BASICTTY AND COMPLEXATION PROPERTTES 0F SUBSTITUTED TETRAZOLES Thesis for the Degree of Ph‘ D. MTCHTGAN STATE UNTVERSTTY THOMAS C. WEHMAN 1968 Limit 113;): "i A‘iichigan Stage I u o v . University , 3W4; / THESIS This is to certify that the thesis entitled Basicity and Complexation Properties of Substituted Tetrazoles presented by Thomas C . Wehman has been accepted towards fulfillment of the requirements for Ph.D. Chemistry degree in 5/ 7 1;“ Major professor Date August 8, 1968 0-169 1“ ‘Whhr‘ I i" BIN-BIHG Iv. ‘5‘ HMS & SUNS' a 800K BINBERY INC. ‘ LIIRARY BINDERS ; ABSTRACT BASICITY AND COMPLEXLTION PROPERTIES OF SUBSTITUTED TETRLZOLES By Thomas C. Webman The complexing ability of the tetrasole ring was investigated in order to determine the‘n-doner ability of tetrasoles. It has been shown that 5-substituted and 1,5-disubstituted tetrasoles readily form charge-transfer complexes with “hacids such as tetracyanoethylene, tetracyanoquinodimethane, pychloranil, 1,3,5-trinitrobensene and 2,h,7-trinitrofleurenone. Formation constants of these complexes were measured spectrophotometrically in dichloromethane at 25'. There is a reasonable agreement between the magnitudes of the formation con- stants and the inductive effects of the substituent groups on the tetra- zole ring. The formation constant of the pentamethylenetetrasole complex with 1,3,5-trinitrobensene has also been measured by nuclear magnetic resonance techniques. The proton accepting properties of tetrasole and substituted tet- razoles were studied using the protogenic solvent, formic acid. Conduc- tance measurements made on formic acid solutions show that tetrazele, Sémethyl-, 5-phenyl-,and pentamethylenetetrasole behave as weak bases with pxb values ranging from 1.78 to 2.06. Precise conductance measurements were also carried out at 25' on solutions of twelve lsl electrolytes in anhydrous formic acid, in particular on alkali metal perchlorates, formates, chlorides and bro- mides. It was found that iodides, nitrates, thiocyanates and fluorides 1 Thomas C. Wehman were rapidly solvolyzed by formic acid. Most inorganic salts are completely dissociated in formic acid solution. Hydrogen chloride, however, behaves as a weak electrolyte with a dissociation constant of l.l x 10““. Single-ion limiting conductances were calculated from the conductance data. These data also indicate that the upper limit for the self-ionization constant of formic acid at 25’ should be 2.2 x 10". Preliminary studies of pentamethylenetetrazole (PMT)-transition metal complexes were performed in formic acid to observe the effect of tetrazole protonation upon complex-coordination. The following seven perchlorate hexahydrate salts were used in this investigation: chromium(III), manganese(II), iron(III), iron(II), cobalt(II), nicxel(II), and copper(II). It was found that the coordinating ligands about the transition metal ion are not PMT or protonated PMT molecules, but rather formate ions. Since these investigations involved tetrazoles which could not be obtained commercially, several different syntheses were used in their preparation. The syntheses of thirteen 5-substituted tetrazoles, three 5,§-ditetrazolyls, five 1,3-disubstituted tetrazoles, and three substitut- ed pentamethylenetetrazoles are described. BASICITY AND COMPLEXATION PROPERTIES OF sunsr ITUTED nrsizous By Thomas C. wehman A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1968 ACKNOVLEDGMENTS The author wishes to extend his appreciation to Professor Alexander I. Papov for his guidance, advice and encouragement throughout this study. Thanks are also extended to Professor Paul G. Sears of the University of Kentucky and Dr. Joseph A. Caruso for their suggestions concerning the conductance studies. The helpful discussions with Professors R. M. Herbst and E. Legoff of this Laboratory are gratefully acknowledged. Thanks are also ex- tended to E. Dean Butler for his many helpful comments during this investigation. Appreciation is expressed for the financial assistance obtained from the National Institute of Health for partial support of this investigation. Deep appreciation is given in memory of Sue and Mark for their patience and understanding during the early years of graduate study. To the author's wife, Patricia, goes a very sincere gratitude for her encouragement, inspiration and help in the completion of this work. ii TABLE or coursurs CHAPTER I INTRODUCTION TE Two“ 3 O O O O O O O O O O O O O O O O O O O O O O O O 0 lo General e e e e e e e e e e e e e e e e e e e e e IIe ACId-B‘SG PrOperties e e e e e e e e e e e e e e III. Complexation - Coordination . . . . . . . . . . . OBJECTIVES OF INVESTIGATION . . . . . . . . . . . . . . . . I. II. III. IV. CHAPTER II SYNTHETIC HORK 5-SUBSTITUTED TETRAZOLES . . . . . . . . . e . . . . . . A. Procedure (A) — Aromatic Substituents . . . . . . 3. Procedure (B) -Alkyl Substituents . . . . . . . DITETRAZOLYLS e e e e e e e e e e e e e e e e e e e e e~ l ’5'DISUBSTITUTED mtmoms O O O O O O O O O O O O O O SUBSTITUTED PENTAMETHYLENETETRAZOLES . . . . . . . . . . A. Preparation 1 — Pentamethylenetetrazole . . . . . 1. CYCthexanone MethOd e o o o o e o e e e e e 2. e-CaprOIRCtam MethOd e e o e o o e e e e e e B. Preparation II -l0-Chlor0pentamethylenetetrasole C. Preparation III —.8-Chlor0pentamethylenetetrazole D. Preparation IV -6-Chlor0pentamethylenetetrazole CHAPTER III STUDY OF n-COMPLEXATION OF SUBSTITUTED TETRAZOLES HISTORICAL O O O O O O O O O O O O O O O O O O O O O O O O THEORET ICAL O O O O O O O O O O O O O O O O O O O O O O O 0 iii Page ¢~uamo 10 12 14 14 14 17 18 18 20 26 30 TABLE OF CONTENTS -- Continued EXPERIMENTAL O O O O O O O O O O O O I. II. III. Solvents . Reagents . Apparatus RESULTS AND DISCUSSION . . . . . e . CHAPTER IV STUDIES IN FORM IC H IS TOR ICAL O O O O O O O O O O O O 0 lo General e e e e e e e e e II. Purification e e e e e e e III. Acid-Base Equilibria . . . IV. Electrolytic Conductance . THEORET 1C“! 0 O O O O O O O O O O O I. General Conductance . . . II. Shedlovsky Iteration EXPERMNT“: O O O O O O O O O O O O I. II. III. IV. Solvent . Reagents e Apparatus Procedures RESULTS AND DISCUSSION . . . . e . e I. II. Conductance . . . . . . . CONPIOX‘CIOD e e e e e e e A. B. C. D. Capper(II) Perchlorate Nickel(II) Perchlorate Cobalt(II) Perchlorate Chromium(III) Perchlorate Studi s ACID SOLUTIONS Technique e e e e e e O O O O O O O O 0 O O O O O O O O O O O O O O O O O O O O O I 0 e e e e e e e e e e e e e e e e Studies 0 e e e SCUdIOS e e e 0 Studies a e e e C e e REFERENCES 0 O O O 0 O O O O C O O O O O O O O O O 0 APPENDIX C O O O .D O O O O O O O O O O O O O O O C 0 iv Page 34 34 35 36 38 71 71 72 72 74 76 76 78 81 81 81 83 85 85 108 111 127 133 I37 143 149 TABLE I. II. III. IV. V. VI. VII. VIII. IX. X. XI. XII. XIII. XIV. XVI. XVII. LIST OF TABLES 5-Substituted tetrazoles . . . . . . . . . . . . . . Ditetrazolyls. . . . . . . . . . . . . . . . . . . . N-Substituted amides . . . . . . . . . . . . . . . . 1,5-Disubstituted tetrazoles . . . . . . . . . . . . Substituted pentamethylenetetrazoles . . . . . . . . flbAcids Tetracyanoethylene complexes with tetrazoles . . . . Charge-transfer complex of TCNE with 5-benzy1tetrazole . . Comparison of complex strengths SpectrOphotometric determination of K for the trinitro- benzene-pentamethylenetetrasole complex in carbon tetra- Chloride...OOOOOOOOOOOOOOOOOOO Equivalent conductances in anhydrous formic acid . . Limiting equivalent conductances in anhydrous formic Salts which react with formic acid . . . . . . . . . Limiting equivalent conductances of single ions . .- Basicity constants of some tetrazoles in anhydrous acid solutions.. e o o o o e o o o o o o o o o o o 0 Transition metal spectra in formic acid and water . Transition metal perchlorate spectra in formic acid acid. formic Page 11 13 15 16 24 37 55 57 65 68 86 92 102 104 107 109 110 FIGURE 1. 2. 3. 4. 5. 8. 9. 10. 11. 12. LIST OF FIGURES Infrared absorption spectra of A, PMT using a KBr pellet; B, 6-chloropentamethylenetetrazole using KBr plates . . . . Beer's law study of TCNE in: A, dichloromethane (268 mp); B, 1,2-dichloroethane (263 mu); C, methanol (235 mp); D, nitromethane (278 mp), and E, ethanol (240 mp) using l-cm-pathlength 03118 e o o o o o o o o o o o e o o o o o Beer's law study of: A, TCNQ (400 mm); B, TNF (280 mp); and C, TNB (260 my) in dichloromethane using l-cm-path- length 09118. e o o o o e o o o o o e o o o o o o o o e e o Beer's law plot of pychloranil (373 mm) in dichloromethane using S l-cm-pathlength C011 o e o e o o o o o o o o o o e SpectrOphotometric study of 5.0 x 10‘"5 §_TCNE (A) with PMT mole ratios of B, 1085; and C, 4330, in dichloromethane using a l-cm-pathlength 0911 o o o o e o o o o o o o o o o SpectrOphotometric study of 1.0 x 10"3 §,TCNE with various mole ratios of PMT in dichloromethane using a l-cm-pathlength cell o o o . e o o e o o o o o e o o o e o e o o o o o o e SpectrOphotometric study of 2.0 x 10'3 §_TCNE with various mole ratios of l-cyclohexyl-5-ethyltetrazole in dichloro- methane using a l-cm-pathlength cell . . . . . . . . . . . Spectrophotometric study of 4.0 x 10"3 §.TCNE with various, mole ratios of l-cyclohexyl-5-methyltetrazole in dichloro- methane using a l-cm-pathlength cell. . . . . . . . . . . . Spectrophotometric study of 4.0 x 10'3 ! TONE with various mole ratios of l-methyl-5-cyclohexyltetrazole in dichloro- methane using a l-cm-pathlength C811 o o o o o o o o o e o Spectrophotometric study of 4.0 x 10'3 §,TCNE with various mole ratios of 5-propyltetrazole in dichloromethane using a l—cm-pathlength cell. o o o e o o o o o o o o e o o o o o o SpectrOphotometric study of 1.0 x 10"3 !,TCNE with various mole ratios of 1,5-diphenyltetrazole in dichloromethane using R I—cm-pathlength cell. o o o o o o e o o o o o o o o SpectrOphotometric study of 4.00 x 10"3 g TCNE with various mole ratios of l-ethyl-S-phenyl tetrazole . . . . . . . . . vi Page 23 39 42 44 45 46 47 48 49 50 51 52 LIST OF FIGURES -- Continued Page 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. Spectrophotometric study of 4.0 x 10'3 §|TCNE with various mole ratios of l-phenyl-5-ethyltetrasole in dichloromethane using R I—cm—pathlength cell. o o o o o o o o o o e o e o e o 53 Spectrophotometric study of 5.00 x 10"3 g TCNE with various mole ratios of 5-benzyltetrazole in dichloromethane using a 5-cm-path1ength cell. e o o 0 o o o o o e o o o o o o e o o e 5‘ Ketelaar plots for PMT complexes of: A, TCNQ (283 mp); B, CA (315 up); C, TNB (270 mp); D, TNF (300 our); and E, ’I‘CNE (293 mp) in dichloromethane . . . . . . . . . . . . . . . . . 58 SpectrOphotometric study of e51 x 10'4 §_TCNE' in acetonitrile using a l-cm-pathlength cell. . . . . . . . . . . . . . . . . 59 Spectrophotometric study of 2.5 x 10"5 §_pychloranil with various mole ratios of EMT in dichloromethane using a l-cm- pathlength 0011 o o o o o o e o o o o o o o o o o e e o o o o 61 SpectrOphotometric study of 2.0 x 10’“ §,l,3,5-trinitrobenzene with various mole ratios of PMT in dichloromethane using a l-cm-pathlength cell o o e e o o o o o o o e o o o e e e o o 62 Spectrophotometric study of 5.1 x 10'4 g tetracyanoquinodi- methane with various mole ratios of PMT in dichloromethane using C I-cm-pathlength Cell o o o e o e o o o o e o o o o o 63 SpectrOphotometric study of 2.0 x 10"5 §_2,4,7-trinitro- fluorenone with various mole ratios of PMT in dichloromethane using R I—cm-pathlength Cell e o o o o o o o o o o o o o o o 64 NMR spectroscopic study of TNB proton shift with increasing concentration of EMT in carbon tetrachloride . . . . . . . . 66 Shedlovsky plots: A, sodium chloride; B, sodium perchlorate; C, potassium chloride; D, potassium bromide; 8, sodium formate, infomic‘CId...OOOOOOOOOOOOOOOIOOO...93 Shedlovsky plots: A, lithium chloride; B, sodium bromide; C, hydrochloric acid; D, tetramethylammonium bromide; and E, ammonium chloride, in formic acid . . . . . . . . . . . . 95 Shedlovsky plots: A, triisoamylbutylammonium tetraphenylborate; and B, triisoamylbutylammonium perchlorate, in formic acid. . 97 Shedlovsky plots: A, 5-phenyltetrazole; B, tetrazole; C, 5-methyl tetrazole; and D, pentamethylenetetrazole, in fomicaCid'OOOOOOOOOOOOOOOO‘OOOOOOOO. 99 vii LIST OF FIGURES -- Continued Page 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. Beer's law plots: A, chromium(III) perchlorate hexahydrate (420 mu); 8, capper(II) perchlorate hexahydrate (815 mp) C, cobalt(II) perchlorate hexahydrate (520 mp); and D, nickel(II) perchlorate hexahydrate (397 mp), in formic ‘CIdOOOOOOOOOOOOOOOOOOOOOOOOOCOC 112 Spectrophotometric study of 1.0 x 10'”3 M_Cu(0104)2-6820 with various mole ratios of PMT in formic acid using a lO—cm cell 114 Spectrophotometric study of 1.0 x 10"3 M,Cu(C104)2°6820 with various mole ratios of PMT in formic acid using a 10ecm cell 115 Spectrophotometric study of 1.0 x 10'3 M Cu(C104)°6H20 with various mole ratios of PMT in formic acid using a lO—cm CGll»o o o o o o o o o e o e o o o o o o o o o o e o o . o o 116 Mole ratio study of A, sodium formate; B, PMT with copper(II) perchlorate hexahydrate in formic acid (700 mp) . . . . . . 118 Continuous variation study of capper(II perchlorate hexahy- drate withEA, sodium formate (2.0 x 10' M); B, PMT (4.5 x 10 M) in formic acid using a 1-cm cell . . . . . . .119 Infrared absorption spectra using KBr pellets of capper(II) precipitate from Cu(C104)206H20 solutions of: A, PMT; B, sodium formate; and C, pyridine, in formic acid . . . . . 120 Infrared absorption spectra using KBr pellets of A, Cu(HC00)206H20; B, H0008 and C, H20, in the liquid state . . 121 Spectrophotometric study of 1.0 x 10"3 g Cu(0104)2°6i-l20 with various mole ratios of sodium formate in formic acid using a lO-cm cell e e o o o e o o o o o o o o o o o o o o o o o o 123 Spectrophotometric study of 2. 0 x 10"3 M Cu(0104) 2°6H20 (C) with: A, 8. o x 10-3 M sodium formate; '5', 8. o x io-3M sodium formate; B, 8. o x 10-3 M PMT; D, 8. o x 10-3 M sodium formate + 8. 0 x 10'3 perchloric acid; E, 8. 0 x 10'3_M_PMT + 8.0 x 10'3 M perchloric acid in formic acid using a 5-cm ce11.125 Spectrophotometric study of 1.2 x 10'3 M Cu(C104)2-6HZO (C) with: A, 2.4 x 10'2 M,pyridine; B, 1.2 x 10'1 M_pyridine; D, 2.4 x 10"2 M_pyridine e 2.4 x 10"2 M perchloric acid in formic acid using 5-cm cells . . . . . . . . . . . . . . . . 126 Spectrephotometric study of 5.5 x 10‘”2 M_nickel(II) per- chlorate hexahydrate with various mole ratios of PMT in form- IC CCId using a l-cm-pathlength 0911 o o o o o o o o o o o o 128 Mole ratio study of PMT -nickel(II) perchlorate hexahydrate in formic 361d e e e o o o o o o o o e e o o o o o e o e o o o 130 viii LIST OF FIGURES -- Continued Page 39. 40. 41. 42. 43. 44. 45. 46. 47. Continuous variation study of nickel(II) perchlorate hexa- hydrate - PMT with a total concentration of (2.2 x 10'1 M) in formic acid using a l—cm-pathlength cell . . . . . . . . 131 Infrared absorption spectra using KBr pellets of nickel(II) precipitate from Ni(C104)2'6H20 solutions of: A, PMT; B, sodium formate; and C, pyridine, in formic acid . . . . 132 Spectrophotometric study of 6.0 x 10'”3 M,cobalt(II) per- chlorate hexahydrate with various mole ratios of PMT using 5 cm pathlength C6118 in formic SCIdo o o o o e o o e e e o 134 Molar ratio study of PMT with cobalt(II) perchlorate hexa- hydrate in formic acid o o o o o o o o e o e o o o e e o o 135 Continuous variation study of cobalt(II) perchlorate hexahy- drate - PMT with a total concentration of (7.5 x 10"2 M) in formic acid using a l-cm-pathlength cell . . . . . . . . 136 Infrared absorption spectra using KBr pellets of cobalt(II) precipitate from Co(C104)2'6H20 solutions of: A, sodium formate, and B, pyridine in formic acid . . . . . . . . . c 138 SpectrOphotomotric study of 2.0 x 10"3 M Cr(C104)3'6820 (C) with: A, 2.0 x 10'1 i_i_ m; B,‘ 4.0 x 10:2 M PMT; D, 4.0 x 10-2 g PMT 4- 4.0 x 10»2 perchloric acid in formic acid "Bing ‘ 5.0-cm cell o o o o o o o o o e o o e e o o o o o e 139 SpectrOphotometric study of 2.0 x 10'3 i_i Cr(C104)3°6H20 (c) with: A, 2.0 x 10"1 i_i sodium formate; B, 4.0 x 10'2! sodium formate; D, 4.0 x 10'”2 M sodium formate «t- 4.0 x 10"2 M perchloric acid in formic acid using 5-cm cells . . . . . . 140 Mole ratio study of A, sodium formate; B, PMT, with chrom- ium(III) perchlorate hexahydrate in formic acid.. . . . . . 