UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The synthesis and structural studies of tin(IV) and organotin(V) sulfonates Yeats, Philip Allen 1973

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

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata


831-UBC_1973_A1 Y42_8.pdf [ 8.58MB ]
JSON: 831-1.0060221.json
JSON-LD: 831-1.0060221-ld.json
RDF/XML (Pretty): 831-1.0060221-rdf.xml
RDF/JSON: 831-1.0060221-rdf.json
Turtle: 831-1.0060221-turtle.txt
N-Triples: 831-1.0060221-rdf-ntriples.txt
Original Record: 831-1.0060221-source.json
Full Text

Full Text

171 V ? THE SYNTHESIS AND STRUCTURAL STUDIES OF TIN(IV) AND ORGANOTIN(IV) SULFONATES BY PHILIP A YEATS B.Sc. (Hons.), University of British Columbia, 1969 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in the Department of CHEMISTRY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA July, 1973 In presenting t h i s t h e s i s i n p a r t i a l f u l f i l m e n t of the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree that the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r reference and study. I f u r t h e r agree that permission f o r extensive copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the Head of my Department or by h i s r e p r e s e n t a t i v e s . I t i s understood that copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l gain s h a l l not be allowed without my w r i t t e n permission. Department The U n i v e r s i t y of B r i t i s h Columbia Vancouver 8, Canada - i i -ABSTRACT The systematic investigation of the solvolysis of methyltin(IV) chlorides of the type (CH„) SnCl, , 1 < n < 4, in strong monobasic 3 n 4-n protonic acids, in particular HSO^F and HSO^CF^, has resulted i n the preparation of new trimethyltin(IV), dimethyltin(IV) and methyl-chlorotin(IV) sulfonates. Preferential cleavage of Sn-Cl over Sn-C bonds was noted resulting in mono- or bisubstitution on tin depending on the mole ratio of the reactants. Additional methylchlorotin(IV) fluorosulfates could be obtained by non-statistical ligand redistribution reactions with chloride and fluorosulfate exchange between suitable substrates, e.g. (CH 3) 2Sn(S0 3F) 2 + (CH 3) 2SnCl 2 • 2(CH ) 2ClSnS0 3F Inorganic fluorosulfates Cl 2Sn(S0 3F) 2 and Br 2Sn(S0 3F) 2 were obtained by similar ligand scrambling of either SnCl^ or SnBr^ in Sn(S0 3F)^. Sn(S0 3F)^ was found to undergo complete SC>3F-C1 exchange with TiCl^ resulting in the formation of Ti 3Cl^g(S0 3F) 2 which appears to have rather unique tetradentate SC^F groups. Complexation reactions of the Lewis acid Sn(S0 3F) 4 with CIC^SC^F, I(SC>3F)3 and Br(SC"3F)3 2- + + result in the formation of Sn(SO.F)^ salts with C10„ , I<S0-F)o and Br(SC> 3F) 2 + cations. Nitrosonium and a l k a l i metal hexakisfluorosulfato-stannates are obtained by the reaction M 2 S n C 1 6 + 3 S2°6 F2 M 2 S n ( S ° 3 F ) 6 - i i i -The reactions of the Lewis acids AsF c and SbF_ with some of the halogen fluorosulfates mentioned above is also discussed. 119 Structural studies are based on vibrational and Sn Mossbauer spectra with the latter providing proof of the identity of the methyl-2-chlorotin(IV) fluorosulfates and the Sn(SO^F)^ ion. 2-Except for the Sn(S0„F), compounds, the t i n is either hexa-J o coordinated or occasionally pentacoordinated with the sulfonate groups acting as bidentate bridging links between the t i n moieties. Two structural types emerge: the octahedral bisfluorosulfates XYSnCSO^F^ with X, Y = CH^, Br, CI, and SO^F in trans octahedral positions and the trigonal bipyramidal X2YSnS0.jF. Successful correlations of isomer shifts vs. the sum of the Pauling electronegativities of X and Y or quadrupole splitting vs. the Taft a for X and Y indicate isostructural compounds for the bisfluorosulfates. The monofluorosulfates allow no 2-such correlations. In Sn(SO.F), monodentate SO F groups give rise J D -> to an octahedral environment for t i n and hence a zero quadrupole splitting. - i v -TABLE OF CONTENTS Page ABSTRACT 1 1 LIST OF TABLES i x LIST OF FIGURES x i i ACKNOWLEDGEMENT x i v I. INTRODUCTION 1 A. General Remarks 1 B. Stereochemistry 2 1. Complexes 3 2. Autocomplexation 4 (a) Bridging ligands 4 (b) Chelating ligands 6 (c) Carboxylates and thiocarbamates 6 3. Clusters 8 C. Bonding 9 1. General considerations 9 2. Methyltin cations 11 3. Fluorosulfates as anion 12 D. Fluorosulf ates 13 1. Synthetic routes to fluorosulfates 13 2. Structural studies on fluorosulfates 15 3. Previous work on ti n fluorosulf ates 16 E. Vibrational Spectroscopy 17 1. Tin-carbon 17 2. Fluorosulf ates 17 3. Other sulfonates 21 - v -Page F. Mossbauer Spectroscopy 23 1. Principles 24 2. Isomer shift 26 3. Quadrupole splitting 28 4. Room temperature effect 29 G. HSO^F as a Nonaqueous Solvent 30 II. EXPERIMENTAL 34 A. Apparatus 34 1. Drybox 34 2. Vacuum lines 34 3. Reaction vessels 35 (a) Glass 35 (b) Metal 38 4. Fluorosulfuric acid 38 5. Conductometry 38 6. Vibrational spectroscopy 39 7. Mossbauer spectroscopy 40 8. Nuclear magnetic resonance spectroscopy 40 9. Melting points 41 10. Analyses 41 B. Chemicals 41 1. Commercial sources 41 2. Literature preparations 41 C. Suppliers of Materials and Equipment 44 - v i -Page III. METHYLTIN SULFONATES 46 A. Introduction 46 1. Previous work 46 2. Promising routes to sulfonates 47 (a) Solvolysis 47 (b) Ligand redistribution 48 B. Preparations 49 1. (CH 3) 2Sn(S0 3X) 2 49 2. (CH3>2Sn derivatives of oxyacids 52 3. CH 3ClSn(S0 3X) 2 52 4. (CH 3) 3SnS0 3X 54 5. (CH 3) 2ClSnS0 3F and CH^Cl^nSC^F .- 57 6. (CH 3) 2Sn(S0 3F)(S0 3CF 3) 58 7. CH 3Sn(S0 3F) 3 58 C. Structural Studies 59 1. (CH 3) 2Sn(S0 3F) 2 59 2. Other methyltin sulfonates 67 (a) (CH 3) 2Sn(S0 3X) 2 67 (b) (CH 3) 3SnS0 3X 70 (c) CH 3ClSn(S0 3X) 2, CH^Cl^nSO^ and (CH 3) 2ClSnS0 3F 77 D. Discussion 79 1. Syntheses 79 2. Spectra 82 3. Bonding 87 - v i i -Page IV. INORGANIC TIN FLUOROSULFATES 95 A. Introduction 95 B. Preparations 96 1. Sn(S0 3F) 4 96 2. Cl 2Sn(S0 3F) 2 97 3. Cl 3SnS0 3F 99 4. Br 2Sn(S0 3F) 2 100 C. Structural Studies 101 1. Mo'ssbauer spectra 101 2. Vibrational spectra 103 D. Reactions of Sn(S0 3F) 4 105 1. Introduction 105 2. Ligand redistribution reactions 106 (a) GeCl. and SiCl. 106 4 4 (b) T i C l 4 107 3. Complexation reactions 110 (a) NOCl I l l (b) N02F and C1F3 112 (c) Adduct formation 112 V. CORRELATIONS OF MOSSBAUER PARAMETERS FOR TIN FLUOROSULFATES 114 A. Correlations for Tin Fluorosulf ates 114 B. Other Possible Correlations 119 C. Point Charge Model 120 - v i i i -Page VI. THE HEXAKISFLUOROSULFATO STANNATE(IV) ION 123 A. Introduction 123 B. Preparations 124 1. (C10 2) 2Sn(S0 3F) 6 124 2. [Br(S0 3F) 2] 2Sn(S0 3F) 6 125 3. [I(S0 3F) 2] 2Sn(S0 3F) 6 126 4. K 2Sn(S0 3F) 6, Cs 2Sn(S0 3F) 6 and (NO) 2Sn(S0 3F) g . 126 C. Mossbauer Spectra 129 D. Vibrational Spectra 132 E. Solution Studies in HS03F 142 VII. REACTIONS OF LEWIS ACIDS WITH FLUOROSULFATES 147 A. Introduction 147 B. C102S03F Reactions 148 1. C10„S0_F + SnF. 148 2 3 4 2. C102S03F + SbF5 148 3. C102S03F + AsF 5 153 4. C102S03F + BF 3 155 C. KS03F + SbF5 156 D. BrS0 3F and Br(S0 3F) 3. Reactions 157 VIII. GENERAL CONCLUSIONS 159 A. Summary 159 B. Suggestions for Further Work 161 REFERENCES 163 - ix -LIST OF TABLES Table Page 1 Methyltln Halides 9 2 Conformations of Fluorosulfates 19 3 Methyltin Vibrations 19 4 S-0 Stretching Frequencies for Monodentate SO^F ... 19 5 Stretching Frequencies for Bi- and Tridentate S03F. 22 6 Previous Assignments for SO^CF^ 22 7 Trimethyltin Carboxylates 31 8 Physical Properties of HS03F 31 9 Chemicals 42 10 Dimethyltin Sulfonates - Analyses 51 11 Elemental Analyses of (CH^Sn Salts 53 12 Other Methyltin Sulfonates - Analyses 55 13 Reported Interatomic Distances and Selected Bond Angles for some Fluorosulfates .. 61 14 Vibrational Spectra of (CH 3) 2Sn(S0 3F) 2 63 15 Electrical Conductivity of (CH 3> 2Sn(S0 3F) 2 in HS03F at 25° 66 16 Mossbauer Data for (CH 3) 2Sn(S0 3X) 2 66 17 Vibrational Frequencies of (CH 3> 2Sn(S0 3X) 2 Compounds 68 18 Dimethyltin Derivatives of Oxyacids - Spectral Results 69 19 Vibrational Frequencies for (CH^SnSO^ 71 119 20 Sn Mossbauer Data for (CH 3) 3Sn(IV) Compounds at 80°K 74 - x -Table Page 21 Specific Conductance of (CH^SnSC^F and (CH 3) 2Sn(S0 3F) 2 75 22 Vibrational Spectra of (CH.) CI Sn(SO.X). 78 r 3 x y 3 4-x-y 23 Mossbauer Data for (CH„) CI Sn(S0oX). 79 3 x y 3 4-x-y 24 Preparative Routes to (CH„) CI Sn(SO„F), 80 r 3 x y 3 4-x-y 25 Vibrational Modes of the S03F Group in Various Tin and Methyltin Fluorosulfates 84 26 Tin-carbon and Tin-chlorine Stretching Modes in the Methyltin(IV) Chloro-sulfonate Compounds 85 27 Analytical Results for Inorganic Tin Fluorosulfates 98 28 Mossbauer Parameters for Inorganic Tin Fluorosulfates 102 29 Vibrational Spectra of Inorganic Tin Fluorosulfates 102 30 Vibrational Spectra of T i 3 C l 1 ( ) ( S 0 3 F ) 2 and TiCl 2(S0 3F> 2 109 31 Point Charge Predictions for some Methyltin(IV) Compounds 122 2-32 Analytical Results for Sn(S0„F), Compounds 128 o o 2-33 Mossbauer Spectra of Sn(S0„F),. Compounds at 80°K. 130 3 b 34 Vibrational Spectra of Alk a l i Metal Hexakisfluoro-sulf atostannates(IV) 133 35 Vibrational Spectra of Heterocation Hexakisfluoro-sulf atostannates (IV) 134 36 Vibrational Spectra of Halogen Bisfluorosulfates Hexakisfluorosulfatostannates(IV) 135 37 Br(S0 3F) 2 + Vibrational Modes 139 38 Literature Values for vBr-0 and vBr-F 141 - x i -Table Page 39 Br-0 Stretches in Fluorosulfates 141 40 Electrical Conductance of K oSn(S0_F) c and 2. 3 D (C10o)oSn(S0„F) , in HSO.F at 25 0;yValues 143 I I 3 b 3 41 Vibrational Spectra of CIO Sb2F 150 42 Vibrational Spectra of Sb„F ~ and SbF ~ Salts 151 i 11 o 43 Raman Spectrum of ClC^AsF^C^F 154 44 Raman Spectrum of Br(S0 F) SbF 158 - x i i -LIST OF FIGURES Figure Rage 1 Structures of Dimethyltin(IV) Fluoride and Chloride 5 2 Structural Features of some Organotin(IV) o Carboxylates and Dithiocarbamates, Bond Distances in A 7 3 Correlation Diagram for S03F 20 4 Decay of ^gSn^^ m 25 5 Isomer Shift, 6 (mm/sec) 25 6 Distribution of 6 for Sn(IV) 27 7 Quadrupole Coupling A (mm/sec) 27 8 Reactors 37 9 The Crystal Structure of (CH 3) 2Sn(S0 3F) 2 60 10 E l e c t r i c a l Conductivity of (CH 3) 2Sn(S0 3F) 2 in HS03F at 25° 65 i i q 11 Sn Mossbauer Spectrum of (CH 3) 3SnS0 3CH 3 at 80°K. 74 12 Specific Conductivities in HS03F at 25.0°C 76 13 Comparison of Mossbauer Parameters of Dialkyltin(IV) Difluorides Bisdifluorophosphates, Bisfluorosulfates and Bistrifluoromethylsulfonates 83 14 Structural Features of (CH 3) 2Sn(S0 3F) 2 88 15 pTr-dTr Bonding in Fluorosulf ates 90 16 S-0 Bond Lengths and TT Bond Orders 90 17 Structural Features of Trimethyltin fluorosulfate.. 92 18 Correlation Between Isomer Shifts and Pauling Electronegativities of the Ligands X in the Series X 2Sn(S0 3F) 2 117 - x i i i -Figure Page 19 Correlation Between Taft's Inductive Constants and Quadrupole Splitting in the Series X 2Sn(S0 3F> 2 and X 2SnF 2 118 20 Raman Spectrum of [Br(S0 3F) 2] 2Sn(S0 3F) 6 136 21 Conductivities of K„Sn(S0„F)r and (CIO^) Sn(S0 F) . 144 xiv ACKNOWLEDGEMENTS I am very grateful to Dr. F. Aubke for suggesting this topic and for his constant encouragement and advice during the course of this work. I would also like to thank Dr. J.R. Sams for help with respect to the MSssbauer spectra and Mrs. A. Sallos for technical assistance with these spectra. The work of Messers. S. Rak and J. Molnar of the glass blowing shop i s appreciated. I would also like to thank Ms. Diane Johnson for typing the thesis. I am also grateful for a MacMillan Family Foundation Fellowship awarded to me from 1971-1973. - 1 -CHAPTER I INTRODUCTION A. General Remarks In recent years there has been an increasing interest in the chemistry of tin. This increase can be attributed to several factors. These include: 1) the a b i l i t y of tin to form bonds with almost a l l the elements, particularly the f a i r l y stable bonds formed with the more abundant elements oxygen, nitrogen, carbon, hydrogen, and the halogens, to produce an extraordinary wealth of t i n and organotin compounds; 2) the applicability of an unusually wide range of preparative routes to the synthesis of t i n and organotin 119 compounds; 3) the advent of physical techniques such as Sn and to a lesser extent ^^Sn nuclear magnetic resonance spectroscopies 119 and Sn Mossbauer spectroscopy; 4) the fact that tin and organotin compounds are readily available and not overly expensive; as well as 5) the possible industrial application of a wide variety of tin compounds. The increased interest in the area of organotin chemistry in particular i s best documented by a number of books on the subject 12 3 which have appeared . recently in f a i r l y short succession. ' ' 2 -The fact that a data colA^-ion on organotin compounds compiled in 1967^ must now be considered as outdated and the number of papers which have appeared since che work was started on the specific area with which this thesis is concerned are indicative of the interest. As is apparent from the more recent review articles on organotin chemistry the main interest in the chemistry of tin centers around three important problems. These are the stereochemistry and structure 5 6 6 of t i n compounds ' as deduced from X-ray diffraction studies, Mossbauer studies,^'^'^ vibrational spectroscopy,^ and nuclear magnetic resonance spectroscopy,^ the bonding between the t i n atom . . ... 5,11,12 . . . „. . 13,14 and i t s neighbours, and the reactivity of tin compounds. Almost a l l of this interest i s centered around compounds of tetra-valent t i n . Compounds of divalent t i n , especially organometallic derivatives, have received relatively l i t t l e attention,''""' perhaps with the exception of SviF^ which is widely used in fighting tooth decay and has been the subject of many c l i n i c a l studies."^ The general references, monographs, reviews, and data collections mentioned above are a l l useful sources for information on the various aspects outlined here. A number of individual topics relevant to the intention and scope of this study w i l l be discussed in greater detail in the following sections. B. Stereochemistry A f a i r l y wide variety of coordination numbers and geometries have been observed for compounds of tetravalent t i n . The coordination numbers range from four in simple tetrahedral molecules such as - 3 -tetramethyltin or tin tetrachloride to the more common five or six and the rarer seven or even eight coordination. Higher coordination can be achieved in three general ways and some reported structures are given as examples of each type. 1. Complexes The coordination number can be expanded to five or six by complexation with neutral molecules, L, to form molecular complexes or adducts, or with anions, X , to form anionic complexes: SnX, + nL SnX, + nX SnX.(L) 4 n [ S n X ( W n-An example of a pentacoordinated adduct i s the trigonal bipyramidal (CH^^SnCl.py''"^ in which the three methyl groups occupy the three 18 equatorial positions in the bipyramid. SnCl^.2SeOCl2 and 19 SnCl4«2py are both examples of hexacoordinated adducts in which the stereochemistry at tin i s octahedral. The selenium oxychloride adduct has the two SeOC^ molecules cis to one another and the pyridine adduct the pyridines are trans. In a l l instances of SnCl^ adducts either the more common cis, or the trans adducts are formed, but never both isomers for one ligand. Anionic complexes in which the tin i s either five or six coordinate are also known. 3-Chloro-l,2,3,4-20 tetraphenylcyclobutenium pentachlorostannate has a trigonal bipyram-21 idal . SnCl c anion and sodium hexafluorostannate an octahedral SnF - 4 -2- 2- 22 g anion. Studies on mixed anion complexes, e.g. SnCl^Br^ 23 are also reported. Cs^(CH^)^SnCl^ i s an example of an anionic organotin complex shown by infrared results to have trans methyl 2-groups in an octahedral (CH^^SnCl^, anion. 2. Autocomplexation The coordination number of tin can also be expanded by auto-complexation. In autocomplexation a nominally monodentate ligand such as NO^  or F is involved in further coordination, either 2-bridging or chelating. Bidentate ligands such as SO^ can also autocomplex by acting as t r i - or tetradentate ligands. (a) Bridging ligands Examples of hexacoordinated structures with bridging ligands are found for SnF 4, 2 4 ( C H ^ S n F ^ 2 5 and ( C H ^ S n C l ^ 2 6 Both fluorides form sheetlike polymers with fluorine bridges, and in each case the bridging tin-fluorine distances are both equal. As shown in Figure 1, methyl groups in (CH3)2SnF2 occupy trans positions in the octahedron with linear CH.-Sn-CH„ resulting in D,, symmetry (I4/mmm for the J j 4n crystal). (CH^^SnC^, on the other hand, has a distorted tetrahedral o environment around tin with weak chlorine bridges, unequal (1.1 A difference) tin-chlorine bond lengths, and an angular CH^-Sn-CH^ grouping. These two compounds ill u s t r a t e the two possible extremes in the extent of autocomplexation. - 5 -Figure 1. Structures of Dimethyltin (iv) Fluoride  and Chloride (CH^ SnCl 2 Trigonal bipyramidal complexes with bridging ligands are also known. The most common examples of this type of coordination are 27 28 t r i a l k y l t i n compounds such as (CH^SnNCS, (CH^SnCN, or 29 (CH^^SnF, which have planar (CH^^Sn moieties joined together by bidentate NCS, CN or F bridges to form chain-like polymers. Trimethyltin isothiocyanate is definitely an isothiocyanate, i.e. i t has a short tin-nitrogen bond and a relatively longer tin-sulfur bond. In trimethyltin cyanide, on the other hand, the CN group i s found equidistant from the two neighbouring tin atoms. Trimethyltin fluoride, an example of five coordinate structure with a single atom bridge, has a disordered structure with two possible solutions. In - 6 -both po s s i b i l i t i e s the tin-fluorine-tin grouping is non-linear and the two tin-fluorine bond lengths are unequal. (b) Chelating ligands Some examples of complexes with chelating ligands are dimethyltin 30 31 bis 8-hydroxyquinolinate, methyltintrinitrate, and t i n tetra-32 nitrate. (CH^^SnCCgHgNO^ is six coordinate with the nitrogen and oxygen atoms of each 8-hydroxyquinolate occupying adjacent positions on the octahedron around t i n . The methyl groups are also cis. In methyltin trinitrate the t i n atom is seven coordinate with three bidentate chelating nitrate groups and one methyl group and in tin tetranitrate the eight oxygen atoms of the four bidentate chelating nitrate groups form a slightly distorted dodecahedron around the tin atom. (c) Carboxylates and thiocarbamates Triorganotin carboxylates and thiocarbamates form an interesting i l l u s t r a t i o n of the possible variations in tin coordination because these two closely related anions form both tetrahedral and trigonal bipyramidal structures. Examples of these different stereochemistries 33 are shown in Figure 2. Tribenzyltin acetate has a five coordinate trigonal bipyramidal structure with three benzyl groups equatorial 34 and the bridging acetates axial. Tricyclohexyltin acetate, however, has a distorted tetrahedral structure with a monodentate acetate group. The difference i n coordination is attributed to the increased steric hindrance of the cyclohexyl groups. Similar changes - 7 -Figure 2 Structural Features of Some Organotin (iv)-Carboxytates o and Dithiocarbamates,Bond Distances in A. a) R I ? 14 2 65 O S n — O l T j , v > V O - " ~ ' " " S n R. R R= Benzy l 0 R, R Sn . ' «J.^ -.s<5o C N R= Methyl b) ^ 2.95 „ O S n S n C R R d) R — p S n ^ J \ > R = Cyclohe>cyl R = Methy l - 8 -from five to four coordination observed when the size of the substituent on the 3 carbon of the carboxylate i s increased are also 35 attributed to steric hindrance. The N,N-dimethyl-dithiocarbamate derivatives also exhibit both tetrahedral structures and a penta-coordinated structure, the latter with a chelating dithiocarbamate 36 group. Trimethyltin N,N-dimethylf.dithiocarbamate occurs in three different crystal forms a l l of which have a distorted tetrahedral arrangement about the tin atom £nd dimethylchlorotin N,N-dimethyl-37 dithiocarbamate in a distorted trigonal bipyramidal structure with two methyl groups and one sulfur in the plane and the chlorine and second sulfur atom axial. The change in coordination is presumably caused by the increased acceptor strength of the tin atom in the (CH^ClSn derivative. 3. Clusters A third distinct type of compound which contains t i n with an expanded coordination number is cluster compounds in which t i n is bonded to either transition metals or other elements. An example of a transition metal cluster with six-coordinated t i n atoms is 38 (C H ) Pt„(SnCl_)0. The structure which has been proposed for the o 1/ 3 3 5 2. 39 carborane l-stanna-2,3-dicarba- closo- dodecaborane (11), BgC^SnH.^, has a pentacoordinate tin at the apex of an icosahedron. - 9 -C. Bonding 1. General considerations The electron configuration for the ground state of t i n is 10 2 2 [Kr]4d 5s 5p and the corresponding valence state would be 10 3 3 [Kr]4d 5s5p . This suggests sp hydridization for compounds of the R^Sn type. A rehydridization of the tin valence orbitals resulting in greater s electron density in orbitals involved i n bonding to carbon w i l l occur in substituted organotin compounds with substituents more 40 electronegative than carbon. Evidence for this change is found in the 1 119 higher tin-carbon stretching frequencies, larger H- Sn nmr coupling 41 42 119 constants ' and the larger Sn Mossbauer isomer shifts for compounds with more electronegative ligands. These trends are illustrated i n Table 1. Tin-carbon bond lengths should also decrease Table 1. Compound v Sn-C a v Sn-C s v Sn-Ca avg <5 (mm/sec)^ J % 1 1 9 S n C (CH 3) 4Sn ^524 4 3 43 '508 J 520 44 1.29 41 53. 8* (CH 3) 3SnCl 545 4 5 514 4 5 535 1.43 4 6 58.5 4 1 (CH 3) 2SnCl 2 563 4 5 524 4 5 544 1.5426 69.7 4 1 CH 3SnCl 3 548 45 548 99.5 4 1 (CH 3) 2SnF 2 598 4 6 532 4 6 576 Calculated according to Lehman's Rule. Mossbauer isomer shift relative to Sn02. Nmr spectra recorded in CC1. solution at 31°. - 10 -but unfortunately insufficient reliable data are available to indicate any trends. The observed tin-carbon bond lengths are in the f a i r l y o wide range of 2.08 to ^2.21 A. This trend to higher s character around t i n is most obvious for the octahedrally coordinated dimethyltin difluoride which has the o shortest tin-carbon bond length (2.08 A) and the highest value of the 3 2 halides for vSn-C. A simplistic view of bonding using sp d orbitals on tin would suggest ^17% s character and correspondingly longer bonds and a decrease in isomer shift. These expectations are not confirmed by experiment. Two views can be quoted to explain the experimental observations. 2+ (1) (CH.j)2SnF2 can be regarded as an ionic solid with (CH^^Sn -and F ions and the bonding in the C-Sn-C skeleton can be described 25 3 2 using an sp orbital on t i n . (2) Rehybridization of the t i n sp d 40 polyhedron according to Bent whereby bonding to the least electro-negative ligands has a l l the s electron density, i.e. ^sp orbitals 2 2 for bonding to carbon, and the SnF^ segment has p d character has been 48 suggested. The overall structure of (CI^^SnF,^ is better described as polymeric with fluorine bridges than as ionic. This view is supported by comparison to chain-like R^SnY compounds where Y i s a bridging group (C02CH3, CN, F, etc.), by the physical properties (high melting point, insolubility even in ionizing solvents, nonvolatility), by the occurrence of the Mossbauer effect at room temperature, and by the vibrational spectrum. Vibrational spectroscopy is particularly useful for studies of similar compounds with anions or anionic groups of potentially high symmetry, e.g. CIO. . - 11 -2. Methyltin cations The question of whether or not organotin cations, specifically 2+ + R^Sn and R^Sn , may exist now becomes interesting and two general routes appear suited to this investigation. These are the study of solids of formula R2SnX2 and R^SnX with R an alkyl group, preferably methyl to eliminate electronic repulsions, and X is a monobasic anion of a very strong acid and the study of R,,SnX2 and R^SnX and other suitable solutes in strongly ionizing solvents such as protonic acids. The previous views on organotin cations are expressed i n review articles 11 12 by R.S. Tobias and H.C. Clark. Tobias reviewed evidence for the existence of organotin cations in aqueous solution obtained by measuring the equilibrium constants for hydrolysis and st a b i l i t y constants for complexes of the cations with organic ligands as well as determinations of the structures of the solvated cations. Using vibrational spectroscopy, particularly Raman 2+ results, the aquated (CH.j)2Sn cation was shown to be octahedral with linear C-Sn-C skeleton and four water molecules coordinated in the equatorial plane, and the (CH^^Sn"1" cation trigonal bipyramidal with planar equatorial (CH^^Sn group and two axial water molecules. Nmr measurements indicated there i s a high degree of metal s character in the metal-carbon bonds of the solvated cations, and that the metal solvent bonds are highly polar. Tobias then suggests on the basis of the stereochemistry in the crystalline state that the bonding in the crystal i s also ionic with 2H~ ~l" 2 3 a bonded (CH^^Sn and (CH^^Sn cations using sp and sp hybrids respectively, joined to the anions by highly polar three center bonds using p orbitals on t i n . The extent of covalent interaction in these - 12 -tin-anion bonds is considered to be small and observed splitting of degenerate vibrational modes is thought to be due to site symmetry effects. H.C. Clark suggest on the basis of crystallographic and vibrational data that the extent of covalent bonding in the solid state i s much greater than suggested by Tobias. The main points supporting this view are: (1) the Sn-F...Sn bridges in (CH^^SnF are nonlinear, (2) splittings of degenerate infrared bands are too large for site symmetry effects and are also observed in methanol solutions of (CH^SnClO^,-and (3) the splittings for (CH^SnCrO^ are larger by an order of magnitude than those of (NH^^CrO^. In general, true interaction between anionic and cationic groups is found for a whole range of anions. 