand looked like a cross between the spectra of 17s and p—aminophenol. The proton resonances of the OR and NH groups appeared as sharp singlets at 10.2 and 9.3 ppm, respectively. The bridgehead hydrogens appeared at 7.5 and 5.9 ppm, showing a large downfield shift from the bridgehead proton signal observed at 5.2 ppm for triptycene78. Two sets of sharp doublets due to the ortho and meta protons on the phenolate ring occur at 7.4 and 6.6 ppm, 42 respectively. The resonance at 7.6 ppm (doublet) arises from protons at the 3, 6 and 15 positions of the triptycene; at 7.3 ppm (doublet) from the 4, 5, 16 positions; and at 7.2 ppm (doublet of doublet) from the 2, 7, 14 positions. The ’ ’0 NMR spectra of the trisubstituted triptycenes typically showed two signals in the region of 43.3-54.7 ppm due to the bridgehead carbons, four signals between 147.8-141.4 ppm due to the carbons next to the bridgeheads, and eight aromatic ring signals at ~ 126 ppm. For some trisubstituted triptycenes, fewer signals were observed than predicted, presumably due to coincidental overlap of chemical shifts. Most of the ”0 resonances observed for the (HO),-tripod ligand can be tentatively assigned. The u0 NMR spectrum for the (H0),-tripod ligand is shown in Figure 17. The aldehyde carbons are observed at 165.9 ppm. The shifts of the carbons on the free aromatic rings can be estimated by applying the principle of substituent additivity to incremental shifts reported for monosubstituted benzene rings79. The incremental shifts are added to the shift observed for benzene carbon atoms (128.5 ppm), to give the calculated shifts shown in Table III. The remainder of the peaks are due to the carbons of the triptycene aromatic rings. The furthest downfield aromatic peak at 146.5 ppm is most likely due to the carbon atoms at the 1, 8 and 13 positions, since these are closest to the aldehyde group. The resonance at 141.2 ppm is probably due to the carbons next to the bridgehead carbon (on the same side as the substituents), and the peak at 129.9 ppm is assigned to carbons at the 3, 6 and 15 positions. The remainder of the resonances at ~ 125 ppm are not separated by more than 1 ppm and are difficult to assign. 43 I 1 I T f I T I r I I I 1 I I I I I r U r l I j V 160 140 120 100 80 60 Figure 17. "C NMR (360 MHz) of the (HO),-tripod ligand in MeaSO-d ‘0 44 Table III. Calculateda and observed "C MIR shiftsb for aromatic resonances of the (HO),-tripod ligand. 9‘3 to NH 1 I 02 O 3 3 4 4 OH O atom O atom Calculated Observed shift shift shift shift 1. - 7.3 + 11.1 132.3 133.4 2. + 1.4 - 9.9 120.0 123.8 3. + 12.7 + 0.2 116.0 114.6 4. + 26.9 - 5.6 149.8 154.0 ‘based on incremental shifts of aromatic carbon atoms of monosubstituted benzenes (benzene at 128.5 ppm). bvs TMS in Me.so (ppm). 45 8. Binding study of the (HO),-trimd ligand to Jan Fe._S_. core 1. Sznthetic approach Two routes to preparing an oxygen-ligated iron-sulfur tetramer were employed. The 'first method utilized the fact that an equilibrium exists in the reaction of thiolate-ligated clusters and mercaptans, as shown in equation 3. It has been shown that when R is an alkyl group [Fe.S.(SR).]" + 4 R'SH F— [Fe.S.(SR')4]" + 4 RSH (3) it can be replaced by an R'SH group, where SR’ is an aromatic thiolao. The ligand replacement can be correlated to the acidity of the SR’ group. The more acidic aromatic thiols will displace aliphatic thiols which have higher pKa values. Kinetics data81 suggest a mechanism in which the rate determining step is protonation of the coordinated ligand followed by rapid separation of the alkylthiol and coordination of the arylthiolate. In the case of phenolates, an equilibrium exists where a small fraction of thiolate is displaced as shown in equation 4. [Fe.S.(SR).] + 4 R'OH _ [Fe.S.(OR')‘] + 4 RSH (4) The reaction can be driven to the right by removing volatile thiol (R = Et, t-Bu) in vacuo. The tridentate ligand should replace three of the thiolates to produce the proposed model for the P-clusters. The (H0),-tripod ligand was reacted with [Pr‘N],[Fe.S.(SEt).] in 46 Measo-d, under a dynamic vacuum (to remove EtSH) for 12 h. The color of the solution changed from dark brown to red-brown. A red coloration is usually indicative of complexation by an oxygen-containing ligand”. The ‘H NMR spectrum of this reaction mixture was recorded and is discussed in the next section. An alternative method of attaching the (HO),-tripod ligand employed a ligand exchange reaction between the tetraethylammonium salt of [Fe.S.Cl.]"’ and the trisodium salt of the (HO),-tripod ligand in Measo-d‘, as shown in equation 5. This method makes use of the [Fe.S.Cl.]"‘ + Na,(O,-tripod) —-’ [Fe.S.(O,R)(Cl)]" + 3 NaCN (5) precipitation of NaCl as a driving force for complete substitution. The trisodium salt of the ligand was synthesized by dissolving the (HO),-tripod in DMA and reacting it with a solution of sodium methoxide (NaOMe) in MeOH. The DMA and MeOH were removed from the reaction mixture in vacuo, and the resultant solid was reacted with [Eth],[Fe.S.Cl.] in Me,SO—d.. The results of the ‘H NMR studies are discussed below. 2. Proton nuclear magnetic resonance studies The ‘H NMR spectra of the solutions of (HO),-tripod ligand and 89.8. cluster show paramagnetically (isotropically) shifted peaks. Paramagnetic shifts are the difference in shift between the observed frequency and the analagous diamagnetic reference frequency. Shifts downfield from the internal standard were assigned positive values. 47 Isotropic shifts were calculated using equation 683. (AH/Rhea = (AH/mobs - (AH/Hm. (6) Because the interpretation of the NMR data depends upon existing theory of paramagnetic shifts, a brief outline of important factors is discussed. Isotropic shifts are due to two components, pseudocontact shifts and scalar shifts, as illustrated in equation 733. Pseudocontact h’iso = h’scalar + ”pseudocontact (7) or dipolar shifts are observed in the presence of a paramagnetic atom. These shifts are due to a through-space contribution to the observed field which arises from the proton spin orientation with respect to the paramagnetic center (in the magnetic field). Scalar or contact shifts are due to the change in observed field when an unpaired electron spin is delocalized onto a proton nucleus. A series of [Fe.S.(SR.)]" clusters has been studied by ‘II NMR spectroscopy84, and it has been shown that the dipolar contribution to shift is negligible compared to the contact contribution. Contact shifts observed for an alkyl ligand show that the ligand 0 molecular orbitals are involved in the spin delocalization. A characteristic 0 delocalization pattern shows a decrease in intensity of the observed signals as the distance from the paramagnetic center increases. In addition, all- shifts are downfield from their diamagnetic references. This a mechanism is consistent with what is observed for [Fe.S.(SR)4]" when R = alky184. Unpaired spin can also delocalize into the 1! system of an aromatic 48 ligand. This phenomenon is evidenced by the fact that shifts tend to alternate in sign between adjacent protons. In addition, the magnitude of the shift varies randomly as the distance increases between the resonating proton and the coordination site. For [Fe.S.(SR).]"’ (where R = aryn84 and [Fe.s.(oph).]=- 82, spins delocalize through the ligand 11 system causing a downfield shift of ortho and para protons but an upfield shift of meta protons. Spin density is stabilized at the ortho and para positions of the aromatic ring as seen in the resonance forms of a phenolate substituent shown in Figure 18. Reported isotropic shifts of the phenyl protons of the [Fe.S.(OPh).]" 45 complex and other arenethiolate analogs34 allowed prediction of approximate chemical shifts for the isotropically shifted resonances of an [Fe,S.(O,R)(SR)]" cluster, as shown in Table IV. The ‘H NMR spectra of the product species from the reaction of [Fe.S.(SEt).]" with the (H0),-tripod ligand after 4h and 12h are shown in Figure 19. The spectrum of the free (HO),-tripod ligand is shown at the top of Figure 19. The assignments of the diamagnetic resonances are discussed in section II.A.2. The triptycene bridgehead proton resonances are marked as T3 and the peaks due to aromatic protons are marked as TA. Only the aromatic region is pictured since the peaks upfield from 6 ppm are due only to cation and solvent protons. The resonances observed for protons bound to the Fe-S cluster show broadened signals due to the fact that paramagnetic ions induce efficient relaxation mechanisms and linewidth is inversely proportional to the efficiency of relaxation. The ortho peak is broader than the meta peak due to the fact that dipolar broadening has an r" 49 8~s'=o~t<:>=o~ o 1 Figure 18. Schematic drawing showing spin density stabilization at the ortho and para positions of an aromatic ring. 50 Table IV. Isotropic shifts8 for ligand protons in [FO‘S‘(03-trip0d)(SEt-)]z- and [FO‘S‘(03-tripod)3]" species in Measo-d‘ solution. Observed Diamagnetic Reported Predictedb Observed Observedc proton shift Isotropic shift shift Isotropic shift shift o-II (tet) 7.4d - 2.3f 5.1 5.5 - 1.9 r8 (tet) 6.64 + 2.2f 8.8 9.1 + 2.5 0-3 (hex) 7.4d - 5.3a 2.1 - - rll (hex) 6.6d + 5.02 11.6 11.6 + 5.0 cu. (EtS’) 2.58 + 10.0h 12.5 12.5 + 10.0 on, (my) 1.3e + 1.111 2.4 2.3 + 1.2 avs. TMS (ppm). bdiamagnetic shift + reported shift for similiar proton. cobserved shift - diamagnetic shift. dshift of free O—tripod ligand. 9shift of free ethanethiol (ref. 84) fisotropic shift of respective protons from [Fe.S.(OPh).]"’ (ref. 47). ‘isotropic shift of respective protons from [Fe‘S.(0Ph).]" (ref. 86). hisotropic shift of respective protons from [Fe,S.(SEt)4]" (ref. 84). 51 Figure 19. ‘H NMR spectra (360 MHz) of the (HO),-tripod ligand and its reaction with [Pr.N],[Fe,S,(SEt).] in Measo-d. after a 4 h period and a 12 h period. 52 (H013-TRIPOD LIGAND M3 0:1 O-H m-H TA 7‘13 Te Ts JLw‘L J ml“) _- ...r.-a-,....,e...,.- fl..- )0 9 8 7 6 PPM T CPr~4NJZLFe4S4CSED41 + CH013 -TRIPOD Pd: 4h CH2CE+S‘) NH ( m-H OH I Lfl T8 NH-+C* mal'et I I?" KT.) ‘H 048* ,..,...,...,...T..e,..:f 16 I4 12 IO a 12x 5 PPM r—H 12 h TB m-hex TezH ”H" user J" (353 J .42 w M max, ...,-,.,...,..~:,. ,...j‘ M 12 10 a 5 pp... Figure 19 53 dependence85. The alternation in signs of the isotropic shifts for the ortho and meta protons and the decrease in intensity of the signal the closer the proton is to the paramagnetic center, are confirmatory signs of dominant contact interaction via a n delocalization mechanism. Table IV shows that the observed isotropic shifts are very close in magnitude and sign to those shifts reported for related compounds. The spectrum recorded after a 4h reaction period shows peaks at 5.5 ppm and 9.9 ppm which are assigned to the contact shifted ortho and meta protons of the ligand attached to tetramer (noted as o-tet and m-tet). The isotropically shifted resonances of the ethanethiolate group appear at 2.3 ppm (CH,-EtS‘) and 12.5 ppm (Cflz-EtS"), but the peak intensity is too large for only one EtS' group attached to a cluster. An analogous ligand substitution reaction between [Pr.N],[Fe.S.