141 ix CHAPTER I INTRODUCTION TETRAZOLES I. General Substituted tetrazoles have been the object of numerous inves- tigations because of their interesting physiological and physiochemical prOperties. Since the first tetrazole was prepared in 1885 by the Swedish chemist, J. A. Bladin (1), over 400 members of this class of nitrogen heterocycles have been synthesized and characterized (3). The structures 1, II and III represent S-monosubstituted and 1,5-disubstituted tetrazoles of particular interest. (5) (1) (5) (1) R c N H R c N R (4)N N(2) (4)N N(2) \ / N N (3) (8) (3) CH2 (1) ,/’/' \x\ (11) (7)ca2 fflz(9) (6)082 faz(10) (5)fi r(l) (4)11 /N(2) \\\\\N’/// (3) (III) Pentamethylenotetrazole (PMT), structure III, represents a special case of the 1,5-disubstituted tetrazoles. In this structure a carbon and a nitrogen of the totrazole ring are connected by a five membered methyl- ene chain forming a fused ring system. The pharmacological and physio- chemical properties are dependent upon the substituent groups on the tetrazole ring or the groups on the methylene chain of PMT. 3 In a series of studies, Gross and Featherstone (2) have shown that the neurological activity of these tetrazoles range from strong stimulation of the nervous system to a depressant action. It is conceivable that the neurological activity is related directly to the physiochemical prop- erties of the tetrazoles. II. Acid-Base Properties The most comprehensive review of tetrazole chemistry to date was published in 1947 by F. R. Benson (3). A discussion of 2-substituted, 1-substituted, and 2,5-disubstituted derivatives is also included in this review. The S-monosubstituted tetrazoles were first shown to behave as weak acids in aqueous solution by Olivera-Mandala (4). He states that their acid strength is approximately equal to that of acetic acid (Kama x 10's). Later investigations by Herbst and co-workers (5) were in fair agreement with Mandala's work. However, due to variation of the solvent mixtures in which pxa values were determined, the relative acid strengths of the tetrazoles are in question. In a recent publication by Caruso 35.51.. (6) the effect of substituent groups on 5-monosubstituted tetrazoles was studied by conductance methods. The basic solvent, 1,1,3,3-tetramethylguan- idine, was used in order to enhance the acidity of the tetrazoles. Be- cause of the low dielectric constant of this solvent (p - 11.00 at 25') extensive ion-pair formation occured. As a consequence, the resulting acid-base equalibria‘were complex and only the overall dissociation con- stant could be calculated. The inductive effect of the substituent groups on the acidity of the 5-substituted tetrazoles is demonstrated by a linear Taft plot. l. The disubstituted tetrazoles would be expected to have some proton affinity since all four nitrogen atoms in the ring contain an electron pair available for coordination. However, neither Dieter (7) nor Zwikker (8) observed any basic character in aqueous PMT solutions. when PMT is dissolved in a strongly protogenic solvent,such as acetic acid, it behaves as a weak monOprotic base and can be titrated with perchloric acid (9). The basicity of PMT and other substituted tetrazoles is further accentuated by the stronger protogenic medium, formic acid (21). III. 025213555ion-Coordination Since the tetrazole ring itself has nucleOphilic preperties it is able to form moderately strong complexes with a number of Lewis acids. Various complexes have been reported with halogens (28) and transition metal ions (39). Also, salt-like compounds have been prepared with 5-substituted tetrazoles by simple neutralization with relatively strong bases (10). The first silver and copper salts of tetrazole and 5-substituted tetrazoles were prepared by Bladin (11). Recently, transition metal complexes with l-substituted and 5-substituted tetrazoles have been re- ported by Brubaker and co-workers (12, 13, 14). Brubaker and Daugherty (13) have suggested that the bonding in these complexes may occure either through one or two nitrogens of the tetrazole ring or that the central metal ion may coordinate to the flaelectron system of the tetrazole ring. The last alternative was originally favored by Jonassen and co-workers (15) in their preparation of microcrystalline iron (11) complexes. However, in a more recent study of transition metal complexes, Jonassen (l6) cites stronger evidence for a M5" band from a nitrogen to the metal ion. The first crystalline halogen coordination compound to be reported was the iodine monochloride-PMT charge-transfer complex (40, 41). This 5 complex was studied along with several other iodine monochloride complexes using infrared techniques by Person g£_gl.(40). They noted systematic changes in infrared spectra of the iodine monochloride stretching vibra- tions as the halogen formed complexes of increasing stability. Using this relationship it was concluded that the donor properties of PMT were slightly weaker than those of pyridine. This study was expanded by PapOV'g§_§;, (41) to include iodine and iodine monobromide complexes. The formation con- stants of these halogen-PMT complexes were determined spectrophotometrically in carbon tetrachloride solution. Although solid iodine monobromide-PMT and iodine-PMT complexes could not be obtained,-their stoichiometry in solution was shown to be the same as the 1:1 iodine monochloride-PMT complex. In a later paper by Vaughn gt £1. (42) the formation constants of iodine monochloride complexes of 7-methy1, 8-gggebutyl, and 8-gybutyl PMT were also determined spectrOphotometrically. While solid compounds could not be prepared, the complexes were all shown to be more stable than the unsubstituted PMT complex. Recently, two independent structure determinations have been carried out on the iodine monochloride-PMT complex by the use of x-ray crystallo- graphic techniques (43). The two determinations are in complete agreement with regard to the overall structure and specific parameters. The PMT acts as a unidentate ligand with the iodine of the halogen molecule bound to 4-nitrogen of the tetrazole ring. The nitrogen-iodine monochloride group is linear and coplanar with the flat tetrazole ring, while the sevenumembered methylene ring of PMT is in a chair conformation. There have been several reported preparations of PMT-bmetal complexes(9) in acetonitrile solutions. Although all of these complexes had the same general formula, (PMT)214’, and approximately the same stability constants, 6 only the unsubstituted PMT-silver nitrate complex could be prepared in solid form. Polarographic studies also showed that the cadmiumfithallium-, and cobalt-PMT complexes in aqueous solutions were almost completely dissociated. In a recent series of papers by Papov and co-workers (39, 62), anhydrous transition metal complexes of PMT have been prepared and char- acterized. These complexes were prepared with iron(II), manganese(II), cobalt(II), nickel(II), and zinc(II) by treating PMT with the respective hexaaquo transition metal perchlorates in 2,2-dimethoxypr0pane solutions. The composition and structure of the complexes have been shown to be MII(PMT)6(0104)2 in an octahedral configuration. Electron spin resonance spectra of several of these complexes show that the metal-ligand bonds are highly ionic with only small distortions from the octahedral symmetry. OBJECTIVES OF INVESTIGATION Uith the exception of the iodine monochloride-PMT complex (43), the manner in which the tetrazole ligand is bonded to other Lewis acids is not clear (13, 16). It seemed to us that an investigation of the elec- tron donor properties of the tetrazole ring would indicate the bonding nature of these complexes. Thus, a study of possible complex formation between various tetrazoles and Lewis'n-acids should indicate the extent of the fibelectron donor ability of the tetrazole ring. Chapter III of this thesis, then, deals with the complexes between substituted tetrazoles and a few strong'n-electron acceptors. Complexation between several tetra- zoles and tetracyanoethylene were studied since it is one of the strongest fl;acids known (44). Also, for comparison, complexes with other‘n—acids such as tetracyanoquincdimethane (45); pychloranil (46); 1,3,5-trinitro- benzene (46, 48); and 2,4,7-trinitrofluorenone (47) were investigated. The proton accepting properties of tetrazole and substituted tetrazoles may be studied using a protogenic solvent to enhance their basicity. An- hydrous formic acid was chosen as the solvent in this study since it's properties remain largely unexplored. Thus, in Chapter IV, preliminary studies using this acidic medium were performed on PMT-transition metal complexes in order to observe the effect of tetrazole protonation upon complex-coordination. Also, described in Chapter IV, are conductance measurements which were carried out on solutions of 1:1 electrolytes and tetrazole derivatives. Since the above investigations all involved tetrazoles which could not be obtained commercially, many had to be synthesized using the methods described in Chapter II of this thesis. CHAPTER II SYNTHETIC WORK With the exception of pentamethylenetetrasole, all of the tetra- zoles used in this investigation were synthesized by the following methods: I. S-Substituted Tetrazoles .;.:Procedure (A)— ‘Aromatic Substituents These tetrazoles were obtained by treating.sodiul aside with the corresponding nitriles following the general reaction (27): LiCl or NHACI N \N/ The reactions were allowed to proceed in dinethylformamide for 20 to RCN e NaN3 *: R in Dimethylformamide 120' 2:0 40 hours using either lithium chloride or ammonium chloride as a Lewis acid catalyst. Since the synthesised tetrazoles were very insoluble- in aqueous solution, they could be easily recovered from the reaction mixture by the addition of a dilute mineral acid. The crude product was filtered from the solution and purified by several recrystalli- nations. 3. Procedure (32 -- Alkyl Substituents Because of appreciable solubility of the 5-alkyl tetrazoles in aqueous solution, Procedure (A) could not be used for their prepara- tion. .A method described by Mihina and Herbst (38) which makes use of 10 a sealed tube reaction was then successfully employed. Benzene RCN e HN3 >: R-—-—-C-———-N-———-H 5 days 140' H | N\\’//N N After the reaction was completed, the sealed tubes were opened and their contents poured into an evaporating dish. The mixture was then brought to dryness in vacuo, and, finally, the crude product was purified by sublimation at reduced pressure. The synthesised compounds and their properties are recorded in Table I. The two new compounds which had not been previously reported were sent out for elemental analysis. Anal. Calcd fer Ste-Chlorobensyltetrazolex C, 49.34; H, 3.62; N, 28.80. Found: C, 49.08; H, 3.61; N, 28.70. Anal. Calcd for 5- -Methy1bensy1tetrasole: C, 62.07; H, 5.78; N, 32.15. Found: C, 6 .304 H, 5.378 N, 32.10e II. Ditetrasolzls The ditetrasolylssl,4, bis(5-tetrasolyl)—butane and 1,3 bis (S-tetrasolyl) propane were prepared by procedure (A) above using adiponitrile and glutaronitrile respectively. LiCl or N8481 ON (0112),“ 4- Nafl3 > Dimethylfornamide 120‘ H N+—* N H l i \V \/ Where: X I 1, 2’ 3.e e e 4°32}; 2—0 z—o 11 .couueauunsml.uasm ”onenueouounoueuu.filuuouu «noueaeun aunsou>eaa yonlmmz ”suspend oclmzs .snsm .au: eu a: amz mz nu A~ H T r i T-—— N 37 Dimethylformamide 120° V "\V \f The results are listed in Table II. III. 1 S-Disubstituted Tetrazoles The overall synthesis involved a two step procedure (29). /0 (I) (II) _ PCl R HN R c/ NHRI- 5 > [ \CZNRI] 3 Ar Benzene 10° C{ Benzene 20° (a) (b) “ IT T “1 N\\‘//N N (c) Ste I The starting amides (a), could either be purchased commercially or obtained by reacting an acid chloride with the apprOpriate amine. O Benzene 25° 0 / R—cf e ulna ; R c/ NRIH + Pyridine-HCI Cl Pyridine (a) The pyridine hydrochloride was filtered off, the solvent was evaporated 13 Table II. Ditetrazolyls. Recrys. Compound mp Lit. mp 1 Yield Solvent 5,51-Ditetrasolyl 255-0 2540s (3) 42: 952 Ethanol 1,3-bis-(5-Tetrazoly1)- 197-200. Not prev. prep. 94% 95% Ethanol Prepanej l,4-bis-(5-Tetrasolyl)- 205-206. 205-205'0 (27) 89% 95% Ethanol Butane, The following compounds were sent out for elemental analysis: Anal. Calcd for 1,3-bis-(5-Tetrazolyl)-pr0pane: C, 33.33; H, 4.48; N, 62.20. Found: C, 33.31; H, 4.50; N, 62.21. Anal. Calcd for 1,4—bis-(5-Tetrazolyl)-butane: C, 37.11; H, 5.19: N, 57e70. Found: C, 37e103 H, 5e253 N, 57e90e 14 and the crude amide was purified by recrystallization. Ste 11 The imide chlorides (b) were never isolated since upon comple- tion of reaction (I), hydrosoic acid was immediately added to the mixture. Solutions of hydrosoic acid in benzene were prepared by reacting sulfuric acid with sodium aside in benzene.(30). The final tetrazole was recovered and purified by repeated extrac- tions and recrystallizations of the reaction mixture after step (II) was completed. The amides used in the syntheses are listed in Table III while the final tetrazoles are listed in Table IV. IV. Substituted Pentamethylenetetrasoles Although many alkyl-substituted pentamethylenetetrasoles have been reported and characterized, (31, 32) no halogenated derivatives have been prepared. The following methods were used in an attempt to pre- pare monochlorinated PMT. A. Preparation I-—-Pentamethylenetetrasole In order to test several possible procedures for the synthesis of monochlorinated derivatives, the following methods were used to prepare unsubstituted PMT. l. Cyclohexanone Method (33) C32 c CH CH ‘\\ba 2 2 2 / K (I) / \ (II) J | CH2 cnz NHZOH ‘ sz 1H2 NaN3 4). r2 $82 CH2 CH2 HzO-CH3CHZOH CH2 CH2 ClSO3OH; c N ‘ cn2c1cnzc1 u I ii“ (I: /" ‘3 L-—on \\\N (a) (b) The overall synthesis involves a two step procedure. The purchased start- ing material, cyclohexanone (a), was converted into an oxime (b) by 15 eousom Hedoueaeoo confidueOduhusmLm.z anon nomad ahaoum_m.+ moundsoaaoum ooze-a a. e.u= -- -- Ros .eauoseu ascosaoum -mrsaaoum-z -- .eoa .~os -- oouaom Hesouoaaoo oesssauaoaaoum I... mfiod mom d I: wousom Hum “Yugo 00m .— «Gian—0m _ ec—adumnuu e nouns-ao:aeuaz .n.me .mo-ao use .eauoseoamoaaou oeaaeuaoaaseuu-z uaa>aom as .usu ea. use.» a :osuaueaoua oesa< .uuhuoem .aoeaa< eouausuueamuz .aua oases i 6 I aosuasaauuan auuosuoaua Avon—auno ado oases—V any zaseuoa-n-ssuam-s ceases—sumac aasouuouum Acoasauao use cascade Ase . nauseoun-aaaoua-a eseuaonum -ouuuooe manna AoNV .oe .oe aha Ame -aasuo-n-ssaoem-s u.».3.~ocunuoz sous .nes-eea .oe_-ees «so “No asacoeasa-n.a assuaon-m “say .--on .momse and Add zasceea-nuassuuua .uneenom ea .uuq as cued» a ended wsuuueum euouauuea .uuhuoem ||| |l .aoaoaeuuou eouauauaesuse.n.s .>H manna treating it with hydroxylamine. 17 The cyclohexanone oxime (b) was separated from the react ion mixture by filtration and was recrystallized from an ethanol-water mixed solvent before it was used in reaction (II) up 88"” lit. ‘9 9°. (34)e Step (II) involved the slow addition of an 1,2-dichloroethane solution of the oxime to a mixture of sodium aside and chlorosulfonic acid in the same solvent. After the reaction was completed, the mixture was neutral ised with aqueous sodium hydroxide solution. V8000 e The 1,2-dichloroethane layer was separated and dried .19. The final product was obtained from the remaining residue and purified by repeated extractions and recrystallisations with ethyl ether. 2. e-Caprolactam Method This synthesis involves a two step procedure which is similar to the preparation of 1,5-disubstituted tetrazoles described above. /0 /CH2—' i312 er C112 C112 (3) (I) PC15 Benzene 10' CH2 (II) (:82 \c‘n2 HN J CH 3 {2 r2 Benzene N 20'-80' I] L N The starting material, 6 -caprolactam (a), which may be purchased commercially, is a cyclic amide in contrast to the straight chain amides encountered previously. described for the 1,5-disubstituted tetrazoles. The reaction conditions were the same as those The final product was separated and purified by repeated extractions with ethyl ether and subl imat ion. 18 B. {generation II-—-l0-ChlorOpentamethylenetetrazole The same general method described in the cyclohexanene - PMT pro- cedure was used for this synthesis. following two step procedure: CH /’/’ *z\\ cnz CH2 CH2 mm | o (a) (I) mazes cu Nzo-cn3cazoa cu --0H 6%:0 2 t I 2 The overall synthesis involved the .caz { \ca (II) T 2 \2 NaN3 on, cac1 _____) ‘ cnzc1cuzc1 11 ———-T N \N/" (Major product) The starting material, 2-chlorocyclohexanone (a), was purchased from commercial sources. ‘After reaction (I) was completed, the 2-chlorocyclohexanone oxime (b), was recovered but could not be purif- ied. However, the oxime was used in its' impure form for reaction (11). After reaction (II) was completed, the mixture was treated in the same fashion as described for the cyclohexanone-PMT preparation. Unfor- tunately, only a few grams of black oily material were obtained which could not be purified or identified. C. Preparation lII- 8-Chloropentamethylenetetrasole The same general method described in the preparation of 'IO-chloro- Dentamethylenetetrazole above was used for this synthesis. Cl (0.) $1 CH (1) N820}! A c112 c112 61 ca (11) / \ NaN3 (132 c‘az cu c1cn c117 cu ca c1 03an U \ 2 N II I N N 19 Since 4-chlorocyclohexanone (a) could not be purchased from commercial sources, the following procedure was used in its preparation (35, 36). 