Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Towards the synthesis of transition metal derivatives of carbohydrates Slee, Arlene Marie 1973

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-UBC_1973_A6_7 S54_5.pdf [ 5.65MB ]
JSON: 831-1.0061837.json
JSON-LD: 831-1.0061837-ld.json
RDF/XML (Pretty): 831-1.0061837-rdf.xml
RDF/JSON: 831-1.0061837-rdf.json
Turtle: 831-1.0061837-turtle.txt
N-Triples: 831-1.0061837-rdf-ntriples.txt
Original Record: 831-1.0061837-source.json
Full Text

Full Text

i . TOWARDS THE SYNTHESIS OP TRANSITION METAL DERIVATIVES OF CARBOHYDRATES by ARLENE M. SLEE B.Sc. (Hon)., University of Sydney, 1 9 6 9 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE 'DEGREE OF MASTER OF SCIENCE in the Department of CHEMISTRY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA January 1 9 7 3 In presenting t h i s thesis i n p a r t i a l f u l f i l m e n t of the requirements for an advanced degree at the University of B r i t i s h Columbia, I agree that the Library s h a l l make i t f r e e l y available for reference and study. I further agree that permission for extensive copying of t h i s thesis for scholarly purposes may be granted by the Head of my Department or by h i s representatives. It i s understood that copying or publication of t h i s thesis for f i n a n c i a l gain s h a l l not be allowed without my written permission. Department of The University of B r i t i s h Columbia Vancouver 8, Canada Date ABSTRACT The synthesis of organometallic derivatives of carbohydrates was approached in two different ways. The f i r s t method was nucleophilic attack by vitamin B , cyclopentadienyl I d. s iron dicarbonyl and cyclopentadienyl tungsten and molybdenum tricarbonyl anions on the carbohydrate sulphonate in an attempt to synthesise direct metal linkage to the carbohydrate via a carbon metal bond. The reaction was found to be most successful when applied to primary carbohydrate sulphonates. L i t t l e success was encountered with secondary tosylates arid epoxides. The main problem enountered here was the d i f f i c u l t y in isolating the organometallic derivatives. Another series of compounds were synthesised in an attempt to extend this study of metal derivatives of carbohydrates, to metal-carbohydrate derivatives bonded through a heteroatom. Model systems were studied of six-membered thio complexes of tungsten, manganese and platinum. From the tungsten and manganese complexes i t was found that the tungsten complex was 'locked' only by a.Jb-butyl system, whereas the manganese complexes appeared to be 'locked' even without the _t-butyl group. The six-membered rings in each case appeared to be puckered chairs. The configuration about the sulphur was not determined, although i t appears to be either the diequatorial configuration, or a fast equilibrium between the axial^equatorial^ diequatorial^ equatorial-axial configurations. Studies on the platinum thio complexes indicate that for a six-membered ring, the average coupling constant of Pt-S-C-H was 5 8 Hz, while the trans coupling of platinum through sulphur was found to be at least 7 3 Hz, with J + J , = 120 Hz. trans gauche i i i TABLE OF CONTENTS GENERAL INTRODUCTION . . . . . . 1 PART ONE INTRODUCTION 6 RESULTS AND DISCUSSION 14 Section A: Reaction with vitamin B. „ . 15 12s • Section B: Reactions with cyclopentadienyl iron . 19 dicarbonyl anion, cyclopentadienyl molybdenum . tricarbonyl anion and cyclopentadienyl tungsten tricarbonyl 1) . Reaction of primary sugar tosylates 2) . Reaction of secondary tosylate and epoxide PART TWO INTRODUCTION 3 7 RESULTS AND DISCUSSION 41 Section A: Preparation of thio ligands 4 2 Section B; Tetracarbonyl complexes of Group V I metals 4 5 Section C: Manganese complexes 6 2 Section D: Platinum complexes 71 GENERAL CONCLUSION 7 8 EXPERIMENTAL . . . . . 81 GENERAL METHODS 81 PART ONE 8 4 PART TWO 1 1 0 REFERENCES 1 2 9 IV. LIST OF TABLES TABLE 1 Electronic absorption spectra of some cobalamins 18 TABLE 2 Chemical shift values for metal derivatives of 25 6-deoxy-1,2;3,5-di-0-methylene-*-D-glucofuranose TABLE 3 Coupling constants for metal derivatives of 26 6-deoxy-1,2;3,5-di-0-methylene-* -D-glucofuranose TABLE 4 Chemical shift values for derivatives of 31 1 ,2-0-isopropylidene-3»5 benzylidene- <x -D-glucofuranose TABLE 5 Coupling constants for derivatives of 32 1 ,2-0-isopropylidene-3,5 benzylidene- c*-D-glucofuranose TABLE 6 Analytical data and carbonyl infra-red spectra 46 of Group V1 metal-sulphur complexes TABLE 7 N.m.r. parameters of Group V1 metal-sulphur complexes 49 TABLE 8 Analytical data and carbonyl infra-red spectra of 64 manganese-sulphur complexes TABLE .9 N.m.r. parameters of manganese-sulphur complexes 65 TABLE 10 N.m.r. parameters of platinum complexes 73 V. LIST OP FIGURES FIGURE 1 Electronic absorption spectra of a) vitamin B._ ' 1 2 b) (6-deoxy-1,2;3,5-di-0-methylene-c<-D-glucof uranose )-6-C_-cobalamin c) (6-deoxy-1,2; 3,5-di-0-methylene-<*-D-glucofuranose)-6-C_-cobalamin in .05 M hydrochloric acid. FIGURE 2 Infra-red spectra of a) iron dimer b) (6-deoxy-1,2;3,5-di-0-methylene-<x-D-glucofuranose )-6-C_-(cyclopentadienyl iron dicarbonyl) FIGURE 3 A) Normal H n.m.r. spectrum of (6-deoxy-1,2;3»5-di-0-methylene- o< -D-glucofuranose)-6-£-(cyclopentadienyl iron dicarbonyl) B) Computer simulation of the normal spectrum 1 FIGURE A Partial H n.m.r. spectrum of 3,5-0-benzylidene-6-deoxy-1,2-0-isopropylidene- oc-D-. glucofuranose-5-ene FIGURE 5 A) Partial H n.m.r. spectrum of ( c o ) 4 w S(CH 3)CH 2CH[C(CH 3) 5]CD 2S(CH 3) with irradiation of the deuterium ... . B) Computer simulation of the normal spectrum FIGURE 6 ' Possible conformations of 17 FIGURE (C0) 4 W S(CH5)CH2CH[C(CH3);5]CD2S(CH3) and ( c o ) 4 w S(<())CH2CH[C(CH3)3]CD2S (j) 7 A) Partial H n.m.r. spectrum of (CO) ¥ S((j))CH2CH2CD2s((j)) B) Above spectrum with irradiation of deuterium C) Computer simulation of the normal spectrum FIGURE 8 Possible conformations of (C0) 4 ¥ S(CH5)CH2CH2CD2S(CH5) (CO) ¥ S((]))CH2CH2CD2S (j) and (C0) 4 ¥ S(CH2(|))CH2CH2CD2S(CH2(|) 23 24 30 48 52 57 60 v i . •t FIGURE 9 A) Partial H n.m.r. spectrum of 66 Br(C0)„ Mn S(CH,)CH_CH0CD0S(CH_) B) Above spectrum with irradiation of the deuterium C) Computer simulation of the normal spectrum 1 FIGURE 10 A) Partial H n.m.r. spectrum of Br(C0) 3 Mn S(CH^)CH2CH[c(CH^) ]CD2S(CH^) 70 B) Computer simulation of the normal spectrum FIGURE 11 Mass spectrum of (6-deoxy-1,2;3,5-di-0-methylene- 94 -D-glucofuranose)-6-C_-(cyclopentadienyl iron carbonyl) v i i . LIST OF FLOW SHEETS FLOW SHEET 1 Preparation of ligands R S CH^CH^D^ R 111 FLOW SHEET 2 Preparation of ligands R S CH 2CH[c(CH 3) 5]CD 2S R 115 FLOW SHEET 3 Primary carbohydrate tosylates 124 FLOW SHEET 4 Secondary tosylates and epoxides 125 FLOW SHEET 5 Ligands 126 FLOW SHEET 6 Manganese and Group V1 complexes 127 FLOW SHEET 7 Platinum complexes 128 v i i i . ACKNOWLEDGEMENT I am very grateful to Dr. L. D. Hall for his continual encouragement and help throughout the course of this work. I would also like to thank Dr. P. Legzdins and Mr. A. Crease for their advice concerning the organo-metallic syntheses and Mr. R. B. Malcolm for his assistance regarding the conformational and n.m.r. problems. 1 . GENERAL INTRODUCTION 2. General interest in both organometallic and biological compounds has increased significantly over the past twenty years. The f i r s t (1,2) naturally occurring transition metal organic complex to be recognised was the vitamin coenzyme which has a very stable cobalt-carbon bond to an adenosyl group, the linkage being directly to the carbohydrate moiety. Another coenzyme of this group, methyl cobalamin, also contains a cobalt-carbon bond, this time to a methyl group. Although not directly related to the problem i t is interesting to note that i t has been possible to synthesise other carbohydrate derivatives with direct metal-carbon linkages. Methoxymercuration ( 3 , 4 , 5 ) of unsaturated sugars has produced mercury-carbon bonded monosaccharides which can be isolated as the chloride salt. Reduction of these salts gives good yields of specific deoxy sugars. It has also been possible to prepare carbohydrate derivatives with tin-carbon, lead-carbon and silicon-carbon bonds ( 6 ) using lithium triphenylphosphine derivatives of the appropriate metal. The Oxo reaction ( 7 ) has been used to synthesise specific branch chain sugars from unsaturated carbohydrates. It is thought that the reactive species is HCo(C0)4 (8,9), which is formed by the reduction of Co 2(C0) Q by R"2. The hydride then reacts with the olefin forming a cobalt-carbon bond. It w i l l be noted that the number of carbohydrate derivatives containing a metal-carbon bond are very few, and the only derivatives of the transition metal elements are the methoxy-mercury compounds and the vitamin 2 coenzyme and the cobalt complex postulated in the Oxo reaction. Of course the formation of direct metal-carbon bonds is not the only way that sugars can be bonded to metals in biological systems since the formation of complexes via donor atoms on the sugar moiety is a viable alters native; e.g. virtually a l l enzymes that transfer phosphate require a metal (10,11), most use magnesium, although some use manganese. Many of the enzymes of interest use adenosine triphosphate as the substrate or one of the sub-strates and the binding of the metal ions to this substrate is through the phosphate groups, i.e. the complexing to the carbohydrate is through an O*atom Magnesium ions also bind very strongly to neutral nitrogen donors, but potassium, sodium and calcium ions bind most strongly to oxygen donor sites in enzyme catalysis reactions. In the carbohydrate area complexes of sugars with inorganic salts in solution are known and are thought to be formed reversibly.(12) Pew complexes have ever been isolated due to the relatively weak interaction of the metal and oxygen. The f i r s t example of this class of complexing was in Reeves (13) investigation of copper-ammoni a complexes of free sugars via the hydroxyl groups. By studying various physical properties that accompany the formation of these coloured complexes in solution Reeves and his coworkers were able to make significant contributions to the conformational analysis of carbo-hydrates. Angyal and coworkers (14, 15, 16) have shown that various metal ions complex with sugars provided the hydroxyl groups are in the required axial-equatorial-axial conformation. * In some places in this thesis common chemicals w i l l be designated by symbol 4. / 2+ - 2+ 2+ + + Using these metal ions, (Ca , Sr > Mg , Na > K in order of complexing strength) Angyal has been able to show that the position of the equilibrium in solution of the cx and ft anomers of the sugars w i l l change i f , as in the case of D-allopyranose, the <x anomer has the required conformation and the (i anomer has not. Similar results have been shown for furanose rings provided the hydroxyls form a cis-cis sequence. Complete change in ring conformation from the stable C 1 form to the less stable 1 C form was also observed in some cases. With the above considerations in mind the general aim of this thesis was the synthesis of sugar-metal complexes with the bonding either through one of the sugar-carbon bonds directly to the metal or through some hetero-atom such as oxygen, nitrogen, sulphur or phosphorus present on the sugar molecule. For convenience, this thesis i s divided into two main sections. The f i r s t section deals with the synthesis and attempted syntheses of various carbohydrates containing metal carbon bonds. The second section discusses the synthesis of various novel heterocyclic sulphur-metal compounds, with a detailed analysis of the ring conformations. 5 . PART ONE 6. INTRODUCTION 7. For a variety of reasons our interest was i n i t i a l l y directed towards the vitamin B coenzyme and i t s analogs. CH2-C0-NH2 ^ 2 "JCV ,CH2-CO-NH2 CH2-CH2-CO-NH2 Co ®© "CH H2N-C0-CH2 NH-C0-CH2~CH2 CH3 CH2 I HC-CH3 I 0 30-P 0 OH O H ' J H H \.. CH3 CH3 CH2-CH2-CO-NH2 0 CH3   HO-CH2^o H Vitamin Bj2 <C63HS0Ol4 N)4 PCO) The coenzyme has been synthesised (17,18) by reduction of the vitamin (cyanocobalamin) or aquacobalamin using sodium borohydride and reaction with the required adenosine tosylate. The reduction occurs via two steps - the f i r s t i s a one electron reduction to vitamin B.0 which undergoes a further 8. one electron reduction to B,„ , which may be considered either as the mono-12s J valent cobalt anion or as trivalent cobalt hydride (19,20). It i s thought to exist mainly as the anion in neutral or alkaline solutions ( l ) . Vitamin B^2S i s the strongest anion known to exist in aqueous solutions (10 times as nucleophilic as I ) ( 2 l ) . Co(l ) in vitamin B „ i s a planar spin-paired 8 d complex and the highest occupied atomic orbital is the d z2 orbital which forms a centre of high polarizability and charge density on the cobalt atom in the z direction, thus making i t a very strong nucleophile (22). The high nucleophilicity of the anion results in easy displacement via an S^2 (21,23) mechanism, of the tosylate from 2'3'-0-isopropylidene-6'-0-tosyl-adenosine resulting (after hydrolysis) in the coenzyme (17,18). Other vitamin B ^ derivatives have been synthesised by attack of the anion on simple alkyl and acyl halides (1 , 2 4 ) and also by opening epoxide rings(23). C H 2 O H C H , I .< C o C O O E t C H 3 Qi C H o C 0 0 H I I C o o r %$0 B r C H a C H 2 C 0 0 H _ _ v i t B H C = C H C H o = CHC00H A ^ C H 0 = C H B r I % C H III C H g II C H C o Vit 12 & c o e n z y m e I C o + C H 9. In a preliminary experiment (vide i n f r a ) i t was found that the product obtained from the r e a c t i o n of 1,2;3,5-di^O-methylene-6-0-tosyl-<* -D-gluco-furanose with the vitamin B anion had a very complex n.m.r."spectrum, and since one of the main aims of t h i s project was to i d e n t i f y the products using t h i s method i t was decided that a simpler system was required. For some time the unusual n u c l e o p h i l i c i t y of vitamin B ^ w a s believed to be due to the e l e c t r o n i c e f f e c t s of the c o r r i n r i n g system on the metal atom ( l ) . Hence many model cobaloximes have been prepared and have s i m i l a r properties to those of-vitamin i n c l u d i n g the n u c l e o p h i l i c displacement reactions with a l k y l halides (25,26,27). Although these model compounds would have been quite good as a l t e r n a t i v e s to vitamin i t was decided that a more general syn-thesis of s u g a r - t r a n s i t i o n metal compounds should be found using metal anions that would i n t e r f e r e very l i t t l e i n the n.m.r. spectra. Thus i t was decided to synthesise carbohydrate d e r i v a t i v e s using metal carbonyl anions as the n u c l e o p h i l i c reagents. Most d e r i v a t i v e s of metal carbonyls containing a metal-C bond are a i r s e n s i t i v e , as opposed to the cobalamins, but i t was f e l t that the o v e r a l l g e n e r a l i t y of the r e a c t i o n would counteract any synthetic d i f f i c u l t i e s that might be encountered. Nucleophilic displacement reactions using metal carbonyl anions give higher y i e l d s than e i t h e r reactions with Grignard reagents(28) or diazomethane (to form methyl d e r i v a t i v e s ) ( 2 8 ) . Metal carbonyl anions have been known f o r some time and are the main precursors to the synthesis of metal-carbon sigma bonds,(29,30,31,32,33) where the carbon containing group may be normal alkyl(28,34,35,36), acyl(37), or aryl(28) d e r i v a t i v e s , or they may be perfluoro groups(38,39,40,41)(which tend to be more stable than the corresponding hydocarbon compounds),sigma-allyl derivatives(42)(which u s u a l l y can be e a s i l y converted to the p i - a l l y l i c or p i -ethylenic complexes) and epoxide derivatives(43,44,45). A l l n.m.r. spectra i n t h i s t h e s i s were proton n.m.r. The nucleophilicities of the anions vary enormously and in trying to rationalise this various effects must be considered (29). (a) A coordination number of 6 is generally more favourable for transition metals which form stable carbonyl derivatives, than one of 5 or 7. Since a nucleophilic displacement reaction of a metal carbonyl anion increases the coordination number of the metal by one, an anion which can attain a co-ordination number of 6 by nucleophilic • displacement: -will be more, nucleophilic than one that cannot. (b) If the electronic configuration of the metal is that of an inert gas then the complexes formed are very stable. However, i f these two were the only factors determining the chemistry of these anions then M(CO),. (Mn, Tc, Re) and Cp-'MtCCO^  (Fe, Ru, Os) would a l l have the same reactivity. However, size and pi-acceptor strength of the ligands also appear to be contributing factors. (c) The heavier (i.e. the larger) the metal atom becomes the more reactive i t i s , which is true for Mn and Re and also for CpM(CO)g where M=Cr, Mo, W. (This rule does not seem to apply for CpFeCCO)^, which is more nucleophilic than the corresponding Ruthenium compound). (d) The pi-acceptor strength of the ligands attached to the central metal atom can also affect the nucleophilicity of a metal carbonyl anion. Since the carbonyl group is a stronger pi-acceptor than almost a l l other ligands which bond to transition metals, replacement of the carbonyl groups with weaker pi-acceptor ligands such that cyclopentadienyl and triphenylphosphine usually increases the negative charge on the central metal atom and thus i t s nucleophilicity. Hence the relative nucleophilicity of certain metal anions can be predicted with a high degree of accuracy. * Cp is interchanged equally with C„H as an abbreviation for cyclopentadienyl. 