(SEt).] and the (HO),-tripod ligand was allowed to run for 12 h. The ‘H NMR spectrum shows no evidence of free ligand and less intense peaks of the ethanethiolate group, indicating that the reaction has gone to completion. Besides the paramagnetically shifted peaks assigned to the tetrameric cluster, peaks that could be assigned to the formation of a hexameric Fe.S. cluster were observed. Hexameric clusters typically show larger isotropic shifts than their tetramer analogses. The meta proton of the [Fe‘S.(O,-tripod),]" compound (noted as m—hex) appeared at 11.6 ppm, the ortho proton was expected to appear at 2.1 ppm, but no signal was observed in this region. The ortho proton resonance may be too broad for observation or it may be obscured by the cation resonances in this region. There are three large signals (at 9.8, 10.4 and 11.3 ppm), that are 54 shifted downfield but are not accountable as an ortho or meta resonance of the [Fe.S,(O,-tripod),]" or [Fe.S,(O,-tripod)]" species. The broad peak at 9.8 ppm (noted as H) increases in intensity as the m-hex peak intensity increases. The peak at 10.4 ppm (noted as T32) is present (even when the reaction has not yet gone to completion and appears too sharp to be contact shifted. This signal (T32) may be the bridgehead proton of the triptycene (the proton on the same side of the molecule as the substituents). As discussed earlier, the change in shift of the bridgehead protons is extremely sensitive to the nature of the substituents on the aromatic rings. The shift observed upon binding to the cluster could be a through-space interaction due to the unpaired electrons of a sulfur atom on a cubane core being pointed towards the inner bridgehead proton. The intensity of the T32 peak is the same as the that of peak TB, which is unshifted from its position in the diamagnetic ligand. The third large peak at 11.3 ppm (noted as NB—tet) is assigned to the contact shifted proton of the para NH group on the ligand. Although this peak has a narrow linewidth for an scalar shifted resonance, it is far enough away from the paramagnetic center that any broadening should be minimized. The intensity of the NH-tet peak is half that of the m-tet peak, as would be expected. Cleland, et a1.“7b noted certain drawbacks in synthesizing [Fe.S,(OPh).]" by removal of volatile thiol; one of which was that a new type of cluster compound with significantly larger isotropic shifts was observed. This new type of cluster was later reported to be phenoxide-ligated hexamer36. A second method of ligand exchange reaction between [Fe.S.Cl.]"‘ and sodium phenoxide was reported to minimize the likelihood of forming a hexameric complex47b. The ligand 55 substitution method was attempted on the chloro tetramer and the trisodium salt of the tripodal ligand since the first synthesis method was giving a mixture of products. The ‘H NMR spectra of Na,(O,-tripod) and the product species from its reaction with [Fe.S.Cl.]" are shown in Figure 20. The absence of the OH signal in the ‘1! NMR spectrum of the Na, (Os-tripod) ligand confirms that the salt was formed. The spectrum of Na,(0,-tripod) + [Fe.S.Cl.]" is essentially the same as the spectrum of (HO),-tripod + [Fe.S.(SEt).]"’, proving that both methods of ligand substitution are producing a mixture of tetrameric and hexameric cluster species in solution. The spectrum of [Fe.S.Cl,]" and Na,(0,-tripod) shows the ortho proton peak of the ligand bound to the tetrameric core at 5.1 ppm and the meta proton at 9.2 ppm. The meta proton of the ligand bound to the hexameric cluster was observed at 11.6 ppm. The reactions between the chloro tetramer and the ligand were allowed to stir for 8 h before the NMR was recorded. Several peaks in the spectrum can be assigned to free ligand (OH, NH), but since there is no evidence of free ligand in the spectrum of Na, (0,—tripod) another explanation is necessary to account for these resonances. If the reaction mixture is stirred for 24 h the spectrum shows the disappearance of para- magnetically shifted peaks and an increase in free ligand resonances. It is possible that traces of water in the Megso-d‘ (which is exceedingly difficult to dry completely) promoted the generation of free ligand by hydrolysis of the trisodium salt; however, no evidence of free ligand was observed in the ‘H NMR spectrum of the Na,(O,-tripod) taken in the same solvent as the ligand exchange reactions were performed. It is more probable that the high-coordinating ability of 56 Figure 20. ‘H NMR spectra (360 MHz) of Na,(O,-tripod), reaction between Na,(O,-tripod) and [Bt.N],[Fe,S.Cl.], and reaction of [Pr.N],[Fe,S.(SEt)4] with (HO),-tripod and five equivalents of PhSH in Me,SO-d.. 57 1.7: m-H TA TA M! Ta J TB ——T——' I [r—v'vtrvvvl ffi l0 9 8 7 6 ppu TA ,————-q [E+4N 125345404] + N03C03-TRIPOD) Ta n-H \ \ o-m - 4 v 1 . v u r I u r . r—u . xvi—u-fo—"I‘-r—1 l7 '0 '8 6 4 PPM E P—‘hfi W4N12£Fe4S4CSEfl41 + CHO)3 #21900 + 5PhSH m-H m—S-+e+ W TB . 81 "1 P-S-+e+ r'v Ta 0 CW 1 T I v T [—T v W—I V v 7f! I l I’ V I V v f T [ v Y—ffi' IO 9 a 7 6 s 4 ppu . Figure 20 58 the Measo-d. solvent (in the presence of water) causes partial salvation of the cluster to produce some free (HO),—tripod ligand as shown in equation 8. This type of equilibrium reaction has been Me SO [FO.S.(0,R)(CI)]" 3—3" [Fe.S.(MeaSO)a(Cl)l* "' (HO),-tripod (8) H30 observed for [Fe.S.Cl,(0Ph),]"’ 49, where the ’H NMR spectrum recorded in Measo-d. showed only the presence of free phenol and hence "complete" salvation of the cluster. The lability of coordinated phenoxide ligands has been demonstrated with [Fe.S.(OPh).]" 82, and it appears that the (H0),-tripod is labile enough to result in salvation of the mixed-ligand cluster. It should be noted that the reaction of [Pr.N],[Fe.S.(SEt).] with the (HO),-tripod ligand shows little evidence of free ligand resonances. One explanation for the absence of free ligand could involve the different preparation of the [Fe.S.(SEt).]" + Na,(O,-tripad) NMR sample compared to that of the chloro tetramer reaction with the Na,(O,-tripad) ligand. When [Fe,S.(SEt).]"‘ was reacted with the ligand a small aliquot of solvent was injected into the reaction flask and then removed to dryness in vacuo, new solvent was then reinjected and the process repeated for a 12 h time period. The cluster mixture was in contact with the Measo-d‘ for only short periods of time. This short exposure to the solvent limits the possibility of salvation of the cluster. The insolubility of the Na, (0,-tripod) ligand made it impossible to carry out the ligand binding reactions in nan-coordinating solvents. Attempts to bind the ligand to an Fe-S cluster in CD,CN, in hopes that the coordinated 59 cluster would be more soluble in CD,CN than the individual starting materials, were unsuccessful. To insure that the (HO),-tripod ligand ar the Fe.S, unit was not decomposing under the reaction conditions, five equivalents of thiophenol was added to the reaction mixture of [Fe.S.(SEt).]"’ and the (Elms-tripod. The resultant spectrum is shown on the bottom of Figure 20. The more acidic thiophenol quantitatively displaces coordinated (H0),-tripod ligand to yield [Fe.S.(SPh).]" and free ligand. New paramagnetic resonances appear at 5.2 (p-H), 5.7 (o—H) and 8.2 (m-H) ppm, which correspond to the reported thiolate Fe-S cluster shiftss“. The remaining peaks in the spectrum were due to the diamagnetic ligand. A series of samples in which the ratio of cluster to ligand was varied, in order to maximize formation of either the tetrameric or hexameric farms of the iron-sulfur clusters, was examined by 1H NMR. Figure 21 shows the spectra that resulted from these experiments and the ratios used. The formation of hexamer should be maximized when the ratio of cluster to ligand is 3:4, since three tetrameric units rearrange to form two hexameric units and each hexameric unit requires two ligands. When the ratio of cluster to ligand is 1:1, the formation of tetrameric cluster should be maximized. The changes in intensities of the o-tet, m-tet, NH-tet, m—hex and H peaks were compared in the stacked plat of spectra. The ratio of ligand to cluster present did effect the amount of tetrameric to hexameric species observed. Irrespective of the ligand to cluster ratio used, the reaction produced a mixture of hexamer and tetramer. The general trend seen in the spectra was that as the peaks H and m-hex grew in intensity, Figure 21. 60 1H NMR spectra (360 MHz) of the reaction between [Et,N]3[Fe.S.Cl,) and Na,(0,-tripad) in Mezso-d‘. .The ratio of cluster to ligand (C:L) used in the reactions is indicated next to each spectrum. 61 LE+4I\DZEFe4S4C :41 + News-TRIPOD) Figure 21 62 the peaks m—tet, o—tet and NH-tet decreased in intensity. This trend suggests that peak H is associated with a hexameric structure. Figure 22 shows a schematic of the possible species present in solution which is consistent with the NMR experiments. The (HO),-tripad ligand is binding to an Fe.S. care, but in solution the mixed-ligand cluster, [Fe.S.(O,R)(Cl)]", readily converts to other clusters. When two mixed-ligand tetramers interact, an [Fe.S.(O,R),]"’ species can be generated in addition to a metastable [Fe,S,Cla(salvent),]" complex which will rapidly rearrange to farm [Fe.S.Cl.]3". The chloro tetramer species can then react with Na,(O,-tripod) to give back the original [Fe.S.(O,R)(Cl)]"‘ complex. When four mixed-ligand complexes interact in solution, a mixture of [Fe.S.(O,R),]"‘ and chloro tetramer will be observed. When three tetrameric mixed-ligand clusters interact, [Fe‘S.(O,R),]"’ can be generated along with a mixed-ligand hexameric cluster, [Fe.S.(O,R)(Cl),]"’. The H peak is assigned to the meta proton of this mixed-ligand hexameric species. In the presence of excess ligand, the [Fe.S.(O,R)(Cl),]" should, in principle, convert to [Fe.S.(O,R),]". The spectrum taken at a cluster to ligand ratio of 1:2 showed a decrease in peak H intensity while the m—hex peak increased, providing further evidence of the assignment of H. Additional evidence of the existence of a mixed-ligand hexameric species is seen in the reaction between [Pr.N];[Fe.S.(SEt)4] and (HO),-tripod. A very broad peak at 14.7 ppm is observed and can be assigned to the CH, protons of the ethanethialate group in [Fe685(0,-tripad)(SEt),]°". The mixture of Fe-S species present in the ‘H NMR studies are a result of a complex rearrangement of the mixed-ligand [Fe,S.(O,-tripod)(Cl)]" cluster in 63 r» [Fe6SGEO3R1213' ~_"—=_3—.. ere4satosRIcnz‘.-J~_=-fi-n 2EFe686EO3R1213 “ ‘9.» + x=2 ,9 + “’00. —[Fe6$6 105110131“ [FeGSGEO3R1213‘ [Fe4s4c1412‘ + [[FeZSZCIZESOLvt-zmlzlz’] Figure 22. Schematic of possible Fe-S clusters present from interconversians of an [Fe.S.