0H Cl Cl I I I C CH CH CH2 CH2 Sealed Tube CH2 fflz KZCr207 082 fflz #; ————) CH C32 80'-90'; CH CH B so H CH 2 2 2 a. a 2 2 Fuming HCl 2 \K \c/ \c/ I | H OH OH O (a) (b) (0) Commercial cyclohexane - 1:4 diol (a) was placed in a sealed tube with fuming hydrochloric acid at 90' for 24 hours. (After reac- tion (I) was completed the h-chlorocyclohexanol (b) was separated by fractional distillation under reduced pressure with a yield of 56%, 16' 16' n 2,1,49603 lit. (35) 3_ 2.1.4964. This material was then ox- idised to h-chlorocyclohexanone (c) by an acidic potassium dichromate solution, the final product being obtained by fractional distillation under reduced pressure with a yield of 14%, n21 '2 1.4878; lit, (36) 321.2.l.4867. .Upon separation, the pure material was used to form h-chlorocyclohexone oxime in Step I of the 8-chlor0pentamethylenetetrasole synthesis. Although the oxime could not be purified, it still was used in the final reaction (II). After the final reaction (II) was completed, the end product was obtained by repeated extractions and a vacuum sublimation. Only a trace amount of an impure brown oil was recovered from the reaction mixture. However, an infrared spectrum of this material indicates that there is a tetrazole ring present because of the characteristic absorp- tion in the optto.12}lregion. Also, strong absorption in the 13.3;1to 14.3;Lregion indicates an aliphatic carbon-chlorine stretching vibration. 20 No further characterization of the final product could be made. D. Preparation IV- 6-Chlor0pentamethylenetetraeole The same general method described in the e -caprolactam preparation of PMT was used for this synthesis. Cl 0 'Cl Cq \H—C/ \CHf-C/ CH2 / \ . m / \\ (I!) / \ cnz NH P615 cnz N 8N3 932 ca2 LB CH 3 10:» LB 3; B ‘ki CL 2 /// 2 ensene 2 2 enzene ‘ 2 2 20'-80' \”“§ L \“5 - :l 1f N N Sincecz- chloro-Ccaprolactam (a), could not be purchased commercially, it was prepared by the following three step procedure (37): 1° (1) <0 (II) //Cfiz-13 0 CH§—» C) T82 ?H C6HSC-Cl 4; H2 \NJC6H5 $02612 ‘ C§&\b C82 N,N,Dimethylaniline igg\~ //’AH2 Cyclohexane 60' csz/ on (a) (b)2 C1 0 Cl 0 \CH—/ 0 (III) /\CH—C{/ ca/2 \Ngfia5 a A “—" C82 32 40' H2 CH2 CH 'C (c)2 (c3 Ste 1 - Preparation of N-BenzoyleE-Caprolactam (b) To a mixture of 90.4 g (0.80 mole) ofe -caprolactam and 114 ml (0.88 mole) of N,N-dimethylaniline was slowly added 102 ml (0.88 mole) 21 of benzoyl chloride. The mixture was heated and stirred at 90‘ for four hours, cooled to 60’, and poured into 800 ml of dilute hydrochloric acid (0.05 molar). The crystalline lumps were broken up, filtered, air dried and recrystallized from a methanol-water solution giving 146 g (84%): mp 70.0-70.5‘; lit. mp 68-70.59 (37). Step (I!) - Preparation of N-Bensoyl-opChloroqs-Caprolactam (c) To a suspension of 87 g (0.4 mole) of N-benzoyl-e-caprolactam (b) in 22 ml of carbon tetrachloride and 66 ml of cyclohexane was added 34 ml (0.42 mole) of sulfuryl chloride. The mixture was heated and stirred at 40' for 36 hours. The material was then evaporated to dryness in vacuo and recrystallized from isOpropyl alcohol giving 64 g (63%): mp 122-123'; lit. mp 120-121.5' (37). Step (III) - Preparation of a-Chloro-€-Caprolactam (d) To 90 ml of concentrated sulfuric ac id was added, in portions, 64 g (0.26 mole) of N-benzoyl-a-chloro-e-caprolactam (c). The mixture was heated and stirred at 50' for three hours, cooled to 25' and poured onto 500 g of ice. After neutralization with concentrated ammonium hydroxide solution, the mixture was extracted with three 250-m1 por- tions of chloroform. The chloroform solution was evaporated, and the residue was recrystallized from petroleum ether giving 33 g (88%): mp 90-92‘; lit. mp 91.5-93‘. The6 -chlor0pentamethylenetetrazole was then prepared in the following manner: To a mixture of 140 ml of dry benzene and 29.6 g (0.20 mole) of u-chloro-C -caprolactam (d) was added 41.7 g (0.20 mole) of phosphorus pentachloride over a period of one hour. The temper- ature was maintained at 15' by use of an ice bath. After all of the phosphorus pentachloride was dissolved, 360 ml of 0.67.n_hydrazoic acid (0.24 mole) in benzene was slowly added. The mixture was allowed 22 to stand overnight at room temperature and was then refluxed at 80' for two hours. The yellow solution was evaporated to dryness in vacuo, redissolved with 50 ml water and again brought to dryness. The residue was extracted several times with chloroform and a light brown oil was obtained. The final product, a clear viscous liquid, was obtained by frac- tional distillation of the oily residue at reduced pressure (70., 0.2 mm pressure). Several attempts were made to crystallize this material by seeding and freezing, but,apparentlybit is a liquid at room temperature. The boiling point at atmospheric pressure could not be determined because of extensive decomposition of the tetrazole at temperatures above 150’. The final yield was 5 g (14%): A§_1. Calcd for6»-chloropentamethylenetetrasolez C, 41.75; H, 5.26; N, 32.46; Cl, 20.54. Found: C, 41.86; H, 5.36; N, 32.12; Cl, 20.60. The ir spectrum of 6~chlor0pentamethylenetetrasole along with a spectrum of unsubstituted pentamethylenetetrazole may be seen in Figure I. A comparison of the two spectra show the similarity in the tetrasole region (6p-12p), while only in Spectrum (8) are there strong absorptions at 13.3p to 14.3p, presumably from a carbon-chlorine stretching vibration. Although mun-spectra show small amounts of impurities in the final pro- duct, they are in essential agreement with the 6-chloropentamethy1ene- tetrazole structure. A summary of all the results may be seen in Table Y. 23 ,V WA - c-C’ AA W - c- N" I! “I, ”\N’ 4 v T l ' 6 8 10 12 14 3 Figure 1. Infrared absorption spectra of A, PMT using a KBr pellet; B, 6-ch1oropentamethylenetetrasole using KBr plates. (wavelength in p) 24 cozuomuuunmv venue: ~32 uouum .. 1 mi 532083.. e c.2206 bonus: :3 2...: Sam 1. .i m 3am 832308 azmaoé venue: ; souuoaumcsm nu nu uosvouo oz scoxezozozu 92m~0noH E a; vofi a: motum “atom soo soo vouoaouuu oz aduouuouosouw 92m muouh monuoz menuu wanna soc eon veuoaeuuo oz ecoxeso~Oho 92m ume>~om .uuzuoez ea .uuq ea cued» R uuuezumam odouduuos codenamuausm if: I“ .meuowuuuouecefizzueldudom meusuuuucsm .> ounua CHAPTER III STUDY OF fl-COMPLEXATION OF SUBSTITUTED TETRAZOLES 25 HISTORICAL In recent years there have been substantial developments in the investigation of molecular complexes, particularly for those that can be studied spectrophotometrically. These coordination compounds have stabilities which range from the very stable, predominately ionic complexes such as those found between ammonia and boron trifluoride (83) to the weak association complexes of solvent molecules found in liquid benzene and nitroanaline (84). These complexes are often referred to as donor-acceptor complexes with the donor molecules be- ing grouped into two categories (85); ”n" donors, such as alcohols, organic sulfides, and nitrogen bases which donate a non-bonded electron pair and, secondly, ”n” donors such as alkenes, alkynes and aromatic hydrocarbons which contain‘n-molecular orbitals. The accep- tor component of the complexes, of which there are numerous types, include metal ions, halides (n- or'n- bonding), and niacids. The nature of the intermolecular forces between the components of a molecular complex is a matter of considerable controversy. The heats of formation are generally small, and there is abundant evi- dence that the bonding forces are smaller than those of covalent bonds (85). The equilibria involved in complex formations are generally so rapio that kinetic studies can not be made by ordinary procedures (110). Bennett and Willis (86) were the first to nonethat the bonding forces in these complexes arose from some type of electron donor-acceptor interaction and proposed the actual formation of covalent bonds. How- ever, it has since been shown that the bond distances are far greater than those corresponding to covalent bonds (87), but still less than those expected from Van der Hall's forces alone (96). 26 27 Pfeiffer (88), upon observing complexes of aromatic hydrocar- bons with inorganic and organic compounds, assumed that the complex bond arose as a result of unsaturated secondary valences present in the aromatic ring. 0n the other hand, Lewis (89) treated these complexes in terms of acid-base theory and proposed that bonding occures by a base donating an electron pair to the recipient acid. Briegleb and Schachowskoy (90) observed that the heats of formation for 1:1 complexes of nitro-compounds' with aromatic hydro- carbons decreased with a decrease in the polarizability of the hydrocarbons. The above trend in the heats of formation led Briegleb (91, 92) to postulate that these complexes are formed as a result of electrostatic attraction between molecules with permanent dipoles and non-polar species which can be polarized by induction. However, this theory does not explain complexation involving substances lacking even partial dipole movements such as those found in iodine- benzene complexes. It has long been noted that marked spectralchanges accompany complex formation, especially in the case of aromatic complexes (93). Hammick and Yule (94) and Gibson and Loeffler (95) suggest that these spectralshifts arrise from electron transfer between the complex components during normal collisions if the molecules are preperly oriented. Orgel and Mulliken (97) later expanded upon this work showing that in some cases spectralshifts do indeed arrise from short lived collision interactions which they called "contact charge- transfer" transitions. Weiss (98) postulated that spectralshifts arrise by the forma- tion of singly charged ions from the donor molecule (0:), and the acceptor (A), which are then held together by electrostatic forces, i=3. 28 D: 4- A—>[D.]‘[A5]' He suggested that the stability of the complex should be dependent on the ionization potential of the donor and the electron affinity of the acceptor. However, since the heats of complex formation are generally too low for ion formation (99), Brackmann (100), proposed that a complex formed from an electron-transfer is actually a resonance hybrid of a non-bonded and a dative bonded structure. The most widely accepted theory of charge-transfer complexation is that preposed by Mulliken (101). Hulliken considers a 1:1 donor- acceptor complex to have a stable ground state,‘f';, which is made up of contributions from a no-bond state,‘f’o (DA), and a dative polar state, Wfl (D*,.A'), and is given by the expression: 9’; - W0 (DA) + H’, (1)”. A‘). 1. The complex in the excited state,fl3{, is: ‘1’; . v, (9*.1') + bi}; (DA) 2. The coefficients a and b, which are approximately equal, are generally small when compared to unity.’ The charge-transfer absorption band of the complex arrises from the transition Y’3-?‘f;. The energy of this transition is related to the ionization potential of the donor (D) and the electron affinity of the acceptor (A). Several semi-emperical relationships relating the charge-transfer absorption frequency (2%E) to the donor ionization potential (In) and the electron affinity of the acceptor (EA) have been proposed (102-108). McConnell, Ham and Platt (102) found that the equation {'th "IP‘EA'V. 3. where H - dissociation energy of charge-transfer excited state fits 29 the data they obtained for a series of iodine complexes and a plot of “$2: Iggggg Ip of the donors yields a straight line. Halkley g£_gl. (107) expanded these studies to include iodine complexes of "n” donor and ”n” donor solvents obtaining similar linear relation- ships. Linear relationships between “EL: and donor Ip have also been reported for aromatiodn;acid complexes of‘grchloranil (46), “1,3,5-trinitrobenzene (46, 48, 103-105), and 2,4,7-trinitrofluorenone (103-105). The equation suggested by Merrifield and Phillip (44), to fit the data they obtained from tetracyanoethylene-methylbenzene complexes, :19“ - 0.487 1p - 1.30 a. was later expanded to include other aromatics by Voigt and Reid (106). Mulliken and Person have recently published a very good review of charge-transfer complexes and related calculations (108). THEORETICAL The basic equation used in quantitative spectrophotometric cal- culation is Beer's law, A" III 95C 5, where absorbance at a given wavelength molar absorptivity at the same wavelength pathlength in centimeters concentration in moles per liter. own» The importance of this relationship lies in the fact that at a given wavelength and pathlength the absorbance (A) is directly proportional to the concentration of the absorbing species. In order to spectrophotometrically calculate equilibrium constants, or more accurately equilibrium quotients, since activity coefficients are not used, the method originally reported by Benesi and Hildebrand (111) or some variation (115) of this procedure is generally used. The method was originally derived for 1:1 complexes between iodine and organic solvents, but it can be altered to include complexes of different stoichiometries. For the donor-acceptor complex (DA) the equilibrium constant K is given by: D+A:DA (mu) (CD-cm) (CA'CDA) where CD - initial concentration of donor CA.- initial concentration of acceptor CDAF concentration of complex at equilibrium. 30 31 let CD>>CA; thus CD>>CDA and C K - . DA 7. CD(CA-CD") from equation 5: %-MA-P4m .m. . If a wavelength is chosen such that - 0 then by substituting Equation 7 into Equation 6 and rearranging we obtain the final equation: 91 1 1 zT-ce * DKDA 9. em where A.r - total a sorbance of solution at same wavelength. 1 C .A plot of [Ct] versus [CD for a series of solutions gives a straight line with a slope of [IIKCDJ and an intercept of l ,8 . In cases where 5‘ l 0, a new equation, which is similar to Equation 9, can be derived (112); l l l l I e 0 e loo . - c . st 6; em $1 “a m A where ‘1: Ct - —— x —— for a weak complex. (CA 4 CDA) CA 32 A plot of (l/ et-CA) m (l/CD) for a series of solutions gives a straight line with a slope of (USDA'EA 0 UK) and an intercept of (ll 6 DA" 6 A)' Person (113) has shown that these equations give linear plots with meaningful slopes and intercepts only if the donor concentration (CD) is greater than 0.1 (l/K). Also, Johnson and Bowen (114) have pointed out, by the use of synthetic data for the complex systems DA, DzA and DAZ, that a linear plot from the Benesi-Hildebrand and Ketlelaar equations (Equations ‘9 and 10) is not always indicative of the complex stoichiometry assumed in the original derivation. The true test of complex stoichiometry is the agreement between calculated equilibrium constants at several different wavelengths. Recently, a new method for the determination of complex formation constants in solutions using nuclear magnetic resonance spectroscopy has been proposed by Hanna and Ashbaugh (116). They have shown that for a 1:1 charge-transfer complex, the chemical shift of the acceptor protons is related to the strength of the donor-acceptor interaction. They derived a relationship analogous to that of Benesi and Hildebrand (Equation .9) l l l —. a.- __ KQCD A0 __ 11. 