11. ANION RELATIVE NTJCLEOPHILICITY (46) C 5H 5Fe(CO)~ * 70,000,000 C J U h i C c o ) : 0 0 2 7,500,000 C^H^NiCco) -o o • 5,500,000 Re(CO)" 0 25,000 C^W (CO)- * 500 Mn(CO)" 0 ' 77 CcHcMo(CO)I 0 0 0 67 C[rHcCr(CO)I OO 0 4 Co(CO)~ •) ** CN Cr(CO)" ) ) CN Mo(CO)~ ) « .01 CN W (CO)~ ) 0 * Other nucleophiles such as 0^B± (7 x 10^) and j^As a r e stronger. ** Arbitrarily designated 1 12. I t i s important to note that the reactions of metal anions with halides have us u a l l y been most successful with simple a l k y l h a l i d e s , and the methyl d e r i v a t i v e s appear to be the most stab l e , although there i s a very wide range of s t a b i l i t y of even these d e r i v a t i v e s ; e.g. CH^Co(Co)^ decomposes above -35° ( 4 7 ) while CH^WCpCco)^ i s stable above 145° and i n the a i r i n d e f i n i t e l y ( 2 8 ) . Some halides are not very r e a c t i v e to the anions e.g. iodobenzene reacts only with CpFe^O),^ to give a 2% y i e l d of the corresponding phenyl d e r i v a t i v e (28). V i n y l c h l o r i d e i s s i m i l a r l y unreactive towards the anions (48). The methyl d e r i v a t i v e s are so stable and so e a s i l y prepared because they do not have a hydrogen atom i n a ft p o s i t i o n r e l a t i v e to the metal. However, i n cases where there i s such a P hydrogen an elimination r e a c t i o n can r e a d i l y occur according to the sequence ( 2 9 ) . H 1 Wf-C-H 1 H _ M —1 + H R V V MH + .RHC = C H „ « M — 1 | i J c M 2 + H 2 Thus the s t a b i l i t y of alkyliron derivatives of the type RFe^O^Cp decreases in the following order. R = methyl> ethyl> isopropyl> tert-butyl(36). In the case of the alkyliron derivatives, the resulting hydride decomposes readily to the very stable iron dimer [FeCp(C0) 2] 2 at room temperature (i.e. > 20°) and i s especially a i r sensitive.(36) 13. It appears then that metal anions would be by far the best way to effect this displacement of groups from the sugar molecule. The strongest anion known in this series, i.e. FeCp^O^ was used i n i t i a l l y as the anion and the anion producing the most stable known products of the alkyl metal carbonyl type i.e. CpW(CO)" was also us ed (28). The sugar derivatives used were the toluenesulphonic esters (tosyl) since these are readily prepared and i n most cases have been displaced by nucleophilic groups ( 4 9 ) ' I n one case the iodo-sugar was prepared in the hope that in this particular case the reactivity would be increased since the sugar halides do have greater reactivity towards nucleophilic displacement. The limitations of the reactions should already be obvious from the previous discussion. Since in a l l cases there was a p hydrogen which could be eliminated i t appeared that olefins could possibly be major interfering products. It i s also possible, though admittedly unlikely, that in the case of the primary tosylates, the anion would not be nucleophilic enough to displace the required group. The following discussion deals with the successful and unsuccessful reactions involving sugar tosylates and metal carbonyl anions. 1 4 . RESULTS AND DISCUSSION 1 5 . (A) REACTION WITH VITAMIN B,_ 1 2s The intent o f t h i s study was to prepare metal-carbohydrate sigma bonds. Since vitamin does form a very strong nucleophile and from i t s very obvious b i o l o g i c a l implications i t i s here that the study began. Vitamin B,| ^ . C o O 11 ),was reduced to vitamin B ^ C o ( 1 ) ~ , using sodium borohydride i n a nitrogen atmosphere. The r e s u l t i n g very strong nucleophile was then reacted with 1 , 2 : 3 > 5-di - 0-methylene - 6 - 0-tosyl-.c* - p_-glucofuranose ( 3 ) and the product ( 3 a ) p u r i f i e d as described i n the l i t e r a t u r e (1 7 , 1 9 ) R = O T s (3) The compound decomposed on exposure to l i g h t (a c h a r a c t e r i s t i c of the coenzyme and i t s analogs) so care was taken to protect i t during work-up and p u r i f i c a t i o n procedures as much as possible. The product was i d e n t i f i e d by the change i n the el e c t r o n i c absorption spectrum* from the s t a r t i n g cyanocobalamin. * The e l e c t r o n i c absorption spectrum of cyanocobalamin (vitamin B ^ ) does not change very much when the cyanide ligand i s replaced by other ligands such as H^O, OH , NH^ or pyridine ( 1 9 ) - However, when the cyanide ligand i s replaced by an a l k y l group there i s considerable change i n the spectrum as was noticed f o r ( 3 a ) . In d i l u t e mineral a c i d vitamin B ^ coenzyme and i t s analogs are protonated and change colour from red to yellow. In the protonated species the 5,6-dimethyl benzimidazole i s protonated and no longer co-ordinated to the cobalt and water i s co-ordinated i n i t s place. This r e s u l t s i n an increase i n electron density on the cobalt and a general s h i f t of the absorption bands to a shorter wavelength. (19 ) This can be seen most e a s i l y on the f i r s t absorption at ca. 500 nm. 16. The n.m.r. spectrum of (3a)t although decidedly different from that of vitamin w a s f a r too complicated (due to the porphyrin ring and two sugars groups) to be of any analytical use and hence i t was decided not to pursue this avenue any further. * Fourier transform spectrum in D^ O carbohydrate - vitamin B ^ (3a) 200 scans vitamin B.„ 60 scans Spectrum kindly run by Mrs. Caroline Preston, Dept. of Chemistry, U.B.C 1.0-09-\ WAVELENGTH (nm) FIGURE 1 El e c t r o n i c absorption spectra of Vitamin B]2 (2 x 1G~5 M) (6-deoxy-1 ,2; 3 ,5-cli-O-methylene--x-D-glucofuranose )-6-£-cobalamin ( 3 c ) (2 x 10~^ M) (6-deoxy-1,2;3,5-di-0-methylene- \ -D-glucof uranose )-6-C_-cobalamin ( 3 c ) (2 x 10~ 5 M) i n .05 M HCl. a) b) " c) 18. TABLE ONE LIGAND CN -CH3 (19) Electronic absorption spectra of some cobalamins PRINCIPAL ABSORPTION BANDS, (ran) 1 -4 \ ( £ x 10 in parentheses) 518 (0.95) 548 (1.0) 362 (3-1) 324 (0.8) 306 (1.0) 290 (1 .4) 278 (1 528 (0.79) 377 (1.05) 340 (1.27) -1 ,2:3,5-di-0-methylene-c< -D-gluc 0 f urano s e Protonated 1,2:3,5-di-0-methylene-<* -D-glucof uranose 525 (1.0) 376 ( 1 . 5 ) 364 (1.55) 290 (2.05) 282 (2.35) 263 (2.5) 465 (1.45) 368 (0.62) 296 (1 . 7 ) 287 (1.8) 265 (2.4) 19. (B) REACTIONS WITH CpFe(CO)~, Cp Ko(CO)^, Cp W(CO) Since i t was already known that the anions CpFe^O^, CpMo(CO)3> and CpW(CO)3 are excellent nucleophiles and that the substituent groups of these anions should not interfere unduly with the n.m.r. spectra of any sugar-metal products that might be formed, these anions were next investigated. The anions used were cyclopentadienyl iron dicarbonyl anion, CpFe (CO cyclopentadienyl tungsten tricarbonyl anion CpW(C0)3, and cyclopentadienyl molybdenum tricarbonyl CpMoCco)^. The iron anion was prepared by reduction of the dimer using 1$ sodium amalgam. Due to the ease with which the'dimer can be reformed'and the d i f f i c u l t y in removing i t from the reaction mixture, i t was decided to try to prepare the anion from cyclopentadienyl iron dicarbonyl iodide by reduction with sodium dispersion. It was f e l t that such a reduction and corresponding reaction of the anion would verify that any dimer obtained during an attempted nucleo-phili c displacement had come from the reaction i t s e l f and not from uncleaved dimer as i s possible in the sodium amalgam reduction. However, the reduction was not successful due to the sodium being coated with the sodium iodide formed and being unable to react further. The tungsten and molybdenum anions were both prepared by refluxing sodium cyclopentadiene (prepared by reduction of cyclopentadiene with sodium dispersion) with the appropriate hexacarbonyl (28). CpFe (co) 2 2 Na/Hq » 2Na + CpFe (co) 2 NaCp + W(C0) + 3 C O g This reaction is,however, exclusive to Cr, Mo and W. 20. (1) REACTION OF PRIMARY SUGAR TOSYLATES The o v e r a l l success of reac t i n g the above anions with carbohydrate d e r i -vatives was disappointing, not because of no n - r e a c t i v i t y , but because of the d i f f i c u l t y i n i s o l a t i n g the required product from the reac t i o n mixture. The main d i f f i c u l t y here was the a i r s e n s i t i v i t y of the products and the s i m i l a r i t y i n properties of the s t a r t i n g materials and side products. However, i t was possible to i s o l a t e one pure product and characterise i t , as well as obtaining s p e c i a l evidence f o r one other, and a d e r i v a t i v e f o r another. The 1,2;3»4-di-0-isopropylidene-6-0-tosyl-c* -D-galactopyranose ( 9 ) d e r i -v a t i v e did not react with e i t h e r of the metal anions [CpW(CO),] or [CpFe (C0)~] 0 T S ( 9 ) with which i t was reacted. This n o n - r e a c t i v i t y was not s u r p r i s i n g since i t has been shown that t h i s p a r t i c u l a r sugar t o s y l a t e reacts only slowly with nucleo-ph i l e s such as f l u o r i d e (50), methoxide (51) and iodide (52), although i t does react r e a d i l y with such n u c l e o p h i l i c reagents as potassium thioacetate, sodium azide and l i t h i u m diphenyl phosphine to give the corresponding 6-thioacetate (53), 6-azid.e (54) and 6-diphenyl phosphine oxide (6) d e r i v a t i v e s . The success of the displacement reactions of ( 9 ) appears to depend very 1 strongly on the charge character of the nucleophile (55). Since the replacement re a c t i o n occurs v i a an S 2 mechanism, the formation of the highly polar 21 . t r a n s i t i o n state i s subject to polar f a c t o r s from other electronegative sub-'stituents on the molecule. The most favourable t r a n s i t i o n state e x i s t s when the r i n g oxygen dipole and the t r a n s i t i o n state dipole are at 90°. However, i f there i s a large e l e c t r o n i c a x i a l substituent present at C-4 there would be a l o t of s t e r i c overcrowding and unfavourable dipole interactions.'. This could be r e l i e v e d by r o t a t i o n but then the dipoles i n v o l v i n g the r i n g oxygen and the t r a n s i t i o n state dipole would no longer be at 90° and thus would be unstable. Hence .the use of anionic nucleophiles f o r galactopyranose sulphonate or iodo (which although more re a c t i v e due to the better leaving group are also subject to the same e l e c t r o n i c e f f e c t s ) d e r i v a t i v e s would i n very few cases be success-f u l , whereas more "neutral" nucleophiles such as sodium azide g r e a t l y enhance displacement of galactose probably due to a change i n the p o l a r i t y of the new bond, r e s u l t i n g i n an a t t r a c t i v e i n t e r a c t i o n i . e . lower energy. The h i g h l y n u c l e o p h i l i c anions CpFe^O)^ and CpW(CO)^ were thus not successful i n d i s -p l a c i n g the t o s y l group from the galactose d e r i v a t i v e ( 9 ) . More success was encountered with the 1,2 ; 3 > 5-di - 0-methylene - 6 - 0-tosyl-o< -D-glucofuranose ( 3 ) reactions with the metal nucleophiles, the most success-f u l being the reaction with CpFe^O)^. The p u r i f i c a t i o n of the product was quite d i f f i c u l t , due to the presence of unreacted t o s y l a t e ( 3 ) and i r o n dimer which appeared to be present e i t h e r because of incomplete cleavage i n the i n i t i a l reduction or as a biproduct due to el i m i n a t i o n from the carbo-hydrate moiety. The l a t t e r (although no o l e f i n was recovered) i s the most l i k e l y explanation for the presence of the dimer. It was found imperative to use a 2:1 molar ratio of carbohydrate to dimer and not an excess of tosylate (as is used with more volatile substrates such as methyl iodide) due to the d i f f i c u l t y in removing the unreacted carbohydrate from the desired product. A major method for purification of organometallic compounds is sublimation and thus this was the technique f i r s t attempted using I.R. spectroscopy to investigate the purity of the product (no bridging CO at 1800 cm \ no iron dimer). However, sublimation, using a variety of different sublimation probes (water-cooled, dry-ice and tube) was not successful. Independent of temperature and pressure the product appeared to decompose, with dimer subliming to the probe and the required product subliming to the walls of the apparatus. Decomp-osition appeared to also occur during chromatography on an alumina column but recrystallisation from benzene/hexane/cyclohexane was successful. Some decomp-osition did occur, but after several recrystallisations a pure compound was obtained. Figure 2 shows the I.R. spectra of the resulting compound superimposed on which is the spectrum of the dimer. It i s interesting to note the positions of the protons from Table 2 and Figure 3. The protons are both found at higher f i e l d than is normally the case in sugar derivatives, This results from the shielding effect of the metal bound directly to the carbon atom on which these hydrogen atoms are also bound. This result was not surprising since the corresponding P ( 0 ) ^ , Sn(^, S i ^ and Pb^ derivatives have the protons occurring at chemical shift values ranging from 1.99 to 2.37 (6). This upfield shift of the protons was the only way of determining that reaction had occurred in the case of (3c). Table 3 shows that the coupling constants for compounds (3b) and (3c) (which unfortunately could not be purified) between and H g and H g 2 are quite different, one being quite considerably larger than the other. This 23. FIGURE 2 Infra-red spectra of a) Cyclopentadienyl i r o n dicarbonyl dimer b) (6-deoxy -1,2;3,5-di-O-methylene-f<-D-glucofuranose) 6-C_-(cyclopentadienyl i r o n dicarbonyl) (3b) i n CH^C^ f II ^H, H 2 FIGURE 3 A) Normal 1H n.m.r. spectrum of (3b) in CDCl^ B) Computer simulation of normal spectrum from H„ to H, Cp methylene protons | H 7 H C 66 64 • r V l i methyl tene protons A 4yv mpurity . ^ l . H 6 2 impurity 4 ^ TABLE 2 Chemical shifts (c ) for derivatives of ( 3 ) containing metallic substituents COMPOUND H1 H 2 H 3 ' H4 H 5 H 6 ' Cp (3)f 6.15 4.63 4.4 4.08 4.4 4.4 5.1 (s) (2) 5.0 (q) (2) (3b)^f 6.1 4.43 4.26 3.92 3.87 1 .58 1 .43 4-79 (3c)*' ~ 6.2 . 4.3 3-9 1 .82 1 .59 5.46 5.04(d) (2) 4.91(d) (1) 4.70(d) (1) VJ1 The spectrum of this compound was subjected to computer analysis. f In CDC1, solution 3 * This derivative was not obtained pure, and only these values that could be discerned from the spectrum are reported N.B. A LAjZSc00N 111 computer programme was used to check the assignment of chemical shifts and coupling constants. This programme was originally provided by Dr. A. A. Bothner - By, Chemistry Dept., Mellon Institute, Pittsburgh, Pa., but was further modified by Mr. R. B. Malcolm, Dept. of Chemistry, U.B.C. for use on an I.B.M 360/67 computer. TABLE 3 Coupling constants (Hz) for various sugar derivatives containing metallic substituents COMPOUND ; i \ x 2 JH 3H 4 4 5 J H5 H61 J H5 H62 J H61 H62 3.6 ~ 0.0 (3b) * f 3.8 ~ 0.0 2.5 2.0 6.5 9.0 -11.5 (3=)* f ^ 0.0 2.5 2.2 5-8 9.0 -11 .9 Subjected to computer analysis * Again only values that could be obtained from the spectra are inserted here. f In CDCL, 27. would suggest that the favoured C-5 C-6 rotamer for these two compounds would be one where the dihedral angles between vicinal hydrogens are approximately 180° and 60° (eclipsed rotamers are not being considered). However, i t must be realized that for these particular compounds one cannot be sure quite what effect the metal would have on these couplings (56). More information concerning conformations of the sugar ring can be ob-tained from the coupling constants. The value of L , i s ca. 0 for (3b) which is characteristic of a dihedral angle of approximately 90°. The magnitude of J . 0 and J _ . i s similar so that the possible conformations are restricted to the 5 T 2 and 5V forms.*(57) * The letters T and V refer to twist and envelope conformations respectively, while the subscripts and superscripts are used to indicate which atoms are below and above the "plane of the other ring. 28. The mass spectrum was obtained as additional proof of the structure of compound ( 3 b ) . Since the analysis of this spectrum i s not meant to be rigorous i t i s given in the Experimental Section. The parent ion appears at m/e 364 and i s 2 . 1 $ of the base peak. The base peak occurs at m/e 5 6 . This is probably due in part to combinations of C,H,0 e.g. C^ H^ O* but probably is also due to some 56 „< Fe . This isotope is 9 1 . 6 4 % and has a relatively low ionisation potential of 7.87 ev . ( 5 8 ) The mass spectrum of jZSCH^FeCpCco)^ shows a peak at m/e 56 which is 2 9 $ of the base peak.(59) Iron pentacarbonyl shows a base peak at m/e 56 with the next most abundant ion being the monocarbonyl metal ion. ( 5 8 ) As has already been discussed the reaction of CpW(C0)3 with compound ( 3 ) was successful,in .that Hg was evident in the n.m.r. spectrum at high f i e l d . However, despite several different attempts to purify the product, i t was • impossible to remove the starting tosylate and side products. The tungsten anion is not as good a nucleophile as the iron anion so that, despite long reaction times (7 days) the reaction did not go to completion as was the case for the iron reaction, thus resulting in the d i f f i c u l t i e s in removing the starting material. The nucleophilicity of the CpMo(C0)3 i s even less than that of the analogous CpW(C0)^.