(O,R)(Cl)]” cluster in solution. 64 solution. Similarity of solubility properties in polar solvents prevents isolation of either the tetrameric or hexameric species. The structure of [Fe‘S‘Cl.]"’ 37 has been recently reported to be a distorted hexagonal prism in which two cyclahexane-chair F6383 units are eclipsed relative to one another. The distance between iron atoms that lie within a plane perpendicular to the c, axis is 3.8 A, while the distance between iron atoms of the Peas, rhombic side units is 2.7 A. The 2.7 A distance is identical to the separation of iron atoms in a tetrameric Fe-S coresa. The (HO),-tripod ligand is designed to span the corners of an Fe.S. cluster, but there are two possible coordination sites in which the ligand can bind to an Fe.S. care. The (HO),-tripod ligand can coordinate either to three iron atoms that lie within the hexagonal bases of the prism along the C, axis (1), or to iron atoms on different hexagonal bases such that the ligand is bound almost perpendicular to the C, axis (11). A schematic of the two possible structures is shown in Figure 23. Crystallographic data and CPR models were used to estimate the distance necessary for coordination between the terminal oxygen atoms on the ligand and the two possible hexameric structures. The structural parameters for [Et.N],[Fe,S‘Cl.]87 were used to calculate the distance between terminal Cl atoms and these distances were then corrected for the difference between Fe-Cl and Fe-O bond lengths using simple trigonometric relationships. The results show that there has to be a 6.7 A distance between oxygen atoms in order for the. ligand to bind to iron atoms which lie within a hexagonal base and a 4.9 A distance to bind to Fe atoms on alternate bases. The 0-0 distance necessary to span a cubane core is 5.9 A. 65 Figure 23. Schematic drawing of two possible coordination modes of the (H0),-tripad ligand to an Fe.S. hexagonal core. 66 Coordination of the ligand as in structure _I_I should result in the loss of 3-fold symmetry of the molecule and consequently an inequivalence of meta proton resonances should be observed. The six meta protons on each ligand would split into two sets of resonances with relative intensities of 2 to l. The peaks labelled a and c, in the spectra shown in Figure 21, simultaneously increase in intensity along with the meta signals H and m—hex. The a and c resonances could be due to the inequivalent set of protons from the ligand bound to the sides of the hexagonal prism. Qualitatively, it appears as if the more stable hexameric structure is the one where the ligand binds along the C, axis. Steric and electronic interactions are minimized in this configuration because the sulfur atom of the cluster is’ not situated within the ligand cavity and the iron atoms are furthest apart. The observed ratio of ~ 4:1 for peaks H:a and m-hex:c suggests that a rapid interconversion between the two modes of binding may be occurring in solution and the more stable structure dominates. The peaks labelled b and d in Figure 21 could be the NH resonances of the [Fe‘S.(O,R),]" and the [Fe.S.(O,R)(Cl),]" species. Another set of NH peaks is possibly obscured by other resonances in the region between 11 and 12 ppm. The system is too complicated to definatively assign these resonances. The accumulated data indicates the possible existance of five Fe-S clusters in solution. The inability of the (H0),-tripod ligand to stabilize a 3:1 mixed-ligand Fe.S. cluster in solution has been demonstrated. The Fe-O bonds of tetrameric clusters are known to be covalent in nature and are capable of donating electron density into the [Fe.S,]"" core47187. Conversion to the [Fe.S,]’+ core probably 67 allows the spin density to delocalize, as evidenced by the larger contact shifts for the hexameric structures. It is concluded from the NMR experiments that the Fe.S. species are thermodynamically or kinetically stabilized by the terminal oxygen ligation of the (HO),-tripod ligand. C. lfl¢MaI§aMoSa_J__lFe 0A0 21 1. Smthesis Oxygen-ligated molybdenum-iron-sulfur complexes have been synthesized by two methods 19:22. The first method involved a ligand exchange reaction on a preformed S,MoS,Fe cluster. This method was the one used to form [S,MoS,Fe(0Ar),]“ (Ar = Ph, p-tolyl)21. The second method involves a direct self-assembly reaction from tetra- thiomolybdate and an appropriate iron( II) complex. Synthesis of [SzuosaFewAchP' was accomplished using the self-assembly approach. The tetraethylammonium salt of tetra- thiomolybdate was reacted with anhydrous ferrous acetate in acetonitrile as shown in equation 9. The resulting dark brown [Et‘NJQMOS‘ + F9‘03CCH3)3 —_'_" [Eth]:[SgMOSaF6(OaCCH3)3l (9) solution was reduced in volume and addition of diethyl ether precipitated a red-brown solid in 75% yield. Attempts to crystallize this compound from acetonitrile or propionitrile resulted in the formation of a black solid, whose optical spectrum in CH,CN revealed 68 Am at 584, 518, 460, 412, 310 and 288 nm, which resembles the spectrum of [Et.N],[Fe(MoS.),]28. This suggests that the bis(tetra- thiamolybdate)iron trianion is thermodynamically more stable in solution than the oxygen-ligated dimer. The trianion species may be formed due to the acetate groups dissociating in solution and allowing the iron to reduce the molybdenum. Synthesis of [SzMosaFe(O,C,O,C.H,)]" was attempted by ligand substitution reaction of [SaMosaFeClalh with sodium phthalate (NaO,C,O,C.H,). This also resulted in formation of [Fe(MoS.),]"' rather than the desired complex. 2. Physical Propgrties a. Optical The electronic absorption spectrum of [SatioSaFe(OAc),]"’ is shown in Figure 24. Peak positions and molar absorptivities are presented in Table V along with those reported for [SaMoSQFe(OPh),]" and [s,uos,re01,]=-. The spectrum of the iron-molybdenum-acetate dimer shows intense absaptions between 520 and 400 nm. Both phenoxide- ligated and chloro-ligated dimers show the same type of absorption pattern. Any cluster containing a tetrathiomolybdate unit shows peaks in this region due to sulfur to molybdenum charge transfer transitions89, which are observed at Am = 470 nm for the free [MoS.]"'. It can be deduced that the band at 464 nm, in the case of the acetate dimer, is a S 9 Mo (1! -> d) transition. The peak at 432 nm of the chloro dimer has been assigned to 69 0.5'] 0.4- Lu - “2’ £0.3- Cl: 0 . ‘éi O.2~ O.l« 200 ' 500 ' €300 I 800 ROW) Figure 24. Electronic spectum of [Et.N] , [S 3 MoS , Fe(0Ac) ,] in CH,CN. 70 Table V. Electronic spectral features and molar absorptivitiesa for [SgMOSzF9(OAC)ala- 3 [SaMOSaFOCla]3- b and [S ,MoS ,Fe(OPh ) 3 ] 3" ‘3 complexes. Complex 1 (102 8) [SaflosaFe(OAc)3]" 286 (78) 310 (100) 428 (sh) [SaMosaFe(OPh)3]” 232 (197) 314 (147) 391 (75) [8,MoS,FeCI,]" 290 (110) 314 (124) 432 (48) 464 (53) 516 (sh) 471 (98) 530 (sh) 469 (64) 528 (sh) ain acetonitrile solutions; values in ms, with molar absorptivities in parentheses. bReference 21. °Reference 19. 71 another sulfur to molybdenum charge transfer band which arises from a splitting of degenerate energy levels of the tetrathiomolybdate when the iron binds to the sulfur”. Coordination of the iron lowers the symmetry of the [MoS.]" unit, and hence two bands are observed. The shoulder peak at 428 nm observed in the acetate dimer may be due to this same phenomenon. The shoulder peak at 516 nm of the acetate dimer can be assigned as a sulfur to iron charge transition since no such low energy absorption is observed in the spectra of [MoS.]3' anions. The oxygen ligation does not change the spectrum much from that of the chloro-ligated dimer; but the peaks are blue-shifted from other thiolate analogs 19. b. Proton nuclear magnetic resonance The variable temperature ‘H NMR spectra of [s,uos,re(ono),]=- recorded in the range of -25 'C to 40 ‘C in CD,CN are shown in Figure 25. Table VI shows the isotropic shifts based on a diamagnetic shift of 2.01 ppm for free acetic acid. The complex shows one broad peak due to the methyl protons of the acetate group. The resonances due to the tetraethylammonium cation are broadened slightly but remained unshifted with the change in temperature. As the temperature is increased the isotropic shift (UH/Him,» of the acetate protons is decreased. A decrease in shift as the temperature is increased was observed by Silvia91 for dominant contact interactiOns of the thiophenolate and thiotosylate-ligated Fe-Mo-S dimers. By analogy, it can be assumed that the paramagnetic shifts for the acetate dimer are due to dominant contact interactions. The equation for contact 72 -———+ l I I i T l l r 1 1 1 I l l l ‘ ' T 80 60 “+0 20 PPM Figure 25. ‘H NMR spectra (360 MHz) of [Et.N],[S,MoS,Fe(OAc),] in CD,CN at various temperatures. 73 Table VI. Isotropic proton nuclear magnetic resonance shifts“ ((AH/Hhao) observed for the OAc group of [Et.N],[S,MoS,Fe(OAc)sl at various temperatures in CD,CN. Temperature ('C) (AH/H)isob (99-) — 25 74.96 - 6 63.94 + 5 57.93 + 25 49.88 + 40 45.07 “relative to 'I‘MS . , l’dianagnetic resonance of OAc group at 2.01 ppm. 74 shift (AH/HM” is usually written as a function of the electron spin-nuclear spin coupling constant, A; (equation 10)83; (All) 3 A! = AiEavE.S_.(§_+_1.l (10) ( Hliao V gufiu3kT where v is the probe frequency and Av is the frequency separation of the paramagnetic and diamagnetic shifts in Hz. The magnetic susceptibility for a molecule can be expressed as in equation (11), where x = fig'pfits + 1) (11) 3kT N is Avogadro’s number. It can be seen from comparing equation 10 and 11 that x is directly proportional to (AH/misc. A plot of (AH/11);“, versus l/T should produce a straight line with a slope proportional to A1 if the system obeys Curie Law behavior. A plot of (AH/11);” versus VT for [SaMosaFe(OAc),]", shown in Figure 26, has a slope A; = 2.79 ° 10"5 K“, and proves that the complex obeys Curie Law magnetism. c. Magnetic Susceptibility The magnetic susceptibility of [Et.N],[S,MoS,Fe(0Ac),] has been measured on a solid sample from 10.9 to 320 K, using a superconducting quantum interference device (SQUID) magnetometer. There are two possible oxidation state combinations for the FeMoS. cluster; either a high-spin Fe(III)(d’)-Mo(V)(d‘) couple, or high-spin Fe(II)(d‘)-— Mo(VI)(d°). A plot of x versus T is shown in Figure 27. The decaying 75 3.1 I I I I I I I I I I 3.3 3.5 3.7 39 4.1 1/ T 003 K) Figure 26. Plot of isotropic shift ((AH/H)iso).versus l/T for [Et.N],[S,MoS,Fe(OAc), ]. XM 76 ’- I ’- I 2. 05-001)— : i )- l. SE-UOl— ' r— L- I l. 0E‘00 l— I : i y. I b A 5. 