25 where A is the difference between the observed shift of the acceptor protons in the presence of the donor and the chemical shift of the uncomplexed acceptor;£§°}is the difference between the shift of the acceptor protons in the pure complex and the shift of the uncomplexed acceptor; K is the formation constant of the complex; and C is the D total concentration of the donor. Just as in the case of Optical spectra, one measures the chemical shift of the acceptor protons in a 33 series of solutions containing varying concentrations of the donor and plots (l/A) versus (l/CD). A straight-line plot is obtained in the case of 1:1 complexes and the values oon and K are obtained from the slope and the intercept. The NHR technique was used very successfully by Foster and nye in the study of the dinitrobenzene and trinitrobenzene complexes with aromatic compounds (117). EXPERIMENTAL I. Solvents The following five procedures were used to purify the solvents in this investigation. Procedure A'3..' Cyclohexane, 1,2-dichloroethane, carbon tetrachloride, dichlorometh- ane, n-heptane, and benzene were purified in the following manner: §532_;. The solvent was washed several times with concentrated sulfuric acid until the acid layer remained colorless. §§gp_g, The acid was drained off and the solvent was washed with water, sodium bicarbonate solution (0.5 5) and, then, water again. £15323. The water was drained off, and the wet solvent was dried for two days over anhydrous barium oxide. §532_fl. The solvent was then refluxed for six hours over fresh barium oxide and fractionally distilled at atmospheric pressure through a 30 in.column packed with glass beads, collecting only the middle portion. Procedure 8 , Triethylamine, piperidine, 2,6-lutidine, pyridine, and nitro- methane were all purified by first drying the solvents over anhydrous barium oxide for two days. They were then vacuum (0.10 mm) fraction- ally distilled through a 30 in.Vigreaux column using fresh anhydrous barium oxide as the drying agent. Procedure C Acetone was purified by simply refluxing it for six hours over anhydrous calcium chloride and then fractionally distilling it at atmospheric pressure through a 30 in.column packed with glass beads, 34 35 collecting only the middle portion. Stronger dehydrating agents could not be used because of possible acetone condensation; Procedure D Methanol was purified by the following procedure: fisgn_1. Solid iodine and concentrated aqueous sodium hydroxide solution were added to the methanol and allowed to stand for several hours to remove any aldehydes or ketones present. §£32__. A large excess of silver nitrate solution (0.1 m) was added to the mixture to remove the unreacted iodine and any other reducing impurities. §§32_;. The above mixture was fractionally distilled to remove most of the water present. The remaining solvent was refluxed over magnesium ribbon for six hours. It was then fractionally distilled at atmospheric pressure through a 30 in.column packed with glass beads, collecting only the middle portion. Procedure E . . Acetouitrile was purified in the following manner: §§gn_1. The impure solvent was allowed to stand over calcium hydride for two days. It was then fractionally distilled at atmospheric pressure through a 30 in.column packed with glass beads, collecting only the middle portion. §§gg_;. The distilled solvent was next refluxed for 24 hours over phosphorus pentoxide and finally fractionally distilled, collecté ing only the middle portion. A II. Reggents Substituted tetrazoles used in this investigation were synthesized and purified according to the procedures described in Part II of this thesis. Pentamethylenetetrazole, however, was obtained from 36 Knoll Pharmaceutical Co. and purified by several recrystallisations with anhydrous ethyl ether: mp 60-61‘; lit. mp 61' (28). The 1,2,4-triazole was purchased from Cslbiochem and recrystallized twice from dichloromethane—methanol. Tetracyanoethylene was originally obtained from the du Pont Co. and subsequently from Eastman. It was recrystallized twice from chlorobenzene and then sublimed twice in an inert atmosphere. The potassium salt of the tetracyanoethylene anion radical was prepared and purified using the method of Webster gg_gl. (118). Tetracyano- ethane was obtained by reducing tetracyanoethylene with hydrogen iodide and was purified by recrystallization from ethyl acetate-hexane (119). The remaining‘n-acids used in this study are listed in Table VI. III. Apparatus Absorption measurements in the visible and ultraviolet regions were made on a Cary recording spectrOphotometer Model 14, in silica cells of 1.00, 5.00, and 10.00 L 0.01 cm pathlengths. Measurements were done at room temperature of approximately 25‘. All NMR measurements were made on carbon tetrachloride solutions with a Varian Associates A-60 spectrometer operating at 60 Hz/sec. The probe temperature was approximately 35'. Tetramethylsilane was used as the reference for chemical shift measurements. Infrared measurements were made on a Beckman recording spectro- photometer Model IR-S, using potassium bromide pellets. 37 mouuaummou campus dun £28 Anew ecwuunnu eosannnu Icuu vewuauduumuoea .oo suauusm ecosouosauouuuuauunn.o.~ gossnuu Amzuv Away ego one noun vowaaueuomuoea .oo ceaueeu easemeeouu—cuuuun.m.~ ecunueaouofinodn 33 .53 .NS not esfiiuuauooe .8 cabana 38 25.35% . “cacao Anew eooauoou soon awou ease um vesuuueo .oo seem so ocunuoauuocusoouemouuuoa moauunso use osoeueeouo~no Annuav Aeev eocu sou“ sown oeuuuasuuhuoox .oo ucom so emoaanuoosumoeuuea an add ma convocaumumm common mason-cu .23 of: .3 «Sue RESULTS AND DISCUSSION The possible complex formations between tetrazoles and Lewis fl; acids were studied in order to indicate the extent of'n - electron donor ability of the tetrazole ring. The greater part of this work was performed using tetracyanoethylene (hereafter abbreviated as TONE) since it has been shown to be one of the strongest flzacids avail- able (44). However, the complexing tendencies of the tetrazoles were studied with other flzacids such as tetracyanoquinodimethane (TCNQ), pychloranil (CA), 1,3,5-trinitrobensene (TNB), and 2,4,7-trinitrofluorenone (run). The choice of solvents for these studies had to take into account two major factors; the general insolubility of the substituted tetra- zoles and the reactivity of the fleacids, especially TONE. Solubility studies were made with various tetrazoles,and it was found that they dissolved to an appreciable extent in several chloronated hydrocarbon solvents and alcohols. In order to test for‘fl-acid-— solvent interaction, Beer's law studies were preformed in the following solvents; dichloromethane, 1,2-dichloroethane, methanol, acetonitrile, and absolute.ethanol (Figure 2). Solutions of TOME in all five solvents adhered to Beer's lawgspectra were obtained immediately after making up the solutions. However, drastic spectral changes were observed within a short period of time in methanol, ethanol and acetonitrile; thus, the use of these solvents was discontinued. Because of greater.solubility of the tetrazoles in dichloromethane than in 1,2-dichloroethane, the former solvent was chosen for the re- mainder of this study. Beer's law studies were run on the rest of the 38 Figure 2. 39 Beer's law study of TCME in: .L, dichloromethane (268mm); B, l,2-dichloroethane (263 mp); C, methanol (235 mp); D, nitromethane (278 mp), and 3, ethanol (240 mp) using l—cm-pathlength cells. 40 .~ assess nno~ a.” gc«Uduumeo:ou GHQ Rn o-ON DO." ,\ ‘ . . . O . . . s. C . G O O . U . n . vo.u sc.n re.“ 01 x eoueqsosqv 41 flbacids (TCNQ, CL, TNB, and TNF) in dichloromethane (Figures 3 and 4). All four compounds obeyed Beer's law and their solutions showed no spectral changes with time. (a series of solutions with TONE and PMT were made in dichloro- methane and their absorption spectra were determined. ‘As-the PMT/TONE mole ratio increases, the TCNE absorption band at 268 mp gradually de- creases in intensity and broadens at the base. The series of curves (FigureS) pass through a rather poorly defined isobestic point, indi- cating the presence of only two absorbing species. In order to make quantitative spectral measurements, higher con- centrations of TOME Gel:x 10-3) had to be used. Since the tetrazoles did not absorb radiation in the region studied (300-450 mp), spectral measurements were taken directly from the TONE absorption band. It is interesting to note that in all the tetrazole-TCNE complexes the absorption maxima of charge-transfer bands were not observed, but rather the absorption band of the TONE broadened considerably and extended to lower frequencies (Figures 6-13). Absorption measurements were, consequently, carried out on the rather steep side of the new _ absorption band with consequent loss of accuracy. An absorption maximum, however, was obtained in the case of S-benzyltetrasole (Figure 14). The experimental data were evaluated according to the method of Kelelaar, g g. (112) using EquationlOp 31. Since this treatment involves the use of a least squares analysis, a Fortran computer pro- gram was used on a Control Data Corporation 3600 computer. The results plus their corresponding average deviations are listed in Table VII. The values of the formation constants are independent of the wavelength 42 Figure 3. Beer's law study of: A, TCNQ (400 mp); B, TNF (280 9:); and C, TNB (260 mp) in dichloromethane using lacm— pathlength cells. .n 0.33..— e.— xzum gaudhudeocou 43 n Geo Ce” i. ON“. Oe-M O-d l .\ OJ e e o .. o.n e e e e . e D r c.n ue , e me 4 01 x ooueqzosqv 44 . oumw .dueo summeunumAnaon~ a woman ossnuoaouo~:o«v :. “at many fiumeuounOLm no soda and u.uoon u up sod x x moaueuumoosoo 92 9.2 own . o.N 1 o.o );o.o 01 x eoueqsosqv Absorbance x 10 45 7.0T 5.0 ' 3.0 l 1.0 fl 2§o 360 3&0 wavelength in q: Figure 5. Spectrophotometric study of 5.0 x 10's! TCNIKA) with PMT mole ratios of B, 1085; and C, 4330, in dichloro- methane using a Lem-pathlength cell. Absorbance x 1 0 46 7.0 ‘ V 5.0 3.0 d 1.0 ‘1 Figure 6e 360 32'0 340 360 Wavelength in mp SpectrOphotometric study of 1.0 x 10'3 M,TCNB with various mole ratios of PMT in dichloromethane using a l-cm-pathlength cell. Absorbance x 10 1:7 7,0 ‘ 5.0 " 3,0 1.0 “ 360 3fo ' 3io 3§o Wave length in Is: 3 Figure 7. Spectrophotometric study of 2.0 x 10- M TONI: with various mole ratios of ‘L-eyclohexyl-S-ethyltetrasole in dichloromethane using a hem-pathlength cell. Absorbance x 10 7.0 4 5.0 3,0 ‘ 1.0 ‘ 48 310 350 310 . 3 o Wavelength in I): Figure 8. Spectrophotometric study of 4.0 x 10"3 M TCNI: with » various mole ratios of l-cyclohexyl-S-methyltetrasole in dichloromethane using a Lem-pathlength cell. Absorbance x 10 49 7.0 ' V 5.0 ,< 3.0 J 5 5 1 1.0 1 1 3io 356* 350 340 flavelength in mp Figure 9. Spectrophotometric study of 4.0 x 10‘3 5 mm: with various mele ratios of 1-methyl-5-cyclohexyltetrasole in dichloromethane using a Lem-pathlength cell. Absorbance x 10 fl 50 7.0 ‘ 5.0 ' 3.0 ' 1.0“ 310 310 350 340 wavelength in up Figure 10. Spectrophotometric study of 4.0 x 10‘3 !,TCME with various mole ratios of 5-prepyltetrasole in dichloro- methane using a tom-pathlength cell. Aw- !. 'm:vll'nl Ihilll< Absorbance x 10 51 7.0 . ~ 5.0 1 3,0 “ 1.0 " 3 0. 9 g 1,5-diphenyltetrasole 4‘ I A\ 360 31'0 350 30 Wavelength in q: Figure 11. Spectrophotometric study of 1.0 x 10'3 M TCME with various mole ratios of 1,5-diphenyltetrasole in dichloromethane using a I-cm pathlength cell. Absorbance x 10 52 7.0 ‘ 5.0 ‘ 3.0 T 1.0 ‘ 3'30 3‘10 3&0 350 wavelength in mp Figure 12. Spectrophetometric study of 4.00 x 10'3 5 mm: with various mole ratios of l-ethyl-S-phenyl tetrasele in dichloromethane using a l—cm pathlength cell. Absorbance x 10 53 7.0 .. 5.0 . 3,0 1.0 q 360 “- 310 3&0 330 Wavelength in q: Figure 13. Spectrephetometrdc study ef 4.0 x 10"3 M TOMS with various mole rat ies of l-phenyl-S-ethyltetrasole in dichloromethane using a l-cm-pathlength cell. Absorbance x 10 54 7.0 I Benzene l x 10.23!) 5.0 . \ ‘ - ‘x 1 2 3.0 1 1.0 .5 360 3% a o , 45'0 Wavelength in up Figure l4. Spectrephetemetric study of 5.00 x 10"3 M TONE with various mole ratios of 5-bensyltetrasele in dichloromethane using a S-cm-pathlength cell. 55 .onouauueu I as ”massed I an ”demoed I an ”mucus I um .nhnuom I ems .o—om_uea uueumu cu ce>uw ewe euueuucoo codes-home 1w .suwce~e>s3 oen can o~a one coon o no.o Hm.” e~.m cu.” e~.~ -.~ an.“ e neun-n-nm-s one can oee cue m so.o oe.s nn.~ me.~ on.“ ee.s u sena.n-uu.s coca can” onus coca o . ms.o oa.o .ma.o om.o eo.o ma.s u nunmso-n.s oe.~ oo.~ co.n o . o~.o so.s ne.~ on.” cc.“ e seam-n cos co. ones o e~.o ~o.~ no.~ o~.~ so.s u ,nessxoeomoao.n-ox.s coo on~s onoN e o~.c on.~ mn.~ an.m an.~ a aaomun-~sxono~omous coo one cnu u ac.o mm.s oo.s «o.s oo.a a paeuucn-~axoeosoaoss ooe ooh cuss anon couu m ~o.o an.m m~.s cm.~ «n.s ~n.~ nn.~ a emu nunlunull u can man can nNn . cum man can mom can odouauuoa .uedouuuuea new: monogamoo eceuhnueocumowuueh .~n> eaeua 56 at which the measurements were made indicating the presence of a 1:1 complex (see p 32 for discussion). The only exception seems to be the complex of the 1,5-diphenyltetrasole, but here again the trend is not outside the expected experimental error. In the case of the complex between TONE and S-bensyltetrasole it was noted that there were spectral shifts with time, possibly from an addition reaction. However, for short periods of time (approx. one hour) the absorption maximum was stable and was similar to the maximum obtained for the benzene-TONE complex (Figure 14). The experimental data are given in Table VIII. Hhile there is little doubt that the appearance of the new band is due to complex formation, in view of relatively low molar absorptivity it is doubtful that this is a charg- transfer band. In all cases the Ketelaar plots were linear as may be seen in the representative graphs of Figure 15. Since TCNE is very reactive chemically, it was important to deter- mine whether the spectral changes observed in the TCME-tetrasole systems were not due to side reactions, Tetracyanoethylene readily forms the radical anion TONE0' or it can undergo a reduction to HZTCNE. Both of these substances were synthesised,and their spectra were determined. Their spectra (Figure 16) showed good agreement with the data in the literature (118). From comparison of these spectra with the results obtained in the TCNE-tetrasole systems it is evident that the spectral changes observed in these systems are not due to the formation of either the radical anion or of the reduction product. It is also known that TOME can react irreversibly with strong bases (118). Solutions of TOME were prepared in methylene chloride containing piperidine, triethylamine, or pyridine. The spectra of these 57 on so an as o ~o.o e~.~ e~.~ an.~ a~.~ o«.~ e .muu one can can can at mums—3253 .euowuuueuahnmemun 52. many no Ned—Boo heuemeuuoewuumu ounu> manna. 58 10.0 4 8.0 . B 6.0 _1_ ec'ea 4.0 2.0 C 0 O I: O 1,0 7- 2.0 3,0 Concn. of Fifi (E) Figure 15. Retelaar plots for PMT complexes of: A, TCNQ (283 up); D, CA (3l5 g1); C, TNB (270 mp); D, TNF (300 up); and E, TCNE (293 I’l) in dichloromethane. Absorbance x 10 59 7.0 GP 5.0 ‘ 3.0 1 leo ‘ 300 530 400 Wavelength in 91 Figure 16. Spectrephetemetric study of s l x 10'4 M TCNE' in acetonitrile using a l—cm—pathlength cell. 60 solutions were found to be time dependent and irreversible formation of colored adducts were noted. The spectral changes were entirely different from those observed with the tetrazoles. An attempt to study TONE-1,2,4-triasole complexation was not successful because of insolubility of the triasole in any of the non- donor solvents. Spectral shifts similar to those obtained in TCNE-tetrasole systems were also observed when.PMT was added to solutions of other‘n- acids, namely trinitrobensene (TNB), gychloranil (CA), trinitnOfluor- enone (TNF), and tetracyanoquinodimethane (TCNQ), (Figures 17-20). Formation constants of the resulting complexes were calculated by the same technique and the results are given in Table IX. It is obvious from Table 1x that the n—acids listed are much weaker‘fl-electron acceptors than TCNE. The absolute values of formation constants of the PMT complexes with CA, TNF, and TNF are not very accurate and should not be heavily weighted. Since the formation constants are small, measurable spectral changes can only be obtained with very large excesses of one of the reagents, and this greatly mag- nifies the experimental error (113). However, the data at least indicate the relative strengths of the PMT complexes. As seen from Table IX the strength of the PMT complexes follows closely the trends of the other two donors, benzene and hexamethylbensene. This is another indication that tetrazoles form‘nacomplexes with fl;acids. The nuclear magnetic resonance method of Hanna and Ashbaugh (116), describedgon.p»32 of this thesis, was used in the determination of the PMT-TNB complex formation constant in carbon tetrachloride solutions. A plot of (lflh) versus [l/(PMT)] gave a satisfactory straight line (see Figure 21). .All chemical shifts were measured from the standard Absorbance x 10 61 7.0 4 x 5.0 q 3.0 4 1 70H 1.0 “ LI 0.08 M PMT - :‘=———J 210 230 2d0 300 Have length in 1,: Figure 17. Spectrephetometric study ef 2.5 x 10"5 M pochleranil with various mole raties of PMT in dichloromethane ’ using a l-cm-pathlength cell. Absorbance x 10 62 7.0 0 5.0 " 3.0”- 3 0 3 o 50 1.0- 5550 250 £50 360 Wavelength in q: Figure 18. Spectrophotometric study of 2.0 x 10" g 1,3,5-trinitro- bensene with various mole ratios of PMT in dichloro- methane using a l-cm-pathlength cell. Abserbance x 10 63 7.0 \ \ 5.0 .. 3.0 - I i 1.0 fl 0 :70 2'30 ‘ 230 300 Wavelength in q Figure 19. Spectrophotometric study of 5.1 x 10" M tetracyano- quinodimethane with various mole ratios of PMT in dichloromethane using a loom-pathlength cell. Abserbance x 10 3.0 .4 1.0 w 230 360 350 Wavelength in 9.1 Figure 20. Spectrophotometric study of 2.0 x 10‘5 M 2,4,7-trinitro- fluorenone with various mole ratios of PMT in dichloro- methane using a loom-pathlength cell. 65 Awuv eoseeouome .eesuoe use nude.eeoamoo ou cums: anodes wouuu cu oeuuo>coo one oeoceweuou emu mu co>~w euuouocoo souuomuou no eesno> one .50 a .eoou ..um~oo .ooouosouh new ..o=u .honsceouo: s.huuoumomu cucomuo cu «exonamou nonwoouozs .ueuoou .x .m one weekend .H .A menu oom~o>o eun.s Amman. moose nn.c --- oe.