Ill-l-i-I. i Ii'i N . O“ SPECTROSCOPIC STUDIES OF METAL COMPLEXES IN SOLUTION Dissertation for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY AZAM RAHIMI 1977 f 52.1.: 4' . Chm-rfi'mr‘fi‘i’f LID v n ' 5.: J.“ a 2 a Y [VJ/J ‘ c A- «Pf-A" ? " ' ~. L V111C>z;11 \J b‘rvLe mvcrsfiy ~""“ This is to certify that the thesis entitled SPECTROSCOPIC STUDIES OF METAL CQVIPLEXE‘S 1N SOLUTIW presented by AzamRalfimi has been accepted towards fulfillment of the requirements for Ph.D. degree in Chemistry ”flatwaw Major professor Date JULY 25. 1977 0-7639 ABSTRACT SPECTROSCOPIC STUDIES OF METAL COMPLEXES IN SOLUTION By Azam Rahimi This thesis reports some exploratory studies on the use of scandium-H5 and silver-109 nmr probes of the im- mediate chemical environment of scandium(III) and silver(I) ions in solutions. Scandium-HS chemical shifts for scandium salt solutions in tetrahydrofuran were found to be strongly dependentcni the nature of the counterion. This influence of the anion was not observed for scandium salts in aqueous solutions in the same range of concentration. It is concluded that in the THF solution Sc3+ salts exist as ion pairs or higher ionic aggregates while in water Sc3+ or Sc(OH)2+ ion is completely solvated by the solvent molecules. Studies of “58c chemical shifts in water-THF mixtures confirmed the above conclusions. Attempts were made to study Sc(III) complexes with crown and cryptands by “580 nmr. The results were largely inconclusive due to very low solubilities of the Sc(III) complexes and the relatively low sensitivity of "530 signal. Azam Rahimi Silver-109 nmr studies were first carried out on aqueous solutions of AgNO and AgClOu. The 109Ag chemical 3 shifts moved upfield with increasing concentration of the two salts. The shifts were somewhat anion-dependent but extrapolated to the same frequency at infinite dilution. This frequency is indicative of 109Ag resonance of a com- pletely hydrated Ag+ ion. Similar studies were carried out in several nonaqueous solvents and, in particular, in acetonitrile. Again an up- field shift with increasing salt concentrations was observed. The 109Ag chemical shift for solvated Ag+ ion in acetonitrile is considerably further downfield than in water, indicative of stronger interaction of Ag+ ion with acetonitrile. The results from 109Ag nmr studies in mixtures of these two solvents supported the above conclusion. Silver-109 nmr was also used to study complexation re- action of Ag+ ion with benzene, pyridine and their deriva- tives, in several nonaqueous solvents. Measurements were carried out in acetone, methanol, tetrahydrofuran, propylene carbonate, pyridine and acetonitrile solutions. Solvent dependence of these complexation reactions was observed. The effect of substituents on the complexation reaction and the effect of temperature on complexation of pyridine with Ag+ ion was also studied. It was found that the Ag+- pyridine complex is more stable at lower temperature. In view of the low solubility of the scandium salts and extremely weak sensitivity of the 109Ag resonance, only Azam Rahimi qualitative information could be obtained with the instru- ments used in this study. SPECTROSCOPIC STUDIES OF METAL COMPLEXES IN SOLUTION By Azam Rahimi A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1977 To my parents as a symbol of my endless appreciation 11 ACKNOWLEDGMENTS The author wishes to express her sincere gratitude to Professor Alexander I. Popov for his guidance, encourage- ment and friendship throughout this study and Professor Gordon A. Melson for his sincere friendship and his in- troduction to research. Professor Andrew Timnick is acknowledged for his help— ful suggestions and discussions as second reader. The author acknowledges the financial support of Iran, through the Ministry of Science and Higher Education. Gratutude is extended to Mr. Frank Bennis and Mr. A. Wayne Burkhardt, without their cooperation, the NMR investi- gations would have been much more difficult, to Janet Berger for her expenditure of time in proofreading this thesis, and to members of the research group for their friendship and helpful discussions. Special thanks to my friends in the U.S.A., in particu- lar Iranian and Greek for many good times. Deep appreciation to my family; without their love this work would not have been possible. To them I dedi- cate this Thesis. 111 TABLE OF CONTENTS Chapter LIST OF TABLES. . . . . . . . . . . . . . LIST OF FIGURES . . . . . . . . . . . Part I I HISTORICAL O O O O O O O O O O O 0 0 II III CHEMISTRY OF SCANDIUM. . . . . . . NUCLEAR MAGNETIC RESONANCE STUDIES . COMPLEXATION WITH CROWN AND CRYPTAND MACROCYCLIC LIGANDS. . . . . . . . . CONCLUSION . . . . . . . . . . . . . EXPERIMENTAL PART. . . . . . . . REAGENTS . . . . . . . . . .’. . . . SALTS. . . . . . . . . . . . . . . . SOLVENTS . . . . . . . . . . . . LIGANDS. . . . . . . . . . . . . . SAMPLE PREPARATION . . . . . . . . . INSTRUMENTAL MEASUREMENT . . . . . . SCANDIUM-us NMR STUDIES OF SOLVATION COMPLEXATION OF SILVER(I) ION IN SOLUTION O O O C O O O C . O O O O 0 INTRODUCTION . . . . . . . . . . . . A. Scandium-A5 NMR Studies of Solu- tions of Scandium Salts in Tetra- hydrofuran and in a Mixture of water and Tetrahydrofuran. . . B. Scandiumsus NMR Studies on Com- plexation of Scandium with Crown Ethers and Cryptands in Solution . CONCLUSION . . . . . . . . . . . . . iv Page vii 10 l2 l3 1“ 1h 1“ 15 15 17 19 20 20 29 3H Chapter II III Part 2 HISTORICAL O O O O O O O O O O O O O 0 O O CHEMISTRY OF SILVER. . . . . . . NUCLEAR MAGNETIC RESONANCE STUDIES . CONCLUSIONS 0 O O O O O O O O O O O O O O 0 EXPERIMENTAL PART. . . . . . . . . . . . . SALTS. . . . . . . . . . . . . . . . SOLVENTS . . . . . . . . . . . . . . . . . LIGANDS. . . . . . . . . . . . . . . . . . INSTRUMENTAL MEASUREMENT . . . . . . . . . NMR STUDIES OF SOLVATION AND COMPLEXATION OF SILVER(I) ION IN SOLUTION . . . . . . INTRODUCTION 0 O O O O O O I O O O O O O A. NMR Studies of Silver Salts in salutions O O O O O O O O O I O O l. Silver-109 nmr Study of Solutions of Silver Salts in Water,Aceto- nitrile and Binary Mixtures of These Solvents . . . . . . . . . 2. Proton nmr Studies of Solutions of AgNO in water,acetonitrile and Binary Mixtures of These SOlvents O O O O O O O O O O O O O 3. Carbon-13 nmr Studies of Solu- tions of AgNO in Acetonitrile and Acetonitri e-water Mixtures. . B. NMR Studies of Ag(I) Complexes in SOlution O O O O O O O O O O O 0 O l. A 109Ag nmr Stud on Complexation Reaction of Ag(I Ion with Benzene and Taluene O I O O O O O O I O O O 2. A 109Ag nmr Study on Complexation Reaction of Ag+ Ion with Pyridine. V Page 36 37 AZ ”5 H6 #7 "7 NB MB 53 5H 5H 5H 66 76 82 82 96 Chapter CONCLUSIONS. REFERENCES. Complexation Reaction of Silver(I) Ion with Substituted Pyridine Derivatives Carbon-13 nmr Studies of Complexation of Ag+ Ion with Benzene and Pyridine vi Page . 110 115 . 118 121 Table II III IV VI VII VIII IX LIST OF TABLES Scandium-H5 NMR Data for Some Scandium Salts in Tetrahydrofuran Solution . . . . . . . . . . Scandium-N5 NMR Data for Some Scandium Salts in Aqueous Solution . . . . . . . . . . . . Scandium Data for ScCl3 in Water— Tetrahydrofuran Mixtures . . . Scandium-HS NMR Data for Sc(ClOu)3 - 6H2O/Cryptand Mixtures in Aqueous SOlution O O O C O O O O I O O O Scandium-U5 NMR Data for Sc(ClOu)3 ' 6H20/Crown Mixture in Aqueous Solution . . . . . . . . . . 1°9Ag NMR Data for AgClOu and AgNO3 in Water . . . . . . . . . Silver-109 NMR Data for AgNO3 in Acetonitrile . . . . . . . . . . 1°9Ag NMR Data for 1 M AgClOu in Different Solvents . . . . . . . 1°9Ag NMR Data for AgNO3 2 M in Water-Acetonitrile Solutions. Proton-1 NMR Data for Solution of AgNO3 in Acetonitrile. . . . . vii Page 21 23 26 30 33 55 59 62 6A 67 Table XI XII XIII XIV XVII XVIII XIX Proton-l NMR Data for Aqueous Solu- tion of AgNO3. . . . . . . . . . . . . Proton-l NMR Data in Acetonitrile- Water Mixtures . . . . . . . . . . Proton-l NMR Data of Solutions of AgNO in Acetonitrile-Water Mixtures 3 Carbon-l3 Data of Solution of AgNO3 in Acetonitrile. . . . . . . . . . . . Carbon-13 Data of Solution of AgNO3 in Acetonitrile-water Mixtures . Silver-109 Chemical Shiftuas a Function of Benzene/AgClO2 Molar Ratio in Various Solvents. . . . . . . Silver-109 Chemical Shifts as a Function of Toluene/Ag0102 Molar Ratio in Different Solvents. . . . . . . . . Silver-109 Chemical Shifts as a Func- tion of Temperature. A) For Regular Reference Solution B) for Reference Solution in Dewar Type Tub . . . . . . Silver-109 Chemical Shifts as a Func- tion of Pyridine/AgClOu Molar Ratio in Various Solvents and Different Temperatures . . . . . . . . . . . . . viii Page 70 72 73 77 80 83 86 97 98 Table XX XXI XXII XXIII Page Silver-109 Chemical Shifts as a Function of 2-Cyanopyridine/AgClOu Molar Ratio in Different Solvents. . . . . . lll Silver-109 Chemical Shifts of AgClOu with Various Ligands in Tetrahydro- furan and Dimethylsulfoxide. . . . . . . . . 11h Carbon-l3 Chemical Shifts as a Function of Ag+/Benzene Molar Ratio in Different Solvents. . . . . . . . . 116 Carbon-13 Chemical Shift as a Function of Ag+/Pyridine Molar Ratio in Various Solvents. . . . . . . . . . 117 ix Figure LIST OF FIGURES Page Structure of macrocyclic ligands cryptands and crowns. . . . . . . . . . . 11 Variation of the chemical shift of the scandium—A5 resonance as a function of solvent composition for binary solvent water and tetrahydrofuran . . . . . . . . 28 Reference solution in dewar-type tube. . . . . . . . . . . . . . . . . . . 50 Silver-109 chemical shifts of silver salts in water. . . . . . . . . . . . . . 57 Silver—109 chemical shifts of silver nitrate in acetonitrile . . . . . . . . . 60 Variation of the chemical shifts of the silver-109 resonance as a function of solvent composition for binary solvent water and acetonitrile . . 65 Proton-l chemical Shifts (CH CN) as a 3 function of mole fraction of AgNO3 in acetonitrile solutions of AgNO3 . . . . . 68 Proton-l chemical Shifts (H20) as a function of AgNO3 concentration in water . . . . . . . . . . . . . . . . . . 71 Figure Page 9 Proton-l chemical shifts (CH CN) as 3 a function of water mole fraction in water-acetonitrile mixture in absence (a) and presence (b) of the salt. . . . . 7U 10 Proton-l chemical shifts (H20) as a function of acetonitrile mole fraction in acetonitrile-water mixtures in the absence (a) and presence (b) of salt. . . 75 ll Carbon-l3 chemical shifts as a function of silver nitrate concentration . . . . . . . . . . . . . . 79 12 Carbon-l3 chemical shifts as a function of water mole fraction in mixtures of water-acetonitrile. . . . . . 81 13 Chemical shifts of 109Ag as a function of benzene/AgClOu molar ratio in various solvents. . . . . . . . . . . . . . . . . 88 1h Chemical shifts of 109Ag as a function of ligand/AgClOu molar ratio in propylene carbonate. (Ligand is either benzene or toluene.). . . . . . . . . . . 89 15 Chemical shifts of 109Ag as a function of ligand/AgClOu molar ratio in acetone (Ligand is either benzene or toluene.). . 90 xi Figure 16 17 18 19 2O 21 22 Page Silver-109 chemical shifts as a function of ligand/AgC10u molar ratio in methanol. (Ligand is either benzene or toluene.) . . . . . . . . . . . . . . . . 91 Silver-109 chemical Shifts as a function of ligand/AgClOu molar ratio in tetrahydrofuran. (Ligand is either benzene or toluene.). . . . . . . . . . . 92 Silver-109 chemical shifts as a function of ligand/AgClOu molar ratio in dimethylformamide and in dimethyl sulfoxide. (Ligand is either benzene or toluene) . . 93 Silver-109 chemical shifts as a function of ligand/AgClOu molar ratio in pyridine. (Ligand is either benzene or toluene.). . . . . . . . . . . . . . . 9" Chemical shifts of 109Ag as a function of pyridine/AgClOu molar ratio in various solvents. . . . . . . . . . . . . 101 Silver-109 chemical shifts as a function of temperature A) for reference solution in dewar type tube, B) for reference solution in regular nmr tube. . 102 Chemical shifts of 109Ag vs pyridine/ AgClOu molar ratio in propylene carbonate at different temperatures 103 xii Figure 23 2A 25 26 27 28 29 Page Silver-109 chemical shifts as a func- tion of pyridine/AgClOu molar ratio in acetone at different temperatures. . . 10h Silver-109 chemical Shifts as a func- tion of pyridine/AgClOu molar ratio in tetrahydrofuran at different temperatures. . . . . . . . . . . . . . . 105 Chemical shifts of 109Ag as a function of pyridine/AgClOLl molar ratio in aceto— nitrile at different temperatures . . . . 106 Silver-109 chemical shift as a func- tion of pyridine/AgClOu molar ratio in dimethylsulfoxide at different temperatures. . . . . . . . . . . . . . . 107 Chemical Shifts of 109Ag v_s_ pyridine/ AgClOu molar ratio in dimethylsul- foxide at different temperatures. . . . . 108 Silver-109 chemical Shift as a func- tion of 2-cyanopyridine/AgClOu molar ratio in various solvents . . . . . . . . 112 Silver-109 nmr resonances of aqueous solution of l E.A8010u doped with l/SO Ii Fe+3 by DA60 and Bruker 180 Instruments . . . . . . . . . . . . . . . 