06-002— ' . , : ' ' ' ' I +- ' I I , ‘ ' ' ' ' " 1 1 1 i i F 0- 0‘5‘00c u L in} ' IIIJU IéU ZLIIU .350 300 T C K) Figure 27. Plot of x versus- T for temperature-dependent magnetic susceptibility data on [Eth];[S3MOSaF9(OAC)]o 77 curve indicates paramagnetism. The effective magnetic moment at room temperature, which has been corrected for diamagnetic contributions of the ligands and cations through the use of Pascal’s constants83, is 4.96 up. This is consistent with the presence of four unpaired electrons (S = 2) and indicates an Fe(II)-Mo(VI) system. The up value is within the range of monomeric high spin-iron(II) complexes (4.9 - 5.5 up)”. A strongly coupled Fe(III)-Mo(V) system was eliminated as a possible description because this oxidation state couple would show a magnetic susceptibility of S = 2 at low temperatures and S = 3 at high temperatures. No evidence of a transition between S = 2 and S = 3 was observed over the temperature range measured. A plot of x versus 1/T is a straight line from which an effective moment was calculated to be 4.6 lip/formula unit. The Curie-Weiss formula x = (C/T + O) was fitted to susceptibility data above 40 K by a linear least squares routine to give (an a per mole basis) 0 = 9.93 K, C = 3.14 (emu ° K)/mol. 0 corrects the temperature for the non-zero intercept. The [SauosaFe(SPh),]" cluster shows similiar values of 0 = 4.2 K and C = 3.19 (emu ° K)19. III. EXPERIMENTAL A. Materials Manipulations involving air sensitive products (i.e. iron-sulfur clusters) were performed under a dry argon atmosphere. Argon was purified by passage through columns of hot BASF R 3-11 catalyst (copper(II) oxide) and Aquasorb (supported phosphorus pentoxide). MeOH and CH,CN were purified by distillation under dinitrogen from magnesium methoxide and calcium hydride, respectively. DMA and M9380 were allowed to stand over activated 4 A molecular sieves prior to use. The tetraethylammonium salts of [MoS.]" 31 and [Fe.S.Cl.]"‘ 92 were supplied by Paul E. Lamberty and Walter E. Cleland, respectively. 1,8-Dichloroanthracene55, 2-amino-6—chlorobenzoic acid73, p—aminobenzene p—xylene sulfide93 and ferrous acetate94 were synthesized by literature procedures. 2-Amino-6-methyl benzoic acid was recrystallized from a 50/50 mixture of EtOH/Hao. 4-Aminophenol was recrystallized twice from CH,CN prior to use. All other reagents were obtained from Aldrich Chemical Company, Inc. and were used without further purification, unless otherwise stated. The silica gel used for column chromatography was obtained from Baker and was 230-400 mesh. 78 79 B. Physical methods ‘11 and ‘ ’C NMR spectra were recorded on a Nicolet NTC-36O instrument (360 MHz). Chemical shifts are reported in parts per million (6) relative to internal standard (CH,).Si. Anaerobic NMR samples were prepared in a side-arm flask and transferred via cannulla into a 9 inch NMR tube (which had been degassed and flushed with argon three times). The tubes were either sealed under argon or used with a serum stopper. Low temperature NMR data were obtained by cooling the probe with dinitrogen passed through coils emerged in an isopropanol/dry ice bath. IR spectra were obtained on a Perkin-Elmer model 1430 ratio recording spectrophotometer, using polystyrene for calibration. Optical spectra were obtained on a Cary 219 spectro- photometer, where any solvent absorption was subtracted from the spectra. Mass spectra were measured at 70 all on a Finnegan MAT model 4600 GC/MS. High resolution mass spectra were obtained on a Finnegan MAT model 8230 GC/MS. Uncorrected melting points were determined either on a Thomas—Hoover capillary apparatus (< 200 ’C) or a Laboratory Device Mel-Temp apparatus (> 200 '0). Variable temperature magnetic susceptibility measurements were performed on a 8.8.8. Corporation SQUID susceptometer. HPLC analyses, using a Waters Associates p—Porasil column, were performed with a Waters Associates model M-45 solvent delivery system equipped with a model U6K injector and model 440 absorbance detector. Semi-preparative scale HPLC ' separations were done using a Whatman Partisil 10 Magnum 9 (50 cm length) silica column. Radial thin-layer chromatography was performed on a 4mm silica plate using a Harrison Research model 7924 80 Chromatotron. Microanalyses were performed by Atlantic Microlab Inc., Atlanta, Georgia. C. Pre ation of 1 disubstituted anthra uinones l,8-Diczanoanthraguinone (10): The procedure for synthesis of l-cyanoanthraquinone from l-chloroanthraquinone was followed95, with 1,8-dichlaroanthraquinone as the starting material. l,8-Dichlora- anthraquinone (10.0 g, 36 mmol) and CuCN (9.2 g, 0.10 mol) were slurried in DMA (50 mL) and refluxed under Ar for 3 h. The hot brown solution was poured onto ice (700 g), and the brown-green precipitate was filtered and washed with water. The copper complex was decomposed with 3N ENG, (500 mL) at 60 ‘C for 4 h. The brown solid was filtered, washed with water and air dried. This procedure afforded crude 10 (8.2 g, 88%): mp 402-406 ‘0 Hit.“ mp > 390 '0); ‘H NMR (Me,SO-d.) 6 8.51 (2H, d, J = 7.92 Hz), 8.44 (2H, d, J = 7.56 Hz), 8.12 (2H, dd, J = 7.74, 7.74 Hz); IR (KBr) 2216 (m, -CN), 1678 (s, -C=O) cm"; MS (EI) m/e (relative intensity) 258 (M‘, 100), 230 (55), 211 (82), 175 (44), 149 (11), 101 (31), 87 (13), 75 (41). Anthraguinone-l&dicarboleic gold (11): The procedure of Waldmann and mum” was followed with modifications. Crude 10 (8.2 g, 30 mmol) was refluxed in 70% H380. (500 mL) for 1 h. The hot . solution was poured onto ice (800 g) to precipitate crude 11 as a brown solid (8.2 g, 87%): mp 294-300 ‘C (lit.68 mp 316 ’C); 1H NMR (Me,SO-d‘) 6 13.40-13.10 (2H, br, s), 8.28 (2H, d, J = 7.56 Hz), 7.96 (2H, dd, J = 7.92, 7.92 Hz), 7.85 (2H, d, J = 7.20 Hz); IR (KBr) 3400-2750 (s, -OH), 1710-1665 (br, s, -C=O) cm"; MS (EI) m/e (relative intensity) 296 81 (16", 0.5), 279 (7), 252 (71), 234 (100), 208 (16), 180 (32), 150 (47), 139 (27), 75(39). D. Preparation of 1,8-disubstituted anthracenes 1,8-Diczanoanthracene (3): The procedure of Akiyama, et al.56 was followed with modifications. 1,8-Dichloroanthracene 255 (6.7 g, 30 mmol) and CuCN (8.1 g, 90 mmol) were slurried in distilled quinoline (70 mL) and refluxed for 24 h under Ar. The warm black solution was poured into IM H01 (600 mL), producing a black solid that was filtered and washed with water. The solid product was partitioned between 1M NH.OH (300 mL) and CH,C1, (300 mL) and stirred vigorously for 6 h. The blue aqueous layer was separated and fresh 1M NH.OH (300 mL) was added to the organic phase and allowed to stir for another 6 h. This procedure was repeated until the aqueous layer was no longer blue. Rotary evaporation of the organic layer left a brown oil which was chromatographed (Rf 0.32, silica gel, CH,Cl,) to afford pure 3 as a yellow solid (2.5 g, 417;): mp 300-303 ‘c an.“ mp 304-306 '0); in NMR (CDCl,) 6 9.16 (1H, s), 8.65 (1H, s), 8.31 (2H, d, J = 10.8 Hz), 8.08 (2H, d, J = 7.2 Hz), 7.63 (2H, dd, J = 10.8, 10.8 Hz); IR (KBr) 2216 (s, -CN) cm"; MS (EI) m/e (relative intensity) 228 (M‘, 100), 201 (11), 175 (5), 100 (10), 87 (11), 74 (5). Anthracene-1,8-dicarboleic 921d ()3): The procedure of Waldmann and mum“ was followed with modifications. Crude 11 (8.2 g, ' 30 mmol) and Zn dust (30 g, 0.5 mol) were refluxed with stirring in 20% NH.OH (350 mL) for 4 h, during which the color changed from dark red to yellow. The solution was filtered to remove excess Zn, and water 82 (500 mL) was added to the yellow filtrate. The filtrate was cooled to 0 ’C, and 10% HCl was slowly added until a yellow precipitate formed. Filtration of the solid and air drying yielded crude 12 (5.9 g, 79%): mp 345-347 'C (dec.) (lit.68 mp 345 'C (dec.)); In NMR (Me,so-d.) a 13.30-13.00 (2H, br, a), 10.47 (1H, s), 8.78 (1H, s), 8.34 (2H, d, J = 8.28 Hz), 7.62 (2H, dd, J = 7.74, 7.74 Hz); IR (KBr) 3300-2450 (br, s, -OH), 1712 (s, -C=O) cm"; MS (EI) m/e (relative intensity) 266 (M’, 100), 249 (7), 236 (2), 221 (14), 204 (40), 192 (5), 176 (4), 166 (27), 150 (6), 139 (4), 124 (4), 110 (6), 97 (5), 82 (9), 69 (6). 1,8-§is(methoxycarbonzlhnthracene Q4: The procedure of Akiyama, et al.69 was followed with modifications. Compound 12 (2.9 g, 10 mmol) was refluxed in MeOH (400 ml.) with concentrated H380. (4 mL) for 16 h. Water (100 mL) was added to the warm brown solution, and the product was extracted with CH,Cl, until the organic layer was no longer yellow. Concentration of solvent by rotary evaporation left a brown oil which was chromatographed (Rf 0.57, silica gel, 03.01,) to yield pure 13 (1.8 g, 61%): mp 101-103 ’C (lit.69 mp 104-105 ’C); ‘H NMR (CDCl,) 6 10.71 (1H, s), 8.49 (1H, s), 8.28 (2H, d, J = 6.84 Hz), 8.18 (2H, d, J = 8.28 Hz), 7.52 (2H, dd, J = 7.20, 7.20 Hz); ”C NMR (CDCl,) 6 167.35, 133.10, 130.97, 130.89, 129.14, 127.31, 127.10, 123.90, 123.65, 51.95; IR (KBr) 1707 (s, -C=O) cm"; MS (EI) m/e (relative intensity) 294 (M"’, 100), 263 (52), 235 (25), 220 (19), 203 (18), 176 (16), 164 (9), 150 (4), 131 (5), 123 (11), 116 (19), 102 (20), 88 (27), 75 (8). 83 E. Prepa_ration of the a—methoxzcarbonzlbenzzne precursor l-Methzl-2-hzdrogen-S-nitrophthalate (15): The procedure reported by Nagai, et al.70 was followed with some modifications. Reagent grade 3-nitrophthalic acid (50 g, 0.24 mol) was dissolved in anhydrous MeOH (200 mL) and was filtered through a fine glass frit to remove small black impurities present in the starting material. The resulting yellowish solution was cooled to 0 ’C, and dry HCl gas was bubbled through at a rate of approximately 1 bubble per sec for 20 min. The resulting colorless solution was refluxed for 2-3 h and poured while hot into ice water (900 ml.) to precipitate the product 15. The white microcrystals were filtered, washed with cold water, and dried in vacuo at 60 ‘C. This procedure afforded pure 15 (41.3 g, 77%): mp 162-164 'C (lit.70 mp 160-162 'C); in mm (Me,so-d.) a 14.10-13.75 (III, br, s), 8.33 (1H, d, J = 8.17 Hz), 8.22 (1H, d, J = 7.72 Hz), 7.82 (1H, dd, J = 8.01, 8.01 Hz); IR (KBr) 3400-3000 (br, s, -OH), 1766 (s, -C=O ester), 1700 (s, :0 acid), 1541 (s, -NOz), 1352 (s, -N02) cm"; MS (EI) m/e (relative intensity) 225 (M"’, 2), 208 (68), 194 (100), 181 (37), 164 (22), 151 (79), 136 (46), 119 (36), 104 (63), 92 (41), 75 (68), 63 (30). 