~ on.“ Assoc a. no deconcoemhnuomoxem ms.o Angus. sug>v . n~.o an.o no.3 Aesoo a. so oocowmom an.a co.c os.o on.o -.o an.s Amsowmo as so use ~H man may «a .azoa maps season encouocoo couusmuom i .omuwceuum Neda-cu no moouuoaaou .xu edema 66 J 5.0 a o H x m <6 3.0 . 1.0 . I ' I 1.0 2.0 3.0 l m Figure 21. NMR spectroscopic study of TNB proton shift with in- creasing cencentration of PMT in carbon tetrachloride. 67 tetramethylsilane signal with an estimated accuracy of $0.5 cps. The formation constant of the complex was found to be X - 1.3, which is higher by an order of magnitude than the constant determined spectro- photometrically. This is not surprising, however, since the latter was determined in dichloromethane which is a polar solvent, as compared to carbon tetrachloride, and can itself participate in the complexation re- action (120). In order to verify this point, the formation constant of the complex was also determined spectrophotometrically in carbon tetrachloride solutions. Once again, no charge transfer band was observed and the measurements had to be made at the broad absorption tail as in the case of TONE-tetrazole complexes. The results are summar- ised in Table X. Good agreement of the two values obtained by quite different techniques seems to confirm the validity of our assumption that the increase in absorption such as shown in Figures 5-20 is due to complex formation. It seems reasonable to conclude from the above data that the tet- razole ring does indeed possess some hedonor ability and is capable of forming charge-transfer complexes with neacids. .As seen from Table VII, VIII, and IX there is a reasonable agreement between the electron inductive effect of the substituent groups and the stability of the complexes. It is also interesting to compare the formation constants of the l-cycldhexyl-5-methy1tetrasole—TCNE complex with that of l-methyl-5- cyclohexyltetrasole-TCNE complex as well as the complexes formed by l-pheny1-5-ethyltetrasole and l-ethyl-5-phenyltetrazole. It is seen that when the larger group is in the l-position of the tetrazole ring, the stabilities of the complexes are less than when the groups are reversed. 68 00: came camp 0 019 .3..— ns... an." noon x hfi . M n.:.~ no~h~ nfiou qr . 59.3 253 1" El 3 .eouuousoouuou scenes 3 lean—loo euooouueuaegmuolousen neuooceoouumcuuu on» won a uo souuoculuoueo ouuueaouoiouuoeam .N egos 69 It has been shown that the nitrogen and carbon bonds which extend from the l- and 5- positions of the tetrazole ring respectively, are coplanar with the ring itself in PMT (43). However, it seems reasonable to assume that these bonds, particularly the nitrogen bond, become slightly distorted and consequently out of plane with the tetrazole ring when large groups such a cyclohexyLéand phenyl-are substituted on the l- or 5- positions. Thus, a large substituent group in the 1- position would extend above and below the tetrazole ring to a greater degree than in the 5- position. This large group would then partially shield the ring from TONE or other flbacids with corresponding lower- ing of the stability of the complex. It is interesting to note that the formation constants of the benzene- TCNE and PMT-TONE complexes are comparable (Table IX). 0n the other hand the formation constant of iodine monochloride complexes with PMT is larger by three orders of magnitude than that of benzene-I01 complex (28). These results indicate that in the former case, the complexation occures through one of the nitrogen atoms on the EMT ring. The crystal structure of the PMT-ICl complex has been recently resolved (28); show- ing unambiguously that the complexation does indeed occur through one of the nitrogen atoms on the PMT ring. CHAPTER IV STUDIES IN FORMIC ACID SOLUTIONS 70 HISTORICAL 1.. 9.29.221 The use of formic acid as a non-aqueous media offers several advantages when compared to other acidic solvents? It has a con- venient liquid range and readily dissolves most organic and many inorganic compounds (17, 49). It is one of the strongest carboxylic acids known, being a much stronger acid than glacial acetic acid (18). A highdielectric constant of 56.1 at 25. (19) essentially eliminates formation of ion-pairs in dilute solutions. Thus, the ionic equilibria in formic acid are much less complex than those in acetic acid (P . 6.13) (20), or in many other acidic solvents. Unfortunately, a relatively large self-ionisation constant narrows its "pH" range as compared with that of water or of acetic acid (18). There are several inherent problems involved in the use of this solvent. It is difficult to purify initially and to retain in the purified state due to spontaneous decomposition by the following mechanisms: atom—+32 .- co2 HCOOH—->H20 ¢ (:0 Hinshelwood, Hartley and Topley (22) studied the influence of temperature on these two decompositions and found that formic acid is best purified and stored at lower temperatures. Since the anhydrous acid is extremely hygroscOpic, caution must be excer- cised to minimise its contacts with the atmosphere. The reader is refered to the Vol. 4 of "Chemistry in nonaqueous solvents" (54) 71 72 for a complete discussion of the solvent properties and chemistry in anhydrous formic acid. 11. Purification The first discussion of purification methods is given by Garner (23) and co-workers in 1911. They suggested that the solvent be distilled at 50' under reduced pressure from anhydrous copper sulfate. Several authors state (23, 24) that the use of phosphorus pentoxide and sulfuric acid as drying agents should be avoided in the purification procedures because of excessive formic acid decomposition. A review of several other procedures (25) indicates that the limiting factor involved in the purifica- tion of formic acid is the retardation of spontaneous decomposi- tion. This may be done by avoiding strongly acidic or basic drying agents and maintaining a temperature close to the freezing point of formic acid, 8.4! (21). A promising purification was reported recently by POpov and Marshall (21). This procedure involves a double vacuum distilla- tion of previously dried (by anhydrous CuSOA) formic acid from boric anhydride at room temperature. Although this method was quite time consuming and cumbersome, it did yield a product of ' very high purity. The acid obtained consistently melted between 8.3‘C and 8.5.0 with an average specific conductivity of 6.6 x FO'S mho/cm. These results are in essential agreement with the best previously reported values (24, 26). III. Acign- Base Eggilibria The strong protogenic nature of formic acid renders it quite useful for the study of solutions of very weak bases. The first study of acid - base equilibria in anhydrous formic acid was 73 reported by Hammett and Diets in 1930 (18). Using a quinhydrone electrode for titration measurements these authors showed that sodium formate, triphenyl carbinol and urea act as strong bases in formic acid while bensenesulfonic acid behaves as a strong acid. In a later study by Hammett and Deyrup (51) the pxb‘values of several weak bases were determined using a system of indicators. In a series of papers by Shkodin, Ismailov, and Dsyuba (50) the authors point out that formic acid both enhances and levels the strength of weak bases. Using quinhydrone and glass indicator electrodes with aqueous calomel as the reference electrode they titrated several weak acids and bases. They found that most inor- ganic acids in formic acid solutions remain largely dissociated while most of the organic acids were too weak to be titrated. The pKa and pr values that they determined for the weak acids and bases are questionable because of the method of calculation that was used. The titration curves of several weak organic bases in formic acid were also studied by Tomicek and Vidner (52) using indica- tors to determine the endpoints. However, their results are questionable because of the use of acetic anhydride to remove water in the purification of formic acid. The most recent work reported using formic acid as the solvent media for acid - base study was that of Popov and Marshall (21, 53). The purification and use of anhydrous formic acid was done using a closed glass system in an inert atmosphere. Potentiometric titrations were successfully carried out on several substituted tetrazoles using quinhydrone electrodes and‘gptoluenesulphonic acid as the titrant. Attempts to use hydrogen electrodes yielded 74 erratic results because of catalized decomposition of formic acid. Using a quinhydrone - sodium formate concentration cell the relative basicity constants were obtained for caffeine, theobromine, urea, sodium sulphate, pentamethylenetetrazole and nine substituted pentamethylenetetrazoles. 1v. Electrolytic Conductance The first conductance measurements in formic acid Imre reported by Zinninovitch-Tessarin in 1896 (55). He found that most inorganic salts were completely ionized while solutions of hydrochloric acid were not dissociated to any measurable degree. However, Schlesinger and Calvert (26) showed that Tessarin's results were in error because of improperly purified solvent. These authors studied the conductance of ammonia solution and concluded that ammonia is completely converted to the dissociated ammonium formate salt. They also showed, contrary to Tesaarin's results, that hydrochloric acid was dissociated to a moderate degree in formic acid. The second paper (37) in a series of studies by Schlesinger and co-workers describes the conductances of sodium, potassium, phenylammonium and ammonium formate in anhydrous formic acid. Also, the dissociation constant for hydrochloric acid in formic acid was calculated using the Ostwald dilution law. This work was later re-evaluated by Schlesinger and Coleman (58) who again showed that these data obeyed the mass action law only if corrections for viscosity were not made. The conductances of alkaline earth formates in formic acid were studied by Schlesinger and Mullinix (56) and later by Schlesinger and Reed (59). The latter authors also developed a method for calculating the degree of dissociation for mixed 75 electrolytes in solution. The transference numbers of several ions in solution were determined in the last study of the series (60). In this paper, Schlesinger and Bunting used the transference numbers to calculate single-ion equivalent conductances at infinite dilution. Kendall, Adler and Davidson (63) extended the conductance work in formic acid to include 1:2 electrolytes such as magnesium and barium formates. A later conductance study by Lange (61) deals with the com- parison between formic acid solutions and aqueous solutions. Equiv- alent conductances of several electrolytes including potassium chloride, tetramethylammonium chloride, potassium picrate, and methylene blue were determined at infinite dilution. Lange also calculated the partial molal volumns of these solutions. In the latest conductance study to appear in the literature, written by Johnson and Cole (19), a proton transfer mechanism is postulated to explain the large equivalent conductance of pure formic acid. Also, a large dielectric constant of 56 at 25' indicates that the molecules of formic acid do not exist as non-polar dimers in solution. This is in contrast to acetic acid solutions where the molecules exist mainly as non-polar dimers. THEORETICAL I- We The high dielectric constant of formic acid, 56.1 at 25' (19), essentially eliminates formation of ion-pairs in dilute solutions. Thus, the dissolution of a weak base in formic acid may be treated by simply assuming the following equilibria s + Hcooa;:::an* ¢ m000' where the basicity constant, Kb, is given by: (HB*)(HCOO') . 1. Kb (3) If we let (33*) - (ncoo‘) . can; and (a) . (Cs‘acs) where CB equals the initial molal concentration of the base, 8, we can then substitute these values into Equation 1 to obtain the Ostwald dilution law: ozca lid 2. Kb- The basic equation used to calculate equivalent conductances is given by: lOOOLs vN. - ---- 3. C The specific conductance, Ls, is defined as K L8 -— 4. R 76 77 where K is the cell constant and R is the resistance in ohms. Arrhendus 'was the first to show that the degree of dissociation, u, of a weak binary electrolyte can be approximately obtained from the expression A a-—_ 5e ‘lVe whereJ\b is the equivalent conductance at infinite dilution. Sub- stituting Equation 5 into Equation 2 we obtain the following rela- tionship: \Q‘? A0 «yo-w) C 6. Kb- By rearranging Equation 6 we get: ‘IVC - (xfifl' - xfiA. 7. 02) .6» o It then follows that a plot of (NC) Eggggg (1(a) should give a straight line, and the values ofJNb and ‘8 can be obtained from the slope and intercept. However, this treatment is only valid for very dilute solutions since it does not take into account inter- ionic effects. Kohlrausch postulated the following imperical relationship A-A. -b JC— 8- and showed that for strong electrolytes whens/Vis plotted against VG? the curve (phoreogram) approaches linearity. In 1927 Onsager (64) derived an equation which was similar to Equation 8 above, taking into account relaxation, electrophoretic and Brownian effects in solution: A -~/\r°'(5/)b + 9) JC— 9. 78 0. and B are defined as 8.204 x 10'5 u - (relaxation effect) - 10. MDT)’: 82.43 B - (electrophoretic effect) .- —— 11, main}: where D is the dielectric constant, qthe viscosity, and T the absolute temperature. However, Shedlovsky, in 1932 (65) showed that A0 was not constant over any appreciable concentration range. Consequently, he rearranged Equation 8 defining a new funct ion/\g given as: , A-r BC‘: A .. __ 12. ° 143.68 A new value oon, which he called the true limiting equivalent I conductance, was then obtained by plotting A0 versus 0 from the equation: I eA'o I'A'o * BC-- 13. L1. Shedlovsky Iteration Techniug In 1938 Shedlovsky expanded the work of Fuoss and Xraus (66) and proposed the following equation for weak electrolytes (67) ‘A' are)” 14. A-XAo-S 0 the Onsager slope, S, is given by S - (LAO + a where a and p are described in Equations 10 and 11“ respectively. The function St, was defined as I Saul-sleaze... 15. 79 where Z is giv. as: s(c )’5 —————— 10. ./»1% 0 2- Substituting these equations into Equation 14 and rearranging we get: 17. Equation 17 may be combined with Ostwald's dilution law, Equation 2, and rearranged to give the final equation: 1 1 cszrg/x —— I —-- + -—-§- 18o A52 W0 KB“ If (ll/\Sz) is plotted versus (CJVSsz) a straight line is obtained with an intercept of (lfl/\°) and a slope of (l/XbJNB2). In order to use this treatment a value forJNB is assumed, and the corres- ponding values of Z and 82 are calculated. These values are used in Equation 18 to obtain a new'value ofJ\b which is then used to recalculate Z and $2. The process is repeated until consecutive JNb values agree within set lhmits. This procedure was used quite effectively by Fuoss and Shedlovsky (68) for the evaluation of A0 values and dissociation constants for several weak electrolytes. More recent conductance treatments by Fuoss and Onsager (69) and Fuoss and Accascina (70) take into account two additional inter- ionic forces. These two corrections deal with osmotic and static viscosity effects in the solution due to the presence of ions. However, in dilute solutions their results are in essential aggree- ment with those obtained by the Shedlovsky tteration.method. 