(46) Hence i t is not surprising that i n this case the reaction did not proceed at a l l and no upfield protons corresponding to Hg could be seen. C H o R 29. The ease with which organometallic compounds can eliminate olefin i f there is a p hydrogen present i s well known, this tendency i s especially marked in the case of C p F e C c O ^ R compounds due to the high s t a b i l i t y of the resulting iron dimer. This was illustrated in this series of reactions by the reaction of 1 ,2-0-isopropylidene-3 ,5-0-benzylidene-6-0-tosyl-^-D-glucofuranose(5) with C p F e C c o ) ^ . The only product isolated from the reaction mixture apart from the dimer was 6-deoxy -1,2-0-isopropylidene-3,5-0-benzylidene-°<-D-glucofuranose -5-ene ( 5 a ) . It would appear from these observations that the metal derivative ( 5 b ) did form during this reaction but because of the large bulky benzylidene group, there would have been steric crowding for the CpFe(C0)2 group and the derivative would have readily decomposed to the olefin and the iron hydride, C p F e C c O ^ H , which i s very unstable and decomposes very readily to the dimer. The reaction with 1,2,3 , 4-tetra-0-acetyl - 6-0-tosyl- p> -D-glucopyranose was not successful due to the basic reaction conditions which resulted in de-0 acetylation. The amalgam was removed before addition of the sugar, but the metal anion (CpFe^O),,) i s also highly basic and so decomposition of the carbohydrate moiety occurred rather than nucleophilic attack on the tosylate. Mi ao 5.5 5.0 4.5 FIGURE 4 P a r t i a l H n.m.r. spectrum of 6-deoxy-1,2-0-isopropylidene 3,5-0-benzylidene-i * -D-glucof uranose-5-ene (5a) i n CDC1„ TABLE 4 Chemical shift values (i ) f o r compounds ( 5 ) and ( 5 a ) ^ COMPOUND H1 ! H 2 H 3 ' H4 H 5 H 6 P^Me °^Me • (5) 6.1 4.65 4.47 • 4.08 4.4 4.4 5-72 1 -52 (s) (3) 1-34 (s) (3) (5a) * 6.25 4.68 4.52 • 4.44 4.95 4.79 ' 5.56 1-56 (s) (3) 1.34 (s) (3) ^ I n CDCl^ s o l u t i o n • ^ Values taken d i r e c t l y from spectra without computer a n a l y s i s * Spurious s i n g l e t at S 1 .28 TABLE 5 Coupling constants (Hz) for compounds ( 5 ) and (.5a) .COMPOUND • • - \ n 2 J H3 H4 J h61 H62 ( 5 ) 3.8 ~ 0.0 2.2 ( 5 a ) ' 3 .7 ~ 0.0 2.1 1.0. i in CDCl^ solution Taken directly from the spectra without computer analysis. 33. (2),-- REACTION OF SECONDARY TOSYLATES AND EPOXIDES. Generally the n u c l e o p h i l i c displacement of secondary t o s y l a t e s i s d i f f i c u l t to achieve (55). This decreased r e a c t i v i t y at the secondary p o s i t i o n s has been a t t r i b u t e d to both s t e r i c e f f e c t s and to an unfavourable alignment of the d i p o l e s i n the t r a n s i t i o n s t a t e . For carbohydrate systems 2-sulphonates are not normally d i s p l a c e d by n u c l e o p h i l e s due to an unfavourable alignment of the d i p o l e s i n the t r a n s i t i o n s t a t e f o r both the <x and p> anomers.(55). The e f f e c t on formation of the t r a n s i t i o n s t a t e of the d i p o l e s i s not as high i n the case of the 3- and 4- sulphonates, but here the other groups i n the r i n g appear to e f f e c t the ease of displacement, e.g. when a p trans a x i a l s u b s t i t u e n t i s present w i t h respect to the sulphonate group displacement i s impaired. used i t was reacted w i t h 1,2;5 ,6-di-0-isopropylidene-3-tosyl-o<-D-allofuranose(l 1 However, the r e a c t i o n was unsuccessful due to the reasons mentioned above. RS0 3' Since the i r o n anion i s the strongest n u c l e o p h i l e of the three being Me X 0 - C H 2 Me' Me •Me R =OTs (11) 34. Due to previous success w i t h the opening of sugar epoxides (6) i t was decided to attempt to react the metal n u c l e o p h i l e s w i t h some epoxides. Before attempting the r e a c t i o n w i t h sugars however, i t was decided to t r y cyclohexene oxide to see i f i t would open us i n g C p F e C c o ) ^ or CpW(CO)^. However, i n n e i t h e r case was the r e a c t i o n s u c c e s s f u l ; t h i s was probably due, i n the case of the tungsten i o n to the f a c t that the anion i s not a s t r o n g enough n u c l e o p h i l e . In t h i s case the hydride was used so that the product would be i s o l a t e d as the a l c o h o l . The r e a c t i o n w i t h CpFe(C0)2 was done d i f f e r e n t l y due to the d i f f i c u l t y i n s y n t h e s i s and high i n s t a b i l i t y (36,60) of the hydride. The anion was reacted and then the product was t r e a t e d w i t h a molar equivalent amount of concentrated a c e t i c a c i d . •; . In light of the very recent work discussed below the acid i n a l l probability decomposed the product by acidifying the iron moiety to form the hydride which subsequently decomposed to the dimer. At the time of writing this thesis the opening of cyclohexene oxide has f i r s t been achieved (45). The authors did not protonate using acid, but instead used water. The result-ing hydroxyl compounds are very unstable and labile with elimination occurring very rapidly. 35. In concluding this section of the thesis i t seems appropriate to mention other methods for forming carbon-metal bonds that may be applicable to sugars. It i s possible to react a primary halide with the Grignard reagents of the metal carbonyl (6l); e.g. -LX + Mg —•... L—Mg-X L—Mg-X + RX — — • RL+ M g X g |_ = Mn(C0) 5, Fe(C0) 2 Cp. ..Unfortunately there i s very l i t t l e detail on this reaction but i t would appear that no matter what method of synthesis was used the elimination reactions would s t i l l occur. The reaction of the Grignard reagent of the sugar (62) with the metal halide i s another possibility although for simple halides this reaction does not give as high a yield as does the reaction with the anions. 36. PART TWO INTRODUCTION 3 8 . The i n i t i a l intention of the work described in this section was to pursue the synthesis of carbohydrate-metal derivatives by u t i l i s i n g some heteroatoms on the carbohydrate moiety as donor atoms to form metal com-plexes. It was decided that a preliminary investigation of model compounds would be advisable and so the desired ligands were synthesized and the con-formations of the resulting complexes studied. However, sufficient time was not available to continue the study to investigate the appropriate carbo-hydrate complexes. It was decided.that similar metals should be used for these studies as for the studies discussed in Part One, i.e.. transition metals in ap-proximately 'the middle of the Periodic Table (e.g. chromium, molybdenum, tungsten and manganese - a l l of which form monomeric complexes with a chelating ligand) as well as platinum, from which i t was f e l t extra inform-ation could be obtained from the n.m.r. spectra since one isotope, platinum 1 9 5 , has spin ( 3 3 - 7 $ natural abundance). The decision to use this selection of metals led to a minor problem in deciding which would be the best heteroatom to use. Platinum i s classified as a "soft acid" ( 6 3 , 6 4 ) , and hence requires a soft base to form stable complexes.' Manganese i s a.border line metal as are molybdenum and tungsten, with chromium being classified as a hard acid. (All metals are being con-sidered i n their normal valence states) ( 6 5 ) . HARD ACIDS SOFT ACIDS BORDER REGION "-• , . -59. • . Since 'Isoft acids" prefer to associate with "soft bases"(63) i t was necessary now to find a suitable soft base to co-ordinate with these metals*. It i s known that the "softness" of a metal appears to increase as a group is descended (hence chromium i s hard, molybdenum and tungsten are border-line with tungsten being softer than molybdenum) and decreases the higher the valence state of the metal** The hardness of a given acceptor atom i s also a function of the other groups attached to i t . Since carbon monoxide is a soft base i t w i l l tend to make the metal to which i t i s co-ordinated (e.g.•chromium, molybdenum, tungsten or manganese) softer. The effect i s one of reducing the positive charge on the metal, i.e. decreasing the valence state.-Thus the chosen metals'should a l l form stable complexes with "soft" bases, with the possible exception of chromium. ; The overall order of st a b i l i t y of "soft acid" metals for various bases f a l l s in the approximate sequence: S-C > I> Br> Cl> N> 0> F ' with this order strongly reversed for the "hard acid" metals.(63) Since carbon and sulphur are the best ligands for "soft acid" metals i t was decided to make six-membered-ring, sulphur complexes of the afore mentioned metalsand study the n.m.r. spectra obtained so that an analysis of the s i x — membered-rings could be made. There has been considerable work done on the conformational analysis of six-membered heterocycles especially the 1 ,3 dioxalanes and the 1,3 d i -thianes (67-69,72,73) * It i s not meant here that a soft acid w i l l not co-ordinate with a hard base but that the resulting complex w i l l not a) be readily formed and b) be a very stable one. ** For this reason i t has been possible to prepare alkoxide derivatives of tungsten (v), since in this valence state i t i s a hard acid and w i l l complex with oxygen, a hard base (66). Also for this reason Cu(l) i s a soft acid, while Cu(ll) i s borderline. 40. However, six-membered chelate rings of sulphur have not been extensively studied - indeed very few have been made - although the preparative routes to five-membered sulphur chelates have been thoroughly investigated and such compounds are well known (74-80) Six-membered sulphur chelates to tungsten have been synthesised but no conformational studies have been done.(81) Such studies have been done on some sulphur heterocycles besides 1,3 dithiane but in these cases the heteroatom in the 2 position was a phosphorus ..atom (82,83)* 1 Inspection of the H n.m.r. spectra of other six-membered hetero-cycles indicated clearly that spectral analysis would be d i f f i c u l t i f not impossible unless attempts were made to simplify the problem. Accordingly i t was decided to prepare a series of specifically deuterated sulphur ligands. This section of the thesis i s thus a conformational study, using specifically deuterated chelating agents, of some six-membered heterocyclic rings containing two sulphur atoms and various metal atoms. * The conformations of six-membered-ring metal diamines have been studied from a theoretical basis (84,85) and from crystallographic(86-89) and n.m.r. experiments.(90-95) Diarsino metal complexes have also been studied for the transition metals, chromium, molybdenum, tungsten and manganese (96,97) 41 • RESULTS AND DISCUSSION 42. 1, A) Preparation of ligands As already mentioned the 'H n.m.r, spectra of the non-deuterated complexes were impossible to solve, so the deuterated ligands were prepared as summarised here. Preparation of 2,2-dideutero-1,5-dimethyl-1,5-dithiapentane (20) 2,2-dideutero-1,5-diphenyl-1,5-dithiapentane (21) 2,2-dideutero-1,5-dibenzyl-1,5-dithiapentane (98) (22) and 1,1-dideutero-1,3-propanedithiol (83) (28) CH-S CD0CH0CH0S CH_ 3 2 2 2 3 (j) CH„S.CD0CH CH0S CH_(j) 1 2 2 2 ^ 2 T (20) <|)S CD2CH2CH2S (j) (22) H S CH2CH2CD2S H (21) (28) & propiolactone HO CH2CH2CD20H BrCD2CH2CH2Br LiAlD 4 PBr„ RSH HO CH2CH2CD20H BrCD2CH2CH2Br RS CD2CH2CH2SR ( 1) ( 2) (.3) BrCD2CH2CH2Br '1) thiourea 2) KOH R=CH (20), R=(j) (21), R=CHj|>(22) HS CD2CH2CH2SH (28) 4 3 . Preparation of 2,2-dideutero-3-t-butyl-1,5-dimethyl-1 ,5-dithiapentane ( 2 4 ) 2,2-dideutero-3-i-butyl-1 ,5-diphenyl-1 ,5-dithiapentane ( 2 5 ) 2,2-dideutero-3-t-butyl-1 ,5-dibenzyl-1 , 5-dithiapentane (96 ,99) ( 2 6 ) ' 2 ( M ) OOH C tt OH (»•) L i A I D . .OH .OH 2 _OH -OH p- to luene su lphony l ch lo r ide .OTs OTs JOTS .OTs RSH SR SR R = CH ( 2 4 ) , R = ^ ( 2 5 ) , R = C H 2 ^ ( 2 6 ) 4 4 . It was found that attempts to prepare the dibromo-_t-butyl ligand were unsuccessful using phosphorus tribromide, probably due to elimination occurring It was also found that the dichloro-t_-butyl compound obtained by reacting the dio l with thionyl chloride could not be displaced by any of the required thiols For these reasons the ditosylate was made and displaced successfully to form the required dithio ethers. [ ( 2 4 ) , ( 2 5 ) and (26)]. 45. B. Tetracarbonyl complexes of Group V1 I n i t i a l l y the methods followed for the preparation of the required chromium, molybdenum and tungsten complexes were the same as those for the corresponding five-membered rings e.g. heating the hexacarbonyl with an excess of ligand (74) or refluxing the hexacarbonyl and ligand in methyl-cyclohexane.(100) Neither reaction was very successful with the 1 , 5-dialkyl-1 ,5-dithiapentane ligands, although a product was obtained from chromium hexacarbonyl, which was extremely a i r sensitive and l a b i l e . The required tungsten complexes were eventually prepared by u.V. irradiation of a solution of tungsten hexacarbonyl and ligand i n n-hexane for 1y hours. In view of the synthetic d i f f i c u l t i e s further attempts to prepare the corresponding chromium and molybdenum complexes were not made. Fortunately studies on the analogous diarsine complexes have shown that there i s very l i t t l e conformational change in the ring for the series chromium, molybdenum and tungsten,(96) so failure to obtain the chromium and molybdenum sulphur complexes seemed unlikely to seriously hamper this conformational study. Some physical properties of the sulphur, carbonyl complexes of tungsten that were prepared are given in Table 6. The number of bands in the carbonyl stretching region of the I . R . spectra obtained for those compounds soluble in cyclohexane is consistent with C^v symmetry about the metal. The number of bands obtained in dichloromethane varied indicating solvent effects. Unfortunately both the cis and trans arrangements of the ligand would have symmetry so that the I . R . data cannot be used for any definitive structural conclusions. In beginning the discussion of these sulphur, carbonyl complexes of tungsten, i t seems appropriate to begin with the conformationally most simple compounds, i.e. those that are locked almost exclusively in one conformation. Hence the 5-substituted t-butyl derivatives (44) and (45) w i l l be discussed f i r s t in some considerable detail. TABLE 6 ANALYTICAL DATA AND CARBONYL INFRA-RED SPECTRA CHEMICAL ANALYSIS COMPOUND :l Calc. (CO),W S (CH,)CH_CH0CD_S(CH,) 4 3 2 2 2 3 (41) (co) w S(P)CH 2CH 2CD 2S(6) (42) fCO),VJ 's(CH2<j>)CH2CH2CD2S(CH2<j)) (43) 'CC),W S(CH )CH CH[C(CH ) jCD S(CK ) (44) (C0)AW S (6)CH2CH[C (CH ) ]CD2S(|) (45) Found INFRA-RED * Dichloromethane solution **• Cyclohexane solution ||C, 25.0 ji iH, 2.8 C, 41.01 jH, 2.88 | 24-5 2.57 40 .90 3-10 2015 (w), 1895 (s), 1850 (s) **2015 (w), 1899 (s), 1880 (s) 1860 (s). | [ I 2010 (w), 1895 (s), 1850 (s). \ .[ |C, 43.15 i JH, 3.4 i 42.26 3.59 2030 (w), 1912 (s), 1903 (s), 1862 (s). **2022 (w), 1924 (s), 1895 (s) ! 1880 (s) i jC, 31.8 i H, 3.67 i i 31.72 3 .92 • 2020 (w), 1900 (s), 1858 (s) j **2022 (w), 1922 (s), 1895 (s) | 1880 (s) C, 44.95 jH, 3-58 I • 1 44-48 3.98 ! 2015 (w), 1895 (s), 1885 (s), 1 1860 (s). | , I 7 | 123° (decor 135° (decomp-140-141°(decc~p• O / \ 105 (deconp; 119-190°(deccEp) Not soluble i n cyclohexane for I.R. bands to be discerned. 47-CO CO s R CO R=CH 3 ( 4 4 ) , R=jzT ( 4 5 ) , The n.m.r. spectra of the complex (44) i s shown in Figure 5- There appears to be no cross-ring hydrogen-deuterium coupling since no change occurred in the downfield portion of the spectrum on decoupling the deuterium. The spectrum was readily analysed as an ABX system with the AB portion of the spectrum downfield, since both protons and are closest to the sulphur and hence least shielded. The coupling constants and chemical shifts, for complexes (44) and (45) are given in Table 7- Both spectra are very similar indicating l i t t l e difference in conformation is obtained by increasing the size of the substituent on the sulphur atoms. In determining the conformation of these complexes i t may be assumed that the six-membered ring adopts a chair (a) a symmetrical boat (b) or a skew boat conformation (c). chair (a) boat (b) skew-boat (c) 48. FIGURE 5 A) .: Partial 1H n.m.r. spectrum of (C0)4 W~S(CH^)CH 2CH[c(CH^) 3 ]CD 2S(CH^) in CDCl^ with irradiation of the deuterium (44) B) . Computer simulation of the normal spectrum TABLE 7 N.M.R. PARAMETERS OF COMPLEXES OF GROUP VI METAL-SULPHUR COMPLEXES* f COMPOUND : •81 S2 S3 S4 J1.2 J1,3 '1.4 J2,3 Jo A 2,4 J3,4 R S-t-Bu (CO)4W S(CH 3)CH 2CH 2CD 2S(CH 5) (41) 3.06 3.06 2.29 2.29 -12.58 7.57 3.77 3.77 7.57 -12.6 C L 4.47 3 (CO) W S((|)CH2CH2CD2S((()) (42) 3-37 3-37 2.31 2.31 -11 .9 7.64 3.57 3-57 7.64 -12.33 0 7.4 (CO)4W S(CH2^))CH2CH2CD2S(CH2(|)) (43) 2.81 2.81 2.12 2.12 -12.8 7.61 3.87 3.87 7.61 -12.8 0 7.86 CH 2 4-5 (co) w s(CH 3)CH 2CH[C(CH 3) 5]CD 2S(CH 3) (44) 2.43 3.27 1 .66 -12.3 10.1 1 .77 CH^ 2.65 0.70 (C0) 4¥ S ((j))CH2CH [C (CH3) ]CD2S(j) (45) 2.95 3.72 1 .78 -12.2 10.4 1 .8 i 0 7.4 , 0.65 * A l l spectra run i n CDCl^ t A l l spectra computer analysed 50. Since the energy of the symmetrical boat conformation i s much larger for octahedral complexes, then either the chair or the skew-boat forms, i t may be assumed that i t i s too unstable to be the conformation present here, so that the possible conformations are those of the chair (a) and the skew-boat(c). Some progress in deciding which of these i s favoured comes from con-sideration of the "R values" obtained from each complex. The R value method of Lambert(101) and Buys(102) is such that R is independent of the electro-negativities of the heteroatoms, even though the individual couplings may be dependent on the nature of these heteroatoms. T, J. J + J , 0 2 R = trans = aa ee = 5-2cos \ J . J '+ J . 