120 xiii AC ACN PC THF DMF DMSO Bz To Py LIST OF ABBREVIATIONS Acetone Acetonitrile Propylene Carbonate Tetrahydrofuran Dimethylformamide Dimethylsulfoxide Benzene Toluene Pyridine Ligand Molarity xiv PART I CHAPTER I HISTORICAL Chemistry of Scandium Scandium was one of the missing elements in the original periodic table of Mendeleev published in 1871. Mendeleev called this element "eka-boran" and estimated its atomic weight to be “5. In 1879, Mendeleev's predictions of this missing element were confirmed by its discovery by Nilson. The element was named scandium, because the minerals containing it were found in Scandinavia. This new rare earth was different from the lanthanons by its weak basicity, atomic weight of less than 131 and by its Spark spectrum. Scandium was obtained in a small quantity from gadolinite and euxenite minerals after separation from lanthanons and fractional crystalliza- tion and precipitation. Since scandium was obtained in a very small quantity, Nilson was not able to study it in detail. For many years not much attention was paid to this element. Later in 1950, when the problem of obtaining this element and its compounds in a good quantity and high purity was overcome, the interest in scandium and its compounds increased rapidly. During recent years dif- ferent compounds of scandium have been synthesized and characterized. Many intermetallic compounds and alloys of this metal with interesting properties are available. Application of different instrumental techniques such as neutron activation analysis, flame and emission 3 spectroscopy, made it possible to detect and determine small traces of this element in rocks, biological materials, etc. Different inorganic and organic compounds of scandium were synthesized and characterized by different methods. Stable complexes of scandium in trivalent states with anionic and neutral ligands were prepared in the solid and solution state. A comprehensive review of all aspects of scandium has been provided by Horovitz, Gschneidner, Melson, Young- blood, and Schoch in a recent monograph (1). Scandium, with electronic configuration of [Ar] 3dlhs2, is a group III element and the first member of Sc, Y, La Ac group. Its ionic radius is ~0.7 A which is smaller than the radii of lanthanides and in its compounds co- ordination numbers exceeding 6 are not known. Scandium chemical behavior is intermediate between that of aluminum and of the lanthanides. The usual oxidation state of this element in its compound is III. Many inorganic salts of scandium have been synthesized. Organo-scandium compounds which contain scandium-carbon bonds are well characterized. Coordination compounds of scandium containing nitrogen-, oxygen-, phosphorus- and sulphur-donor ligands have been of physicochemical measurements on these systems. Physicochemical studies on scandium complexes in solu- tions are particularly rare. This fact might be due to the 3d0 configuration of scandium in its +3 oxidation state, which makes useless some of the techniques, such as electronic spectroscopy and magnetic susceptibility measurements. Nuclear magnetic resonance spectroscopy seems to be a potentially useful technique for such studies. Nuclear Magnetic Resonance Studies During recent years nuclear magnetic resonance (nmr) spectroscopy has been extensively used as a powerful tool for the elucidation of the molecular structure and investi- gation of environmental effects on a given nucleus. In particular the application of Fourier transform technique to the nmr spectroscopy has introduced the advantage of obtaining spectra in a shorter period of time. In the past decade the nmr of metallic nuclei with I > 0 has become an interesting and useful extension of the more common nmr spectroscopy of nuclei such as 1 H. 13C, 31P, and 19F. A lot of work has been done in this area specially on alkali metal nmr (3), but very few reports about other nuclei have been published. Scandium- A5 spectroscopy has Shown the sensitivity of the chemical shift to the variation of the coordination environment of the Sc(III) ion and suggests that this technique may be useful for such studies. Scandium-us nuclide with Spin of 7/2 has 100% natural abundance and a relative sensitivity of 0.3 compared to an equal number of protons at constant field. Very few reports of the application of “58c nmr spectroscopy to the study of the scandium complexes have been published. Proctor and Yu (A) measured the magnetic moment of scandium with a nuclear induction spectrometer. They measured the resonance frequency of “58c in an aqueous solution of Sc(N03)3. By comparing this frequency with that of 23Na from a saturated aqueous solution of NaCl and taking for the scandium spin the value of 7/2, they calculated the magnetic moment of scandium to be, uu58c=u.7fl97i0.008 uN (nuclear magneton). The magnetic moment of scandium was also measured by nuclear magnetic resonance absorption (5). Proton resonance was used for comparison. The value obtained for the magnetic moment was n.799l6 1 0.00012 which agrees quite well with the previous results. Scandium-N5 nucleus has also been used as a secondary standard for measure- ments of the nuclear magnetic moment of other nuclei (6,7). Lutz (8) measured the Larmor frequency of scandium- AS in a dilute solution of scandium chloride in acidified heavy water and calculated the magnetic moment of scandium to be, "“5Sc = “.7A87ll uN. Lutz also investigated the relative chemical shifts of ”58c in solutions of Sc(N03)3, $0013 and Sc(ClOu)3 in heavy water and light water as the solvents. It was found that for acidified solutions of scandium chloride in water, there is no Shift as the concentration of the salt is increased, and the observed linewidth was narrow, 28 to 35 Hz. In the case of acidified Sc(ClOu)3 solution, there was a shift to lower frequency as the concentration of the salt increased, and the linewidth was also narrow, 30 Hz. He also noticed that at low concentration, there is no difference in the scan- dium chemical shift between acidified solution of ScCl3 and Sc(ClOu) The linewidth for Sc(NO3)3 solution was 3. broader than that of the above salts. He observed a dif- ference in the Larmor frequencies of scandium-A5 in solu- tions of light and heavy water, which was related to the difference in shielding of the “58c nucleus by light and heavy water ("solvent isotope effect"). Buslaev gt_al. (9) investigated the scandium complex formation in aqueous solution. This work was done by study of the magnetic resonance of "58c nucleus as a function of the concentration of the salt and the nature of the anion: 01‘, N03, and 010;. It was observed that the linewidth for Sc(NO3)3 is greater than those of ScCl3 and Sc(ClOu)3 and related it to the following order of the complexing ability of the anions: N03 > 01' 1.010; This order is unexpected since Cl' usually forms stronger complex with metal ions than N03. More recently, Buslaev gt El- (10) reported a study on the linewidth and chemical shift of the "53c resonance in the solutions of various scandium salts. They found that the linewidth of "58c resonance in the nitrate solution depends on the concentra— tion of the salt. In concentrated solutions the line was very broad (Av = 2630 Hz for the saturated solution which is 80% by weight of the salt (11)). The line broadening was explained in terms of formation of an outersphere ion pairs and higher aggregate formed in concentrated solution. They did not observe the change in the chemical shift in the concentration range of 2% to saturated solution of Sc(NO3)3. In an aqueous solution of (NHu)3ScF6 in the presence of ammonium fluoride, fine structure in the “58c nmr spectrum was observed. The spectrum consists of seven lines whose relative intensities indicate the interaction of “580 Spin with six equivalent fluorine nuclei in Sch' anion which has an octahedral structure. They also observed a Shift of “58c resonance toward higher field in HCl solu- tion of ScCl3 relative to an aqueous solution of ScCl3. They suggested that in these solutions various aquochloro complexes of scandium are formed, but the exchange is fast relative to the nmr time scale so that only one resonance is observed. In a solution of ScBr3 in HBr, the exchange is slow and two resonances were observed due to aquo— and bromo complexes of scandium. They also ob- served a change in the chemical shift of ”58c resonance with respect to the nature of the counterion. The chemical shift decreased as the anion changed from chlorine to bromine and to iodine. Melson gt_al (12) reported scandium-M5 nmr spectra for several scandium salts in aqueous solution. They observed the concentration and anion dependence for the chemical shift and linewidth of Sc-resonance in the solutions of the scandium chloride, perchlorate, and nitrate. They noticed that at low concentration for chloride, perchlorate and bromide solutions of scandium, the chemical shifts approach the same value. In other words the limiting chemical Shift is independent of the counterion. They suggested that in these solutions with low concentration the species present might be hydrated [Sc(OH)]2+ or [Sc(H20)6]3+ ions. They also observed the concentration and anion dependence of the chemical Shift and linewidth of the Sc- resonance in aqueous solutions of the different scandium salts. In the case of chloride and bromide,the presence of the chloro and bromo species in the solution was sug— gested. Other investigators (13-15) also studied these systems and identified the presence of the hydrated species of ScC12+, Sc01%+, 2+, and ScBr;+ in the aqueous solu- tions. They measured the stability constant of these com- ScBr plexes by various methods such as potentiometry and cation- exchange. Their values do not agree with each other. For perchlorate as anion, the observed changes in the chemical Shift and linewidth of Sc-resonance was related to the formation of [Sc(OH)]2+ ion at low concentration and polymeric hydroxy species at higher concentration (10). Melson gt a1. (12) observed different behaviors for SC(N03)3 and Sc2(80u)3 in aqueous solution. The chemical shifts in these cases did not approach the limiting value at lower concentration which indicates that the species 10 present in the solution include the anions. The line- width increased with increasing concentration of the salts, which was indicative of the formation of a complex (or contact ion pair) of nitrate and sulfate with scandium ion in concentrated solutions. The stability constants for the formation of Sc(III)-nitrate complexes and related thermo- dynamic parameters have been measured by solvent extraction method (16). For sulfate complexes of scandium in aqueous solution the log k, AH° and AS° values were also determined calorimetrically (17). These evidences support the con- clusion of Melson 33 El- (12) obtained from "58c nmr. Complexation with Crown and Cryptands Macrocyclic Ligands The discovery of cyclic polyether (crown) by Pederson (18) and diazapolyoxamacrocyclic (cryptands) by Lehn and his coworkers (19) introduced into different areas of research a new class of compounds with very interesting and unusual binding properties. Figure 1 shows the structure of some of the crowns and cryptands. These macrocycles contain flexible frameworks with different cavity Sizes which enable them to bind a wide variety of cations (20). These compounds form strong complexes with alkali and alkaline earth metal ions (21) and they have been used as models for ion carrier molecules of biological interest. Crown ethers form two dimensional complexes, and cryptands form three—dimensional complexes. The strength of the ll flo/w cm N/‘W Om ONfl Nmo MN NL/OOUN K/OU 0d "211" "221" /’\O/\' @E0 o @E: :I@ UV (L03 "benzo—lS—crown-S" "dibenzo- 18-crown-6" WO/W (1::33 DJ "dicyclohexyl- 18-crown—6" Figure 1. Structure of macrocyclic ligands cryptand and crown. l2 complexes largely depends on the relation of the cavity sizes of the polyethers to the diameter of the ions. Although these macrocyclic polyethers have been used extensively as complexing agents for a variety of metal ions, the studies on complexation of these ligands with Sc(III) ion has Just been started. Olszanski (22) has done some studies on reaction of ScCl3 and Sc(NCS)3 with the benzo-lS-crown—S and dibenzo-l8-crown-6. He found that the stoichiometry of the complex is a function of both the salt and the solvent used in the course of the reaction. A more extensive historical discussion of these macrocyclic ligands and their complexes with metal ions in particular alkali metal ions can be found in the Ph.D. thesis of E. H. Mei, Michigan State University (1976). Conclusion From the above discussion it is evident that “58c NMR spectroscopy may be used as a sensitive probe for the detection and the study of scandium (III) ion solvation and complexation in solution. The first part of this thesis reports an investigation of Sc(III) salt solutions and of complexation of this ion with macrocyclic ligands crowns and cryptands by scandium-N5 nmr spectroscopy. CHAPTER II EXPERIMENTAL PART 13 Experimental Reagents Scandium oxide was purchased from Research Organic/ Inorganic Chemical Corporation and was 99.9% pure. Salts Anhydrous ScCl was prepared as described by Stutz 3 and Melson (23); ScCl3°5H20, Sc(NO °3H20, Sc(ClOu)3° 3)3 6H20, and Sc13°6H20 were prepared from Sc203 according to the described methods (29-28). These compounds were dried over P205 under vacuum at room temperature. The Karl Fischer titration method was used to measure the numbers of water molecules accompanying each salt. For the hydrated ScCl and Sc(N03)3 cases, the number of water molecules is 3 different from that reported earlier (25,26). The dif- ference may be due to differences in drying procedures (reported values for the number of water molecules accom- panying ScCl is 6 and for Sc(N03)3 it is A). 3 Solvents Tetrahydrofuran (THF), (Burdick and Jackson Laboratories, Inc.) was dried over metallic sodium and benzophenone by refluxing. Its water content, measured by the Karl Fischer method, was less than 100 ug/ml. 19 l5 Ligands The cryptands C211 and C221 were obtained from E. M. Laboratories, Inc. and were used as received. Crown ethers dibenzo-l8-crown-6, dicyclohexyl-l8-crown-6 and benzo-lS-crown-S were purchased from Aldrich Chemical Company. Sample Preparation Tetrahydrofuran solution of scandium salts were pre- pared by the following methods: 1. The required amount of anhydrous salt sample was introduced into a volumetric flask and diluted to the mark by tetrahydrofuran or by a mixture of tetrahydrofuran and water in desired ratio. 