80 All of these conductance methods are best handled by using the method of least squares. Hence evaluation of.l»° and ‘8 are easily obtained by use of modern digital computers. EXPERIMENTAL I. Solvent Formic acid (98+ percent, Baker and Adamson) was dried for 24 hours over anhydrous copper (II) sulfate. The solvent was then fractionally distilled through a 30 in.Vigreaux column under 1 mm pressure at room temperature. In order to keep the vapor pressure of the distilled solvent low, the receiving vessel was cooled with dry ice. The distilled product was further purified,;under a dry nitrogen atmosphere, by batch fractional freezing. A modified, sixoliter,separatory funnel with provisions for vacuum removal of impure solvent was used for this process. ’Usually five or six successive fractional freezings were required to obtaina.product with a constant mp range of 8.25! to 8.30. (ave. 8.27.). Specific conductance of the purified formic acid varied between 5.9 x 10-5 to 6.3 x 10"5 ohm"1 cm"1 with an average value of 6.08 x 10-5 ohm'1 cm'l compared to the best literature value of 6.6 x 10"5 ohm“1 cm"1 (21). The solvent was recovered for reuse by distilla- tion.) A Cyclohexane, nitromethane and pyridine were all purified by vacuum fractional distillation through a 30 in.Vigreaux column using anhydrous barium oxide as the drying agent. II. Reagents The inorganic salts used for conductance work, with the excep- tion of sodium formate, were of reagent quality and were used without further purification. They were, however, dried before use at 110'. Sodium formate was recrystallized three times from conductance water. Anhydrous hydrogen chloride was prepared by 81 82 slowly dropping concentrated sulfuric acid on to analytical grade sodium chloride. The gas was dried by passage through several magnesium perchlorate drying columns and then bubbled into the anhydrous formic acid. The concentration of the stock solution was determined by the standard silver chloride precipitation (82). The organic salts, triisoamylbutylammonium iodide (TABI) and triisoamylbutylammonium tetraphenylborate (TABTPB), were prepared and purified using the method of Coplan and Fuoss (71). Eastman Grade tetramethylammonium bromide was used without purification but was dried in vacuo at 50' for two hours. Triisoamylbutylammonium perchlorate (TABClOa) was prepared by a metathetical reaction between silver perchlorate and the iodide of the organic salt. -The obtained material was recrystallized three tines from a methanol-water mixture. The final product melted at 94-96.. Substituted tetrazoles used in this investigation were synthes- ized, and purified according to the procedures described in Part II of this thesis. Conductance water for potassium chloride solutions was prepared by passing distilled water through a mixed resin bed obtained from Crystalab Research Laboratories. Matheson, Coleman and Bell "Reagent A.C.S." grade potassium chloride was fused in a platinum crucible, ground in an agate mortar and dried at 110'. The transition metal perchlorate hexaquo salts used in the complexation study were reagent grade, purchased from G. Frederick Smith Chemical Co. These salts were dried in vacuo at 20' to constant weight prior to their use in the preparation of stock solutions. The waters of hydration were removed when necessary by the addition of calculated amounts of Fisher Certified Reagent grade 83 acetic anhydride. 111. .ABELEEA The A. C. conductance bridge used in this investigation has been previously described (72). It was operated at a frequency of 1000 cps. The conductance cells were similar to those described by Daggett, Bair, and Xraus (73). Platinum electrodes were used but they could not be platinized since platinum black catalyzed the decomposition of formic acid. All equipment used for con- ductance work was cleaned with sodium hydroxide solution to re- move traces of oil, thoroughly rinsed with hot distilled water and then steamed for one half hour on a steaming apparatus. .After steaming, the cells were rinsed with pure acetone and dried in a stream of purified nitrogen. The constants of the three cells used in this investigation were measured by the standard procedure using XCl solutions and the Lind, Zwolenik,and Fuoss equation.(74). They were found to be: 0.4427 t't0,.0012; 1.446 :t 0.005; 3.988 1: 0.013. The temperature of the cell solution was maintained at 25.00 g 0.029 by a Sargent S-84805 thermostatic bath assembly filled with? light mineral oil. All melting points were taken on a Fisher- Jones melting point block for which the usual stem corrections were made. Absorption measurements in the visible, ultraviolet and near- infrared regions were made on a Cary recording spectrophotometer Model 14, in glass stoppered cells of 1.00, 5.00 and 10.00:t 0.01 cmppath-lengths. ,Measurements were done at room temperature of approximately 25'. Infrared measurements were made on a Beckman recording spectro- photometerbModel IR-S, using potassium bromide pellets. 84 IV. Procedures A given amount of pure formic acid was weighed into each conductance cell, allowed to equilibrate, and its specific con- ductance determined. Next, known amounts of a stock solution were added to the formic acid from a weight buret and the contents of the cells were thoroughly mixed. After temperature equilibration, the resistance was recorded, the solutions were then remixed and the resistance taken a second time. The addition of stock solutions to the conductance cells were made under normal laboratory condi- tions since brief exposure to the atmosphere caused no observable. changes in resistance readings. The concentration range of solutions used in the conductance studies varied from 5 x 10'3M,to 2.0 x 10'15, 'The upper limit of concentration was determined by the Fuoss equation. cmax - 3.2 x 10'7D3, where D is the dielectric constant, since at higher concentrations the simple laws of dilute solutions of electrolytes are no longer obeyed (75). The lower limit was taken such that the specific conductance of the solvent would be less than 10% of the specific conductances of the most dilute solution. The solvent correction was made by subtracting the specific conductance of the solvent from that of the respective solution. A The spectrophotometric methods, including mole ratio and continuous variation studies were preformed in a manner which has been previously described (76) in Part III of this thesis. RESULTS AND DISCUSSION I. Conductance A Conductance data in formic acid were obtained for eleven salts, four substituted tetrazoles and anhydrous hydrochloric acid, Attempts to study several other salts and tetrazoles were not successful either due to their reaction with formic acid or to the lack of sol- ubility. The measured equivalent conductances of the solutions and their respective concentrations are given in Table XI. The experimental conductance data were evaluated according to the method of Fuoss and Shedlovsky (67, 68). Since this treatment involves the use of an iteration procedure and a least squares analysis, a Fortran computer program was used on a Control Data Corporation 3600 computer. The results and the corresponding average deviations are listed in Table XII. It should be noted that the precision of the results is not as great as those often found in conductance measure- ments. At least in part the experimental scatter is due to the diffi- culties encountered in working with formic acid as a solvent, namely, its instability, especially in the presence of platinum, its extreme hygroscopicity and its high degree of self-ionization. The Shedlovsky data, which were taken directly from the computer print-out, are plotted in Figures 22 to 25. It should be noted that while all the graphs are linear, the reproducibility varied between two extremes as represented in Figure 22. In the one case, the points from separate runs fall on the same line as seen in the sodium chloride and sodium formate plots. In the other case the two sets of data formed separate lines as demonstrated by the potassium chloride plots. No explaination, other than experimental error, is presented for 85 86 Table XI. Edu ivalent conductances in anhydrous formic acid. (Superscripts designate series of determinations) = I = 1020 /\ 1020 /\ 102t /\ TABCIOA 0.3023 24.60‘ 0.4502 38.11‘L 1:167 47:88a 0.3773 24.51, 0.6317 37.37 2.289 45.68 0.4345 24.44 0.8683 36.57 3.350 44.23 0.4853 24.31 1.051 36.06 4.204 43.27 0.5464 24.23 1.226 35.60 5.141 42.42 0.5858 24.16 1.389 35.20 5.957 41.69 0.6431 24.04 1.519 34.92 6.771 41.02 0.6855 23.93 1.684 34.58 7.470 40.53 0.7333 23.80 1.928 34.09 8.180 40.07 0.7720 23.57 2.047 33.86 9.466 39.23 0.3034 24.62b 0.5458 36.90b 10.035 38.94 0.3597 24.51 0.7651 36.25 1.180 47.00b 0.4003 24.42 0.9789 35.65 2.189 44.84 0.4429 24.35 1.146 35.26 3.266 43.18 0.4887 24.29 1.313 34.93 4.205 42.18 0.5237 24.24 1.474 34.45 5.135 41.15 0.5690 24.13 1.638 34.09 5.881 40.48 0.6035 24.04 1.761 33.82 7.418 39.41 0.6338 23.91 1.890 33.53 8.155 38.87 0.6790 23.77 2.019 33.27 9.456 38.04 2.171 32.99 10.165 37.67 cont inued Table XI-o- Continued. 87 1 I 102C /\ 1020 /\ 1020 /\ 11.0104 NaHCOO 1.899 42.24‘ 2.450 63.87‘ 0.6400 75.06‘ 3.429 39.94 32.60 63.26 1.388 65.79 5.683 37.74 4.169 62.47 2.006 60.18 7.458 36.50 4.738 62.04 2.509 56.50 8.995 35.54 5.351 61.73 3.086 53.23 10.314 34.81 5.963 61.14 3.557 50.94 11.023 34.47 6.549 60.66 3.997 49.05 1.229 43.30b 7.109 60.33 4.860 45.97 2.644 40.37 7.621 59.90 5.238 44.75 3.876 38.65 8.022 59.71 5.576 43.74 5.209 37.20 2.694 62.28 5.947 42.74b 6.323 36.25 3.658 60.99 0.6828 72.18 7.022 35.69 4.183 60.38 1.374 63.43 7.593 35.29 5.277 59.28 2.072 57.27 0.0590 49.04c 5.740 58.81 2.628 53.47 0.1350 48.17 6.194 58.36 3.159 50.71 0.2260 47.58 6.460 58.12 3.513 49.12 0.3610 46.68 6.937 57.71 4.027 46.99 0.4930 45.96 7.408 57.36 4.479 45.88 0.6080 45.49d 2.313 63.46°‘ 4.835 . 44.16 0.0830 48.79 3.256 62.63 5.210 43.09 0.1670 47.76 4.062 61.91 5.566 42.09 0.2600 47.15 4.725 61.28 5.962 41.03 0.3980 46.35 5.365 60.82 0.5300 45.76 2.011 64.89d HCl (Schlesinger) 0.6680 45.17 3.176 64.19 0.7800 43.23 4.166 63.48 0.8000 70.00 5.039 62.82 1.100 69.20 NaHCOO (Schlesinger) 5.791 62.28 1.500 62.30 2.530 64.10° 1.950 58.90 6.418 61.70 3.364 63.51 2.760 53.70 8.375 60.54 4.323 62.78 3.280 51.80 9.682 59.75 5.061 62.20 5.180 45.20 7.550 55.94 5.706 61.79 9.510 37.40 20.68 54.67 10.10 35.40 23.37 53.72 17.401 30.00 25.28 53.00 18.10 26.00 27.34 52.28 31.53- 50.94 continued Table XI ---Continued. I 88 1020 /\ 102C /\ 102C /\ NaBr KBr 1101 1.877 42.57‘ 1.029 47.52‘L 1.621 43.86a 2.695 41.45 2.018 45.43 2.314 42.58 3.404 40.64 2.833 44.26 3.176 41.30 4.056 40.01 3.515 43.43 3.853 40.54 4.786 39.31 4.222 42.73 4.649 39.68 5.496 38.76 4.827 42.20 5.169 39.18 6.048 38.35 5.514 41.56 5.941 38.57 6.660 37.95 6.105 41.13 6.426 38.25 7.194 37.56 6.766 40.67 7.067 37.92 7.694 37.28 7.242 40.43 7.853 37.39 8.225 36.97 7.784 40.20 8.531 36.92 1.863 41.83b 8.272 39.85 1.534 44.56b 2.635 40.60 1.376 45.83b 2.394 43.27 3.379 39.56 2.327 43.81 3.049 42.48 3.943 38.98 3.294 42.36 3.625 41.87 4.556 38.36 3.995 41.62 4.298 41.29 5.212 37.78 4.942 40.61 4.776 40.87 5.840 37.27 5.543 40.03 5.401 40.36 6.300 36.93 6.243 39.42 5.898 40.00 6.868 36.54 6.942 39.17 6.290 39.71 7.260 36.27 7.492 38.78 7.216 39.11 7.926 35.80 8.204 38.46 7.639 38.87 8.733 38.25 1.533 43.05c 9.345 37.73 3.122 41.14 4.593 39.91 6.228 38.85 7.160 38.56 7.479 38.16 8.267 37.86 8.667 37.53 9.420 37.19 9.731 37.08 continued 89 Table XI --- Continued. 10 c /\ 10 c /\ 102C /\ NaCl MeéNBr LiCl 1.568 41.23‘ 0.9590 46.94a 0.8872 41.72‘ 3.134 39.64 2.370 44.05 1.701 40.22 4.830 38.09 3.134 42.94 2.487 39.18 6.215 37.10 3.860 42.08 3.317 38.29 8.713 35.62 4.419 41.54 3.932 37.71 9.434 35.24 5.002 40.96 4.512 37.24 1.664, 41.17 5.571 40.52 5.162 36.77 3.032 39.26 6.065 40.10 5.752 36.42 4.467i 37.70 6.578 39.81- 6.161 36.17 5.632 36.71 7.057 39.49' 6.743 35.80 6.874 35.84 7.451 39.201) 7.143 35.57 7.737 35.33 1.260 45.67 7.571 35.27 8.548 34.90 2.100 43.78 1.520 40.18b 1.053 42.17c 3.043 42.27 2.341 38.77 2.072 40.36 3.669 41.48 3.259 37.68 2.926 39.29 4.270 40.90 4.113 36.73 4.068 38.06 5.035 40.00 4.965 36.01 5.187 37.09 5.692 39.43 5.589 35.59 6.016 36.52 6.241 39.03 6.423 35.13 6.822 36.03 6.788 38.65 6.978 35.00 7.323 38.60 7.624 34.60 7.831 38.23 8.334 34.37 8.392 37.89 8.916 34.30 9.545 33.31 continued Table XI ---Continued. 1020 A 102C A 1020 A Tetrazole 5-Methyl Tetrazole 5-Phenyl Tetrazole 0.7413 11.76‘ 1.428 28.15‘ 1.742 8.898 1.593 ”9.947 1.698 26.76 2.305 8.196 2.458 8.754 3.091 22.54 2.606 7.886 3.188 8.038 3.691 21.13 3.131 7.422 3.624 7.700 4.346 20.08 3.497 7.148 4.227 7.301 5.028 19.01 3.914 6.875 4.848 6.955 5.456 18.53 4.345 6.619 5.407 6.726 6.298 17.60 4.940 6.315 5.889 6.468 6.481 17.36 5.183 6.206 6.311 6.304 7.136 16.81 1.595 9.079b 6.862 6.114 7.723 16.27 2.050 8.439 7o192 6.008b 8e813 ISoSIb 2:520 7e907 1.738 9.641 0.9373 30.57 2.956 7.498 2.462 8.682 1.846 25.84 3.376 7.155 3.161 7.993 2.209 24.44 3.782 6.886 3.708 7.574 3.215 21.64 4.030 6.725 4.317 7.185 3.663 20.81 4.347 6.538 4.816 6.904 4.130 19.85 4.661 6.373 5.333 6.648 4.993 18.57 4.969 6.225 5.793 6.442 5.153 18.47 6.241 6.267 5.741 17.64 6.703 6.107 6.366 17.10 7.097 5.975 7.725 15.91 8.849 15.17 continued Table XI ---Continued. 91 1020 /\ 1020 /\ 102C /\ PMT “» a b c 0.8592 36.75 1.485 31.62 1.288 34.15 1.350 33.42 2.063 30.17 1.511 32.67 2.732 28.13 2.825 26.83 3.019 28.33 3.034 26.74 4.063 24.46 3.183 26.48 4.358 24.24 4.401 23.44 4.343 24.61 5.151 22.36 5.404 22.13 4.713 23.65 5.585 22.25 5.620 21.62 5.661 21.90 6.511 20.51 6.708 20.39 6.182 21.49 6.955 20.53 6.826 20.19 6.618 20.57 7.665 19.24 7.673 19.37 7.194 20.32 :006 19.48 8.012 19.11 7.518 19.63 8.856 18.19 8.190 18.92 7.927 19.25 8.963 18.65 9.153 18.12 8.163 19.36 10.20 17.33 Table XII. Limiting equivalent conductances in anhydrous formic acid. Run Ave. Salt No. 1 No. 2 No. 3 No. 5 Average Dev. TABTPB 24.61 24.71 24.66 0.05 r180104 42.03 41.33 41.68 0.35 N80104 50.33 50.90 49.63 50.22 0.04 NaHCOO 70.74 70.38 71.04 70.83 70.92 0.32 NaHCOO Schlesinger and Reed Value (57) 66.23 (69.39)‘‘ NaCl 47.34, 47.10 47.73 47.39 0.23 NaBr 49.03 49.31 49.17 0.15 KBr 51.88 52.69 52.29 0.40 KCl 49.66 50.60 50.58 50.28 0.41 LiCl 46.16 45.60 45.88 0.28 NHacl 53.89 53.16 53.53 0.36 MeéNBr 52.22 51.61 51.92 0.30 801 107.75 104.55 106.15 1.60 HCl Schlesinger and Martin Value (59) 80.0 (104.99)‘l aRecalculated values are given in parenthesis 93 Figure 22. Shedlovsky plots: A, sodium chipride; B, sodium perchlorate; C, potassium chlor de; D, potassium bromide; E, sodium formate, in formic acid. 1 94 «N enemas on x Noam/xv one. ease one. one oHN v Coma .— use ouch. u use soon 0 fl. and 2.06:} 9 I 01 6:; a 6.8 3 ll? I OeQN 95 Figure 23. Shedlovsky plots: A, lithium chloride; 8, sodium bromide; C, hydrochloric acid; D, tetramethyl- ammonium bromide; and E, ammonium chloride, in formic acid. 96 nu..aeuaa on x Noam/\o ones cues once ofle o .r o.- no ¢ . . . c.8s I O Q L H.‘ .0. . C . C 3 00.0000. w o a . c.c~ 97 Figure 24. Shedlovsky plots: A, triisoamylbutyla-onium tetraphenylborate; and B, triisoamylbutylammonium perchlorate, in formic acid. 40.0 ‘ 36.0 '1 32.0 n 28.0 d 24.0 98 WW 110 210 2 CASzf x 10 Figure 24 3.0 4.‘0 99 Figure 25. Shedlovsky plots: A, 5-phenyltetrazole; B, tetrazole; C, 5-methyl tetrazole; and D, pentamethylenetetrazole, in formic acid. 100 n« sham—h. S a nem‘oxv + 7‘001‘ - 50.51; experimental value for xc1 is A0 - 50.28. The agreement between the two values is within experimental error. Schlesinger and Bunting (60) used Hittorf's method to determine the transference numbers of sodium and potassium ions in solutions of corresponding formates. Using their value of 0.220 for the sodium ion,the corresponding limiting equivalent conductance of 15.6 was calculated from our sodium formate data. The discrepancy between T‘bl. XIVe 104 Limiting equivalent conductances of single ions. Ion ’A o Ion )\ o 113* 12.33 acoo‘ 50.05 11* 19.36 0104' 29.33 u.‘ 20.97 Br' 28.30 2+ 23.99 c1“ 26.52 una* 27.01 193' 12.33 4. 3* 79.63 105 their value and our value of 20.97 may be explained by the fact that they used Ostwald's method (Equation 7, p 77) to treat their conductance data in order to calculate the original transference numbers. Comparison of the limiting ionic conductances in Table XIV leads to some interesting observations. It is seen that both 8* and 0800‘ have abnormally high conductances, which is indicative of a proton-jump conductance mechanism similar to the one found in aqueous solutions. 0n the other hand, the mobility of the alkali metal cations do not differ appreciably from that of the tetra- methylammonium ion-oer ,9 for that matter, among themselves. mile the limiting ionic conductance‘of Li’ in aqueous solution is only half of that of K‘ in the same solvent, the respective values are 19.36 and 23.99 in formic acid. It seems reasonable to conclude that this fact is due to the much smaller ability of formic acid to solvate ions. These conclusions are in essential agreement with those postulated by Johnson and Cole (19). ‘ Based upon the limitinglequivalent conductances of the formate and hydrogen ions from Table XIV along with the average value of the specific conductivity of pure formic acid (6.08 x 10"5 ohmflcm'l), the autoprotolysis constant of the pure solvent was calculated. The obtained value of 2.2 x 10"7 seems to indicate a purer solvent than the value of 5 x 10'7 reported by Hammett and Diet: (18). The value of 2.2 x 10"7 at least represents the upper limit of the autOprotolysis constant of formic acid. 106 In a continuing study of the physicochemical properties of substituted tetrazoles, the investigation of their pr values in formic acid appeared to be of interest. The pr's and limiting equivalent conductances of several tetrazoles were determined using Equation 1, p 76. It should be noted that in view'of the high di- electric constant of formic acid the concentration of ion-pairs should be quite insignificant and,in contrast to acetic acid solu- tions, there is no distinction between the ionization and the dissociation constants. The results are listed in Table XV. It is interesting to compare the value of the pr for penta- methylenetetrazole obtained from conductance work to the value ob- tained potentiometrically (21). The agreement between the two values is exceptionally good considering that two different methods were used in their determination. It is surprising to note that the five- substituted tetrazoles act as bases of approximately the same strength as the 1,5-disubstituted tetrazoles. This fact indicates that not only is the dissociation of the acidic proton on the tetrasble ring suppressed but the ring itself or a nitrogen atom on the ring acts as a proton acceptor. 