2 cis ae ea 4cos % where "if. i s the ring dihedral angle. or cos -v. = 2+4R H For ideal chair forms R should have a value of 1,9-2.2, for puckered chairs R should be greater than 2.5, and for flattened or flexible (i.e. boat forms) R should be less than 1.8. Using this method i t i s thus possible to calculate the ring dihedral angle, or torsional angle for the S-C-C-C fragment from the vicinal proton-proton coupling constants. This method i s somewhat easier and better than using the Karplus relation(l03) for these heterocyclic systems although 2 i. JHH = A c 0 S THH ~ B c 0 S <t )HH + ° 51 , the (Rjl)-) relationship i s derived using i t . § ^ i s the dihedral angle between coupling protons and A,B,C, are constants which may vary considerably from system to system. The R values calculated for complexes (44) and (45) were found to be 5-70 and 5-77 respectively, indicating that there i s considerable puckering in the ring when i t i s locked at one end by a _t-butyl group, and complexed to a metal at the other. The ring dihedral angles calculated here were ca. 7 0 ° . For a perfect chair the ring dihedral angle would be 6 0 ° . While i t now appears that the complex assumes a puckered chair con-formation, there s t i l l remain numerous such conformations that the complex can prefer, and since the jt-butyl group i s locking the ring the complex must assume one of these. The possible conformations for complexes (44) and (45) are shown in Figure 6. Since the jt-butyl group w i l l prefer to s i t equatorial there are only four possible conformations. The only difference between these conformers is the configuration about the sulphur atom, i.e. do the methyl or phenyl groups s i t diaxial, diequatorial, axial and equat-o r i a l , or i s an equilibrium between the above pos s i b i l i t i e s occurring? It was for assistance in deducing the configuration about sulphur that the different compounds (44) and (45) were prepared with different substituents on the-sulphur. It was felt.that - the" sulphur-methyl complex ( 4 4 ) would possibly form a complex similar to (A) with diaxial substituents* while the * The possibility that the methyl groups would prefer the diaxial positions and thus eclipse leaving the lone pairs to assume a diequatorial orientation arose from the controversy surrounding the actual size of the lone pairs i n various amino compounds. Dipole and polarizability studies on N-methylpiper-idine indicate that the lone-pair requires more space than a hydrogen atom and the requirement appears to be similar to that of a methyl group.(104) I t was f e l t that similar results might be obtained here since the sulphur atom has a pyramidal configuration as does nitrogen in the above case, especially since there i s a large repulsive interaction between the axial lone-pairs which would probably be larger than any interaction between axial methyl groups. 52. 53. sulphur-phenyl complex (45) would prefer the diequatorial conformation (B) or the axial-equatorial configurations ( c ) and (D). Since there i s almost no difference in the two spectra obtained for the two complexes (44) and (45) i t i s assumed that both complexes have the same configuration about the sulphur i.e. (B), (C) or (D ) . '.. Assuming the conformer (B) i s the only one present in solution, the interaction between the diaxial lone-pairs would be large* and hence this would hardly be expected to be a very stable conformer. Since i t appears that both, (A) and (B) are unlikely, and since conformers (c) and (D) would relieve a l l strain due to lone-pair, lone-pair repulsions and steric inter-actions- of axial groups, i t would appear that these would be the most l i k e l y conformers to exist in solution. If, however, the complexes do exist i n either conformation ( c ) or (D) two methyl peaks would be expected for the axial and'equatorial groups** in complex (44)..Such was not the case here, even at -100° only one methyl peak could be seen, thus indicating that the complex is either inverting configuration about the sulphur atom by dissociation or conformations ( c ) and (D) are not present in solution. * Studies on 1,3-dimethyl-1,3-diazanes(l05) indicate that the methyl groups would prefer to exist so that one i s equatorial and the other axial, so that the"rabbit ear" interaction' of the syn-axial lone-pairs on the heteroatom are eliminated. In making this analogy with the diazanes i t must be remembered that a) these compounds do not have a metal substituent, while in the case being considered here the metal does have a large axial substituent (the carbonyl group) which w i l l interact with an axial group on the sulphur, b) The carbon-sul-phur and sulphur metal bond distances are much greater than those of nitrogen, so that the eclipsed lone pairs w i l l not be in such close proximity and c) an X-ray study (106) on Cl(C0) MnAs(CH ) CR^CB^CH^'s (CH ) shows that the axial methyl groups on the arsenic are pushed away and out from the ring, which would tend to indicate that the lone-pairs in conformer (B) would also be pushed out from the ring, thus increasing.the distance between them. ** Two methyl peaks corresponding to the axial and equatorial configurations were always seen in the spectra of the analogous arsenic compounds ( 9 6 ) 54. If conformers ( c ) and (D) were static forms (as they have been assumed to be thus far) i.e. no dissociation i s occurring in solution, then quite complicated spectra would be expected arising from the effects of the axial and equatorial substituents on the adjacent protons, i.e. since the ligand is unsymmetrical, i f ( c ) and (D) are not interconverting, there would be two different isomers present i n solution depending on the position of the deuterium atoms in the ring. From the data in hand i t appears that the complexes (44) and (45) exist either in a static conformation (B), such that both groups on the sulphur are equivalent, and thus only one peak would be expected, or a very rapid equilibrium between conformers (B), (C) and (D) i s occurring, i.e. the sulphur-metal bond i s breaking and reforming very rapidly - faster than the n.m.r. time scale and thus a single peak is seen for the substituent on the sulphur. Such an equilibrium has been seen in the case of a f i v e -membered ring complex of platinum, (107) where two triplets were seen for the methyl groups, with a coalescence temperature of 80°. It was somewhat surprising that i t was not possible to obtain a coalescence temperature for these tungsten complexes (44) and (45). However, rate studies (81 ) of the complex (x) where the sulphur chelate was displaced by phosphites have shown that the rate of the f i r s t step (dissociation of one end of the ligand to form a five co-ordinated complex) i s independent - 3 - 1 of phosphite and i s 3-8 x 10 sec thus indicating the relative ease with which dissociation of one end of the ligand can occur. From the _t-butyl compounds (44) and (45) the skew-boat form can be eliminated, not only from the R values obtained, but also the skew-boat forms in a locked position would result in two isomers depending on the positions of the -CR^  -CD^ - in the ring. Also i f the skew-boat forms were undergoing fast interconversion then average couplings of the J , J and 0,3. 63, J values would be obtained, which were not seen, ae . The axial proton (H^) on the complexes (44) and (45) was about 75 -85 Hz upfield from the equatorial proton (H^)- The chemical shift difference between the axial and equatorial protons in cyclohexane i s much less than this, being 48 Hz.(l08) The reasons for these shielding differences appears to be due to the influences of the carbonyl groups on the metal, the t_-butyl group and the heteroatoras present in the complex. ''••"For the analogous arsenic compounds (96) the difference was of the order of 95 - 100 Hz and for platinum and palladium substituted trimethylene diamine complexes chemical shift differences of up to 65 Hz were found,(90,91) for geminal protons with the axial proton upfield. Dr. R.S. Thompson suggested that "metal-sulphur-carbon bond angles may well be close to 90° in which case arguments concerning axial and equatorial configurations about S are meaningless". If this were so then the sub-stituents must assume a cis relationship with respect to the t-butyl group as the most stable configuration. 56. The n.m.r. spectra* obtained f o r the non-locked compounds ( 4 1 ) , ( 4 2 ) and ( 4 3 ) R = ^ (43) R= CH^zf ( 43 ) were more d i f f i c u l t to analyze than the locked compounds (44) and (45), since r i n g i n v e r s i o n now occurrs between the two possible c h a i r forms, and t h i s i s now a four spin system and would thus be expected to be more complicated. A l l the spectra obtained f o r these three compounds (41), (42) and (43) were very s i m i l a r , that of the sulphur-phenyl complex (42) shown i n Figure 7 being a t y p i c a l example. The spectra are in.each case that of a deceptively simple four spin system, inherent i n the system i t s e l f , since running the sample i n d i f f e r e n t solvents r e s u l t e d i n no change i n the spectra. The a n a l y s i s of the system could only be done by. computer analysis since none of the couplings could be obtained by f i r s t order a n a l y s i s . It w i l l be noted that the computer plot gives the middle band i n each h a l f of the spectra greater i n t e n s i t y than i t should have indi c a t e d by the experimental spectra. The spectra are, i n each case, symmetrical about t h e i r centres and «{ A l l spectra r e f e r r e d to above are n.m.r. spectra. 57, FIGURE 7 - JL , ,1 ,, I , ,,,1 I 1 1.. 3.55 3.05 255 g 2.05 i i ) Partial H n.m.r. spectrum of (CO)^ ¥ S(s2f)CH2CH2CD2S(0)(42) in CDCl^ ) Above spectrum with irradiation of deuterium ) Computer simulation of the normal spectrum 58. y i e l d s olutions i n d i c a t i v e of AA'BB' systems. The coupling constants and chemical s h i f t s of the compounds (41), (42) and (43) are l i s t e d i n Table 7. The symmetrical' n.m.r. spectra i n d i c a t e that the six-membered rings are under-going f a s t conformational i n v e r s i o n , since, i f t h i s were not the case, more complicated spectra due to four inequivalent protons would be obtained. Again the skew-boat conformer can be eliminated by i n v e s t i g a t i n g the "R v a l u e s " ( l O l ) of the complexes (41 ), (42) and (43) having a -CH 2CH 2-fragment. For the tungsten complexes (41), (42) and (43) R has values of 2.0, 2.14 and 1.96 r e s p e c t i v e l y , i n d i c a t i n g an average i d e a l chair conformation f o r these complexes r e s u l t i n g i n a r i n g dihedral angle of 58 -57°. There did not appear to be any c o r r e l a t i o n i n the s i z e .of the sub-s t i t u e n t on the sulphur with the value obtained f o r R. The values obtained f o r R f o r these complexes are considerably smaller than f o r the analogous 1,3 dithiane heterocycles (67, 109) which are found to be puckered chairs with R values of 3-23 and r i n g dihedral angles of 63° even i f the r i n g i s locked i n the two p o s i t i o n by a phenyl group. These r e s u l t s would tend to i n d i c a t e that the influence of the metal substituent and. the substituents on the sulphur atom i s such that the r i n g takes up an almost perfect chair conformation i n s o l u t i o n , on the average. Since i n t h i s case there i s rapid equilibrium betv/een the two r i n g conformations there are consequently twice as many possible conformations 59. that the ring may assume, compared with the locked compounds (44) and (45). The possible conformations for compounds (41 ), (42) and (43) are shown in Figure 8. Conformer (E), (F), (G) and (H) are analogous to (A), (B), (C) and (D) for the locked ring. The remaining four conformers in Figure 8 apply to the alternate ring conformation, with similar configurations about sulphur as depicted above for (E), (F), (G) and (H). For the complexes (41), (42) and (43) di fferent substituents were again placed on the sulphur to determine i f possible, the configuration about the sulphur atom. Since again no difference was observed i t would appear that while rapidly under-going ring inversion these complexes either assume diequatorial configur-ation around the sulphur, or, are undergoing fast equilibrium with the equatorial-equatorial and both axial-equatorial configurations thus inverting the configuration about the sulphur, analogous to the locked rings already discussed. 60. FIGURE 8 POSSIBLE CONFORMATIONS OF: 62. C. Manganese complexes It was f e l t that some interesting results might be obtained by changing the metal to that of another group. Manganese was selected since a previous study in the diarsine series (97) indicated that the Mn(C0),X (where X = c:i Br. and i ) group acts to lock the ring completely in one conformation. Since i t was found that the difference obtained in the spectra for the different halogens for the diarsine complexes was only slight i t was decided to only synthesise the bromo derivatives for the sulphur complexes. Due to the small (almost insignificant) changes obtained from varying the substituent on sulphur for the tungsten derivatives i t was decided only to synthesise two compounds of manganese; the methyl substituted sulphur complex (51 ) and the 3-substituted t_-butyl methyl substituted sulphur complex (52). Various methods were tried in attempting to synthesis these derivatives, but since halogeno pentacarbonyl metal derivatives are much more reactive than the hexacarbonyl derivatives the reaction would usually go too far forming some decomposition product (probably the dimer [Br(C0)^Mn]2 or . some sulphur ligand containing dimer). y.V. irradiation was attempted i n i t i a l l y since this method was successful with the group V1 metals, but decomposition had occurred after y hour. Heating and refluxing in benzene were next attempted, Refluxing was far to vigorous, but providing the reaction was monitored by collecting and measuring the carbon monoxide that was evolved heating the reactants in benzene at 50° was quite successful. The products were sublimed at 50 - 60° for a few hours to remove a l l unreacted starting material [Mn(C0),-Br] and then the product sublimed at 150°. The compounds obtained did not appear to be very air-sensitive but instead were light-sensitive decomposing completely in a few days. The analytical and infra-red data for these compounds are given in Table 8. The coupling constants and chemical shift values are given in Table 9. The I.R. data in the carbonyl stretching range are consistent with the expected cis geometry of the ligands as well as with a possible trans ligand arrangement so they cannot therefore be used for any structural conclusions. The n.m.r. spectrum of the complex (5l) is shown in Figure 9. The four protons for this complex were readily discerned and the spectrum was analyzed as an ABXY system with a l l the protons separated showing that the ring is indeed locked by the Mn(C0)gBr group. The coupling constants could a l l be obtained from the spectra, contrary to the analogous tungsten compounds (4-1), (4-2) and C+3). Again only one methyl peak was v i s i b l e , whereas in the analogous arsenic compounds the axial and equatorial methyl peaks are separated by ca. 6 Hz. Thus i t appears that again the configuration about the sulphur is unknown although i t appears to be a fast equilibrium between the diequat-o r i a l and equatorial-axial configurations as discussed for the tungsten complexes. The compound (5l) has thus far been shown to be locked but the question now arises, into what conformation? The large value indicates that these 2 protons are involved in a vic i n a l axial-axial interaction which would be smaller i f the ring were locked in any skew boat conformer. The possibility of a boat form (as well as a skew form) is excluded by an investigation of the "R" value for this compound (51). It was found to be TABLE 8 CARBONYL INFRA-RED SPECTRA AND ANALYTICAL DATA (MANGANESE COMPLEXES) COMPOUND ANALYSIS Calcd. Found INFRA-RED* m. p.. Br(CO)5Mn StCH^CHgCH^D^CH)^ (51) C, 27-0 1 27.90 H, 3.70 1 3-78 2023 (w), 1950 (s), 1915 (s). 112-113° Br(CO)3Mn S(CH3)CH2CH[C(CH^)?]CD2S CH^ (52) C, 34.9 34.68 H, 4.36 4.60 2025 (w), 1949 (s), 192® (s) 152° * CH2C1 2 • " ** Decomposes on further attempts to purify TABLE 9 N.M.R. PARAMETERS OF MANGANESE COMPLEXES* f COMPOUND . • S1 S2 S S4 S-CH^ S-t-Bu J1,2 J1,3 J1,4 J2.3 J2,4 J3,4 (CO),Br Mn S(CH_)CH_CH_CD.S(CH_) \ 3 3 2 2 2 3 (51) 3.35 2.72 1 .98 2.26 2.5 -12.4 11 .9 2.6 2.4 6.4 -16.0 (CO)3BrMnS(CH^)CH2CH[C(CH^) ]CD2S(CB^) (52) 3.06 2.80 1 .56 2.48 0.97 0.93 -11.96 10.9 1 .6 * A l l spectra run in CDCl^ ^ A l l spectra computer analysed 66. FIGURE 9 A) . Partial H n.m.r. spectrum of Br(C0)3Mn S(CH3)CH2CH2CD2S(CH )(51) in CDC1 B) . Above spectrum with irradiation of deuterium C) Computer simulation of the normal spectrum 67. 