SYNTHESIS AND FHOTOLISIS OF AROMATIC NITRATE ESTERS by BfflE G. CSIZMADIA Dipl. Chem. Eng., Polytechnical University of Budapest, 1956. M.Sc, University of British Columbia, 1959• A Thesis Submitted in Partial Fulfilment of the Requirements for the Degree of DOCTOR OF PHILOSOPHY in the Department of CHEMISTRY We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA SEPTEMBER, 1962. \ In presenting t h i s t h e s i s i n p a r t i a l f u l f i l m e n t o f the requirements f o r an advanced degree at the U n i v e r s i t y of B r i t i s h Columbia, I agree t h a t the L i b r a r y s h a l l make i t f r e e l y a v a i l a b l e f o r reference and study. I f u r t h e r agree that permission f o r extensive copying of t h i s t h e s i s f o r s c h o l a r l y purposes may be granted by the Head of my Department o r by h i s r e p r e s e n t a t i v e s . I t i s understood t h a t copying or p u b l i c a t i o n of t h i s t h e s i s f o r f i n a n c i a l g a i n s h a l l not be allowed without my w r i t t e n permission. Department of The U n i v e r s i t y of B r i t i s h Columbia, Vancouver 8, Canada. Date PUBLICATION 1. I.G. Csizmadia and L„D« Hayward, S t e r i c E f f e c t s i n Ni t r a t e Esters I. The Synthesis and Spectra of 1, 2-Acenaphthenediol Derivatives and the S t e r i c I n t e r a c t i o n of Contiguous Nitroxy Groups. Tetrahedron, submitted f o r pu b l i c a t i o n , The University of B r i t i s h Columbia FACULTY OF GRADUATE STUDIES PROGRAMME OF THE FINAL ORAL EXAMINATION FOR' THE DEGREE OF DOCTOR OF PHILOSOPHY of IMRE G. CSIZMADIA D i p l . Chem. Eng. Polytechnical U n i v e r s i t y of Budapest, 1956 M.Sc, University of B r i t i s h Columbia, 1959 FRIDAY, SEPTEMBER 28, 1962, AT 2:30 P.M. IN ROOM 261, CHEMISTRY BUILDING COMMITTEE IN CHARGE Chairman: F„H„ SOWARD J S E , BLOOR W.A. BRYCE J,B 0 FARMER L„D. HAYWARD J . P. KUTNEY c A 0 MCDOWELL D CE„ McGREER C.-REID External Examiner: P. de MAYO University of Western Ontario SYNTHESIS AND PHOTOLYSIS OF AROMATIC NITRATE ESTERS ABSTRACT N i t r a t e esters of aromatic alcohols were synthesized by e s t e r i f i c a t i o n which involved competition between 0-n i t r a t i o n and aromatic C - n i t r a t i o n . TLC analysis gave a pattern of adsorption a f f i n i t i e s for the ni t r o x y group and other substituents consistent with the molecular conformations. The NMR frequency of the CC-protons showed a l i n e a r c o r r e l a t i o n with the accepted group e l e c t r o n e g a t i v i t i e s of the substituents i n molecules with r i g i d carbon skeletons and gave a value of 4.18 kcal/mole f o r the ni t r o x y group. The symmetric and asymmetric IR st r e t c h i n g frequencies of ni t r o x y groups i n d i l u t e cyclohexane so l u t i o n were s h i f t e d to higher values by s t e r i c i n t e r a c t i o n between contiguous groups when the C-QNQ2 bonds were constrained to coplanarity. The UV spectra showed benzenoid, TC—*-TC*, and n-»ir* bands' and a solvent perturbation e f f e c t assigned to a solvent • —^-solute charge-transfer i n t e r a c t i o n . ;' The n i t r a t e esters reacted with the solvent when i r r a d i a t e d i n s o l u t i o n i n the wavelength range of the n — e x c i t a t i o n . Product analysis indicated that C-C bond cleavage occurred v i a intermediate alkoxyl r a d i c a l s . Rate studies showed the following order of r e a c t i v i t y : benzyl n i t r a t e < dl-hydrobenzoin d i n i t r a t e ^EtOH > kPhH. On the basis of product analysis, rate measurements, estimated quantum y i e l d s and ESR spectra a mechanism for the n i t r a t e ester photolysis was proposed. GRADUATE STUDIES F i e l d of Study; Organic Chemistry Quantum Chemistry..................J.A.R. Coope S t a t i s t i c a l Mechanics ........3......R.F. Snider C r y s t a l Structures ..................K.B. Harvey