107 .Auuv .uuefleumeuoa cauuoaomucouomd eno.~ nun enoneuuouocouasueaducom .«o.~ ~o.~ oo.~ no.“ no.~ nus oo.o mm.on oo.on ~m.oo on.om o,\ odouauuouoaosasuoaacaom ~o.o on.“ ma.a ~m.~ use ma.c n~.as oo.o~ o~.am /\ odoauuuouesaosm-n oo.o oo.~. oo.~ oo.~ nus on.o o~.~n no.sn ca.~n o/\ odouauuoudssuoz.n eo.o m~.~ um.” ch.” ”as on.o -.m~ nu.oa ow.aa Ix odouuuuoa .>en .924 encased n .02 N .02 fl .02 com .umcoo mououmuuoa .ucouusuou odes cannon economnce a“ uoaousuuou seen we eucsuucoo houndusm .>x enema 108 II. Complexation A systematic study involving the first row transition metal per- chlorates was made in order to observe any complex formation with PMT in anhydrous formic acid. The following seven perchlorate hexahydrate salts were used in this investigation: chromium(III), manganese(II), iron(III), iron(II), cobalt(II), nickel(II), and copper(II). These salts were dried in vacuo at room temperature to remove absorbed water, care being taken not to remove any water of hydration. Since iron perchlorate salts could not be dried without under- going partial decomposition, they were not included in this study. Solutions of the five remaining perchlorate salts were made in both formic acid and water to note any spectral differences. The small spectral shifts between the two solvents are recorded in Table XVI. Due to the low molar absorptivity of manganese(II) perchlorate hexahydrate, its visible spectrum could not be obtained without recourse to extremely high concentrations. In order to observe the effects of hydration upon the perchlorate salts in formic acid, spectra were obtained for anhydrous and hydrated solutions of each salt. The waters of hydration were removed from the anhydrous solutions by adding calculated amounts of acetic anhydride. The results may be seen in Table XVII. Although the peak shifts be- tween the anhydrous and hydrated solutions are negligible, there is a small change in the molar absorptivities. Also, a study was made with the solutions containing acetic anhydride to see if there were any spectral changes with time. The reaction between the anhydride and water appeared to be complete within a few minutes after mixing. No spectral changes were observed after several days of standing. 109 Table XVI. Transition metal spectra in formic acid and water. Salt Solvent Amax (q) s Cu(0104)2 . 6H20 320 ' 313 12 Cu(0104)2 - 6820' acoos 820 22 111(0104)2 8 6820 320 405 5 660 2 725 3 1170 4 Ni(0104)2 6 6820 acooa 400 8 670 3 730 .4 1180 5 Co(0104)2 . 6H20 520 276 3 459 3 511 5 00(0104)2 ' 6820 HCOOH ‘(250 46s 7 520 12 576 12 Cr(0104)3 6 6020 30008 426 24 586 29 110 Table XVII; Transition metal perchlorate spectra in formic acid. ' fig: Salt max (9‘) e Cu(ClOz)2 - 6820 820 22 Cu(ClOz)2 4 611203 820 23 Ni(0104)2 - 6820 400 8.6 670 3.2 730 3.4 1190 4.6 Ni(ClOa)2 . 6820a 400 8.7 669 3.3 729 3.5 1190 4.7 C0(C10 ) 3 6H20 520 12 4 2 1300 2.7 C0(C104)2 & 61120a 520 12 1300 2.7 c:(c104)3n- 6H20 420 32 587 44 c:(c104)3 . 611203 425 37 590 52 ‘Acet ic anhydride added to remove water, 111 A series of solutions with varying concentration was made for each transition metal salt. Calculated amounts of acetic anhydride were added to each solution and their spectra were determined. As may be seen in Figure 26 all four of the salts obey Beers law at the concentration range studied. Spectral studies were performed on solutions with and without the addition of acetic anhydride. Since only minute differences were observed, all of the remaining work was performed without the addition of acetic anhydride. A. Co r II Perchlorate studies In order to insure that the hydrated perchlorate salt did not lose any waters of hydration during drying, copper was determined using a standard iodometric procedure (82). m. Calcd for 011(0104)2 ~ 6H20: Cu, 17.15. Found: Cu, 17.11. Hypsochromic spectral shifts were noted when PMT was added to capper(II) perchlorate hexahydrate solutions in formic acid (Figure827éu28). The absorption maxima progressively changed from 12,300 cm‘1 (813 mp) for the copper(II) perchlorate hexahydrate to 14,000 cm'1 (695 mp) at a PMT/Cu(II) mole ratio of 10:1. These spectra are typical of copper(II) complexes which theor- etically have only one allowed transition [ZEg——92T2 (D)]. The absorbance maximum remained unchanged from a mole ratio of 10:1 up to a ratio of 50:1. However, at mole ratios of 50:1 and greater, the maximum again progressively changes to shorter wavelengths (Figure 29). Figure 26. 112 Beer's law plots: A, chromium(III) perchlorate hexa- hydrate (420 an); 8, copper(II) perchlorate hexahydrate (815 mp); C, cobalt(II) perchlorate hexahydrate (520 mm); and D, nickel(II) perchlorate hexahydrate (397 mm), in formic acid. 113 a. .a0 3 as»: «S x m as“ usuucoooeo o.” o.“ 0”" P P (I) O. O t.o.n r c.0u 01 x eoueqzesqv AUGUI’U‘HUU K LU 114 700 in 5.0 1 600 A700 800 900 Have length in sq: Figure 27. Spectrophotometric study of 1.0 x 10"3 g Cu(0104)2 ' 61120 with various mole ratios of PMT in formic acid using a Abserbance X 10 115 7.0 1 5.0 . “' 1.0 ' 660 V 760 860 T 900 Have length in 1;: Figure 28. Spectrophotometric study of 1.0 x 10'3 g Cu(0104)2 . 6820 with various mole ratios of PMT in formic acid using a 10°C. COlle Abserbance x 10 116 7.0 ~ 0 in reference cell 5.0 fl .. Q. -. erence cell 3.0 d O 1.0 I 600 700 860 _ 900 Have length in mp Figure 29. Spectrophotometric study of 1.0 x 10"”3 g Cu(ClOa) ' 6H20 with various mele ratios of PMT in formic acid using a lO—cm cell. 117 These absorbance shifts were accompanied by a hyperchromic effect up to a mole ratio of 20:1 at which point a maximum absorptivity was obtained. At higher concentrations of PMT the absorptivity passed through a minimum and then slowly increased (Figure 29). These results were totally reproducible and may be seen by comparing Figures 27, 28, and 29. Maximum absorptivity at a 20 fold excess of PMT is clearly seen in a plot of molar absorptivity y;;ggg_mole ratio PMT (Figure 30). A continuous variation study (76) of PMT with copper(II) perchlor- ate hexahydrate shows a maximum at a ligand mole fraction of approx- imately 0.67 (Figure 31). This would correspond to a stoichiometry of 2:1 (ligand/metal) for the solution complex. In several of the above solutions a light blue crystalline mat- erial precipitated out after 24 hours. This precipitate could be re- dissolved by the addition of dilute perchloric acid. Larger quantities of this material were prepared by carefully controlling the concentra«3 tion and temperature of the PMT-copper(II) perchlorate solution in order to avoid super-saturation. An infrared spectrum of the precipi- tate, after it had been filtered and dried in vacuo (Figure 32A) showed no evidence of a tetrazole ring (of. Figure l, p.23). However, a comparison of this spectrum with those of ccpper(II) formate tetra- hydrate, water and formic acid (Figures 33A, 8 and C) shows that the precipitate is copper(II) formate without any solvated water or formic acid. The loss of solvation molecules is also substantiated by the fact that the precipate's weight decreased as it changed from a light blue to a light green‘material upon drying. The dried material was analized for copper using a standard iodometric procedure (82). Anal. Calcd for Cu(HCOO)2: Cu, 41.38. Found: Cu, 41.41. 118 .ALB canv macs cannon cm ououomcdxo: ouuucfinouoo Aunvuoooco some 925 .m mononucu Beacon .4 no mouse ouumu ego: .on ouswum cumc.uaeo~0vso .cocoo comma“ .cocco 3cm o.2. anon omen cm: 10w O 1 4 ONH 3.531A12dzosqw Jeton 119 .maoo so a a wanna use. cannon scam .-cs x n.ev emu .n can .-o~ x o.~v .uusuou lagoon .<.mu¢m oueuommdxom susuoucoueo annvuscoco no hoses cemueuuse unoccuucoo .mn sound» can! a .8333 £62 “.6 9.6 sue P L 0 .0g 5. h 01 x 7V. 'ponoy - '08qu to.” ro.~ no...” 120 TV 5 5f 7 '1 i3 1 '5 7' 3 1'1 13 1 9 1‘1 1'3 C Infrared absorption spectra using KBr pellets of copper(II) precipitate from Cu(ClO )2 ° 6H20 solutions of: A, PMT: 8, sodium formate; and 3, pyridine, in formic acid. (wave length in 1:) Figure 32. 121 3 5 7' 9 11 1'3 A 3 5 '7 9 1'1 1'3 8 3' 5 7' '9 11 {3 c Figure 33. Infrared absorption spectra using KBr pellets of state. (wavelength in )1) 122 Copper analysis of the solvated copper(II) formate salt could not be obtained because of interferences from reduction of the formic acid molecules. A mole ratio study was run on the copper(II) perchlorate hexa- hydrateesodium formate system in formic acid (Figure 34). A comparison of the results with those of the PMT system (Figures 27 and 30) shows a striking similarity. Just as in the PMT study, hypsochromic shifts were noted with increasing concentration of sodium formatee-changing progressively from 12,300 cm"1 (813 mm) for the uncomplexed solution to 14,400 cm"1 (695 mp) at a sodium formate/Cu(II) mole ratio of 10:1. Again, a maximum absorptivity was obtained at a 20 fold excess of sodium formate (Figure 30). At higher concentrations of sodium for- mate an absorptivity decrease and a new hypsochromic shift was noted. Upon standing, a light blue precipitate was obtained fro-1a concentrated solution of copper(II) perchlorate hexahydrate and sodium formate. The precipitate turned from a light blue to green upon drying in vacuo, An in spectra (Figure 328) showed that it was copper (II) formate. A standard iodometric procedure was used to analyse the percent copper in the sample (82): Anal. Calcd for Cu(HCOO)2: Cu, 41.38. Found: Cu, 41.36. It appears, therefore, from the above observations that the addition of either PMT or sodium formate to Cu(II) solutions in formic acid causes the formation of copper(II) formate. A continuous variation study of sodium formate and copper(II) perchlorate (Figure 31) shows a sharp peak at a ligand mole fraction of 0.67. This would correspond to a stoichiometry of 2:1 (ligand to metal) for the complex. XAU ADSOI’D‘IICQ 123 8.04 6.0 ‘ 4.0 . 2.0 I} 600 760 860 900 Have length in 11,1 Figure 34. Spectrophotometric study of 1.0 x 10"3 g Cu(0104)2 . 61120 with various mole ratios of sodium formate in formic acid using a.-10-cm cell. 124 A further comparison of mole ratio studies (Figure 30) shows that the sodium formate complex solutions have a greater molar absorp- tivity than the PMT solutions. This seems reasonable if the following equilibria are assumed: 1 walo-2 1. PMT-«1- acooH———.amr* + 110007 HCOOH 2. 21mm" 4» Cu”___. Cu'H'(HCOO')2 - x 118008 Thus, at equal molar concentrations of PMT and sodium formato,there would be fewer formate ions available for coordination in the PMT solutions due to the 161112481611 equilibrium of 2111' (Step 1). The coord- ination of the formate ions to copper(II) ion in formic acid solutions should be suppressed by addition of a strong acid. The effect of perchloric acid additions to PMT solutions of copper(II) perchlorate hexahydrate may be seen in Figure 35. As small amounts of perchloric acid are added to these solutions, the spectra are progressively shifted back to that of the capper(II) perchlorate hexahydrate solu- tion (Spectrum C). However, the absorptivities of the perchloric acid solutions are slightly less than that of the original capper(II) perchlorate hexahydrate solution. This indicates that a solution of copper(II) perchlorate hexahydrate in formic acid is coordinated by formate ions which arise in small concentrations from the autoprotol- ysis of the solvent. Pyridine in formic acid solution should act as a strong base and be leveled to an equivalent amount of formate ion. Thus, we would expect to obtain similar spectral shifts to those observed above for solutions of sodium formate and capper(II) perchlorate. Indeed, inspec- tion of Figure 36 shows that pyridine solutions behave in exactly the same .-uu" w—eaavv A Lv 125 0.7 - 0.5 _ 0.3 660 760. 800 900 Have length in nu: Figure 35. Spectrophotometric stud; of 2. 0 x 10'"3 H Cu(0104)2 61103 (C) with: A, 8. 0 x 10" i_i sodium formate; 8, 8:02 x 10' M PMT; 0, 8. 0 x 10-3 M sodium formate + 8. 0 x 10' per- chloric acid; a, 8. o x 10"3 M PMT + 8. 0 x 10-3 M perchloric acid in formic acid using a 5-cm cell. “UGULU‘IIUC A LU 126 0.7 . 0.5 ‘L 0.3 «.4. \ \ 0.1 - 660 700 800 900 Wavelength in 11,: Figure 36. Spectrophotometric stud; of 1. 2 x 10'"3 510u(0104)¥06H20 (C) with: A, 2.4 x 10" g pyridine; 8,-1.2 x 10" pyridine; D, 2.4 x 10"2 11 pyridine o 2.4 x 10"2 g per- chloric acid in formic acid using 5-cm cells. 127 . manner as sodium formate and PMT solutions of copper(II).perchlorate. Again a precipitate was obtained which upon drying was found to be copper(II) formate (Figure 320). Anal. Calcd for Cu(CH00)2: Cu, 41.38. Found: Cu, 41. 35. Attempts to study solutions of copper(II) formate in several polar non-donor solvents such as nitromethane were not successful due to lack of solubility. To summarize, it appears that PMT or protonated PMT does not coordinate with capper(II) perchlorate hexahydrate in formic acid. Apparently the spectralshifts in these solutions at low ligand concen- trations are caused by the coordination of two formate ions with each capper(II) ion. However, as the mole ratio excess of the ligand is in- creased to larger values (Figure 29), new spectral shifts indicate the presence of a different absorbing species. This species is probably a copper(II) ion with additional formate ions coordinated about it. B. NickelSII) Perchlorate studies In order to insure that there was no water of hydration loss upon drying, a Karl Fisher water determination (81) was run on the vacuum dried nickel(II) perchlorate hexahydrate salt. Anal. calCd for N1(C10 )2: 320, 29.61. Found: H2 0, 29. 5%. A mole-ratio study of the PMT-nickel(II) perchlorate hexahydrate system is shown in Figure 37. The spectrum obtained from nickel(II) perchlorate hexahydrate solution (Spectrum A) is characteristic of a weakly coordinated octahedral complex (77). This spectrum shows the following absorption bands: (27,) 8,500 cm'l (1180 up) 1:131:13 from a 3A28——> 31.28 transition: (‘93) which is Split into two bands, 13,700 cm"1 (730 mu) and 14,900"1 (670 mp) arising from a 3A2§-e3T18 128 .uuoo camcousquIHOIa o wcuumgouoo cannon :« 92m no nouuou ouoa emouue> nova economnoxon scone—noueo Anuvuoxoac.m mac“ n n.n «o momma ouuueaouocmouuoocm .un summon 1E_a« newcouobom fl _ ofia _ o? 2: can / 1‘ o.o 01 x ocueqxosqv 129 transition. This band is split into two bands in weak complexes because of spin - orbit coupling between the 3T28(F) and 1E8 states which are very close in energy (78). The last band, (93) 9.000 cm"1 (400 my) is due to a 3‘28_"3T23 transition. As PMT is added to nickel(II) perchlorate hexahydrate solution, hypsochromic spectral shifts occur (Figure 37).' The absorption bands of the weakly coordinated hydrated salt (55) 8,5009cm"1 (1180 mu), (”2) 13,700 cm""1 (730 mp) and 14,900 cm“1 (670 mm), and (2%) 9,000 cm”1 (400 mu)'progressively changed to 10,300 cm"1 (975 mp), 17,000 c111'l (590 11,1), and 27,400 cm'1 (365 191) respectively. The 3123-» 3228 transition (75) is no longer split into two absorption bands since the nickel(II) ion is now in a stronger ligand field (no 2T2g(F)'lES spin-orbit coupling). Even at high mole ratio excesses of PMT (300:1) limiting absorbances could not be obtained for any of the absorption peaks (Figure 38). When perchloric acid was added to these solutions, all of the spectra shifted back to the original spectrum of nickel(II) perchlorate hexahydrate. The above results indicate that just as in the copper(II) solutions, coordination of the metal ion takes place with formate ions, and not with protonated PMT molecules. A continuous variation study of PMT and nickel(II) perchlorate hexahydrate in formic acid showed a clear minimum at a PMT mole fraction of 0.67 (Figure 39). These results would correspond to a stoichiometry of 2:1 (ligand to metal) for the complex. Solid nickel(II) formate was obtained by carefully controlling the concentrations of the nickel(II) perchlorate hexahydrate-PMT solutions in formic acid. This light green precipitate, which could be redissolved by the addition of dilute perchloric acid, was filtered and dried in vacuo. The infrared spectrum of the dried material (Figure 40A) 130 macs unseen cu ououohnoxon ouuaonmauoa auuvuoxodc.u 92m we moose ouuou oaoz onc.