3.66 i n d i c a t i n g that the r i n g i s that of a puckered c h a i r . The r i n g d i h e d r a l angle % was calculated to be 65°. Drawing analogy to the X-ray study performed on the d i a r s i n e complex (106) i t can be assumed that the conformation here i s as shown below. I t was hoped that the complex obtained from rea c t i n g Mn(C0) nBr with o the 2,2-dideutero-1,5 dimethyl-3-t_-butyl-1,5 dithiapentane (24) ligand would r e s u l t i n two isomers since the r i n g i s now locked at both ends and the two isomers depicted by (M) and (N) would be possible*. * Such isomers were obtained i n the analogous d i a r s i n e complexes (97) although most of the mixture was converted on heating to the most stable isomer [thought to be s i m i l a r to (M), since t h i s was the conformer ob-tained from the X-ray study on C l ( C 0 ) 5 MnAs(Me) 2CH 2CH 2CR" 2As(Me) 2] (106) 68. In the _t-butyl complex (52) there was a small amount (ca. 15$) of another isomer present, as evidenced by a small peak on the side of the large _t-butyl peak in the n.m.r. spectrum but the ratio of the two isomers did not change significantly on heating. Hence i t would appear that an equilibrium of the two isomers was formed i n i t i a l l y , the most stable being the one formed in greatest amount, and this ratio did not change very much on sublimation since the sulphur ligands were able to assume the most stable isomer by an on-off equilibrium process under the i n i t i a l reaction conditions used here. The small amount of the least stable isomer did not interfere with the analysis of the spectrum, and i t was not able to be isolated from the main isomer. However, despite the fact that i t was known that there were two isomers present in solution there i s s t i l l only evidence for one sulphur-methyl peak, while four peaks were seen in the diarsine compounds. This can be interpreted in various ways. Since the complexes can be assumed to have ring conformations either (M) or (N) [ignoring the configuration on the sulphur for the present and assuming that the predominant conformer in each case is the one with the Jb-butyl group equatorial], the most stable isomer would most probably be the (M) as was found for the diarsine com-plexes. (97) It would appear that this complex does give rise to some indication that a fast equilibrium is occurring about the sulphur so that a change in configuration occurrs. Since there are two isomers present in solution, such that the _t-butyl group gives rise to 2 peaks, i t would be expected that the S-Me peak would also give rise to 2 peaks at least. However, again only one sulphur-methyl signal is seen, which tends to, indicate that some fast equilibrium i s occurring, so that inversion of configuration about sulphur i s occurring rapidly. 69. The spectrum of the compound ( 5 2 ) i s shown in Figure 10 and the n.m.r. data are given in Table 9. The "R" value obtained for this complex was calculated to be 6.81, thus making i t an even more puckered chair than was obtained for complex (51 )• In the tungsten complexes the effect of the t_-butyl group was to increase the puckering of the ring since the "R" value here increased from ca. 2.0 for the flexible rings (41), (42) and (43) to ca. 5 . 7 for the locked complexes (44) and (45). Here the increase in "R" value i s similar, increasing from 3.66 for (51) to 6.81 for (52). It seems unusual that the _t-butyl group should increase the amount of puckering in the ring and i t would appear that locking the ring at the metal end as for (51) does not have as much effect on the ring conformation as locking the ring at the 3 position as for (44), (45) and ( 5 2 ) . The chemical shift values for protons 3 and 4 i n complex ( 5 1) indicate that the chemical shift value of the axial proton H^  i s at higher f i e l d (ca. 28 Hz) than the corre-sponding equatorial proton which i s in agreement with the results found for the tungsten complexes previously. However, the protons H^  and H^  appear in both (51 ) and ( 5 2 ) with this order reversed. In these complexes the H.j (axial) proton appears at unusually low f i e l d so that i t i s actually downfield of the equatorial (H^) proton by approximately 53 Hz for (51 ) and 26 Hz for ( 5 2 ) . It is possible that in these compounds there is some form of attractive interaction between the axial protons and the halide on the manganese.* * The X-ray study on Cl(CO^Mn As(Me)2CH2CH2CH2As(Me)2 (106) showed that the M-Cl bond i s bent towards the chelate ring, indicating that such an attractive interaction does occur here. FIGURE 1 0 A B o ( 1 A) P a r t i a l H n.m.r. spectrum of Br(C0) 5 Mn S(CH5)CH2CH[c(CH ) ]CD2S(CH ) (52> i n CDC1. B) Computer simulation of the normal spectrum 71 • D. Pt complexes It was f e l t that some information about the metal end of some cyclic sulphur systems might be obtained by using a metal that has a nuclear spin. Since platinum i s a soft acid i t can readily form stable bonds with sulphur, a soft base. Platinum 195 has a spin of \ present in 33.7 % natural abundance, while a l l other isotopes of Pt present naturally do not have any spin. The easiest Pt (11) complexes to prepare are the dichloro com-plexes, prepared by reacting the ligand with K^PtCl^, and f i l t e r i n g off the insoluble complex. However, this i n s o l u b i l i t y of these complexes re-sulted in them being extremely d i f f i c u l t to work with, in that the only solvents in which they are found to be soluble were dimethyl sulphoxide and pyridine. In pyridine i t appeared that the complexes were dissociating to form the pyridine complexes which could precipitate out in ca. 10 minutes. In DgMSO the portion of the n.m.r. spectrum to be investigated occurred in almost every case under or near the water and D^ MSO peak present in even freshly opened vials of D^ MSO and samples prepared and sealed in a glove bag. No changes in the coupling constants could be found for the complexes and the free ligands and there was no evidence for H-C-S Pt couplings. Hence i t would appear here that the complex i s also dissociating in solution although no other evidence for this could be found and there has been re-ported H-C-S-Pt coupling constants for the•five-membered ring case (107), and for non-cyclic systems for the dichloro complexes (111, 112) Since there was very l i t t l e success encountered here with the sulphur chelates i t was decided to investigate some complexes with a covalent bond to the sulphur. In such compounds there was no problem as to the con-figuration of the substituents on the sulphur, since there were none and there was no confusion in picking out the H-C-S-Pt couplings since there 72. was only the ring protons to which the Pt could couple. These compounds were prepared by reacting the required ligand with P ^ P ^ ^ C l ^ in a basic solution for 24 hours. The product obtained was recrystallized several times from acetone to yield yellow crystals i n a l l cases. However, the elemental analyses on these compounds were not good, so one compound (58) was re-crystallized ca. 10 times to obtain an acceptable analysis. Complex (58) was purified to indicate that with considerable d i f f i c u l t y , these complexes can be purified. The contaminant, the starting Pt complex, does not inter-fere with the n.m.r. spectrum so that the purification of the other complexes was not attempted. The n.m.r. data for these compounds are given in Table 10. There was. one rather large disadvantage in using the triphenyl phosphine derivatives in that the P-H couplings complicated the spectrum considerably and while heteronuclear decoupling simplified them i t was not possible to triple irradiate both the D and P nucleii at the same time.* In studying these six-membered-ring platinum complexes i t was hoped that the absolute J, and J . platinum couplings could be obtained trans gauche for a completely locked ring. For this purpose the S-pheriyl derivative, here were disappointing, since the protons a l l f e l l on top of each other so that neither H., H_ or H, could be distinguished and thus their couplings * It would probably, in retrospect, have been easier to interpret the spectrum i f the arsino derivatives had been used. TABLE 10 N.m.r. parameters f o r platinum complexes* 'OMPLEX S VALUES J VALUES JPt-H JP-H >CH2S)2Pt(P<|> ) 2 3) 3 . 7 0 * * ( s a t e l l i t e ) 3.905, 4-106 (s) 42.0 '<j>3)2Pt S CH 2CH 2S (60) 2.182(a) 3-038 3-264 45-2J S1 S2 S 3 i J1,2 J1,3 J2,3 J3,2 1 i 1 '(J>5)pt s CH(CH3)CH2S 1) 3-075 3.80 3-975 CE 3 1.93 -11.5 9.1 4.0 J3-CH, 6.5 5 Cannot be j ca. 0.5 determined ! '45)2Pt s CH(CH2CH3)CH2S 2) 2.80 3-18 3-18 CH_ 1.17 3 CH 2 1.98 -11 -7 10.0 3-9 7-0 " j , ca. 0.5 '(j)5)2Pt's CH 2CH 2CP 2 S * * * 6) 2- 975(s) 3- 265 3-555(s) 2.975(s) 3-265 3-555(s) 2.35 2.35 I - 6.5 i 6.5 i i i i 6.5 J1,4 6.5 58 1 7-4 Pt-H 5-0 j '<|»5)2Pt S CH2CH((|>)CD2S 8) 453.6 (s) 419-8 398.6 (s) 446.6 (s) 419-8 392.4 (s) 419.8 4> 8.3 57-6 5-7 <>J3)2Pt s CD (CH 5 )CD 2CH 2 S 9) 4.14 (s) 3-92 3.705(s) 3-875(s) 3-475 3-12 (s) CH 3 1.775 -12.5 I ^"trans^^*^ gauche 4 5-75 j JPt-H I J2-p 1 2 ' ° 4 j 1-p 5-0 p-H .11 samples run in CDC13 with external T.M.S. lock. ** (s)=Pt 195 s a t e l l i t e . * These values were obtained accurately from the dideutero and tetradeutero analogs (56) and (57) -0 01 into each other also could not be distinguished. i s pulled so far down-f i e l d by the influence of the phenyl group, while both and 11^  are moved upfield due to the influence of the sulphur atoms. However, a phosphorus 3 coupling of 5 . 7 Hz could be distinguished, and the J^, „ coupling was found r t — n to be 5 7 . 6 Hz. r The results obtained for the 2 methyl derivative (pcj^^Pt S CD^H^CT^CHpS ( 5 9 ) were however rather interesting and much more informative than the 3-phenyl complex ( 5 8 ) («S 3 P) 2 P The methyl group appears to lock the ring primarily into one con-formation but.unfortunately i t i s impossible to say whether i t i s completely locked or not. The methyl group, i t i s assumed, w i l l primarily s i t equat-o r i a l . The only couplings assumed to be present would be those of the protons into the Pt and P and to each other. The coupling of the protons to each other i s —12 . 5 , while the coupling into the Pt has two values. The axial (downfield) proton H^  has the smaller coupling of 4 8 Hz while the equatorial proton has a coupling of 7 3 Hz. The equatorial proton i s in a trans orientation to the Pt and hence has the larger coupling.* Thus while i t i s not possible to assume J, of 7 3 Hz is the maximum JJ„, coupling r trans Pt possible (since i t is not known whether the ring i s locked or not) at least 3 one can say that J„, , must be at least. 7 3 Hz. Also the sum of J, Pt trans trans + J , = 120 Hz. , gauche * In analogous platinum - ni trogon aix-mombered-ring complexes g 65 Hz and J . - 22 Hz.(93) gauche 75. The equatorial proton i s coupled to the phosphorus with a coupling constant of 12.0 Hz, much larger than the axial coupling of 5-0 Hz. This i s due to the equatorial proton being able to form a planar W configuration (as shown above) with the Pt whereas the axial proton cannot. Apart from the expected couplings, there were other couplings present in this compound, which can be explained as follows. There i s a small coupling of 5-75 Hz into the methyl group, appearing as shoulders on the side of 4 approximately one-third the intensity of the main-peak. This i s the J_ coupling and as explained earlier this i s due to the planar W formation (as shown above) that the methyl group can assume with a platinum atom. There also appears to be some long range P coupling into the methyl peak, since the two shoulders and the main peak are s p l i t by .a coupling of 1.5 Hz. There also appeared to be some additional coupling into both H^  and E^, which cannot be explained by any sensible reasoning. 3 To obtain the Jp^ . ^ couplings for a non-locked six-membered ring complex (55), (56) and (57) were prepared. Pt(P0Q 3'2 ( 5 5 ) nt = R = H ( 5 6 ) F W R 2 = D ( 5 7) R ^ R 2 = D 3 The Jp^ _ platinum coupling obtained for these non-rigid complexes was 58.0 Hz*. This i s in good agreement with the average coupling obtained. * As expected from a comparison of the J, and J , value previously r trans gauche cited, the coupling through sulphur is again larger than that through nitrogen. For analogous amino complexes Jp^ _ was found to be 43 Hz (92) 76. from J and J , [obtained from complex (59)] of 60.5Hz. 4 The J phosphorus c o u p l i n g obtained f o r these complexes was again q u i t e l a r g e , being 7-4 Hz, again being due to the planar ¥ c o n f i g u r a t i o n that can be assumed with the phosphorus atom. This c o u p l i n g i s s m a l l e r than the average of the phosphorus couplings obtained from complex (59) (8.5 Hz), but the agreement i s not too bad. I t seems reasonable to suppose 4 that a Jp^_ c o u p l i n g should be able to be obtained, provided there i s a planar ¥ c o n f i g u r a t i o n from the platinum to the protons on C^. For complex (56) such c o u p l i n g can be seen as shoulders cn the proton s i g n a l , w i t h a c o u p l i n g of ca. 5-0 Hz. R = H ( 8 0 ) R = C H 3 ( 6 1 ) R = C H . C H (62) o 2 3'2 U n f o r t u n a t e l y , w h i l e i t was c l e a r that some Pt c o u p l i n g was present i n the system (due to the presence of sm a l l peaks adjacent to the other peaks i n the spectrum) the ^Jp^ g couplings c f compounds (61) and (62) 4 could not be determined and there was no evidence f o r J to the CH^ or e t h y l group. However, the co u p l i n g constants could be determined f o r the non-substituted compound (60). The n.m.r. spectrum obtained was a s i n g l e peak as expected, s i n c e the five-membered r i n g i s f r e e to f l i p i n t o t w i s t and envelope conformations w i t h two s a t e l l i t e s f o r the Pt co u p l i n g s . The coup-l i n g s here were 45-2 Hz, somewhat s m a l l e r than the c o u p l i n g obtained f o r the analogous c o n f o r m a t i o n a l l y i n v e r t i n g six-membered r i n g (55)- This i s probably due to the d i f f e r e n c e s i n d i h e d r a l angle f o r f i v e and six-membered r i n g s w i t h the angle f o r the six-membered r i n g being l a r g e r , and hence l a r g e r 4 c o u p l i n g constants. The J phosphorus c o u p l i n g i s a l s o s m a l l e r , which would 7 7 . appear to be because six-membered rings are more able to take up a planar W type conformation so that coupling over four bonds becomes significant(113). For a f i n a l comparison the acyclic benzylidene complex was prepared, so that the average Pt-S-C bond could be obtained and compared with the cyclic systems already discussed. The Jp^ _ coupling obtained for complex (63) was found to be 42 Hz*, which i s smaller than either J couplings for 3 the five or six-membered rings, as would be expected. Hence the J platinum (63) coupling increases from 42 Hz for an acyclic system, to 45 Hz for a f i v e -membered ring to interconvert, to 58 Hz for a six-membered ring free to interconvert, to values for J, and J , of 73 Hz and 48 Hz respect-trans gauche ^ ively where the ring i s no longer free to interconvert so readily. / C H S ffCH S P t ( P ^ 3 ) , * The coupling obtained for the analogous dichloro (111, 112) complex [(cj)CH2 ) 2 S ] 2 P t C l 2 was 43 Hz. 78. GENERAL CONCLUSION 7 9 . In conclusion, I would like to make a few remarks about the two sections of this thesis. From the f i r s t part i t can be concluded that.the reaction of FeCp(C0)2 and WCpCco)^ i s successful when applied to primary tosylates of carbohydrates with the exception of the galactose derivative ( 9 ) . Both the iron and tungsten anions were found to react with the glucose derivative (3) although only the iron adduct could be isolated, besides the Vitamin B.J2 adduct. The iron anion was also found to-react with the slightly sterically hindered glucose derivative ( 5 ) although the product obtained was the olefin. It appears that the secondary tosylates are unreactive to this kind of nucleophilic attack. The main problem besides the unreactivity of the sugar sulphonates in some cases was the d i f f i c u l t y in isolation of the products. Due to the similarity of the products and the starting materials and the high sensi-t i v i t y of the products (especially in solution) and their general instab-i l i t y , purification procedures were not easy. It i s for this reason that I think that the reactions would have been more successful i f the substrates had not been so complicated, i.e. i f instead of the carbohydrate derivatives simple cyclohexane and cyclopentane derivatives had been used. In concluding section two I would make the comment that these reactions appeared highly successful when applied to the model ring systems and i t -would seem that i f the appropriate thio-sugar derivatives were made some 80. conformationallyinteresting and perhaps biologically active compounds could be prepared. It would appear from the study made that the rings in these derivatives are conformationally fixed in the case of a l l the manganese derivatives, the tung-sten derivatives where there i s a _t-butyl or a phenyl group on the 3 position, in the Pt case, a group on the 2 position. The results obtained regarding the conformationsoof the tungsten and manganese derivatives were not very different from those obtained for the analogous arsenic derivatives, the only point of conjecture being the orient-ation of the substituents on the sulphur atom. The results obtained for the Pt complexes indicate that considerable information could be gained from such complexes regarding the actual con-formation about the metal in solution, be i t a square planar arrangement as for platinum ( l l ) derivatives or an octahedral arrangement as for platinum (1V-). The ava i l a b i l i t y of Fourier Transform techniques (since the solubility of these complexes severely hampers routine n.m.r. results) could result in extensive insight being obtained as to the nature of these complexes. The relevance of such studies on model systems i s that they w i l l have applicability for future studies on biological systems containing similar metals or moieties, providing the experimental d i f f i c u l t i e s can be overcome. 81 , EXPERIMENTAL 82. GENERAL METHODS 1. A l l solvents used were dried and d i s t i l l e d as follows; a) Tetrahydrofuran was d i s t i l l e d over lithium aluminium hydride. b) Benzene and toluene were dried over calcium chloride for at least 48 hours, f i l t e r e d and d i s t i l l e d over sodium wire. c) Hexane, cyclohexane and pentane were d i s t i l l e d over lithium aluminium hydride. d) Methanol was dried over molecular sieve and d i s t i l l e d from magnesium turnings. 