L.W. Reeves Physical Organic Chemistry R. . Stewart R.E. Pincock Molecular Rearrangements...... A. Rosenthal Recent Synthetic Methods .R.A. Bonnett D.E. McGreer Chemistry of Polysaccharides G.G.S. Dutton Related Studies: Biochemistry M. Darrach J. Polglase S.H. Zbarsky D i f f e r e n t i a l Equations.. . . . J . Abramowich Computer Programming H. Dempster Analogue Computers E.V. Bohn (ii) ABSTRACT Nitrate esters of aromatic alcohols have been synthesized success-fully by direct esterification which involved competition between ^ -nitration andr aromatic C-nitration. The physical properties of the fully characterised nitrate esters have been studied by TLC and by NMR, IR, and UV spectroscopy,, TLC results gave a consistent pattern of adsorption affinities for the nitroxy group and other substituents and also revealed stereochemical features of the nitrate esters. The NMR frequency of the OC—protons in substituted carbinoljte, R'R"CHOX, showed correlation with the accepted group electronegativities of X i f the system had a rigid carbon skeleton. The correlation failed however when free rotation was possible on the common bond in substituted diols. The correlation also showed the nitroxy group electronegativity to be higher (ca. 4.18 kcal/mole) than hitherto reported. The characteristic infrared frequencies of nitroxy groups were measured in dilute cyclohexane,solutions of the nitrate esters and showed shifts to higher frequencies due to steric interaction between contiguous nitrate ester groups when the C-ONC^ bonds were constrained to coplanarity. The UV spectra of the aromatic nitrate esters showed benzenoid, It—»fC and n — - bands and a solvent perturbation effect which was tenta assigned to a solvent—*• solute charge—transfer interaction similar to that previously reported for aliphatic nitro compounds. Fhotolytic experiments revealed that nitrate esters underwent photochemical reaction when irradiated in the wavelength range of the n—*-1t excitation. Product analysis indicated that C-C bond cleavage occurred ( l l i ) which was explained i n terms of intermediate alkoxyl r a d i c a l formation. Rate studies of the photolysis showed the following order of reactivity? Benzyl Nitrate ^tOH > kPhH On the basis of product analysis, rate and quantum y i e l d measure-ments and ESR spectra a mechanism for the nitrate ester photolysis was proposed. (ma) A D D E N D A Pages 63-66. The 1,2-acenaphthenediol dinitrates were f u l l y character-ised "by the author m a previous research (68). The nitrogen analyses (Table XII) and IR spectra (Figure 9) of the crude by-products (£ , B , C and C ) were examined in an attempt to establish the trans^ cis trans occurrence of ring-nitration and complete characterisation was not undertaken. Pages 101 and 103. The identities of substances E, E' and E'' given on page 101 are not correct. The bands at 2900 and 1726 cm~l (Figure 2k) pointed to the presence of aldehyde groups m conjugated bond systems. Pages 120 and 121. The energy transfer process suggested i n paragraph 3 would only be probable i f the (undetermined) energy levels were related as shown m Figure 30. The observed spectra (68) implied this would be an "uphill" energy transfer. This is a rarely observed phenomenon, but cannot be entirely excluded. Very efficient u t i l i s a t i o n of n-*^ .