2-Amino-6-(methoxzcarbonzl)benzaic £151 fig): Compound 15 (30.0 g, 0.13 mol) was dissolved in MeOH (125 ml.) and placed in a heavy-walled pyrex bottle, to which 5% Pd on charcoal (0.30 g) was added. The rubber stopper used to cover the bottle was lined with Parame to insure that sulfur in the stopper would not poison the catalyst. The reaction flask was pressurized to 40 psi with H, and allowed to shake on a Parr pressure reaction apparatus for 8 h at room temperature. The resultant bright yellow solution was filtered to 84 remove the catalyst. Rotary evaporation of the solvent and subsequent drying in vacuo yielded 16 as a gummy solid (23.8 g, 92%): 1H NMR (Mezso-d.) 6 9.20-8.00 (3H, br, s), 7.21 (1H, dd, J = 7.56, 7.56 Hz), 6.87 (1H, d, J = 8.2 Hz), 6.61 (1H, J = 6.84 Hz), 3.71 (3H, s); 1'-‘C NMR (CDCl,) 6 170.64, 170.40, 135.51, 132.55, 119.31, 116.79, 109.96, 52.56; IR (NaCl) 3490 (w, -NH2), 3380 (w, -NH2), 1712 (m, -C=O ester), 1615 (m, -C=O acid) cm"; MS (EI) m/e (relative intensity) 195 (M"’, 1), 177 (12), 163 (44), 147 (13), 119 (75), 90 (100). F. Premtion of 1,8,13- and 1,8,16—trisubstituted triptycenes General procedure for the premation of 1&,13- and 1,8,16-trisubstituted triptycenes: The 1,8-disubstituted anthracene (5 mmol) was dissolved in a minimal amount of hot DME (10-300 mL). Isoamyl nitrite (1.3 mL, 10 mmol) was added to the refluxing solution and the substituted anthranilic acid (10 mmol), dissolved in DME (10-20 mL), was added dropwise over a 20-min period. The solution was refluxed 20 min, and another charge of isoamyl nitrite (1.3 mL, 10 mmol) was added. A second aliquot of anthranilic acid (10 mmol), dissolved in DME (IO-20 mL), was added over a 20-min period. The solution was refluxed another 40 min, cooled to 0 'C, and 95% EtOH (20 mL) was added. Gold 7.5% NaOH was added until a precipitate formed. The solid was filtered, washed with cold MeOH:H,O (4:1) until there was no brown color in the filtrate, and dried in vacuo at 60 '0. Purification was performed in some cases by subliming off unreacted anthracene. Separation of some isomers by chromatography was performed as indicated. 85 1,8-Diczano-13-methzltriptzcene (5s) and l,8-diczano-lG-methzltriptycene (5a): Reaction of 3 (1.1 g) with 2-amino-6-methylbenzoic acid (recrystallized from 95% EtOH), by the general procedure above, yielded 5a and 5s, and some unreacted 1,8-dicyanoanthracene. Sublimation of the mixture separated the lower melting anthracene from the triptycenes to yield an off-white solid of . 5a and 5s (0.9 g, 57%). Separation by semi-preparative HPLC afforded pure 5s and pure 5s as white solids. 5_a: mp 365-368 'C (dec.), ‘H NMR (CDCl,) 6 7.59 (2H, d, J = 7.20 Hz), 7.45 (1H, d, J = 7.20 Hz), 7.33 (2H, d, J :: 7.92 Hz), 7.14 (2H, dd, J = 7.56, 7.56 Hz), 7.00 (1H, dd, J = 7.56, 7.56 Hz), 6.93 (1H, d, J : 7.56 Hz), 6.26 (1H, s), 5.77 (1H, s), 2.51 (3H, s); ”C NMR (CDCl,) 6 147.54, 145.95, 141.66, 141.53, 132.45, 128.76, 127.96, 127.70, 126.34, 125.93, 122.84, 116.60, 108.83, 50.48, 49.91, 18.49; IR (KBr) 2220 (s, -CN) cm"; MS (BI) m/e (relative intensity) 318 (W, 72), 303 (100), 288 (6), 275 (8), 158 (5), 151 (9), 144 (8), 138 (10), 130 (9), 124 (7); high resolution mass spectrum, exact mass calcd. for C,,H,,N, (11+) 318.1157, found 318.1144; HPLC retention time (CH,Cl,) 10.8 min, flow rate 8.0 mL/min. Anal. Calcd. for CuHuNa: C, 86.77; H, 4.43; N, 8.80. Found: C, 85.67; H, 4.55; N, 8.65. 55: mp 408-411 'C (dec.), ’H NMR (CDCl,) 6 7.50 (2H, d, J : 7.20 Hz), 7.33 (2H, d, J = 7.92 Hz), 7.26 (1H, d, J = 6.12 Hz), 7.14 (2H, dd, J = 7.56, 7.56 Hz), 6.97 (1H, dd, 7.02, 7.02 Hz), 6.94 (1H, d, J : 6.12 Hz), 6.58 (1H, s), 5.54 (1H, s), 2.66 (3H, s); ”C NMR (CDCl,) 6 147.27, 146.36, 143.52, 140.13, 133.75, 128.55, 127.83, 127.70, 126.38, 125.97, 121.79, 116.59, 108.83, 53.73, 46.58, 18.58; IR (KBr) 2224 cm"; MS (EI) m/e (relative intensity) 318 (M"’, 100), 303 (84), 288 (5), 275 (8), 158 (4), 86 144 (7), 138 (7), 132 (8), 124 (5); high resolution mass spectrum, exact mass calcd. for CnHuN, (M"’) 318.1157, found 318.1144; HPLC retention time (CH,C1,) 9.2 min, flow rate 8.0 mL/min. Anal. Calcd. for (C,,H,,N,)°(CH,Cl,),/.: C, 84.29; H, 4.36; N, 8.51. Found: C, 83.90; H, 4.74; N, 8.22. l3-Methzl—triptzcene 1,8—dicarboleic pc__ig _(_6_s_)_ §_r_l_d_ 16-methzl-triptzcene 1,8-dicarboleic acid (6a): A mixture of 5a and 5s (0.75 g, 0.23 mmol) and KOH (6.7 g, 0.12 mol) was dissolved in ethylene glycol (30 mL). The yellow solution was allowed to stir for 4 days at 100 'C and then cooled to room temperature. Water (10 mL) was added to the solution, and HCl (6N) was added dropwise until a white percipitate formed. The solid was collected on a fine glass frit, washed with H.O, and dried overnight in vacuo to yield 611 and 6a (0.77 g, 93%): mp > 450 'C; ‘H NMR (Measo-d.) 6 13.20-13.00 (2H, br, s), 8.09 (1H, s), 7.75 (1H, s), 7.65 (2H, d, J = 7.20 Hz), 7.61 (3H, d, J = 7.20 Hz), 7.44 (3H, d, J = 7.56 Hz), 7.26 (2H, d, J = 6.84 Hz), 7.21 (1H, d, J = 6.48 Hz), 7.08 (3H, dd, J = 7.56, 7.56 Hz), 6.89 (3H, m), 6.02 (1H, s), 5.77 (1H, s), 3.38 (3H, s), 2.56 (3H, s); MS (EI) m/e (relative intensity) 356 (M"', 100), 338 (20), 310 (49), 293 (45), 279 (11), 265 (47), 252 (28), 239 (15), 189 (7), 169 (9), 146 (7), 131 (44), 125 (22), 119 (18), 97 (18), 85 (21). 1,g-Dichloro-13-methzltriptxcene (g) a_r_l_d l,8-dichloro-16-methzltriptzcene (82),: Reaction of 255 (1.2 g) with 2-amino-6-methylbenzoic acid (recrystallized from 95% EtOH), by the general procedure above, yielded a white solid of 8a and 8s (1.3 g, 74%). Pure 8s was obtained by selective crystallization from EtOAc. 83 and 8s: (Data are reported for a 25:75 mixture of 8a and 8s.) 87 mp 332-335 'C; ‘H NMR (Me3SO-d.) 6 7.51 (2H, d, J = 7.20 Hz), 7.46 (2H, d, J = 7.20 Hz), 7.36 (1H, d, J = 7.20 Hz), 7.35 (1H, dd, J = 6.48, 6.48 Hz), 7.16 (2H, d, J = 7.20 Hz), 7.15 (2H, d, J = 7.20 Hz), 7.08 (2H, dd, J = 7.20, 7.20 Hz), 7.07 (2H, dd, J = 7.20, 7.20 Hz), 6.92 (4H, m), 6.60 (1H, s), 6.28 (1H, s), 6.06 (1H, s), 5.84 (1H, s), 2.49 (3H, s), 2.48 (3H, 8); HC NMR (CDCl,) 6 147.79, 147.34, 144.82, 143.10, 142.97, 142.19, 141.88, 141.81, 133.03, 132.10, 129.91, 129.84, 127.26, 127.13, 126.45, 126.40, 125.97, 125.81, 125.25, 122.36, 122.01, 121.51, 54.67, 50.77, 47.32, 43.39, 18.60, 18.49; MS (EI) m/e (relative intensity) 340 (M+ + 4, 4), 338 (M+ + 2, 47), 336 (W, 80), 301 (66), 286 (45), 266 (100), 250 (17), 189 (8), 132 (30), 125 (8); high resolution mass spectrum, exact mass calcd. for C,,H,.Cl, (M"') 336.0473, found 336.0479. Anal. Calcd. for (C,,H,.Cl,)°(CH,Cl,),/,: C, 68.00; H, 3.98. Found: C, 67.72; H, 3.73. .8_s: mp 355-357 'C; 'H NMR (CDCl,) 6 7.26 (2H, d, J = 6.84 Hz), 7.23 (1H, d, J = 7.20 Hz), 7.04 (2H, d, J = 7.92 Hz), 6.93 (2H, dd, J = 7.56, 7.56 Hz), 6.92 (1H, dd, J = 7.56, 7.56 Hz), 6.89 (1H, d, J = 6.84 Hz), 6.73 (1H, s), 5.43 (1H, s), 2.60 (3H, s); "C NMR (CDCl,) 6 147.81, 141.90, 133.04, 129.86, 126.76, 126.09, 125.45, 124.89, 121.64, 121.14, 54.68, 43.40, 18.59; MS (EI) m/e (relative intensity) 340 (M+ + 4, 6), 338 (M"’ + 2, 32), 336 (M"’, 61), 301 (75), 286 (37), 266 (100), 250 (15), 176 (4), 150 (8), 143 (10), 131 (37), 118 (11); high resolution mass spectrum, exact mass calcd. for C,,H,.Cl, (W) 336.0473, found 336.0479. Anal. Calcd. for (C,,H,.Cl,)°(CH,Cl,),/.: C, 72.94; H, 4.13. Found: C, 73.44; H, 4.29. 1,8;Dich1oro-13-dibromomethzl triptycene (93) gr_l_d_ 1 8-dichloro-16- dibromomethyl triptycene (9a): A mixture of 8a and 8a (0.20 g, 0.59 mmol) was dissolved in 001. (60 mL). The solution was flushed with Ar, and N-bromosuccimide (0.21 g, 1.2 mmol) along with a few granules of 88 dibenzoyl peroxide were added to the reaction flask. The solution was stirred under a high-intensity lamp for 12h. The flask was cooled to 0 ‘C in an ice/water bath and precipitated succimide was filtered off through a Buchner funnel. The filtrate was rotary evaporated to dryness to yield 9a and 9s as a white solid (0.24 g, 83%): mp > 450 'C; ‘H NMR (CDCl,) 6 7.46 (2H, m), 7.34 (1H, d, J = 7.20 Hz), 7.28 (3H, m), 7.10 (68, m), 6.99 (5H, m), 6.93 (1H, s), 6.83 (18, s), 6.57 (18, s), 6.47 (1H, s), 5.48 (1H, a); MS (EI) m/e (relative intensity) 494 (M"', 8), 414 (90), 350 (10), 336 (100), 117 (30). 1,8,13-Tris(methoxycarbonylflriptzcene (1h) a_n_d 1,8,1§:tris(methoxycarbonyl)triptzcene (17a): Reaction of 13 (1.5 g) with 16, by the general procedure above, yielded 17a and 17s (1.3 g, 62%). Separation by radial chromatography (Rf 0.24 (17a), Rf 0.10 (17s), silica gel, CH.Cl.) and subsequent recrystallization from 1:1 CH,Cl,: EtOAc afforded colorless crystals of pure 17a and pure 17s. _1_71_l_: mp 256-257 'C; ‘H NMR (CDCl,) 6 8.01 (1H, s), 7.70 (1H, d, J = 7.20 Hz), 7.65 (1H, d, J = 7.92 Hz), 7.61 (4H, d, J = 7.92 Hz), 7.08 (1H, dd, J = 7.56, 7.56 Hz), 7.07 (2H, dd, J = 7.56, 7.56 Hz), 6.95 (1H, s), 4.02 (9H, s); ”C NMR (CDCl,) 6 167.32, 147.40, 146.38, 146.05, 145.94, 129.14, 128.15, 127.10, 126.73, 125.40, 125.20, 52.06, 52.02, 50.23, 46.75; IR (KBr) 1740-1720 (br, s, -C=0), 1600 (s, -C=C), 1300-1250 (br, s, -CO,) cm“. MS (EI) m/e (relative intensity) 428 (M"', 100), 413 (4), 397 (36), 381 (6), 368 (22), 337 (85), 309 (11), 305 (8), 293 (17), 278 (19), 266 (6), 250 (29), 237 (7), 183 (22), 176 (4), 161 (4), 153 (5), 147 (4), 139 (8), 125 (33), 118 (7), 84 (9). Anal. Calcd. for 0,.8,,o.: C, 72.89; H, 4.71. Found: C, 72.80; H, 4.69. 115: mp 287-289 ‘C; ‘H NMR (CD013) a 8.69 (1H, s), 7.57 (3H, d, 89 J = 7.92 Hz), 7.52 (3H, d, J = 7.20 Hz), 7.07 (3H, dd, J = 7.92, 7.92 Hz), 5.53 (1H, s), 4.05 (9H, s); ”C NMR (CDCl,) 6 167.40, 146.73, 144.72, 127.86, 126.97, 126.92, 125.19, 54.23, 52.02, 43.63; IR (KBr) 1740-1700 (br, s, -C:O), 1600 (s, -C:C), 1310 (a, -CO,), 1260 (s, -CO,); MS (EI) m/e (relative intensity) 428 (M+, 100), 413 (3), 397 (31), 368 (19), 337 (27), 310 (13), 293 (13), 278 (12), 266 (5), 250 (19), 237 (7), 198 (13), 183 (9), 147 (5), 125 (26), 118 (9). Anal. Calcd. for C,.H“0.: C, 72.89; H, 4.71. Found: C, 72.43; H, 4.66. Triptzcene 1,8,13-tricarboleic acid (1;) pp}; triptycene 1,8,16-tricarboleic pc_ic_i, (la): Compound 17a or 17s (2.5 g, 5.8 mmol) was dissolved in MeOH (625 mL). 10% KOH (125 ml.) was added to the solution which was then refluxed for 12 h. The reaction mixture was cooled to room temperature, water (200 mL) was added, and the solution was concentrated by rotary evaporation to half the volume of solvent. 6N HCl was added with stirring until a white precipitate formed. The solid was collected on a fine glass frit, and dried in vacuo at 60 'C overnight to yield pure 7a or pure 7s: (2.1 g, 95%). 19: mp 426-428 'C; ‘H NMR (Measo-d‘) 6 13.12 (3H,s), 7.87 (1H, s), 7.64 (2H, d, J = 7.20 Hz), 7.60 (1H, d, J = 7.20 Hz), 7.57 (1H, d, J = 7.92 Hz), 7.50 (2H, d, J = 7.92 Hz), 7.15 (1H, dd, J = 7.20, 7.20 Hz), 7.13 (2H, dd, J = 7.20, 7.20 Hz), 6.90 (1H, s); ”C NMR (Mezso-d.) 6 167.85, 167.81, 146.43, 146.16, 145.68, 145.26, 128.20, 127.91, 127.58, 126.72, 126.45, 125.26, 49.64, 46.19; MS (EI) m/e (relative intensity) 386 (W, 100), 368 (12), 340 (25), 323 (50), 295 (14), 279 (44), 250 (29), 239 (15), 237 (10), 184 (6), 124 (14), 119 (11), 75 (5). Anal. Calcd. for C,,H,.O.: C, 71.50; H, 3.65. Found: C, 71.40; H, 3.76. Ls; mp 432—435 'C; 1H NMR (Mezso-d.) 6 12.91 (3H, s), 8.35 (1H, s), 90 7.63 (3H, d, J = 7.20 Hz), 7.40 (3H, d, J = 7.92 Hz), 7.11 (3H, d, J = 7.56 Hz), 5.87 (1H, s); ”C NMR (Megso-d.) 6 167.85, 146.96, 129.31, 126.62, 125.85, 125.06, 52.58, 43.45; MS (EI) m/e (relative intensity) 386 (M"‘, 24), 324 (21), 279 (10), 250 (10), 239 (8), 84 (100); high resolution mass spectrum, exact mass calcd. for CnHuO, (M"‘ - (CC, + H,O)) 324.0786, found 324.0775. Anal. Calcd. for C,,H,.O.: C, 71.50; H, 3.65. Found: C, 70.64; H, 3.72. Triptycene 1,8,13-tris(carbonzlchloride) (18s) and triptycene 1,8,16-tris(carbonzlchloride) (18a): Compound 7a or 78 (0.2 g, 0.52 mmol) and SOC], (10 mL) were mixed under Ar, and the slurry was refluxed for 12 h. SOC], was removed under reduced pressure to yield 1811 or 188 as a white solid: (0.23 g, 98%). 