2. A l l solvents were degassed before use. This was done either by freezing in liquid nitrogen, evacuating the flask and allowing the solvent to melt or by evacuating the flask without freezing un t i l boiling occurred. Both evacuation procedures were repeated three times - in the f i r s t case the solvent was then ready for use, in the second nitrogen was bubbled into the solution through a sintered glass bubbler for a further thirty minutes. A l l solvents were degassed on the day required for reaction. 3. The alumina used for any columns needed for chromatography of air-sensitive compounds was of neutral grade. It was degassed (114.100) by heating under vacuum at 200° for twenty-four hours. The column was packed under nitrogen and the fractions were collected under nitrogen after being eluted with degassed solvents. 4. A l l reaction flasks were opened in a Vacuum Atmospheres Corporation dry box, and a l l Infra-Red, n.m.r. and analysis samples of air-sensitive com-pounds were prepared in there. 5. A l l Infra-Red spectra were measured on a Perkin Elmer 457 Grating spectro-meter in dichloromethane. 83. 6. A l l proton n.m.r. were measured on a Varian XL 100 spectrometer or on a Varian HA 100 spectrometer in a frequency sweep mode using tetramethyl-silane as either an internal or external standard. Proton chemical shifts are a l l reported in the S scale. 7. Melting points were measured on a Thomas-Hoover capillary melting point apparatus and are not corrected. 8. Microanalyses were carried out by Mr. P. Borda of the U.B.C. Chemistry Department and by Alfred Bernhardt Mikroanalytisches Laboratorium, West Germany. 9. A l l mass spectrometry was measured on a AEI MS9 spectrometer. 10. Optical rotation measurements were made on a P.E. 141 Polarimeter using a 1.0 cm c e l l . 11. A l l Ultraviolet spectra were run on a Carey 15 recording spectrometer. 12. Dicyclopentadiene was cracked by use of a fractionating column at approx. 45° and collected under nitrogen. It was used immediately in reactions so that dimerization did not reoccur. 13. The Whatman Advanced ion exchange celluloses used were DE 22 (Diethyl-amino ethyl celluloses) and CM 22 (carboxymethyl cellulose). These were treated as directed with acid and base and then washed with d i s t i l l e d water un t i l neutral. The celluloses were then packed on 10 x 1.5 cm columns. 14. A l l irradiations were performed using a Hanovia 450 watt immersion ultra-violet lamp. 84. PART ONE 85. 1 . 2 ;5.5-di-Q-methylene-*-r^-glucof uranose *(115) (1) Glucose (40 g), paraformaldehyde (200 g) and a mixture of cone, sulphuric acid '(140 g) and water (140 g) were well mixed and slowly heated in an o i l hath to 130°, when the solid f i n a l l y dissolved. It was then rapidly cooled by running water, ice was thrown i n and then i t was poured into an ice-cold solution (cone.) of potassium carbonate (107 g) and then neutral-ised to pH 7. Any solid present was f i l t e r e d off and washed with ethanol and the f i l t r a t e s concentrated to 100 ml; any solid present was f i l t e r e d off and washed with a l i t t l e water. This separation of the mono and dimethylene compounds was followed by a continuous chloroform extraction for 24 hours. The dried chloroform extract was evaporated under vacuum and then d i s t i l l e d at high vacuum to give a yellow o i l , but some decomposition occurred. Crude Yield: 28.22 g (62.5$), after d i s t i l l a t i o n at 165° 9-77 g (21.6$) (lit...3- .005 mm 122-124° 25$ )." 6-0-acetyl-1 ,2;3,5-di-0-methylene-<x-D-glucof uranose (116) (2) Due to the d i f f i c u l t i e s encountered in extracting the alcohol from the water, the acetate was made which can be readily deacetylated by sodium methoxide in methanol. D-glucose (100 g) was dissolved in water (50 ml) and mixed with gla c i a l acetic acid (400 ml). Paraformaldehyde (110 g) was then added followed by cautious addition with shaking of cone, sulphuric acid (50 ml). The mixture was heated at ca. 80° for an hour, and the clear yellow solution was then cooled and water (400 ml) was then added. The solution was then extracted three times with chloroform and the combined extracts washed with water * A l l synthetic procedures w i l l be outlined here whether literature re-ported syntheses or not. 86. unti l neutral. The chloroform solution was then dried with sodium sulphate and the solvent removed to yield a yellow o i l which crystallised after three days. Crude Yield: 73.2 g (53.8$) 1 ,2:3,5-di-0-methylene-«-D-glucofuranose (1) The acetate (2) (73-2 g) was dissolved in dry methanol (500 ml) to which was added at room temperature, with st i r r i n g , a freshly prepared sodium methoxide solution (200 ml; 1.0 g sodium in 200 ml methanol). When the reaction was complete (ca. 30 minutes, and a test drop added to water dissolves and is alkaline to litmus) the solution was neutralised by s t i r r i n g with Amberlite IR-120(H+). The resin was f i l t e r e d off and the solvent removed to yield a yellow o i l which was not purified further. Crude Yield: 54.8 g '(47.0$ based on «C-D-glucose) 1,2;3,5-di-0-methylene-6-0-tosyl-*-D-glucofuranose (3) The alcohol (1 ) (9-77 g) was dissolved in pyridine (50 ml) and p-toluene sulphonyl chloride (15 g) in pyridine (50 ml) added slowly at room temperature with s t i r r i n g . The solution was then l e f t standing 24 hours after which i t was poured into water (100 ml). The product was extracted into chloroform which was then washed several times with water to remove the pyridine. The solvent was then removed (toluene was used to remove the last traces of pyridine). The o i l was recrystallized twice from methanol using decolourizing carbon to yield white crystals. Yield: 6.7 g (39-1$) m.p. 111°-112° (lit.112) 3,5-0-benzylidene-1 ,2-0-isopropylidene-6-0-tosyl-<x-D-glucofuranose (5) 3,5-0-benzylidene-1,2-0-isopropylidene-x-D-glucofuranose * (4) (1 g) * Kindly supplied by Mrs L. Evelyn, Chemistry Dept. U.B.C. 87, was dissolved in pyridine (20 ml) and p-toluenesulphonyl chloride (2.5 g) in pyridine (10 ml) added very slowly with s t i r r i n g . The resulting mixture was stirred and l e f t in ice overnight after which i t was poured into cold water and extracted with chloroform. The chloroform extract was washed several times with water to remove any pyridine. The chloroform was removed and the resulting o i l recrystallized from methanol to yield white crystals. Yield: 1.1 g (68$) 1,2,3»4-tetra-0-acetyl-6-0-tosyl-p-D-glucopyranose ( 7 ) i ;. ' 1 ,2,3,4-P-D-glucopyranose tetraacetate * (6) (1 g) was tosylated in a similar manner to (4), using p-toluenesulphonyl chloride (2.5 g) in pyridine (20 ml). The resulting mixture was recrystallised from methanol to yield white crystals. Yield: 1.2 g (67.0$) 1,2;3,4-di-0-isopropylidene-K-D-galactose (117) (8) Zinc chloride (129 g) anhydrous D-galactose (108.9 g) and concentrated sulphuric acid (4-3 ml) were placed in a 3 l i t r e flask containing dry acetone (1350 ml) and shaken for four hours. Sodium carbonate (217 g) in water (380 ml) was added, stirred for ten minutes and f i l t e r e d . The solution was evaporated to a syrup and then extracted with water and chloroform. The chloroform layer was dried over sodium sulphate, f i l t e r e d and evaporated to yield a syrup which was not purified further. Crude Yield: 43.2 g (36$) * Kindly donated by Mr. Joffre Berry, Dept. of Chemistry, U.B.C. 88. 1,2;3i4-di-0-isopropylidene-6-0-tosyl-x-D-galactose ( 9 ) Di-O-isopropylidene galactose (8) (20 g) was reacted with p-toluene sulphonyl chloride (22 g) in pyridine, (as. before). The product obtained was crystallized from aqueous ethanol to yield white crystals. Yield: 15.49 g (50.5$) 6-deoxy-6-iodo-1,2;3,4-di-O-isopropylidene-t* -D-galactose (10) ™ S I 1,2;3»4-di-0-isopropylidene-6-0-tosyl-e<-D-galactose ( 9 ) (4.0 g) and sodium iodide (3.0 g) were refluxed in 2,4 pentadiene (20 ml) for six hours. The reaction mixture became very dark, suggesting that decomposition had occurred. No product could be obtained from the reaction mixture and n.m.r. indicated no tosyl group or anomeric proton. The reaction was repeated, but with heating and not refluxing, but in this case no reaction occurred and the tosylate was recovered. 6-deoxy-6-iodo-1,2;3,4-di-0-isopropylidene-°<-D-galactose(118,119) (1 0) 1»2:3»4-di-0-isopropylidene-6-0-tosyl-»c-D-galactose (8) (6125 g) was placed in a sealed tube with sodium iodide (4.5 g) and anhydrous acetone (50 ml). The tube was then degassed and sealed, and heated for 36 hours at 110°C in a pressure tube oven. The solution was allowed to cool and then the tube was opened. Crystals of sodium tosylate were f i l t e r e d off and washed several times with dry acetone. The f i l t r a t e s were combined, the acetone removed and the resulting syrup was then stirred with water containing a few crystals of sodium thiosulphate. The resulting brown o i l was recrystall-ized twice from methanol using charcoal/celite as decolourizing agents to yield white crystals. Yield: 1.42 g (22.7$) m.p. 68-70°. ( l i t . 118 m.p'. 69-71°) 8 9 . (6-deoxy-1,2;3,5-di-0-methylene-oc-D glucofuranose) 6-C-cobalamin ( 3 a ) A solution of cyanocobalamin (100 mg) i.e. Vitamin and cobalt nitrate(l20) (1 mg) in water (10 ml) was placed in a small (25 ml) erlen-meyer flask, f i t t e d with a serum cap and degassed with nitrogen for 10 minutes. Then a degassed solution of sodium borohydride (15 mg) i n water (1 ml) was added. Very slight foaming occurred and the solution immediately turned brown and over some time (ca. 1 hour) the solution changed to a dark grey-blue. Since the product i s light sensitive a l l apparatus was wrapped in f o i l from this point on.(l7,2l). 1 ,2;3,5-di-0-methylene-6-0-tosyl-<*-D-glucofuranose (79.2 mg, 3 x excess) in ethanol (5 ml) was added after i t had been degassed. The solution immediately turned red-brown, and was stirred for a further ten minutes. The cobalamin was purified by extraction through phenol, a procedure to remove water soluble salts from cobalamins. The procedure followed was that of Dolphin ( 9 ) and was as described below: a stack solution of phenol (100 g) in methylene chloride (100 ml) was prepared. . The aqueous solution was extracted into 20$ of i t s volume of phenol solution. The organic layer was separated and the aqueous layer was re-extracted with succesive.aliquots of phenol-methylene chloride u n t i l no further colour was extracted, (ca. 60 ml). The combined organic extracts were then washed with d i s t i l l e d water (2 x 20$ the volume of the organic layer) and the organic layer was diluted with methylene chloride to 10 times i t s original volume (600 mis). The cobalamin was re-extracted from the organic layer with d i s t i l l e d water ( 5 $ the volume of the organic layer) un t i l no colour remained in the organic layer (120 ml). The combined aqueous extracts were then washed with methylene chloride (3 x the volume of the combined aqueous layer) to remove any traces of phenol. The aqueous solution was then reduced in volume to 5 ml using a rotary evaporator, but ensuring that the water bath at no time 9 0 . was hotter than 50 , since decomposition occurs at this temperature. The solution was then placed on a column ( 10 x 1 .5 cm) of diethylaminoethyl cellulose from which i t was eluted ; rapidly with d i s t i l l e d water. The eluate was then concentrated to 5 ml and introduced to a column of carboxymethyl cellulose (10 x 1.5 cm). Both columns had been prepared according to directions from Whatman for advance ion exchange celluloses.* D i s t i l l e d water eluted the required product leaving inchanged Vitamin B ^ absorbed at the top of the column from which i t could be eluted with . 0 2 M HC1. The light absorption spectrum of this eluate in a solution of pH 7 was identical with that of Vitamin B ^ at "the same pH. The aqueous eluate was reduced in volume to 1 ml and treated with acetone u n t i l a faint cloudiness appeared. On standing, red needlelike crystals of product we're deposited. These were f i l t e r e d , washed with acetone and a i r dried. Yield: 92 mg (81 .1$) max H 2 ° (2 x 10"5M) 1.0 cm c e l l A -05NHC1(2 x 1 Q - 5 M ) max 525 nm ( E - 1 x 10"4 376 nm U = 1 .3 x 10 364 nm u = 1 .55 290 nm u = 2 • 05 282 nm u 2 • 35 263 nm (e = 2 • 5 465 nm u 1 .45 x10 368 nm u = .62 296 nm . ( * = 1 .7 287 nm (e = 1 .8 265 nm u = 2 .4 - 4 Methylcyclopentadienylirondicarbonyl CpFe(C0)2CH3 (121) (13) The apparatus ( as used for a l l sodium amalgam reactions was a 500 ml round bottom flask equipped with a stopcock at the base and with a dropping funnel, mechanical s t i r r e r and a reflux condenser attached) was flushed with nitrogen for five minutes and mercury (20 ml) was added during this See General Experimental Methods. 91. period. Sodium (2.0 g) was heated slightly in a crucible while covered with a hydrocarbon solvent. It was then added with vigorous s t i r r i n g in 0.2 g lots to the mercury. The amalgam was allowed to cool to room temperature and then dry tetrahydrofuran (100 ml) and iron dimer [C-pYe(CO)^\ ^  (5 g) were added. Nitrogen was flushed through the system for the entirety of the reaction. The solution changed from the red-purple colour of iron dimer to a dirty yellow and f i n a l l y to a yellowish green. The mercury amalgam was removed after two hours ( any excess sodium present in the amalgam was destroyed by s t i r r i n g with methanol) and methyl iodide (5 g) was quickly added to the stirred solution, and the s t i r r i n g continued for a further eight hours. The s t i r r e r , condenser and dropping funnel were replaced by stoppers and a nitrogen inlet and the solvent removed by the continued flow of nitrogen. A sublimation probe was inserted and the flask evacuated, but the vacuum was not applied continuously since the product is very volatile. The product was collected on a water cooled probe at room temperature.and scraped into a Schlenk tube which was flushed with nitrogen. The methycyclopentadienyl-irondicarbonyl is a yellow-orange waxy solid m.p. 78-80°C which decomposes in a i r after about one hour. Yield: 2.57 g (94-8$) _1 Infra-red analysis showed no bridging at 1800 cm but showed two -1 -1 terminal carbonyls at 2000 cm and 1950 cm N.m.r. showed 2 singlets in CDCl^ at; 4.8 ( 4.7) (122) Cp 5 protons 0.2 ( 0.11) (122) CH^ 3 protons (6-deoxy-1,2;3,4-di-0-methylene-y-D-glucofuranose)6-C_-(cyclopentadienyl iron  dicarbonyl) (3b) The method followed for a l l reactions involving iron dimer [ FeCp(C0)2] ^  was similar to that used for the preparation of MeFeCp(C0)2 although on a smaller scale with a large excess of sodium amalgam being present in a l l cases. 9 2 . Sodium pellets (2.0 g) were added with vigorous s t i r r i n g to mercury (20 ml) followed by addition of dry tetrahydrofuran (100 ml) and iron dimer (0 .5 g). After s t i r r i n g for two hours the mercury amalgam was removed and 1,2:3,5-di-0-methylene-6-0-tosyl-oc- D glucofuranose ( 3 ) (0.9 g) was added in tetrahydrofuran (50 ml). The reaction mixture was stirred overnight and then the tetrahydrofuran was removed. The solid obtained was removed from the flask in a glove bag and then transferred to the dry box. Various methods of purification were attempted, sublimation using a water cooled probe, dry ice acetone probe and tube sublimer were a l l found unsuccessful due to de-composition to the starting dimer as well as some yellow product (impure) being obtained. No dimer was actually present in the reaction mixture as was indicated by t.I.e. of the reaction mixture in benzene, red sublimation product 0.56 yellow sublimation product 0.125 Rg starting sugar 0.13 Iron dimer 0.6 R, reaction mixture 0.12 Column chromatography on alumina also proved unsuccessful due to de-composition occurring on the column. Recrystallization proved to be the most successful purification technique. Yellow crystals were obtained from benzene/cyclohexane/hexane after a week. The compound decomposes in air after 15 minutes to a reddish-brown solid and burns in a i r to leave a red residue. Yield: 100 mg m.p. 73° [ « ] 23 = + 64 --9 ± 0.1 (C, I, CHC15) Anal.Calcd.for C.JL.,OnFe : C, 49-48; H, 4.43; Pe, 15-34 1 5 1 o I Found: C, 49-53; H, 4-60; Fe, 11.85 (123) I.R.: 1900, 2100 cm" (terminal carbonyl bands) no 1800 cm-1 band (bridging carbonyl) 93. S 6.1 $ 4.43 S 4.26 5 3-92 6 3.87 S 1 -58 5 1 -43 I 4.79 X 5.04 6 4.91 J 4.70 d, 1, J. H1 H2 3.8 V d, 1 , J. ca. 0 H 2) d, 1 , J. H3 H4 t, 1, J. 4 5 = 2.5 = 2.0 H 3) V broad , T _ £ c ^ t r i p l e t ' l j JH 56 1 " H j J H g = 9.0 5) H5 62 q» 1 > J-s, 5, d, 2, d, 1, d, 1, H6 =6 1 2 =-11-. 5 H,) 6 1 Cp) methylene on H^H2) methylene on H^ H,-) methylene on H^ H,-) u i n | i i i i M n i | i l n III n u n 1I.M11,. II iiini in fULL (P-2C0)' (P-CO)" in* H , II, II 220 230 240 250 240 270 280 290 300 310 320 330 340 350 360 370 380 390 Mass spectrum of (6-deoxy-1,2;3,5-di-0-methylene-*-D-glucofuranose)-6-C-cobalamin (3c) 95. MASS SPECTRUM 70 e.v. * 365 364 336 334 308 306 304 299 286 285 279 278 248 243 230 192 190 Assignment P + 1 (Pe 5 7 isotope) P (Fe 5 6) P - CO (364 -28) P-CH20 (364 - 30) - 2C0(364 -56 ) - (CO + CH20) (364 - 58) - 2CH 20 5 5 - 77 ) ) (364 - 60) (364 - 65) jo of base peak 0.3 2.1 17.6 3.3 23.1 11 .9 0.5 1 .0 rearrangement of 'Q^JE^+ CE^ 37.0 78 ) to benzene ring (2C0 + CHO) (2C0 + CR"20) (2C0 + 2CR"20) (2C0 + CCH,. 5 5 (2C0 + 2CH20 + H20) [CH 3FeCp(C0) 2]' [Pe(C0) 2C 5H 6 ] ( 14.4 26.0 73.4 35-9 2.9 7.8 25.3 32.5 i . e . - i + CO GO * • This analysis of the mass spectrum i s NOT meant to be rigorous. 96. m/ e Assignment 186 P - (FeCp(C0)2H) . 