3. Fluorosulfate as anion The anion group of choice is the fluorosulfate group for the following reasons. (1) HSO^F i s one of the strongest simple monobasic acids known and extensive work in the HSO^F solvent system has been 49 done. (2) Infrared and Raman spectra of the fluorosulfate anion are known and indicate that the ion has C ^ symmetry. Bridging or chelation using two oxygens w i l l result in a symmetry reduction to C , causing positional changes as well as changes in the number of s bands. (3) A variety of suitable preparative techniques should provide a sufficiently large number of derivatives. This variety is in con- = trast to the situation in R^SnX compounds such as hexafluoroantimonates or arsenates"^ where only a few related derivatives are known. - 13 -(4) Extension of the study to other related molecules should be informative. The related acid groups -SC^CF^ -SC^Cl and -S03CH3 provide an easy means of altering the nucleophilicity of the SC^X group. Extension and comparison to the isoelectronic ClO^ or ^ 2 ^ 2 ions should also be possible in addition to synthesis of purely inorganic tin(IV) fluorosulfates^"'' using other preparative routes. (5) In contrast to the perchlorate derivatives which were found to 52 be explosive, the SC^X derivatives should be safe to handle. (6) Like the isoelectronic ClO^ , SC>3F i s a poor coordinating ligand. This fact should preclude any extensive complexation such as formation of [Me„Sn(SO„F) ] ^ n _ 1 ^ when HSO„F is used as a fluorosulfonating agent. 3 3 n 3 Some disadvantages of fluorosulfates w i l l have to be kept in mind. The S-F bond is sensitive to hydrolysis. This is a pH dependent process, e.g. KS03F can be recrystallized from water but is hydrolyzed rapidly in acidic or basic solution. As a consequence water would not be a suitable solvent for ionization studies. Also, the fluorosulfate group is thermally labile. Two common modes of cleavage are elimination of SC>3 and elimination of S 2 ° 5 F 2 ' S°2 F2 ° r other polysulfuryl fluorides. D. Fluorosulfates 1. Synthetic routes to fluorosulfates The chemistry of fluorosulfates has been f a i r l y extensively studied and their preparation, properties and reactions have been 49 53-55 reviewed recently. ' The chemistry of the fluorosulfate group is in many ways similar to that of the halogens and is found to meet - 14 -a l l but one of the requirements established by Cotton and 56 Wilkinson for a pseudohalogen. Fluorosulfates are usually prepared by one of three methods 57-59 (1) insertion of SO^ into an element-fluorine bond, such as the reaction of calcium fluoride with SO^ at 200° to produce 58 calcium bisfluorosulfate 200° CaF 2 + 2S03 » Ca(S0 3F) 2 (2) reaction of fluorosulfuric acid with metal fluorides, chlorides or c a r b o x y l a t e s , ' ^ e.g. ZnCl 2 + HS03F » Zn(S0 3F) 2 + 2HC1 and (3) reaction of substrates such as chlorides, bromides, oxides or even the elements with the fluorosulfonating agents peroxydisulfuryl dif l u o r i d e , S - C K F - , bromine fluorosulf ate, ^  or chlorine 2 6 2 fluorosulfate. Examples of reactions of the third type are HgO + 2S 20 6F 2 - Hg(S0 3F) 2 + 0 2 + 6 1 SiCl. + 2BrS0_F »- SiCl o(S0 oF)_ + 2BrCl 6 2 4 3 2 3 2 Reaction with the potentially oxidizing S 20gF 2 or BrS03F may not be suitable for synthesis of organotin fluorosulfates because of oxidation of the organyl group. Another rather general preparative - 15 -method commonly used to synthesize oxyacid derivatives - the silver salt method - is not applicable to the synthesis of fluorosulfates because AgSO^F is d i f f i c u l t to obtain. 2. Structural studies on fluorosulfates Structural studies have been reported for compounds containing both ionic and covalently bonded fluorosulfates. Ionic bonding is found for alkali,*' 3 alkaline earth 6 <^ and some other metal fluorosulfates 6 <^ as well as for fluorosulfates of some heterocations such as N0 + or + 64 NC>2 . The crystal structures have been determined for two of these 65 66 salts, KS0oF and NH,SO„F. A crystal structure has also been 3 4 3 6 7 reported for acetate acidium fluorosulfate, CH^C(OH)^SO^F. In this case the fluorosulf ate ion is hydrogen bonded to the CHgCKOH^"1" cation resulting in bidentate fluorosulfate groups. Mostly vibrational spectroscopy has been used to study covalently bonded fluorosulfates. S „ 0 , F o , 6 8 , 6 9 BrS0„F, 7 0 C1S0„F, 6 9 and FSO.F 6 9 L D 2. J J 3 were a l l shown by Raman spectroscopy to be covalently bonded with monodentate fluorosulfate groups. Two examples of bidentate covalent fluorosulf ates have been reported. SbF^SO^F''^ was shown primarily by 19 F nmr measurements to have fluorosulfate bridging, however, no vibrational assignment is made and the infrared and Raman spectra that are reported differ substantially. More recently a report on 72 SeF^SO^F indicates bidentate SO^F groups on the basis of vibrational and nmr spectra. Tridentate fluorosulfates are also possible and 73 one example of such a conformation has been found in Co(S0.jF)2 quite recently. - 16 -3. Previous work on ti n fluorosulfates There are two reports of synthesis of inorganic tin(IV) fluoro-sulfates in the literature. Hayek et a l . reported the reaction of SnCl^ with HSO^F to produce a mixed product most lik e l y made up of 59 Cl 2Sn(S0 3F) 2 and ClSn(S0 3F) 3, and Lustig and Cady reported the reaction of SnCl, with S~0,F_ to produce ClSn(SO_F),. 5 1 Sn(SO_F). 4 z o 2 3 3 3 2 74 has also been prepared. While some of this work i s of a preliminary nature and not even vibrational spectra are reported, i t does indicate that t i n fluorosulfates may be thermally stable and not spontaneously decomposed to fluorides, oxides or sulfates as i s found, e.g. for si l i c o n fluorosulfate derivatives.^ 76 Trimethyltin fluorosulfate has been reported as well as several organotin derivatives of other sulfonic acids. These include (CH 3) 3SnS0 3CF 3, 7 7 (CH ) 2Sn(S0 3CF 3) 2, 7 7 (CH3>3SnSC>3CH3,78 and 79 (CH3)2Sn(SC>3CH3)2. Only replacement reactions with the acids or the silver salt method appear to be applicable to the synthesis of methyl-, trifluoromethyl-, or chlorosulfates as well as to the isoelectronic difluorophosphates. Insertion of S0 3 has also been used on occasion, e.g. in the reaction of S0 3 with (CH )^Sn, to yield (CH 3) 3SnS0 3CH 3. 7 8 (CH 3) 4Sn + S0 3 (CH 3) 3SnS0 3CH 3 S0 3 insertion has also been used in the reaction with SnCl^ presumably 80 to produce SnCl 2(S0 3C1) 2-- 17 -E. Vibrational Spectroscopy In the study of the vibrational spectra of ti n and organotin fluorosulfates attention can be focussed on two main points; the tin-carbon stretching vibrations in the 620-500 cm 1 range and the fluorosulfate anion stretching vibrations which are expected in the range of 1500-700 cm"1. 1. Tin-carbon If we take the compound (CH^^SnSO^F as an example, there w i l l be two possible conformations for the C^Sn moiety, i t can be either planar or pyramidal. Similarly for (CH^^Sn compounds the possi b i l i t i e s are a linear or a nonlinear C^Sn grouping. An example of a molecule with a planar C^Sn group i s (CH^^SnF. This molecule has 29 symmetry and the infrared and Raman activities and intensities of the asymmetric and symmetric Sn-C stretches should be as follows: v Sn-C, weak i n the Raman and strong in the infrared, and v Sn-C, 3. S strong in the Raman and inactive in the infrared. If the C^Sn moiety is nonplanar the symmetry w i l l be C^v and both tin-carbon stretches w i l l be active with comparable intensities in both the infrared and Raman spectra. An example of this case would be (CH^^SnCl. 2. Fluorosulfates The conformations for the fluorosulfate group include the five po s s i b i l i t i e s summarized in Table 2. The SO^F ion has symmetry and six vibrational modes, a l l of which are active in both the infrared and Raman spectra. These modes can be described as asymmetric and - 18 -symmetric S0„ stretches, v SO„(E) and v SO„(A), S-F stretch, vS-F(A), 3 a 3 s 3 asymmetric and symmetric SO bends, 6 SO„(E) and 6 SO (A), and a 3 cl J S J rocking mode T SO^F(E). If the fluorosulfate group is coordinated to the tin atom in a monodentate or bidentate manner, i t s symmetry w i l l be lowered to C g resulting in an increase i n the total number of modes due to SO^F. The coordination over 0 (or F) should also reduce the electron density in the S-0 (or S-F) bonding region which in turn w i l l increase the back donation from the remaining terminally bonded atoms and therefore cause changes in the positions of the vibrational modes. The following diagnostic c r i t e r i a should emerge for distinction between the four covalent p o s s i b i l i t i e s . (1) The total number of SO^F vibrations should allow distinction between monodentate and bidentate on one hand and other possible conformations. (2) An increase in the energy of VS-F would be expected for the tridentate case and a decrease for the tetradentate. (3) The positions of the SO^ stretching modes should be different in the monodentate case than they are i n the bidentate case. For a vibrational analysis of this sort the following factors must be considered. (1) The observed motions should be pure motions and not mixed. 81 This has been found to be the case for fluorosulfates. (2) The vibrations should be easily identified and not obscured. For methyltin fluorosulfates the positions of the methyltin vibrations are well established and the positions of these bands as shown in - 19 -TABLE 2 Conformations of Fluorosulfates Conformation Description Symmetry Ho. of modes No. of ( a l l are i r and stretching Raman active) modes S0 5F ionic c 3 v 6 (3A.5E) 3 0*S02F monodentate c s 9 ( 6 A ' , 3 A " ) 4 02*S0F bidentate c s 9 (6A',3A") 4 tridentate ° 3 v 6 (3A,'3E) 3 05*SF* tetradentate ° 3 v 6 (3A.3E) 3 indicates bidentate 0 or P 0, F indicates monodentate 0 or F - TABLE 3 Methyl-tin Vibrations Mode Position Ir Intensity •v aCH 5 3000 cm"1 VgCHj 2900 5 aCH 3 1400 weak S SCH 5 1200 weak CH-j rock 800 strong . VSn-C 500-600 varied TABLE 4 S-0 Stretching Frequencies for Monodentate SO-^ F Mode Compound CISO^J?6^ BrSO-jF?0 S 2 0 6 F 2 6 9 Xe(S0jF) 2 8 2 HSOjF8? ^ 7 ( aS0 2) 1478CM"' 1438 1498 1425 1445 s S 02) 1 2 25 1206 1248 1238 1230 "^2( SO) 856 884 847 959 960 - 20 -Table 3 should lead to only a minimum of overlap. (3) The s p l i t t i n g or shifting of vibrational modes should be caused by conformational differences and not by site symmetry effects or by weak anion-cation interactions. Some examples are taken from the literature to i l l u s t r a t e these points. The ionic fluorosulfate i s typified by KS03F and the positions of the six vibrational modes of the S0 3F anion in KS03F are shown in the correlation diagram, Figure 3. An example of an ionic fluorosulfate FIGURE 3 Correlation Diagram for SO3P c3v 1285 OK"1 1079 745 594 570 407 C s ^ ? ( A " ) ^ i ( A ' ) \>2W) - f y A ' ) 0 6 ( A " ) ^ 4 ( A » ) ^ 3 ( A ' ) ^ 9 ( A " ) %{k') CISO3F 6 9 1478 1225 856 630 534 4c6 573 3o9 363 which shows sp l i t t i n g of the E modes due to anion-cation interactions 64 or site symmetry effects is N0S03F. In this case the three E modes are s p l i t by between 12 and 32 cm The best examples of monodentate fluorosulfates are the halogen f l u o r o s u l f a t e s ^ ' 7 ^ one of which, C1S03F, i s shown as an example of a C symmetry fluorosulfate in the correlation diagram. The biggest s difference between the spectrum of C1S03F and that of KS03F i s that C1S03F has nine vibrational modes compared to six for KS03F. The spli t t i n g of these E symmetry modes, most obvious for v SO. i s much a J larger for C1S03F than i t i s for N0S03F. In addition the position of - 21 -v S-F is significantly raised in CISO^F. Because of the expected change in electron density in the S-0 region with change in coordination number, the positions of the S-0 stretching modes for monodentate SO-jF, list e d in Table 4, are of interest. Very few vibrational spectra of b i - or tridentate fluorosulfates and no examples of tetradentate fluorosulf ates are known. The only 71 72 reported bidentate fluorosulfates are SbF^SO^F and SeF^SO^F. 19 In both cases the structure is assigned with the aid of F nmr and other physical techniques and the vibrational spectra are not well understood, e.g. for SbF^SO^F large discrepancies between the Raman and infrared spectra are observed. Spectra of more bidenate fluoro-sulfates w i l l be needed before any structural interpretations based on the positions of the S-0 stretching modes can be made. A tridentate 73 fluorosulfate group was identified very recently for CoCSO^F)^ on the basis of infrared, Raman, uv-visible, and magnetic measurements. The SO^F group was found to have symmetry with the position of v S-F raised from that of ionic fluorosulfates by about 100 cm \ 3. Other sulfonates Some work has also been reported on the vibrational spectra of the other sulfonate anions. The frequencies of the SO^CF^ and SO^CH^ ions were assigned with the aid of normal coordinate analyses in two separate 82 83 — studies. ' As shown in Table 6, the two assignments for SO^CF^ disagree considerably. The complications are caused by the extensive mixing of vibrational modes. Vibrational spectra of some monodentate SO^CF^ - 22 -TABLE 5 S t r e t c h i n g F r e q u e n c i e s f o r 3 i - & T r i d e n t a t e SO 3 P M o d e O o m p o x m d K o d e S b P ^ S O ^ F 7 1 SeF3S03F'''2 C o ( S 0 5 P ) 2 ? 3 i r Hainan i r R a m a n i r V.so 1 4 0 0 1 4 3 0 1276 1290 1236 1265 ^ a S 0 3 V a s o 2 1216 1 2 2 4 1237 1020 1 0 3 0 1076 1 0 8 1 1109 ^ s s o 3 865 690 6 2 4 819 807 850 \)s? T a b l e 6 . P r e v i o u s A s s i g n m e n t s f o r SO C F B a n d p o s i t i o n A s s i g n m e n t A g r e e m e n t ( c m - 1 ) T o b i a s 8 2 8 3 B u r g e r M . 2 7 0 v C F _ a 3 v 7 ( E ) v a S 0 3 n o 1 2 3 0 v C F _ s 3 v l ( A ) v s C F 3 y e s 1 1 8 0 v S 0 _ a 3 V 1 0 ( E ) V F 3 n o 1 0 3 8 V ° 3 v . ( A ) v S O . 4 s 3 y e s 7 6 6 6 s C F 3 ( A ) ' v v C - S n o 6 3 0 6 S O -a 3 v 5 ( A ) 6 s S 0 3 n o 5 8 0 6 S O „ s 3 v 8 ( E ) 6 a C F 3 n o 5 2 0 6 a C F 3 v l l ( E ) 6 a S 0 3 n o 3 5 3 PSO3 v 1 2 ( E ) P S 0 3 y e s 3 2 1 v C - S ( A ) n o 2 0 8 P C F 3 v 9 ( E ) p C F 3 y e s - 23 -84 — 86 derivatives have appeared very recently and as expected these spectra show spli t t i n g of the E symmetry SO^ modes. A vibrational assignment has also been made for the SO^Cl anion. In conclusion, vibrational spectra should include Raman as well as infrared spectra with the emphasis on the anion stretching modes and the C-Sn skeletal vibrations. For liquids, polarized Raman spectra should provide additional information and help substantiate assignment of the observed modes. The obtained information should reveal the functionality of the SO^F group but not whether the fluorosulfate acts as a bridging or chelating ligand. To make this f i n a l distinction supporting evidence from other physical techniques 119 such as X-ray diffraction or Sn Mossbauer spectroscopy i s needed. Physical properties, e.g. melting point, v o l a t i l i t y , or solubility i n nonpftj ar solvents, may also provide additional clues. F. Mossbauer Spectroscopy A second physical method which was used extensively in the 119 investigation of the structures and the bonding i s Sn Mossbauer 119 spectroscopy. Since Sn, one of the many nuclei which exhibits the Mossbauer effect, has a reasonable natural abundance, 8.57%, i t s second Y~ r aY energy i s i n a convenient range, and the h a l f - l i f e of 119 the f i r s t excited state i s optimal, Sn MOssbauer spectroscopy has been used extensively i n the study of the chemistry of t i n . An introduction to the use of Mossbauer spectroscopy in chemistry has been written 88 by Greenwood. - 24 -1. Principles The Mossbauer effect for t i n arises from recoilless emissions 119 and resonant reabsorption of y radiation by the Sn atom. 119 Sn decays, as shown in Figure 4, to i t s ground state by two Y emissions and the second decay from the 3/2 nuclear excited state to the 1/2 ground state i s the nuclear transition used to measure the Mossbauer effect for tin. The energy of the Y rays emitted in this decay are 23.875 KeV. This value i s not in the range where extremely low temperatures are needed to effect recoilless emission and absorption, however, 23.8 KeV is a large energy and the source is usually polymeric SnO^ imbedded in a Y inert matrix to prevent recoil. The absorber usually has to be cooled in liquid nitrogen to prevent recoil on the absorption of the y-vays. The h a l f - l i f e of —8 the f i r s t excited state (3/2 state) is 1.84 x 10 sec and is in the range where Mossbauer line widths w i l l be optimal. Short half-lives (< 10 sec) result in lines with wide natural line widths and —6 longer half-lives (> 10 sec) yield lines so narrow that serious experimental d i f f i c u l t i e s are encountered in their detection. The chemical applications of Mossbauer spectroscopy arise from the fact that the spacings of the nuclear energy levels depend on the electron distribution about the nucleus, in particular on the distribution of the valence electrons. Thus in a Mossbauer experiment, the energy difference between the ground state and f i r s t excited state of the source and absorber w i l l be different and resonant reabsorption by the absorber of the Y r aY emitted by the source w i l l not be achieved. In order to bring the source and absorber into resonance, the frequency - 25 -Figure 4 D e c a y of 5 Q S n 119 m S n H 9 m 1 s t E x c . S t a t e 3 / 2 + G r o u n d S t a t e V 2 + •? 2 4 5 d y = 6 5 . 6 6 K e v f I . 8 4 x i o _ s s e c / M = 2 3 . 8 7 5 S n I I9 Figure 5 Isomershift 8 ( mm /sec ) M o s s b a u e r s p e c t r u m Absorption -1 r~ O •! 2 3 A •>• Velocity in m m / s e c Exc i t ed state Ground s ta te — r Source Absorber IS = E - E„ a s L77-Zez[lv// a d s ( O ) l 2 - 1^/ s o u r c e (O) lJ](<Rex - <Rg>) cons t . p r S Iv// (Of €s pos if <Rex>!arger <Rg>as for tin - 26 -of the y emission is modulated by moving the source relative to the absorber thus u t i l i z i n g the Ooppler effect. The resulting Mossbauer spectrum consists of a plot of the count of Y rays transmitted through the absorber against the velocity of the source. At some velocity there i s resonant absorption and the count rate drops. 2. Isomer shift The isomer shift, 6 , i s the measure of the source velocity needed to bring a particular absorber into resonance with the source. The isomer shift depends on the r a d i i of the nucleus and the s electron density at the nucleus according to: A r 2 6 = const • — • A|\JJS(0)| where A r = r . , ^ ^  - r , ^ ^  excited state ground state A 1^8(0) | 2 = |^s(0) | 2 - |^s(0) | 2 1 1 1 absorber1 ' source1 = difference in s electron density at the nucleus, between the source and absorber. This equation shows that the isomer shift i s dependent on a nuclear factor, Ar/r, and an extranuclear factor, A|I(IS(0)| . For tin, Ar/r is positive and therefore an increase i n <S indicates an increase in I^s(0)absorber I ' fc^e s e l e c t r o n density at the absorber nucleus. This means that 6 can be used as a measure of the s electron density at the t i n nucleus. The variation in 6 from SnO^, customarily set at 0 mm/sec, for some tin(IV) compounds is shown in Figure 6. - 27 -Figure 6 Distribution of S for Sn(iv) increas ing bondpolar i ty S n n 4+ S n 5s° 5 p ° -0.433 K 2 S n F | Snf^ 0.42 K S n C i d imethy l t in( iv ) derivatives 1.1 — 1.6 mm/sec 0.80 • 1.29 SnCt S n B r . ( C H , ) . S n • . 4 • 3 4 S n (iv) 5 s 1 5 p 3 1.93 a - S n ( 7 7 ° ) Sri, 1 T 1.5 ( m m / s e c ) 2 .0 r - 0 . 5 —r-O i + 0 . 5 1.0 S n O , Figure 7 Q u a d r u p o l e C o u p l i n g A ( m m / s e c ) A b s o r p t i o n Ve loc i t y in m m / s e c E x c i t e d r _ 3 state Ground j = L s ta te 2 Isomer sh i f t - 7 - + 5 ±1-Quadrupo le coupl ing c o n s t Q n u c l e a r q u a d r u p o l e m o m e n t q e l e c t r i c F i e l d g r a d i e n t n~] A s y m m e t r y F a c t o r f o r a n a x i a l l y s y m m e t r i c s y s t e m : A E Q = c o n s t . V z z Q - 28 -3. Quadrupole spl i t t i n g Quadrupole splittings, A, can arise i f there is an electric f i e l d gradient, q, for any nuclear state which has a quadrupole moment, Q, i.e. for any state with spin greater than 1/2. The f i r s t 119 excited state of Sn has I = 3/2 so t i n compounds w i l l have a quadrupole splitting whenever an imbalance of charge around the t i n nucleus causes an e.f.g. for tin as given by the following equation. AEq = 1/2 e q Q(l + n /3) ' = A V -V where n is an asymmetry factor equal to — ^ — . The V's are the zz components of the e.f.g. tensor. For axial symmetry, n = 0 and A = const.q.,Q.- It can be seen from these equations that both the magnitude and sign of A w i l l depend on the e.f.g., q. For tin compounds i t is generally believed that q arises primarily from inequalities in the polarities of the 44 89 90 a bonds, and not IT bonding effects as was once thought. ' The size of q w i l l depend on two factors, charges on the atoms surrounding t i n , [q.. ], and the imbalance in the distribution of the la t valence electrons on tin, t c l v a ^ ] • These two contributions w i l l have 91 92 opposite signs but t ^ ^ ] > > ^^.at^ a n c* '•Sral'' * S ^ e n c e t* i e dominant factor. The sign and magnitude of [ ^ ^ l W H 1 also be dependent on two factors, geometrical effects due to the different orientations of ligands about t i n and any imbalance in the polarity of the tin-ligand - 29 -a bonds. The influence of geometry on A can be considered most 44 91 93 readily using the point charge approximation. ' ' In this model each ligand is assumed to be a point charge and the contributions of the ligands [X] for the various geometries can be calculated [ x ] = q(x)(l-A(x) r 3 x V = Z(3cos 6-1)[X] zz x q = charge on X A = shielding factor x 'Sn-X and A a V ( l / 3 r ) 2 + l ) 1 / 2 zz 94 * Electronegativities or Taft inductive constants, a , can be used to estimate the a bond polarities. For compounds with the same geometry, as the bond polarity differences between the ligands increase, A w i l l also increase. The point charge model can also be used to predict quadrupole splittings, i.e. i f the partial quadrupole splitting for each ligand can be determined and the geometry is known, then these can be used to predict A. Consistency between experimental A's and ones calculated in this manner can also be used as a check on the validity of the geometry assumed for the calculation. 4. Room temperature effect At room temperature tin compounds generally recoil when emitting or absorbing radiation and since the Mossbauer effect depends on recoilless emission and absorption no spectra are produced. The fact - 30 -that " some tin compounds do give spectra at room temperature has 95 96 been attributed to the polymeric nature of these compounds ' because polymeric compounds w i l l have less freedom to recoil thus exhibiting resonant absorption at higher temperatures than monomeric compounds w i l l . The effect i s usually reported as the parameter R = e o n o/e 0 where the e's are the absorptions at the indicated temperatures. R w i l l be in the range of zero to one but experimental considerations cause the lower limit to be % 0.03. The room temperature effect gives only a rough indication of possible polymeric nature. Exceptions to this rule, both polymeric compounds which do not have R > 0 and non-polymeric molecules with R > 0, are known. This means the presence of a room temperature effect can only be used as corroborative evidence and cannot be used as a basis for structural assignments. The sensitivity of the M8ssbauer parameters, particularly the quadrupole splittings, to changes in the environment around tin can be illustrated by reference to the data for trimethyltin carboxylates, Table 7, which indicate the sensitivity of this technique to changes in the X group in CO^CX^. G. HSO^F as a Non-aqueous Solvent System The use of fluorosulfuric acid as a solvent system has been extensively investigated 1^ 1 (^ 2 and the whole subject of fluorosulfuric 49 acid chemistry has been reviewed. Some physical properties of HSO^F are listed in Table 8. The large liquid range of HSO^F allows measurements and reactions to be performed over a wide temperature range. Although the freezing point i s low cryoscopic measurements can -5 31 T TABLE 7 T r i m e t h y l t i n c a r b o x y l a t e s Compound (0H 3 ) 3 S n C 0 2 C H 3 OO2OH2OI C0 2 CHC1 2 OOoGHoBr OO2OOI3 0 0 2 C F 3 C 0 2 C B r 3 C0 2 N (CHx ) P mm/sec •A nun/s e c r e f e r e n c e 1 . 3 5 • 3 . 6 8 97 1.41 3 . 8 9 98 1 . 3 7 4 . 0 8 98 1 . 3 4 3 . 9 0 93 1 . 4 4 4 . 1 5 98 1 . 3 3 4 . 2 2 93 1 . 4 3 4 . 1 3 93 1 . 3 2 3 - 5 6 99 1 . 2 6 3 . 3 9 99 TABLE 8 P h y s i c a l P r o p e r t i e s o f HSO3F P r o p e r t y V a l u e (temp) R e f e r e n c e B o i l i n g p t ( *0 ) 1 6 2 . 7 100 F r e e z i n g p t A * C ) - 8 8 . 9 8 102 D e n s i t y ( d . ° ) 1 . 7 2 6 (25 ) 100 V i s c o s i t y ( c e n t i p o i s e ) I . 5 6 (25 ) 100 D i e l e c t r i c c o n s t a n t 120 (25 ) 100 S p e c i f i c c o n d u c t a n c e ( o h i n " 1 cm" 1) 1 . 0 6 x 1 0 " ^ ( 25 ) 100 S e l f i o n i z a t i o n e q u a t i o n 2HSO3F ^ = ^ 1 H2S03F"*" + S O 3 F " - 32 -102 s t i l l be made. The low specific conductance, K, of HSO^F allows conductometry because the proton jump mechanism is operative. The relatively low viscosity means contributions from other ions w i l l be more noticeable in HS0„F than in H„SO.. 3 2 4 49 Fluorosulfuric acid i s perhaps the strongest simple acid known and only a very limited number of solutes are expected to behave as acids in this system, i.e. give rise to ^ SO^F"*" ions. Sulfur trioxide as well as the inorganic fluorides AuF^, PtF^, TaF,. and SbF,. are acids but only SbF,. and SbF^-SO^ mixtures"*"^ have been studied extensively. Some examples of bases are metal fluorosulfates such as KSO^F, salts of inorganic acids such as KF or KCIO^, inorganic acids for example I^SO^, as well as other inorganic or organic molecules which can be protonated. Measurements of the specific conductance of a fluorosulfate contain-ing solute in HSO^F at different solute concentrations w i l l give information on the extent of the dissociation of the solute into ions. Because of the high molalities of the H^SO^F"1" and SO^F ions, the specific conductance of the solution w i l l depend on Y, the number of moles of SO^F or ^ SO^F"1* ions produced in solution per mole of solute. Y can be calculated by comparing the molality of a solution which has some specific value of K with the molality of the solution of the fu l l y dissociated base KSO^F which has the same specific conductance. Y w i l l be the ratio of m /m - ^ . . KSO^F solution Some advantages of fluorosulfuric acid as a solvent system are: (1) the common anion should f a c i l i t a t e the interpretation of conductivity results, (2) the acid can be used as both a non-aqueous solvent and a - 33 -reactant, and (3) HSOgF can be easily purified. There has been one study of tin(IV) compounds in 100% sulfuric 103 acid. Tetramethyltin and trimethyltin sulfate were found to dissolve rapidly to form solvated trimethyltin cations. Tetraphenyltin and triphenyltin hydroxide dissolve in ^SO^ with complete cleavage 2-of Sn-C bonds to form the HSn(HS04)6 and SnCHSO^Jg ions. If the expected range of inorganic and organometallic tin fluoro-sulfates can be prepared as well as analogous derivatives of other sulfonic acids, their study should help to clear up controversies mentioned above; such as the existence or non-existence of organotin cations in solution and in the solid state. This wide range of compounds should also be helpful in establishing the chemical s i g n i f i -cance of Mbssbauer parameters as well as in investigating the applicability of vibrational spectroscopy to structural studies of this type. - 34 -CHAPTER II EXPERIMENTAL A. Apparatus Because many of the compounds handled during this investigation were hygroscopic, techniques which avoided the contact of these compounds with air were necessary. For this reason solids and non-volatile liquids were handled in a dry box and volatile liquids and gases, on a vacuum line. 1. Dry box The dry box was a Vacuum Atmospheres Corporation Model HE-43-2 Dri-Lab with a Model HE-93-B Dri-Train. "L" grade nitrogen was used for the atmosphere in the dry box. The dryness of the nitrogen was maintained by circulating i t through a regeneration chamber f i l l e d with Lindes Molecular Sieves. The sieves were regenerated periodically by heating them in the presence of air. 2. Vacuum lines The general purpose glass vacuum line consisted of a 50 cm long piece of 2 cm glass tubing with four outlets connnected to a mechanical rotary vacuum pump via a Fischer and Porter teflon stem stopcock, a standard taper B19 ground glass cone-socket connection for removing - 35 -the line, and two safety traps. The four outlets consisted of teflon stem stopcocks and BIO ground glass sockets, three spaced equally along the lower side of the line and the fourth at the end of the line. A mercury manometer which could be isolated from the rest of the line via a stopcock was also incorporated into the system. A metal vacuum line made of 1/4" diameter monel tubing and equipped with Whitey valves (1KS4316) which worked in a manner similar to the glass line, was also available for handling substances reactive towards glass. 