2. Since the preparation of anhydrous scandium salts was difficult, attempts were made to make tetrahydrofuran solutions of scandium salts by another method. Hydrated scandium salts were dissolved in the dry tetrahydrofuran. The solvent from the solution was refluxed and the condensed solvent was passed through molecular sieves which were in the column of the Soxhlet extraction apparatus. The water concentration during the course of the extraction was monitored by the Karl Fischer method. After a few hours the water concentration was reduced from ~250-900 ug/ml to m20-30 ug/ml. To make solutions with the highest 16 possible concentrations of the salts (which was still very low, 0.02 M for ScCl 0.0017 M for Sc(NO and 0.001“ 3’ 3’3 M for ScBr3,)some of the solvent was evaporated. The IR spectra of the solid dehydrated product (after dehydration process and evaporation of the solvent) did not Show any H20—bands which are present in the spectra of the hydrated scandium salts. The IR spectra of the product in the ScCl3 case was identical to that of the ScCl '3THF 3 adduct, first prepared by Herzog §t_a1. (29). The crystal and molecular structure of this compound was re— cently reported (30). In this compound the scandium ion is hexa-coordinated by three chloride ions and three tetra- hydrofuran molecules. The IR spectra of tetrahydrofuran adducts of other scandium salts have bands in common with those of the ScC13'3THF adduct with some shifts in their positions. Therefore, it is possible that in the dehydra- tion process the H20 molecules around scandium ion have been replaced by tetrahydrofuran molecules. The above dehydration process was used in other solvents besides tetrahydrofuran, such as acetonitrile (CHBCN), dimethyl- formamide (DMF), dimethyl sulfoxide (DMSO) and acetone (CHBCOCH3). In CH3CN hydrolysis took place, and the IR spectrum of the precipitate indicated that it was Sc203. In DMF, DMSO and CH3COCH3 decomposition took place during dehydration process. Therefore, tetrahydrofuran was used as the nonaqueous solvent for this work. Scandium iodide was not used, since the water concentration did not reduce l7 appreciably even after a long dehydration process. The solutions for study of 145Se nmr in mixture of sol- vents were prepared by adding the required volumes of the two solvents to the desired weighted amount of the salt in a volumetric flask. Instrumental Measurement Scandium-95 nmr Continuous wave “58c nmr spectra were obtained for some of the experiments with a highly modified NMRS-MP- 1000 Spectrometer (31,32) operated at 55.1 MHz at a field of 53.3 Kg. The time sharing method of Baker 33 al. (33,34) with frequency sweep was used. Spectra were time averaged for many scans on a Nicolet 1083 computer. Five mm nmr tubes were used. The reference was 3.0 M aqueous Sc(ClOu)3 solution. Increasing chemical shift values (positive chemical shifts) correspond to upfield shift. Magnetic susceptibility correction was not made because of the large chemical shift range. The line- widths are not the natural linewidth since some instru- mental 1ine broadening contributes to them. For some of the experiments, the spectrometer was modified to operate in the Fourier transform mode which saved time (since for continuous wave mode, a lot of scans were necessary to see the Signal which took as long as ~29 h). 18 The “58c nmr studies in mixed solvents were done in both modes and identical data were obtained. Infrared Infrared measurements in the “000-200 cm”1 spectral region were obtained on the Perkin Elmer Model “57 grating spectrophotometer. The mull samples were held between potassium bromide salt plates. NuJol was used for making the mulls. CHAPTER III SCANDIUM—Nfi NMR STUDIES OF SOLVATION AND COMPLEXATION OF SCANDIUM(III) ION IN SOLUTIONS 19 Introduction Scandium-95 nmr Spectroscopy is shown to be a poten- tially useful technique for the study of the nature of the chemical species present in the solution. The work done in this area so far deals with “53c nmr studies in aqueous solutions . This might be due to low solubility of anhydrous scandium salts in nonaqueous solvents which makes it dif- ficult to see uSSc-resonance with these solutions. With this in mind this study was initiated to investigate the salt solutions of Sc(III) salts in nonaquous solvents. In addition since there have been no published reports of any studies involving the “58c nmr studies of crown and cryptand complexes of this ion in solution, it seemed interesting to study complex formation of these complexing agents with Sc(III) ion by “58c nmr Spectroscopy. A: A Scandium-95 NMR Study of Solutions of Scandium Salts in Tetrahydrofuran and in a Mixture of Water and Tetra- hydrofuran The chemical shift and linewidth of the “58c resonances for tetrahydrofuran solutions of scandium salts are recorded in Table I. The chemical Shifts, "5", are referred to an aqueous solution of ScCl3 at infinite dilution. In- creasing values (positive chemical shifts) correspond to 20 21 Table I. ”53c NMR Data for Some Scandium Salts in Tetra- Hydrofuran Solution. Concentration Sc3+ Chemical Shift Line Width Scandium Salt M 6, ppm Hz ScCl3 0.020 -202 660 ScCl3'NH20 0.020 -l9l 800 ScCl3'hH20 0.015 ~190 890 ScClB‘AHzo 0.010 -193 900 ScBr3 0.007 -273 231 ScBrB'AHZO 0.007 -26N 323 Sc(N03)3 0.017 - 10 2h“ Sc(N03)3’3H20 0.017 - 7 216 Sc(ClOu)3°6H20 0.020 - 23 A92 Sc(ClOu)3'5H20 0.010 - 25 A94 22 increasing shielding. For the sake of comparison the data obtained from this study will be discussed in conjunction with those obtained for aqueous solutions by Olzanski gt a1. (22) as listed in Table II. It is obvious from the data in Table I, that for tetra- hydrofuran solutionscfi’scandium salts a large range of chemical shifts is observed. These Shifts can indicate that there is a significant change in the coordination environment of the scandium(III) ion in these solutions com— pared to the aqueous solutions. A study of concentration dependence of the chemical shift was not possible because of (a) low solubility of scandium salts in the tetrahydro- furan and (b) limitation of the instrument sensitivity for observing the signal at lower concentrations. However, in the concentration range in which nmr observations were possible, the chemical shifts were anion dependent and did not ap- proach a limiting value as was observed in aqueous solution of SCC13, ScBr and Sc(ClOu)3 (22). These results indicate 3 that even at such low salt concentrations the concentration of free and solvated Sc(III) ion is very small and the anion is still interacting with the Sc(III) ion. This difference between the two systems might be explained in terms of the dielectric constant of tetrahydrofuran (D = 7.58 at 25°C) with respect to water (D - 78.5h at 25°C). The large Chemical Shifts observed for ScCl3 and ScBr tetrahydro- 3 furan solutions also indicate that there is strong inter- action between the anions and Sc(III) ion and that the 23 Table II. “53c NMR Data for Some Scandium Salts in Aqueous Solutiona. Concentration Sc3+ Chemical S 1ft Line Width Scandium Salt M , ppm Hzc 220 200 178 52 50 AB 56 50 58 181 183 176 1&8 1N6 93 71 139 11“ 85 65 59 143 "7 "5 These data are from Olzanski's thesis and the Sign of the chemical shifts were changed to follow our convention. ScCl ~18. 3 I H O =OOOHNU'IQOOOOOI-‘w O HWQOWOU‘IO I + + 0 LA) ScBr me000 + -+ +-+ + O u: + Sc(ClOu) 3 WO‘OU'IOU'IO OOHHNNwOOOOHNwOOOOI-‘HNNW OMQNNOQNOQHW-fiWOI—‘NI—‘mmmtm + -r 4- + I-‘MWU'ICD O I-' O O 2“ Table II. Continued Concentration Sc3+ Chemical Shift Line width Scandium Salt M 6, ppmb Hzc SC(N03)3 2.0 + 0.1 1655 1.0 - 0.9 100“ 0.6 - 1.6 739 0-3 - “.3 656 0.1 - 4.3 “56 0.03 - 5.0 uuo Sc2(sou)3 2.0 - 6.7 “Sh 1.2 - 7.3 272 0.6 - 8.6 220 0.2 - 9.1 197 0.06 - 9.3 190 0.02 - 9.“ 181 ScI3 3.0 + 5.7 1185 2.5 + 1.8 961: 2.0 - 1.“ 751 1.5 - 2.5 571 1.0 - ".2 60“ 0.6 - “.3 50“ 0.3 - 3.3 377 0.1 - 2.8 295 8.Data obtained at ambient temperature. bRelative to aqueous ScCl3 solution at infinite dilution. Error limits :1 ppm. cNot corrected for instrumental broadening. 25 species present in the solution are contact ion pairs rather than solvent separated ion pairs. The linewidths in tetrahydrofuran solutions are much broader at equal concentrations (Table I) than those found in dilute aqueous solutions (Table II), indicating that the environment of Sc(III) ion is not symmetrical which may be due to contact ion pair formation. In dilute aqueous solutions, the species present is suggested to be predominantly [Sc(H20)6]3+ ion or the hydrated [Sc(OH)]2+ (35.36). Solutions of hydrated scandium salts in tetrahydrofuran were also studied and the results indicate that there is a change in the chemical shift with respect to the anhydrous salt solutions. This study was carried out to understand the role of water in these solutions and the competition between tetrahydrofuran and water for the Sc(III) ion. The ”58c nmr spectra were obtained on a series of the solutions of ScC13 in a mixture of tetrahydrofuran with water. The variation of “530 resonances as a function of solvent composition is shown in Table III and Figure 2. From these data a smooth transition of the Sc—resonance from its value in one pure solvent to the other is observed. The rate of this transition is dependent on the relative solvation ability of the two solvents. The isosolvation point has been defined as a point in which the chemical shift lies midway between the values for pure solvents and where the two solvents compete equally 26 Table III. “58c nmr Data for ScCl a in Water-Tetrahydro- 3 furan Solutions. Mole Fraction Chemical Shift H20 5, ppmb 0.000 -20H 0.001 -200 0.00” -186 0.008 -178 0.011 -177 0.020 —l70 0.039 -150 0.13“ -119 0.206 - 88 0.31u ‘ — 52 0.529 - 12 0.818 - 3 0.976 0 0.988 0 1.000 0 a[Sc3+] = 0.010M bz2 ppm. 27 for the cation in question. Frankel, £3 31. (37) and Bloor and Kidd (38) have examined preferential solvation of 003+ and 23Na respectively in solvent mixtures. Greenberg (39) has studied 23Na nmr in mixtures of several solvents and has used the isosolvation point as a measure of the pref- erential solvation of the solvents in question. In our case, a change in the concentration of the HZO-THF mixture affects not only the solvation shell of the Sc(III) ion but also the ion pair dissociation. There- fore the isosolvation criterion cannot be used in this case. However, from the curvature of the plot of chemical shifts vs mole fraction of tetrahydrofuran (Figure 2) it is obvious that the change in the chemical shift is large, even in solvent mixtures with low mole fraction of H20. These results indicate that the interaction of H20 with So (III) ion is strong and Sc(III) ion is preferentially sol— vated by H 0 which is in accord with the donor number of 2 the two solvents: DNH20 > DNTHF' Donor number which can represent the donor ability of a solvent has been defined empirically by Gutmann (“0) as the enthalpy of complex formation between the given solvent and antimony pentachloride in 1,2-dichloroethane solution. It has been shown by Gut- mann that the donor number is very useful in predicting the trends of complexation reaction in nonaqueous solvents. A comparison of the shifts for the anhydrous and hydrated 28 200 160 40 Figure 2. Variation of the chemical shift of the scandium-NS resonance as a function of solvent composition for binary solvent water and tetrahydrofuran. 29 salts (Table I), leads to the following observations: the chemical shift for anhydrous ScBr3 (0.007g) in tetrahydro- furan h3-273 ppm, for ScBr3, “H20 (0.007!) it is -26H ppm, for anhydrous Sc(N03)3 (0.017M) it is -10 ppm, and for Sc(N03)3, 3H20 (0.017!) it is -7 ppm. The apparent shift of the hydrated salt toward the chemical shift of Sc(III) ion in water indicates the partial solvation of Sc(III) ion by water in these solutions. In all these experiments only one scandium resonance is observed because the rate of exchange of tetrahydro- furan with water in the coordination sphere of Sc(III) ion is very fast with respect to nmr time scale. Therefore, the observed resonance signal corresponds to the average of all the species present in the solution. B: Scandium-“5 NMR Studies on Complexation Reactions of Scandium With Crown Ethers and Cryptands in Solution Scandium—H5 nmr spectra were obtained for a series of solutions of the Sc(ClOu)3-6H20 in water at varying mole ratios of ligand to metal. Different crown ethers and cryptands were used. Cryptands C211 and 0221 (Figure 1) were used, as ligands, because of their cavity radii (0221, 0.8 X; 0221, 1.15 8). Table IV records the chemical shifts and line- widths of scandium resonance as a function of ligand to metal mole ratio. As the ligand is added to the solution 30 Table IV. “53c NMR Data for Sc(Clou)3-6H20/cryptand Mixtures in Aqueous Solution Chemical Line Conc. Cryptand Conc. Sc(III) Shift ‘Width M M 6, ppm Hz Cryptand C211 0.000 0.03 0 120 0.003 0.03 0 160 0.009 0.03 —6 #88 0.018 0.03 -22 >2000 Cryptand C221 0.000 0.05 0 150 0.005 0.05 0 260 0.015 0.05 -10 583 31 the chemical shifts increase as well as the linewidths, which can serve as an indication for complexation of the Sc(III) ion with cryptands. Formation of these kinds of com- plexes distorts the spherically symmetrical electric field around the scandium ion and results in a very appreciable broadening of the scandium resonance. This line broadening is due to the interaction of the quadrupole moment of this nucleus with the fluctuating electric fields around the nucleus. In the case of C211, at mole ratios above 0.6 the signal becomes so broad that the line disappears. In all these cases only one resonance is observed because of the rapid exchange between the coordinated and the free Sc(III) ion