183 P - (2C0 - 2CHo0 - CVHV) 2 5 5 182 P - (2C0 - 2CR"20 - C H g) 178 [FeCp(CO)2H]® 177 [FeCp(C0) 2f 143 P -[ (FeCp(C0)2CH2)® + CH20] 127 P -[ (FeCp(C0)2)® - 2CH20] 134 Fe C..H,-5 6 122 Fe CaEc 5 6 121 Fe CVIL-5 5 113 P -[ (Fe Cp(C0)2CH2)® + 2CH20] 112 Fe(C0)2® 91 Possibly tropylium ion 84 FeCO 78 • C 6 H 6 9 77 C 6 H 5 9 66 °5 H6 + 65 °5 H5 + 56 Fe + + other ions jo of base peak 40.6 10.7 37.7 38.5 55-1 25-1 10.1 73-9 92.9 35-5 8.6 89.0 23.8 4.8 8.9 31 -4 59-4 100 (58,59) 97. Possible pathways for some other ion follow. m/e 125(12.2$) -CO y m/e'71 (90.1$) . -CO m/e 43. (52.8$) 98. 100. 101 . m/e 95 (36.3$) 103. Attempted reaction FeCp(C0) 2 with 1,2,3,4 tetra-O-acetyl - 6-tosyl--D-glucopyranose (7) The reaction procedure was as before using sodium (2.0 g), mercury (20 ml), 0.35 g [FeCp(C0) 2] 2 and tetrahydrofuran (100 ml). The sugar (7) (1.0 g) was added in T.H.F. (100 ml) after removal of the amalgam and stirred overnight. The reaction mixture had turned red immediately on addition of the sugar, and after removal of the solvent and dissolution in benzene and f i l t r a t i o n a red solid was obtained. This solid was shown by IR to contain iron dimer, but the n.m.r. showed no sugar resonances, only a peak at£4.8 corresponding to the dimer cyclopentadiene. It appeared that the sugar had completely decomposed in the alkaline conditions present. Attempted reaction FeCp(C0) 2 with 1,2;3,4-0-isopropylidene 6-tosyl-«c-D-galactopyranose ( 9 ) The reaction was carried out as before using sodium (2.0 g), mercury (20 ml), iron dimer (.1.5 g) and galactose tosylate. ( 9 ) (3.5 g). The reaction mixture slowly turned red and after 2-g- hours was distinctly red. T.I.e.on alumina in benzene, IR and n.m.r. a l l indicated sugar tosylate ( 9 ) and dimer as the only products. The non-reactivity of the galactose tosylate was disappointing and hence the iodide was reacted, since the iodide ion is a better leaving group than the tosylate. However the reaction using sodium (2.0 g), mercury (20 ml), dimer (0.5 g) and iodo sugar (1 .12 g) (10) was again unsuccessful, with IR indicating a strong band at 1800 cm ^ and n.m.r. showing sugar to be present as well. 3,5-0-benzylidene-6-deoxy-1,2-0-isopropylidene- ex -D-glucofuranose-5-ene (5a) This compound was prepared while attempting to prepare the sugar-FeCp(CO) derivative. As in the previous reaction the amalgam was prepared using sodium (2.0 g) [ FeCp(CO)J ? (0.3 g) and T.H.F. (100 ml) were added and stirred for 104. 2 hours and mercury (20 ml). 1,2-0-isopropylidene 3,5-0-benzylidene-6-tosyl-<*-D-glucofuranose ( 5 ) (0.7 g) was added in T.H.F. (50 ml) after the amalgam was removed. The mixture was stirred overnight and had changed to a brownish colour. The solvent was removed and the solid dissolved in the minimum of benzene and f i l t e r e d . Some benzene was removed under vacuum and the pre-cipitate which formed removed. This procedure was continued u n t i l the solution was no longer reddish brown, but was instead a yellowish brown. A l l the solvent was removed to yield a yellow-brown solid (20 mg) in which there was no iron dimer present (no 1800 cm band in IR) and no tosylate present from the n.m.r. From the n.m.r. i t was deduced that the compound was the olefin, but there was a spurious high f i e l d peak in the spectrum which could not be re-moved even after recrystallization from celite/charcoal/methanol: N.m.r. * (CDCl^) S 6.25 S 4.68 5 4.52 6 4.44 i 4-79 S 4.95 S 5.56 S 1.55 S 1.34 (d, 1 (d, 1 (d, 1 (d, 1 (d, 1 (d, 1 (S, 1 v2 = 3-7 H 3 H 4 61 62 benzylidene proton H, H, (S, 3» isopropylidene protons (S, 3> isopropylidene protons * Spurious peak at S 1 .28 105. Preparation of sodium cyclopentadienyl tungsten tricarbonyl (28) 1 . Sodium dispersion 50$ Na/50$ parafin wax (0.46 g) was stirred in dry toluene (20 ml) for five minutes and allowed to settle. The toluene was then removed. This was repeated three times to ensure the sodium i s wax free. 2. Dicyclopentadiene must be freshly cracked before use. This was done by heating under nitrogen and collecting the fraction d i s t i l l i n g at 40°C. 3. The wax free sodium dispersion (.23 g) was stirred i n tetrahydrofuran (30 ml). Cyclopentadiene (0.7 g) was added in tetrahydrofuran (30 ml), solid tungsten hexacarbonyl (3.4 g) was added and the solution refluxed for 12 hours. The solvent was removed under vacuum and a small sublimation probe inserted into the flask, which was then heated to 1.00°C vacuum of 0.05 mm. Heating was continued un t i l no more unreacted carbonyl was sublimed. The sodium salt was obtained as a yellow powder (3.0 g - 88$) Preparation of methyl cyclopentadienyl tungsten tricarbonyl(28) (l4) To check the purity of the tungsten salt, a known compound was prepared. Sodium cyclopentadienyl tungsten tricarbonyl (0.5 g) was dissolved in tetra-hydrofuran and methyl iodide ( 1 ml - large excess) was added and the resulting mixture stirred overnight under nitrogen. The reaction mixture went brown, and the product was sublimed directly from the reaction mixture at room temperature and pressure of 1 mm to yield pale yellow crystals, which were reasonably stable and can be handled for short periods of time in the a i r . Yield: 0.4 g (94$) n.m.r. S7-0 Cp S2.0 CH_ 3 Attempted preparation: Reaction of NaWCp(C0)3 with 1,2;3,5-di-0-methylene- 6-0-tosyl-K-D glucofuranose to form (6-deoxy-1,2;3,5-di-0-methylene-f<-D glucofuranose) 6-C_-(cyclopentadienyl tungsten tricarbonyl) (3c) Sodium cyclopentadienyl tungsten tricarbonyl (1 g) was dissolved in 106. tetrahydrofurari (30 ml). Then the sugar tosylate ( 3 ) (1.g) was added to tetrahydrofuran (30 ml). The reaction mixture was stirred for 36 hours, but s t i l l showed considerable tosylate, so it'was stirred for a further 36 hours. The reaction mixture s t i l l showed considerable tosylate, but i t was decided to remove the solvent and see i f there was any other product present. Some of the reaction mixture was dissolved in benzene, f i l t e r e d and evaporated to dryness. N.m.r. of this product indicated quite a few products - starting sugar tosylate appeared to be ca. 28$, and 33$ of the product (percentages taken from H.j integration) appeared to contain a cyclopentadienyl group and an upfield multiplet corresponding to the two Hg's. (Analogous to the iron sugar). However sublimation was unsuccessful in purifying the product. At 180°C an orange waxy solid appeared. This solid caught f i r e in the a i r indicating that i t probably was a decomposition product. Analysis of the n.m.r. of this compound was impossible. IR indicated 5 carbonyl bands and t . l . c . showed 6 compounds, one of which was the sugar tosylate. Separate sublimation of the sugar tosylate (160° - 180° at 1 mm) and NaWCp(C0)3 solved the problem. The sodium salt sublimes at 180° - 200°C to an orange waxy solid, which was a mixture of five compounds. Hence any tungsten-sugar adduct must de-compose and these products also obtained. Column chromatography was success-f u l in removing the tosylate ( 3 ) from the reaction mixture, but no other product could be eluted off the column, i.e. decomposition occurrs on the column. Recrystallization in this case was not successful either, since there was always tosylate and some other product crystallizing out from the benzene/hexane with the required product. The reaction was repeated for a longer length of time (7 days, i.e. doubling length of reaction) but again there was ca. 30$ of tosylate present, and again the required product could not be isolated. 107. N.m.r. (CDCl^) (As far as can be assigned) * 6.2 (1, H1 ) H4 > H 5 ) 5 4.3 (t, 1, J • = 2.5 4 5 JR\HC = 2 , 2 4 5 6 3.9 (Sextet, 1, = 5-8 5 61 J 5 6 2 = 9'° S 1 .82 (q, 1, J = -11.9 H, ) ( 61 62 61) S 1 -59 (q, 1 , H ) 2 S 5.46 (S, 5, Cp ) Attempted Preparation: Reaction of Na¥Cp(C0) 3 with 1,2;3,4-di-0-isopropylidene-6-0-tosyl-iK-D-galactopyranose. ( 9 ) . The reaction was repeated as for the glucose reaction. The mixture was Na¥Cp(C0) 3 (1 g) in T.H.F. (30 ml) to which was added galactose tosylate (1.1 g) in T.H.F. (30 ml) was stirred for 7 days altogether, each day the mixture was tested using t . l . c . on s i l i c a in ether, but no reaction could be detected. N.m.r. after 7 days on the benzene soluble product showed this to be unreacted tosylate, and sublimation of the benzene unsoluble portion yielded the decomposition products as obtained for glucose compounds as i n -dicated by t . l . c . Attempted reaction NaMoOcoJ^Cp with 1 ,2;3,5-di-0-methylene 6-0-tosyl-tx-D- glucofuranose ( 3 ) NaMoCp(C0)3* is prepared (28)in an analogous manner to Na¥Cp(C0)3-NaMoCp(C0)3 (1 g) was stirred in tetrahydrofuran (50 ml) with glucose tosylate (1.35 g) for 24 hours. The solvent was then removed and t . l . c . indicated only 2 compounds present - i.e. the two starting materials. N.m.r. also indicated tosylate present in the benzene soluble fraction. * Kindly donated by Allan Crease, Dept. of Chemistry, U.B.C. 108. Attempted reaction. FeCp(C0) 2 with 1,2;5,6-di-0-isopropylidene-3-0-tosyl- c<-D-glucof uranose (11 ) The reaction was performed as before using sodium (2.0 g), mercury (20 ml) iron dimer (1.5 g) in T.H.F. (100 ml). Amalgam removed a sugar (3.5 g) added in T.H.F. (100 ml) and the mixture stirred overnight. The solvent was re-moved, the solid dissolved in benzene, f i l t e r e d and benzene removed to yield a red solid which was a mixture of dimer and unreacted tosylate as indicated by n.m.r. and I.R. Reaction of FeCp^O^Na with cyclohexene oxide. As before mercury (20 ml), sodium (2.0 g), iron dimer (0.5 g) in T.H.F. (50 ml). The mixture was stirred for 1-g- hours. The amalgam was removed and cyclohexene'oxide(3 ml - large excess) was added and the mixture stirred overnight. The concentrated acetic acid (0.15 ml) (4 drops) was added. The reaction mixture, which was an olive green, was stirred for another 2 hours and then the T.H.F.was removed, leaving a red reaction product. The reaction mixture showed iron dimer was present. The reaction flask was heated very slowly at pressure 0.05 mm pressure, in an attempt to sublime out any pro-duct. The only product obtained was the dimer as dark purple crystals- as shown by I.R. Preparation of HWCp(CO),. (100) NaWCp(C0)3 (1 g) was dissoved in T.H.F. and stirred with..33 ml (7 drops) glacial acetic acid for 2 hours. The solvent was then removed and the hydride sublimed from the reaction mixture as a pale yellow solid (0.2 g). After a short while the sublimate became pink and was discarded. This pink solid i s the dimer 'W^CO)^, - the hydride i s extremely a i r sensitive and w i l l oxidize to the dimer very readily. luy . Reaction with cyclohexene ;oxide with cyclopentadienyl tungsten tricarbonyl  hydride. . Tungsten hydride (0.2 g) with cyclohexene oxide (1 ml - large excess) was stirred for 2 days in dry T.H.F.. The reaction mixture went black, so the solvent was removed leaving a black o i l behind, in which crystals ap-peared. N.m.r. of these crystals (which were soluble in pentane) showed a singlet at S5-5 but no other peaks. They burned leaving a black residue which turned white, thus indicating a metal was present. These crystals were shown to be the starting hydride, since on sublimation the pink dimer was obtained, i.e. no reaction had occurred with cyclohexene oxide. 110. PART TWO 111. FLOW SHEET 1 P R E P A R A T I O N OF L I G A N D S 2 Li AID. ^ » HOCD CH CH OH <s 2 2 R r M e ( a 0) ( 2 8 ) R=<rf («0 RrzoyzM 8 2 ) 112. 1,2-dideuterc—1^-propanediol (83) p -Propiolactone (10.8 g) was dissolved in anhydrous tetrahydrofuran (40 ml) and added dropwise over a period of 45 minutes to a vigorously stirred suspension of lithium aluminium deuterid.e (3.0 g) and anhydrous tetrahydrofuran (120 ml). After refluxing for 12 hours the solution was treated successively with water (3.0 ml), 15$ NaOH (3.0 ml) and water (9 ml). The suspension was f i l t e r e d and the precipitate washed with boiling tetra-hydrofuran. D i s t i l l a t i o n of the combined tetrahydrofuran fractions yielded 7.0 g (61$) of the desired product as a liquid boiling at 110-112° / 15 mm. 1 ,1-dideutero-1,3-dibromopropane (83) Phosphorous tribromide (8.5 g 33$ excess) was added dropwise at 30°-40° over a period of fifteen minutes to the vigorously stirred diol (2.5 g). The mixture was then heated to 90° for two hours, cooled and treated with water (50 ml). The dibromide was extracted with ether, dried over sodium sulphate and d i s t i l l e d to give 4-5 g (70$) of pure dibromide boiling at 94.- 98° / 90 - 95 mm. 1 ,1-dideutero-1,5-dimethyl-dithiapentane (98) (20) A solution of potassium hydroxide (11.2 g) in ethanol (100 ml) was stirred at 5° and methane thiol (11 g) added. The dibromide (20.2 g) was added dropwise since the reaction i s exothermic at 25° - 45° over a fifteen minute period. Large quantities of white precipitate (KBr) were deposited immediately but the reaction was continued for a further two hours. After this time the mixture was diluted with water (200 mis) and the heavy o i l which formed was extracted with ether (4 x 100 ml) to give a yellow liquid b.p. 181 - 183°. Yield: 11.1 g (82$) 113 . 1 ,1 -dideutero-1,5-diphenyl-dithiapentane (98) (21) This was made analogous to (20).From dibromide (2.0 g), ethanol (20 ml), potassium hydroxide (1.12 g) and phenylthiol (2.1 g) was obtained 2.0 g (77$) of a yellow liquid. 1,1-dideutero-1,5-dibenzyl-dithiapentane (98) (22) The preparation was as previously described. From dibromide (3.01 g), ethanol (30 ml), potassium hydroxide (2 g) and toluenethiol (3.75 g) was obtained 4.1 g (79$) of yellow liquid. 1,1-dideutero-1,3-propanethiol (83) (28) A mixture of the dibromide (1.5 g), thiourea (1.68 g) and water (1 .20 ml) was refluxed and stirred for 4-5- hours to form a solution. Then a solution of potassium hydroxide (2.2 g) in water (4.0 ml) was added and refluxing and sti r r i n g continued for a further eight hours. The upper layer of di t h i o l was d i s t i l l e d directly, separated from water in the d i s t i l l a t e and dried. The product was a yellow, foul smelling liquid. Yield: .73 g (84$) The other ligands used for the Pt complexes were obtained from Aldrich Chemicals. They were 1,2 ethanethiol ( 3 2 ) , 1,2 propanethiol (33), 1,2 butane-thiol (34), ethanethiol (35) and toluenethiol (36). The remaining ligands, (30) and (30a), (31 ) and (31a) were kindly donated by Dr. J. R. Campbell, formerly of the Dept. of Chemistry, U.B.C. Preparation of 2-hydroxymethyl-3,-dimethyl butyric acid (96,99) The reaction was carried out under nitrogen. n-Butyl lithium (1 mole 64 g) was added dropwise to diisopropylamine(l mole 104 g) in freshly dried T.H.F. (400 ml) at 0°C. After the addition was complete 3,3-dimethyl. butyric acid (58 g) was added slowly, with heat being evolved. Paraformaldehyde (15 g) was then added and the solution stirred overnight at room temperature. The 114. reaction mixture was then poured into a large quantity of water and extracted with ether. HC1 was then added unt i l the solution was just acid and then extracted with dichloromethane. On removal of the solvent the product obtained was found to be almost exclusively the starting acid (7 g). The water solution was then acidified further and extracted several times with ether, which on drying and evaporation yielded the required hydroxy acid. Crude Yield: 45 g (70.0$) m.p. 148 - 150 . The product was reduced without any purification. 115. FLOW SHEET 2 P R E P A R A T I O N OF L I G A N D S B u L i + HN A + A v Li (N ) Y r .GOOH v C H O H CKjCOOH C H 2 0 ,COO LI AID, . O H P B r ELIMINATION OCCURS .OH p - to luene s ulphony I ch lor jde SOCL OTs OTs RSH - C l RSM H Y D R O L Y S I S OCCURS Cl S R S R R=CH 3 ( 2 4 ) , R = / ( 2 5 ) , R = C H 2 j / ( 2 6 ) 116, Preparation of 1 , 1-dideutero-2-_t-butyl-1 ,3-propane diol (96) The acid alcohol prepared above (20.0 g) in T.H.P. (250 ml) was added dropwise to a solution of lithium aluminum deuteride (7.05 g) in T.H.P. (200 ml) at such a rate as to maintain gentle reflux. After addition was complete the reflux was maintained for a further 3 hours. After cooling, water (7 ml), sodium hydroxide (7 ml 15$) and water (21 ml) were added to the stirred reaction mixture. The resulting gelatinous precipitate was f i l t e r e d off and dissolved in dilute HCl which was then extracted with ether. The T.H.F. and ether layers were evaporated and the resulting o i l dissolved in ether and dried over sodium sulphate. On re-moving the ether a white solid was obtained which was recrystallized from ether petroleum ether to give transparent crystals. Yield: -15 g (82$) m.p. 56 - 58°C. " [.. Preparation of 1 ,3-dichloro-2-t,-butylpropane (96) The diol (5.8 g) was dissolved in pyridine (6.9 g) and thionyl chloride (10.5 g) was added over a period of five minutes. Fumes appeared i n i t i a l l y on addition of each drop, and a white precipitate formed in an amber coloured solution. After the addition the flask was heated slowly and at 80° sulphur dioxide began to evolve, The heating was continued between 85 - 90° u n t i l evolution ceased (about one hour). The water layer was further washed with ether and the combined ether layers were then washed with acidified water followed by 10$ sodium carbonate. The ether layer was then dried over sodium sulphate. Evaporation of the ether gave the product. Yield: 6.63 g (89$). Attempted preparation of 1,5-disubstituted-2-_t-butyl-1,5-dithiapentane Potassium hydroxide (l g) was dissolved in ethanol (50 ml) and the solution stirred at 0°C.