3. Reaction vessels Several types of glass reaction vessels and one type of metal reactor were used during this work. (a) Glass The simplest type was a two part glass reactor which consisted of a 50 ml round bottomed flask with a standard taper B19 ground glass cone and an adapter top consisting of a teflon stem stopcock between a B19 socket for attachment to the flask and a BIO cone for attachment to the vacuum line. This type of reactor had the advantages that i t could be readily opened to put in solids or liquid reactants and also to take out products, and i t could also be attached to the vacuum line for the addition or removal of volatile liquids. The main disadvantages of this reactor were that because of i t s two part construction i t could not be used with gases which were at a pressure of greater than one atmosphere and because i t had a ground glass joint which had to be greased, i t could not be used whenever grease attack - 36 -would be a problem. These two disadvantages could be overcome by using a one part reactor which could be sealed closed after solid reactants had been added. Reactors of this type with several different shapes were used. The simplest one consisted of a test tube with a constriction and B19 cone at the top and a side arm leading to a BIO cone via a teflon stem stopcock. After solids were added (in the dry box, i f necessary) through the constriction and the B19 cone capped, the vessel could be flame-sealed at the constriction and volatile reactants d i s t i l l e d into the resulting one piece grease free reactor on a vacuum line. Two other reactors which worked on the same principles were also used. These were the same as the one part reactor described above except that greater volumes were attained by using 50 ml round bottomed flasks or 125 ml erlenmeyer flasks instead of simple tubes. For reactions between two volatile reactants, a simpler vessel without the constriction and B19 cone could be used. These reactors consisted of test tubes or bulbs with teflon stem stopcocks and BIO cones attached directly to the top of the tube or bulb. Similar vessels were also used to store volatile liquids. The one part reactors could only be used once and had to be broken open inside the dry box to remove solid products. However, they allowed a complete exclusion of stopcock grease and would hold a positive pressure of several atmospheres. Teflon coated sti r r i n g bars could be used to s t i r reaction mixtures externally. - 37 T F i g u r e 8 R e a c t o r s Lid "V. r = 0 h o k e V a l v e ( N o 431) • M o n e l M e t a l T u b e Sof ts to S e c u r e L i d to B o t t o m V e s s e l C o n d e n s e r Inlet B o t t o m • C o n d e n s e r Inlet fvtonel Meta l R e a c t i o n Vesse l ( 1 5 0 ml ) Monel Metal a - P e r t Reaction Vessel B 19 Cone 135 ml P y r e x -Erler.m'eyer F l a s k Two P a r t G l a s s E e a c t o r 1 — S I O Ccrs. F ische r c n i Por te r Te f lon V c l v e One P a r t G l a s s E e a c t o r s - 38 -( b) Metal A suitable metal reactor used in this study consisted of a cylindrical monel pot of % 100 ml volume connected with six bolts to a l i d equipped with a Hoke valve (#431). A vacuum tight seal was achieved with a teflon ring inserted into a groove between the pot and the l i d . This reactor could be attached directly to the metal vacuum line using a swagelok connector. 4. Fluorosulfuric acid The apparatus and techniques involved in the purification have been d e s c r i b e d . T e c h n i c a l grade HSO^F was doubly d i s t i l l e d and for conductivity work d i s t i l l e d directly into the conductivity c e l l . Before d i s t i l l a t i o n L grade was passed through the c e l l and s t i l l for several hours and the entire apparatus was flamed out periodically to remove moisture or traces of HF. HSO^F with a specific conductivity -4 -4 -1 -1 of between 1.1 x 10 and 1.3 x 10 ohm cm was obtained in this -4 -1 -1 manner. The lowest reported value is 1.08 x 10 ohm cm 5. Conductometry The design of the conductivity c e l l has been d e s c r i b e d . I t is a three electrode c e l l with c e l l constants of ^ 3, 7 and 10 cm ^ 104 with platinum electrodes platinized periodically with platinum black. The c e l l was calibrated using aqueous K C l . ^ ^ Conductivity measurements were made using a Wayne Kerr Universal Bridge, B221A, while the c e l l was kept at a constant temperature of 25.00 + 0.01° (Brookly precision thermometer) in an o i l bath with a volume of ^ 9 1. with a Sargent Model ST Thermonitor. - 39 -Determination of the specific conductivity of a solid solute over a range of concentrations was done by making several additions of known amounts of the solute from a dropping funnel designed for the addition of solid samples in the absence of moisture to a known weight of HSO^F in the conductivity c e l l and recording the conductance after each addition. This funnel was simply a large bore glass stopcock with an enclosed cavity for the sample above the tap and a narrow tube and BIO cone below. 6. Vibrational spectroscopy Infrared spectra over the range of 4000-250 cm ^ were recorded on a Perkin Elmer 457 grating spectrophotometer. Solid samples were run as nujol mulls, or in the cases where the samples attacked nujol, as thin films, using AgCl, AgBr, KRS5 (TIBr-TlI), or Csl windows obtained from Harshaw Chemicals. The KRS5 and Csl windows are transparent over the whole spectral range of the spectrometer, however, these windows are not very inert and could not be used for many samples due to window attack. In these cases AgBr, transparent to ^300 cm \ or AgCl, transparent to ^400 cm \ were used. Spectra of gases were obtained using a monel metal gas c e l l of 7 cm path length with AgCl windows. This c e l l could be attached to a vacuum line via a Whitey valve and B10 cone. Infrared spectra were calibrated using a polystyrene film. Raman spectra were recorded on a Cary 81 spectrophotometer with a o Spectra Physics Model 125 He-Ne laser light source (X = 6328 A). Spectra of solid samples were recorded with the sample sealed into 6 mm OD glass tubes with a f l a t end. - 40 -7. MBssbauer spectroscopy The Mossbauer spectrometer was of the constant acceleration type and consisted of a TMC Model 305 velocity tranducer driven at constant acceleration by a TMC Model 306 wave form generator which also synchro-nized the 400 channel analyzer. The source, BaSnO^ enriched with ^"^mSn, was mounted on the transducer. y-Rays transmitted through the absorber were detected by a Reuter-Stokes RSG 60 proportional counter and fed into a Nuclear Chicago Model 33-15 single channel analyzer and then to a 400 channel memory. The output was displayed on a Hewlett Packard Model 120B oscilloscope and also printed out on an IBM Model 44-16 typewriter. The output of the spectrometer was fit t e d to a Lorentzian curve on an IBM 360/67 computer and information about peak intensity, isomer shift and quadrupole splitting were obtained in terms of channels. The velocity was calibrated with a National Bureau of Standards sodium nitroprusside crystal absorber with 119 a A of 1.726 mm/sec. Isomer shifts were calibrated with an Sn enriched SnO^ absorber. Samples were placed in brass cells with mylar windows and spectra were recorded with the absorber either at 78° or 298°K. Cells fi t t e d with teflon windows were used for samples which reacted with the mylar. Isomer shifts are reported relative to Sn02 with an estimated precision of + 0.03 mm/sec. 8. Nuclear magnetic resonance spectroscopy 19 1 F and H nmr spectra were both obtained on a Varian HR 60 19 spectrometer. For F nmr CFCl^ was used as an external standard and for "4l nmr (CH^^Si was used, also external. Samples were placed in 5 mm diameter tubes and either flame sealed or capped. - 41 -9. Melting points Melting points were determined using a Thomas Hoover capillary melting point apparatus in which both the sample under study in a glass capillary and the thermometer bulb are heated in an o i l bath. The melting points are reported uncorrected. 10. Analyses Elemental analyses for Sn, F, S, As, Sb, and CI were obtained from A. Bernhardt Microanalytical Laboratories in Germany. Carbon and hydrogen analyses were carried out by Mr. P. Borda of the Chemistry Department, University of British Columbia. For some compounds, tin and chloride were determined in our laboratory. Tin was estimated gravimetrically as SnO^ and chloride by potentiometric titration using a Radiometer Model 26 pH meter. B. Chemicals 1. Commercial sources Most of the chemicals used in the preparations were obtained from commercial sources and were of reagent grade or the highest grade available. The chemicals, sources and any remarks are list e d in Table 9. 2. Literature preparations Some of the starting materials which were not commercially available were prepared by methods described in the literature as outlined below. - 42 -T A B L E 9 C h e m i c a l s C h e m i c a l S o u r c e R e m a r k s ' ( C H 3 ) 4 S n ( C H ^ U S n C l ( O H ^ ) p S n C l p ( C H v ) B n C l , 4 S n C l S n B r S n F ^ S 1 C 1 4 T i C l 4 G e C l 4 S b F 5 AsFp-B F 3 J C s C l NO N 0 ? c i 2 B r S S O , BaCS03CF3) H S 0 3 F HSO3CI H3O3CE3 H S 0 v C 2 H 5 NOSO4H s o 2 c i 2 P e n n i n s u l a r C h e m R e s A l p h a . V e n t r o n O z a r k M a h o n i n g , P . C . R . A l p h a V e n t r o n B . & A . , B . D . H . A l p h a V e n t r o n R e s e a r c h I n o r g C h e a A l p h a V e n t r o n B . D . H . O z a r k M a h o n i n g O z a r k M a h o n i n g . M a t h e s o n B . D . H . M a t h e s o n M a t h e s o n M a t h e s o n B . D . H . A l l i e d A l l i e d 3M A l l i e d i ' l a t h e s o n E a s t m a n E a s t m a n C o l u m b i a O r g a n i c C h e m E a s t m a n , B . D . H v . r e c r y s t a l i z e d f r o m CHCI3 v a c u u m d i s t i l l e d d i s t i l l e d i n N 2 s t r e a m t o r e m o v e HP T e c h , d o u b l y d i s t i l l e d n o t p u r i f i e d , s e e r e f 106 o b t a i n e d a s a l i q u i d , p r o b a b l y d u e t o e x c e s s H 2 S 0 4 - 43 -Peroxydisulfuryl difluoride, S„O^F , was prepared by the AgF0 Z b Z Z catalyzed reaction of fluorine with sulfur trioxide at 180-220° using the method described by Cady and S h r e e v e . T h e catalytic reactor and method were essentially the same as reported, however, the following changes were made to increase the yield. A larger reactor 120 cm long (vs. 90 cm) and a reaction temperature of 180° was found to give better results. The S0^ was also heated to 50° (vs. 25°). To avoid the condensation of potentially hazardous FSO^F the last trap was cooled to only -78°C (dry ice) and not -183° (liquid 0^) as done in the literature preparation. Unreacted SO^, i f present, was removed from the crude $2^6^2 ^ washing with 96% HL^ SO^  in a separatory funnel in a fumehood. Impurities of FSO^F were removed by pumping on the S„0,.F- at -78° for extended periods of time. The purity of Z o Z 19 the f i n a l product was checked by infrared and F nmr. Bromine monofluorosulfate was prepared by the direct reaction 108 of excess Br_ with So0,,F- as reported by Cady. Bromine trisfluoro-Z Z b Z sulfate was prepared by the reaction of bromine with excess S„0,F_ Z O Z 108 at room temperature. Iodine trisfluorosulfate was prepared by the similar reaction of iodine with excess S_0,F also at room temperature. Z b Z Chlorine dioxide was prepared from potassium chlorate, oxalic 109 acid and sulfuric acid using the method described by .Brauer. The CIO2 produced was purified by pumping on i t at -78° to eliminate CO2 and by trap-to-trap d i s t i l l a t i o n . Because of i t s explosive nature, the CK>2 was not allowed to warm up but was reacted immediately with S.0,F„ to produce C10_S0„F.^® A l l operations involving C10_ were Z O Z Z 5 Z carried out in a fumehood with blackened windows. - 44 -Nitrosyl chloride was prepared fron nitrosyl hydrogen sulfate, NOSO^ H, and sodium chloride following the procedure described in Brauer's Handbook.^ "'""'' Potassium hexachlorostannate and cesium hexa-chlorostannate were prepared from aqueous SnCl^ and an excess of the 112 al k a l i metal halide following the procedure for K^SnCl^. Nitrosyl 113 hexachlorostannate was prepared from N0C1 and anhydrous SnCl^. Trifluoromethanesulfonic acid was prepared from BaCSO^CF^)^ and fuming sulfuric acid by J.R. Dalziel in our laboratory. More recently HSO^CF^ has become available from the 3M Company. C. Suppliers of Materials and Equipment The suppliers for some of the materials and equipment used in this work are listed here. Swagelock f i t t i n g s , Columbia Valve and Fitting Co., Vancouver, B.C. Metal fittin g s using metal or teflon seals to join pieces of metal tubing. Whitey valves, 1KS4, Columbia Valve and Fitting Co. and Hoke valves, #431, Hoke Inc., Creskill, N.J., high vacuum metal values. Whitey valves had the advantage that they could be disassembled for cleaning or refurbishing. Fischer and Porter 4 mm teflon stem stopcocks, Fischer and Porter, Warminster, Pa., Kontes valves, Kontes of I l l i n o i s , Franklin Park, I l l i n o i s , and Rotaflow valves, Quickfit and Quartz, Staffordshire, England. Greaseless taps for glass systems which produced vacuum tightness via a teflon to glass seal. - 45 -Teflon coated s t i r r i n g bars, Canlab, Vancouver. Molecular sieves, Linde Air Products, distributed by Fisher Scientific. Infrared c e l l windows, Harshaw Chemical Corp., Cleveland, Ohio. AgCl, AgBr, KRS5, and Csl windows 0.5 cm x 2.5 cm diameter and AgCl and AgBr sheets 1 mm thick. - 46 -CHAPTER III METHYLTIN SULFONATES A. Introduction 1. Previous work Organotin derivatives of a wide range of oxyacids including 114,115 . 116-118 118 . . . . t 119-121 sulfates, nitrates, perchlorates, and carboxylates are known. These compounds were usually prepared by the silver salt method, e.g. 118 (CH 3) 3SnCl + AgC104 *- (CH ) 3SnC10 4 + AgCl Most of these compounds are dimethyltin or trimethyltin derivatives, 31 however, some monomethyltin compounds such as methyltin trinitrate 119 121 121 and mixed methylhalotin derivatives ' such as (CH^)^ClSnCOOCH^ are reported. The number and variety of sulfonic acid derivatives i s more limited. Before our study began (CH 3)2Sn(S0 3CH 3)2 had been prepared 79 by the silver salt method and (CH3) 3 S n S 0 3CH 3 by insertion of SQ3 78 into a tin-carbon bond in (CH^^Sn. Some other dialkyltin bismethane- and bisethanesulfonates were prepared by reaction of 79 dialkyltin oxides with the appropriate sulfonic acids, and the reaction of (C^Hg^SnO with paratoluenesulfonic acid was used to - 47 -122 prepare dibutyltin bistoluenesulfonate. Since our work began, the preparations of (CH^SnSC^F and (CH3) 3SnSC>3CF3 by the reactions of 76 ((CH 3) 3Sn) 20 with and ^ ^ ( ^ ^ r e s P e c t i v e l y w e r e reported as well as the much simpler preparations of (CH3)3SnSC>3CF3, (CH 3) 2Sn(S0 3CF 3) 2 and (CH 3) 2Sn(S0 3C 2F 5) 2 by the solvolysis of (CH 3) 4Sn in the appropriate amounts of the sulfonic a c i d s . 7 7 The following comments can be made about the previous work on methyltin sulfonates. (1) Only dimethyltin bisulfonates or trimethyltin sulfonates are reported, there are no reports of monomethyltin sulfonates or mixed derivatives such as methylhalotin sulfonates. (2) The preparative routes are usually tedious and often require starting materials which are d i f f i c u l t to obtain or handle. (3) A general unified preparative method is lacking. (4) No structural studies have been attempted. 2. Promising routes to sulfonates Two promising preparative routes which should complement each other are solvolysis reactions in the sulfonic acids and ligand  redistribution reactions. (a) Solvolysis Acid solvolysis has found previous application in organotin chemistry . ,. 119,123,124 , , . ... 125,126 using carboxylic acids or hydrogen halides as solvents and tetraalkyltins or tetravinyltins as substrates. These reactions generally result in cleavage of one or two tin-carbon bonds, 127 but in exceptional cases a l l four Sn-C bonds are broken. The - 48 -mechanism for the substitution was shown in the case of the hydrogen 126 halides to be two competitive consecutive second order reactions k l R,Sn + HC1 — R „ S n C l + RH 4 3 k2 R 3SnCl + HC1 — R 2 S n C l 2 + RH At the outset of this study mainly cleavage of tin-carbon bonds had been reported and no systematic study involving methyltin halides, for example, had been made. More recently, the solvolysis of triorgano-tin chlorides in perfluorocarboxylic acids resulting in the preferential cleavage of the tin-carbon bond to form dialkylchlorotin perfluoro-carboxylates has been reported."'"28 (b) Ligand redistribution Ligand redistribution reactions are an area of preparative chemistry which has been extensively studied for many organometallic systems including organotin compounds. Much of this work has been reviewed f a i r l y recently and the general topic of ligand redistribution 13 129 has been discussed extensively. ' A common type of ligand redistribution reaction is one, exemplified by mixtures of tin halides, in which a s t a t i s t i c a l distribution of a l l possible combinations is produced, e.g. 130 SnCl. + SnBr. — - SnCl.Br + SnCl.Br. + SnClBr. 4 4 — — 3 L I 3 - 49 -A second type, more appealing from a preparative point of view, is a ligand redistribution reaction in which only one product is formed preferentially. Examples of reactions of this type are the preparations of alkyltin and aryltin halides from tin tetrahalides and tetraalkyltins 131 132 or tetravinyltins, ' e.g. 3R,Sn + SnCl. 4 4 4R3SnCl 132 In these reactions single products could be obtained by using the appropriate concentrations of the reagents. Only one example of a ligand redistribution reaction of an oxyacid is known, the redistribution 121 reaction of an organotin carboxylate with a diorganotin dichloride. R 2SnCl 2 + R^SnCOOR" R2SnClCOOR" + R^SnCl and R 2SnCl 2 + 2R^SnC00R" R2Sn(COOR")2 + 2R^SnCl B. Preparations 1. (CH 3) 2Sn(S0 3X) 2 Dimethyltin bisfluorosulfate was prepared by the solvolysis of either (CH^Sn, (CH^SnCl or (CH^SnC^ in an excess of HS03F at room temperature. Methane was a byproduct when (CH^^Sn was used, methane and HC1 when (CH^SnCl was used and HC1 with ( C H ^ S n C l ^ HSO F (CH 3) 4Sn + 2HS03F - ( C H ^ S n ^ F ^ + 2CH4 HSO F (CH 3) 3SnCl + 2HS03F ^ (CH3) 2Sn(S0 3F) 2 + CH4 + HC1 - 50 -HSO F (CH 3) 2SnCl 2 + 2HS03F ^ (CH 3) 2Sn(S0 3F) 2 + 2HC1 A typical reaction with (CH3>3SnCl was carried out by d i s t i l l i n g an approximately five-fold excess of HS03F onto one to two grams of dried (CH3>3SnCl i n a two part round bottom flask reactor. The (CH^SnCl reacted immediately with the evolution of HC1 to produce a clear solution. Approximately ten minutes later methane was evolved, and clear plate-like crystals precipitated out. These crystals were then fi l t e r e d and washed with HS03F. If a larger excess of HS03F was used in the reaction the product remained in solution and could be isolated by removing the HS03F by vacuum d i s t i l l a t i o n . In this case the yield is virtually 100%, e.g. in one preparation 1.83 g (9.20 mmoles) of (CH 3) 3SnCl yielded 3.22 g (9.29 mmoles) of (CH 3) 2Sn(S0 3F) 2. The (CH 3) 2Sn(S0 3F) 2 forms as hygroscopic colorless plate-like crystals which melt with decomposition at 253°. The stoichiometry was established by elemental analysis of C, H, Sn, S, and F. This reaction could not be arrested after the f i r s t step. This was tried both by using a greater excess of HS03F in an attempt to form a stable solvated (CH 3) 3Sn + cation which might not react further and by running the reaction at lower temperatures. (CH 3) 2Sn(S0 3CF 3) 2, (CH 3) 2Sn(S0 3Cl) 2, ( C H ^ S n ^ C H ^ and (CH 3) 2Sn(S0 3C 2H^) 2 can be prepared by analogous reactions using the appropriate sulfonic acids but higher reaction temperatures and longer reaction times have to be used for the less reactive methane- and ethanesulfonic acids. Previous reports on dimethyltin bissulfonates gave melting points of 327° 7 7 for (CH 3) 2Sn(S0 3CF 3) 2 and 325° 7 9 for (CH 3) 2Sn(S0 3Me) 2. - 51 -Table 10. Dimethyltin Sulfonates. Compound mp (CH 3) 2Sn(S0 3F) 2 253 (CH 3) 2Sn(S0 3CF 3) 2 336-7 (CH 3) 2Sn(S0 3Cl) 2 370-5 (CH3)2Sn(SO CH )-2 312 (CH 3) 2Sn(S0 3C 2H 5) 2 272 (CH 3) 2Sn(S0 3C 6H 4CH 3) 2•2H 20 (CH 3) 2Sn(S0 3C 6H 4CH 3) 2 >300 Analysis ( % ) Calculated Found C 6.92 7.01 H 1.72 1.80 Sn 34.22 34.54 S 18.49 18.67 F 10.96 11.09 C 10.74 10.70 H 1.35 1.50 F 25.51 26.52 C 6.06 5.04 H 1.59 3.04 CI 18.67 15.4 C 18.54 18.37 H 4.67 4.59 C 19.62 19.78 H 4.39 4.11 C 36.45 36.18 H 4.58 4.13 C 39.10 39.25 H 4.11 4.0 - 52 -(CH 3) 2Sn(£-S0 3C 6H 4CH 3) 2«2H^O can be prepared by reacting (CH 3) 2SnCl 2 and paratoluenesulfonic acid in aqueous solution at room temperature and collecting the resulting insoluble solid product on a f r i t . The anhydrous salt can be obtained by heating the hydrate to 120°. A reaction similar to the one used to prepare (CH„)_Sn(S0„C-H.CH„) • J 2 3 o 4 3 z 2H20 using (CH 3) 2SnCl 2 and HS03CH3 produced [(CH 3) 2SnCl]^ and not (CH3)2Sn(SC*3CH3)2. The product was identified by comparing i t s infrared spectrum and melting point to those of [(CH 3) 2SnCl] 20. 2. (CH 3) 2Sn Derivatives of oxyacids The dimethyltin derivatives of some other oxyacids were synthesized either by literature methods"'"22> 133,134 ^ ^ m i x i n g saturated aqueous solutions of (CH 3) 2SnCl 2 with aqueous solutions of the acids or the sodium salts of the acids. The (CH 3) 2Sn salts a l l precipitated out of solution as very fine precipitates and could be f i l t e r e d and then dried in an oven at 120°. Several salts precipitated out as the basic salts, i.e. as (CH 3) 2SnA 2'OSn(CH 3) 2. Analytical results are lis t e d in Table 11. 3. CH 3ClSn(S0 3X) 2 Methylchlorotin bisfluorosulfate could also be prepared by a solvolysis reaction in HS03F. When an excess of HS03F was d i s t i l l e d onto 0.500 g (2.08 mmoles) of CH 3SnCl 3 in a round bottomed flask, HC1 was evolved and a clear solution formed. If the HS03F was removed by vacuum d i s t i l l a t i o n , 0.765 g (2.08 mmoles) of CH 3ClSn(S0 3F) 2 was isolated as a hygroscopic white crystalline solid which decomposed at 162°. - 53 -Table 11. Elemental Analyses of (CH ) Sn Salts. Compound C ( % ) H C/o) Calculated Found Calculated Found (CH 3) 2Sn(C 6H 5P0 3H) 2 36.30 36.24 3.92 3. 83 (CH 3) 2Sn(C 6H 5P0 2H) 2 1 2 2 39.02 39.24 4.21 4. 29 [(CH 3) 2Sn] 3(P0 4) 2 11.61 11.11 3.01 2. 80 122 (CH 3) 2Sn(H 2P0 2) 2 8.61 8.59 3.62 3. 54 133 [(CH 3) 2Sn] 3(As0 4) 2 J 9.95 9.32 2.52 2. 62 (CH 3) 2SnC 6H 5As0 3 1 3 3 27.53 27.66 3.18 3. 03 (CH 3) 2SnC 6H 5CH 2As0 3 29.77 29.90 3.61 3. 52 133 (CH 3) 2Sn[(CH 3) 2As0 2] 2- L J J 17.05 17.20 4.29 4. 00 134 (CH 3) 2SnC 4H 40 4-(CH 3) 2SnO 3 * 22.19 22.36 3.76 3. 47 (CH 3) 2SnC 8H 40 4 38.38 37.10 3.22 2. 91 (CH3)2SnW04-(CH3)2SnO 10.83 10.68 2.72 2. 80 (CH 3) 2SnMo0 4 1 3 4 7.77 8.15 1.96 2. 05 - 54 -HSO F CH 3SnCl 3 + 2HS03F CH 3ClSn(S0 3F) 2 + 2HC1 The stoichiometry was established by elemental analysis. Solutions of CH 3SnCl 3 in HS03F in one part erlenmeyer reactors were heated eventually to 140° i n attempts to prepare CH 3Sn(S0 3F) 3 > These attempts were a l l failures, and the products were identified as Cl^SnCSO^)^ presumably via: HSO F CH 3SnCl 3 + 2HS03F CH 3ClSn(S0 3F) 2 + 2HC1 HSO F CH 3ClSn(S0 3F) 2 + HC1 — C l 2 S n ( S 0 3 F ) 2 + CH4 CH 3ClSn(S0 3CF 3) 2 was prepared by a similar solvolysis of CH 3SnCl 3 in HS03CF3. When a solution of CH 3SnCl 3 in HS03CF3 was heated a tin(II) compound tentatively identified as Sn(S0 3CF 3) 2 was produced. No attempts were made to extend this work by preparing derivatives of HS03C1, HS03CH3 or HS0 3C 2H 5. 4. (CH 3) 3SnS0 3X Trimethyltin fluorosulfate was prepared by slowly adding HS03F from a burette to an approximately three-fold excess of (CH^.