in the solution. The “58c resonance observed is the av- erage for Sc(III) ions in all different sites. For cryptand 0221 at a mole ratio above 0.3, a white precipitate was formed which precluded the study of the complexation at higher mole ratios. This precipitate was found to be the perchlorate of the protonated ligand. This conclusion was based on elemental analysis and the IR spectrum of the solid which remained after the solvent was evaporated and the residue was dried over P205 under vacuum. Results from elemental analysis are as follows: Calculated for 0221' 2HClOu: C, 36.02; H, 6.37; N, 5.25; CI, 13.32. Found: C, 35.66”; H, 6.12; N, 5.17; Cl, 12.97. The IR spectrum of this solid product shows a NH band at 31h0 cm-1 1 and Clou' bands at 1080 cm” and 620 cm'l. The above data indicate that complexation takes place 32 in the solution but it is very weak because in aqueous solution, water molecules coordinate strongly to the Sc(III) ion and compete with the ligand. The formation constants of the complexes could not be evaluated because of follow- ing experimental difficulties: (a) broadening of the “58c resonance line which causes the disappearance of the signal at higher mole ratios (in case of C211). (b) Protonation of the ligand in aqueous solution (in case of C221). (0) Low solubility of the Sc salts in other solvents of interest. The combination of low solubility of scandium salts in nonaqueous solvents such as tetrahydrofuran and broad line width results in signal-to-noise ratios below a usable limit. Crown ethers dicyclohexyl-lB-crown-6, benzo-lS-crown-S and dibenzo-lB-crown-6 (Figure 1) were used as complexing agent for Sc(III) ion. To study complexation of dicyclo- hexyl-lB-crown-6 a series of aqueous solution of Sc(ClOu)3° 6H20 with varying mole ratios of ligand to metal was pre- pared. The uSSc nmr spectra of these solutions were obtained. This experiment was done at different concen- trations of Sc(ClOu)3'6H20. Results are shown in Table V. No appreciable changes in the chemical shifts or linewidths of the Sc—resonance in these solutions were observed. This shows that no complexation reaction has taken place in the solution. Similarly no evidence for complexation of benzo-lS-crown-S with Sc(III) ion was obtained for aqueous solution of Sc(ClOu)3'6H20. 33 Table V. “58c NMR Data for Sc(ClOu)3'6H20/Crown Mixtures in Aqueous Solution Conc. Sc(III) Chemical Shift Line Width M L/M* 6, ppm Hz 0.02 0.00 -19 138 0.02 0.20 -17 136 0.02 0.25 -16 129 0.02 0.30 -18 0.02 0.50 -17 127 0.02 1.00 -17 13“ 0.02 2.00 -17 102 0.03 0.00 -17 137 0.03 0.20 —16 120 0.03 0.75 -17 110 0.03 0.50 -17 117 0.03 1.00 -16 85 0.01 0.00 -20 0.01 0.20 -19 183 0.01 0.25 ~19 159 0.01 0.50 -20 207 0.01 1.00 -19 156 0.01 2.00 —20 158 *L/M == [Dicyclohexy1-18 crown 6]/[Sc(III)]. 3“ Since the solubility of dibenzo-18-crown-6 was very low in water, the complexing ability of this ligand toward Sc(III) ion could not be measured. From above data it can be concluded that these ligands cannot displace water molecules which surround the Sc(III) ion. This indicates the coordination abilities of these ligands are lower than those of water. Conclusion This study shows that “53c spectroscopy is very useful technique in directly providing qualitative information about the nature of the chemical environment about Sc(III) ion in its solutions. However due to low solubility of the salts quantitative data could not be obtained. The sensitivity of scandium resonance to the variation of the coordination environment of the Sc(III) ion makes this technique useful for study of Sc(III) ion solvation, ion pair formation, and complexation of this ion with different ligands in solution. In particular the range of the chemical shift for l"SSc-resonance is large,on the order Of hundreds of part per million, which makes it more sensitive to detect small changes in the chemical environ- ment of this nucleus. PART II 35 CHAPTER I HISTORICAL 36 Chemistry of Silver Silver with electronic configuration of [KrJleisl is a group B element and the second member of Cu, Ag, Au group. Its normal and dominant oxidation state in its compounds is +1. Oxidation states of +2 and +3 have also been identified in many silver compounds.“1 Silver ion shows a great tendency to form complexes with a wide variety of ligands. Organosilver compounds in which silver interacts with one or more carbon atoms have been characterized. In these compounds the organic moiety can be a-bonded to Ag+ ion as in alkyl , alkenyl, and alkynyl silver compounds, or a- and u-bonded as in alkene-, a1kyne—, and arene-silver coordination complex- es.u2-uu Coordination complexes of Ag(I) with aromatic and hetero- aromatic compounds have been a subject of interest for many years. In these compounds there is an interaction be- tween the silver ion and the n-electronic system of the aromatic moiety. This interaction includes both C- and u-type bond— ing. The O-type bonding is as a result of donation of w-electrons from ligand to metal ion and the w-type bonding is as a result of back donation of d-electrons of metal to ligand. ‘ In 1922 the first silver arene complex, C6H6Ag+,was detected and isolated from a solution of AgClOu in ben- zene.u5 Silver-arene complexes with other stoichiometries have also been reported.“6 37 38 Equilibrium constants for predominant stoichiometries of complexes of Ag+ ion with benzene have been determined from solubility measurements of benzene in aqueous silver nitrate solutions“7 and from measurements of the chemical shift of the benzene resonance in aqueous solution of benzene in the presence of varying excess concentrations “8 The following numerical values of of silver nitrate. equilibrium constants from the two measurements are similar but cannot be compared directly because the two types of measurements were made at different temperatures: K K 1 2 from solubility“7 2.u1 l--mol"1 (25.0°) 0.212 l-mol'1(25.0°) from nmr”8 2.30 kg/mol-l(33.5°) 0.u8 kg-mol'1(33.5°) In 1/1 and 1/2 (benzene/Ag+) complexes Ag+ ion was suggested to be on the six-fold symmetry axis of the aro- matic ring at one or both sides respectively.“7 This idea was supported by the results of Raman spectroscopic studies?9 But from theoretical considerations based on symmetries of quantum-mechanical wave functions an unsym- metrical geometrical configurations of these complexes was suggested. In these complexes silver ions lie away from the symmetry axis of the aromatic ring and above and between adjacent carbon atoms.50 This conclusion was supported by the results from x-ray diffraction studies on the solid complex of C6H6A301Ou which shows the silver ion 51 interacts primarily with the two nearest ring carbons. Further x-rav investigation suggested52 that the 39 discrete carbon atoms undergo a small amount of thermal motion and the benzene molecules possibly undergo some type of reorientation (highly hindered rotation). From proton nmr studies on the benzene-silver perchlorate complex at 77 K and 298 K,53 it was concluded that at lower temperatures the benzene molecule in the complex is more or less rigidly held in the crystal lattice, while at room temperature it is rotating more or less freely about the six-fold molecular symmetry axis. The apparent conflict- ing result from x-ray and proton nmr studies was attributedSu to the relative time scales involved in the two types of experiments. The proton nmr spectrum of the aqueous solution of A8N03 with benzene exhibits one single absorption due to aromatic protons.55 The observation was explained in terms of rapid exchange between several different species in equilibrium with one another, and it was not established whether or not all protons are equivalent. Complexation of several substituted benzene derivatives with silver ion has also been studied and the equilibrium constants of these complexes have been determined from solubility measurements of these aromatic compounds in aqueous solutions of silver nitrate.55'58 The results of these studies indicate that the stability of the complexes increases with increasing electron donating abilities of the substituent groups and suggest that o-type bonding is more significant than w-type bonding in the formation of these “0 complexes. The complexation reaction of silver ion with hetero- cyclic aromatic compounds such as pyridine and its substi- tuted derivatives has also been studied by many investi- gators. In pyridine and its derivatives, the electron pair present on the heterocyclic atoms is more available for in- teraction with Ag+ ion than are the aromatic n—electrons. In these complexes there will be a donation of electrons from the ligand to silver ion, 1,2,, o-bonding and donation from silver to ligand, 3121' n-bonding. Formation constants, enthalpy and entropy of the complex formation between silver ion and pyridine (pyridine/Ag+: 1/1 or 2/1) have been measured by different techniques 59’60 ultraviolet absorption spec- 62,63 such as calorimetry, 61 and potentiometry. troscopy The obtained values from different methods are in good agreement. log K1 log K2 -AH1 -AH2 (kcalgmol.) (kcal/mol) From calori- metricsg’60 measurement (at 25°, in H20) 2.0010.0h 2.11:0.08 h.83i0.05 6.5110.06 -ASl(e.u.) —A82(e.u.) 7.0 10.3 12.2:0.5 Al -AH1(kcal/mol) -ASl(e.u.) log K: (1 mol-l) From UV ab- sorption61 mea- surement (at 26° in CH3CN) 2.037 u.5 a) The K1 was reported61 as log K1 for comparison. log K1 from potentiometric measurement62 (at 1.97 25° in H20) Above values are for following reactions Ag+ + pr 1 Ag(PY)+ (x1, AHl. A81) Ag(PY)+ + PY : Ag(PY); (x2, AH2, A82) log K 5.8 to be 109 but is tabulated here 2.38 Formation constant and thermodynamic data for complexes of silver ion with substituted pyridine have been determin— ed62-6ll from potentiometric and calorimetric measurements and a linear relationship between the logarithm of formation constants and basicity of the ligands was obtained.63 92 Nuclear Magnetic Resonance Studies In recent years the use of magnetic resonance of metallic nuclei provided a sensitive probe in the study of metal ion reactions in solutions. In particular the range of the chemical shift for many metal nuclides is very large,on the order of hundreds to thousands ppm,which makes the chemical shift of metallic nuclei very sensitive to small changes in the chemical environment of the metal nucleus. The major disadvantage of the use of magnetic resonance of metal ion for chemical studies is the lower sensitivity of the resonances compared to the proton or fluorine. Both silver isotopes 107Ag and 109Ag have weak nmr signals which makes the measurements practically difficult. Silver-109 has a slightly higher sensitivity than 107Ag which makes it preferable for nmr studies. Silver-109 nuclide with the spin of 1/2 has “9.65% natural abundance and the low sensitivity at constant field of 1.01 x 10‘" compared to proton (assuming sensitivity of l for proton). In addition silver-109 has very low resonance frequency of 2.7916 MHz at a field of 1.N092 Tesla. Further difficulty of 109Ag nmr studies is its extremely long relaxation times (T1 on the order of minutes) due to small magnetic moment and zero quadrupole moment (I - 1/2) which results in very weak interaction of the silver nucleus with its surroundings. As a result of the experi- mental difficulties described above very few reports have 43 been published on 109Ag nmr studies. Most of the work done in this area carried out by various investigators report either studies on silver alloys,65'66 57,58 solid solutions, silver metal69 or on measurement of the magnetic moment or Larmor frequency of this nucleus.70'73 To our knowledge there are only two published papers on silver-109 nmr studies in solutions. One is by Burges, gt a.” and the other one is by Jucker gt £1.76 In addition some studies are being carried out at this time by Maciel and his group75 but have not been published as yet. Burges gt‘al, measured the ratio of Tl/TZ for 109 Ag in 9.1 molal aqueous solution of AgN03:T1/T2 = 12. They found that the addition of a paramagnetic ion decreases this ratio. They measured relaxation times for 8.3 molal aqueous solution of AgF (Tl/T2 - 611) were T1 - (“9:20) sec and T2 . (8.212) sec. They also measured the shift of the 107Ag Larmor frequency due to paramagnetic ion in a 9.1 molal aqueous solution of AgNO3 with varying concentration of a paramagnetic ion as Fe(N03)3 from 0-1.2 molal. The maximum observed shift was 8.3 ppm and the shift of 107Ag resonance was not a monotonous function of the concentration of paramagnetic ions. The concentration dependence of the Larmor frequencies of the 107Ag and 109Ag was determined in AgClOu, AgNO3 (107Ag) and AgF (109Ag) in the range from 0.