1,8-Bis(methoxycarbonyl)-13-methzltriptzcene (23s) a_n§ 1,8-bis(methoxzcarbonzl)-16-methzltriptzcene (23a): Reaction of 13 (1.5 g) with 2-amino-6-methylbenzoic acid (recrystallized from 95% EtOH), by the general procedure above, yielded 23a and 23s (1.1 g, 58 %). Separation by semi-preparative HPLC afforded pure 23a and pure 23s as white solids. 233: mp 215-216 'C; 'H NMR (CDCl,) 6 7.90 (1H, s), 7.58 (2H, d, J = 7.92 Hz), 7.50 (2H, d, J = 7.20 Hz), 7.39 (1H, d, J = 7.20 Hz), 7.03 (2H, dd, J = 7.56, 7.56 Hz), 6.92 (1H, dd, J = 7.56, 7.56 Hz), 6.85 (1H, d, J = 7.56 Hz), 5.71 (1H, s), 4.01 (3H, s), 2.50 (3H, s); "C NMR (CDCl,) 6 167.41, 146.80, 146.35, 143.86, 143.23, 131.69, 128.94, 127.26, 127.03, 126.90, 126.63, 125.22, 124.90, 122.85, 52.02, 50.63, 47.02, 18.35; MS (EI) m/e (relative intensity) 384 (M"’, 71), 362 (5), 353 '(14), 337 (13), 331 (4), 324 (16), 310 (4), 303 (4), 293 (100), 278 (5), 265 (25), 263 (29), 250 (19), 239 (7), 189 (7), 176 (5), 146 (5), 132 (19), 125 (10 ), 75 (4); HPLC retention 91 time (CH3C13) 13.3 min, flow rate 8.0 mL/min. Anal. Calcd. for C35H300.: C, 78.11; H, 5.24. Found: C, 78.15; H, 5.26. _2_§_s: mp 231-233 'C; 1H NMR (CDCl,) 6 8.32 (1H, s), 7.58 (2H, d, J = 7.92 Hz), 7.49 (2H, d, J = 7.20 Hz), 7.22 (1H, d, J = 7.20 Hz), 7.02 (2H, dd, J = 7.56, 7.56 Hz), 6.90 (1H, dd, J = 7.20, 7.20 Hz), 6.87 (1H, d, J = 7.20 Hz), 5.47 (1H, s), 4.01 (3H, s), 2.68 (3H, s); ”C NMR (CDCl,) 6 167.40, 147.31, 145.98, 145.03, 142.25, 134.04, 127.26, 127.14, 126.84, 126.70, 125.13, 124.93, 121.19, 54.51, 52.00, 42.94, 18.59; MS (EI) m/e (relative intensity) 384 (M"', 100), 352 (59), 324 (37), 309 (12), 292 (91), 278 (7), 265 (59), 263 (57), 250 (29), 239 (12), 226 (4), 189 (16), 176 (10), 161 (13), 146 (7), 132 (24), 125 (10 ), 118 (6), 84 (21), 75 (6); HPLC retention time (CH,Cl,) 9.7 min, flow rate 8.0 mL/min. Anal. Calcd. for C,,H,oo,: C, 78.11; H, 5.24. Found: C, 77.70; H, 5.20. ‘ 1.8,13-Trichlorotriptycene (2_4_§)_ egg 1,8,16-trichlorotriptzcene L23_a_)_: Reaction of 255 (1.2 g) with 2-amino-6-chlorobenzaic acid7 3, by the general procedure above, yielded 24a and 24s (0.48 g, 27%). This pair of isomers was not separated. pg pan—d 2_4s: mp 355-358 'C (dec.), ‘H NMR (Me,SO-d,) 6 7.58 (2H, d, J = 7.92 Hz), 7.55 (3H, d, J = 7.92 Hz), 7.51 (1H, d, J = 7.20 Hz), 7.21 (2H, d, J = 8.64 Hz), 7.20 (1H, d, J = 8.64 Hz), 7.20 (3H, d, J : 7.92 Hz), 7.12 (1H, dd, J = 7.92, 7.92 Hz), 7.12 (3H, dd, J = 7.92, 7.92 Hz), 7.11 (2H, dd, J = 7.92, 7.92 Hz), 6.82 (1H, s), 6.38 (1H, s), 6.17 (1H, s), 5.98 (1H, s); MS (EI) m/e (relative intensity) 360 (M+ + 4, 9), 358 (M+ + 2, 27), 356 (16+, 33), 321 (42), 286 (100), 250 (34), 176 (8), 160 (7), 143 (20), 125 (22), 112 (5); high resolution mass spectrum, exact mass calcd. for CaoH,,Cl, (M"’) 355.9926, found 355.9950. Anal. Calcd. for (Cagng 1013).(CH3013)‘/0: c, 65s63; H, 3.07. Found: c, 65014; H, 20980 92 1,8-Bis(methoxycarbonyl)-13-chlorotriptycene (25s) and 1,8-bis(methoxycarbonyl)-16-chlorotriptzcene (25a): Reaction of 13 (1.5 g) with 2-amino-6-chlorobenzoic acid”, by the general procedure above, yielded 25a and 25s (0.40 g, 20%). Separation by semi- preparative HPLC yielded pure 25a and pure 25s as white solids. Zia: mp 218-219 ’C; ‘H NMR (CDCl,) 6 7.99 (1H, s), 7.62 (2H, d, J = 7.92 Hz), 7.57 (2H, d, J = 7.20 Hz), 7.44 (1H, d, J = 7.20 Hz), 7.08 (1H, d, J = 7.92 Hz), 7.05 (2H, dd, J = 8.28, 8.28 Hz), 6.95 (1H, dd, J : 7.56, 7.56 Hz), 5.97 (1H, s), 4.02 (6H, s); ”C NMR (CDCl,) 6 167.25, 146.61, 146.01, 145.94, 142.59, 129.30, 127.75, 127.23, 127.01, 126.72, 126.04, 125.26, 123.48, 52.07, 50.69, 46.94; MS (EI) m/e (relative intensity) 406 (M"' + 2, 38), 404 (M"‘, 100), 373 (61), 368 (11), 344 (54), 337 (45), 312 (83), 301 (6), 293 (13), 286 (57), 278 (41), 273 (6), 266 (34), 249 (67), 238 (34), 223 (12), 210 (9), 186 (12), 175 (30), 169 (11), 142 (13), 138 (21), 124 (37), 118 (29), 111 (11), 85 (9), 83 (78); high resolution mass spectrum, exact mass calcd. for CuH, ,O.Cl (M"’) 404.0815, found 404.0796; HPLC retention time (CH,Cl,) 17.3 min, flow rate 6.0 mL/min. Anal. Calcd. for (ONE,,O,Cl)-(CH,Cla),/.: C, 69.74; H, 4.18. Found: C, 69.73; H, 4.33. . _2§g: mp 204-205 ‘C; ‘H NMR (CDCl,) 6 8.41 (1H, s), 7.63 (2H, d, J = 7.56 Hz), 7.52 (2H, d, J = 7.20 Hz), 7.28 (1H, d, J = 7.20 Hz), 7.08 (2H, dd, J = 7.56, 7.56 Hz), 7.07 (IH, d, J = 7.56 Hz), 6.95 (1H, dd, J = 7.56, 7.56 Hz), 5.51 (1H, s), 4.06 (6H, s); ”C NMR (CDCl,) 6 146.81, 144.85, 127.76, 127.31, 127.25, 126.73, 126.69, 126.41, 125.35, 121.86, 54.53, 52.17, 51.90; MS (EI) m/e (relative intensity) 406 (M"' + 2, 40), 404 (M"’, 100), 373 (80), 367 (25), 344 (42), 336 (66), 312 (67), 305 (7), 300 (24), 293 (34), 286 (74), 277 (32), 273 (8), 265 (43), 249 (58), 237 (43), 22 (14), 93 209 (12), 185 (16), 175 (30), 161 (8), 142 (17), 138 (28), 131 (29), 124 (35), 118 (26), 111 (24), 83 (78), 71 (9); high resolution mass spectrum, exact mass calcd. for CuH, ,0,Cl (M"‘) 404.0815, found 404.0796; HPLC retention time (CH,Cl,) 18.1 min, flow rate 6.0 mL/min. Anal. Calcd. for (CuH,,O.Cl)-(CH,CI,),/.: C, 68.36; H, 4.14. Found: C, 68.65; H, 4.29. 1,8-Dichloro-13-(methoxycarbonylhriptzcene (426;) ;a_n_d 1,8-dichloro-16-(methoxycarbonyl)triptycene (26): Reaction of 255 (1.2 g) with 16 by the general procedure above, yielded 26a and 26s, and some unreacted 2. Sublimation of the mixture separated the more volatile anthracene from the triptycenes yielded 26a and 26s (0.90 g, 47%). Separation by semi-preparative HPLC yielded pure 26a and pure 26s as white solids. _2§p: mp 256-258 ‘C; ‘H NMR (CDCl,) 6 7.67 (1H, d, J : 7.92Hz), 7.65 (1H, d, J = 7.92 Hz), 7.36 (2H, d, J = 7.20 Hz), 7.10 (1H, dd, J = 7.56, 7.56 Hz), 7.06 (2H, d, J = 7.20 Hz), 6.95 (2H, dd, J = 7.56, 7.56 Hz), 6.91 (1H, s), 6.45 (1H, s), 3.98 (3H, s); "C NMR (CDCl,) 6 167.20, 147.07, 146.94, 145.14, 141.84, 129.76, 128.51, 127.29, 126.71, 126.11, 125.78, 125.21, 122.84, 52.10, 50.10, 50.42, 47.10; MS (EI) m/e (relative intensity) 384 (14" + 4, 9), 382 (M+ + 2, 51), 380 (M"’, 96), 349 (22), 345 (8), 320 (33), 313 (89), 286 (100), 278 (12), 266 (16), 250 (63), 224 (6), 211 (6), 176 (20), 156 (9), 143 (13), 139 (16), 125 (61), 112 (11), 84 (13); HPLC retention time (CH3C13) 7.8 min, flow rate 4.0 mL/min. Anal. Calcd. for C,,H,.O,Cl,: C, 69.31; H, 3.70. Found: C, 69.30; H, 4.01. _2_§__s: mp 322-323 'C ‘H NMR (CDCl,) 3 7.81 (1H, s), 7.69 (1H, d, J = 7.92 Hz), 7.54 (1H, d, J = 7.20 Hz), 7.27 (2H, d, J = 7.92 Hz), 7.10 (1H, dd, J = 7.56, 7.56 Hz), 7.07 (2H, d, J = 7.92 Hz), 6.95 (2H, dd, J = 7.56, 7.56 Hz), 5.48 (1H, s), 4.08 (3H, s); ”C NMR (0001,) 6 167.22, 94 147.47, 146.60, 144.59, 141.42, 130.56, 127.65, 127.43, 126.71, 126.26, 125.42, 122.85, 121.96, 54.52, 52.29, 43.82; MS (EI) m/e (relative intensity) 384 (M"' + 4, 8), 382 (M" + 2, 44), 380 (M+, 91), 349 (20), 344 (8), 320 (34),. 313 (91), 286 (100), 278 (11), 266 (11), 250 (83), 246 (8), 224 (6), 211 (5), 176 (14), 157 (10), 143 (11), 139 (10), 125 (37), 112 (7); HPLC retention time (CH3013) 8.1 min, flow rate 4.0 mL/min. Anal. Calcd. for C,,H,.O,Cl,: C, 69.31; H, 3.70. Found: C, 69.05; H, 3.78. 1,kDiczano-l3-(methoxycarbonzlnriptzcene (fl): Reaction of 3 (1.1 g) with 16, by the general procedure above, yielded 27s and some unreacted 3. Sublimation of the mixture separated the anthracene from the triptycene yielded 27s (0.69 g, 38%). The compound was further purified by semi-preparative HPLC. This procedure afforded pure 27s as white crystals: mp 366-369 'C; ‘H NMR (CDCl,) 6 7.77 (1H, d, J = 7.92 Hz), 7.73 (1H, d, J = 7.92 Hz), 7.69 (2H, d, J = 7.56 Hz), 7.36 (2H, d, J = 7.56 Hz), 7.17 (1H, dd, J = 7.56, 7.56 Hz), 7.16 (2H, dd, J : 7.56, 7.56 Hz), 7.08 (1H, s), 6.32 (1H, s), 4.00 (3H, s); ”C NMR (CDCl,) 6 161.40, 149.38, 147.21, 145.84, 145.63, 143.63, 129.03, 128.98, 128.61, 127.97, 126.65, 125.95, 116.52, 108.79, 52.25, 50.23, 49.47; MS (EI) m/e (relative intensity) 362 (M13 83), 347 (20), 330 (30), 302 (100), 275 (16), 201 (5), 181 (5), 165 (6), 151 (21), 138 (29), 124 m(35), 11(8), 84 (8), 75 (11); high resolution mass spectrum, exact mass calcd. for C,.H,.N,O, (M"’) 362.1055, found 362.1055. Anal. Calcd. for C,,H,.N,O,: C, 79.55; H, 3.89; N, 7.73. Found: C, 78.77; H, 4.46; N, 7.39. 95 G. Preparation of 1.8.13- and 1,8,16-trisl(N-substituted)carboxamido|- triptycenes General procedure for the premation of 1,8,13- and l 8 16-tris- N-substituted carboxamido tri t cenes: Method 1: The tris(carbonylchlaride) 18a or 18s, (0.23 g, 0.51 mmol) was made in situ and kept under Ar. The appropriate amine (1.5 mmol) in CH3Cl, (20 mL) or CH,CN (30 ml.) was syringed into the reaction flask, followed by Et,N (0.22 mL, 1.5 mmol). Upon addition of the reagents, some reaction mixtures formed solutions, while others remained as slurries. Each reaction mixture was refluxed 18 h under an inert atmosphere, allowed to cool, and the volume reduced to 15 mL under reduced pressure. A white precipitate was collected on a fine glass frit and washed with water, followed by 10% HCl. The products were dried in vacuo overnight at 60 '0. Method 2: The tris(carbonylchlaride) 18a or 18s, (0.23 g, 0.51 mmol) was made in situ and kept under Ar. The appropriate amine (3.6 mmol) in CH3Cl, (50 mL) or CH,CN (80 mL) was syringed into the reaction flask. The reaction mixture was refluxed 24 h under inert atmosphere. The cooled reaction mixture was filtered through a fine glass frit, and the solid collected was washed with water, 10% HCl and 95% EtOH. The products were dried in vacuo overnight at 60 'C. 1,8,13-Tri_s—[(N-4-hydroxyphenzl)carboxamido |triptycene (19s) and 1,8,16- gig-l(N-4-hydroxzphenzl)carboxamido|triptzcene (19a): Reaction of 18s or 18a with p—aminophenol in CH,Cl,, by the general procedure above, yielded pure 19s or 19s as a white solid: (0.32 g, 95%). 195: mp 420-424 'C (dec.); ‘H NMR (Me,so-d.) a 10.18 (38, s), 9.29 96 (3H, s), 7.94 (3H, d, J = 7.20 Hz), 7.50 (1H, s), 7.40 (6H, d, J : 8.64 Hz), 7.29 (3H, d, J = 7.56 Hz), 7.15 (3H, dd, J = 7.56, 7.56 Hz), 6.63 (6H, d, J = 8.28 Hz), 5.93 (1H, s); "C NMR (Me,SO-d.) 6 165.90, 153.99, 146.44, 141.84, 133.43, 129.93, 125.43, 125.10, 124.61, 123.81, 114.56, 52.72, 43.42; MS (EI) m/e (relative intensity) 659 (M137), 551 (59), 442 (8), 282 (4), 250 (40), 109 (100), 86 (17), 80 (79); high resolution mass spectrum, exact mass calcd. for C,,H,,N,O. (M"’) 659.2056, found 659.2042. Anal. Calcd. for C.,H,,N,O,: C, 74.65; H, 4.43; N, 6.37. Found: C, 73.92; H, 4.55; N, 6.38. L95: mp 413-416 'C (dec.); ‘H NMR (Me,SO) 6 10.12 (1H, s), 10.06 (2H, s), 9.30 (1H, s), 9.29 (2H, s), 7.60 (1H, d, J : 8.64 Hz), 7.57 (2H, d, J = 7.56 Hz), 7.49 (6H, d, J = 8.64 Hz), 7.30 (1H, d, J = 7.56 Hz), 7.25 (2H, d, J = 7.56 Hz), 7.13 (1H, dd, J = 6.48, 6.48 Hz), 7.11 (2H, dd, J 7.20, 7.20 Hz), 6.80 (1H, s), 6.78 (2H, d, J = 8.28 Hz), 6.71 (4H, d, J 8.28 Hz), 6.30 (1H, s); IR (KBr) 3368-3270 (8, br, -NH amide), 1652 (s, -C=O), 1620 (111, -NH amide) cm"; MS (EI) m/e (relative intensity) 659 (M"', 5), 551 (2), 250 (2), 149 (3), 109 (100), 91 (2), 80 (98). 