* energy in the photodecomposition might shift the equilibrium toward this form. Alternatively, and more probably, there may be transfer to some state of lower energy than the T T * state. This might be a charge-transfer state, from which direct decomposition is possible. Page 142. Other factors including the presence of the relatively heavy atoms 0 and N and the altered symmetry of the molecule could also be expected to reduce the lifetime of the t r i p l e t state sufficiently to make the ESR signal undetectable. Pages 152 and 153• The identities of Fractions A, B, and C were not satisfactorily established since the elementary analyses were either lacking or did not agree with the calculated values. ( X V > ACKNOWLEDGMENTS I wish to express my very sincere thanks and appreciation to Dr. L. D. Hayward for his help and encouragement throughout the course of the research and the preparation of this thesis. I wish also to express sincere gratitude to the Head of the Department, Dr. C. A. McDowell for his continuing interest i n the work and for making available the ESR f a c i l i t i e s of his laboratory. My thanks are also due to Dr. J . B. Parmer of this Department for help with the analysis of the ESR spectra, to Dr. D. H o l l i s of Varian Associates, Palo Alto, who at miy; request, kindly recorded the f i r s t ex-ploratory ESR spectrum of an irradiated nitrate ester, and to the students and s t a f f of the Department ESR laboratory for their valuable technical assistance. For helpful discussion on theoretical and spectroscopic questions I am indebted to Dr. J * Bloor of the B.C. Research Council and to Drs. D. McGreer and L. ¥. Reeves of this Department. For mass-spectral analysis I wish to thank Dr. D* C* Frost. F i n a l l y , I wish to express my appreciation to the National Research Council of Canada for a studentship for the period 1960-62. TABLE OP CONTENTS TITLE PAGE ( 1 ) ABSTRACT (n) ACKNOWLEDGMENTS^ ....*.... (iv) TABLE OP CONTENTS (v) LIST OP FIGURES (vn) LIST OF TABLES W GENERAL INTRODUCTION 1 HISTORICAL INTRODUCTION 4 The Chemistry of the Nitrate Esters 5 I. (R1) XN02 Compounds 6 II. Properties of the Nitrate Esters 13 III, The Structure of Nitrate Esters 15 IV. Syntheses of Nitrate Esters 16 V. Reactions of Nitrate Esters 20 A. Electrophi]ic Substitution on Oxygen (Sg„) 23 B. Nucleophilic Substitution on Carbon (S^/ and Nitrogen (S ) 26 C. Olefin (EQJQ) and Carbonyl (E C_ Q) Elimination Reactions .TT; 32 D. Homolytic Decomposition of Nitrate Esters........... 36 The Photochemistry of the Nitrate Esters and Related Compounds ....•••••.»••••••.......... 41 RESULTS AND DISCUSSION 58 I. Synthesis of Aromatic Nitrate Esters 59 II. Chromatography of Nitrate Esters • 67 III. Analysis of Aromatic Nitrate Esters 75 IV. Spectra of Aromatic Nitrate Esters 76 A. Nuclear Magnetic Resonance: Spectra (NMR) 76 B. Infrared Spectra (IR) 81 C. Ultraviolet Spectra (UV) 84 (vi) V. Photolysis of Aromatic N i t r a t e Esters 90 A. Preliminary Experiments ............... 90 B. I d e n t i f i c a t i o n of Photolysis Products • 97 C. K i n e t i c Study of the Photolysis .... .......... 113 D. ESR Study of N i t r a t e Ester Photolysis 130 E. Summary of Proposed Reaction Mechanism 142 EXPERIMENTAL 146 I . Materials 147 A. Solvents * 147 B. Reagents 148 C. Reference Compounds 148 D. Aromatic N i t r a t e E s t e r s . . . 153 ( i ) S t a r t i n g Materials 153 ( i i ) Aromatic N i t r a t e Esters v i a D i r e c t N i t r a t i o n 158 ( m ) Aromatic N i t r a t e Esters v i a Exchange Reactions 162 I I . Analyses 165 A. Melting Point Determinations 165 B. Elementary Analyses 165 I I I . Spectra 165 A. U l t r a v i o l e t Spectra (UV) 165 B. Infrared Spectra (IR) 165 C. E l e c t r o n Spin Resonance Spectra (ESR) 166 D. Nuclear Magnetic Resonance Spectra (NMR) 166 IT. Chromatography 167 V. Photolyses • 169 A. Light Sources and Apparatus 169 B. Preliminary Experiments 174 C. K i n e t i c Experiments 177 D. I s o l a t i o n and I d e n t i f i c a t i o n of Photoreaction Products 179 REFERENCES 185 ( v n ) LIST OP FIGURES 1. Calculated Ionic Character of X-NO2 Bonds 11 2. The Structure of the Organic N i t r a t e Group 17 3. Resonance Structures of the Nitroxy Group 17 4. Modes of S c i s s i o n of the Nitroxy Group 22 5. A. Energy Levels of Molecular'Orbitals and Possible E l e c t r o n i c Transitions f o r N i t r i t e Esters 44 B. Typical E l e c t r o n i c Spectrum of a N i t i * i t e Ester (2-butyl n i t r i t e i n ether (95) ) 44 6. A. Energy Levels of Molecular O r b i t a l s and Possible E l e c t r o n i c Transitions f o r N i t r a t e Esters 44 B. T y p i c a l E l e c t r o n i c Spectrum of a N i t r a t e Ester (2-butyl n i t r a t e i n ethanol (95)) 44 7. C o r r e l a t i o n o f Z-ONO and X-N02 Bond Length i n ( R X ) n XN02 Compounds 52 8. Thin-layer Chromatography of N i t r a t i o n Products from c i s -and trans-l f2—Acenaphthenediols. 