“ molal to nearly saturated solutions. For AgClOu and AgNO3 an up- field shift was observed with increasing the concentration UN but for AgF, the shifts were toward lower field with in- creasing the concentration. Jucker and his group very recently reported?6 the 109Ag nmr studies of some silver complexes. They investigated solutions of silver salts in acetonitrile, propionitrile, pyridine, and ethylenediamine and also in aqueous solu- tions of Na23203 and ethylamine. From the data the forma- tion of several complexes in these solutions were suggested. They also observed that the chemical shifts of mixtures of AgCl and AgBr and also of AgCl and AgI dissolved in 70% aqueous solutions of ethylamine are linear functions of anion mole fraction. Maciel and his group75 simultaneously with our work used 109Ag nmr spectroscopy to study complexation of thiourea and tetramethyl thiourea with Ag+ ion in DMSO solution. They examined solutions containing 1 M silver nitrate in dimethylsulfoxide with varying ligand concentra- tion. The addition of the ligand to the silver ion solution results in a downfield shift of the 109Ag resonance. Thus the existence of the sulfur bonded Ag(I) complexes was postulated. From the study of the curvature of the plot of mole ratio of ligand to metal ion g§_chemical shifts, complex species including highly aggregated forms with different stoichiometries were suggested. This group has also studied solution chemistry of high concentrationscd‘AgNO in dimethylsulfoxide, acetonitrile, 109Ag 3 water and binary mixtures of these solvents using “5 nmr technique . The concentration and anion dependence of the 109Ag chemical shift was measured. The observed up- field shift with increasing the salt or anion concentration in the solution was related to the strong cation-anion interactions. They also found evidence for Ag(I) complexes with dimethylsulfoxide and acetonitrile from 109Ag nmr studies in these systems. Conclusions Thus it is apparent that in spite of experimental dif— ficulties that one encounters in silver-109 nmr measure- ments, this technique showed some promise for the studies of silver ion complexes in solution. The second part of this thesis reports the use of this technique in the I study of (A) silver salt solutions and (B) complexation reaction of this cation with aromatic and heteroaromatic compounds, such as benzene, pyridine and their derivatives. CHAPTER II EXPERIMENTAL PART N6 Experimental Salts Silver nitrate (Fisher, certified reagent) was dried under vacuum in an oven at 65°C. Silver perchlorate was purchased as hydrated or as anhydrous salt (0. Frederick Smith Chemical 00.). Both salts were of reagent grade quality. The hydrated salt was first recrystallized from an aqueous solution and then dried over P205 under vacuum at room temperature. The anhydrous salt was also dried by the same procedure. All the salts were kept in.the dark to prevent decomposition. Solvents Acetone (Fisher) was distilled over Drierite and further dried over molecular sieves. Acetonitrile (Matheson Cole- man & Bell)) was refluxed over calcium hydride and then fractionally distilled over granulated barium oxide. Di- methylformamide (Fisher) was vacuum distilled over P205. Dimethyl sulfoxide (Fisher) was dried over molecular sieves for two days. Methanol (Fisher) was first fractionally distilled from calcium hydride and then from magnesium turnings in a nitrogen atmosphere. Propylene carbonate (Aldrich) was dried over activated molecular sieves for two days. Molecular sieves used were activated by heating "7 H8 them at 500°C under dry argon in an oven (Lindberg) for 12 hours. Water content of salts and solvents were measured by an automatic Karl Fischer titrator Aquatet II from Photovolt Corp. Ligands Benzene (Matheson Coleman & Bell) and toluene (Analyti- cal reagent, Mallinckodt) were purified and dried by re- fluxing and fractionally distillating over sodium metal and benzophenone under nitrogen atmosphere. Pyridine (Fisher) was refluxed over granulated barium oxide and then fractionally distilled under nitrogen atmosphere. Ligand 2—cyanopyridine (Aldrich) was purified by fractional freezing technique.77 The purity of 2,5-dibromopyridine (Aldrich), 2,6-dichloropyridine (Aldrich), h-cyanopyridine (Aldrich) and 2,5-dichloropyridine (K and K Laboratories) were checked by their melting points (9h-95°C, 87-88°C, 78-79°C and 59-60°C, respectively), which were in agree- ment with literature values. Instrumental Measurement Silver-109 NMR Silver-109 nmr measurements were made on a Fourier transform instrument using the magnet of a Varian DA-60 49 nmr spectrometer equipped with a wide-band probe, computer78 controlled rf pulse generation and data collection as 78'80 1H field look was described previously. An external used to maintain field stability. The instrument was operated at a field of l.b092 T (mlu KG) and at a frequency of 2.7916 MHz for 109Ag. The measurements at room temperature were made with reference to an external standard which was a saturated aqueous silver perchlorate solution (~12.5 M) doped with Fe+3 salt in a 15 mm Wilmad nmr tube. At other temperatures the same reference solution was placed in a dewar type tube (Figure 3) which was used in the -110 to +130°C ranges. The reference solution was in.a 10 mm precision Wilmad nmr tube which was concentrically sealed in a 15 mm precision Wilmad nmr tube. The space between the tubes was evacuated 6 to 10' torr and then sealed. The secondary reference was a 6 M_aqueous solution of AgClOu doped with Fe+3. The mole ratio of Fe+3/Ag+ was 1/100. The chemical shift of this reference gs that of the primary reference was -59 ppm. The line width of the 109Ag resonance for the reference was 5.8 - 8.h Hz and for the sample solutions was 1.1 - 5.8 Hz. No systematic change occurred in the line width for the various experiments that were done. Since the relaxation time of 109Ag nucleus is extremely long to reduce it the sample solutions were doped with small amounts of paramagnetic ions such as Ni+2 (since Fe+3 precipitates in acetonitrile solutions of AgNO3). 50 10 mm Wilmad nmr tube K Vh‘ro vacuum v .1 Q” 'l“JA Saturated AgClOd‘ Aqueous solution 15 mm Wilmad nmr tube Figure 3. Reference solution in dewar-type tube. 51 In all cases the ratio of Ag+ ion to that of paramagnetic ion was at least 50/1. The lower limit of detection with the present instrument is 0.5 M. In our convention the upfield shifts correspond to the positive values. The reported chemical shifts are not corrected for change in diamagnetic susceptibility since the range of chemical shift is so large that it would not produce any problem. Carbon-l3 NMR Carbon-13 nmr measurements were made on a Varian CFT20 Fourier transform nmr spectrometer equipped with computer controlled pulse generation and data collection. The instrument was operated at a field strength of 18.682 KG and at a frequency of 20 MHz.81 For 130 nmr studies on solutions of Ag(I) salts, DMSO was used as an external reference (since TMS has some effect on chemical shift) and D20 was used for locking the system. Both DMSO and D20 at l/l volume ratio were in a capillary tube which was coaxially centered on the 8 mm nmr tube containing the sample solution._ For 13C nmr studies on complexation reactions, the chemical shifts were referred to the solvent peak which changed the least upon addition of salt (within 0.1 ppm change). The sample solution was in a 5 mm nmr tube which was coaxially centered in the 8 mm nmr tube containing the external reference acetone with D20 for locking the system. 52 A positive shift from the reference is downfield. The reported chemical shifts for this case were not corrected for diamagnetic susceptibility because the solvent peak were chosen as a reference. Proton NMR Proton nmr measurements were made on Varian A-56/60 D and T60 nmr spectrometers with operating frequency of 60 MHz and magnetic field strength of 1h.092 KG.82’83 The measurements were made with internal reference TMS for acetonitrile solutions and D88 (sodium 2,2-dimethy1-2- silapentane-S-sulfonate) for aqueous solutions. The sample solutions were held in a 5 mm 0.D. Wilmad nmr tube. The positive shifts from the reference is downfield.. CHAPTER III SILVER-109 NMR STUDIES OF SOLVATION AND COMPLEXATION OF SILVER(I) IN SOLUTIONS 53 Introduction It was mentioned previously that the NMR of some metal nuclei was shown to be a very useful probe for the study of these ions in solutions. It was of interest to us to explore the possible use of 109Ag nmr for the study of silver chemistry in solutions and in particular, in nonaqueous solvents. It should be mentioned that the experi- mental difficulties described earlier (page H2) and the low sensitivity of the 109Ag nmr signal did not augur well for the successful use of this technique at least with the previously available instrumentation; nevertheless, a detailed study of the 109Ag chemical shifts in various solvents at different concentrations and in the presence or absence of complexing agents would be a worthwhile initial exploration of silver nmr. A: NMR Studies of Ag(I) Salts in Solution 1. A Silver-109 NMR Study of Solutions of Silver-salts in Water, Acetonitrile, and Their Mixtures The chemical shifts of the 109Ag resonance for aqueous solutions of silver nitrate and silver perchlorate were determined. The data are presented in Table VI. The chemical shift "6" is referred to saturated aqueous solu- tion of AgClOu doped with Fe+3. 5H 55 Table VI. 109Ag NMR Data for AgClOu and AgNo3 in Water Conc., M 6(ppm) 0.5 A3010” -101.5 1.0 - 97.0 2.0 - 89.20 ”.0 - 71.7 6.0 - 53.u 8.0 - 3”.9 10.0 * - 12.2u 10.7 - 7.” 55193 1.0 - 92.8 2.0 - 81.9 3.0 - 73.8 ”.0 - 67.7 5.0 - 62.1 6.0 - 56.8 56 Figure ” shows the variation of 109Ag chemical shift as a function of concentration of AgNO3 and AgClOu. As is the case of Sc(ClOu)3 and Sc(N03)3 in aqueous solu- tion (Table II ) the chemical shift moves upfield with increasing concentration of silver salt. VanGeet and Templeman85 observed similar behavior for sodium per- chlorate in aqueous solution and they suggested that up— field shifts occur when a water molecule in the cation solvation shell is replaced by an anion which results in a decreased electron density around alkali cation. Deverell and Richards noted86 the same phenomenon for aqueous alkali metal nitrate solution and suggested a direct interaction between the ions is the predominant cause of the chemical shift of the cations resonance in the solution. Bloor and Kidd observed similar behavior for aqueous solu- tion of potassium nitrate and they proposed that since the chemical shift arises from the overlapping of the outer electron orbitals of ions during random ionic collisions, the upfield shift of the cation resonance is due to weaker overlap interaction between the anion with cation than water-cation interaction.87 It is possible that the same explanation applies to the upfield shift observed in aqueous solution of silver salt. Since the 109Ag chemical shifts reflect changes in the immediate environment (solvation sphere) of the cation and the shifts for the two anions extrapolate to the same value of the chemical shift which is indicative of the 57 0 ’g 40 CI. 3 CI 07: .9 60 ' so :20»— J J l l l l 2 4. 6 8 l0 I2 MOLARITY OF 119* SALT Figure ”. Silver-109 chemical shifts of silver salts in water. 58 chemical shift of free solvated Ag+ ion in aqueous solu- tions, it is possible to assume that the upfield shift with increasing concentration is due to increasing cation- anion interaction and ion pair formation. This assump- tion is also in accord with the conclusion proposed by Lee and Wilmshurst from their Raman studies on aqueous silver nitrate solution.88 The silver-109 chemical shift was also studied in several nonaqueous solvents. The resultsiku'acetonitrile solutions of silver nitrate is presented in Table VII. The variation of the chemical shift as a function of concentration is shown in Figure 5. There again an up- field linear shift was observed with increasing concen- tration as was observed with silver perchlorate in aqueous solution. In this case, the 109Ag chemical shift for the solvated Ag+ ion is considerably further downfield than in the case of water (chemical shift of ~-””0 ppm 1s -100 ppm) and also the variation of the chemical shift with concentration covers a much larger range in aceto- nitrile than in water. These data indicate that in these solutions strong interactions exist between the cation and the solvent as well as the cation and the anion. This ob- servation is in accordance with the conclusion of Janz 22.21-89’90 that in these solutions complexes Ag(NCCHB); and Ag(NCCH3)+ are formed and there is an equilibrium be- tween contact ion pairs and these complexes. They arrived at this conclusion from the results of their ir, Raman, and 59 Table VII. Silver-109 NMR Data for AgNO3 in Acetonitrile. Conc. of AgN03, M dppm 0.5 -”29.6 1.0 -”l2.8 2.0 -397.2 3.0 -379.” ”.0 -362.5 5.0 ~3”1.5 6.0 -317.0 60 5; 3 300— 320 - 4’ 340 r- / 360 - -3I09A9(F>Pm) 3807- 420+- L l l l 2 4 6 MOLARITY or: 1193103 Figure 5. Silver—109 chemical shift of silver nitrate in acetonitrile. 61 nmr spectroscopic studies on this system. The experiment could not be repeated with A3010, due to low solubility of the salt in acetonitrile (5.1 M). Other solvents of interest,propy1ene carbonate, acetone, tetrahydrofuran, dimethylsulfoxide, dimethylformamide, acetonitrile, and pyridine were chosen for 109Ag nmr studies since 1 M solution of AgClOu could be made for 109Ag nmr studies, and the solutions were stable over a long period of time. Concentration studies were not possible because of low solubility of AgClOu (ml M) and low level concentra- tion, 0.5 M detection limit of 109Ag for the instrument. Silver nitrate could not be tried because of low solu- bility of this salt in the above solvents. For concentra- tion levels in these solvents with which experimental work was possible, 109Ag chemical shifts were obtained. The results are shown in Table VIII, as are the donor numbers of the solvents used. The data from this study indicate that the extent of low field shift is larger for solvent with higher donor number which is understandable since these solvents have higher ability to interact with Ag+ ion and decrease the interaction of perchlorate ion with Ag+ ion. The low field shift is largest for pyridine and acetonitrile is next lowest even though acetonitrile has the lowest donor number. This observation might be explained in terms of presence of w-bonding in these cases in addition to the o-bonding which is present in the case of other solvents.62’63 62 Table VIII. 109Ag NMR Data for l M_AgClOu in Different Solvents Donor Solvent Number 6(ppm) PC 15.1 + 5 AC 17.0 - 61 THF 20.0 - ”1 CH3OH 25.7 - 55 DMF 26.6 -130 DMSO 29.8 -258 CHBCN l”.1 -529 PY 33.1 -556 63 The variation of 109Ag chemical shift for ZM'AgNO3 in water-acetonitrile mixturescfl'varying composition was also studied as a function of solvent composition. The data are presented in Table IX. From the curvature of the plot of chemical shift vstole fraction of acetonitrile as shown in Figure 6, it is obvious that the changes in chemical shift are large even in solvent mixtures with low mole fractions of acetonitrile. This behavior indicates that the interaction of acetonitrile with silver(I) ion is strong. The isosolvation point occurs at 0.075 mole fraction of acetonitrile which indicates a strong preferen- tial solvation by this solvent. Similar behavior was observed in studies in which transference, conductance, and potentiometric measurements were made.91 From the results of these studies selective solvation of Ag+ ion with aceto- nitrile and preferential accumulation of water in the environment of the nitrate ion was concluded. It is in— teresting, however, that the plot is not monotonic but shows a definite and reproducible minimum at No.6 mole fraction of acetonitrile. Maciel gt 31.92 observed the same behavior and the minimum in the curve can be explained as follows: upon initial addition of acetonitrile to the aqueous solution of AgNOB, the silver ion interacts with the organic solvent while the nitrate ion remains hydrated, as a result the 109Ag resonance shifts to lower field. At the mole fraction at which the minimum occurs, 6” Table Ix. 1°9Ag NMR Data for AgN03 2 M’in Water-aceto- nitrile Solutions. MCH3CN/MH20 5‘99”) 0.000 - 82.2 0.018 -139.0 0.037 ~197.2 0.079 -277.8 0.128 ‘330-5 0.187 -36”.9 0.257 '385.5 0.381 -uou.0 0.””7 -”1”.5 0.587 -”18.9 0.757 -”1”.3 0.970 '395.” 