^Sn-.cooled to -90' in a 50 ml round bottomed- flask. The burette was attached to the flask by a B19 ground glass union in order to keep the reactants and products isolated from the atmosphere. After addition of the HS03F the burette was replaced by a teflon stem stopcock adaptor through which the flask could be attached to the vacuum line. These components - 55 -Table 12. Other Methyltin Sulfonates, Compound mp Analyses Calculated Found CH 3ClSn(S0 3F) 2 161-2 CI S F 9.68 17.46 10.35 9.84 17.70 10.33 CH 3ClSn(S0 3CF 3) 2 178-80 S F 14.04 24.50 13.85 24.62 (CH 3) 3SnS0 3F 109-11 S F 12.20 7.23 12.29 7.04 (CH 3) 3SnS0 3CH 3 144-5 C H 18.15 4.67 18.59 4.48 (CH 3) 2ClSnS0 3F 108 Sn S F 41.90 11.32 6.71 42.25 11.49 6.31 CH 3Cl 2SnS0 3F 112-5 Sn S CI F 39.47 10.56 23.35 6.26 38.7 10.78 23.09 5.96 - 56 -reacted exothermically at -90° to produce (CH^)3SnS03F, which was insoluble in the excess (CH^^Sn and CH^. -90° (CH 3) 4Sn + HS03F (CH 3) 3SnS0 3F + CH4 The product was isolated by d i s t i l l i n g off the excess (CH 3) 4Sn. (CH 3) 3SnS0 3F is a hygroscopic white solid which melts without decomposition at 108° and can be sublimed under vacuum at 70-75°. This reaction also goes to completion, e.g. in one preparation, 1.50 ml (2.62 g, 26.1 mmoles) of HS03F reacted with (CH^Sn to yield 6.87 g (26.1 mmoles) of (CH 3) 3SnS0 3F. Trimethyltin fluorosulfate i s soluble in HS03F and reacts with HS03F to produce (CH 3) 2Sn(S0 3F) 2, providing yet another route to this compound. Elemental analysis results for (CH 3) 3SnS0 3F are listed in Table 12. (CH 3) 3SnS0 3CF 3 was prepared by Schmeisser in a similar manner and this method could be extended to the production of (CH 3) 3SnS0 3CH 3 from (CH 3) 4Sn and HS03CH3 at 15°. Attempts to make (CH 3) 3SnS0 3Cl using HS03C1 and the same method surprisingly yielded (CH 3) 3SnS0 3CH 3 as the sole product. The reaction scheme is probably: (CH 3) 4Sn + HS03C1 *- [(CK) 3SnS0 3Cl] + CH4 [(CH 3) 3SnS0 3Cl] + CH4 —*- (CH3) 3SnS0 3CH 3 + HC1 The product was identified as (CH 3) 3SnS0 3CH 3 by elemental analysis and comparison of the infrared and Mossbauer spectra with those of - 57 -(CH ^SnSO^Hg. The melting point of (CH 3) 3SnS0 3CF 3 prepared here i s 76-78° in good agreement with the previous report of 74-75°. 7 7 The 76 previously reported mp for (CH3)3SnSC>3F was > 100° again in agreement with our results. 5. (CH 3) 2ClSnS0 3F and CHgCl^SnSO F Dimethylchlorotin fluorosulfate was prepared in a ligand redistribu-tion reaction when 0.492 g (1.42 mmoles) of (CH 3) 2Sn(S0 3F) 2 was reacted with 0.311 g (1.42 mmoles) of (CH 3) 2SnCl 2 in dry chloroform at room temperature for several hours in a round bottomed flask attached directly to a glass f r i t via a B24 ground glass union. (CH 3) 2Sn(S0 3F) 2 + (CH 3) 2SnCl 2 — ^ 2 (CH^ClSnSC^F An excess of (CH 3) 2SnCl 2 could also be used in this reaction. The (CH 3) 2ClSnS0 3F i s insoluble in CHC13 and could be isolated by f i l t r a t i o n followed by washing with CHCl-j. (CH 3) 2Sn(S0 3F) 2 is also insoluble in CHC13 and could have been a contaminant i n the product, however, the reaction went 100% to completion and no (CH 3) 2Sn(S0 3F) 2 was evident in the product. Methyldichlorotin fluorosulfate was prepared in a similar reaction of CH 3ClSn(S0 3F) 2 with CH^nCl^ CH 3ClSn(S0 3F) 2 + CH 3SnCl 3 —>• CH Cl 2SnS0 F Both of these products are hygroscopic white solids. Attempts to - 58 -prepare the analogous SO^CF^ derivatives by similar methods were a l l unsuccessful, as were attempts to prepare any of these derivatives by solvolysis reactions. 6. (CH 3) 2Sn(S0 3F)(S0 3 C F p Mixed acid derivatives could be prepared by reacting a trimethyltin sulfonate with one of the other sulfonic acids. Dimethyltin fluoro-sulfate trifluoromethylsulfate was prepared in this manner by adding an excess of HS03CF3 to (CH3)3SnSC>3F in a two-part reactor in the dry box. Methane was released in the reaction and the (CH3)2Sn(SC>3F) (S0 3CF 3) was isolated by removing the excess HS03CF3 by vacuum d i s t i l l a t i o n . HSO CF (CH 3) 3SnS0 3F + HS03CF3 - — ^ (CH^SnCSO^) (S0 3 C F 3 ) (CH 3) 2Sn(S0 3FX(S0 3 CF 3 ) is a hygroscopic white solid with a decomposition point of 315°. The elemental analysis results are Sn calc, 29.91; found, 30.26 and F calc, 19.15; found 18.86. 7. CH 3Sn(S0 3F) 3 Methyltin trisfluorosulfate i s the only member of the series of compounds described by the formula (CH„) CI Sn(SO.F). which could r J 3 x y 3 4-x-y not be prepared. As already mentioned, in one attempt to prepare CH 3Sn(S0 3F) 3, CH 3SnCl 3 was heated to 140° in the presence of HS03F in order to replace the last Cl~ by SO-jF-, but only Cl 2Sn(S0 3F) 2 was isolated. In another attempt CH 3ClSn(S0 3F) 2 was reacted with peroxydisulfuryl difluoride in a small one-part reactor with a teflon - 59 -stem stopcock. This reaction, as well as reactions of (CH^J^SnCl^ and (CH„)0SnCl with S.CvF., produced only Sn(SO.F).. Because S.0.Fo is 3 3 2. o 2. 3 H z b z an oxidizing agent and the substrates contained oxidizable methyl groups, only very small quantities 200 mg of substrate) were used and the vessels were l e f t in a safe place while the reactions proceeded. One of the several reactions of this type which were attempted did explode. C. Structural: Studies 1. (CH 3) 2Sn(S0 3F) 2 Of these compound*, dimethyltin bisfluorosulfate is the most thoroughly investigated and therefore i t can be used as a model for discussion of the structural and bonding possi b i l i t i e s of the others. The crystal structure of (CH 3) 2Sn(S0 3F) 2, which has been 135 25 determined by X-ray diffraction, i s similar to that of (CH 3) 2SnF 2 and consists of sheet-like polymers with two equivalent bidentate bridging S0 3F groups. As shown in Figure 9 the geometry at each tin is octahedral with the methyl groups above and below the Sn(S0 3F) 4 . plane, i.e. in a trans octahedral arrangement about tin . The tin-carbon bond lengths are short and a l l three S-0 bonds are equivalent within experimental limits and equal to the value of r for the ionic KS0 3F.^ The Sn-0 bond is longer than is usually found for covalent ° 18 Sn-0 bonds, e.g. in SnCl^-2SeOCl2 Sn-0 is ^ 2.12 A. Both the short Sn-C bond length and the equivalence of the S-0 bond lengths suggests 2+ that the crystal consists of (CH 3) 2Sn cations and S0 3F anions. Other structural information, however i s inconsistent with this fomulation. - 60 -Figure 9 The Crystal Structure of (CH3)2Sn(S03F)* One polymeric layer viev.'ed along a* 1 K2) © S n © S O o © F o C Bonddistances A S n - C S n - O( l ) S n - 0 ( 2 ) S - O O ) S - 0(2) S - 0(3) S - F 2 . 0 6 5 (9) 2 . 2 7 0 (6) 2 . 2 7 1 (7) 1 . 4 3 7 (8) 1 . 4 2 8 (7) 1 . 4 2 2 (11) T . 5 0 4 ( 8 ) * F H Allen, d.A.Lerbscher and 3.Trotter. O.Chem. Soc. (A) 2507,(1971). - 6 1 - -T a b l e 13 R e p o r t e d i n t e r a t o m i c d i s t a n c e s a n d s e l e c t e d b o n d a n g l e s f o r s o m e f l u o r o s u l f a t e s B o n d p a r a m e t e r * ' K S O ^ F ' " ( C H ^ S n C S O F ) 2 ~ " R c „ [ A ] 1 . 5 8 1 . 5 0 S - F R g _ 0 [ A ] 1 . 4 3 1 . 4 2 R S n - 0 [ l ] 2 ' 3 0 2 x R s _ Q , b ) [ & ] 1 . 4 3 1 . 4 3 $ O - S - 0 ( a v . ) 1 1 2 . 9 ° 1 1 1 . 1 ° * 0 - S - F ( a v . ) 1 0 5 . 8 ° 1 0 6 . 4 ° d e n o t e s u n c o r r e c t e d f o r t h e r m a l m o t i o n ( a v ) d e n o t e s a v e r a g e 0 ' d e n o t e s t h e o x y g e n a t o m b o n d e d t o b o t h S a n d S n - 62 -Both the vibrational spectra and the conductivity results indicate a more covalent interaction between the (CH^^Sn and the SO^F. The vibrational results with an assignment of the peaks are listed in Table 14. The assignment of the nine SO^F modes is made by comparison of these spectra to those of a large number of C symmetry s f l u o r o s u l f a t e s ^ 9 , 7 ^ some of which are Raman spectra of liquids which can be analysed more ful l y by making use of polarizability measurements. The infrared spectrum of (CH )_Sn(SO F) shows C symmetry for the j /. j c. S SO^F group (9 modes) with a large splitting of the previously degenerate E symmetry SO^ stretching mode. The splitting i s much larger than is found for ionic salts whose symmetry is reduced by crystal effects, 64 e.g. NOSO^F, and i t s magnitude i s similar to the size of the splittings found in other covalent fluorosulfates. The other two E modes, v c and v, in the ionic, are also s p l i t by large amounts 5 b compared to the splittings found in NOSO^F. The positions of the SO^  stretching modes (1350, 1180 and 1080 cm"1 vs. 1408, 1206 and 884 cm"1 for BrSO^F) are indicative of bidentate SO^F and since bidentate bridging is confirmed in this case these relative positions of v SO^'s can be used as a guide for assignment of the vibrations of other tin fluorosulfates. Another indication of covalency is the position of vS-F. In (CH 3) 2Sn(S0 3F) 2, vS^ -F i s shifted from the position of 750 cm"1 6 3 —1 found for ionic fluorosulfates to the 800-900 cm region where vS-F is found in covalent f l u o r o s u l f a t e s . ^ 9 The higher energy of the S-F bond is consistent with the shortness of the S-F bond as shown by the crystal structure. In the vSn-C region, mutual exclusion of i r - 63 -Table 14. Vibrational Spectra of (CH ) Sn(SO F) IR Ccm"1) Raman (c«* Assignment 3048 m V H 3 2952 mw 2944 m V H 3 2860 vw 2871 m 2425 w 1455 mw 1350 vs,br 1354 s vS0 3 (A") 1222 w 1180 vs,br vS0 3 (A') 1088 m,sh 1088 s VS03 (A*) 1072 s,br 827 m,sh 826 vS-F (A') 798 vs,br p(Sn)-CH3 650 w 620 m,sh 610 6S03(A") 590 ms 584 6S03 (A') 576 s v Sn-C a 554 ms 551 mw,sh 531 vs 6S03(A') v Sn-C s 417 s 420 w PS02 (A") 360 w 367 mw pS03F (A') 304 w 320 mw vSn-0 s = strong V stretch m = medium <5 bend w = weak P rock br= broad a asymmetric sh= shoulder s symmetric - 64 -and Raman peaks is observed for the two tiri^-carbon stretching modes as expected for a linear C-Sn-C arrangement. The positions of the Sn-C vibrations are consistent with those of other octahedral (CH^^Sn compounds. The el e c t r i c a l conductivity study on (CH3)2Sn(SC>3F)2 in HSO^F shows the expected large increase in conductivity when a fluorosulfate is added, however, the increase i s only one-half as great as is observed for the same concentrations of KSC^F. If (CH 3) 2Sn(S0 3F) 2 were ionic, i t would be expected to dissociate completely in HS03F and the increase in conductivity should be twice that of KSC^F. The observed increase means that only approximately one-quarter of the S03F groups in (CH 3) 2Sn(S0 3F> 2 are dissociated to SC^F in HSC^F and therefore there must be considerable covalent interaction between (CH 3) 2Sn and SG^F to hold these units together in solution. The values of the Mossbauer parameters for (CH 3) 2Sn(S0 3F) 2 are isomer shift (6), 1.82 mm/sec and quadrupole splitting (A), 5.54 mm/sec. There is also an appreciable room temperature effect, R = 0.089. The isomer shift and quadrupole splitting are exceptional. 5.54 mm/sec is the largest A known for tin and 6 is among the largest for organotin compounds. In order to explain these results a bonding scheme which predicts a large s electron density at t i n is needed to explain the high isomer shift. The s electron density must also be concentrated along the z axis to produce a large electric f i e l d gradient and hence a large A. - 66 -Table 15. El e c t r i c a l Conductivity of (CH ) Sn(SO F ) 2 in HSC^F at 25°. K (ohm cm ) m Y 2.0 X 10"3 7.0 x 10~4 0.60 6.0 x 10"3 4.2 x 10~3 0.48 1.0 X 10"2 8.8 x 10~3 0.41 1.4 X lO" 2 1.47 x 10~2 0.36 1.8 X lO" 2 2.12 x 10~2 0.32 Table 16. Mossbauer Data for (CH„)0Sn(SO_X) Compound 6 (mm/sec) A (mm/sec) R (CH 3) 2Sn(S0 3F) 2 1.82 5.54 0.089 (CH 3) 2Sn(S0 3CF 3) 2 1.79 5.51 0.129 (CH 3) 2Sn(S0 3Cl) 2 1.75 5.20 0.197 (CH 3) 2Sn(S0 3CH 3) 2 1.52 5.05 0.129 (CH 3) 2Sn(S0 3C 2H 5) 2 1.52 4.91 0.117 (CH 3) 2Sn(S0 3C 6H 4CH 3) 2 1.51 4.85 0.18 (CH 3) 2SnF 2 1.23 4.52 (CH_)„SnS0. 3 2 4 1.57 4.94 90 (CH 3) 2Sn(N0 3) 2 1.62 4.13 V - 67 -2. Other methyltin sulfonates The vibrational and MSssbauer results for (CH^)^SnCSO^F)^ can now be used as a guide in the investigation of the other methyltin sulfonates. (a) (CH 3) 2Sn(S0 3X) 2 A series of compounds with the general formula (CH 3) 2Sn(S0 3X) 2 where X = F, CF 3 > CI, CH3, , and C^R^CU^ have been prepared. The MSssbauer parameters for these compounds are shown in Table 16. Both 6 and A values - for this series f a l l into f a i r l y narrow ranges and a l l compounds have noticeable room temperature effects. These similarities suggest common possibly polymeric structures for these compounds similar to that of (CH 3> 2Sn(S0 3F) 2. The decrease in both 6 and A as X changes from F through CF 3, CI, CH3» and C ^ to C^R^CH^ appears to 136 proceed parallel to the decrease i n electronegativity of X for the same series. Infrared and when possible Raman spectra were recorded for the members of this series. Because of experimental d i f f i c u l t i e s and the more complex nature of some of the sulfonate groups, a complete analysis of the spectra could not be made. However, the same pattern for the S0 3 stretching modes indicating bidentate sulfonates could be seen i n a l l cases. The MSssbauer results and vibrational frequencies in the Sn-C region for the Me2Sn derivatives of some oxyacids are lis t e d in Table 18, and as i s shown in the table the isomer shifts vary from 1.03 mm/sec for (CH 3) 2Sn((CH 3) 2 A s°2^2 t 0 1 , 5 1 m m / s e c f o r t h e P a r atoluene-- 68 -Table 17. Vibrational Frequencies of (CH )_Sn(SO X) Compounds. (CH 3) 2Sn(S0 3CF 3) 2 (CHg) 2Sn(S0 3Cl) £ (CH 3) 2Sn(S0 3CH 3> 2 ( C H 3 ) ^ ( S O ^ H ^ 2 IR Raman IR IR IR Raman 1405 w 1340 s 1300 m 1314 w vS03(A") 1324 s 1325 1320 s,br 1240 s,br 1250 s,br 1260 w 1292 w 1231 s 1225 1212 m 1195 vs vS03(A') 1150 s,br 1155 1160 s,br 1090 s,br 1200 s 1196 m 1090 sh 1080 sh vS03(A*) 1035 s 1036 s 1050 m 1030 s ,br 1035 s 982 w 965 sh 975 m 990 w 905 m 935 w 821 vs 825 s 804 s 810 s 774 mw 783 775 s 778 s 790 sh 745 s 735 w 645 s 642 625 m 630 s 600 s 591 w ,m 598 m 583 m 580 575 s 572 s 555 s 534 m 530 sh 531 vs 532 s 520 m,sh 522 sh 514 m 512 s 520 m 480 sh 406 s 385 m,sh 383 390 s 385 s 370 ms 351 342 m 339 m 326 331 322 280 w 220 - 69 -Table 18. Dimethyltin Derivatives of Oxyacids. Compound MSssbauer (mm/sec) A (mm/sec) R Vibrational IR (cm Raman (cm (CH 3) 2Sn(C 6H 5P0 3H) 2 1.34 4.57 0. 26 590m' 522s. (CH 3) 2Sn(C 6H 5P0 2H) 2 1.17 4.30 0. 19 582nr 518s [(CH 3) 2Sn] 2(P0 4) 2 1.23 3.72 0. 16 573m,528m 575w,531s (CH 3) 2S n(H 2P0 2) 2 1.23 4.35 0. 10 580s 518s [(CH 3) 2Sn] 3(As0 4) 2 1.15 3.36 0. 23 573m,526w 572w,529s (CH 3) 2SnC 6H 5As0 3 1.06 3.20 0. 23 572m,526w 579w,525s (CH 3) 2SnC 6H 5CH 2As0 3 1.06 3.10 0. 18 573m,526w 572w,529s (CH 3) 2Sn[(CH 3) 2As0 2] 2 1.03 3.73 0 580m (CH 3) 2SnC 4H 40 4.(CH 3) 2SnO 1.18 3.43 0. 15 581m,527s 582w,533s (CH 3) 2SnC 8H 40 4r 1.41 3.94 0 582m (CH3)2SnW04-(CH3)2SnO 1.17 3.17 0 572s,529w 566w,531s (CH3)2SnMo04 1.37 4.19 0.56 575s 534s (CH 3) 2SnC 20 4.H 20 X J° 1.55 4.65 (CH 3) 2SnW0 4 1 3 8 1.39 3.53 (CH 3) 2SnMo0 4 1 3 8 1.42 4.10 (CH 3) 2SnCr0 4" L J° 1.28 2.98 - 70 -sulfonate. This variation i s much larger than would be expected i f the 2+ same tin containing ion, (CH^^Sn , existed in a l l these compounds. 2-For comparison the SnF ion has 6's ranging from -0.48 to -0.40 with b + + 137 2-five different cations from Na to C10„ and SnCl, has 6 from 2 6 +0.42 to +0.52 mm/sec with four different cation ranging from NH^"*" to NO +. 1 3 7 (b) (CH 3) 3SnS0 3X (CH 3) 3SnS0 3F, (CH 3) 3SnS0 3CF 3 and (CH 3) 3SnS0 3CH 3 are a l l hygroscopic white solids at room temperature. The geometries of these compounds, to be consistent with (CH^SnF, (CH3> ^ nClO^, the dimethyltin bis-sulfonates, and other similar compounds, would be trigonal bipyramidal with planar CH3Sn groups and axial S03X bridging groups forming linear polymeric solids. The infrared spectral frequencies which are listed in Table 19 confirm this expectation. This result disagrees with the report by Schmeisser who claims that i r data, not reported, for 77 82 83 (CH 3) 3SnS0 3CF 3 agree with spectra of ionic trifluoromethylsulfonates. ' Looking f i r s t at the (CH 3) 3SnS0 3F spectrum, the nine fundamentals expected for a fluorosulfate with C g symmetry are observed and the S0 3 stretches are at 1335, 1200, and 1068 cm ^ i n very similar positions to the corresponding bands in (CH 3> 2Sn(S0 3F) 2. The other six S03F modes also agree with those of (CH 3) 2Sn(S0 3F> 2. They are the S-F stretch at 820 cm \ S0 3 bending modes at 630, 596 and 578 cm ^ and S03F rocking and torsion modes at 410 and 370 cm ^. Only one band in the 500-600 cm ^ region can be assigned to tin-carbon stretching modes. This mode, v Sn-C at 555 cm \ i s found in the same position a Table 19. Vibrational Frequencies of (CH3)3SnS03X. (CH ) SriSOgF (CH 3) 3SnS0 3CF 3 (CH 3) 3SnS0 3CH 3 1408 w, 6 CH_ a 3 1355 s, vSO (A11) 1400 1319 w, 6 CHL a 3 vs,br, vS03(A") 1418 1345 w, 1405 w,sh 6 CH„ a 3 m,sh, 1337 m, 6 CH„ s 3 1335 m, 6 CHQ s 3 1226 vs, vCF 3(E) 1266 s, vS03(A") 1218 s, vS03(A') 1254 vs, 1196 v s 1179 s,br, vCF 3(A 1) 1207 1196 vw, vw, 1123 w, 2 x 555 1145 s, vSO (A») 1112 vs, vS03(A') 1068 vs, vS03(A') 1026 s,vS03 (A') 1035 s, vS03(A') 820 s,sh, vS-F (A') 968 w, 778 vs, CH3 rock 796 s, CH3 rock 810 s,sh, CH3 rock 771 m, 'WS-C (A 1) 793 s, vS-C 630 m, 6S0 3(A M) 633 ms, 6S03(A") 562 s, 6S03(A') 596 s, 6S03(A') 577 ms, 6CF3 (E) 550 vs, v Sn-C a 578 s, 6S03(A') 531 s, 5S03(A') 555 s, v Sn-C ' a 516 ms, 6S03(A') 555 s, v Sn-C a 530 m,sh, 6S03(A') 425 mw, 410 ms, S03F rock (A") 517 356 ms, 6S03(A') ms, S0 3CF 3 rock 352 346 m,sh S0„CHo rock (A") m, 3 3 370 m,br, S03F rock (A') 347 330 s, <^5CF3 (A2) m, S0 3CF 3 rock 330 m, vSn-0 298 ms, vSn-0 317 m, vSn-0 275 m,sh, S02CH3 rock (A*) - 72 -118 as i t is in other similar five-coordinate trimethyltin compounds. There are no IR peaks i n the 520 cm 1 region where the symmetric stretch would be expected. The absence of v Sn-C in the infrared i s s consistent with a trigonal bipyramidal geometry because for only the asymmetric stretch would be IR active. Both tin-carbon stretches should be Raman active but Raman spectra could not be obtained. The infrared spectrum of (CH^) .jSnSO^CF^ ± s n o t a s easily assigned as the spectrum of (CH^)^SnSO^F is because the SO^ and CF^ modes are found i n the same spectral regions. This is particularly obvious for ionic trifluoromethanesulfonates where the symmetries of both of these components are the same and extensive coupling results. It is therefore not surprising that two independent studies, both using normal coordinate analysis and extensive comparison to SO^CH^ , S03CD3 and S03CC13 arrive at different assignments for S0 3CF 3 . However, for the covalent S0 3CF 3 group found here, the symmetry of some of the S0 3 modes w i l l be reduced and this should lead to a clearer distinction between the S0 3 and CF 3 modes. Tentative assignment of a l l peaks in the infrared spectrum of (CH 3) 3SnS0 3CF 3 has been made using the (CH 3) 3SnS0 3F spectrum as a guide. Vibrational modes for the S0 3CF 3 group are found >at approximately 40 cm 1 lower energy than the corresponding S0 3F modes as found earlier for the pairs CF 3S0 2F and 139 FS02F and CF3S02C1 and FS02C1. As in the fluorosulfate, only the asymmetric tin-carbon stretch i s evident in the IR and not the symmetric stretch. The spectrum of (CH 3) 3SnS0 3CH 3 can be similarly assigned. It i s interesting that a duplication of a l l CH3 modes is observed here, as i s - 73 -expected with two different CH^ groups in the molecule. The results of the Mossbauer spectra are shown in Table 20 along with values for some similar compounds. No room temperature effect is observed for any of the trimethyltin sulfonates indicating perhaps weaker polymeric association than is found i n the corresponding dimethyltin derivatives. This absence does not rule out the possibility of polymeric association altogether. The isomer shifts of the sulfonates, in the narrow range of 1.43 to 1.52 mm/sec are among the highest reported for (CH^^SnClV) compounds and the quadrupole splittings are 98 140 also among the largest known, larger than haloacetates or sulfates for example. As was found for the dimethyltin bissulfonates, both <5 and A decrease with decreasing electronegativity of X. The behaviour of (CH^SnSO^ in HS03F is of interest in view of previous reports of a solvated (CH^^Sn"*" cation in H^SO^."^3 The plot of specific conductance vs. molality, shown in Figure 12, indicates far smaller conductance values than would be expected for a 1:1 electrolyte. The observed behaviour in solution suggests that there is incomplete breakdown of the polymers into ions. At higher concentrations the dissociation decreases further and some solvolysis of the tin-carbon bonds accompanied by CH^ release occurs to give dimethyltin bisfluorosulfate. The quantitative conversion of (CH^J^SnSO^F into (CH 3) 2Sn(S0 3F) 2 in HS03F according to: (CH 3) 3SnS0 3F + HS03F — ^ ( C H ^ S n C S O ^ + CH4 is easily accomplished in more concentrated solutions. It must be concluded that, in contrast to the findings in H.SO., the stronger T a b l e 2 0 •^^Sn M b s s b a u e r D a t a f o r ( C H - ) 3 S n ( I V ) C o m p o u n d s a t 8 0 K C o m p o u n d S ( n a n / s e c ) / \ ( m m / s e c ) ( C K 3 ) 3 S n S 0 3 F 1 . 5 2 4 . 6 1 ( ' c H 3 ) 3 S n S 0 3 0 P 3 1 . 5 2 4 . 5 7 ( C H 3 ) 3 £ n S 0 3 C H 3 1 . 4 3 4 . 2 1 ( C H 3 ) 3 S n S b ? 6 1 . 4 7 4 . 7 5 ( C H 3 ) 3 S n C 0 2 C ? 3 9 3 1 . 3 8 4 . 2 2 ( C H 3 ) 3 S n 2 S 0 4 1 4 ° 1 .37 4 . 0 6 ( C H 3 ) S ; i P 1 4 1 1 . 1 8 3 . 4 7 Figure 11 " s n MOSSBAUER S P E C T R U M of ( C H 3 ) 3 S n S 0 3 C H 3 at 8 0 ° K —i 1 1 1 1 1 1 > r - 6 - 3 O 3 6 DOPPLER VELOCITY ( mm / s ) - 75 -Table 21 Specific E l e c t r i c Conductance of (CH3>3SnS03F and (CH^ 2Sn(S0 3F) i n HS03F at 25°C (CH 3) 3SnS0 3F c (CH 3) 3SnS0 3F (CH 3) 3Sn(S0 3F) 2 [moles [ohai-l cm-1 ohm-1 CE-1 ohm-1 cm-1 2.5x10 -3 4.82x10 -4 4.65x10 -4 4.15x10 -4 5. u x i O 8.26x10 7 . 8 0 x 1 0 6.18::10 7.5x10 -3 11.22x10 -4 10.64x10 - A 8.75x10 -4 1.0x10 -2 14.20x10 -4 13.30x10 -4 10.50x10 -4 1.25x10 17.01x10 -4 15.74x10 -4 12.40x10 measured immediately after addition measured after 30 minutes. - 76 -- 77 -protonic acid HSO^F is not suited to the formation of (CH 3) 3Sn +(solv) via a simple solvolysis. (c) CH 3ClSn(S0 3X) 2, CH^l^nSC^F and (CH 3) 2ClSnS0 3F The results of the vibrational spectra of CH 3ClSn(S0 3F) 2, CH 3ClSn(S0 3CF 3) 2, CH3Cl2SnSC>3F and (CH 3) 2ClSnS0 3F are listed in Table 22. In a l l cases the position of the S03X vibrational modes agree with those of the corresponding (CH^SnSC^X and (CH 3> 2Sn(S0 3X) 2 spectra indicating that these compounds also have bidentate SO^K groups. This would lead to hexa-coordination for CH 3ClSn(S0 3X) 2, most lik e l y with the methyl and chlorine groups above and below the Sn(S n 3X) 2 plane. In CH 3Cl 2SnS0 3F both the asymmetric and symmetric tin-chlorine stretches are observed in both the IR and Raman indicating a non-linear SnCl 2 arrangement. The same result i s found for C,,Sn in (CH 3) 2ClSnS0 3F. The structures of these two compounds w i l l likely be trigonal bipyramidal with axial S03F groups and trigonal planar (CH 3) 2ClSn groups for (CH^ClSnSO^F and C l ^ C l ^ n group for CH 3Cl 2SnS0 3F in the equatorial plane, resulting in polymers much like those suggested for (CH 3) 3SnS0 3F, however, as indicated by the molecular structure of (CH 3) 2SnCl 2 (Figure 1) chlorine bridging i s also possible. The Mdssbauer parameters of these four compounds are list e d in Table 23 and the magnitudes of these parameters are consistent with the geometrical arrangements suggested. They certainly unambiguously indicate the presence of true compounds rather than reactant mixtures. Table 22. Vibrational Spectra of (CH„) CI Sn(SOoX). 3 x y -? 4-x-y Assignment CH 3ClSn(S0 3F) 2 CH3C12 SnS03F (CH 3) 2ClSnS0 3F CH 3ClSn(S0 3CF 3) 2 Assignment IR Raman IR Raman IR Raman IR 5. CHQ 1440 m 1405 vw 1403 w 1400 w 6 CH„ a 3 a 3 vS0 3(A M) 1361 s 1350s 1360w 1343 s 1320 w 1320 s,br vS03(A") 6 CH_ s 3 1220 sh 1220 sh 1213 w 1228 s, sh vCF 3(E) vS03(A') 1165 vs,br 1250 vs ,br 1250 m 1190 s,br 1195 s vCF3(A) vS0o(A') 1072 s 1095 s 1080 m 1072 s 1074 w 1155 s,sh vSO (A') 1065 sh 1022 s vS03(A') vSF 830 s,sh 825 m,sh 820 m,sh 825 w 816 ms PCH3 PCH3 812,vs,br 806 vs,br 795 vs,br 774 m,sh vS-C 1 640 vw 640 sh 6S03 » 6S03 610 m,sh 623 m 605 ms 610 w 607 s 603 w 628 m 6S03 6S03 590 s,sh 603 w 588 s 590 m 594 w 585 m 5CF3(E) v Sn-C a 578 m 582 m 575 m vSn-C vSn-C 578 s 585 m 576 vs 6 S 0 3 555 s 562 m 555 ms 560 555 s 562 m 518 m,sh 6S03 v Sn-C s 527 m 535 s 510 s 6CF3(A) P S O 3 420 m 423 m 405 w 405 w 409 m 416 w 375 s vSn-Cl v Sn-Cl a 385 ms 390 s 384 s 387 sh 362 s PS0CF3(A") vSn-Cl 345 m 345 m 350 m,sh vSn-0 v Sn-Cl s 366 s 368 s 325 mw PS03CF3(A') PS0,F 305 mw 308 mw 300 w 305 w 306 m 322 mw 250 m 247 w - 79 -Table 23. Mossbauer Data for (CH„) CI Sn(SO X) Compound 6 (mm/sec) A (mm/sec) R CH 3ClSn(S0 3F) 2 1.23 3.77 0.12 CH 3ClSn(S0 3CF 3) 2 1.19 3..70 0.20 CH 3Cl 2SnS0 3F 1.14 3.25 0.55 (CH 3) 2ClSnS0 3F 1.58 4.69 0.05 D. Discussion 1. Syntheses An extensive series of methyltin fluorosulfates could be prepared using one of two simple and versatile preparative routes; solvolysis of a methyltin substrate in HS03F or ligand redistribution reactions of two methyltin compounds. In some cases a compound could be made by either method. The solvolysis route could be extended to the preparation of other methyltin sulfonates by substituting the appropriate sulfonic acids for HS03F. (CH^SnSO^, (CH 3) 2Sn(S0 3F) 2 and CH3SnS03F were prepared by acid solvolysis and (CH3) 3SnS0 3F, (CH^ClSnSO^ and CH 3Cl 2SnS0 3F by ligand redistribution. These reactions are summarized in Table 24. (CH 3) 3SnS0 3CF 3, (CH 3) 2Sn(S0 3CF 3) 2 and CH 3ClSn(S0 3CF 3) 2 could be prepared by solvolysis reactions i n HS03CF3, analogous to those in HS03F, but ligand redistribution reactions using these trifluoro-methanesulfonates as substrates and similar experimental conditions were a l l totally unsuccessful. This lack of success i s possibly - 80 -Table 24. Preparative Routes to (CH„) Gl Sn(SO.F), 3 x y 3 4-x-y Solvolysis Reactions 1. ex(CH 3) 4Sn + HSO F " 9 0 ° * (CH ) SnSOgF + CH4 2. (CH 3) 4Sn + exHS03F 2 5 >- (CH 3> 2Sn(S0 3F) 2 + 2CH4 3. (CH 3) 3SnCl + exHS03F 2 5 \ (CH ) 2Sn(S0 3F) 2 + HC1 + CH4 4. (CH 3) 2SnCl 2 + exHS03F 2 5 *> (CHj) 2Sn(S0 3F) 2 + 2HC1 5. CH 3SnCl 3 + exHS03F — *- CH 3ClSn(S0 3F) 2 + 2HC1 6. (CH 3) 3SnS0 3F + exHSO F 2 5 -»» (CH 3) 2Sn(S0 3F) 2 + CH4 Ligand Redistribution Reactions 7. (CH 3) 2Sn(S0 3F) 2 + ex(CH3>4Sn 2 5 »• (CH ^SnSOjF 8. Sn(S0 3F) 4 + ex(CH 3) 4Sn 2 5 — (CH^SnSC^F 9. (CH 3) 2Sn(S0 3F) 2 + ex(CH 3) 2SnCl 2 cHCl^ (CH3>2ClSnS03F 10. CH 3ClSn(S0 3F) 2 + exCH 3SnCl 3 CHC1* CH 3Cl 2SnS0 3F - 81 -caused by sterlcal hindrance exerted by the bulky CF^ groups. Several general comments can be made about the systematic synthetic study of the solvolysis of methyltin compounds in HSO^F and ligand redistribution reactions of these compounds. (1) Only mono- or disubstitution i s observed, and (2) chlorine i s cleaved in preference to carbon. (3) The solvolysis reactions are simple clearcut reactions leading to single products. (4) Ligand redistribution reactions can be used to synthesize compounds which could not be prepared by acid solvolysis as well as to prepare some compounds which could also be made by solvolysis. The solvolysis reactions of methyltin compounds in HSO^F are similar i n some respects to the solvolysis reactions of the same 142 compounds in anhydrous hydrogen fluoride, however, these two systems differ in many respects. The main similarities are that in both cases reactions occurred with a l l (CH^^SnCl^^^ compounds (n > 1) to form a variety of compounds and i n a l l cases tin-chlorine bonds were cleaved before tin-carbon bonds were attacked. In the HF study i t was possible to control the course of the reaction by altering the reaction conditions such as time, temperature or mole ratios much more effectively than i t was in the HSO^F study. It was also possible to prepare CH^SnF^ but CH^SnCSO^F)^ was not obtained. Solvolysis reactions in HSO^F could also be used to make other dialkyltin bisf luorosulf ates using (C2H,.) ^ SnCl or ( C 2 H ^ ) 3 S n C l 2 or other alkyltin chlorides in place of (CH^^SnCl. Diethyltin bisfluoro-sulfate and dipropyltin bisfluorosulfate were prepared in this manner - 82 -but w i l l not be discussed here. Dialkyltin bisfluorosulfates and dialkyltin bistrifluoromethanesulfonates have been investigated in 143 conjunction with J.R. Dalziel in our laboratory. The solvolysis method can also be extended to the solvolysis of alkyltin compounds 143 in other strong monobasic acids such as HPO2F2. Mossbauer results for these derivatives as well as those of dialkyltin difluorides are shown in Figure 13 in order to demonstrate the similarities between these derivatives and the trends within the series. For a more detailed discussion see reference 143. (CH 3) 2Sn (S0 3F) 2 and CsSC^F were mixed in HSC^F in an attempt to 2-prepare Cs 2(CH 3) 2Sn ( S 0 3F) 4, in analogy to the (CH^SnCl^ salts, but no reaction occurred and only mixtures of CsS0 3F and (CH 3)2Sn(S0 3F)^ were isolated. In a similar attempt, this time to prepare (NO)2(CH 3) 2Sn(S0 3F) 2 Cl2, N0C1 was reacted with (CH 3> 2Sn(S0 3F) 2, but NOS03F and (CH 3 ) 2 SnCl2 were the only products identified. 