(ice salt bath). Methane thiol (1 g) was added. The dichloride (1 g) was added in ethanol (10 ml) at 25°C with vigorous s t i r r i n g over about 10 minutes. No precipitate appeared as would be expected, and t . l . c . and n.m.r. indicated that hydrolysis had occurred to the d i o l . The 117. above reaction was repeated using sodium dispersion (1 g) to make the sodium salt of methane t h i o l . Subsequent reaction with the dichloride again yielded the d i o l . Reaction with phenyl thiol was also unsuccessful. Attempted preparation of 1 ,3-dibromo-2-ti-butylpropane. The diol (1.32 g) was reacted in benzene (20 ml) with PBr^ (2.71 g) and refluxed overnight (6 products) for 1-g- hours (4 products) and in an ice bath (0°C) for 2 hours (3 products). It appeared that monobromination, dibromination and elimination products were being formed. Preparation of 1 ,1-dideutero-2-_t-butyl-1 ,3 propaneditosylate (98) The diol (1 g) was dissolved in pyridine (20 ml) and stirred vigorously at 0°. p-Toluenesulphonyl chloride (3' g) was then added in pyridine (20 ml) and the mixture stirred overnight. The reaction mixture was then treated with 7N sulphuric acid (15 ml) and extracted with ether. T.l.c. ( s i l i c a plate in CHCl^) showed two compounds were present, i.e. the mono and di-tosylates. The reaction was repeated and after another 24 hours the reaction mixture was worked up as before and the o i l obtained showed one spot on t . l . c . and sol i d i f i e d . The product was recrystallized from methane to yield white cry s t a l s / Y i e l d : 2.3 g (68$) 1 ,1-dideutero-2-_t-butyl-1 ,5-diphenyl-1 ,5-dithiapentane (25) Phenylthiol (0.3 g) was dissolved in an alcoholic ethanol solution (0.16 g KOH in 50 ml ethanol) and the ditosylate added (0.55 g). The mixture was refluxed for three hours and then concentrated, diluted with water and extracted with ether. The ether extract was dried over sodium sulphate and concentrated to give a yellow o i l (.31 g 66$) which was shown by n.m.r. to be the required product. 118. 1,1-dideutero-2-t-butyl-1,5-dimethy1-1,5-dithiapentane (24) This was obtained as" for the benzyl and phenyl analogs except for the added precaution of adding methane diol (.65 g) to an ice cold solution of potassium hydroxide (1 g) in ethanol (100 ml). The ditosylate (3 g) was then added and the mixture refluxed and worked up as before to yield a yellow o i l . Yield: 4 g (78$) 1 ,1-dideutero-2-_t-butyl-1 ,5-dibenzyl-1 ,5-dithiapentane (26) This was obtained as for the diphenyl analog using toluenethiol (1.7 g), ditosylate (3 g), potassium hydroxide (1 g) and ethanol (100 ml) to yield a yellow o i l . Yield: 2.3 g (71$). 119, COMPLEXES The tungsten complexes were prepared [after unsuccessful attempts to prepare them by mixing and heating the reactants neat (74), refluxing in methylcyclohexane (100) and heating in benzene in a sealed tube (96) ] , by irradiation (80, 81 ) in degassed n-hexane for 12" hours using a slight excess of ligand. W(C0) G LIGAND 1.5 g. | 0 . 7 5 s (20) (20a) . Complex obtained (CO). ¥ S(CH_) CH0 CH_ CD„ S (CH-) (41) 4 3 2 2 2 3 , , . m.p. 123° (decomp.) Anal, Calcd for C ^ QD^WO^: C, 25-0; H, 2.80. Found: • C, 24-5; H, 2.57 Infra-red carbonyl stretching frequencies in dichloromethane 2015 (w), 1895 (s), 1850 (s) cm - 1. 1.5 g . 1 ' 3 9 g ( 2 1 ) complex obtained (C0) 4 W S(<j)) CH2 CH2 CD2 S (<|>) (42) m.p. 135° (decomp) Anal. Calcd for C i g H u D 2 S 2 ¥ 0 4 : C, 41.01; H, 2.88 Found: C, 40.90; H, 3-10 Infra-red (cm - 1) 2010 (w), 1895 (s), 1850 (s) 1"° s . 1.03 (22) complex obtained (CO) V S(CH2$) CH2 CH2 CD2 ^(CH^) (43) m.p. 140-141°(decomp) Anal. Calcd for C„,H, QD oS o¥0,: C, 43-15; H, 3-4 i i i l o d d. 4 Found: C, 42.26; H, 3-59 ' Infra-red (cm - 1) 2030 (w), 1912 (s), 1903 (s), 1862 (s) 120. W(C0)g ' ' LIGAND 0.75 g 0.41 (24) I ) 4 WS(CH3) CH2 CH[C(CH 3 / 3 ( ( complex obtained (CO  (CH H H[C(CH ) ] CD S CH (44) ;,.', m.p. 105° (decomp) Anal.' Calcd. f o r C j ^ g D ^ W O ^ C, 31.8; H, 3-67 Pound; C, 31.72; H, 3.92 Infra-red (cm - 1) 2020 (w), 1900 (s), 1858 ( s ) , 1 0.66 g 0.6 g (25) complex obtained (C0) 4 WS($) CH~2 CH[C(CH5) ] CD2 S((j)) (45) m.p. 119-120° Anal. Calcd. for C^H^D^WC^: C, 44-95; H, 3-58 Pound: C, 44-48; H, 3-98 Infra-red (cm - 1) 2015 (w), 1895 (s), 1885 (s), 1860 (s). The reaction mixture after irradiation varied in colour from pale-yellow or green to dark brown. The reaction mixture was concentrated. Any crystals in the reaction vessel were washed out with toluene and the product recrystallized several times from toluene/hexane. Only CIL^ S CH2CH2CH2(CH3)S W(C0)^ was able to be sublimed (l35°/-05 mm), severe decomposition occurring in a l l other cases. The products were obtained as yellow crystals, that decomposed when heated and decomposed very rapidly in chloroform, (about 15 minutes), acetone (45 minutes) and benzene (12 hours) 121 . • MANGANESE COMPLEXES I 1 BrMn(CO)5S(CH)5CH2CH2CD2S Me (51 ) {r . Irradiation with ultraviolet light, refluxing i n benzene and s t i r r i n g at room temperature in benzene were unsuccessful. In the f i r s t two cases the reaction mixture contained many products, indicating the reaction had proceeded too far and in the second case there was s t i l l a large amount of manganese pentacarbonyl bromide unreacted. The reaction was carried out as follows:-Manganese pentacarbonyl bromide (0.5 g) and 1,1 dideutero-2,5,dithiahexane (20) (0.25 g) were stirred and heated in benzene at 50° unti l the required (measured) amount (88 ml at 22 c) of carbon monoxide had been evolved. The solvent was then removed and the resulting solid sublimed at 40 - 70°(0.05 mm) for several hours. After the unreacted starting material [(Mn(CO)^Br sublimes at'50-60°/l mm](l00) had a l l sublimed the product was sublimed (150 /0.05 mm) and obtained as orange-yellow crystals. Yield: .35 g (55$) m.p. 112-113°. Anal. Calcd. for C _D0S JfaO^Br: C, 27.0; H, 3-70 o 1U 2 2 3 Pound: C, 27-9; H, 3-38 Infra-red (cm - 1) 2023 (w), 1950 (s), 1915 (s). The corresponding _t-butyl complex (52) was prepared in a similar fashion, Mn(C0)3Br (0.5 g), ligand (24) (0.353 g) in benzene (50 ml). The product in this case sublimed at 155 -160 /0.05 mm as yellow crystals. Yield: .38 g (51$) m.p. 152°. Anal. Calcd. for ^ D^MnOjBr: C, 34-9; H, 4-36 Found: C, 34.68; H, 4.60 Infra-red (cm - 1) 2025 (w), 1949 (s), 1920 (s). 122. Pt COMPLEXES In a l l cases Pt Cl2(P<|>^)2 (0.395 mg) was reacted in anhydrous ether (lOO ml) and triethylamine (1 g) with a slight excess of ligand. The mixture was stirred overnight and the product recrystallized from chloroform to yield yellow crystals in a l l cases, Due to d i f f i c u l t y in purifying these compounds (due to starting material) only one was purified to the point of purity by elemental analysis (this took ca. 10 crystallizations). It was found that, any thiol ligand with a hydroxyl group present would not react to give a crystalline derivative, but instead was an o i l , i.e. a mixture of compounds. 1,3 propane di t h i o l (27) 55 mg complex obtained p t s CH2CH2CR"2 S' m.p. 255° Anal. Calcd. C, 56.7; H, 4-4 Found: C, 55-63; H, 4-59 1,1-dideutero -1,3-propanedithiol (28) 55 mg complex obtained (p§^2 P t S C H 2 C H 2 C r i 2 S ^ 5 6 ^ m.p. 255° Anal. Calcd. C, 56.6; H, 4-1 Found: C, 54-59; H, 4-52 1,1,3,-dideutero-1,3-propanedithiol* (29) 55 mg complex obtained ( P ^ ) 2 pTT^H^CE^CD^S1 (57) m.p. 255° Anal. Calcd. C, 56.4; H, 3-9 Found: C, 51-02; H, 4-29 * Kindly donated by Dr. J. R. Campbell, formerly of Dept. of Chemistry, U.B.C. 123. 2-phenyl-1,3-propanedithiol* (30a) 92 mg complex obtained ( P ^ ^ F t S C H 2 C H W C H2 S' (58a) m.p. 255° Anal. Calcd. C, 59-9; H, 4-4 Found: C, 59-60: H, 4-45 1,1 dideutero-2-phenyl-1 ,3-propanedithiol*(30) 92 mg complex obtained ( P ^ p Pt S CH2CH(())) CD2S' (58) 1-methyl-1,3-propanedithiol* (31a) 62 mg complex obtained ( P ^ ^ 'Pt S CH2CH2CH(CH3) S (59a) 1-methyl-1,2,2-trideutero-1,3-propanedithiol*(31) 62 mg complex obtained (P<]>5)2 P't S CH2CD2CD(CH3) S' (59) 1,2-ethanedithiol (32) 47 mg complex obtained ( P c i ^ Pt S CH^ CH^ S' (60) m.p. 240° Anal. Calcd. C, 56.2; H, 4-1 Found: C, 54-9; H, 4-45 1-methyl-1,2-ethanedithiol (33) 54 mg complex obtained ( ? § ^ 2 P t s CE^CHtCH^) S' (61 ) 1-ethyl-1,2-ethanedithiol (34) 61 mg complex obtained ( F ^ ) 2 Pt S CH2CH(CH2CH3) S' (62) m.p. 250° Anal. Calcd. C, 57.2; H, 4-5 Found: , C, 57-00; H, 4.65 Toluenethiol (35) 124 mg complex obtained (P<j>5)2 Pt (S CH2(J))'2 (63) * Kindly donated by. Dr. J. R. Campbell, formerly of Dept. of Chemistry, U.B.C. 124. FLOW SHEET 3 PRIMARY CARBOHYDRATE TOSYLATES R = OTs ( 9 ) R = I (10) 125. FLOW SHEET 4 SECONDARY TOSYLATES AND EPOXIDES 126. FLOW SHEET 5 L IGAN DS -SR SR •SR '2 R - Me ( 2 0 ) R=rf ( 2 1 ) R=CKyzT ( 2 2 ) SR R = C H 3 (24) R = ^ ( 2 5 ) R = C H 2 J ^ (26) NON-DEUTERATED LIGANDS ARE DEPICTED BY THE SUBSCRIPT (a) IN THE TEXT •SH -SH D SH H -SH -SH (27 ) •SH -SH (30 ) •SH •SH ( 3 2 ) (28 ) SH •SH CH_ (33) C H 3 D Et (29 ) •SH -SH (31) •SH •SH 0 C H SH ' 2 (34) (35) 127 . FLOW SHEET 6 MANGANESE AND GROUP VI COMPLEXES R - - C H 3 ( 4 1 ) R = J * ( 4 8 ) R = C H ^ ( 43 ) C O R=CH g ( 4 4 ) , R=jrf (51) CO NON -DEUTERATED ANALOGS ARE DEPICTED BY THE SUBSCRIPT (a) IN F L O W SHEET 7 PLANAR PLATINUM COMPLEXES P t ( P0 ) 3 2 (55) R t = R = H (58) R 1 = H 1 ( R 2 (57) R 2 =D (6 0) (61) f62) H 3 ( ^ 3 P ) 2 P H P t ( p r f ) , 3'2 / C H S ^ C H S 129. REFERENCES 130. 1. G, N. SCHRAUZER. Acc. Chem. Res. 1, 98 (1968). •2. D. CR0WF00T-H0DGKIN. Proc. Roy. Soc.(London) A288, 294 (1965). 3. L. D. HALL and P. R. STEINER (to be published). 4. G. R. INGLIS, J.' P. SCHWARZ and L. McLAREN. J. Chem. Soc. 1014 (1962). 5- D. HORTON, J. M. TARELLI and J. D. WANDER. Carb. Res. 23, 440 (1972), 6. P. R. STEINER. PhD. Thesis. Dept. of Chemistry, U.B.C. (1972). '7. A. ROSENTHAL. Adv. Carb. Chem. 23, 59 (1968). '8. M./ORCHIN,' L. KIRCH and J. GOLDFARB. J. Amer. Chem. Soc. 78, 5450 (1956). 9. R.-F. HECK and D. S. BRESLOW. J. Amer. Chem. Soc. 83, 4023 (1961). 10. B. .L. VALLEE and R. J. P. WILLIAMS. Chemistry in Britain, 4, 397 (1968). 11. R.'-J.-P. WILLIAMS. Quart. Rev. 2±, 331 (1970). 12'.: J r A," RENDLEMAN ( j r ) . Adv. Carb. Chem. 2±, 209 (1966). 13- R/E. REEVES. Adv. Carb. Chem. 23_, 59 (1968). .14. S . J . ANGYAL and K. P. DAVIES. Chem Comm. 500 (1971). 15- S. J. ANGYAL. Aust. J. Chem. 25_, 1957 (1972) 16. S. J. ANGYAL and M. E. EVANS. . Carb. Res. 23, (1972). 17. A. W. JOHNSON, L. MBRVYN, N. SHAW and E. SMITH. J. Chem. Soc. 4146 (1963). 18. K. BERNHAUER, 0. MTJLLER and F. WAGNER. Angew.Chem.(int.Ed.Engl.) 3, 200 (1964). 19. D. DOLPHIN, A. W. JOHNSON and R. RODRIGO. Ann. N.Y. Acad. Sci. 1_T2, 590 (1964) -20. G. N. SCHRAUZER and R. J. HOLLAND. J. Amer. Chem. Soc. 21, 4060 (1971 ) . 21. G-. N. SCHRAUZER and E. DEUTSCH. J. Amer. Chem. Soc. 2 1 , 3341 (1969) -22. G. N. SCHRAUZER, E. DEUTSCH and R. WINDGASSEN. J.Amer.Chem.Soc. 9^ , 2441 0968). 23. F. R. JENSEN, V. MADAM and D. H. BUCHANAN J.Amer.Chem.Soc. <%L, 1414 (1970). 24. R. BONNET. Chem. Rev. 6J5, 573 (1963) . 25. G. N. SCHRAUZER and R. J. WINDGASSEN. J.Amer .Chem.Soc. §2, 1999 0967) • 131 . 26. P. CALDERAZZO and C. FLORIANI. Chem. Comm. 137 (19&7). 27. G. COSTA and G. MESTRONI. J. Orgmet. Chem. 1_1_, 325, (1968), 28. T. S. PIPER and G. WILKINSON. J. Inorg. Nucl. Chem. 3.' 1 °4 0956). 29. R. B. KING. Acc. of Chem. Res. 3_, 417 (1970). 30. R. B. KING. Adv. Organomet. Chem. 2, 157 (1964). 31. G. W. PARSHALL and J. J. MROWCA. Adv. Organomet. Chem. 7, 157 (1968), 32. R. F. HECK. Adv. Organomet. Chem. 4, 243 (1966). 33. M. I. BRUCE. Adv. Organomet Chem. 10, 274 (1972). 34. E. W. ABEL, A. SINGH and G. WILKINSON. J. Chem. Soc. 1321 (i960). 35. R. D. CLOSSON, J. KOZIKAVOKI and T. H. COFFIELD. J.Org.Chem. 22, 598 (1957). 36. M. L. H. GREEN and P. L. I. NAGY. J. Organomet Chem. 1, 58, (1963), 37. R. B. KING and M. B. BISNETTE. J. Organomet. Chem. 2, 15 (1964). 38..R. B. KING and M. B. BISNETTE. J. Organomet. Chem. 2, 38 (1964) . . 39. R. B. KING, S. L. STAFFORD, P. M. ZREICHEL and F. G. A. STONE. J. Amer. Chem. Soc. 83_, 3604 (1961 ) . 40. D. W. McBRIDE, E. DUDEK and F. G. A. STONE. J. Amer. Chem. Soc. 1752 (1964), 41. W. R. McCLELLAN. J. Amer. Chem. Soc. 83, 1598 (1961). 42. M. L. H. GREEN and P. L. I. NAGY. J. Chem. Soc. 189 (1963). 43. R. F. HECK. J. Amer. Chem. Soc. 85_, 1460 (1963). 44. G. N. SCHRAUZER and R. J. WINDGASSEN. J; Amer. Chem. Soc. 89_, 143 (1967). 45. W. P. GIERING, M. ROSENBLUM and J. TANCREDE. J. Amer. Chem. Soc. 94, 7172 (1972)^ 46. R. E. DESSY, R. L. POHL and R. B. KING. J. Amer. Chem. Soc. 88, 5121 (1966), 47. W. HIEBER, 0. VOHLER and G. BRAUN. Z. Naturforsch. J J b . , 192, (1958), 48. M. L. H. GREEN, N. ISHAQ and T. MOLE. Z. Naturforsch. 20b, 598, (1965). 49. D. H. BALL and P. W. PARRISH. Adv. Carb. Chem. 24, 139 (1969), 50. N. P. TAYLOR and P. W. KENT. J. Chem. Soc. 872 (1958), 132. 51. J- H. WESTWOOD, R. C. CHALK, D. H. BELL and L. LONG (Jr) J.Org.Chera. 22, 1643 (1967 52. R. S. TIPSON. Adv. Carb. Chem. 8, 107 (1953), 53. J. COX and L. N. OWEN. J. Chem. Soc. ( c ) 1121 (l967). 54. W. A. SZANEK and J. K. N. JONES. Can. J. Chem. 43, 4223 (1965)-55- A. C. RICHARDSON. Carb. Res. 10, 395, (1969). 56. T. D. INCH. Ann. Rev. of n.m.r. Spectroscopy. 2, 35 (1969). 57. L. D. HALL, P. R. STEINER and C. PEDERSON. Can. J. Chem. 48, 1155 (1970), 58. R. E. WINTERS and P. W. KISER. Inorg. Chem. 3., 699 (1964). 59- M. I. BRUCE! Inorg. Nucl. Chem. Letters. 3, 157 (1967). 60. M. L. H. GREEN, C. N. STREET and G. WILKINSON. Z. Naturforsch Mb, 738 (1959) 61. J. M. BURLITCH and S. W. ULMER. J. Organomet. Chem. 19_, P21 (1969). 62. W. A. BONNER. Adv. Carb. Chem. 6, 251 (1951 ) . 63. R. G. PEARSON. Science. V§1_, 172 (1966), 64. R. G. PEARSON. J. Amer. Chem. Soc. 85., 3533 (1963). 65. S. AHRLAND, J. CHATT and N. R. DAVIES. Quart. Rev. V2, 265 (1958), 66. D. P. RILLEMA, W. J. REAGAN and C. H. BRUBAKER Jr. Inorg. Chem. 8, 587 (1969). 67. H. R. BUYS. Rec. Trav. Chim. 89_, 1244 (1970). 68. M. J. 0. ANTEUNIS. Conformational Analysis 31, (1971 ) : 69. F. G. RIDDELL. Quart. Rev. 21., 365 (1967). 70. R. U. LEMIEUX, J. 5. STEVENS and R. R. FRASER. Can. J. Chem. 40, 1955 (1962)r 71. E. L. ELIEL and M. C. KNOEBER. J. Amer. Chem. Soc. $0, 3444 (1968). 72. H. J. KALFF and E. HAVINGA. Rec. Trav. Chim. 85, 467 (1966). 73. C. ROMERS, C. ALTONA, H. R. BUYS and E. HAVINGA. Topics in Stereochem. 4, 39 (196°) 74. H. G. E. MANNERSKANTZ and G. WILKINSON. J. Chem. Soc. 4454 (1962), 75. S. E. LIVINGSTONE. Quart. Rev. 12, 386 (1965). 133. 76. L. F. LINDOY. Co-ord. Chem. Rev. 4, 41 (1969)-77. C. K. JORGENSEN. Inorg. Chim. Acta. Rev. 2, 65 (1968). 78. E. W. ABEL. Organomet. Chem. Rev. 2, 443 (1967). 79. R. DOBSON and L. W. HOUK. Inorg. Chim. Acta. 2, 287 (196?). 80. G. R. DOBSON. Inorg. Chem. 8, 90 (1969). 81. G. R. DOBSON and G. C. FABER. Inorg. Chim. Acta. 4, 87 (1970), 82. R. 0. HUTCHINS and B. E. MARYANOFF. J. Amer. Chem. Soc. 21, 3266 (1972), 83. L. D. HALL, J. R. CAMPBELL, R. B. MALCOLM and B. DONALDSON (to be published). 84. J. R. GOLLOGLY and C. J. HAWKINS. Inorg. Chem. H , 156 (1972). 85. C. J. HAWKINS. "Absolute Configuration of Metal Complexes". Wiley Interscience N.-.T. (1971). 86. R. J. GEUE and M. R. SNOW. J. Chem. Soc. (A) 2981 (1971 ), 87. K. MATSUMOTO, S. 001 and H. KUROYO. Bull. Chem. Soc. Japan. 4_3_, 1903 (1970). 88. T. NOMURA, F. MARUMO and Y. SAITO. Bull. Chem. Soc. Japan. £2, 1016 (1969)-89. A. KOBAYASHI, F. MARUMO, Y. SAITO, J. FUJITA and F. MIZUHAMI. Inorg. Nucl. Chem. Letters. 2 » 777 (1971 ). 90. T. G. APPLETON and J. R. HALL. Inorg. Chem. <£, T 8 0? (1970), 91. T. G. APPLETON and J. R. HALL. Inorg. Chem. 10, 1717 (1971). 92. T. G. APPLETON and J. R. HALL. Inorg. Chem. jM, 112 (1972). 93. T. G. APPLETON and J. R. HALL. Inorg. Chem. Vi., 117 0 972). 94. T. G. APPLETON and J. R. HALL. Inorg. Chem. 1J_, 124 (1972), 95- I. R. JONASSON, S. F. LINCOLN and D. R.' STRANKS. Aust. J. Chem. 23_, 2267 (1970) 96. W. R. CULLEN, L. D. HALL, J. T. PRICE and G. SPENDJIAN. To be published. 97. W. R. CULLEN, L. D. HALL, J. T. PRICE and G. SPENDJIAN. To be published. 98. J. R. CAMPBELL. Private communication. 99. T. DURST. Zet. Lett. 4171 (1971)« 100. R. B. KING and J. J. EISCH. Organomet Syntheses J_, (1 965), 134. 101. J. B. LAMBERT. J. Amer. Chem. Soc. 8g_, 1836 (1967). 102. H. R. BUYS. Rec. Trav. Chim. 88, 1003 (1969)-103. M. KARPLUS. . J. Amer. Chem. Soc. 85_, 2870 (1963). 104. M. ARONEY and R. J. W. Le FEVRE. J. Chem. Soc. 2161 (i960). 105. R. 0. HUTCHINS, L. D. KOPP and E. L. ELIEL. J. Amer. Chem. Soc. £0, 7174 (1968)„ 106. C. A. BEAR and J. TROTTER. To be published. 107. E. W. ABEL, R. P. BUSH, F. J. HOPTON and C. R. JENKINS. Chem. Commun. 58 (1966). 108. J. A. POPLE, W. G. SCHNEIDER and H. J. BERNSTEIN. High Resolution N.M.R. McGraw-Hill, N.Y. (1959) -109. H. R. BUYS. Rec. Trav. Chim. 89_, 1253 (1970), 110. E. W. ABEL and. G. WILKINSON. J.,'Chem. Soc. 1501 (1959)-111. P. HAAKE and P. C. TURLEY. J. Amer. Chem. Soc. 8°,, 4611 (1967),, 112. P. HAAKE and P. C. TURLEY. J. Amer. Chem. Soc. 89_, 4617 (1967), 113. M. A. THOMAS. Ann. Rev. N.M.R. Spec. 1_, 43 (1968). 114. D. F. SHRIVER. The manipulation of Air Sensitive Compounds. McGraw-Hill (1969), 115- 0. Th. SCHMIDT, A. DISTELMAIER and H. REINHARD. Chem. Ber. 86, 741 (1953). 116. L. HOUGH, J. K. N. JONES and M. S. MAGSON. J. Chem. Soc. 1525 (1952). 117. 0. Th. SCHMIDT. Methods in Carb. Chem. 2, 324, (1963). Ed. R.L.Whistler and M. L. Wolfram. (Acad. Press) . 118. R. S. TIPSON. Methods in Carb. Chem. 2, 253 (1963). Ed. R. L. Whistler and M. L. Wolfram. (Acad. Press), 119. A. L. RAYMOND and E. F. SCHROEDER. J. Amer. Chem. Soc. 20, 2785 (1948)„ 120. D. DOLPHIN. Methods Enzymol. 18, (Pt.c) 34 (1971 ), 121. R. J. ANGELICI. Synthesis and Techniques in Inorganic Chemistry. 136 (1969) W. B. Saunders Co. (1969)_ 122. A. DAVISON, J. A. McCLEVERTY and G. WILKINSON. J. Chem. Soc. 1133 (1963), 123. J. M. OTTOWAY, D. T. COKER, W. B. ROWSTON and D. R. BHATTARAI. Analyst 21, 567 (1970). 124. N. K. KOCHETKOV and 0. S. CHEZHOV. Adv. Carb. Chem. 21., 39 (1966), 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items