65 80g |20~ .4: CH3CN - H20 Mixture ISOL- ZM AgNOs 200»? g. . Q 240 '- VD q P" 8 ‘6 280 '- l _ 320L 360“ 400 - 41/? l l a O 0.2 0.4 0.6 0.8 LO mole fraction CHscN Figure 6. Variation of the chemical shifts of the silver- 109 resonance as a function of solvent composition for binary solvent, water and acetonitrile. 66 3 tion sphere of N03 ion is removed and since N0 N0 is still hydrated but at higher mole ratio, the hydra— 3 solvated by acetonitrile, the interaction between Ag+ is poorly with N03 increases and the shift toward higher field occurs. In all solutions used in 109Ag nmr studies, only a single nmr line was observed, because the rate of exchange of species interacting with Ag+ ion is fast, therefore, the observed resonance signal corresponds to the population average of all the Ag-species present in the solutions. 2. A Proton NMR Study of Solutions of AgN03_in Water, Acetonitrile and Their Mixtures The variation of lH(CH3CH) chemical shift as a function of AgNO3 mole fraction was measured in solutions of AgNO3 in acetonitrile. The measurements were made with two instruments, T60 and Varian 56-60. The results from the two sets of measurements are in good agreement. The chemi- cal shifts are referred to TMS resonance. The data are presented in Table X, and a plot of chemical shift vs mole fraction of AgNO is shown in Figure 7. From the 3 data, it becomes apparent that as the mole fraction of salt increases from 0.026 to 0.30 the chemical shift moved downfield by 1313 Hz. For the same change in mole fractions, Janz89 observed 28 Hz and Schneider91 10 Hz downfield shift in the 1H chemical shift. The observed downfield 67 .Apco>aom peony 20mmo when no unfinm Hmofisono on» op oopmaon popped“ some one mza op commence pmsfim one; mpmanm Headsono n.m 0.m 0.0a 0.0 0.0a H0m.0 0.s 0.0 0.0a 0.H 0.00 000.0 0.0 0.m 0.0a 0.0 0.0a 00N.0 0.0 0.0 0.0a 0.H 0.0a 000.0 0.0 o.m 0.0 m.H 0.0 oma.o o.m 0.0 0.e 0.H 0.5 H00.0 0.0 0.0 0.0 0.H 0.0 m00.0 0.H 0.N 0.0 0.0 0.0 000.0 0.0 as 0.0 an 0.0 we 0.0 nm 0.0 000.0 0.0 100-00 scasce aov 1009 eev +00 no a peace; ocaq-eeaem Hooasono scopes osaaupeaem Hmoasono someones .oooo mac: mafihpficopmo< CH mozm¢ ho mCOdeflom flow puma mzz HIGOQOAm .xn wands 68 80-12) Figure 7. Proton-l chemical shifts (CH3CN) as a function of mole fraction of AgNO3 in acetonitrile solu- tions of AgNO3. 69 shift on addition of the salt to the solution indicates that as the concentration increases there is a decrease of electron density around hydrogen atoms which might be due to the complexation of more acetonitrile with Ag+ ion. In all solutions only one resonance line was observed due to fast exchange of molecules. The same experiment was repeated in H20 as a solvent. The results are presented in Table XI and Figure 8. Chemical shifts are referred to DSS (page 52) resonance since TMS is not very soluble in H20. As the mole fraction of the salt was increased, almost no systematic change in the 1H(H20) resonance was observed which may be due to a weaker interaction of H20 with Ag+ ion than with that of CH3CN. The variations of 1H(CH3CN) and lH(H20) resonances as a function of mole fraction of either solvents were also measured in solution of constant concentration of AgNO3, [2 M], and different composition of the two sol- 1H nmr studies were conducted vents. For comparison the in solution with different composition in the absence of Ag+ salt. Data are presented in Table XII and XIII. The plots of chemical shift vs concentration for different cases are shown in Figures 9 and 10. The results indicate that in the absence of Ag+ ion, as the mole fraction of H20 increases there is a downfield shift of 1H(CH3CN) resonance which might be due to the interaction of the hydrogen atom of H20 with nitrogen atom of acetonitrile which causes deshielding of acetonitrile. 70 Table XI. Proton-l NMR Data for Aqueous Solution of AgN03. Concentration Chemical Shifta Line width of Salt, M, Hz Hz 5 278 ”.5 ll 279 3.0 2 279 2.0 1 279 1.2 O 276 1.0 aChemical shifts are referred to DSS. 71 I i I I l 288 r- ‘ 3 +____.__.__.___+——o— (HZ) 268- '- 1 g l I l I 2 M .3 4 5 A94” Figure 8. Proton-1 chemical shifts (lHZO) as a function of AgNO3 concentration in water. 72 Table XII. Proton-l NMR Data in Acetonitrile—water Mix— tures. X X Chemical Chemical CH3CN H2O Shift (Hz) Shift (Hz) 1H(CH CN) 1H(H 0) 3 2 0.000 1.000 ___ 290 0.018 0.982 138 290 0.079 0.921 131 27” 0'257 0‘7"3 122 2”6 0.”50 0.550 121 229 0.580 0.”20 120 211 0.756 0.2”” 117 198 1.000 0.000 Table XIII. 73 Proton-l NMR Data of Solutions of AgNO in Acetonitrile-Water Mixtures. 3 Chemical Chemical Shift (Hz) Shift (Hz) XCH3CN XH20 1H(CH3CN) 1H(H2o) 0.000 1.000 --- 290 0.018 0.982 1”1 286 0.079 0.921 137 28” 0.257 0.7”3 128 25” 0.050 0.550 128 236 0.580 0.”20 127 222 0.756 0.2”” 126 219 0.870 0.130 125 217 1.000 0.000 125 --- 7” I47 - I37 — 8(HZ) (b) :27 - . .. H7 : 0.2 ' X H 20 Proton-l chemical shifts (CH3CN) as a function of water mole fraction in water-acetonitrile mixtures (a) in the absence and (b) presence of the AgN03. Figure 9. 75 307‘ i 1 l I o 247- 8(HZ) ._ o (b) - (a) l8 0 _ L l l l (D (3.2! (1‘! C165 (D.E3 X CHCN 3 Figure 10. Proton-1 chemical shifts (H20) as a function of acetonitrile mole fraction in acetonitrile- water mixtures (a) in the absence and (b) presence of the AgN03. ' 76 When the ASN03 is present, the same phenomenon is taking place but in addition there is also a downfield shift caused by interaction of Ag+ ion with the nitrogen atoms. For 1H(H20) in the absence of the salt there is an upfield shift of lH(H20) resonance when the mole fraction of aceto- nitrile is increased. This shielding of proton might be due to interaction of hydrogen atom with nitrogen atom of acetonitrile. This shielding is less when the salt is present in the mixture which indicate due to complexation of acetonitrile with Ag+ ion, less acetonitrile molecules are present to interact with H 0. At high mole fraction 2 of acetonitrile all water molecules contribute in the interaction with acetonitrile and the 1110120) chemical shift remains constant. Schneiderg1 has done similar studies in these systems. He explained the downfield shift of 1H(CH30N) resonance to the formation of [Ag(CHBCN)2]+, complex and the upfield shift of 1H(H20) resonance on the addition of acetonitrile to the ruptures of the hydrogen bonds between water molecules. This ef— fect was not as great when the silver salt was present. 3. A Carbon-13 NMR Study of Solutions of AgNO.2 in QEBCN and CHBCN-HZO Mixtures The variation of the 130 chemical shift as a function or ASNO3 concentration in acetonitrile solutions was measured. The results are presented in Table XIV. The 77 Table XIV. Carbon-13 Data of Solutions of AgNO3 in Acetonitrile Conc. of Chemical Shift Chemical Shift A3N03’ fl (-13CN) ppm (-13CH3) ppm 0.0 117.89 1.3” 0.1 117.97 1.38 0.2 118.06 1.”0 0.5 118.27 1.”7 0.7 118.”2 1.52 1.0 118.6” 1.57 2.0 119.30 1.79 3.0 119.91 1.97 ”.0 120.”5 2.06 5~0 121.01 2.25 6.0 l2l.”9 2.3” 7.0 121.80 2.”9 78 plot of chemical shift vs concentration is shown in Figure 11. For both carbon atoms: -l3CH3 and 13ON a downfield shift was observed as the concentration of salt was in- creased. The extent of Shift was more for the -13CN than the -CH3. These results are not unexpected since aceto- nitrile is known to interact with the Ag+ ion through the nitrogen atoms. The 13C chemical shifts were also measured for solu- tions 0f A8N03 in acetonitrile-water mixtures. The results are presented in Table XV. The variation of chemical shift XE mole fraction of water for both acetonitrile carbons is shown in Figure 12. The data indicate that as the mole fraction of water is increased a low field shift of both carbons is observed. The extent of shift is larger for -13CN than ~13CH3. This low field shift was also observed for protons in acetonitrile and as it was explain- ed it is due to interaction of water with acetonitrile through nitrogen atoms. If one compares the results from three different tech- niques: 1H, 130, 109Ag nmr, it is obvious that the 109Ag has the largest chemical shift range. a) For the concentra- tion range of l to 6 M for solutions of AgNO3 in aceto- nitrile, the change in the chemical shift of: 1H(-CH3) is ~0.2 ppm, l3C(-CH3) is ~0.77 ppm, 13C(-CH3) is ~0.77 ppm, 13c(-CN) is ~2.85 ppm and for 109Ag(AgN03) is ~96 ppm. b) For the concentration range of l to 5 Miof AgNO 3 in aqueous solutions, the change in the chemical shift 79' I2I.8 CH3CN I I 12|.O |20.2 . I ”9.4" 8 '3CIppm) ||8.6 l we) 2.81 2.0 '- p- l l 1 l I o I 2 3 ‘1 5 6 MOLARITY of AgNO3 Figure 11. Carbon—13 chemical shifts as a function of silver nitrate concentration. 80 Table XV. Carbon-l3 Data of Solutions of AgNO3 in Aceto- nitrile-water Mixtures. Chemical Shift Chemical Shift xH20 XCH3CN -13CN -13CH3 0.000 1.000 119.28 1.81 0.130 0.870 119.”8 1.8” 0.2”” 0.756 119.70 1.87 0.”20 0.580 120.08 1.96 0.550 0.”50 120.33 2.03 0.793 0.257 120.83 2.16 0.921 0.079 l21.”5 2.35 0.982 0.018 121.67 2.39 81 12I.s- CH3CN -H20 mixture 2M AgNO3 |2|.4- -130N l2|.O |20.6 120.2 ”9.8 S'BCIOpm) ”9.4 I . - '9025 _13CH 2.6- 3 I. 22'- L8 1 L 1 1 1 I 1 1 1 1 O <12 <14 (16 (18 L0 MOLE FRACTION H20 Figure 12. Carbon—l3 chemical shift as a function of water mole fraction in mixtures of water- acetonitrile. 82 of: 1H(H20) is No.0 ppm and 109Ag(AgN03) is ~30 ppm.. 0) For solutions of AgNO3 in a mixture of acetonitrile- water, the change of the chemical shift for mole fraction change of acetonitrile from 0 to l are for: 1H(-CH3) ~0.26 ppm, for 13C(-CH3) 0.58 ppm, for l3C(-CN) ~2.39 ppm and for 109Ag(AgN03) N300 ppm.. B. NMR Studies of Ag(I)gComplexes in Solution 1. A 109Ag nmr Study of the Complexation Reaction of Ag+ Ion with Benzene and Toluene Silver-109 nmr spectra were obtained in a series of solutions of 1 M AgClOu with varying mole ratios of ligand to the metal ion in different solvents. Benzene and toluene were used as the ligands. Complexation reactions were studied in propylene carbonate, dimethylformamide, di- methyl sulfoxide, acetone, methanol and pyridine solu- tions. These solvents were preferred because they dissolve Ag+ salts, they are miscible with benzene and toluene, dissolve the doping agent, and the solutions in these solvents are stable. The data are presented in Tables XVI and XVII and plots of chemical Shift as a function of mole ratio of ligand to metal are shown in Figures 13-19. The data indicate that as the mole ratio of ligand to metal ion increases in the solutions, the 109Ag resonance 83 Table XVI. 109Ag Chemical Shift as a Function of Benzene/ AgClOu Mole Ratio in Various Solvents. 29. 1.0. Ile/[As*la appm [BZJ/[Ag+]b oppm 0 + 5.2 0 - 61.3 1/10 - 7.0 1/10 - 59.5 1/7 - 8.5 1/1 - 91.” 1/3 - ”0.2 2/1c —129.u 1/2 - 57.2 3/1 -165.3 1/1.5 - 78.7 5/1 -218.2 1/1 -l27.5 7/1 -282.2 2/1 -205.5 8/1 -303.3 3/1 -262.5 9/1 -3u1.1 ”/1 -289.5 10/1 -369.1 5/1 -323.2 6/1 -3”0.2 7/1 -352.” 8/1 -366.5 9/1 -375.3 10/1 -389.1 Table XVI. Continued. THF. 91139.89. [Bz]/[Ag+] Gppm [le/[Ag+] dppm 0 - l”.9 0 - 55.1 1/2 - 31.5 1/10 - 65.6 1/1 - 79.6 1/2 - 97.1 2/1 -113.7 1/1 —l”6.5 3/1 -152.2 2/1e -188.9 ”/1 -191.5 3/1 —2””.9 5/1 -233.6 ”.22/1 ~283.u 7/1 -277.7 5/1 ~300.0 8/1 -311-3 7/1 -335-0 9/1 -3uu.6 DMSOf QMEE 0 -257.5 0 -l30.3 1/2 -256.2 1/2 -136.u 1/1 -256.” 1/1 -l”3.0 2/1 -255.u 2/1 -15u.8 3/1 -251.2 3/1 -178.u ”/l -2u9.2 ”/1 -195.0 5/1 -200.3 6/1 -211.6 85 Table XVI. Continued. 333 [BZJ/[As+1 ppm 0 -556.2 1/2 ~55”.5 1/1 -55”.5 2/1 -551.8 3/1 -5”7.5 ”/1 -5”3.1 a’bAt mole ratios higher than 10/1, both salt and doping agent were not soluble enough. c Signal could be seen only after 5 h scanning, while for the rest of the samples 1 h was enough. d Above mole ratio of 7, two phases were formed in the solu— tion which might indicate the mixability of benzene in CH OH is limited, so chemical shift of 09Ag in these so utions was not measured. e Signal could be seen after 5 h scanning, while for the rest of the samples 1-2 h was enough. f Above mole ratio of ”/l 8 Bz/Ag+, the benzene and DMSO were not mixable. g Decomposition took place in all the solutions after a while so the experiments should be done fast in these solutions. h At L/M = ”/1 solution seems to be saturated, so higher mole ratio was not examined. 