2. Spectra The vibrational spectra of methyltin fluorosulfates indicate that the fluorosulfate groups are a l l bidentate and as shown in Table 25, the nine fundamental SO-^ F bands for a l l these compounds are in very similar positions. By making use of group theory predicitions on the presence or absence of vibrational modes in the Raman and infrared, the vibrational spectra can be used to help predict the geometries of the methyltin fluorosulfates. As shown in Table 26 the tin-carbon stretching frequencies i n the 620-520 cm ^ region show mutual exclusion for - 83 -Figure 13 C O M P A R I S O N O F M O S S B A U E R P A R A M E T E R S O F D i A L K Y L T I N ( i v ) D I F L U O R I D E S , B I S — D l -F L U O R O P H O S P H A T E S , B I S - F L U O R O S U L P H A -T E S A N D B I S — T R I F L U O R O M E T H Y L S U L P H O -N A T E S . Vibrational modes of the'S0_F group Cor.-ound VS03(A") vSO (A') vso30 (CH3)2Sn(S03F)2 1350 1180 1076 (CH3)3Sn(S03F) 1355 1207 1068 (CK3)2SnC£(S03F) 1343 1190 1072 CH3SnCi(S03F)2 1361 1165 1072 Cr^ Sntt (SO F) 1350 1250 1080 Tabic 25 ,n various tin and methyltin(IV) fluorosulfates vSF SS03F(A?) 6S03F(A") SSO^ A') S02 rock 827 620 590 554 417 820 630 596 ' 555 410 820 607 590 555 409 830 620 590 555 420 825 605 588 555 405 - 85 -Tabic 26 Tin-carbon and lin-chlorine stretching tr-odes in the me11:y 11 i.n(IV) chloro-sulfonnte compounds. Compound VSn-C V Sn-C V Sn-C x>S:\-CZ V Sn-C£ v Sn-C sym a sym (CH 3) 2Sn(S0 3F) 2 IR RA 576 531 (CH 3) 2Sn(S0 3CF 3) 2 IR RA 58^  530 (Ch*3)3SnS03F 2 ) IR RA 580 523 CH 3SnC£ 2S0 3F IR RA 576 384 3S7 366 368 (Ch* 3) 2SnCUS0 3F) IR RA 578 582 527 535 345 345 (CH 3)SnC£(S0 3F) 2 IR RA 578 585 385 390 (CH 3)SnCUS0 3CF 3) 2 IR 575 384 - 86 -(CH ) Sn(SO F) as expected for C-Sn-C, and absence of v Sn-C in the IR for (CH^^SnSO^F indicating a trigonal planar C^Sn moiety, but both asymmetric and symmetric Sn-C stretching modes are found in IR and Raman for (CH^)2ClSnS03F so this molecule has a nonlinear (CH^^Sn arrangement. Similar bent C^Sn grouping is expected for CH^C^SnSO^F which also has both asymmetric and symmetric Sn-Cl stretches active in both the IR and Raman spectra. The Mossbauer isomer sh i f t , quadrupole splittings and room temperature effects are best explained i f the fluorosulfates are a l l bridging, i.e. i f octahedral structures occur for both (CH^)2Sn(S0.jF)2 and CH^ClSnCSO^F)2 and trigonal bipyramidal structures with axial SO^F groups for (CH^SnSC^F, (CH^ClSnSC^F and CH^C^SnSC^F. These polymeric five- and six-coordinated structures are needed to explain the large quadrupole splittings observed for a l l these compounds and to explain the room temperature effects. Both the isomer shift and the quadrupole splitting decreases in the (CH 3) 2Sn (S0 3X) 2 series as fluorine i s replaced by CF 3 > CI, CH3, C„H C, or C-H.CH„. These results are explained by considering the withdrawal of p electron density around tin by the electronegative S03X groups. The smaller electronegativity of the S0 3C1 groups for example w i l l cause less withdrawal of electron density, more shielding of the tin s electrons and therefore a smaller isomer shift. The smaller electronegativity difference between SO^Cl and CH3 than between SC^F and CH3 results in a smaller A for the chlorosulfate than the fluoro-sulf ate. Similar electronegativity effects cause an analogous trend in the (CH^SnSC^X series. - 87 -3. Bonding Dimethyltin bisfluorosulfate w i l l be used as an example in the discussion of the proposed bonding scheme because interatomic distances are available and can supplement vibrational and Mossbauer data. 135 The structure of (CH^^SnCSO^F^j as shown previously in Figure 9 suggests two dimensional sheet-like polymers with only van der Waals interactions between the various sheets. It w i l l be convenient to discuss three distinct segments of the molecule; (1) the linear C-Sn-C moiety with very short Sn-C bonds using 119 the X-ray structure, the vibrational spectra and the Sn Mossbauer data as suitable probes into the bonding, (2) the distorted tetrahedral SO^F groups using the structural data as well as the vibrational spectra, in particular the removal of degeneracy of the E modes, and (3) the interaction between the two segments as reflected in 119 the observed Sn-0 distances, the Sn Mossbauer quadrupole splittings and the room temperature effect. A simplified model as shown in Figure 14, with the z axis along the C-Sn-C bonds and the four oxygen atoms in the equatorial plane around tin placed on the positive and negative branches of the x and y axes is helpful. The high isomer shift and short Sn-C bond distances indicate an extraordinarily high s-character in the tin-carbon bonds and therefore a high 5s electron density around tin. These results suggest a sp^ hybrid orbital i s utilized i n the bonding. A rather surprising departure from the previously discussed 33 34 acetate structures ' is noted for the fluorosulfate group. A l l - 88 -Figure 14 Structural Features of (CH 3) 2 Sn ( S 0 3 F ) 2 i i All bond distances are uncorrected - 89 -three S-0 bond distances are identical within the limits of error and the nonbonded sulfur-oxygen distance i s not shorter as found for 33 the C-0 bonds of tribenzyltin acetate. The S-0 distances are 65 identical to those found for KSO^F and apparently only the S-F bond is shortened. These results are also confirmed by the vibrational 47 spectra. The average S-0 stretching frequency using Lehman's Rule is identical for KS03F and (CH 3) 2Sn(S0 3F) 2 > but vS-F is raised in the ti n compound. There i s , however, a substantial sp l i t t i n g of the E symmetry S0 3 stretching mode by vL70 cm ^ in (CH 3) 2Sn(S0 3F) 2. Bonding in tetrahedral sulfur compounds is best described using 144 models involving i piT-dTr bonding proposed by Cruickshank. Using 144 145 a relationship between bond distances and IT bond order ' (see Figure 16) a IT bond order of between 0.5 and 0.6 per S-0 bond seems logical. In addition a survey of S-F bond distances indicates that ir back donation into 3d orbitals on sulfur also takes place here. As a consequence one can postulate that electronic charge removed from the (CH 3) 2Sn region and residual lone pairs on the nonbonding oxygen and on fluorine are used in strong back donation. The interaction between the electron withdrawing S03F groups and the positive (CH 3) 2Sn groups involves? P x and p^ orbitals on t i n as bonding orbitals. The electron density in the Sn-0 region results in four identical Sn-0 distances which are slightly longer than those observed previously. Previously reported Sn-0 distances are a l l ° 2 80 around 2.10 A, ' e.g. in SnCl^.2SeOCl2 the Sn-0 bond length i s ° 18 2.12 A. The resulting electron imbalance around tin due to the high electron density in the z direction and low density in the xy plane is - 90 -Figure 15 P77-—d77 Bonding in Fluorosulphates 0.0 0.2 0.4 0.6 0.8 1.0 77--bond orders. - 91 -reflected in the extremely large quadrupole splittings for (CH^^SnCSO^F^ and the other dimethyltin bissulfonates. The interaction between the (CH^^Sn moiety and the SO^F groups can be described using a three center four electron approach involving p^ and p^ orbitals on t i n , " ' ' 1 1 thus avoiding any necessity to use 5d orbital contributions to o bonding. This view may serve as an alternative to the often invoked rehybridiza-3,2 „ 40 tion of sp d according to Bent. The noted similarities between the interatomic distances of KSO^F and those of (CH^^SnCSO^F^ may give rise to an alternative view of the bonding. (CH^)2Sn(S03F)^ can be viewed as consisting of tightly 2+ packed linear (CH^^Sn cations and tetrahedral SO^F anions arranged in a most economical fashion. Although this situation appears to be closely approached in (CH^^SnCSO^F^ and i t s SO^CF^ analogue, the chemical evidence (Sn-0 distances, room temperature effect and splitting of E modes) points towards strong covalent interactions. A similar model based on a trigonal bipyramidal structure can be invoked for (CH^)^SnSO^F. This compound w i l l have the three methyl 2 groups in the xy plane bonded to tin using sp 'hybridization at tin, and the two oxygen atoms in the axial positions of the trigonal bipyramid bonded to tin using the p z orbital on tin and a three center four electron bond. As with (CH^)2Sn(S03F)^ the Mossbauer and vibrational results are consistent with this model. Both 6 and A are among the very highest for trimethyltin compounds, and large splittings of the E modes and energy increase for vS-F are observed in the vibrational spectra. As shown, these bonding models can be used to explain the X-ray structural data for (CH^^SnCSO^F^ and Mossbauer and vibrational Figure 17 Structural Features of Trimethyltin ( iv ) fluorosulf ate z I - 93 -results for (CH 3) 2Sn(S0 3F) 2 and (CH 3) 3SnS0 3F. The other experimental results also f i t in with these models. (1) The only partial dissociation of (CH 3) 2Sn(S0 3F) 2 and (CH 3) 3SnS0 3F into ions in HSC^F solution is consistent with a formula-tion involving considerable covalent interaction. (2) The wide range of isomer shifts and quadrupole splittings for oxyacid derivatives of dimethyltin(IV) also indicates covalency because, i f these compounds were ionic, changes in the anion should only cause minor changes in the electronic environment of the 2+ (CH 3) 2Sn cation. (3) Increasing the acceptor a b i l i t y of the tin atom by substituting CI, Br or SO^F for one or both of the methyl groups w i l l increase the covalency of the interaction between the tin and the fluorosulfate provided the structural framework is maintained. The resulting inorganic fluorosulfates Cl2Sn(SC>3F)2, Br 2Sn(S0 3F) 2 and SnCSC^F)^ are discussed in the next chapter. These changes have the following effects on the Mbssbauer and vibrational spectra; both the isomer shift and quadrupole splitting decreases and the magnitude of the room tempera-ture effect increases as more electronegative substituents are placed on tin and the covalency of the S03F group, reflected in the E mode splittings, increases. (4) Increasing the donor a b i l i t y of the acid group by replacing the fluorine in SC^F by less electronegative substituents on sulfur such as CF„, CI, CH„, C„H_, or C^H.CH„ also increases the amount of 3 3 2 5 6 4 3 covalent interaction between the sulfonate and t i n as indicated by decreases in 6 and A and increased room temperature effects for - 94 -(CH 3) 2Sn(S0 3CF 3) 2 > (CH^Sn'CSC^Cl) 2 > etc. Similar effects are also seen for the (CH ) 0Sn derivatives of weaker acids. - 95 -CHAPTER IV INORGANIC TIN FLUOROSULFATES A. Introduction Many metal fluorosulfates have been reported over the years, and the synthetic routes to these compounds have been discussed earlier. Although many binary fluorosulfates of formula MCSO^F)^ are known, almost a l l of these are mono, bis or trisfluorosulfates. The only I 62 previous examples of tetraJc .isf luorosulf ates are CCSO^F)^ which is unstable at room temperature and SiCSO^F)^ 2CH.jCN in which stabiliza-tion by two coordinated ligands can be assumed. Neither of these compounds is well characterized. The only inorganic t i n fluorosulfates to be reported are ClSn(SO„F)„ prepared by the reaction of SnO,F„ with anhydrous SnCl. 3 3 2 o 2 H below 100° and a product of the very approximate composition 59 Cl 2Sn(S0 3F) 2~Cl 3SnS0 3F formed in the slow interaction of SnCl^ with HS0„F at elevated temperatures. Since S„0..Fo doesn't decompose in 3 2 6 2 glass vessels above 100° except for the reversible dissociation into 146 •S03F radicals and a small amount of interaction with the glass walls, further heating of ClSn(S0„F)„ with S„0,.F,, should, providing the 3 3 2 o 2 resulting product i s sufficiently stable, enable us to replace the last chlorine by S03F. This was found to be the case by Poh Bo Long - 96 -147 of this lab who f i r s t prepared SnCSO^F)^ by this method in 1968. The synthesis of SnCSO^F)^ and Cl 2Sn(S0 3F) 2 and their Mossbauer and vibrational spectra were f i r s t investigated by Poh Bo Long and 148 subsequently by myself, however these compounds were not fu l l y characterized at that time, their relationship, to other fluorosulfates of t i n was not immediately recognized and their chemistry was not investigated. Therefore, although the preparative procedures and 147 148 Mossbauer and vibrational spectra have been reported ' they are included briefly here for completeness and for comparison to some related preparations which were attempted later. B. Preparations 1. Sn(S0 3F) 4 As reported previously"' 1 a mixture of SnCl^ and a 5-10 fold excess of S„0^F„ reacted instantaneously at room temperature to form a white 2 o 2 solid and chlorine gas. If the chlorine, excess S 0 F„ and any other 2. o 2. volatile byproducts were removed by d i s t i l l a t i o n to avoid formation of CISO-F or ClO-SO-F at higher temperatures, fresh S„0.Fo added and 3 2 J 2 o / the resulting solution heated gradually to 120° further reaction occurred and a white solid identified as tin tetrafciisf luorosulf ate was formed. The overall reaction scheme would be 120° SnCl. + 2So0,F. — Sn(S0 oF). + 2C1„ 4 2 6 2 3 4 2 However, since S„0-Fo was present in excess and reaction conditions were 2 o 2 149 similar to those used for the formation of CISO^F, i t s formation is - 97 -lik e l y and was indeed observed. Further oxidation may then have given rise to the formation of ClC^SOgF probably via the intermediate C10SQ3'F. SnCSO^F)^ is a nonvolatile hygroscopic white solid which melts with decomposition at 217° and is insoluble in the polar solvent HSO^F. The identity of the product was checked by elemental analysis of fluorine, t i n and sulfur. Fluorine and sulfur were determined by Bernhardt Microanalytical Laboratories and tin gravimetrically as SnC^ ."''"^  The analytical results are listed in Table 27. 2. Cl 2Sn(S0 3F) 2 When finely powdered Sn(S0 3F) 4 or any reaction intermediate of the composition CI Sn(SO„F), with n > 2 was mixed with an excess of n 3 4-n anhydrous SnCl^ at room temperature and the resulting suspension stirred for about 15 minutes; a reaction occurred with the uptake of one mole of SnCl^ per mole of Sn(SC>3F)4 to form a product of composition Cl 2Sn(S0 3F) 2. The Mossbauer-spectrum of this compound indicates that there i s only one environment for t i n ruling out possible formation of an SnCl^-Sn(S0 3F) 4 adduct. The reaction to form dichlorotin bisfluoro-sulf ate, similar to reactions discussed in Chapter III, proceeded, according to Sn(S0 3F) 4 + SnCl 4 »- 2Cl2Sn(SC>3F) 2 Sulfur and fluorine were again analyzed by Bernhardt. Chlorine and tin were determined by precipitating SnS2 from a hydrolyzed sample - 98 -Table 27. Analytical Results. Compound mp Elemental Analyses Element Calculated Found Sn(S0 3F) 4 217 F 14.75 14.90 Sn 23.05 23.0 S 24.90 24.65 Cl 2Sn(S0 3F) 2 207 CI 18.29 18.15 F 9.80 9.55 Sn 30.61 30.5 S 16.55 16.4 Br 2Sn(S0 3F) 2 142-5 Br 33.53 33.84 F 7.97 7.84 Sn 24.90 25.17 ClSn(S0 3F) 3 CI Sn 32.81 36.60 32.95 36.3 - 99 -of Cl^SnCSO^F)^ on addition of I^S solution. The SnS2 solution was filt e r e d and the f i l t r a t e was heated to boiling to remove excess I^S. Chloride could then be determined by potentiometric titration of the f i l t r a t e with AgNO^ and t i n gravimetrically as the oxide once the SnS was heated in the presence of air to convert i t to SnG^- This analytical procedure was necessary because simple precipitation of SnC^ from the hydrolyzed sample and titration of the f i l t r a t e for CI gave results for Sn and CI which were always about 10% low, possibly due to the formation of chlorotin anions in solution. 3. Cl 3SnS0 3F Trichlorotin fluorosulfate was produced when an excess of anhydrous SnCl. was reacted with S-0,,F at room temperature. 4- 2 o 2 SnCl. + 1/2S O F - Cl-SnSO.F + 1/2C1„ 4 2 6 2 J J z This reaction i s exothermic (on one occasion the reaction mixture exploded rather unexpectedly) and the Cl 3SnS0 3F produced is a very viscous colorless liquid which is immiscible with the SnCl^. This reaction was performed by d i s t i l l i n g an excess of SnCl^ ('v 6 g, 24 mmoles) onto a small amount of S^ O^ F,, (0.667 g, 3.36 mmoles) in a 2 o 2 small one part reactor with teflon stem stopcock, and allowing the mixture to warm up gradually to room temperature. The resulting mixture was l e f t to react for approximately one' hour at room temperature and the excess SnCl^ was then removed by vacuum d i s t i l l a t i o n to yield 2.194 g (6.75 mmoles);of liquid Cl 3SnS0 3F. This compound is - 100 -only moderately stable and decomposes on standing at room temperature for a period of several days. The decomposition products are SnCl^ and C^SnCSO^F^. Chlorine and t i n analyses were carried out using the SnS2 method described previously in the section on C^SnCSO^F)2» 4. Br 2Sn(S0 3F) 2 Dibromotin bisfluorosulfate was prepared in a ligand redistribution reaction similar to the one used to make Cl 2Sn(S0 3F) 2 but because of the higher melting point of SnBr^ the reaction had to be run either with the aid of a solvent or i n molten SnBr. at 40°. 4 SnBr4 + Sn(S0 3F) 4 - 2Br 2Sn(S0 3F) Typically, an excess of SnBr4 (2.70 g, 6.16 mmoles) and Sn(S0 3F) 4 (0.263 g, 0.511 mmoles) were ground together to a fine powder in the dry box and then placed in a glass tube with a B19 cone and a teflon stem stopcock adaptor top was placed on the tube. The reactor and contents were heated to 40° (mp of SnBr 4 is 31°) for 24 hours during which time a cream coloured solid was formed. The product was purified by vacuum subliming the excess SnBr4 onto a cold finger until no further sublimation was observed (y 3 days). Several times during the sublimation the unsublimed material;;was ground in the dry box in order to break up the large lumps of solid. After the sublimation was completed white solid Br 2Sn(S0 3F) 2 remained. A l l attempts to synthesize I 2Sn(S0 3F) 2 in a similar manner were unsuccessful, iodine was liberated in a l l cases. F 2Sn(S0 3F) 2 has been 142 obtained previously from the reaction of Cl 2SnF 2 and S 20gF 2 > - 101 -C. Structural Studies The physical properties of SnCSO^F)^, i t s high decomposition point, i t s insolubility in polar solvents and i t s lack of v o l a t i l i t y suggest that SnCSO^F)^ has a polymeric structure not unlike those of 133 2 A ^ H3^2 S n^ S°3 F^2 ° r ^ n F4 rather than a monomeric structure either 32 tetrahedral like SnCl^ or with chelating SO^F groups like SnCNO^)^. This proposal is confirmed by both the vibrational spectra which show vibrations due to both terminal and bridging covalent fluorosulfates and the Mossbauer spectra which show a negative isomer shift, small quadrupole spl i t t i n g and room temperature effect, as expected for this type of structure. The Mossbauer results are very similar to those of 95,151 SnF. 4 Cl 2Sn(S0 3F) 2 and Br 2Sn(S0 3F) 2 both have physical properties very similar to those of Sn(S0 3F) 4 and are easily prepared from Sn(S0 3F) 4 under mild reaction conditions. A structural proposal for these compounds which retains the similarities to Sn(S0 3F) 4 is one, similar to the structure determined for (CH 3) 2Sn(S0 3F) 2, in which linear X-Sn-X groups (X = CI, Br, CHg) are joined by bidentate bridging SO^F groups to form a structure similar to SnCSO^F)^ but with the terminal SO^F groups replaced by X. Both Mossbauer and vibrational spectra are consistent with this proposal. 1. MBssbauer Spectra The Mbssbauer results are shown in Table 28. Both 6 and A increase as the terminal fluorosulfate, (SO^F)^ is replaced by the progressively less electronegative CI, Br and CH» to form Cl 9Sn(S0 F) , Br„Sn(S0 F) - 102 -Table 28. Mossbauer Parameters of ) C 2Sn(S03p) 2 Compound 6(mm/sec) <^>(mn/sec) H Sn(S0jP)4 -0.27 1.34 0.42 Cl 2 S n ( S 0 5 P ) 2 0.34 2.29 0.45 Br 2Su(S0 5P) 2 0.58 2.42 0.53 (CH 3) 2Sn(S0 5F) 2 1.82 5.54 0.09 S n F 4 -0.26 1 5 1 l.dO 9 3 0.73 9 5 Table 29. Vibrational Spectra. Assignment Cl 2Sn(S0 3F) 2 Br 2Sn(S0 3F) 2 Cl^nSC^F IR Raman IR Raman IR Raman 1730 v? vS03(A") 1385 s 1389 s 1366 s 1350 w 1375 s vS03(A) 1130 vs 1125 w 1140 s,br 1118 s vS03(A) 1087 s ,br 1089 1061 vs m 1070 s,br 1085 m 1060 s 1073 1004 w w vS-F 864 s 870 s 842 s 865 m 845 s 815 m 6S03 628 m 632 s 613 s 624 609 s 618 m 650 w 6 S ° 3 5S6 s 587 ms 578 s 584 m 587 m <sso3 555 s 552 m 548 s 553 w 554 m 564 w Pso3 446 420 s w, sh 442 424 s m 406 m 423 w 420 m 423 m v Sn-Cl a 411 s 400 m 407 383 w si v Sn-Cl s 356 vs 369 v: pS03F 312 s 312 s 316 V7 v Sn-Br s 222 165 m w 132 w - 103 -and (CH^^SnCSO^F^ respectively. These trends are expected because as the electronegative (S0.jF)t is replaced by less electronegative groups the s electron density at tin would be expected to increase causing the isomer shift to be larger. The electronegativity difference between X and the bridging SO^F groups would also increase causing larger electric f i e l d gradients and hence larger quadrupole splittings. A room temperature effect i s observed for a l l these compounds as expected for polymeric structures. The Mossbauer spectrum of Cl^SnSO^F at liquid nitrogen temperature, which was more complicated than any of the other Mossbauer spectra were, consisted of an unsymmetrical three line pattern. This result suggests that there is more than one tin environment in Cl^SnSO^F in the solid state. The three absorption maxima occur at -0.36, +0.85 and +1.62 mm/sec from Sn02> and could be due to one of several possible combinations of singlet and quadrupole s p l i t absorptions. 2. Vibrational spectra SnCSO^F)^ has approximately twice as many peaks in the region of the infrared and Raman spectra in which SO^F modes are expected as Cl 2Sn(S0 3F) 2 or Br 2Sn(S0 3F) 2 does. For Cl 2Sn(S0 3F) 2 the nine modes for a Gg symmetry fluorosulfate are observed in the infrared at 1385 cm 1 (vSO), 1130 cm"1 (v S0„), 1087 cm"1 (v S0„), 864 cm"1 (vS-F), and a 2 s 2 bending and deformation modes at 628, 586, 555, 446, and 312 cm 1. Br 2Sn(S0 3F) 2 has similar absorptions at 1366, 1140, 1070, 842, and 613, 578, 548, and 406 cm The positions of these bands are consistent with those found for other compounds with bidentate fluorosulfates such - 104 -as (CH 3) 2Sn(S0 3F) 2. Sn(S0 3F) 4 shows peaks due to bidentate S0 3F at 1411, 1130, 1085, 850 cm"1, etc. as well as peaks at 1438 (v SO ), \2"b2 (v S0„), 920 (vSO), and 832 cm"1 (vS-F) due to monodentate S0„F. S 2 5 This i s the expected pattern for monodentate fluorosulfates. Gaseous fluorosulfates have v S0„ in the 1480-1500 cm 1 range and v S0 o a 2 s i at 1230-1250 cm"1, and BrS03F which also has monodentate S03F has v S0„ at 1438 cm"1, v S0„ at 1206 cm"1 and vSO at 884 cm"1, a 2 s 2 In the tin-halogen stretching region, Cl 2Sn(S0 3F) 2 has an IR active vibration at 411 cm 1 and a Raman active one at 356 cm 1. The mutual exclusion rule applies for these two peaks indicating linearity for the Cl-Sn-Cl group. Neither of these peaks is observed in the spectra of Sn(S0 3F) 4 and their positions are close to the positions of the Sn-Cl stretching modes in SnCl^ which are at 403 cm 1 and 368 cm 1 . The symmetric Sn-Br stretch in Br 2Sn(S0 3F) 2 is found at 222 cm 1 in the Raman spectrum in good agreement with v Sn-Br at 220 cm 1 in s SnBr 1 5 2 4 The infrared spectrum of Cl 3SnS0 3F shows that this compound also has bridging S03F groups and eight of the nine modes expected for C symmetry S0„F are observed in the IR spectrum. Unfortunately good S .5 Raman spectra could not be obtained and only a few of these bands could be seen in the Raman spectrum of Cl 3SnS0 3F. The three S0 3 stretching modes are observed at 1380, 1118 and 1060 cm \ in good agreement with those found for Cl 2Sn(S0 3F> 2, (CH 3) 2Sn(S0 3F) 2 and other bidentate fluorosulfates. The Sn-Cl stretching frequencies for Cl 3SnS0 3F show the absence of v g Sn-Cl in the IR spectrum expected for a trigonal planar SnCl 3 105 -group. In the IR spectrum where v Sn-Cl would be expected two strong c l overlapped peaks are observed, one at 420 cm 1 and the second at 400 cm No other absorptions are observed down to 300 cm 1 which i s the transparency limit of the AgBr windows. The Raman spectrum has a medium intensity peak at 423 c i \ a weak peak at 407 cm \ a very strong peak at 369 cm 1 with a shoulder at 383 cm 1 and a weak absorption at 316 cm"1. The peak at 420 1 cm"1 (IR) and 423 cm 1 (Raman) is assigned to the SO^ rocking mode, the 400 cm 1 (IR) and 407 cm 1 (Raman) peak to v Sn-Cl and the 369 cm 1 peak in the Raman only to v Sn-Cl. c t S The Raman absorption at 316 cm 1 would likely be the ninth SO^F mode. D. Reactions of Sn(S0^F)^ 1. Introduction During the course of this work on Sn(S0 3F) 4. several types of reactions were investigated. These included redistribution reactions with t i n tetrahalides as well as with some other metal tetrachlorides, and complexation reactions both adduct formation and addition reactions with other fluorosulfates and with chlorides and fluorides. The ligand redistribution reactions of Sn(S0 3F) 4 with SnCl^ and SnBr^ have already been discussed in this chapter and the complexation reactions with C102S03F, BrS03F and I(S0 3F) 3 are discussed i n Chapter VI. As mentioned in Chapter III, the reaction of Sn(S0 3F) 4 with an excess of (CH^^Sn was found to yield (CH.