1,§,_1_6-Tri§-[(N-propzl)carboxamido|triptzcene (20a): Reaction of 18a with propylamine in CH,Cl,, by the general procedure above, yielded pure 20a as a white solid (0.25 g, 96%): mp 369-402 'C (dec.), ‘H NMR (Measo-d.) 6 8.48 (2H, t, J = 5.40 Hz), 8.33 (1H, t, J = 5.40 Hz), 7.48 (2H, d, J = 6.84 Hz), 7.38 (1H, d, J = 7.20 Hz), 7.13 (3H, d, J : 7.56 Hz), 7.06 (3H, dd, J = 7.20, 7.20 Hz), 6.63 (1H, s), 6.25 (111, s), 3.32 (6H, m), 1.62 (6H, m), 0.97 (68, t, J = 7.20 Hz), 0.95 (3H, t, J = 7.20 Hz); 1’C NMR (Measo-d.) 6 167.46, 145.73, 145.27, 143.90, 142.39, 133.11, 132.55, 125.46, 125.34, 124.85, 124.68, 124.10, 123.57, 49.85, 46.72, 41.00, 40.74, 22.41, 22.35, 11.51, 11.43; MS (EI) m/e (relative intensity) 509 (MI, 24), 451 97 (14), 424 (16), 394 (18), 365 (10), 279 (10), 250 (13), 168 (20), 125 (15), 100 (8), 86 (29), 73(100); high resolution mass spectrum, exact mass calcd. for CgangsNgog (M+) 509.2678, found 509.2680. Anal. Calcd. for C,,H,,N,O,: C, 75.41; H, 6.92; N, 8.24. Found: C, 74.66; H, 6.80; N, 8.08. 1.131.16-TrisL-l (N-4-methzlphenzl)carboxamido|triptzcene (21;): Reaction of 18a with p—toluidine in CH,CN, by the above procedure, yielded pure 21a as a white solid (0.32 g, 97%): mp 371-374 'C (dec.); ‘H NMR (Measo—d.) 6 10.28 (38, s), 10.17 (3H, s), 7.72 (2H, d, J : 7.92 Hz), 7.59 (2H, d, J = 6.48 Hz), 7.57 (4H, d, J = 7.92 Hz), 7.52 (1H, d, J = 7.20 Hz), 7.33 (18, d, J = 7.56 Hz), 7.27 (2H, d, J = 9.00 Hz), 6.78 (1H, s), 6.29 (1H, s), 3.35 (6H,s), 2.31 (3H,s); ”C NMR (Measo-d.) 6 166.19, 166.09, 145.89, 145.36, 144.10, 143.27, 136.72, 136.49, 133.28, 132.59, 132.42, 128.99, 128.73, 126.31, 125.60, 124.81, 124.04, 123.80, 120.52, 119.98, 49.86, 46.68, 20.55, 20.52; MS (EI) m/e (relative intensity) 653 (M"', 10), 547 (81), 413 (10), 394 (5), 384 (10), 366 (5), 355 (7), 341 (6), 307 (5), 278 (5), 250 (38), 239 (5), 221 (15), 207 (11), 198 (8), 184 (10), 176 (22), 171 (20), 149 (10), 133 (5), 125 (19), 106 (100), 91 (37), 86 (46), 79 (51), 64 (23); high resolution mass spectrum, exact mass calcd. for C,,H,,N,O, (M+) 653.2678, found 653.2681. Anal. Calcd. for (C..H,,N,O,)-(CH,CI,),/,: C, 78.74; H, 5.30; N, 6.23. Found: C, 79.07; H, 5.37; N, 6.21. 1 8 16-Tris- N-benzene- S-x lene carboxamido tri t cene 1%)! Reaction of 18a with p—aminobenzene p-xylene sulfide93 in CH3C13, by the general procedure above, yielded 85% of 22s as a white solid.“ mp 300-304 'C (dec. ; 18 NMR (Me,so-d.) a 10.40 (18, s), 10.30 (28, s), 7.77 (2H, d, J = 8.28 Hz), 7.60 (2H, dd, J = 6.84, 6.84 Hz), 7.53 (1H, d, J : 7.20 Hz), 7.36 (4H, d, J = 8.28 Hz), 7.31 (1H, d, J : 8.28 Hz), 7.20 98 (1211, m), 7.08 (11H, m), 6.76 (1H,s), 6.29 (1H, s), 4.20-4.15 (6H, br, s), 2.27 (3H, s), 2.25 (3H, s); MS data not available because M"' > 1000 m/e. H. Reaction of |Pr.m, Fe._S_._Ql. with H0 -tri d li and In a dry box, 19s (0.15 g, 0.023 mmol) and [Pr.N],[Fe.S.(SEt).] (0.022 g, 0.023 mmol) were weighed out and placed in a side-arm flask equipped with a serum stopper. Measo-d. (1.5 mL) was syringed into the reaction vessel, and the resultant red-brown solution was stirred for 4h, with periodic evacuation of the flask to remove volatile EtSH. An aliquot of solution was transferred to a 5 mm NMR tube via cannulla, and the tube was sealed under Ar. The sample was frozen until the ‘H NMR could be obtained. A second sample was prepared as outlined above, but allowed to stir for 12 h under dynamic vacuum. After ~4 h, all solvent had evaporated, and another 1.5 ml. aliquot of us,so-d. was added to the flash. This procedure was repeated one more time before the sample was sealed in an NMR tube. 1. Reaction of mum, Fe._S_,C_l.| with Na,(Q,-trigd) 1. Premration of Na,_(Q,-trimd) A stock solution of NaOMe (0.435 M) was prepared by dissolving Na (1.0 g, 43.5 mmol) in MeOH (100 mL). Compound 198 (0.010 g, 0.015 mmol) was placed in a side-arm flask under Ar and dissolved in DMA (2 mL). NaOMe in MeOH (0.10 mL) was syringed into the solution, 99 causing an immediate color change to a bright yellow-green. After 6 h, the solvents were removed in vacuo to give a yellow solid of Na,(O,-tripod) (0.011 g, 98%): ’H NMR (Measo-dg) 6 9.87 (3H, s), 7.69 (1H, s), 7.56 (3H, m), 7.25 (9H, m), 7.13 (3H, m), 6.31 (6H, s), 5.81 (1H, s). 2. Reaction of chloro tetramer with the trisodium salt of ‘HO!3‘trimd The Na,(O,-tripod) (0.011 g, 0.015 mmol) isolated by the procedure outlined above, was washed twice with CH,CN (1 mL) under Ar, and dried overnight to remove excess solvent. A stock solution of iron-sulfur cluster (1.38 - 10'2 M) was prepared by dissolving [Et.N],[Fe.S.Cl.] (0.05 g) in Mezso-d. (4.8 mL). Various amounts of iron-sulfur cluster solution were syringed into the ‘ flask containing the sodium salt of the ligand. Five reactions were done and the amount of cluster used in each was 0.020, 0.015, 0.011, 0.010 and 0.008 mmol, respectively. The solution was stirred at room temperature for 8 h before an aliquot was removed and sealed in an NMR tube under Ar. J. PregLation of |Et.N_1,f§,MoS,Fe(OAC),1 Tetraethylammonium tetrathiomolybdate (1.0 g, 2.0 mmol) was placed in a side-arm flask and flushed with Ar. CH3CN (80 mL) was added to dissolve the solid. Ferrous acetate (0.39 g, 2.3 mmol) was transferred in the dry box to a side-arm flask equipped with a serum stopper. CH,CN (20 mL) was syringed into the flask containing the Fe(OAc) 2 to 100 form a slurry. The orange tetrathiomolybdate solution was cannulled into the white Fe(OAc), slurry; this resulted in an immediate color change to wine-red. The solution was stirred at room temperature for 2 h and then filtered anaerobically through a fine glass frit. The filtrate volume was reduced to 30 mL under vacuum. Diethyl ether (50 ml.) was added to precipitate a red-brown solid of '[Et.N].[S.MoS,Fe(OAc),] (1.0 g, 74 %): Optical spectrum Amax (3(M" cm“)) 286 (7795), 310 (10000), 428 (sh), 464 (5270), 508 (sh). Anal. Calcd. for CaoH“N,O.S.FeMo: C, 36.47; H, 7.04; N, 4.25. Found: C, 35073; H, 6082; N, 4.00. IV. CONCLUSIONS A macrocyclic tridentate ligand ((HO),-tripod) based on a symmetrically trisubstituted triptycene with terminal phenoxide substituents has been designed and prepared. Interactions of the ligand with a 4Fe-4S cluster were examined in order to investigate the possibility of the presence of oxygen-ligated iron-sulfur clusters in nitrogenase. En route to preparing this ligand, twenty-one new trisubstituted triptycenes were isolated and characterized by 1H NMR, ”C NMR and IR spectroscopies, as well as by mass spectrometry. Fifteen of the trisubstituted triptycenes were synthesized from the Diels-Alder cycloadditon of a 1,8-disubstituted anthracene with an ortho-substituted benzyne. The other six trisubstituted compounds were prepared from conversion reactions of preformed triptycenes. The syntheses of the starting materials 1,8-dicyanoanthracene, 1,8-dicyanoanthraquinone and anthracene 1,8-dicarboxylic acid were improved from that of literature procedures. A high yield synthesis of a synthetically useful ortho-substituted benzyne precursor, 2-amino-6-(methoxycarbonyl)benzoic acid, was also developed. The triptycenes were isolated as mixtures of syn (1,8,13) and anti (1,8,16) trisubstituted triptycenes. These were purified by sublimation and separated by HPLC to afford pure isomeric products. The (ratio of syn to anti isomers obtained was found to depend on the electron- donating or withdrawing ability of the substituents on the benzyne and anthracene units. 101 102 Complexation of the tridentate ligand to an F6434 cubane core was investigated by ‘H NMR spectroscopy. Two methods of ligand exchange reactions were used: (1) the (HO),-tripod ligand was reacted with [Fe.S.(SEt).]"‘ and (2) the trisodium salt of the (HO),-tripod ligand was reacted with (Fe.S.Cl.]". Proton magnetic resonance spectra show isotropically shifted peaks that are mainly contact in origin and occur through a s delocalization mechanism. The paramagnetically shifted resonances were assigned by comparison of chemical shifts with those reported for previously known iron-sulfur clusters. Evidence of free ligand was observed in the spectra, suggesting that in highly-coordinating solvents the (HO),-tripod ligand is labile and is being displaced by solvent. Analysis of the proton resonances indicated that a mixture of tetrameric (Fe.S.) and hexameric (Fe.S.) forms of the ligated cluster are generated in solution. The ratio of tetrameric species to hexameric species was altered by changing the ratio of ligand to cluster in the NMR samples. The (HO),-tripod ligand binds to form a mixed-ligand [Fe.S.(O,-tripod)(L)]"' (L = SEt, Cl) cluster, but rapidly rearranges to hexameric [Fe.S‘(O,-tripod),]"' and [Fe.S.(O,-tripod)(L),]" (L : SEt, Cl) complexes. The results presented in this dissertation show that under the reaction conditions used, oxygen ligation stabilizes the hexameric Fe-S species. In addition to investigating a mixed—ligand cluster as a possible model for P-clusters, an oxygen-ligated model of the FeMo-cofaCtor was prepared. A linear Mo-Fe-S cluster with acetate ligands, [Et.N],[S,MoS,Fe(OAC)-.-l. was synthesized by reacting [Et.N],[MoS.] with Fe(0Ac),. In solution the oxygen ligands are labile, and 103 conversion to [Fe(MoS.),]” was observed. The optical spectrum of the acetate-ligated dimer exhibits characteristic absorbances of an FengoS, care. The ‘H NMR spectrum shows a paramagnetically shifted methyl resonance, and variable temperature studies indicate Curie Law magnetic behavior. Magnetic susceptibility measurements shows simple paramagnetism that follows the Curie Law, in agreement with the NMR data. An effective moment of 4.96 up was observed and is consistent with an S = 2 system, and indicates an Fe(II)-Mo(VI) couple. The magnetic properties of this cluster show no major differences from that of the chloro- or phenoxide-ligated dimers. The FesaMoS, unit may resemble a structural portion of the cofactor, although a simple linear array does not agree with preliminary physical data obtained on the enzyme“. The study of this compound has served to extend the general knowledge of iron-molybdenum-sulfur clusters of this type. REFERENCES l. 2. 3. 5. 6. 8. 9. REFERENCES Hardy, R. W. F.; Burns, R. C. "Iron-Sulfur Proteins"; Lovenberg, W., Ed.; Academic Press: New York, 1973; Vol. 1, pp 65-110. Holm. R. H.; Ibers, J. A. "Iron-Sulfur Proteins"; Lovenberg, W., Ed.; Academic Press: New York, 1977; Vol. 3, pp 206-281. Orme-Johnson, W. H.; Davis L. C.; Henzl, M. T.; Averill, B. A.; Orme-Johnson, N. R.; Munck, E.; Zimmerman, R. "Recent Develop- ments in Nitrogen Fixation"; Newton, W. E., Postgate, J., Rodriquez-Barrueco, J., Eds.; Academic Press: New York, 1977; pp 131-178,. Orme-Johnson, W. H.; Davis, L. C. "Iron-Sulfur Proteins"; Lovenberg, W., Ed.; Academic Press: New York, 1977; Vol. 3, pp 15-60. Shah, V. K.; Brill, W. J. Proc. Natl. Acad. Sci. USA 1977, 74, 5468. Wang, G. B.; Kurtz, D. M.; Holm, R. H.; Mortensan, L. 8.; Upchurch, R. G. J. Am. Chem. Soc. 1979, 101, 3078. Zimmerman, R.; Munck, 8.; Brill, W.; Shah, V.; Henzl, M.; Rawlings, J.; Orme-Johnson, W. H. Biochim. Biophys. Acta 1978, 537, 185. Averill, B. A. Structure and Bonding 1983, 53, 61. Orme-Johnson. W. H.; Lindahl, P. A.; Nelson, M. J. "Advances in Inorganic Biochemistry"; Eichhorn, G., Marzilli, L., Eds.; Elseveir Biomedical: New York, 1981; Vol. 4, pp 1-35. 