86 Table XVII. 109Ag Chemical Shifts and Line width as a Function of Toluene/AgClOu Mole Ratio in Different Solvents .139: 19 [Toll/[Ag+1 dppm [TOlJ/[A8+] dppm 1/2 - 77.0 1/10 - 57.7 1/2 -1u6.0 1/2 - 57.7 2/1 —232.6 1/1 - 93.6 3/1 -292.1 2/1 -132.9 ”/1 -315.7 3/1 -179.3 5/1 -336.2 ”/1 -215.1 6/1 -3u9.8 5/1 -2u5.0 0 5.0 6/1 -268.5 7/1 -305.2 28.3911.” 112° 0 - 55.5 0/1 - 15 1/10 - 70.0 1/2 - ”1 1/2 -ll7.2 1/1 - 78 1/1 -l32.0 2/1 -l30 2/1 -200.3 3/1 ~17” 3/1 -257.1 ”/1 -21” ”/1 -288.5 5/1 -2”8 5/1 -307.8 6/1 -282 6/1 -325.3 87 Table VII. Continued. DMSOd BEE [Toll/[Ag]d dppm [Toll/[Ag]e dppm 0 -257.5 0 -l30.3 1/2 -256.0 1/2 -137.0 1/1 -255.8 1/1 —iuu.o 2/1 -253.6 2/1 -l60.9 3/1 —251.8 3/1 ~175.” ”/1 -186.7 5/1 -20u.2 £1 [Toll/[AH]f dppm 0 -556.2 1/2 ~552.7 1/1 -552.7 2/1 -5ua.u aAt above mole ratio of 6/1 the solubility is not enough (at L/M = 7, solution is not clear), it might be due to low solubility of AgClOu in toluene. b the low mixability of the toluene with CH30H. cAt mole ratio above 6/1 solution seems saturated. dAt mole ratio L/M = 3/1, the toluene and DMSO are not mixable. eAt mole ratio of 6/1, two phases were formed. fThe solution was saturated at [Tol]/LAg+]= 2/1 and above that precipitation took place. At mole ratio above 6/1 two phases were formed because of 88 I I I I I l O -I I -- PC 50 {-.\ --—--THF __ r\ \ ----<: H30.H Is... -\. “““PY .. '\_. “‘"DMSO 2 0 O “' ~ WDMF ‘ -3 \ \-. mp” 250— \. 0 \ \, \\ \\ 3 5 0 _.. \, ._ .IJ¢¢V\_ 4-0 8-0 12-0 MBz/IVIAg Figure 13. Chemical shifts of 109Ag as a function of benzene/AgCloh molar ratios in various solvents. (In this plot experimental points are not dis- played to prevent crowding. The same plots are shown in the following figures in more detail.) 89 IOO (ppm 200 300 Figure 1”. I l ' ' I ._ I' o 82 __ \, . To \ r- . PC 5 1 l I ' 20 6.0 \0.0 ML/MAg Chemical shifts of 109Ag as a function of ligand/ -A8010u molar ratio in propylene carbonate. (Ligand either benzene or toluene) (In this fig- ure and following figures with the same point sizes the error range is less than the point sizes used.) ' 90 O BZ I00 0 TO CHCOCH -3 200- 3 3 " (ppm) 300 - ‘ I l l l | 2 . O 6.0 I0.0 ML/MAg Figure 15. Chemical shifts of 109Ag as a function of ligand/ AgClOu molar ratio in acetone. (Ligand is either benzene or toluene.) ' 91 IOO Figure 16. Silver-109 chemical shifts as a function of ligand/AgClOu molar ratio in methanol. (Ligand is either benzene or toluene.) 92 - I I I I I o BZ -8 . 4 (ppm) 200 r- ‘ THF 0 300*- . " 1 1 1 1 1 2 O 6.0 10.0 Figure 17. Silver-109 chemical shifts as a function of ligand/AgClOi molar ratio in tetrahydrofuran. (Ligand is e ther benzene or toluene.) 93 | l l 1 l I ° 82 IOO- _. 0 TO 6 -8 (ppm) DMF 200- _ DMSO 5 300- ' 1 2.0 4.0 60 M M L.// l”g Figure 18. Silver-109 chemical shifts as a function of ligand/AgClOII molar ratio in dimethylformamide and dimethylsulfoxide. (Ligand is either benzene or toluene.) 9” 5I0 I fl 1 l 0 82 0T0 --8 550 "‘ (PP M) P Y 590 -- "‘ I l I I0 2.0 3.0 ML/M Ag Figure 19. Silver—109 chemical shifts as a function of ligand/AgC10 molar ratios in pyridine. (Li— gand is eith r benzene or toluene.) 95 shifts down field. Maciel gt_al,75 also observed a similar behavior when thiourea or tetramethyl thiourea were added as ligands to solutions of Ag+ ion in DMSO. They related this downfield shift to the deshielding of 109Ag as a result of complexation of Ag+ ion with these ligands. However, the extent of the shift to lower field due to the interaction of Ag+ ion with ligand is different in different solvents (Figure 13) which indicates that the solvent plays an important role in these complexation processes. In the case of solvents with low or moderate donor number such as PC, THF, AC and CH OR the shift to lower field is larger. 3 For solvents with higher solvation ability such as DMF and DMSO, the extent of shift is less. In pyridine, which is known to form a stronger complex with Ag+ ion, the change in the chemical shift is very small. From the comparison of the results for the two ligands (benzene and toluene) one notices that the downfield shift is more pronounced for the toluene than benzene, in propylene carbonate, acetone, methanol, and tetrahydro- furan. In these solvents the complexation of Ag+ seems to be stronger with toluene than benzene, from the curvature of the plots of chemical shifts !§_the mole ratio of ligand to metal ion (Figures 13-17). The same conclusion was reach- ed from solubility measurements“2 , a larger stability constant was found for toluene than benzene and the stronger inter- action Ag+ ion with toluene was related to the inductive effect of -CH3 group in the ring. This difference in 96 stability is leveled off in dimethylformamide and dimethyl- solfoxide which are better solvating agents than former solvents. 2. A 109Agnmr Study on Complexation Reaction of Ag+ Ion with Pyridine. Silver-109 nmr spectra were obtained in a series of l M AgClOu solutions with varying mole ratios of pyridine to metal ion in different solvents such as propylene carbonate, dimethylformamide, dimethyl sulfoxide, acetone, and meth- anol. The recorded values of the chemical shifts for each solvent case are shown in Table XIX and the plots of chemical shift gs molar ratio of pyridine to metal ion are presented in Figure 20 and also in detail in Figures 22-27. The data indicate that the 109Ag resonance shifts downfield with increasing concentration of pyridine which is due to the interaction of Ag+ ion with pyridine. In all solvents used, the chemical shifts approach the same value at higher mole ratio of ligand which shows that the chemical shift is independent of solvent used and the Ag+ ion is all complexed by pyridine. In the case of acetonitrile as a solvent a pronounced break is observed in the plot of chemical shift is mole ratio of pyridine to metal ion at mole ratio of 2:1 which can be an indication of the stoichiometry of the complex which is a known Ag+-pyridine complex species. 97 Table XVIII. Silver-109 chemical shifts as a function of temperature. A for regular reference solu- tion, B for reference solution in dewar type tube, Treference solution is saturated a solution of AgClOu ~12.5 M doped with Fe qgeous A B Temperature, °Ca Gppm Temperature, °C appm ~106 0.0 -31 -3.5 - 88 0.0 -23 -3.0 - 73 —0.9 -13 -l.5 - 55 0.0 - 3 -0.8 - 36 0.0 12 0.9 - 16 0.0 20 0.0 19 0.0 36 0.8 37 0.0 ”1 1.0 61 0.0 ”9 0.8 87 0.8 51 0.9 91 0.8 53 1.0 116 0.8 66 1.5 71 3.0 80 2.0 8” 3.5 93 1.7 96 3.5 aBelow -3l°C solution froze. 98 Table XIX. Silver-109 Chemical Shifts as a Function of Pyridine/AgClOu Molar Ratio in Different Sol- vents at Different Temperatures. 5520 55° 28° -”8° [PY]/[Ag+] dppm [PYJ/[Ag+] dppm [PYJ/[As+l dppm 12.38 -538.7 0.00 12.2 0.00 - 1.0 0.99 -l78.” 0.99 -l9l.5 1.98 -360.3 1.98 -377.8 3.99 -”80.1 3.99 -5”2.3 5.99 -525.6 5.99 -586.8 8.99 -5”7.5 8.99 -593.8 12.38 -553.0 12.38 -595.6 Acetone 28° ~22° -36° 0.00 -56.8 0.00 - 70.0 0.00 - 71.7 0.99 -212.5 0.99 -223.0 5.99 -572.9 2.97 -”28.5 2.97 -”53.0 8.99 -586.9 ”.00 —”7”.0 ”.00 -516.0 12.38 —589.5 5.99 -517.7 5.99 -560.6 8.99 -5u2.2 8.99 -577.2 12.38 -55”.5 ” 12.38 ~582. Table XIX. Continued. 99 DMSO 28° 3. [PY]/[Ag+] dppm [PYJ/[As+] Gppm 0.00 -258.0 0.00 -265.9 0.99 -3u1.0 0.99. —3u7.2 2.50 -u30.3 2.97 -u52.1 3.99 -”68.8 5.99 -519.” 5.99 ~503.7 8.99 -551.8 7.92 -521.7 12.38 ~567.6 8.99 -535.2 9.90 -5”2.2 12.38 -555.0 DMF 28° -uu° 0.00 ~130-3 0.00 -132-0 0.99 -272.0 0.99 -285.1 1.98 -393.5 1.98 —”13.6 3.99 -”77.5 3.99 -539.8 7.”0 -527.3 5.99 -571.9 9.90 -5””.5 7.”0 -592.5 12.38 -551.5 12.38 —593.9 Table XIX. Continued. 100 THF 28° 22° [PYJ/[Ag+] dppm [FYI/[As+] Gppm 0.000 - 7.8 0.000 - 8.8 0.2”7 - 80.5 0.195 -101.u 0.195 -116.7 3.998 -u81.0 3.”66 -”59.3 5.990 -519.5 3.998 -”77.5 8.987 -5”7.5 5.990 -51u.3 12.380 -556.3 7.920 -533.5 9.900 —5”6.6 12.380 -553.0 CH3CN 28° 21° 0.00 -529.1 0.00 -535.2 0.99 -505.6 0.99 -509.9 0.99 -50”.6 1.98 -”88.0 1.98 —u85.” 3.99 -51”.2 2.97 -”96.7 6.93 ~537.8 3.99 -510.2 9.90 -550.1 6.93 -532.6 9.90 -5”9.2 6.93 -533.5 12.38 -556.2 9.90 -5”6.6 12.38 -553.6 101 I 1 H I F l 0 r' 3 ——DMSO \\ “‘PC .IOOL'I‘ _ “AC . \I‘ --O M F '\I\ “C H3CN I2()C)P§ M -I -8 I :1 ‘1 (P PM) \ soo- I. \\ 4001— \. .\\\ \1 50 O - °= \\ e; C) l _I I I l 1 O 20 4O 60 '80 I00 IZO MPY / MAg Figure 20. Chemical shifts of 109Ag as a function of pyri- dine/AgClOu molar ratio in various solvents. (The experimental points are not displayed in this plot to prevent crowding, but the same plots are shown in following figures in more detail.) 102 I F I L 1 I 1 I 1 3 -IIo -30 +50 +I30 (PPM) I I I I I I I '+”D'_.r1r_4y—Ch—4r—‘F—CP—CHICF-4V.F4IDH. "1 -l0 —- B . '7 1 i J_ l l 1 l —20 +20 +60 +I00 o T Figure 21. Silver-109 chemical shifts as a function of temperature: A) for reference solution in dewar type tube, B) for reference solution in regular nmr tube. 103 4.0 8.0 l2 .0 M M PY/ A9 Figure 22. Chemical shifts of 109Ag Kg pyridine/AgClOu molar ratio in propylene carbonate at dif- ferent temperatures. 10” I I I I I I I C) _ I 0 28° 9 o --2 2 200e, _ _8 a -36° IPPM) 400- — ° C HCO . 3 C8 31\ . 600- . — J I 1 1 I I 1 20 6%) MM) 140 M /M PY Ag Figure 23. Silver-109 chemical shift as a function of pyridine/AgClOu molar ratio in acetone at dif- ferent temperatures. 105 O -I o 280 O - o 22 ‘I ‘I 2001‘ " \\ -8 \\ hppnfl \b\ \\ _J 400- \\ \ THF 600,— "" I 1 I I 1 1 I 40 8-0 I210 M M PY/ Ag Figure 2”. Silver-109 chemical shift as a function of pyridine/AgClOu molar ratio in tetrahydro- furan at different temperatures. 106 o 22‘ —-8 (PINTOSOO 600— ACN I I I“ I I 1 I 40 8'0 l2-O M PY A9 Figure 25. Chemical shifts of 109Ag as a function of pyridine/AgClOu molar ratio in acetonitrile at different temperatures. 107 '00 I I I I I I ._ (38m) 500 —I 700 - - DMSO I 1 1 1 1 I 4.0 8 -0 I20 M M PY Ag Figure 26. Silver-109 chemical shift as a function of pyridine/AgClOu molar ratio in dimethylsul- foxide at different temperatures. 108 ZOO GOO DMF 4.0 8.0 [2.0 M M PY Ag Figure 27. Chemical shifts of 109 vs pyridine/AgClOu molar ratio in dimethylfOrmamide at different temperatures. 109 The influence of temperature on the 109Ag chemical shift was studied. The measurements were made using an insulated reference solution (in a dewar type tube). This arrangement allowed the determination of true temperature effects on the 109Ag chemical shifts. It is seen from Table XVIII and Figure 21 that the chemical. shift of the insulated reference solution is much less temperature- dependent than the chemical shift of the uninsulated sample. The results of silver-109 nmr studies on the l M AgCHX‘solution with varying molar ratios of pyridine in different solvents and temperatures are presented in Table XIX. The plots of chemical shift 19 the molar ratio of ligand to metal ion for each solvent case are shown in Figures 22-27. The results indicate that in all solvents tried, the shift to lower field is larger at lower temperature and the effect of temperature is more on the Ag+ complexed with pyridine than on the solvated Ag+ ion. In addition, the approach of the chemical shift to the limiting value is more pronounced (the steeper slope) at lower tempera- ture which might be due to a greater stability of the complex at lower temperatures. This observation is in agreement with the results of calorimetric studies on this system (page ”0). From this measurement, a negative enthalpy for the reaction was calculated.59’60 110 3. Complexation Reaction of Ag+ Ion with Substituted Pyridine Derivatives Since a linear relationship between the logarithms of formation constants and the basicities of the ligands in complexation of Ag+ ion with substituted pyridine has been found,63 it seemed interesting to find out how lo9Ag nmr reflects this phenomenon. Silver-109 nmr spectra were studied in l M AgClOn solutions with substituted pyridine compound such as 2-cyanopyridine, ”-cyanopyridine, 2,5- dichloropyridine, 2,5—dibromopyridine, and 2,6-dichloro- pyridine. Complexation of 2-cyanopyridine was studied in dimethyl sulfoxide, dimethylformamide and pyridine. The other solvents of interest could not be tried because of low solubility of this ligand. The chemical shift was measured as a function of 2CNPy to metal ion molar ratio in these solvents. A limited number of molar ratios could be tried because of either the low solubility of the ligand or precipitation of the complex at higher molar ratio of ligand to metal ion. Results are presented in Table XX and the plot of chemical shift gg molar ratios of ligand to metal is shown in Figure 28. From the data it is clear that the extent of downfield shift for this ligand in DMF and DMSO is less than in pyridine for the same molar ratio range. In pyridine, little change in the chemical shift of 109Ag- resonance was observed on the addition of 2CNPy. For other mentioned substituted pyridines, 109Ag nmr spectra were run in a series of 1 E A8010” solutions with a ligand to metal 111 Table xx. 1°9Ag Chemical Shifts and Line Widths as a Func- tion of 2CNPy/AgClOu Molar Ratio in Different Solvents DMSOa DMFb PYC [2CNPy1/[As+1 Gppm [2CNPyl/[As+] Gppm [2CNPy1/[As+] Gppm l/2 -268.5 0.”l -172.2 1/2 -552.7 1/1 -277.2 0.50 -181.0 1/1 -553.1 2/1 -295.0 0.83 -209.8 2/1 -5”9.6 1.00 -221.3 a’cAbove L/M = 2/1, precipitation took place. bAt L/M = 2/1 molar ratio, precipitation took place. 112 '" I30 I I I I- D M F 200 7' _3 250 (PPM) 300_ DMSO «I 550 PY EYTC)‘ I 1 1 l O 2.0 4.0 11.0 01,, Figure 28. Silver-109 chemical shift as a Agfunction of 2-cyanopyridine/Ag010u molar ratio in various solvents. 