^SnSO^F according to 3(CH 3) 4Sn + Sn(S0 3F) 4 4(CH 3) 3SnS0 3F - 106 -However, considering the ease with which (CH^)^SnSO^F is formed in the solvolysis reaction and the length of time required to synthesize SnCSO^F)^, this route cannot be considered an attractive alternative. 2. Ligand redistribution reactions Since Sn(S03F>4 reacted so readily with SnCl^ to yield Cl 2Sn(S0 3F) in a ligand redistribution reaction i t was hoped that Sn(S0 3F) 4 might react with other group IV tetrahalides to form halofluorosulfato compounds or possibly even polymers containing both t i n and the other group IV element in alternation joined together i n one molecule by fluorosulfate bridges. (a) GeCl 4 andJ3iCl 4 In the case of germanium an excess of GeCl 4 was reacted with approximately 0.3 g of Sn(S0 3F) 4 i n a two part reactor. The mixture was stirred for about one half hour at room temperature without any evidence of reaction except that the vapor pressure in the reactor increased from 50 mm Hg for GeCl 4 alone to 61 mm Hg. These gases were shown by their infrared spectrum to be GeCl 4 and S i F 4 > When the volatiles were removed the weight of the flask plus Sn(S0 3F) 4 was virtua l l y unchanged, i.e. no reaction had occurred. The reaction of Sn(S0 3F) 4 with S i C l 4 attempted under similar experimental conditions was also unsuccessful. - 107 -(b) TiCl 4 When a large excess of TiCl^ was added to 0.869 g (1.69 mmoles) of Sn(S0,jF)4 i n a two part reactor and stirred for approximately one half hour a largely increased volume of a yellow solid was formed. When the excess Ti C l ^ was removed a weight increase to 1.347 g was noted. This weight corresponds to an uptake of four moles of TiC l ^ per mole of Sn(S0 3F) 4. Some of the TiC l ^ could be removed by extended pumping while slowly raising the temperature to 60° but no definite product of lower TiCl^ composition comparable to those found in the SnCl 4~Sn(S0 3F) system could be obtained. The yellow solid was unusual in that i t was completely devoid of tin. Chloride analysis indicated that the stoichiometry was Cl^gTi^SO^F),^ and microanalytical results for T i , S and F confirmed this result. These analyses are calculated, T i , 20.60; S, 9.21; CI, 50.90; and F, 5.46; and found, Ti, 20.86; S, 9.48; CI, 50.71; and F, 5.51. Although we were f i r s t inclined to regard the product as a rather unusual anomaly, subsequent routes to products of identical properties 153 and composition were found. These routes are T i C l 4 (large excess) + C l ^ T i ^ S O ^ + C l 2 TiCl^ (large excess) + HS03F C l 1 ( ) T i 3 ( S 0 3 F ) 2 + 2HC1 and C l 2 T i ( S 0 3 F ) 2 + 2TiCl 4 C l ^ T i ^ S O ^ - 108 -Another unusual feature of this compound is that there are no peaks in the IR or Raman spectra, Table 30, in the region in which the S-F stretching frequency is expected. However, there i s a strong absorption at ^ 660 cm 1 which can only be due to the S-F stretch of a fluorine atom which is involved in further bidentate bonding. In order to achieve six-coordination for a l l three titanium atoms in this molecule, both SO^F groups would have to be tetradentate, i.e. coordinating through a l l three oxygens and the fluorine, or some of the chlorines would have to be bidentate. If the tetradenate SO^F's did exist i t would explain the unusual position of the S-F stretch as well as the lack of similarity between the positions of the SO^  stretching modes in Cl^jTigCSO F ) 2 and those of other fluorosulfates. Supporting evidence for the proposed structure i s found in HSO^CF^-TiCl^ and HSO^-CH^-TiCl^ systems where tetradentate sulfonate groups cannot occur and under identical conditions (large excess of TiCl^) the compounds formed that are highest in chlorine content were Cl^TiSO^CF^ and C l 3 T i 3 S 0 3 C H 3 . 1 5 3 Addition of donor atoms (adduct or complex formation) should remove the necessity for fluorine bridging and allow the fluorine to revert to i t s more normal coordination and vS-F should reappear in the 800-900 cm"1 region of the infrared spectrum. Both N0C1 and P0C13 were added to C l ^ H ^ S O . ^ ^ in attempts to check this possibility but the results were inconclusive due to experimental d i f f i c u l t i e s encountered in isolating the products. In the NOCl reaction mixtures containing (N0)„TiGl, were formed. .2 o In order to further investigate this compound attempts were made to prepare titanium fluorosulfates by the more straightforward route of T 109, T TABLE 3 0 Vibrational Spectra of TiC£ (SO„F) 1 JZ and TiC£ 2CSO.F) 2 IRfcm"1] CS0 3F) 2 Raman[cm-1] KS03F Raman[cm-*] Assignment For KS03F T i C i 2 ( S 0 3 F ) 2 IR 2 3 [cm - 1] 1248 vs,br 1205 vw 1188 vw 1260 ms | 1248 m / 1285 vS0 3(E) 1380 m,sh) 1342 s | 1195 m,sh • 1082 m,sh \ 1070 s f 1082 s 1079 vSOjCAp 1080 ms,sh 1020 s,b 870 vw 650 s ) 643 s,sh / 676 ms 755 vSFfA^ 846 ms 732 vs 592 m 590 ms 586 6S0 3(E) 630 s 579 m 570 ms 570 616 m,sh 495 ms 464 s 430 vs 415 vs 461 m,sh 444 vs • 578 m,sh) 555 ms y 445 s (448)* 390 ms 390 407 pS03FCE) 418 sh 390 m* 363 ms* a) good correspondence between Raman and IR frequencies was noted. - 110 -adding S„O^F„ to TiCl.. TiCl. and So0,F„ reacted in one part glass 2 6 2 4 4 2 6 2 reactors at room temperature or above to produce glassy products of various compositions presumably containing fluoro-fluorosulfates or oxyfluorosulfates of titanium. However, i f an excess of S-O^F^ was added to 0.406 g (2.14 mmoles) 2 D 2 of T i C l ^ in a one part glass vessel and stirred at -20° for several hours, 0.720 g (2.27 mmoles) of a yellow solid identified by i t s IR 154 spectrum as C^TitSO^F^ was produced. -20° T i C l 4 + S 20 6F 2 — * C l 2 T i ( S 0 3 F ) 2 + C l 2 If the excess So0,.Fo was removed the solid product was stable at room 2 o 2 temperature. This compound can also be produced by the reaction of 154 T i C l 4 with HS03F. J H When 9.45 g (29.9 mmoles) of C l 2 T i ( S 0 3 F ) 2 was reacted with a 10 fold excess of T i C l 4 for 20 hours, C l 1 Q T i 3 ( S 0 3 F ) 2 was produced in a ligand redistribution reaction. C l 2 T i ( S 0 3 F ) 2 + 2TiCl 4 ~ C l 1 ( ) T i 3 ( S 0 3 F ) 2 3. Complexation reactions Several attempts were made to prepare complexes in which the Sn(S0 3F)^ acted as an acceptor molecule. One of these was the preparation of (CIO.)oSn(S0„F)£ discussed in Chapter VI. In this 2 2 J o case Sn(S0 3F) 4 acted as an acceptor of S03F groups. Attempts were also made to add both chloride using N0C1 and fluoride using N02F 2- 2-or C1F3 to form Sn(S0 3F) 4Cl 2 and Sn(S0 3F) 4F 2 ions respectively. - I l l -Although none of these reactions succeeded, some interesting side products were formed. Addition of neutral donors such as POCl^, CH^ CN or (CF^^CO to SnCSO^F)^ to form donors-acceptor adducts were also unsuccessful. (a) N0C1 Nitrosyl chloride, N0C1, reacts with either SnCSC^F)^ or Cl 2Sn(S0 3F) to produce NOSO_F and (N0)oSnCl,. For the reaction of N0C1 with j z o Cl 2Sn(S0 3F) 2, 0.384 g (0.993 mmoles) of Cl 2Sn(S0 3F) 2 were added into a glass reactor in the dry box and the side arm for addition of solid was sealed off. Excess N0C1 was added by vacuum d i s t i l l a t i o n and the reactor was warmed to room temperature. An immediate vigorous reaction which produced a yellowish solid occurred and when the excess N0C1 was removed 0.618 g of the solid remained. This weight corresponds to an uptake of 3.6 moles of N0C1 per mole of Cl 2Sn(S0 3F) 2 > No volatiles other than N0C1 were observed in the gas phase and the solid product was shown by i t s infrared spectrum to consist of N0S0„F and (N0)oSnCl,. j z o The reaction i s presumably 4N0C1 + Cl oSn(S0.F) o 2N0S0oF + (NO)-SnCl, Z 3 z J z o The reaction of N0C1 with Sn(S0 3F) 4 was carried out in the same way. In this case 0.408 g (0.792 mmoles) of Sn(S0 3F) 4 reacted with excess N0C1 to produce 0.633 g of solid (uptake of 4.3 moles of N0C1 per mole of Sn(S0 3F) 4). Again only N0C1 was observed in the gas phase and the solid was shown to be a mixture of N0S0,F and (N0)„SnCl,. These two - 112 -components could be separated by subliming the (NO^SnClg. 6N0C1 + Sn(SO„F). »• 4N0S0oF + (NO).SnCl, 3 4 3 2 o It appears that Sn(SO.jF)4 undergoes rapid exchange with NOC1 rather than the expected complex formation. (b) N02F and C1F3 Neither n i t r y l fluoride, N02F, or chlorine trifluoride reacted 2-with Sn(S0 3F) 4 to produce a complex of the Sn(S0 3F) 4F 2 anion. The reaction with N02F produced N02S03F and that with C1F3 surprisingly yielded (ClO.).SnF-, even though the reaction was carried out in a 2 2 o metal reactor. It must be borne in mind however that similar experiences 154 when handling chlorine fluorides are not uncommon. Both of these reactions were attempted i n the same manner. Approximately 1 g (2 mmoles) of Sn(S0 3F) 4 was placed i n a metal reactor in the dry box, the reactor was then evacuated and an approximately ten fold excess of C1F3 or N02F was added by d i s t i l l a t i o n . After reaction had occurred at room temperature the excess C1F3 or N02F was removed and the identity of the solid products was checked by IR. (c) Adduct formation SnF^ interacted with neutral donor molecules such as (CH3)2SO to form hexa-coordinated donor-acceptor complexes S n F ^ ' 2 D . S n ( S 0 3 F ) 4 might be expected to form similar complexes so Sn(S0 3F) 4 was mixed with POCT3, CH3CN and (CF 3) 2CO in attempts to form such complexes. When excess P0C13 was d i s t i l l e d onto about one gram of Sn(S0 3F) 4 a - 113 -greenish-yellow gas (C^?) was produced, indicating the failure of any intended addition. Likewise, addition of CH^ CN resulted in the formation of a viscous yellow o i l after the excess CH^ CN was removed in vacuo from a clear colourless solution of SnCSO^F)^ in CH^CN. (C7^)^C0, which would be expected to be a rather poor donor, did not interact with Sn(SO„F). when these two components were mixed together. - 114 -CHAPTER V CORRELATION OF MOSSBAUER PARAMETERS FOR TIN FLUOROSULFATES A. Correlations for Tin Fluorosulfates In the preceding chapters the synthesis and structural characterization of a number of tin(IV) and organotin(IV) bisfluoro-sulfates of the general type XYSn(S0 3F) 2 where X and Y are CH3, CI, Br, F, or S03F has been described. Vibrational spectroscopy has been used to identify the bidentate S03F group and the linearity of the X-Sn-X groups when possible and Mossbauer has been used to establish the presence of hexa-coordination and.the linkage into polymers. Combining information from both techniques has allowed a f a i r l y accurate description of the trans-octahedral geometry around t i n , with confirmation available from the subsequently reported X-ray 135 diffraction study on (CH 3) 2Sn(S0 3F) 2. Based largely on these data, a basic description of the bonding in these and related compounds has emerged. The general contention that a l l XYSn(S0 3F) 2 compounds have 24 25 identical structures, the same as e.g. SnF^ and (CH 3) 2SnF 2, which when simplified suggests the following environment for t i n : - 115 -should result in two predictable trends in the principle Mossbauer parameters 6 and A. (1) Assuming a constant background is being provided by the four oxygen (two from each SO^F group) the observed isomer shi f t , should be a function of the sum of the electronegativities of ligands X and Y. (2) The quadrupole s p l i t t i n g , A, w i l l be a function of the inductive constant a , the Taft constant, of the ligands X and Y. The principal conditions for these two expected trends are; (1) changes in the electronic environment around tin are governed solely by X and Y, i.e. redistribution of electron density into the SO^F groups does not interfere, and (2) a l l the structures are reasonably regular, i.e. contributions from the asymmetry parameter, n, to A are neglected. Precedents for similar correlations abound in the literature. Some examples of linear relationships between <5 and electronegativity are 2- 2- 22 90 156 hexahalostannate anions SnX, and SnX.Y„ ' ' and the tin o 4 2 156 * tetrahalides. Examples of linear relationships between A and a have been found for R^SnBr' s, 1"^ triphenyltin h a l o a c e t a t e s t r i m e t h y l t i n - 116 -9ft 44 acetates, and R3SnX' (R = CH , C H 5 and X = F, CI, Br, I). A 142 similar trend is found for the related XYSnF^ series, a different slope reflecting the different electron dispersing a b i l i t i e s of the F atom and the SO^F group both in bridging positions. As shown in Figures 18 and 19 such trends are found for both 6" and A for the fluorosulfates and the basic accuracy of the above stated assumptions is evident. One would suspect that these correlations are not restricted to the bisfluorosulfates but are also applicable to (1) the isostructural 142 oxyacid derivatives or as shown in part to the fluorides, to (2) monofluorosulfates of the type XYZSnS03F where X, Y and Z are CH3, CI, F SO-jF, etc. and of course to (3) other fluorides and oxyacid derivatives. Two problems are faced here. First, for no other oxyacid derivatives are sufficiently large numbers of examples known or even obtainable - see the failure to extend the use of ligand redistribution reactions to S0 3CF 3 derivatives - to allow similar conclusions, and secondly, monofluorosulfates where one of the ligands i s chlorine need not be s t r i c t l y penta-coordinate because as the crystal structure 26 of (Cl^^SnC^ indicates, chlorine apparently can act as a weakly bidentate ligand, capable of distorting any idealized geometries assumed for such a correlation. Even a superficial inspection of data obtained on monofluorosulfates indicates a failure to rationalize both <5 and A. A similar conclusion had been reached previously for some monofluorides. Finally the fluorosulfates are exceptional when compared to some reported structures of carboxylates 3 3' 3^ or thiocarbamates 3^' 3 7 (see Figure \ B I 1 1 1 1 1 3 . 0 4 . 0 5 . 0 6 . 0 7 . 0 8 . 0 2 of Paul ing's E lec t ronegat iv i t ies F i g u r e 13 <X ( C H 3 ) 2 S n ( S 0 3 F ) 2 5 . 0 -1 . O Correlation between laft 's inductive constants and the quadrupole Splitting in the series X 2 S n ( S Q 3 F ) 2 and X 2 SnF 2 ^ C H 3 C l S n ( S 0 3 F ) 2 I—1 I-1 00 I B r , S n ( S O . F ) . C l 2 S n ( S 0 3 F ) 2 b > S n ( S 0 3 F ) 2 ( S 0 3 F ) 2 S n ( S 0 3 F ) 2 O 1 . 0 2 . 0 3 . 0 4 . 0 5 . 0 6 . 0 7 . 0 8 . 0 CT - 119 -Figure 2). The carboxylates and thiocarbamates are anisobidentate, i.e., there are differences in the Sn-0 or Sn-S bond distances, whereas 135 basing our claim on (CH^) 2Sn(S0 3F) 2 and the successful correlations for 6 and A, we can regard the SO^F group in these molecules as iso-bidenate. It is obvious that anisobidentate groups where the tin-oxygen (or tin-sulfur) bond distances depend on the electronic and steric properties of other ligands bonded to tin, do not provide a similar background for similar correlations. B. Other Possible Correlations Two additional correlations may be considered for the compounds discussed in this study. For the dimethyltin bissulfonates of general formula (CH 3) 2Sn(S0 3A) 2 with A = F, CF 3 > CI, CH 3 > C ^ , or pj-CgH^ CH.^ , the isomer shift, 6, should be a function of the electronegativity of * A and the quadrupole splitti n g , A, a function of the Taft constant, a , of A. Similar correlations had been detected for trimethyltin halo-98 159 acetates, ' (CH3) .jSnCO^ CX.^  with X = H, F, Br, or I. However, for the sulfonates major d i f f i c u l t i e s are encountered. Group electro-• * ^ , • ,160-162 negativities for organic functional groups are often very controversial even when based on thermochemical data, e.g. for the trifluoromethyl group the values reported differ widely from 2.86 ^ ® over 3.10 to 162 3.29, and are sometimes unavailable as e.g. for p-C-H.CH„. ^ D 4 3 Taft constants are available for most systems, however, as a brief * 94 glance at the a values of 3.08 and 2.58 for F and CF 3 respectively, shows the expected differences in the A values is not found experimentally for either (CH 3) 2Sn(S0 3A) 2 or (CH^SnSO^ where, as shown in Tables 16 and - 120 -20, 6 and A values for fluorosulfates and trifluoromethanesulfonates are identical within error limits. These experimental findings also preclude any meaningful isomer shift-electronegativity correlations. It appears safe to say that both 6 and A are very sensitive to changes in electronegativity or inductive constants of the immediate substituents X and Y, however, long range effects are depicted in only rough and approximate trends in <5 and A. The a b i l i t y of some of the A substituents, F and CI, to participate in pir-drr backv donation and the presence of a non-coordinated oxygen, again capable of IT interaction with sulfur, may account for the discrepancy between sulfonates and haloacetates. No consistent data are available, to relate Mossbauer data of trimethyltin or dimethyltin sulfonates to the dissociation constants of the corresponding sulfonic acids. The determination w i l l require work in a non-aqueous solvent, e.g. acetic acid, and from limited data 16 3 available where values of 4.7 and 6.1 are reported for pKa for HSO^CF^ and HSO^F respectively i t must appear doubtful whether a.meaning-f u l correlation can be made here. C. Point Charge Model Correlation of the kind mentioned above involving quadrupole splitting values and the Taft inductive constants of some of the substituents on tin (X and Y) for XYSnCSO^F)^ compounds suggests that these compounds may be suitable model compounds for point charge discussions and provide an application of the concept of partial quadrupole splittings of individual substituents. Extension to other sulfonates of the same stoichiometry and even to other geometries such as X»SnS0oF should be - 121 -feasible provided the coordination polyhedra around t i n are sufficiently regular. A large number of geometries have been treated 164 in this way f a c i l i t a t i n g the prediction of the electric f i e l d gradient tensor with V the component of prime interest, and the zz partial contributions of the individual ligands. Different signs for V should be observed for the two principal z z models, the trigonal bipyramidal XYZSnSO^A and the octahedral XYSnCSO^A)^, discussed here and in Chapter III. By comparison to previous work, for the penta-coordinated species V should be positive"^2»165,166 a ^ ^ zz should be negative for the octahedral compounds. Some experimental and computed A's for a number of supposedly trigonal bipyramidal compounds are listed in Table 31. The computed values are negative as expected for trigonal bipyramidal structures but for a l l methylchlorotin compounds a satisfactory f i t is achieved only when strong contributions from the asymmetry parameter are assumed. The possibility 166 of a sign reversal cannot be ruled out completely and i t is necessary to obtain this information experimentally. For the cases listed above, application of an external magnetic f i e l d at 4°K w i l l 119 remove the degeneracy of the 3/2 excited state of the Sn nucleus. 92 166—168 A number of magnetic perturbation studies have been published ' and work on a series of compounds, whose synthesis has been discussed in this thesis, has commenced, but the discussion of these measurements w i l l exceed the scope of this thesis. - 1 2 2 -T a b l e 31 P o i n t - c h a r g e P r e d i c t i o n s f o r s o a e M e t h y l t i n ( I V ) C o m p o u n d s C o m p o u n d o . ( o b s ) /L\{pre&) < ^ ( p r e d ) m m / s e c m m / s e c m m / s e c ( C H 3 ) 5 S n S 0 ^ C i , 3 4 . 5 7 - 4 . 5 6 0 ( C H 5 ) 3 & n S 0 5 C H 5 4 . 2 1 - 4 . 1 2 0 ( C H - 5 ) 2 C l S n S 0 - j P 4 . 6 9 - 4 . 0 5 0 . 6 4 C H - j C l g S n S O - j P 3 . 2 5 - 3 - 3 0 0 . 8 1 ( C H ^ S n 2 S 0 4 1 4 0 4 . 0 6 - 4 . 0 7 0 ( C H ^ S n F 1 4 2 3 « 9 0 - 4 . 1 0 0 ( C H 3 ) 2 C l S n P 1 A 2 3 . 8 0 - 3 . 5 8 0 . 7 4 C H 3 C l 2 S n P 1 4 2 2 . 6 9 - 2 . 8 4 0 . 9 8 - 123 -CHAPTER VI THE HEXAKISFLUOROSULFATOSTANNATE(IV) ION A. Introduction Although hexahalogen metallates, (MX^)n , of group I l i a , IVa, and Va metals are well known, no examples of analogous fluorosulfates have been reported. For fluorosulfates, the largest number of SO^F groups bonded to a central atom appears to be four observed in the anions ~ — 169 — 163 BrCSO^F)^ and iCSO^F)^ , in the neutral compounds such as 6 2 C(SO.jF)4 and SnCSO^F)^ and in the mixed fluorofluorosulfato anion SbF^CSO^F)^ which was identified in solution only. Judging from the structure proposed for Sn(S0 3F) 4, i t seems likely that tin would be a suitable central atom for a M(S0oF) " anion. 3 o 2-Two synthetic routes to compounds with the Sn(SO„F) anion were 3 o attempted. These are the interaction of SnCSO^F)^ as a fluorosulfate ion acceptor with a suitable fluorosulfate and the complete substitution 2-of CI in the SnCl^ ion by S0oF using S„0,F„ as the fluorosulfonating o J 2 o 2 agent. The f i r s t method was used to make (C10„)oSn(S0„F) 2 2. J O [Br(S0 3F) 2] 2Sn(S0 3F) 6 and [I(S0 3F) 2] 2Sn(S0 3F> 6 from Sn(S0 3F) 4 and the second to make K oSn(S0„F)., CsoSn(S0„F), and (NO)oSn(S0oF), from 2 5 o 2 3 D 2 j o K 0SnCl., Cs 0SnCl, and (N0) oSnCl, respectively. The structures of 2 o 2 o 2 b these compounds were investigated using vibrational and Mossbauer spectroscopies. Solution studies in HSO-jF were of special interest - 124 -because the parent acid HoSn(S0„F),. can be expected to be a strong acid, 2 j o 101 perhaps comparable to HSbF 2(S0 3F) 4 one of the main components of , - 170 super acid. B. Preparations 1. (C10 2) 2Sn(S0 3F) 6 The work described in this chapter was started after low melting solids were obtained accidentally in the attempted preparation of SnCSO^F)^ at temperatures above 100° in the presence of considerable amounts of chlorine or secondary products such as CISO^F and C102S03F. The observed weight increase was in excess of the expected weight for the formation of SnCSO^)^. Pure dichloryl hexakisfluorosulfatostannate(IV) was prepared via a complexation reaction of SnCSO^F)^ with CK^SO^F. 2C102S03F + Sn(S0 3F) 4 > ( C l O ^ S n C S C ^ 2 6 2 This reaction was carried out in a 100 ml erlenmeyer flask equipped with a teflon stem stopcock and containing a teflon coated magnetic stirring bar by adding 0.849 g (5.02 mmoles) of C102S03F onto 0.520 g (1.01 mmoles) of Sn(S0 3F) 4 in the dry box at room temperature. To ensure good mixing about 10 ml of S.0,Fo were then added by vacuum d i s t i l l a t i o n . 2 o 2 The reaction mixture was heated to about 70° and stirred for several hours. 0.793 g (0.96 mmoles) of a bright yellow solid, later identified as (C10.)_Sn(S0„F). were isolated after the S_0,F„ and the excess 2 2 3 o 2 o 2 C10„S0„F were removed by vacuum d i s t i l l a t i o n . (CIO.)_Sn(S0„F), i s a 2 J 2 2 j o bright yellow solid thermally stable up to 130° where i t decomposes to - 125 -a red liquid. It hydrolyses very vigorously to a clear chloride free solution. 2. [Br(S0 3F) 2] 2Sn(S0 3F) 6 [Br(S0.F)„]oSn(S0„F)/, was prepared in two stages f i r s t by the J 2 2 3 o reaction of BrSC^F with Sn(S0 3F) 4 to form a red-brown solid containing 2-Sn(SO.F), possibly Br oSn(S0.F), followed by reaction of this compound J O 2 j o with excess S„0,F„ to produce pale yellow precipitate of 2 o 2 [Br(S0 3F) 2] 2Sn(S0 3F) 6. 2BrS03F + Sn(S0 3F) 4 Sn(S0 3F) 4'2BrS0 3F Sn(S0 3F) 4-2BrS0 3F + 2S 20 6F 2 — - * [Br (SC^F) 2] 2Sn(S0 3F) In a typical reaction, a small excess of BrS03F (^1 ml) was added by d i s t i l l a t i o n onto 0.354 g (0.688 mmoles) of Sn(S0 3F) 4 in a one part round bottomed flask reactor with teflon stem stopcock. These components were reacted at room temperature for three days to produce 0.601 g (0.689 mmoles) of Sn(S0 3F) 4•2BrS0 3F. After the excess BrS03F was removed by d i s t i l l a t i o n approximately five grams (25 mmoles) of S2°6 F2 W a S a d d e d o n t o t h e Sn(S0 3F) 4-2BrS0 3F by d i s t i l l a t i o n . A slow reaction occurred to produce a viscous yellow liquid soluble in S„0,F„ 2 o 2 from which a pale yellow solid slowly separated out. The [Br(S0„F)o]„Sn(S0 F). was isolated on a f r i t when the S_0,F„ solution 3 2 2 3 6 2 o 2 was fi l t e r e d at -40°. The product could not be isolated by removing the S 20 6F 2 by d i s t i l l a t i o n because the [Br(S0 3F) 2] 2Sn(S0 3F) 6 in - 126 -decomposed on pumping. [Br(SO„F)_].Sn(SO.F), melts at 48-50° to a 3 2 2 J O yellow liquid. 3. [I(S0 3F) 2] 2Sn(S0 3F) 6 Stoichiometric amounts of I(S0 3F) 3 and Sn(S0 3F) 4 react at the melting point of I(S0 3F) 3 to produce a viscous yellow liquid which slowly crystallized to pale yellow [I(S0„F)-]„Sn(S0oF),. 5 2 2 J O 2I(S0 3F) 3 + Sn(S0 3F) 4 3 5 ° ^ [I(S0 3F) 2] 2Sn(S0 3F> 6 In this preparation, 1.522 g (1.01 mmoles of Sn(S0 3F) 4 were added in the dry box to 1.72 g (1.03 mmoles) of I(S0 3F) 3 in a round bottom flask with a teflon stopcock adaptor top. These reactants were heated to 35° where a<'. reaction occurred to form a viscous yellow liquid. About 10 grams of S^O^F. were added to f a c i l i t a t e the reaction by providing 2 o 2 a diluting agent and these compounds were stirred at 35° for several hours. The S 0 1 . was then removed and the viscous yellow liquid 2 o 2 remained. This liquid slowly s o l i d i f i e d over a period of about five days at room temperature to yield 2.35 g (1.06 mmoles) of pale yellow [I(S0 3F) 2] 2Sn(S0 3F) 6. 4. K 2Sn(S0 3F) 6, Cs 2Sn(S0 3F) 6 and (NO) 2Sn(S0 3F) 6 The potassium,cesium and nitrosonium salts were prepared in reactors similar to the one used in the (C10o)-Sn(SO.F), preparation 2 2 3 o by a ligand substitution reaction. - 127 -M^SnClg + 3S 20 6F 2 f^ " - M 2 ISn(S0 3F) 6 + 3C12 M1 = K, Cs, or NO In the case of CsoSn(S0„F) , for example, a five-fold excess of S o0 cF o 2 j o 2 o 2 ( 10 g) was d i s t i l l e d in vacuo onto 0.731 g (1.22 mmoles) of carefully dried Cs 2SnClg. The mixture was allowed to warm to room temperature where a slow reaction with evolution of chlorine occurred. The reactor was then heated to 50° for about three hours while stirring the mixture magnetically. When a l l gas evolution had ceased, the excess S 0 F 2 O 2 was removed by vacuum d i s t i l l a t i o n and 1.211 g (1.24 mmoles) of white solid Cs_Sn(S0„F), were obtained as a nonvolatile residue. Both 2 2 6 K 0Sn(S0 oF) £ and (NO) Sn(S0 F) were obtained in the same way, the reaction 2 j 6 2 j 6 temperature was 60° for the former and 50° for the latter. A l l three of these compounds are hygroscopic white solids. 