~Ago~oynz .mamoc\\\ocemau .ccaou own emu om: .mm oumwuh 3 thnyndzosqv Jeton 131 .aa96 suwaoasuoonsoca a was»: oaoo oaauoe as am duos x ~.~v no aoauosoaoocoo moooo ,. m a»; 92m .. summons—axes ououcgcaon 35.3on no house commonest, use—.5230 .2” show; one a.o p 92m couscouu one: 0 n.o ‘ n.o a. r as non mm .To 0.77 .0 (0.01 Z-OI x v<7- °p°"°v - 'psq° O O Io.~.. 132 < 3' 5‘ T 9 {1 f 3 A 3 5 7‘ 9 1‘1 1‘3 B 3 5 7 9 T1 13 C Figure 40. Infrared absorption spectra using KBr pellets of nickel(II) A, PMT: precipitate from Ni(C104)2 - 6H20 solutions of: 8, sodium formate; and C, pyridine, in formic acid. (wavelength in.p). 133 shows the presence of formate ions. No further analysis was pre- formed on this material. Mole-ratio.studies could not be obtained for sodium formate- nicke1(II) perchlorate solutions even in very dilute solutions be- cause of nickel (II) formate precipitation. An ir spectrum (Figure 403) of the precipitate shows that it is nickel(II) formate. However, by the use of lO—cm—pathlength absorbance cells and extremely dilute solutions a continuous variation study was performed. Unfortunately, the absorbance difference between the various solutions were so small that meaningful results could not be obtained. Similarly, solubil- ity difficulties were observed for the pyridine-nickel(II) perchlorate system. An infrared spectrum of the precipitate obtained from pyridine-nickel(II) solutions (Figure 400) is identical to that of nickel(II) formate. C. Cobalt(II) Peggglgrate studies A mole ratio study of PMT-cobalt(II) perchlorate hexahydrate is shown in Figure 41. ,These spectra are indicative of an octahedrally coordinated cobalt(II) ion, giving the following absorption bands: (3)1) 7.700 cm'1 0130 mp) arising froma 4210:4422 transition; (95) 19,200 cm"1 (520 my) arising from a 4T1(F)---’4A2 transition; and (93) 21,300 cm;1 (470 mp) arising from a 4T1(F)--+“‘T1(P) transi- tion (79). As PMT is added to solutions of cobalt(II) perchlorate hexahydrate, slighé hypsochromic spectral shifts with corresponding increases of absorptivity occur. Even at high mole ratios of PMT to cobalt(II), a limiting absorptivity is not reached (Figure 42). Continuous variation studies give a peak at a ligand mole fraction of 0.67 (Figure 43), indicative of a 2:1 stoichiometry (ligandzmetal) for the complex. Although solid cobalt(II) formate 134 .cmoo summon on mauoo cuwcoacuoaia01m moans 92m no uouusu egos usomnm> some ououoznoxo: oueucucouoa AHHvuuencc.m nwo~ x 0.0 no meson ouuuoacuococuuoocm .Ho ousmam 15 cu summouo>m3 8: go oh... pen \ \ . o: .. o.n om .. ca 2: 01 x ooueqzosqv 135 .omos cannon cu ououomcdxo: ououcanouoa Auuvunocco cove 92m no human cause usacz ouzc.~Amo~ovoo .coccoxazm .coooo me on ma .No oaowaa as o~m r o~ Kniniqdzosqe melon 136 .28 Bwaoafioousoua 4 min: 22. 2.83 3 Am «.3 x 25 no 5330588 aouou o to; Ha I ououozcdonoc oueuoucouoo 333.300 we bosom condenser users: uncu .mo snow; HE coduoouu So: one he n..o n._o 5w 0 O o o o o o o o o nu $5 . Q o O O o o o *5 .«m o o r ooe T ed 7 o.- 01 x vv- P°"°v - psq°v 137 could not be recovered from the PMT solutions, addition of perchloric acid showed that these Spectral shifts were reversible. In pyridine and sodium formate solutions of cobalt(II) perchlorate hexahydrate a pink precipitate was obtained which could be redissolved by addition of dilute perchloric acid. The infrared spectra of the two precipitates are identical (Figure 44), both showing the presence of cobalt(II) formate. D. Chromium(IIl) Perchlorate studies A mole ratio study of the chromium(III) perchlorate hexahydrate- PMT system (Figure 45) shows octahedral coordination with the following absorption bands: (15) 17,400 cm"1 (585 mp) .arising from a QAZg-—e4ng transition; and (fig) 23,800 cm"1 (420 mp) 'arising from a 4A2§-e4T18 transition (80). There are only slight hypochromic Spectral shifts and hyper- chromic effects upon addition of PMT to chromium(III) solutions. These same phenomena, which are reversible upon addition of dilute perchloric acid, are also noted for sodium formate-chromium(III) solutions in formic acid (Figure 46). A plot of molar absorptivity versus the molearatio excess of ligand shows no limiting absorptivity (Figure 47). Although several attempts were made to prepare solid chromium(III) formate from PMT, sodium formate,or pyridine solutions of chromium(III) perchlorate hexahydrate in formic acid, they were unsuccessful. In summary, nickel(II), cobalt(II) and chromium(III) perchlorate- PMT solutions in formic acid behave in the same manner as copper(II) perchlorate-PMT solutions, 1.3, the coordinating ligands about the transition metal ion are not PMT or protonated PMT molecules, but rather formate ions. However, there is no evidence of multiple 138 3 5 7 9' 1'1 1'3 A 3' ‘5 7' 9 T1 1'3 B Figure 44. Infrared absorption spectra using KBr pellets of ' 6H20 solutions Cobalt(II) precipitate from Co(ClO )2 of: A, sodium formate, and B, pyr dine in formic acid (wave length in p). x10 508013081108 0.7 0.5 0.3 0.1 q 139 Figure 45. A A B 0 0 I 700 500 600 710 Wavelength in mp Spectrophotometric study of 2.0 x 10"3 !,Cr(0104)3°6H20 (c) with: 12 2.0 x 10- 13 mm 8, 4.0 x 10-2 g m; D, 4.0 x 10‘ §_PMT o 4.0 x 10"2 perchloric acid in formic acid using a 50-cm cell. Abserbance x 10 140 0.7 ie 0.5 . A C D 0.3 . D 0.1 . 400 500 600 100 Wavelength in m Figure 46. Spectrophotometric study of 2.0 x 10"3 1_i Cr(C104)3°6H 0 (C) with: A, 2.0 x 10" g sodium formate; B, 4.0 x l '2 .11 sodium formate; D, 4.0 x 10"2 g sodium formate 4- 4.0 x 10-21_i_ perchloric acid in formic acid using 5-cm cells. so 141 .3 co 3E3 5 economic: cyanogen—om Cass—saloons :33 «BE an ”ouoauou acumen .4 no moose onus." ego: .3 one»: sunfnaoosvao 5388533 £28 2 . on ma 1 on as So I n ? O|| 3H ON.» 4 a £314; :daosqe anon 142 complexat ion as there is in the copper(II) perchlorate hexahydrate systems o E H: a" 1. 3. 4. 5. 6. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. REFERENCES J. A. Bladin, Ber. lg, 1544 (1885). (a) E. G. Gross and R. M. Featherstone, J,_£haza,_£xp._1hgzp. £1. 291 (1946). (b) ibid., 81, 299 (1946). (c) ibi ., fig, 353 (1946). (d) ibid., 23, 323 (1948). (e) bi ., 22, 330 (1948). o. p- Q. F. R. Benson, Chem, Rezs., £1, 1 (1947). E. Olivera - Mandala, Cass Chem, Ital., 35, 175 (1914). (a) To So Mihina and R. M. Herbst, J, Org, Chem., lg, 1082 (l950)e (b) Re Ho HerbSt and Ko,Ro Wilson, ibido, 22, 1142 (1957)o J. A. Caruso, P. G. Sears, and A. I. Popov, J, Phys, Chem., 1;. 1756 (1967). Ao Dieter, J. Ph‘rm‘ 3513., 2’ 190 (1948). J. J. L. Zwikker, Pharm, Weekblad., 1;, 1170 (1934). A. I. Papov and R. D. Holm, J, Amer, Chem, Soc., §;, 3250, (1959). (a) J. Thiele and H. Ingle, Agn., 287 (1895). (b) ibid., 233. (1985). J. Bladin, 325.,‘25, 1413 (1892). Co H. BtUbfiker Jro’ J. Amer, Chem, SOCQ,‘§Z, 82 (1960). C. H. Brubaker and N. A. Daugherty, J, Amer, Chem, Soc., 8;, 3779, (1961). C. H. Brubaker and G. L. Gilbert, Inorg, Chem., 2, 1216 (1963). (a) A. D. Harris, R. H. Herber, H. B. Jonassen, and G. K. Worthein, J, Amer, Chem, Soc., Q2, 2927 (1963). (b) H. B. Jonassen, T. 0. Terry, and A. D. Harris, J, Inorg, Nucl. Chem., 22, 1239 (1963). A. D. Harris, H. 8. Jonassen, and R. D. Archer, M0: 5, 147 (1965). O. ABChan’ Chen: 2., £1, 1117 (1963)o L. P. Hammet and H. Diets, J, Amer, Chem, Soc., 22, 4795 (1930). 143 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 144 Jo F. Jahnson and Re Ho C013, J: at. Chem: 8°C., 7—3.’ 4538 (1951)o I. Kolthoff and S. Bruckenstein, J, Amer, Chem, Soc., lg, l (1956). A. I. P0pov and J. C. Marshall, J, Inorg, Nuclear Chem., 12. 340 (1961). C. N. Hinshelwood, H. Hartley and B. Tepley, Proc, Roy, Soc., (London), 100A, 575 (1922). J. B. Garner, 8. Larton and H. 0. Parker, J, Amer, Chem, Soc., 49, 236 (1911). A. S. Cooledge, J, Amer, Chem, Soc.,‘gg, 2166 (1928). J. C. Marshall, Ph.D. Thesis, University of Iowa, Iowa City, Iowa. H. I. Schlesinger and R. P. Calvert, J, Amer, Chem, Soc., 22. 1924 (1911). W. G. Finnegan, R. A. Henry, and R. Lolquist, J, Amer, Chem, Soc., 89, 3908 (1958). A. I. Popov, C. Castellani-Bisi and M. Craft, J, Amer, Chem, Soc., g9, 6513 (1958). ' E. K. Harvill, R. M. Herbst, E. C. Schreiner, and C. W. Roberts, J, Org, Chem. 15, 662 (1956). Von Bran“. flog fig-Q, 125 (1931). (a) E. K. Harvill, C. W. Roberts, and R. M. Herbst, J, Org, Chem., 1;, 58 (1950). (b) ibid, 671 (1950). R. L. Burke, Ph.D. Thesis, Michigan State University, East Lansing, Michigan, 1952. Hungarian Patent 114, 061, Chem, Abstr., 30, 3947 (1936). Beilstein, 7, 14 (1925). L. N. Owen and P. A. Robins, J, Chem, Soc, I, 329 (1949). "Organic Synthesis”, 2nd ed., Vol. I,-p 340 (1941). R. Tull, R. O'Neill, E. McCarthy, J. Pappas and J. Chermerda, J: Org, Chemo. $2, 2425 (1964) Jo So Mihina and Re Ho Herbst, Jo Org, Chem., g, 1082-92 (1950)o F. M. D'Itri and A. I. Popov, Inorg, Chem., 2, 1670 (1966). 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 59. 145 H. B. Person, R. E. Humphrey, H. A. Deskin and A. I. Pepov, J, Amer, Chem, Soc., £2, 2049 (1958). A. I. Pepov, C. C. Bisi, and M. Craft, J, Amer, Chem, Soc., fig, 6513 (1958). A. I. Pepov, T. C.'Hehman, and J. H. Vaughn, Inorg, Nucl, Chem., 19, 2027 (1964). N. C. Baenziger, A. D. Nelson, A. Tulinsky, J. H. Bloor, A. I. POpov, J, Amer, Chem, Soc., 82, 6463 (1967). R. E. Merrifield and H. D. Phillips,J, Amer, Chem, Soc., §Q, 2778 (1958). . 'u.'fi- Elsa“. L. R. Melby, R. J. Harder, H. R. Hertler, H. Mahler, R. E. Benson and H. E. Machel, J, Amer, Chem, Soc., §§, 3374 (1962). G. Briegleb, J. Czekalla, G. Reuse, Z, Physik, Chem., (Frankfurt), I) .‘fli‘.’ 16'” .2" . ‘“ I M. Orchin and E. O. Heulfolk, J, Amer, Chem, Soc., fig, 1727 (1946). E. E. Castro, L. J. Andrews, and R. M. Keefer, J, Amer, Chem, Soc., ‘89, 2322 (1958). T. Paulopoulos and H. Strehlow, Z, Physik, Chem., 202, 474 (1954). (a) A. M. Shkodin, N. A. Izmailov, and N. P. Dzyuba, Zhur. ObSChei nine, 2_3_, 27 (1953). (b) ibid., 6, 273 (1951). L. P. Hamett and A. J. Deyrup, J, Am_er, Chem, Soc., 24, 2721 (1932). O. Tomicek and P. Vidner, Chem List , 31, 516 (1953). A. I. Pepov and J. C. Marshall, J, Inorg, Nucl, Chem., 34, 1667 (1962). H. Knaver in "Chemistry in Non-Aqueous Solvents", Vol. 4, G. Janders, H. Spandau and C. C. Addison, Ed; John Hiley and Son, InCe New York, No Ye 19630 . H. Zanninovitch-Tessarin, Z, Phys, Chem., 12, 251 (1896). H. I. Schlesinger and R. D. Mullinix, J, Amer, Chem, Soc., £1. 72 (1919). Re I. SChlesiflger ‘nd ‘0 H. Martin. ibid., .22. 1589 (1914). H. I. Schlesinger and C. Coleman, ibid.,.flfl, 271 (1916). H. I. Schlesinger and F. H. Reed, ibid.,_&1, 1921 (1919). 60. 61. 62. 63. 64. 6b. 66. 67. 68. 69. 70. 7a. 75. 76. 77. 78. 79. 80. 81. 146 H. I. Schlesinger and E. N. Bunting,ibid., 32, 1934 (1919). J. Lange, 2. Physik. Chem., A187, 27 (1940). H. A. Kuska, F. M. D'Itri and A. I. Popov, Inorg. Chem., 2, 1272 (1966). J. Kendall, H. Adler, A. H. Davidson, J, Amer, Chem. Soc., 32. 1848 (1921). (8) Le Onsager, Ph281k: Rita, 21’ 388 (1926), (b) ibid., gg, 277 (1927). T. Shedlovsky, J, Amer. Chem. Soc., 23, 1405 (1932). (a) R. M. Fuoss and C. A. Kraus, J, Amer. Chem, Soc., 22, 476 (1933). (b) R. M. Fuoss, ibid., 21, 488 (1935). . A". -. 'l' .rxl-u‘lny-t l T. Shedlovsky, J. Franklin Inst., 225, 739 (1938). III..." .. R. M. Fuoss and T. Shedlovsky, J, Amer. Chem. Soc., 12, 1496 (1949). R. M. Fuoss and L. Onsager, J. Phys. Chem., 22, 668 (1957). R. M. Fuoss and F. Accascina, ”Electrolytic Conductance", Interscience Publishers, Inc., New York, N. Y., 1959. M. A. Ceplan and R. M. Fuoss, J. Phys. Cheg., 22, 1177 (1964). H. B. Thompson and M. T. Rogers, Rev, Sci, Instr., 21, 1079 (1956). Ho Me Daggett, E. J. 8311' and Co A. Kraus, J: met: Chem: 8°C., 22. 799 (1951). J. E. Lind, J. J. Zwolenik and R. M. Fuoss, J, Amer. Chem. Soc., 22, 1557 (1959).- R. M. Fuoss, J. Amer. Chem. Soc., 21, 2604 (1935). F. Rossetti and H. Rossetti, "The Determination of Stability Constants", McGraw-Hill, Inc.,-New York, N. Y. 1961. F. Cotton and G. Hilkinson, "Advanced Inorganic Chemistry” Interscience Publishers, Inc., New York, N. Y., 1962, p 734. ibid., p 736. ibid., p [25. C. J. Balinausen, "Introduction to Ligand Field Theory", McGraw- Hill, Inc., New York, N. Y., 1962, p 235. K. 6. Stone, "Determination of Organic Compounds", McGraw-Hill, 1110., New York, No Y., 1956, p 6e 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. 101. 147 E. Leininger, and K. G. Stone, "Elementary Quantitative Analysis”, Michigan State College Press, East Lanaing, Michigan, 1950, p 102. R. S. Mulliken, J, Amer, Chem, Soc., 13, 811 (1952). C. A. Coulson and P. L. Davies, Trans, Faraday Soc., 22, 777 (1952). L. J. Andrews and R. M. Keefer, ”Molecular Complexes in Organic Chemistry", Holden-Day, Inc. San Francisco, Calif., 1964. s. M. Bennett and c. a. W11115, J, Chem, Soc., 256 (1929). J. S. Anderson, Nature, 140, 583 (1937). P. Pfeiffer, ”Organische Molekulver-bindungen", 2nd Ed., F. Euke Co., Stuttgart, Germany, 1937. G. N. Lewis, J, Franklin Institute, 226, 293 (1938). G. Briegleb and T. Schachawskoy, Z, Phys, Chem., B19, 255 (1932). G. Briegleb, Z, Phys, Chem., B16, 249 (1932). G. Briegleb, ”Zwisehenmolekulare Krafte”, G. Braun Co., Karisruhe, Germany, 1949. D. L. L. Hammick, g£“22., J. Chem. Soc., 171 (1932). 0. L. L. Hammick and R. B. M. Yule, J, Chem, Soc., 1539 (1940). R. E. Gibson and O. E. Loeffler, J, Amer, Chem, Soc., 62, 1324 (1940). J. Rose, ”Molecular Complexes”, lst Ed., Pergamon Press Ltd” London, England, 1967. L. E. Orgel and R. S. Mulliken, J, Amer, Chem, Soc., 12, 4839 (1957). (a) J. Heiss, J, Chem, Soc., 245 (1942). (b) ibid., 322, (1943). M. J. 5. Dewar, ”Electronic Theories of Organic Chemistry", Oxford University Press, 1949. H. Brackmann, Rec, Trav, Chem., 22, 147 (1949). (g) Re So Mulliken, J: her: Chem, 5°C.. 2, 600 (1950). (b) ibid., 15. 811 (1952). (c) ibid., J Chem. Ph 3 .‘lg (1951). (d) ibid., 29, 801 (1952). (e) ibid., 2;, 341 (1954). (f) ibid., 22, 397 (1955). (g) ibid., Rec, Trav, cgim., 12, 845 (1956). 102. 103. 104. 105. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 117. 118. 119. 120. 148 He MCCOMOII’ Jo So Ham, 3.“ Jo Re Platt, J. CEO”. Phl‘o’ El, 66 (1953). M. J. S. Dewar and A. R. Lepley, J, Ameg, Chem, Soc., 22, 4560 (1961). Me J. S. De“: ”d Ae ROgCrS’ iblde. £12, 395 (1962). ‘o Re Lepley, ibido’ _8_4_, 3577 (1962)o E. M. Voigt and C. Reid, J, Amer, Chem, Soc., 8_6,, 3930 (1964). J. Halpley, D. N. Glev, and J. H. Hildebrand, J, Chem, Phy., £3, 107 (1960). R. S. Mulliken “1d H. Be Person, Ann, R9" ERIE. Chae. L3. 107 (1962). Po Job, Ann, Chili. P3113, 2, 113 (1928)o D. L. L. Hammick and G. Sixsmith, J, Chem, Soc., 580, (1935). H. A. Benesi and J. H. Hildebrand, J, Amer, Chem, Soc., 1;. 2703 (1949). Je ‘e ‘e Kecelur, Ce van de 81:01”. Ae GWdB‘it’ ‘nd H. chubas, Rec, Trav,‘ Chim., 11,, 1104 (1952). H. B. Person, J, g», Chem, Soc., 22, 167 (1965). G. De JOhnSOn ‘nd Re Ee Bonn, J: at. ChOlII, SOCe, £1, 1655 (1965). Re L. SCOCC’ Rec, Trav, Chile. 22, 787 (1956)o M. H. Hanna and A. L. Ashbaugh, J, Phys, Chan 92, 811 (1964). R. Foster and C. A. Fyfe, Trans, Faraday Soc., 6_1, 1626 (1965). O. H. Hebster, H. Mahler and R. E. Benson, J, 2r, Chem, Soc., §_£_._, 3678 (1962). w. J. Middleton, a. s. Heckert, s. L. 11661. andIC. c. KIOSpan, J, Amer, Chem, Soc., 22, 2786 (1958). R. Foster and D. L. L. Hammick, J, Chem, Soc., 2685 (1954). APPEND IX 149 COMPUTER PROGRAM A. Explaination The Ketelaar method, used for the spectrophotometric study of charge-transfer complexes, involves determination of the slope and intercept of a straight line. This is conveniently done by using the following modified least-squares Fortran computer program. The data are read into the program in the following manner: Card 1, name card; Card 2, the number of data sets read in; the remaining cards are used for data in the form of donor concentration and corrected total solution absorbance at same wavelength. The output consists of the printed slape, intercept, molar absorptivity, and formation constant of the complex along with their standard deviations. 150 151 moacflucoo he.nau..uezeeexomn<.mxz=muwxxpm Aevw+sz=muwz=m AHvx+xz=muxz=m z.sueonca oe AHVx\o.enAHvx an z.eu-mon on oe.on.onfiavee onuszpm ouuxzam ousxzam onwzsm ouxzpm em A~V»\onaevw Hon z.eue son on Aom.o~evama.Aonvommevea m.o.n-AevauAev>ma xxmvoq99 nam=H- _ .\.e.emm.uueoqm no end nemmeam .\.e.emm.n» maqum 4 mo >me nemm-.x~.omevaexeoe mm.