113 molar ratio of 1/2 in DMSO. At higher molar ratio of ligand to metal ion, precipitation took place. For the 2,5—dichloropyridine study, tetrahydrofuran was also used as a solvent. Other substituted pyridines were not soluble in this solvent. The results are presented in Table XXI. The data for Py and 2CNPy as ligands are also presented for the sake of comparison. The results indicate that the downfield shift of 109Ag resonance upon complexation is considerably greater for pyridine than for the substituted pyridines. However, for these compounds, a decrease of reactivity toward the Ag+ ion is expected due to withdrawing properties of substituents on the ring. As mentioned earlier, a linear relationship between the logarithm of the complexity constant and the basicity of a series of substituted 63’6“ Therefore, it is possible pyridine has been obtained. to conclude that the higher downfield shift for the pyridine complex is due to a stronger interaction of this ligand with Ag+. The data for 2-cyanopyridine and ”-cyanopyridine indi- cate that the downfield shift for Ag+ ion is greater when it interacts with ”CNPy than 2-cyanopyridine. In this case the higher reactivity of ”-cyanopyridine has been explained62 on the basis of increased m-bonding as a result of structures like the following which involves multiple bonding of the silver ion to the ligand which makes 11” Table XXI. Silver-109 Chemical Shifts of AgClOu + Ligand in THF and DMSO. L/M Solvent 6(ppm) Ligand:l 1/2 DMSO -258.8 2-5DiCle 1/2 DMSO -257.l 2-6DiCle 1/2 DMSO -2au.2 ”CNPy 1/2 DMSO -261.5 2-5DiBrPy 1/2 DMSO -268.5 2CNPy 1/2 DMSO ~308.0 Py 1/2 THF -116.7 Py 1/2 THF - 79.2 2-5DiCle 115 this complex more stable than the one formed with 2-cyano- pyridine. ”. Cgrbon-13 nmr Studies of Complexation of Ag+ Ion with Benzene and Pyridine Carbon-l3 nmr spectra were run on a series of solutions of AgC10u with either pyridine or benzene as ligands in different solvents such as THF, DMSO, and PC. These solutions contained constant concentration of ligand (PY: 0.2117 M, Bz: 0.2”9 M) and different molar ratios of metal to ligand such as 0/1 and ”/1 were tried to find out how the carbon resonance of the ligand's carbons respond to this change. The results are presented in Tables XXII and XXIII. As the data indicate, there is a change in the chemical shift of pyridine's carbon atoms as AgClOu is introduced into the solution. The change is most for the y-carbon and least for the B-carbon. The same trend was observed in protonation of pyridine. For benzene no 116 Table XXII. 130 Chemical Shift as a Function of Ag+/Bz Molar Ratio in Different Solvents. Sample 6(ppm) Ada Ag+/Bz 82 + Ag+ in DMSO 129.91 8.0 +0.110.01 82 in DMSO 129.81 0.0 Bz + Ag+ in PC 128.83 8.0 . -o.9io.15 Bz in PC 129.3” 0.0 82 + Ag+ in THF 128.73 8.0 B2 in THF 128.86 0.0 B2 128.60 a : A6 - [6(sample solution with salt) - 6(sample solution without salt)] 1 change of chemical shift for reference solvent peak in these two solutions. 117 :0 owcmso on» 0 flApHMm psonpaz soapsaom magswmvo I Auawm spa: soapsaom 0005mmva 0 o< .xmoa coconohon no phase Hmoaeono map o.n.m 00.000 00.000 00.000 00 0.0 00.0m0 00.000 00.000 00:0 :0 00 m0.0000.0 00.0000.m m0.0000.0 0.0 00.000 00.000 00.000 0020 s0 +m< + as 0.0 00.000 00.000 00.000 see :0 00 00.0000.N 00.0000.0 00.0000.m . 0.: 00.000 00.000 00.000 see s0 +w< + 00 0.0 00.000 00.000 00.000 00 s0 00 00.0000.0 00.000s.0 00.0000.N 0.: 00.000 00.000 00.000 00 s0 +w< + 00 m h .I& m > s mm\+m< m< M< M< “suave Asgavo Asaave mansom .mpco>Hom econommfin :0 o0pcm 60oz em\+m< 0o so0posss s as 00000 0000soeo 000 .HHHNN manwa 118 change is observed within the experimental error. For the case of benzene Just one 13C-resonance was observed which might be due to fast exchange of Ag+-bonded carbons in the ring. In a report of an x-ray investiga- tion, it was concluded that Ag+ interacts unsymmetrically with benzene being closer to two of the carbons of benzene. It seemed interesting to carry out 130 nmr studies on this system at lower temperatures where the exchange might slow down. The experiment was done for a 1.”” M AgClOu solution with benzene at 8.7/1 molar ratio of Ag+ to benzene at -”7°C in THF. This solvent was tried because of its low melting point (-109°C). No splitting of the carbon- resonance took place which indicates that the rate of ex— change is fast compared to the nmr time scale, or the 13C nmr is not sensitive enough to detect small changes taking place in the chemical environment of the 13C nucleus. It thus appears again that 109Ag nmr is a more sensi- tive technique than the 13C nmr for studies of complexation reactions and, in particular, for weak interactions. Conclusions Exploratory study of the 109Ag nmr showed that this technique has a good possibility of becoming a very useful probe of the reactions of the silver ion in solutions. The low sensitivity of the 109Ag signal, however, made it impossible to work at concentrations below 0.5 M_and 119 since the solubilities of silver salts in nonaqueous sol- vents are rather limited, the upper limit of silver salt concentration seldom was larger than 1.0 M, This narrow (and high) operational concentration range precluded quantitative studies of the silver complexes such as have been carried out previously in this laboratory with 7Li, 23Na and 13303 nmr. Recent acquisition of Bruker-180 multinuclear spectrometer with a high field superconducting solenoid should enlarge considerably the scOpe of 109Ag nmr research. Unfortunately, it arrived too late to be of use in this study. Nevertheless the comparison of the sensitivities of the DA-60 and Bruker-180 spectrometers shown in Figure 29, vividly illustrates the much superior performance of the latter instrument. Of course, the progress in the nmr instrumentation is bound to continue and spectrometers with ultra high fields undoubtedly will become available in the future (albeit at a price!). With such spectrometers it should be pos- Sible to study very dilute solutions of silver salts and complexes in solutions and the application of 109Ag nmr to chemical problems should increase manyfold. 120 Bruker I80 WIII WW IIIIIIIIIIIIIIII IIIII I000 Hz IOOOHz Figure 29. Silver-109 nmr resonances of aqueous solution of 1 M AgClO doped with 1/50 M Fe +3 by DA60 and Bruker l 0 instruments (1000 scans). REFERENCES 1. 10. ll. 12. 13. la. 15. REFERENCES C. T. Horovitz, (Editor) K. A. Gschneidner, Jr., G. A. Melson, D. H. Youngblood and H. H. Schock, "Scandium, Its Occurance, Chemistry, Physics, Metallurgy, Biology and Technology", Academic Press, New York, NY 1975. G. A. Melson and R. W. Stotz, Coordination Chemistry Reviews, 1, 133 (1971). A. I. Popov, "The Use of Alkali Metal Nuclear Magnetic Resonance in the Study of Solvation and Complexation of Alkali Metal Ions", Chapter in Solute-Solvent Inter— actions, Vol. II, J. F. Coetzee and C. D. Ritchie, editors, Marcel Dekker, Inc., New York, NY 1976. W. G. Proctor and F. C. Yu, Physical Review, §1(1), 20 (1951). D. M.)Hunten, Canadian Journal of Physics, 22, R63 1951 . Y. Ting and D. Williams, Physical Review, §g(3) 595 (1953). R. E. Sheriff and D. Williams, Physical Review, §g(5), 651 (1951). 0. Lutz, Physics Letters, 225(2), 58 (1969). Yu. A. Buslaev, V. P. Tarasov, M. N. Buslaeva, and S. P. Petrosyants, Dokl, Akad. Nauk. SSSR, 209, 882 (1973). Yu. A. Buslaev, S. P. Petrosyants, V. P. Tarasov, and V. I. Chagin, Russian Journal of Inorganic Chemistry, 12(7). 975 (197"). G. Ya. Pushkina and L. N. Komissarova, Russian Journal of Inorganic Chemistry, §(6), 777 (1963). G. A. Melson, D. J. Olszanski, and Eric T. Roach, Journal of the Chemical Society Chemical Communications, 229 (197“). A. D. Paul, J. Chem. Phys., 66, 12h8 (1962). D. F. C. Morris, G. L. Reed and K. J. Sutton, J. Inorg. Nucl. Chem. 26, 1H61 (196D). G. L. Reed, K. J. Sutton and D. F. C. Morris, J. Inorg. Nucl. Chem. 26, 1227 (196"). 121 16. 170 18. 19. 20. 21. 22. 23. 2h. 25. 26. 27. 28. 29. 30. 31. 32. 122 D. F. C. Morris, J. C. Green, G. L. Reed, Radiochim. Radioanal. Letters, 1(u), 231 (1969). R. M. Izatt, D. Eatough, J. J. Christensen, and C. H. Bartholemew, J. Chem. Soc. A, h7,(1969). C. J. Pederson, J. Amer. Chem. Soc. g3, 7017 (1967). B. Dietrich, J. M. Lehn, J. P. Sauvage, at J. Blanzat, Tetrahedron 22, 1629 (1973). J. J. Christensen, D. J. Eatough and R. M. Izatt, Chemical Reviews, 1fl(3), 351 (197“). B. Dietrich, J. M. Lehn, et J. P. Sauvage, Tetrahedron, a. 16147 (1973). D. J. Olszanski, Ph.D. Thesis, Michigan State University, 1975. R. W. Stotz and G. A. Melson, Inorg. Chem. 11, 1720 (1972). N. P. Shepelev, J. V. Arkhangel'skii, L. N. Komissarova, and V. M. Shatskii, Russian Journal of Inorganic Chemistry, 16(12), 1706 (1971). I. V. Arkhangel'skii, L. N. Komissarova, V. M. Shatskii and N. P. Shepelev, Russian Journal of Inorganic Chem— istry. 11(2). 163 (1972). L. N. Komissarova, G. Ya. Pushkina and V. I. Spitsyn, Russian Journal of Inorganic Chemistry, §fi6),719 (1963). F. Petru and F. Kutek, 2. Chem. 3, N73 (1963). L. N. Komissarova, V. M. Shatskii and G. I. Moiseichenko, Russian Journal of Inorganic Chemistry 19, AG? (1965). Herzog, S., Gustav, K., Kruger, E. et a1., Z. Chem. 3, A28 (1963). J. L. Atwood and K. D. Smith, J. Chem. Soc. Dalton, 921 (197“). M. T. Rogers, E. T. Roach and A. W. Burkhardt, luth Experimental NMR Conference, University of Colorado, 1973. ‘ R. H. Erlich, M. S. Greenberg, and A. I. Popov, Spectro- chim. Acta. 29A, 5M3 (1973), 33. 3h. 35. 36. 37. 38. 39. “0. H1. H2. “3. an. “5. I6. M7. #8. “9. 50. 51. 123 E. B. Baker, L. W. Burd and G. N. Root, Rev. Sci. Instrum., 36, 1995 (1965). E. B. Baker and L. W. Burd, Rev. Sci. Instrum. 3A, 238 (1963). D. L. Cole, L. D. Rich, J. D. Owen and E. M. Eyring, Inorg. Chem. 6, 682 (1969). L. N. Komissarova, N. M. Prutkova and G. Ya. Puskina, Russian J. Inorg. Chem. 16(7), 95” (1971). L. S. quMel, T. R. Stengle and C. H. Langford, Chem. Commun., 393 (1965). E. G. Bloor and R. D. Kidd, Can. J. Chem. , M6, 3M25 (1968) M. i. Greenberg, Ph.D. Thesis, Michigan State University 197 . (a) V. Gutmann and E. wychera, Inorg. Nucl. Chem. Lett., 2: 257 (1965)- (b) V. Gutmann, "Coordination Chemistry of Nonaqueous Solvents", Springer-Verlag, Vienna, 1968. H. N. Po, Coordination Chem. Rev. 26, 171 (1976). C. D. M. Beverwijk, G. J. M. van.Derkerk, Organometal. Chem. Rev. A 2, 215 (1970). F. R. Hartley, Chem. Rev., 13, (2), 163 (1973). E. Sigleton and A. Tuinman, J. Organometal- Chem. 22; 337 (1975). A. E. H111, J.Amer. Chem. Soc., 53, 1163 (1922). L. W. Daasch, Spectrochim. Acta, 2, 726 (1959). L. J. Andrews,and R. M. Keefer, J. Amer. Chem. Soc. 71, 36h“ (19“9)- M. I. Forman J. Gorton and R. Foster, Trans. Faraday, Soc. 66(Pt.9) 2120 (1970). H. J. Taufen, M. J. Murray and F. F. Cleveland, J. Amer. Chem. Soc. 61, 3500 (19fl1). R. S. Mulliken, J. Amer. Chem. Soc. 13, 811 (1952). H. G. Smith and R. E. Rundle, J. Amer; Chem. Soc. 69, 5075 (1958)- 52. 53. 59. 55. 56. 57. 58. 59. 60. 61. 62. 630 6h. 65. 66. 67. 68. 59. 70. 12“ H. G. Smith, J. Chem. Phys. 36, 2h12 (196“). D. F. R. Gilson and C. A. McDowell, J. Chem. Phys. 12, 1825 (1963). D. F. R. Gilson and C. A. McDowell, J. Chem. Phys. 36, 2&13 (196A). J. C. Schug and R. J. Martin, J. Phys. Chem. 66, 155A (1962). L. J. Andrews, and R. M. Keefer, J. Amer. Chem. Soc. 72, 3113 (1950). L. J. Andrews, and R. M. Keefer, J. Amer. Chem. Soc. 13, 36uh (19h9). L. J. Andrews and R. M. Keefer, J. Amer. Chem. Soc. 1g, 503“ (1950). P. Paoletti, A. Valla, and D. Arenare, J. Phys. Chem. P. Paoletti, A. valla and D. Arenare, Coordination Chem. Rev. 3, 280 (1966). Y. Bando and S. Nagakura, Theor. Chim. Acta, 3, 210 (1968). R. K. Murmann, and F. Basolo, J. Amer. Chem. 300- ll, 3uau (1955). M. S. Sun and D. G. Brewer, Can. J. Chem. 25; 2729 (1957). G. Berthon, 0. Enea, and Y. Fuseau, Thermochimica Acta. .19 323. (1976) A. Narath and A. C. Gossard Phys. Rev., 363(2), 391 (1969). ’ J. Sclyom, Z. Physik, 336, 195 (1970). G. A. Matzkanin, J. J. Spokas, C. H. Sowers D. 0. Vangstenburg, and H. G. Hoene, Phys. Rev. 181, 559 19 9 . R. J. Snodgrass, Phys. Rev. B, 3 (11), 3738 (1971). U. El-Hanany, M. Shaham and D. Zamir, Phys. Rev. B 36, 23fl3 (197“). E. Brun, J. Oeser, H. H. Staub, and C. G. Telschow, Phys. Rev. 33, 172 (195A). 71. 72. 73. 7a. 75. 76. 77. 78. 79. 80. 81. 82. 83. an. 85. 86. 87. 88. 89. 125 P. B.Soygo and C. D. Jeffries, Phys. Rev. 23, 17M (l95h). A. Abragam and J. M. Winter, Compts. Rendus, 2A9, 1633 (1959)- W. Sahm and A. Schwenk, Z. Naturforsch, 22a, 1763 197 . C. W. Burges, R. Koschmieder, W; Sahm and A. Schwenk, Z. Naturforsch, 28a, 1753 (1973)- P. M. Henrichs, G. E. Maciel and J. J. H. Ackerman, Personal Communication. K. Jucker, W. Sahm and A. Schwenk, Z. Natureforsch, la. 1537 (1976). C. Berger and L. R. Dawson, Anal. Chem. 35, 99A (1952). D. D. Traficante, J. A. Simms and M. Mulcay, J. Magn. Reson., 33, usu (197A). D. A. Wright and M. T. Rogers, Rev. Sci. Instrum., 33) 1189 (1973)- J. W. Cooper "An Introduction to Fourier Transform NMR and the Nicolet 1080 Data System", Nicolet Instrument Corp., Madison, Wisconsin, 197“. "OFT-20 NMR Spectrometer System", 87-1flu-000 Varian Instrument Division. "A-56/60D General Information", publication Number 87- 103D-100A268, varian Analytical Instrument Division. "AT-60 NMR Spectrometer System", publication Number 87-121-001, Varian Analytical Division. R. A. Komoski and A. C. Levy,Magnetic Resonance Review, 3, 289 (197“). G. J. Templeman and A. L. VanGeet, J. Amer. Chem. Soc., C. Deverell and R. E. Richards, Mol. Phys. 551 (1966). E. G. Bloor and R. G. Kidd, Can. J. Chem., 66, 3926 (1972). H. Lee and J. K. Wilmshurst, Aust. J. Chem. 31, 9H3 (196“). C. B. Baddiel M. J. Tait, and G. J. Janz, J. Phys. Chem. g2, 363A (1965). 90. 91. 92. 126 G. J. Janz, M. J. Tait, and J. Meier, J. Phys. Chem. 1, 963 (1967). H. Schneider, Solute-Solvent Interactions, Marcel Dekker, N. Y. London, 1969. J. J. H. Ackerman and G. E. Maciel, Personal Communication. IIIIIIIIIIIIIIIIII 4 2 4 7 5 7 1 3 0 3 9 2 1 3 IIIIIIIIIII IIIIIIIIIIII