104 10. 11. 12. 13. 14. 15. 16. 17. 18. 105 Lehninger, A. L. "Biochemistry"; Worth: New York, 1975; pp 720-725. Coucouvanis, D. Acc. Chem. Res. 1981, 14, 201. Cramer, S. P.; Hodgson, K. 0.; Gillum, W. 0.; Mortensan, L. E.; J. Am. Chem. Soc. 1978, 100, 3398. Cramer, S. P.; Gillum, W. 0.; Hodgson, K. 0.; Mortensan, L. E.; Stiefel, E. I.; Chisnell, J. R.; Brill. W. J.; Shah, V. K. J. Am. Chem. Soc. 1978, 100, 3814. Antonio, M. R.; Tea, B.-K.; Orme-Johnson, W. H.; Nelson, M. J.; Groh, S.; Lindahl, P. A.; Kauzlarich, S. M.; Averill, B. A. J. Am. Chem. Soc. 1982, 104, 4703. 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. Biochem. Biophys. Acts 1975, 400, 32. a.) Huynh, B. H.; Henzl, M. T.; Christner, J. A.; Zimmerman, R.; Orme—Johnson, W. H.; Munck, E. Biochim. Biophys. Acts 1980, 623, 124. b.) Johnson, M. K.; Thomson, A. J.; Robinson, A. E.; Smith, 8. E. Biachim. Biophys. Acts 1981, 671, 61. Orme-Johnson, W. H.; Lindahl, P. A.; Meade, J.; Warren, W.; . Nelson, M.; Groh, S.; Orme-Johnson, N. R.; Munck, E.; Hunyh, B.; Emptage, M.; Rawlings; Smith, J.; Roberts, J.; Hoffmann, B.; Mims, W. B. J. "Current Perspectives in Nitrogen Fixation"; Gibson, A. H., Newton, W. E., Eds.; Australian Academy of Science: Canberra, 1981; pp 79-83. 19. 20. 21. 22. 23. 24. 25. 26. 27. 106 a.) Tieckelmann, R. H.; Silvie, H. C.; Kent, T. A.; Huynh, B. H.; Waszczak, J. V.; Tea, B.-K.; Averill, B. A. J. Am. Chem. Soc. 1980, 102, 5550. b.) Coucouvanis, D.; Simhon, E. 0.; Swenson, D.; Baenziger, N. C. J. Chem Soc. Chem. Comm. 1979, 361. c.) Coucouvanis, D.; Stremple, P.; Simhon, E. D.; Stremple, P. P. J. Am. Chem. Soc. 1983, 105, 5767. a.) Muller, A.; Tolls, H,-G.; Bogge, H. Z. Anorg. Allg. Chem. 1980, 471, 115. b.) Muller, A.; Jostes, R.; Tolle, H.-G.; Trautwein, A.; Bill, E. Inorg. Chim. Acts 1980, 46, L121. a.) Silvie, H. C.; Averill, B. A. Inorg. Chim. Acta, 1981, 54, L57. b.) Tea. B.-K.; Antonio, M. R.; Tieckelmann, R. H.; Silvie, H. C.; Averill, B. A. J. Am. Chem. Soc. 1982, 104, 6126. Zhuang, B.; McDonald, J. W.; Newton, W. E. Inorg. Chim. Acta 1983, 77, L221. Coucouvanis, D.; Simhon, E. D.; Stremple, P.; Baenziger, N. C. Inorg. Chim. Acts 1981, 53, L135. Coucouvanis, D.; Simhon, E. D.; Baenziger, N. C.; Stremple, P.; Swenson, D.; Kostokas, A.; Simopulos, A.; Petrouleas, V.; Papaefthymiou, V. J. Am. Chem. Soc. 1980, 102, 1730. Garner, C. D.; Acott, S. R.; Christou, G.; Collison, D.; Mabbs, F. E.; Petrouleas, V.; Pickett, C. J. "Nitrogen Fixation: The Chemical-Biochemical-Genetic Interface"; Muller, A., Newton, W. E., Eds., Plenum Press: New York, 1983; pp 245-273. Coucouvanis, D.; Baenziger, N. C.; Simhon, E. D.; Stremple, P.; Swenson, D.; Somopoulos, A.; Kostikas A.; Petrouleas, V.; Papaefthymiou, V. J. Am. Chem. Soc. 1980, 102, 1732. Tieckelmann, R. H.; Averill, B. A. Inorg. Chim. Acta 1980, 46, (L35. 28. 29. 30. 31. 32. 33. 34. 35. 107 a.) Coucouvanis, D.; Simhon, E. D.; Baenziger, N. C. J. Am. Chem. Soc. 1980, 102, 6644. b.) McDonald, J. W.; Freisen, G. D.; Newton, W. E. Inorg. Chim. Acts 1980, 46, L79. c.) Friesen, G. D.; McDonald, J. W.; Newton, W. E.; Euler, W. B.; Hoffmann, B. M. Inorg. Chem., 1983, 22, 2202. - Dahlstrom, D. L.; Kumar, S.; Zubieta, J. J. Chem. Soc. Chem. Comm. 1981, 411. Base, K. S.; Lamberty, P. E.; Kovacs, J. A.; Sinn, E.; Averill, B. A. Polyhedron 1986, in press. a.) Wolff, T. E.; Berg, J. M.; Warrick, C.; Hodgson, K. 0.; Holm, R. H.; Frankel, R. B. J. Am. Chem. Soc. 1978, 100, 4630. b.) Wolff, T. E.; Berg. J. M.; Hodgson, K. 0.; Frankel, R. B.; Holm, R. H. J. Am. Chem. Soc. 1979, 101, 4140. a.) Christou, G.; Garner, C. D.; Mabbs, F. E.; King, T. J. J. Chem. Soc. Chem. Comm. 1978, 740. b.) Christou, G.; Garner, C. D.; Mabbs, F. E.; Drew, M. G. B. J. Chem. Soc. Chem. Comm. 1979, 91. c.) Acott, S. R.; Christou, G.; Garner, C. D.; King, . T. H.; Mabbs, F. E.; Miller, R. M. Inorg. Chim. Acta 1979, 35, L337. d.) Christou, G.; Garner, C. D.; Miller, R. M.; Johnson, C. E.; Rush, J. D. J. Chem. Soc. Dalton Trans. 1980, 2364. Cleland, W. E.; Averill, B. A. Inorg. Chim. Acta 1985, 107, 181. Wolff, T. E.; Berg. J.M.; Power, P. P.; Hodgson, K. 0.; Holm, R. H.; Frankel, R. B. J. Chem. Soc. 1979, 101, 5454. a.) Christou, G.; Garner, C. D. J. Chem. Soc. Dalton Trans. 1980, 2354. b.) Christou, G.; Mascharak, P. K.; Armstrong, W. H.; Papaefthymiou, G. C.; Frankel, R. B.; Holm, R. H. J. Am. Chem. Soc. 1982, 104, 2820. 36. 37. 38. 39. 40. 41. 42. 43. 44. 108 Armstrong, W. H.; Holm, R. H. J. Am. Chem. Soc. 1981, 103, 6246. Armstrong, W. H.; Mascharak, P. K.; Holm, R. H. J. Am. Chem. Soc. 1982, 104, 4373. a.) Palermo, R. E.; Singh, R.; Baskin, J. K.; Holm, R. H. J. Am. Chem. Soc. 1984, 106, 2600. b.) Palermo, R. E.; Holm, R. H. J. Am. Chem. Soc. 1983, 105, 4310. c.) Mascharak, P. K.; Armstrong, W. H.; Mizobe, Y.; Holm, R. H. J. Am. Chem. Soc. 1983, 105, 475. Coucouvanis, D.; Kanatzidis, M. G. J. Am. Chem. Soc. 1986, 108, 337. a.) Laskowski, E. J.; Frankel, R. B.; Gillum, W. 0.; Papaefthymiou, G. C.; Renaud, J.; Ibers, J. A.; Holm, R. H. J. Am. Chem. Soc. 1978, 100, 5322. b.) Laskowski, E. J.; Reynolds, J. G.; Frankel, R. B.; Foner, 8.; Papaefthymiou, G. C.; Holm, R. H. J. Am. Chem. Soc. 1979, 101, 6562. c.) Stepan, D. W.; Papaefthymiou, G. C.; Frankel, R. B.; Holm, R. H. Inorg. Chem. 1983, 22, 1150. Berg, J. M.; Hodgson, K. 0.; Holm, R. H. J. Am. Chem. Soc. 1979, 101, 4586. Hagen, K. 8.; Watson, A. D.; Holm, R. H. Inorg. Chem. 1984, 23, 2984. Cotton, F. A.; Wilkinson, G. "Advanced Inorganic Chemistry: A Comprehensive Text"; John Wiley and Sons: New York 1980; pp 759. Coucouvanis, D.; Kanatzidis, M. G.; Simhon, E.; Baenziger, N. C. J. Am. Chem. Soc. 1985, 107, 1874. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 109 Kanatzidis, M. G.; Baenziger, N. C.; Coucouvanis, D.; Simopoulos, A.; Kostikas, A. J. Am. Chem. Soc. 1984, 106, 4500. Johnson, R. W.; Holm, R. H. J. Am. Chem. Soc. 1978, 100, 5338. a.) Cleland, W. E.; Averill, B. A. Inorg. Chim. Acts 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. Kanatzidis, M. G.; Salifoglou, A.; Coucouvanis, D. J. Am. Chem. Soc. 1985, 107, 3358. Kanatzidis, M. G.; Coucouvanis, D.; Simopoulos, A.; Kostikas, A.; Papaefthymiou, V. J. Am. Chem. Soc. 1985, 107, 4925. Kanatzidis, M. G.; Ryan, M.; Coucouvanis, D.; Simopoulos, A.; Kostikas, A. Inorg. Chem. 1983, 22, 179. Johnson, R. E.; Papaefthymiou, G. C.; Frankel, R. B.; Holm, R. H. J. Am. Chem. Soc. 1983, 105, 7280. Tieckelmann, R. H. "Possible Synthetic Pathways to Models of Metal Cofactors of Nitrogenase"; Michigan State University; Masters Thesis 1978. Lamberty, P. E. unpublished results. Friedman, L.; Logulla, F. M. J. Org. Chem. 1969, 34, 3089. House, H. 0.; Koepsell, D.; Jaeger, W. J. Org. Chem 1973, 38, 1167. Akiyama, S.; Misumi, 8.; Nakagawa, M. Bull. Chem. Soc. Japan 1960, 33, 1293. Skvarchenko, V. R.; Shalaev, V. K.; Klabunovski, E. I. Russ. Chem. Rev. 1978, 43, 951. Mari, 1.; Kodoaka, T.; Yoshiteri, S.; Misumi, 8. Bull. Chem Soc. Japan 1971, 44, 1649. Fuson, R. C.; Rabjohn, N. Org. Synth. 1955, 3, 557. 60. 61. 62. 63. 65. 66. 67. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 110 Berger, G.; Olivier, S. C. J. Rev. Trav. Chim. 1927, 46, 600. Sharai, H.; Yashiro, T.; Sato, T. Chem. Pharm. Bull. 1969, 17, 1564. Trahanovsky, W. 8.; Young, L. B. J. Org. Chem. 1966, 31, 2033. Holum, J. R. J. Org. Chem. 1961, 26, 4814. Sela, T.; Sargent, M. V. J. Chem. Soc. Chem. Comm. 1978, 253. Bartlett, P. D.; Ryan, M. J.; Cohen, S. G. J. Am. Chem. Soc. 1942, 64, 2649. Pearson, R. E.; Martin, J. C. J. Am. Chem. Soc. 1963, 85, 3142. Carey, E. J. Pure Appl. Chem. 1967, 14, 19. Waldmann, H.; Oblath, A. Chem. Ber. 1938, 71, 366. Akiyama, 8.; Misumi, S.; Nakagawa, M. Bull. Chem. Soc. Japan 1962, 35, 1829. Nagai, U.; Abe, E.; Sana, R. Tetrahedron 1974, 30, 25. "Handbook of Chemistry and Physics"; Weast, R. C., Ed.; CRC Press: Clevland, Ohio, 1975; 56th ed., p D-151. Jencks, W. P.; Salvesen, K. J. Am. Chem. Soc. 1971, 93, 4433. Piper, J. R.; Stevens, F. J. J. Org. Chem. 1962, 27, 3134. Sauer, J.; Sustmann, R. Angew. Chem. Int. Ed. Engl. 1980, 19, 779. Exner, 0. "Correlation Analysis in Chemistry"; Chapman, N. B., Shorter, J., Eds.; Plenum Press: New York, 1978; pp 439 - 540. Bromilow, J.; Brownlee, R. T. C.; Lopez, V. 0.; Taft, T. W. J. Org. Chem. 1979, 44, 4766. Topsom, R. D. Acc. Chem. Res. 1983, 16, 292. Kidd, K. J.; Kotowycz, G.; Schaeffer, T. Can. J. Chem. 1967, 45, 2155. 79. 80. 81. 82. 83. 84. 85. 86. 87. 88. 89. 111 Silverstein, R. M., Bassler, G. C., Morill, T. C. "Spectroscopic Identification of Organic Compounds", 4th ed.; John Wiley and Sons: New York, 1981; pp 264-265. Que, Jr., L.; Bobrik, M. A.; Ibers, T. A.; Holm, R. H. J. Am. Chem. Soc. 1974, 96, 4168. Dukes, G. R.; Holm, R. H. J. Am. Chem. Soc. 1975, 97, 528. Cleland, W. E. "The Synthesis, Characterization and Reactivity of Iron-Sulfur and Molybdenum-Iron-Sulfur Complexes with Phenoxide Terminal Ligands"; Michigan State University; Ph. D. Dissertation 1984. Drago, R. S. "Physical Methods in Chemistry"; W. B. Saunders: Philadelphia, 1977; pp 436 - 463. Holm, R. H.; Phillips, W. D.; Averill, B. A.; Mayerle, J. J.; Herskovitz, T. J. Am. Chem. Soc. 1974, 96, 2109. Horrocks, Jr., W. DeW. "NMR of Paramagnetic Molecules. Principles and Applications"; LaMar, G. N., Horrocks, Jr., W. DeW., Holm, R. H., Eds.; Academic Press: New York, 1973; pp 127 - 175. Cleland, Jr., W. E.; Averill, B. A. Inorg. Chim. Acta 1985, 106, L17. Kanatzidis, M. G.; Hagen, W. R.; Dunham, N. R.; Lester, R. K.; Coucouvanis, D. J. Am. Chem. Soc. 1985, 107, 953. Bobrick, M. A.; Hodgson, K. 0.; Holm, R. H. Inorg. Chem. 1977 16, 1851. Diemann, E.; Muller, A. Coord. Chem. Rev. 1973, 10, 79. 90. 91. 92. 93. 94. 95. 112 Tieckelmann, R. H. "The Synthesis and Characterization of the Molybdenum-Iron-Sulfur Clusters [S ,MoS ,FeLala", [SaMosaFesaFeLzP‘ and [SzMongesaFeszMoSfl‘w ; Michigan State University; Ph. D. Dissertation 1981. Silvis, H. C. "The Synthesis, Characterization and Reactivity of Molybdenum and Tungsten-Iron-Sulfur Complexes"; Michigan State University; Ph. D. Dissertation 1981. Wang, G. B.; Bobrick, M. A.; Holm, R. H. Inorg. Chem. 1978, 17, 1093. Frankel, M.; Gertner, D.; Jacobson, H.; Zilkha, A. J. Chem. Soc. 1960, 1390. Hardt, V. H. D.; Moller, W. Z. Anorg. Allg. Chem. 1961, 313, 57. Golden, R.; Stock, L. M. J. Am. Chem. Soc. 1972, 94, 3080. ARiES H .1 11 “1111111111111111111111 3168 8 3 4 8 .As— —- .4 “' . ..~‘ ‘e- r. ‘ I 1. ‘ . >4 . . s 8.5,, s .. . . 3,. 9: 44 WW