2-A l l of the Sn(S0„F)£ compounds, in particular the K and Cs j o salts, show reasonably high thermal st a b i l i t y . This suggests that the preparation of other salts, e.g. RboSn(S0„F), or Na oSn(S0 oF). may be 2 j 6 2 J 6 feasible, but no attempts at their synthesis were made. A l l of these compounds are soluble in HSO^F quite in contrast to Sn(S0 3F) 4. This contrast serves as preliminary evidence that the products are.not merely mixtures of Sn(S0,jF)4 and MSO^F. The melting points and analytical results for the six compounds are listed in Table 32. Another possible route to compounds of this type, the heterocation substitution of C10 2 + by N0+ or N0 2 + was attempted in the reaction of (C10„)oSn(S0„F), with an excess of NO or N0„ at room temperature. The 2 2 j o 2 - 128 -Table 32. Analytical Results. lis ( ^ e») Compound mp Analysi 1 Calculated Found K_Sn(S0oF)/: 235-8 Sn 15.00 14.84 2 3 D S 24.31 24.58 F 14.41 14.20 Cs oSn(S0 oF), 249-53 Sn 12.13 2 J o 11.80 S 19.65 19.60 F 11.65 11.54 (N0) oSn(S0 oF), 94-7 Sn 15.35 15.57 2 3 D S 24.88 25.00 F 14.75 15.02 (C10-)oSn(S0_F), 130 Sn 14.00 14.25 2. 2. 3 D S 22.69 22.84 F 13.44 13.65 CI 8.36 8.11 - 129 -expected formation of CINO^,17''"'172 identified by i t s IR spectrum," 7 3 took place, but the solid residue always contained small amounts of indicating incomplete reaction as well as some ionic SO^F due to the formation of NOSC^F or NC^SC^F. C. Mossbauer Spectra 119 The Sn Mossbauer data recorded for the hexakisfluorosulfato-stannates are listed in Table 33 together with literature values for a number of related compounds. A l l the hexakisfluorosulfato-stannates except Sn(S0 3F) 4«2BrS0 3F gave single line spectra with very similar isomer shift values, however, the wider line widths (r) for the NO, CK^, BrCSOgF^ and ICSO^F^ compounds may indicate some unresolved splittings in these spectra and hence distorted octahedral structures for these compounds. SnCSO^F^^BrSO^F is the only compound whose MBssbauer spectrum differs significantly from the rest. It has both a higher isomer shift and a small but resolved quadrupole splitting. These two differences suggest that the two SO^F groups are not totally abstracted from BrSO^F by SnCSO^F)^ and the resulting complex may resemble an adduct with the previously uncoordinated oxygens of BrSO^F + 174 coordinating to SnCSO^F)^. A Br cation would be unlikely and no evidence for i t has been found. 137 In general, the similarity of the fluoro and fluorosulfato compounds .whose Mossbauer' data are recorded here is apparent, particularly 95 the transition from well resolved quadrupole splittings for SnF^ and SnCSO^F)^ to single line spectra for the anionic complexes. However, some interesting differences between the fluoro and fluorosulfato compounds deserve attention. - 130 -Table 33. Mossbauer Spectra at 80°K. Compound 6 (mm/sec) A (mm/sec) Line Width CO R K 2Sn(S0 3F) 6 -0.26 0 1.14 0 Cs 2Sn(S0 3F) 6 -0.25 0 1.11 (NO) 2Sn(S0 3F) 6 -0.28 0 1.41 (C10 2) 2Sn(S0 3F) 6 -0.30 0 1.28 Sn(S0 3F) 4-2BrS0 3F -0.16 0.87 Li 29,1.26 [Br(S0 3F) 2] 2Sn(S0 3F) 6 -0.23 0 1.29 [I(S0 3F) 2] 2Sn(S0 3F) 6 -0.25 0 1.45 Sn(S0 3F) 4 -0.27 1.34 1.05,1.32 0. 42 SnF. 4 -0.261 5 1 95 1.80 0. i 73 K 2SnF 6 1 3 7 -0.43 0 1.59 0. 61 (C10 2) 2SnF 6 1 3 7 -0.40 1.01 1.46,1.46 0. 71 K 0SnCl, 2 6 +0.48 0 1.35 0. 38 95 - 131 -Although SnF^ ^""^  and SnCSO^F)^ have identical isomer shifts, the isomer shifts of the hexafluorostannates are about 0.10 to 0.15 mm/sec lower than those of the hexakisfluorosulfatostannates. This difference indicates a slightly higher effective nuclear charge on tin for the latter group of compounds, probably caused by a lower electronegativity of the fluorosulfate group. In light of recent i. . 22,156,175,176 ^ . , . , . discussions, i t does not seem meaningful to derive electronegativity values for the SO^F group from the.sedata. The absence of any observable quadrupole splittings for (C10o)„Sn(S0„F),. contrasts with the observation for (C10„)„SnF, where 2 2 j o 2 2 b 137 a splitting of 1.01 mm/sec was observed. Since in (CK^^SnF^ the splitting i s caused by appreciable anion-cation interaction i t must be concluded that a similar interaction is either absent in (C10_)_Sn(S0„F) 2 2 J o or i t s effect i s effectively buffered by the larger SO^F groups and not relayed to the tin atom to the same extent that i t i s in the hexa-fluorostannate ion. Finally, whereas a l l hexafluorostannates give well resolved spectra at room temperature, the fluorosulfate derivatives do not seem to show the same behaviour. No resolvable spectrum could be obtained at room temperature for K 2Sn(S0 3F) 6. (C10 2> 2Sn(S0 3F) 6, [Br(S0 3F)^ 2Sn(S0 3F) 6 and [I(S0„F)o]oSn(S0„F)£ were found to interact with the mylar windows 3 2 2 J O at room temperature. With the bromine and iodine derivatives the reaction with the mylar occurred so rapidly that teflon windows had to be used instead of the mylar ones. - 132 -D. Vibrational Spectra Both Raman and infrared spectra (only down to ^  300 cm 1 because only silver halide windows were found to be inert to attack) were recorded for K 2Sn(S0 3F) 6, Cs 2Sn(S0 3F) , (N0) 2Sn(S0 3F) 6 > and (C10 2) 2Sn(S0 3F) 6 and Raman spectra only for [Br(S0„F)_]oSn(S0„F). and [I(S0„F)„]„Sn(S0„F). j 2 2 J O j 2 z J O because these compounds reacted with even AgCl windows. No reasonable spectra of the dark brown Sn(S0 3F) 4.2BrS0 3F could be recorded because of window attack in the case of IR spectra and lack of transparency i n the case of the Raman spectra. The spectral results are list e d in Tables 34, 35 and 36. 2-As expected, absorption bands due to the Sn(S0 F) ion are found j o in the same regions for a l l ^six compounds. Although reasonably good general agreement with previous work on monodentate S0 3F groups is noted, extensive vibrational coupling and solid state splitting results in complex absorption bands, particularly in the S-0 and S-F stretching regions. These effects are more apparent in the Raman spectra than in the poorly resolved infrared spectra. The spectra of [Br(S0 3F) 2] 2Sn(.S0 3F)^ and [I(S0,F)_]„Sn(S0_F), and to a lesser extent (C10 o) oSn(S0 oF). are 3 2 2 j o 2 2 j o further complicated by the large number of vibrations due to the 2-cations most of which are found in the same region as the Sn(S0„F) j o vibrations are. Since for [Br(S0„F)„]oSn(S0„F), or [I(S0„F)o]oSn(S0„F), j 2 2 J O J 2 2 J O the S03F groups attached to both Br (or I) and Sn are expected to be monodentate extensive overlap is expected and a differentiation between the two w i l l be impossible. Because of the lack of cation vibrations and the unlikeliness of cation-anion interactions, the spectra of K„Sn(S0.F) and CsoSn(S0„F) z J O z j o - 133 -Table 34. Vibrational Spectra of Alkali Metal Hexakis(fluorosulfato)-stannates(IV). K 2Sn(S0 3F) 6 Cs 2Sn(S0 3F) 6 IR Raman IR Raman Assignment 1407 m 1360 br 1390 m 1375 1278 s 1260 1200 br,s 1228 w 1205 1096 s 1080 990 br,s 1002 m 990 859 m 836 m 800 br ,s 823 m 800 620 s 625 s 621 571 sh 582 m 570 550 s 560 w 550 430 m 432 m 416 w 422 360 sh 346 mw 266 mw s,br 1407 1399 } m } v S09 a 2 sh 1270 s ? s,br 1218 w V°2 sh 1091 s s,br 995 m } v SO s,br 828 811 m m } v SF m 625 s SnOS02 bend sh 578 s S0 2 bend s 560 431 s m S0 2 rock w,sh 418 407 } w SF wag SO wag 345 w vSn-0 260 mw S0oF torsion - 134 -Table 35. Vibrational Spectra of Heterocation Hexakis(fluorosulfato)-stannates(IV). (C10 2) 2Sn(S0 3F) 6 (NO) 2Sn(S0 3F) 6 IR Raman IR Raman Assignment 2305 w 2334 m VN0+ 1420 m,sh 1412 w 1398 m Va S°2 1378 1386 s 1360 br 1385 m 1303 w 1308 w v 3C10 2 + 1290 m 1295 w 1265 ms 1272 s 1 1215 s,br 1206 s 1200 s,br 1221 m Vs S°2 1180 m, sh 1092 m 1096 vs ? 1025 1062 s V 1 02 + 1008 990 sh 1000 s,br 1026 w vSO 995 m, sh 1010 w 830 s,br 842 m 842 m , 810 s, sh 820 m 800 s ,br 822 814 sh m vSF 628 m 624 s 620 m 623 s SnOSO. bend 600 sh 604 m z 576 m 582 m 575 s 585 m S0 2 bend 550 m 555 m 555 s 559 s S0 2 rock 519 m 526 m v 2C10 2 + 430 m 429 m 420 428 sh sh 435 w SF wag 402 m 399 w - - 408 w SO wag 348 m 355 347 w vSn-0 264 w 262 w S02F torsion - 135 -Table 36. Vibrational Spectra of Halogen Bisfluorosulfato Hexkis-fluorosulfato Stannate(IV). ,[Br(SO-F) 2] 2Sn(S0 3F) 6 Raman [I(S0 3F) 2J 2Sn(S0 3F) 6 Raman Assignment 1497 w 1420 vw 1384 w 1387 m 1370 w 1248 s 1251 s 1197 m 1213 m 1184 m 1145 m 1092 m 1106 m 1020 w 1010 w 985 w 963 s 865 vs 882 m 830 w 826 w 745 vs 663 vs 652 vs 652 s 640 s 631 m 596 w 580 m 558 w 562 m 530 w 464 vs 456 w 430 vw 435 w 414 m 412 w 386 w 384 vw 367 w 370 w 331 m 309 vs 287 w 264 w 254 w 163 w,sh v S0„ cation a 2 } v SO. a 2 v S0„ cation s 2 } V s S 0 2 ? vSO vSO cation vSF cation vSF v Br-0 a v 1-0 a SnOS02 bend v Br-0 s v 1-0 s S0 2 bend S0 2 rock 6BrOS bend SF wag SO wag vSn-0 6I0S bend SBrOS wag <SI0S wag S0 2F torsion - 9CT -- 137 -should be the simplest. The general features of these spectra are 2-repeated in a l l the other Sn(SO_F), derivatives and serve as a good 3 o starting point for discussions of the vibrational spectra. The three SO^ stretching modes and the S-F stretch are found at ^1400, ^ 1200, ^ 1000, and ^830 cm 1 and are in the regions expected for SO^ and S-F vibrations of monodentate SO^F compounds. The splitting, of these bands is not unexpected, however i t seems rather unreasonable to account for the bands at 1270 and 1090 cm 1 in the Raman spectra as part of the 1200 or 1000 cm 1 systems. The remaining bands at frequencies below 700 cm 1 can be tentatively assigned as shown in Table 34. The absence of any vS-F characteristic of ionic fluorosulfates in the 750-800 cm 1 region is good additional evidence for the existence of 2-discrete Sn(S0„F)£ ions rather than formulation as mixtures of 3 o SnCSO^F)^ and KSO^F or CsSO^F respectively and complements the Mossbauer results nicely. Tne (C10_)-Sn(SO_F). and (NO)_Sn(S0„F). spectra, Table 35, have 2 2 J O 2 3 o 2-absorption bands due to the Sn(SO.F). corresponding to those of the 3 o potassium or cesium compounds as well as bands due to the heterocations. No additional splitting of the SO^F bands is observed. The C102+ vibrations, at 1060 cm \ at 526 cm 1 and at 1308 and 1295 cm 1 + 137 177 are i n agreement with previous observations on ClO^ ' is 35 37 s p l i t into two components presumably due to CI and CI isotopes as observed previously. In (NO)oSn(S0.F),, vN-0 is found at 2334 cm"1 2 j o -1 137 and compares well with 2325 cm in (N0)oSnF^ rather than the value I 6 —1 137 178 of 2207 cm reported for (NO)-SnCl-. ' These findings for the 2 O heterocation complexes are thus consistent with ionic formulations with CK> 2 + and N0+ cations. - 138 -The Raman spectra of [Br(S0 3F) 2] 2Sn(S0 3F) 6 and [I(S0 3F) 2] 2Sn(S0 3F) 6 > Table 36, are considerably more complex than those of the previous four compounds, however, a l l but two of the absorption bands seen in KoSn(S0„F)^ are found here in very similar positions. The unassigned 2 j o band at 1278 cm 1 in K„Sn(SO-F), is missing in both of these complexes 2 3 o and the 625 cm 1 band tentatively assigned to an S0 3 bending mode i s shifted by 25 cm"1 to 652 cm - 1 in both [Br(S0 3F) 2] 2Sn(S0 3F) 6 and [I(S0 3F) 2] 2Sn(S0 3F) 6. There are ten additional bands which could be due to the Br(S0 3F) 2 + cation in the Raman spectrum of [Br(SO„F)„]_Sn(SO_F),. These bands can 3 2 2 j o be assigned with the help of some literature data and the Raman spectrum of Br(SO.F)„SbF (discussed more fu l l y in Chapter VII). Gillespie and 3 2 6 179 + Morton identified the Br(SC>3F)2 cation in the HS03F/SbF5 -solvent system and recorded a Raman spectrum of this solution and attributed three bands to the cation, however, since these bands are also found in B r ( S n 3 F ) 3 , the assignment to the Br(S0 3F) 2 + ion must be regarded as questionable. These values as well as the Br(SC> 3F) 2 + modes of Br(S0 oF) oSbF. and the ten bands in [Br(SO„F).]oSn(S0„F), are listed 3 2 o j 2 2 j o in Table 37. Looking f i r s t at the S n 3 stretching modes, they are in f a i r l y typical positions for monodentate S03F although the 1500 cm 1 band seems to be at f a i r l y high energy for a solid compound. vS-0 is at 1424 cm"1 in "KBr(Sd F).^ 1 6 9 and 1490 cm"1 in B r ( S 0 3 F ) 3 , 1 6 9 so 1497 cm"1 is probably reasonable for Br(S0 3F) 2 +. Only one of these S0 3 stretching 179 modes was observed in the super acid solution., but since the observed - 139 -Table 37. Br(S0 3F) 2 Vibrational Modes. Br(S0 3F) 2SbF 6 Br(S0 3F) 2 [Br(S0 3F) 2] 2Sn(S0 3F) 6 Assignment "1514 1497 v SO. a 2 1254 1242 1248 V°2 980 985 vSO 875 865 vS-F 720 745 v Br-0 a 616 640 v Br-0 s 468 462 464 SOBr bend 320 308 309 SOBr wag - 140 peak i s by far the most intense of the three in [Br(SO.F).]_Sn(S0oF), or 3 z Z J o Br(S0„F)„SbF^, this result seems reasonable. The S-F stretch i s j z o assigned to the band at 865 cm 1 in the tin compound. The remaining bands cannot be assigned with certainty. The di f f i c u l t i e s arise because of the lack of data on Br-0 stretching frequencies and the general uncertainty in the region below 600 cm 1 due to overlap of modes due to the cation with those of the anion. Literature values on bromine-oxygen stretching modes are available for the fluorosulf ates BrSO^F 7<^ and BrCSO^F)^ and the anions Br(S0 3F) 2" 7 0 and B r ( S 0 3 F ) 4 ~ . 1 6 9 In addition values for Br 20 1 8 0 181 - 182 (with bridging oxygen) and Br0 3F and Br0 3 are reported and the assigned stretching modes for a l l these compounds are listed in Table 38. In [Br(S0-F)o]„Sn(S0.F), the only possible assignment for Br-0 3 z z j o stretching modes are two bands at 745 and 640 cm ^. This region is free from S03F vibrations (stretching modes are higher than 800 cm 1 and bending modes should be below 630 cm "*). Two stretching modes would be expected for a bent Br0 2 grouping. Although i t i s reasonable to expect an increase i n Br-0 stretching frequencies when going from the anion Br(S0 3F) 4 over the neutral molecule Br(S0 3F) 3 to a cation as shown by the corresponding trends for Br-F derivatives, i t appears that previous assignments for Br-0 stretching modes in the S03F derivatives are decidedly too low and should be reassigned. For instance, in the case of BrS0 3F and the Br(S0 3F) 2 ion an assignment more consistent with the subsequently 69 published spectra of the halogen fluorosulfates FS03F and C1S03F - 1 4 1 -T a b l e 3 8 L i t e r a t u r e V a l u e s f o r S > B r - 0 a n d i ) B r - F C o m p o u n d V B r - O C c m " 1 ) \ ) B r - F ( c m " 1 ) R e f . B r S 0 3 F 4 6 4 70 B r(S03* ,) 3 4 5 5 , 3 8 4 1 6 9 B r ( S 0 3 F ) 2 ~ 4 3 7 7 0 B r ( S 0 3 F ) 4 ~ 4 4 7 , 399 1 6 9 B r 2 0 587 1 8 0 B r O j F 9 7 4 , 8 7 5 1 8 1 B r 0 3 " 836, 8 0 6 1 8 2 532, 4 5 7 1 8 3 B r F 2 ~ 5 9 6 1 8 4 B r F 3 6 7 4 , 6 1 3 , 5 8 1 1 8 5 B r F 2 + 7 1 5 , 7 0 6 1 8 3 T a b l e 3 9 B r - 0 S t r e t c h e s i n F l u o r o s u l f a t e s C o m p o u n d \ ) B r - 0 6 1 5 6 1 8 6 5 9 721, 6 4 5 , 6 1 2 7 4 5 , 6 4 0 B r ( S 0 3 F ) 4 B r ( S 0 3 F ) 2 BrSOjF B r ( S 0 3 F ) 3 B r ( S 0 3 F ) 2 - 142 -would place vBr-0 for BrSO^F at 659 cm 1 and the Raman active BrO^ stetch for a presumably linear O-Br-0 configuration in BrCSO^F^ at 618 cm ^. In a very similar way assignments for the trivalent deriva-tives Br(SO„F) and Br(SO F) w i l l have to be revised. Strong bands at 721, 645 and 612 cm 1 are more appropropriately assigned to the expected three Br-0 stretching modes in BrCSO^F)^ and for the anion a strong band at 615 cm 1 is at least one of the Br-0 stretches. The remaining frequencies can again be accounted for assuming monodentate SO^F groups but a concrete revision appears to be beyond the scope of this thesis. The new assignments are shown in Table 39 to indicate the trends in vBr-O. The positions of the Br-0 stretching modes in BrCSO^F)^"^ should also be confirmed by positions of corresponding modes for the I(S0.jF)2+ cation and indeed bands at 663 and 631 cm 1 are most conveniently assigned as the asymmetric and symmetric stretches in [iCSO.jF^^Sn^O.jF)^. E. Solution Studies in HSO^F The solutions of K„Sn(S0 oF). and (C10„).Sn(SO„F). in HSO-F are 2 3 o 2 2 3 o J both ionic, but the solutes are less f u l l y dissociated than KSO^F i s . These solutions were shown to be basic by titration of a solution of KoSn(S0„F),, in HSO.F with the standard base KSO_F. They values 2 j o 3 3 _2 calculated for concentrations up to 6 x 10 moles/kg are list e d in Table 40. A plot of concentration vs. specific conductance i s shown in Figure 21. As shown i n the figure, the solutions of (CIO.)„Sn(S0.F)£ are 2 2 3 b slightly less conducting than those of K_Sn(S0„F)£. The same order has 2 3 o been observed previously for solutions of KSO^F and CK^SO^F in HSO^F ^ ® - 143 -TABLE E l e c t r i c a l Conductance of K„ [SnCS0oF),J and l J O CCA0 2) 2[SnCS0 3F) 6J in HSC^F at 25.00°C; Y Values Specific Conductance [ohm-1 cm - 1] x 10 4 y values for K ^ S n C S C y ) ^ Y values for , v CC£0 2) 2lSnCS0 3F) 6J b ; 10 0.73 0.59 15 0.63 0.55 20 0.58 0.51 25 0.55 0.48 30 0.52 0.45 35 0.49 0.44 a) _ , experimental concentration range: 1.6 -4 x 10 to 6.3 x 10~2 moles kg ^^experimental concentration range: 3.0 x 10"5 to 3.5 x 10~2 moles kg <r 144 -Figure 21. Conductivities of K2[Sn(S03F)( and (Cl02)2[Sn(S03F)6] at 25 °C 10 20 30 40 50 60 C [moles kg-]x 103 - 145 -and has been attributed to differences in the mobilities of the solvated cations. The formation of the solvated C l O ^ cation i s indicated by the yellow-red color of the solutions of K^SnCSO^F)^. Clear solutions were obtained even at the higher concentrations 19 needed for F nmr studies. This result excludes the possibility that SnCSO^F)^ which is virtually insoluble in HSO^F may have been formed. 19 For K oSn(S0„F)only a single F nmr line at -41.37 ppm downfield 2 3 o from CFCl^ used as an external standard could be found. In addition the signal for HSO^F is found at -40.76 ppm, virtually unchanged from the position in the neat solvent. This finding excludes any appreciable SO^F exchange between solute and solvent at room temperature. The nmr evidence points to a rather simple mode of ionization without appreciable condensation, polymerization or dissociation. For the three possible modes: HSO F (1) X oSn(S0 oF). 2X (solv) + Sn(S0_F), (solv) 2 3 o J D HSO F (2) X oSn(S0 oF). + HSO^F 2X (solv) + HSn(SO.F), (solv) 2 3 6 3 3 o + SO F~(solv) HSO F (3) X_Sn(S0_F), + 2HS0.F — 2 X (solv) + HoSn(S0.F).(solv) 2 3 0 3 2 j b + 2S03F (solv) X = C102 or K the expected y values would be 0, 1 or 2 respectively. The observed 2-value of ^0.5 indicates an equilibrium with Sn(SO-F),, and 3 b HSn(S0„F), as the dominant tin species. The fact that both are 3 o indistinguishable in the nmr spectrum may be attributed to fast proton - 146 -exchange between the two ions. It appears that the neutral acid HoSn(S0„F)^ behaves as a good proton donor in HS0oF. This is not surprising since HSbE^CSO^F)^"'"^1 is also a rather strong acid or proton donor in this solvent system. It must be concluded that the hexakisfluorosulfatostannate(IV) moiety can exist in solutions of strong protonic acids without appreciable alteration. - 147 -CHAPTER VII REACTIONS OF LEWIS ACIDS WITH FLUOROSULFATES A. Introduction The reaction between CK>2S03F and Sn(S03F>4 f a l l s into the general category of donor-acceptor reactions which proceed with complete anion transfer. As in most reactions of this type described in the literature only one type of anion, common to both the donor and the acceptor, i s present. We decided to study a similar type of reaction in order to obtain i f possible 'mixed ligand complexes. The most likely acceptors would be SnF^, SbF,., AsF,. or BF 3 and the donor of choice, C102S03F, for the following reasons. (1) The C10 2 + cation i s well characterized and i s e.g. formed readily in the reaction of 187 FC10o + AsF c >- ClO.AsF, , 2 5 2 6 (2) C102S03F has convenient physical properties for this purpose and (3) the two anions F and S03F are expected to have similar electronegativities. If may be noted that similar interactions of a fluoride donor with a - 148 -fluorosulfate as acceptor, e.g. FNC^ + SnCSO^F)^, had failed. There are very few relevant precedents for this type of compound. Anions of the [SbF (SO.F)^ ] type have been identified in so-called n 3 6-n super acid solutions."'"^"'" They are evidently formed by SO^ insertion into SbF,. and subsequent or concurrent solvent interaction with HSO^F. B. C102SO^F Reactions 1. C10„S0oF + SnF. 2 3 4 CK^SO^F reacts with SnF^ at room temperature to produce a gas, identified as C102 by infrared spectroscopy, and a yellow solid shown by i t s infrared spectrum to consist mostly of (C102)2Sn(S0,jF)g. C102 may well be formed from C102F originally produced by reaction with traces of moisture. In this reaction approximately 3.09 g (18.1 mmoles) of C102S03F were added to 0.832 g (4.27 mmoles) of SnF^ in a one part reactor in the dry box. These two components reacted to produce C10o and (C10o)oSn(S0„F) plus some unreacted solid, probably unreacted 2 2 2 3 6 SnF. or (CIO.)-SnF,.. The total weight of nonvolatile products was 4 2 2 o 4.07 g. This reacton was not investigated further once (C10 2) 2Sn(S0 3F)g was identified as a major product, indicating that this route would not 2-lead to the expected SnF 4(S0 3F) 2 . C102S03F + SbF5 C102S03F reacted with SbF5 with the transfer of F from C102S03F to SbF5 to produce C l O ^ b ^ ^ and SbF^SO.^ according to C10„S0„F + 3SbFc C10.Sb oF 1 1 + SbF.SO.F. 2 3 5 2 2 11 4 3 - 149 -In a typical reaction an excess of SbF,. was added to 1.021 g (6.14 mmoles) of C102S03F in a two part reactor in the dry box... The CK>2S03F slowly dissolved i n the SbF,. and then reacted to produce a yellowish solid and some clear, moderately volatile liquid. Separation was accomplished by d i s t i l l i n g off a l l volatiles. SbF^SO^F was identified in the d i s t i l l a t e by comparing i t s properties and infrared spectrum with those of SbF 4S0 3F published by G i l l e s p i e , 7 1 although there are some discrepancies in Gillespie's vibrational spectra. After a l l volatiles were removed in vacuo at 40-50° for several days 3.19 g of a white solid, m.p. 57-58° without decomposition, was obtained. The solid which was free of S03F (test with BaCl 2 after hydrolysis) was later identified as C10 2Sb 2F^ by elemental analysis. The elemental analysis results are; calculated for C10 2Sb 2F^, Sb, 46.74% and F, 40.31%, and found, Sb, 46.51% and F, 40.26%. The vibrational spectral results for C K ^ S b ^ ^ are shown in Table 41. The three vibrational modes for C10 2 + are observed; v 1 at 1050 IR and 1051 Raman, v 2 at 513 IR and 514 Raman and the v 3 doublet 1310 and 1295 IR and 1310 and 1294 Raman, in agreement with + 137 previous work on C102 . The remaining peaks would be assigned to vibrations of the S b 2 F ^ anion and are in positions similar to those 187 188 189 found for XeFSbjF , N 2 F 3 S b 2 F l l ' a n d V 0 2 S b 2 F l l ' A n approximate 189 assignment is given by Wiedlein and Dehnicke and further assignment w i l l not be attempted here. These spectra differ significantly from the spectra of hexafluoroantimonates in two respects. The SbFg compounds exhibit no SbF^ vibrations above about 675 cm 1 even when the symmetry of the anion is lowered from 0^ by anion-cation interactions - 150 -Table 41. Vibrational Spectra of C102Sb F.^. Infrared Raman Assignment 1310 m 1310 m 1295 m ' 1294 w 1150 w,br 495 m 290 s,br 305 w } V3 C 1 ° 2 + <V 1050 w 1051 s v CIO + (v ) X £ - S 702 m 690 vs,br 677 s 660 sh 647 s 626 w 590 w 595 m 513 s 514 m v 2 C10 2 + (6) - 151 -Table 42. Vibrational Spectra of Sb^F a n d s b F g Salts. O N F 2 S b 2 F n 1 8 8 N 2 F 3 S b 6 F n 1 8 8 XeFSb 2F 1 1 1 8 7 Vt^Sb^ 1 8 9 699 680 614 296 284 725 695 665 298 220 693 682 662 654 765 730 705 700 685 523 278 275 225 BrF„SbF^ 2 6 183 ClF„SbF_ 2 6 183 KSbF, 183 678 641 552 523 493 284 270 662 644 596 542 537 292 282 267 661 575 294 278 - 152 -and a l l SbF, spectra have an absorption between 550 and 575 cm 1 6 which i s not seen in any of the Sb2F.^ complexes. The f i r s t stage of the reaction of CK^SO^ with SbF,. probably involves the transfer of S03F from C102S03F to SbF5 to form C10 2SbF 5S0 3F as an intermediate. Either C102SbF,.S03F cleaves off S0 3 > which may subsequently react with another molecule of SbF,. 7 1 according to C10_SbFcS0oF *• C10oSbF£ + S0 o 2 5 3 2 6 3 ClO-SbF, + SbFc *~ ClO.Sb-F.. Z D 5 2 2 11 SO. + SbFc >- SbF.S0„F 3 5 4 3 anion replacement may take place: C10oSbFcS0.F + SbFc >- C10„SbF, + SbF.S0„F 2 5 3 5 2 6 4 3 C10oSbF, + SbF_ ->• ClO-Sb.F.,., 2 o 5 2 2 11 or C102SbF^S03F may react with another SbF,. molecule prior to S0 3 cleavage: C10oSbF_S0oF + SbF' =»- ClO.Sb.F. „S0oF 2 5 3 5 2 2 10 3 C10 2Sb 2F 1 0S0 3F ^ C10 2Sb 2F n + S0 3 S0 3 + SbF5 >- SbF 4S0 3F - 153 -The reaction with CK^SO^F in excess was subsequently attempted but removal of a l l residual CIO SO^F after reaction had occurred was not possible even with prolonged pumping. However no SbF^SO^F or SO^ could be detected indicating that neither of these possible byproducts was produced. A more feasible route to ClO^SbF^SO^F could possibly be the reaction of C10„F with SbF.SO.F but this reaction was considered 2 4 3 beyond the scope of this thesis. 3. C10 2 S0 3F + AsF 5 Gaseous arsenic pentafluoride reacts with CIC^SO^F to produce ClO2AsF._SO.jF. For this reaction a two part reactor was charged with 0.345 g (2.08 mmoles) of CH^SO^F in the dry box, evacuated in a vacuum line, and then the flask and vacuum line were pressurized to 300 mm Hg with AsF,. at room temperature. Over a period of about four hours the red color of CK^SO^F slowly changed to yellow. If the excess AsF,. was removed and the remaining nonvolatile yellow liquid cooled in liquid nitrogen a yellow glass was formed and on warming to room temperature the glass crystallized, with the evolution of a small amount of gas, to a yellow solid. 0.721 g (2.20 mmoles) of CIO AsF S03F which melted at 38° to an orange liquid was produced in this reaction. The composition was established by elemental analysis: calculated for C10 2AsF 5S0 3F, CI, 10.54%, F, 33.89% and S, 9.53% and found, CI, 10.40%, F, 34.10% and S, 9.67%. The vibrational spectra listed in Table 43 indicate the presence of C102 + , weakly bridging SO^F, and AsF^. The CK^ "** modes are in the same regions as found previously and the higher frequency AsF modes - 154 -Table 43. Raman Spectrum 0 f ClC^AsF^C^F. Raman Assignment 1359 s V S 0 3 1307 m V 3 C 1 0 2 + 1291 w 1205 vs V S 0 3 1060 sh v CIO + 1057 vs 1 I 1034 w vS0 3 856 w 849 m vSF 700 s vAsF 668 vs vAsF 641 w 602 sh 596 m 559 m 524 s 451 sh >remaini 448 w 429 m 382 m 366 sh 353 s 315 m 298 sh 272 w v 3C10 2 + (5) - 155 T-(the only ones which could be assigned with any certainty) are in the 190 191 positions expected from comparisons to spectra of C10oAsF , AsF , 2 o _> 192 and BrF(.. The positions of the SO^F modes, particularly the band at 1034 cm 1 , are somewhat unusual. This band which can only be due to the third SO,, stretching mode is at higher frequency than would be 70 2-expected for a monodentate SO^F group (see e.g. BrSO^F or Sn(S0.jF)g complexes) and is rather reminiscent of the lowest frequency SO^ stretch in bidentate compounds (see Table 25). This suggests that the SO^F group may be weakly bridging between the AsF,. and C102 groups, i.e. that there is some anion-cation interaction between the C10 2 + and the fluorosulfate group. Anion-cation interactions have been postulated previously for C10 2 + compounds on the basis of anion dependent Cl-0 stretching modes and non-zero quadrupole splitting in the Mossbauer 137 spectrum of (C10„)_SnF^. However, the shifting of the S0 o stretch 2 2 o J -1 2-by approximately 25 cm from i t s position in Sn(S0„F), compounds j o i s perhaps too small to allow definite conclusions. 19 The F nmr spectrum of C102AsF,.S03F in HS0..CF3 shows two fluorine environments in addition to the solvent resonance. One at -40.3 ppm downfield for CFC13 is in the region in which fluorosulfate fluorides are found (HS0.-F is -40.7 ppm) and the other, a broad resonance at +49.1 ppm, is due to the fluorines on arsenic. No fine structure was detected at room temperature. 4. C102S03F + BF 3 A reaction similar to the reactions of C102S03F with SbF^ and AsF,. was attempted with C102S0.-F and BF 3 by adding a 300 mm Hg pressure - 156 -of BF^ to CIO SOgF at room temperature but there was no evidence of reaction. The same reactants were mixed overnight at -78° using 550 mm Hg pressure of BF^ and 0.455 g (2.74 mmoles) of CK^SO^F. Some orange coloured solid which melted at -25° to a red liquid was produced. At room temperature 0.498 g of red liquid remained after a l l volatiles were removed. 2.74 mmoles of ClO2BF.jSO.jF would weigh 0.640 g. This suggests that either very l i t t l e reaction occurred or the product is unstable at room temperature and decomposed to C102S03F and BF^. Only BF^ was detected by IR in the gas phase. C. KS03F + SbF5 The reaction of KSO^F with SbF^ also produced some unusual results. When an excess of SbF,. was added to 0.681 g (4.94 mmoles) of KSO^F in the dry box and the solution was allowed to react for two days a white, crystalline, moderately hygroscopic solid which was insoluble in the excess SbF,. and melted over the broad range of 110-130° was produced. The liquid phase was shown by i t s IR spectrum to contain SbF^SO^F.71 The IR spectrum of the solid indicated that i t contained unreacted KS03F, KSbFg (confirmed by the presence of SbF^SO^ in the liquid) and some compound containing covalent fluorosulfate groups. The S0 3 stretching frequencies of this covalent fluorosulfate are found at 1410, 1220 and 1012) cm"1 and the S-F stretch at 822 cm"1. The weaker peaks at 582 and 554 cm 1 would possibly be S0 3 deformation modes. Reactions of KS03F and SbF<. for shorter periods of time also produced both solid products. - 157 -D. BrSO^F and BrCSO^F) Reactions BrSO^F reacts with either a defi c i t or an excess of SbF,. to produce a deep red coloured liquid and no very volatile byproducts or solid precipitates. This red colour i s characteristic of the B r 2 + 193 cation. An infrared spectrum of the red liquid formed in the reaction of BrSO^F and excess SbF^ was recorded and i t indicated the presence of both bidentate SO^F and possibly monodentate SO^F as well as SbF . x Br(SO.jF)3 reacted with SbFc to produce a pale yellow coloured solution from which white crystals slowly separated. When an excess ;(15.;4gg,;71.0 mmoles) of SbF5 was d i s t i l l e d onto 3.80 g (10.1 mmoles) of Br(S0.jF)3 a reaction occurred with gas evolution to produce a brown coloured intermediate solution and then a yellow solution. When the excess SbF,. was removed at 50° no crystallization occurred and 5.51 g (10.7 mmoles i f Br(S0 F)„SbF ) of the yellow solution remained. On 3 2 6 standing for several days some solid precipitated out, however, attempts to separate this solid were unsuccessful. SbF^SO^F was identified by i t s IR spectrum i n the liquid removed from the reactor. A Raman spectrum, Table 44, of the nonvolatile yellow liquid was recorded. This spectrum shows Br^O^F^"*" peaks very similar to those + 179 of the cation in [Br(S0„F)„]„Sn(S0„F). and Br(S0 oF). /super acid 3 2 2 3 o 3 2 and SbF." bands at 660, 564, 299, and 283 cm"1 indicating that o Br(S0„F)„SbF. i s present in the liquid. 3 2 o - 158 -Table 44. Raman Spectrum of Br(SO F) SbF Mode Assignment 1514 m Va S°2 1425 vw 1254 s Vs , S 02 1080 w 985 w vSO 897 w,sh 875 m vS-F 720 vs v Br-0 a 705 w, sh 660 vs vSb-F (v±) 616 w v Br-0 s 564 s vSb-F (v 2) 530 w, sh S0 2 rock 468 m SOBr bend 435 w SF wag 320 w SOBr wag 299 w SSb-F } ( v 283 w 6Sb-F 230 w, sh - 159 -CHAPTER VIII GENERAL CONCLUSIONS A. Summary The application of two simple synthetic methods, the acid solvolysis of organotin(IV) chlorides by strong protonic acidsj in particular fluorosulfuric acid, and the rather unique nonstatistical redistribution reactions of organotin(IV) chlorides with fluorosulfates, has resulted in the preparation of novel methyltin(IV) chloride sulfonates. The second method, ligand scrambling, could also be extended to reactions of Sn(SO.jF)4 with tin(IV) halides resulting in several tin(IV) fluorosulfate derivatives. With TiCl^, complete SO.-F transfer was observed and the structurally interesting compound Ti^Cl^(SO..F)^ with possibly tetradentate fluorosulfates was obtained. The investigation of the complexing a b i l i t y of Sn^O^F)^ resulted 2-in the detection of the [Sn(SO F) ] ion. Ligand substitution of CI J o 2-in SnCl^ by SO„F using S 0 F provided an alternative route to this 6 J 3 2 6 2 ion. The surprising thermal st a b i l i t y of the hexakisfluorosulfato stannate ion allowed the formation of the new halogen bisfluorosulfato cations I (SO^F) 2 + and Br(S0.-F)2+. Attempts to obtain these and similar cations resulted in the identification of the anion [AsF^SO^F] in an offshoot of the original study. - 160 -Besides occasional solution studies of interesting solutes in 119 HSO^F our primary interest focussed on the use of Sn Mossbauer spectroscopy, Laser Raman spectroscopy and infrared spectroscopy in elucidating the identities and structures of the compounds synthesized. Two basic structural types emerged. The f i r s t was trans octahedral bisfluorosulfates (and other related sulfonates) of the type XYSn(S0 3F) 2 (X, Y = CH<3, CI, Br, F, and S03F) with the fluorosulfate groups functioning as bidentate bridging groups with coordination through oxygen. This postulate was originally based on vibrational c r i t e r i a such as the apparent symmetry lowering from C ^ to C resulting in an increase from six to nine fundamentals a l l found s in well defined postions and was later confirmed by an X-ray diffraction 135 study of (CH 3) 2Sn(S0 3F) 2. These results together with the Mossbauer data allowed good insight into the bonding in these compounds. The postulated model using 3 center 4 electron bonds and 5p and 5p orbitals x y on tin for tin-oxygen bonding leaving a 5s5p hybrid orbital for bonding z to X and Y, is consistent with both the X-ray diffraction results and the Mossbauer parameters. Correlations such as between the electronegativ-i t i e s of the ligands X and Y with the observed isomer shifts and between their Taft inductive constants and the quadrupole splittings were rather successful. The second group, supposedly trigonal bipyramidal monofluorosulfates (and other sulfonates) of the type X2YSnS03F (X, Y = CH 3 > CI), could be rationalized as having bidentate bridging fluorosulfates as before. Results for these compounds could not be correlated nearly as well because a tendency towards hexacoordination around t i n via potentially - 161 -bridging chloride ligands w i l l cause distortions. 2-Structural evidence for Sn(SO_F)., rests primarily on Mossbauer 3 b results, in particular on the absence of any quadrupole splitting. The vibrational spectra, i n particular for [X(SO„F)„]„Sn(SO„F), compounds, i l l j o are rather complex and defy any rigorous assignment in the SO^F stretching and bending regions. As is so often the case in the stereochemistry of tin(IV) and organotin(IV) compounds, the "normal" coordination number of four was not found for any of the newly synthesized compounds. B. Suggestions for Further Work A number of possi b i l i t i e s for expansion emerge as a result of this study, some of which are currently investigated. These possibilities are mentioned here in no particular order. (1) The extension of protonic acid solvolysis to other suitable organometallic derivatives of some main group elements such as Ge, Pb, Sb, or Bi may be contemplated. The use of other organotin(IV) halides with larger alkyl groups, vinyl groups or aromatic groups has been partially 143 142 explored in connection with this study as has the use of HF or 143 H P 0 2 F 2 a S a c i d s " (2) The use of similar ligand redistribution reactions may prove f r u i t f u l . Recently completed work on ligand rearrangement reactions of 153 titanium(IV) chloride-sulfonates supports this view. (3) The identification of bidentate bridging sulfonate groups in 194 other systems should be facilitated. The vibrational study of Ga(SO.jF)3 may serve as an example here. - 162 -(4) Refinement of the Mossbauer data by experimental determination of the sign of the electric f i e l d gradient tensor i s necessary as a basis for detailed point charge calculations and to arrive at a more definite view on the supposedly trigonal bipyramidal compounds. This study i s currently in progress. (5) The possible use of SnCSO^F)^ as a Lewis acid deserves more attention. Interest could center around other possible fluorosulfato cations. (6) The search for other halide-fluorosulfate anions like [AsFj-SO^F] may be successful. (7) Last but not least, i t i s hoped that some of the postulated structures may be confirmed by X-ray diffraction studies. - 163 -REFERENCES 1. W.P. Neumann, "The Organic Chemistry of Tin", John Wiley and Sons, London, 1970. 2. R.C. Poller, "The Chemistry of Organotin Compounds", Academic Press, New York, 1970. 3. A.K. Sawyer (Editor), "Organotin Compounds" (3 vols), Marcel Dekker, Inc., New York, 1971. 4. M. Dub (Editor), "Organometallic Compounds", Vol. 2, Springer Verlag, New York, 1967. 5. R. Okawara and M. Wada, "Structural Aspects.of Organotin Chemistry", Adv. Organometal. Chem., 5_, 137 (1967). 6. B.Y.K. Ho and J.J. Zuckerman, "Structural Organotin Chemistry", J. Organometal. Chem., 49, 1 (1973). 7. V.I. Goldanskii, V.V. Khrapov, O.Y. Okhlobystin and V.Y. Rochev, 119 " Sn Metal Organic Compounds", in V.I. Goldanskii and R.H. Herber (editors), "Chemical Applications of Mossbauer Spectroscopy", Academic Press, New York, 1969, p. 336. 8. J.J. Zuckerman, "Applications of 1 1 9 m S n Mossbauer Spectroscopy to the Study of Organotin Compounds", Adv. Organometa1. Chem., _9, 21 (1970). 9. P.J. Smith, "Mossbauer Parameters of Organotin Compounds", Organometal. Chem. Rev., A, 5_, 373 (1970). 10. J.W. Emsley, J. Feeney and L.H. Sutcliffe, "High Resolution Nuclear Magnetic Resonance Spectroscopy", Vol. 2,Pergamon Press, Oxford, 1966, p. 1082. 11. R.S. Tobias, "o-Bonded Organometallic Cations in Aqueous Solution and in Crystals", Organometal. Chem. Rev., 1, 93 (1966). - 164 -12. H.C. Clark, "Organometallic Cations", i n E.A.V. Ebsworth, A.G. Maddock and A.G. Sharpe (editors), "New Pathways in Inorganic Chemistry", University Press, Cambridge, 1968, p. 1. 13. K. Moedritzer, "The Redistribution Reaction", in E.I. Becker and M. Tsutsui, "Organometallic Reactions", Vol. 2, Wiley Interscience, New York, 1971, p. 1. 14. A.J. Bloodworth, "Tin and Lead", in B.J. Aylett (editor), "Organo-Metallic Derivatives of the Main Group Elements", University Park Press, Baltimore, 1972, p. 275. 15. J.D. Donaldson, "The Chemistry of Bivalent Tin" in F.A. Cotton (editor), "Progress in Inorganic Chemistry", Vol. 8, Interscience, New York, 1967, p. 287. 16. H.C. Hodge and F.A. Smith, "Effects of Fluoride on Bones and Teeth", in J.H. Simons, "Fluorine Chemistry", Vol. 4, Academic Press, New York, 1965, p. 377. 17. R. Hulme, J. Chem. Soc, 1524 (1963). 18. Y. Hermodsson, Arkiv. For. Kemi., 31, 73 (1969). 19. I.R. Beattie, M. Milne, M. Webster, H.E. Blayden, P.J. Jones, R.C.G. Killean, and J.L. Lawrence, J. Chem. Soc. A, 482 (1969). 20. R.F. Bryan, J. Amer. Chem. Soc, 86, 733 (1964). 21. G. Hebecker, H.G. Von Schnering and R. Hoppe, Naturwiss., 53, 154 (1966). 22. R.H. Herber and H. Cheng, Inorg. Chem., 8, 2145 (1969). 23. J.P. Clark and C.J. Wilkins, J. Chem. Soc. A, 871 (1966). 24. R. Hoppe and W. Dahne, Naturwiss., 49, 254 (1962). 25. E.O. Schlemper and W.C. Hamilton, Inorg. Chem., 5_, 995 (1966). 26. A.G. Davies, H.J. Milledge, D.C. Puxley and P.J. Smith, J. Chem. Soc. A, 2862 (1970). - 165 -27. R.A. Forder and G.M. Sheldrick, J. Organometal. Chem., 21, 115 (1970). 28. E.O. Schlemper and D. Brltton, Inorg. Chem., _5, 507 (1966). 29. H.C. Clark, R.J. O'Brien and J. Trotter, J. Chem. Soc, 2332 (1964). 30. E.O. Schlemper, Inorg. Chem., 6, 2012 (1967). 31. G.S. Brownlee, A. Walker, S.C. Nyburg, and J.T. Szymanski, Chem. Comm., 1073 (1971). 32. CD. Garner, D. Sutton and S.C. Wallwork, J. Chem. Soc. A, 1949 (1967). 33. N.W. Alcock and R.E. Timms, J. Chem. Soc. A, 1873 (1968). 34. N.W. Alcock and R.E. Timms, J. Chem. Soc. A, 1876 (1968). 35. B.F.E. Ford, B.V. Liengme and J.R. Sams, Chem. Comm., 1333 (1968). 36. CS. Harreld and E.O. Schlemper, Acta Cryst. B, 27, 1964 (1971). 37. K. Furue, T. Kimura, N. Yasuoka, N. Kasai, and M. Kakudo, Bull. Chem. Soc. Jap., 43, 1661 (1970). 38. L.J. Guggenberger, Chem. Comm., 512 (1968). 39. R.L. Voorhees and R.W.J. Rudolf, J. Amer. Chem. Soc, 91, 2173 (1969). 40. H.A. Bent, Chem. Rev., 61, 275 (1961). 41. J.R. Holmes and H.D. Kaesz, J. Amer. Chem. Soc., 83, 3903 (1961). 42. N. F l i t c r o f t and H.D. Kaesz, J. Amer. Chem. Soc, 85, 1377 (1963). 43. W.F. Edgell and CH. Ward, J. Amer. Chem. Soc, 77, 6486 (1955). 44. R.V. Parish and R.H. Piatt, J. Chem. Soc., A, 2145 (1969). 45. W.F. Edgell and CH. Ward, J. Mol. Spect., 8, 343 (1962). 46. J.C H i l l , R.S. Drago and R.H. Herber, J. Amer. Chem. Soc, 91, 1644 (1969). 47. W.J. Lehman, J. Mol. Spect., 7_, 261 (1961). 48. B. Gassenheimer and R.H. Herber, Inorg. Chem., J3, 1120 (1969). - 166 -49. R.C. Thompson in G. Nickless (editor) "Inorganic Sulfur Chemistry", Elsevier, Amsterdam, 1969, p. 587. 50. H.C. Clark and R.J. O'Brien, Inorg. Chem., 2, 1020 (1963). 51. M. Lustig and G.H. Cady, Inorg. Chem., 1, Ilk (1962). 52. J.R. Dalziel, private communication. 53. J.K. Ruff in W.L. Jolly (editor), "Preparative Inorganic Reactions", Vol. 3, Wiley,. New York, 1966, p. 35. 54. S. Williamson, Prog. Inorg. Chem., ]_, 39 (1966). 55. A.A. Woolfe, in E.A.V. Ebsworth, A.G. Maddock and A.G. Sharpe (editors), "New Pathways in Inorganic Chemistry", University Press, Cambridge, 1968, p.327. 56. F.A. Cotton and G. Wilkinson, "Advanced Inorganic Chemistry", 2nd Edn., Interscience, New York, 1966, p. 560. 57. E. Hayek, A. Czaloum and B. Krismer, Mh. Chem., 87, 741 (1956). 58. E.L. Meutterties and D.D. Coffman, J. Amer. Chem. Soc, 80, 5914 (1958). 59. E. Hayek, J. Puschman and A. Czaloum, Mh. Chem., 85, 359 (1954). 60. J. Goubeau and J.B. Milne, Can. J. Chem., 45, 2321 (1967). 61. J.M. Shreeve and G.H. Cady, J. Amer. Chem. Soc, 83, 4521 (1961). 62. D.D. Des Marteau, Inorg. Chem., ]_, 434 (1968). 63. D.W.A. Sharpe, J. Chem. Soc, 3761 (1957). 64. A.M. Qureshi, H.A. Carter and F. Aubke, Can. J. Chem., 49, 35 (1971). 65. K. 0'Sullivan, R.C. Thompson and J. Trotter, J. Chem. Soc. A, 2024 (1967). 66. K. 0'Sullivan, R.C. Thompson and J. Trotter, J. Chem. Soc. A, 1814 (1970). 67. A. Kvick, P. Jonsson and I. Olovsson, Inorg. Chem. , 8_, 2775 (1969). 68. .F.B. Dudley and G.H. Cady, J. Amer. Chem. Soc, 79, 513 (1957). 69. A.M. Qureshi, L.E. Levchuk and F. Aubke, Can. J. Chem., 49, 2544 (1971). - 167 -70. A.M. Qureshi and F. Aubke, Inorg. Chem., 10, 1116 (1971). 71. R.J. Gillespie and R.A. Rothenbury, Can. J. Chem., 42, 416 (1964). 72. R.J. Gillespie and W.A. Whitla, Can. J. Chem., 47_, 4153 (1969). 73. J.M. Taylor and R.C. Thompson, Can. J. Chem., 49, 511 (1971). 74. K. Mailer, Ph.D. Thesis, University of British Columbia, 1970. 75. CM. Harbort and D.H. O'Brien, J. Organometal. Chem., 24, 327 (1970). 76. H.W. Roesky and H. Wiezer, Ber., 104, 2258 (1971). 77. M. Schmeisser, P. Sartori and B. Lippsmeier, Ber., 103, 868 (1970). 78. H. Schmidbauer, L. Sechser and M. Schmidt, J. Organometal. Chem., 15, 77 (1968). 79. H.H. Anderson, Inorg. Chem., 3_> 1 0 8 (1964). 80. I. Lindquist, "Inorganic Adduct Molecules of Oxo-compounds", Academic Press, New York, 1963. 81. S.P. So, Mol. Phys., 23, 1147 (1972). 82. M.G. Miles, G. Doyle, R.P. Cooney and R.S. Tobias, Spect. Acta, 25A, 1515 (1969). 83. H. Burger, K. Burczyk and A. Blaschette, Mh. Chem., 101, 102 (1970). 84. M. Wechsberg, P.A. Bulliner, F.O. Sladky, R. Mews, and N. Bartlett, Inorg. Chem., 11, 3063 (1972). 85. C. Stepien, B.Sc. Thesis, University of British Columbia, 1973. 86. J.R. Dalziel and F. Aubke, submitted to Inorg. Chem., 87. R.J. Gillespie and E.A. Robinson, Can. J. Chem., 40, 644 (1962). 88. N.N. Greenwood, Chem. Brit., _3> 5 6 (1967). 89. T.C. Gibb and N.N. Greenwood, J. Chem. Soc. A, 43 (1966). 90. N.N. Greenwood and J.N.R. Ruddick, J. Chem. Soc. A, 1679 (1967). 91. R.V. Parish and R.H. Piatt, Inorg. Chim. Acta, 4_, 65 (1970). - 168 -92. R.V. Parish and C.E. Johnson, Chem. Phys. Lett., 6_, 239 (1970). 93. B.W. Fitzsimmons, N.J. Seeley and A.W. Smith, J. Chem. Soc. A, 143 (1969). 94. R.W. Taft, in M.S. Newman (editor), "Steric Effects in Organic Chemistry", John Wiley and Sons, New York, 1956, p. 556. 95. R.H. Herber and H.A. Stockier, "Applications of the Mossbauer Effect in Chemistry and Solid State Physics", Int. At. Energy Agency, Vienna, 1966, p. 110. 96. R.C. Poller, J.N.R. Ruddick, B. Taylor, and D.L.B. Toley, J. Organometal. Chem., 24, 341 (1970). 97. B.F.E. Ford, B.V. Liengme and J.R. Sams, J. Organometal. Chem., 19, 53 (1969). 98. C. Poder and J.R. Sams, J. Organometal. Chem., 19, 67 (1969). 99. R.F. Dalton, Ph.D. Thesis, University of Manchester, 1969, quoted in Ref. 9. 100. J. Barr, R.J. Gillespie and R.C. Thompson, Inorg. Chem., 2> H49 (1964). 101. R.C. Thompson, J. Barr, R.J. Gillespie, J.B. Milne and R.A. Rothenbury, Inorg. Chem., 4_, 1641 (1965). 102. R.J. Gillespie, J.B. Milne and R.C. Thompson, Inorg. Chem., 5_, 468 (1966). 103. R.J. Gillespie, R. Kapoor and E.A. Robinson, Can. J. Chem., 44, 1197 (1966). 104. A.I. Vogel, "Textbook of Quantitative Inorganic Analysis", 3rd. Edn., Longmans, London, 1961, p. 972. 105. J.E. Lind, J.J. Zwolenik and R.M. Fouss, J. Amer. Chem. Soc, 81, 1557 (1959). - 169 -106. J.A. Ciruna and E.A. Robinson, Can. J. Chem., 46, 1715 (1968). 107. G.H. Cady and J.M. Shreeve, Inorg. Synth., _7, 124 (1963). 108. J.E. Roberts and G.H. Cady, J. Amer. Chem. Soc, 82, 352 (1960). 109. G. Brauer (editor), "Handbook of Preparative Inorganic Chemistry", 2nd. Edn., Academic Press, New York, 1963, p. 301. 110. H.A. Carter, A.M. Qureshi and F. Aubke, Chem. Comm., 1461 (1968). 111. Ref. 109, p. 511. 112. Ref. 109, p. 731. 113. H. Rheinboldt and R. Wasserfuhr, Ber., 60, 732 (1927). 114. H.C. Clark and R.G. Goel, Inorg. Chem. , 4., 1428 (1965). 115. H.C. Clark and R.G. Goel, J. Organometal. Chem., ]_, 263 (1967). 116. C.C. Addison, W.B. Simpson and A. Walker, J. Chem. Soc, 2360 (1964). 117. K. Yasuda and R. Okawara, J. Organometal. Chem., _3, 76 (1965). 118. H.C. Clark and R.J. O'Brien, Inorg. Chem., 2y 740 (1963). 119. R. Okawara and E.G. Rochow, J. Amer. Chem. Soc, 82, 3285 (1960). 120. V. Peruzzo, G. Plazzogna and G. Tagliavini, J. Organometal. Chem., 18, 89 (1969). 121. A.D. Cohen and C.R. Dillard, J. Organometal. Chem., 25, 421 (1970). 122. D. Seyferth and F.G.A. Stone, J. Amer. Chem. Soc. , 79, 515 (1957). 123. R. Sasin and G.S. Sasin, J. Org. Chem., 20, 770 (1955). 124. G.S. Sasin, A.L. Borror.and R. Sasin, J. Org. Chem., 23, 1366 (1958). 125. D. Seyferth, J. Amer. Chem. Soc, 79., 2133 (1957). 126. V. Bade and F. Huber, J. Organometal. Chem., 24, 387 (1970). 127. A. Henderson and A.K. Holliday, J. Organometal. Chem., 4-, 377 (1965). 128. CS. Wang and J.M. Shreeve, Chem. Comm., 151 (1970). 129. J.C Lockhart, "Redistribution Reactions", Academic Press, New York, 1970. - 170 -130. J.J. Burke and P.C. Lauterbaur, J. Amer. Chem. Soc, 82, 326 (1961). 131. K.A. Kozeschkow, Ber., 62, 996 (1929). 132. K.A. Kozeschkow, Ber., 66, 1661 (1933). 133. B.L. Chamberland and A.G. McDiarmid, J. Chem. Soc, 445 (1961). 134. E.G. Rochow, D. Seyferth and A.C. Smith, J. Amer. Chem. Soc, 75_, 3099 (1953). 135. F.H. Allen, J.A. Lerbscher and J. Trotter, J. Chem. Soc, A, 2507 (1971). 136. L. Pauling, "The Nature of the Chemical Bond", 3rd. Ed., Cornell University Press, Ithica, N.Y., 1960. 137. H.A. Carter, A.M. Qureshi, J.R. Sams, and F. Aubke, Can. J. Chem., 48, 2853 (1970). 138. M. Vucelic, Croat. Chim. Acta, 40, 255 (1968). 139. E.A. Robinson, Can. J. Chem., 39, 247 (1961). 140. B.F.E. Ford, J.R. Sams, R.G. Goel, and D.R. Ridley, J. Inorg. Nucl. Chem., 3_3, 23 (1971). 141. R.H. Herber and G.I. Parisi, Inorg. Chem., _5, 769 (1966). 142. L.E. Levchuk, J.R. Sams and F. Aubke, Inorg. Chem., 11, 43 (1972). 143. T.H. Tan, J.R. Dalziel, P.A. Yeats, J.R. Sams, R.C. Thompson, and F. Aubke, Can. J. Chem., 50, 1843 (1972). 144. D.W.J. Cruickshank and B.C. Webster, in G. Nickless (editor), "Inorganic Sulfur Chemistry", Elsevier, Amsterdam, 1968, p. 7. 145. R.J. Gillespie and E.A. Robinson, Can. J. Chem., 41, 2074 (1963). 146. F.B. Dudley and G.H. Cady, J. Amer. Chem. Soc, 85, 3375 (1963). 147. P.A. Yeats, B.L. Poh, B.F.E. Ford, J.R. Sams, and F. Aubke, J. Chem. Soc, A, 2188 (1970). - 171 -148. P.A. Yeats, B.Sc. Thesis, University of British Columbia, 1969. 149. W.P. Gilbreath and G.H. Cady, Inorg. Chem., 2_, 496 (1963). 150. Ref. 104, p. 503. 151. R.H. Herber, H.A. Stockier and W.T. Reichle, J. Chem. Phys., 42, 2447 (1965). 152. D.A. Long, T.V. Spencer, D.N. Waters, and L.A. Woodward, Proc. Roy. Soc, A, 240, 499 (1957). 153. J.R. Dalziel, R.D. Klett, P.A. Yeats, and F. Aubke, submitted to Can. J. Chem.. 154. K.O. Christie and W. Sawodny, Inorg. Chem., 8, 212 (1969). 155. E.L. Meutterties, J. Amer. Chem. Soc, 82, 1082 (1960). 156. J.E. Huheey and J.C. Watts, Inorg. Chem., 10, 1553 (1971). 157. E.V. Bryuchova, G.K. Semin, G.I. Goldanskii, and V.V. Khrapov, Chem. Comm., 491 (1968). 158. B.F.E. Ford and J.R. Sams, J. Organometal. Chem., 31, 47 (1971). 159. N.W.G. Debye, D.E. Fenton, S.E. Ulrich, and J.J. Zuckerman, J. Organometal. Chem., 28, 339 (1971). 160. E. Constantinides, Proc. Chem. Soc, 290 (1964). 161. A. Yingst, Chem. Comm., 480 (1965). 162. F. Gallais, D. Voigt, and J.F. Labarre, Compt. Rend., 260, 128 (1965) 163. A. Engelbrecht, and B.M. Rode, Mh. Chem., 103, 1315 (1972). 164. J.R. Sams, in C.A. McDowell (editor), "Magnetic Resonance", University Park Press, Baltimore, 1972, p. 85. 165. R.H. Herber and S. Chandra, J. Chem. Phys., 53, 3330 (1970). 166. R.V. Parish and C.E. Johnson, J. Chem. Soc., A, 1906 (1971). 167. N.E. Erickson, Chem. Comm., 1349 (1970). - 172 -168. B.W. Fitzsimmons, J. Chem. Soc, A, 3235 (1970). 169. H.A. Carter, S.P.L. Jones and F. Aubke, Inorg. Chem., 2485 (1970). 170. R.J. Gillespie, Acc. Chem. Res., JL, 202 (1968). 171. H. Martin and T. Jacobsen, Angew. Chem., 67, 524 (1955). 172. H. Martin, Angew. Chem., 70, 97 (1958). 173. R.H. Miller, D.L. Bernitt and I.C. Hisatsune, Spect. Chim. Acta, 23A, 223 (1967). 174. R.J. Gillespie and M.J. Morton, Quart. Rev., 25, 553 (1971). 175. C.A. Claussen and M.L. Good, Inorg. Chem., 9_, 817 (1970). 176. H. Cheng and R.H. Herber, Inorg. Chem., 9_, 168° (1970). 177. K.O. Christie, C.J. Shack, D. Pilipovich, and W. Sawodny, Inorg. Chem., 8, 2489 (1969). 178. D.W.A. Sharpe and J. Thorley, J. Chem. Soc, 3557 (1963). 179. R.J. Gillespie and M.J. Morton, Inorg. Chem., 11, 586 (1972). 180. C. Campbell, J.P.M. Jones, J.J. Turner, Chem. Comm., 888 (1968). 181. H.H. Claassen and E.H. Appelman, Inorg. Chem., 9_, 622 (1970). 182. C. Rocchiccioli, Compt. Rend., 249, 236 (1959). 183. H.A. Carter and F. Aubke, Can. J. Chem., 48, 3456 (1970). 184. T. Surles, L.A. Quarterman and H.H. Hyman, J. Inorg. Nucl. Chem., 35, 668 (1973). 185. H.H. Claassen, B. Weinstock and J.G. Malm, J. Chem. Phys., 28, 285 (1958). 186. A.A. Woolfe, J. Chem. Soc, 4113 (1954). 187. F.O. Sladky, P.A. Bulliner and N. Bartlett, J. Chem. Soc, A, 2179 (1969). 188. A.M. Qureshi and F. Aubke, Can. J. Chem., 48, 3117 (1970). - 173 -189. V.J. Weidlein and K. Dehnicke, Zelt. Anorg. Chem., 348, 278 (1966). 190. H.A. Carter, W.M. Johnson and F. Aubke, Can. J. Chem., 47, 4619 (1969). 191. L.C. Hoskins and R.C. Lord, J. Chem. Phys., 46_, 2402 (1967). 192. G.M. Begun, W.H. Fletcher and D.F. Smith, J. Chem. Phys., 42. 2236 (1965). 193. A.J. Edwards, G.R. Jones and R.J.C. S i l l s , Chem. Comm., 1527 (1968). 194. A. Storr, P.A. Yeats and F. Aubke, Can. J. Chem., 50, 452 (1972). Publications: 1. P. A. Yeats, 3. F. E. Ford, J. R. Sams, and P. Aublce, "Synthesis and Mossbauer Spectra of Dimethyltin Derivati of Konobasic Sulfonic Acids", Chem,Commun, 7 9 1 , (19o9). 2. W. Reed, D. V<. Secret, R. C. Thompson, and P. A. Yeats, " i i l e c t r i c a l Conductivity Studies of Fluorosulf u r i c Acid and Tetramethylammoniuiu Fluorosulf ate i n Nitrobenzene", ganad J Chem, 4J_, 4275, (1969). 3. P. A. Yeats, J. R. Sams, and F. Aubke, " T i n l l 9 Mossbauer Spectra of Tin(IV) Chloride Complexes with Oxygen Donor Ligands", Inorg Chem, 9,. 740, (1970). 4 . P. A. Yeats, 3. L. Poh, 3. F. E. Ford, J. R. Sams, and F. Aublce, "Synthesis and Mossbauer and Vib r a t i o n a l Spectra of Tin(IV) Fluorosulfate and Dichlorotin(IV) Fluorosulfate", J Chem Soc, A, 2168, ( 1 9 7 0 ) . 5« P. A. Yeats, J. R. Sams, and F. Aublce, "The Synthesis and Mossbauer and V i b r a t i o n a l Spectra of some Triraethvltin(IV) Sulfonates", Inorg Chem, 10, 1877, (1971). 6 . A. Storr, P. A. Yeats, and F. Aublce, MGallium(III) T r i s -f l u o r o s u l f a t e " , Canad J Chem, 5 0 , 452, (1972). 7. T. H. Tan, J. R. D a l z i e l , P. A. Yeats, J. R. Sams, R. C. Thompson, and F. Aublce, "The Synthesis and Structural Studies of some Dlalicyltin( IV) Salts of Strong Monobasic Acids", Canad J Chem, 5.0, 1843, (1972). 8. P. A. Yeats, J. R. Sams, and F. Aublce, "Synthetic and St r u c t u r a l Studies of some Methyltin(IV) Chloro-Sulfinato Compounds", Inorg Chem, 11, 2634, (1972). 9 . P. A. Yeats, W. W. Wilson, and F. Aubke, "Dibromolodine(III) and Dichloroiodine(III) Fluorosulfate", Inorg Mucl Chem Lett, 10. P. A. Yeats, J. R. Sams, and F. Aubke, "The Hexa kisfluoro-s u l f a t o Stannate(IV) Ion", Inorg Che in, V T ^ S ^ L g - (\<i"T3) 


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