65 9. iE-Spectra of c i s — and trans-1,2-Acenaphthenediol N i t r a t i o n Products 66 10. Chromatographic Patterns of Representative N i t r a t e Esters 72 11. C o r r e l a t i o n o f foijj with Group E l e c t r o n e g a t i v i t i e s i n (A) trans-1,2-jCyclohexane- (B) Benzyl- (C) trans and (D) cis-1,2-Acenaphthenyl-Denvatives 78 12. Tentative C o r r e l a t i o n of Tbc^ and Mechanism i n the Reaction of N i t r a t e Esters with Pyridine at 25 80 13. UV Spectrum of iso—Amylnitrate m Methanol ............ 85 14. Benzenoid Absorption of Benzyjj&lcohol (A) Benzyl-Hitrate (B) i n Hexane Sol u t i o n 86 15. Difference Spectrum of Benzyljtiitrate and Benzyl|&lcohol m various Solvents 88 16. C o r r e l a t i o n of the Frequency of Band I with the Solvent I o n i z a t i o n P o t e n t i a l 89 17. C o r r e l a t i o n of the Frequency of Band I I with the Solvent I o n i z a t i o n P o t e n t i a l 89 18. Chromatographic Separation of Photolysis Products of Ni t r a t e Esters 92 (vm) 19. Chromatographic Separation of Photolysis Products of N i t r a t e Esters 93 20. Chromatographic Separation of Photolysis c/f Products of N i t r a t e Esters 96 21. Chromatographic Separations of Products from Photolysis of meso-Hydrobenzom D i n i t r a t e i n Benzene Sol u t i o n .... 99 22. Phenolic Products Isolated from Photolyzed i i n C^H^) me s o-Hydr obenzo i n D i n i t r a t e 100 23. Infrared Spectra of meso-JIydrobenzoin D i n i t r a t e (A) Nitrobenzene (B) and Photolysis Products of A(C and D) 102 24. Spectra (IR) of 2,4-Dinitrophenol (A) and Photolysis Products from meso-Hydrobenzom D i n i t r a t e (E, E 1 and E") .....•••«•»••»•.•».......•........•.•.••....••••... 103 25. Infrared Spectra of 2,6-Dinitro-4-Phenylphenol (A) and P h o t o l y t i c Products from meso-Hydrobenzom D i n i t r a t e (F, K and N) • 104 26. NMR Spectra of 1,2-Diphenyl ethane d e r i v a t i v e s ........ 106 27. Possible Mechanism of Photolysis of meso-Hydrobenzoin D i n i t r a t e 112 28. Primary Photochemical Reactions of N i t r a t e Esters (NE). 115 29. Rates of Photoreactions of (A) Benzyl N i t r a t e , (B) d l - and (C) mesD~Hydrobenzpin J)initraje£, (D) tr a n s - and 7E) cis-1,2-Acenaphthenediol D i n i t r a t e s i n Benzene Solu t i o n at 24.2°C 117 30. T r i p l e t - T r i p l e t Energy Transfer Between Benzophenone and Naphthalene and S i n g l e t - S i n g l e t Energy Transfer w i t h i n 1,2-Acenaphthenediol D i n i t r a t e s 121 31. Rates of Photoreactions of meso-Hydrobenzoin D i n i t r a t e (A) and Benzyl N i t r a t e (B) i n Ethanol and of meso-Hydrobenzom D i n i t r a t e (C) i n Ether at 34.2 C 123 32. Energy Level Diagram f o r N i t r a t e Ester - Solvent Complex E x c i t a t i o n 125 33. Light Energy Emitted by Source (A) and Absorbed by meso-Hydrobenzom D i n i t r a t e ^B) and Solvent Benzene (C) i n the Photoreaction at 24.2 C 127 34. Steady State ESR Spectra of I r r a d i a t e d cis-(A) and t r a n s -(B) 1,2-Acenaphthenediol D i n i t r a t e s and meso-(C) and dl-Hydrobenzom D i n i t r a t e s i n Benzene.Solution.at.Room.. Temperature............. 131 (ix) 35. ESR Signals of Irradiated trans-1,2-Acenaphthenediol Dinitrate Obtained Initially (A) and After Ten Days in the Dark (B and C) 132 36. ESR Spectrum of NOg (A) in Solid Argon and (B) Generated from trans-1f2-Acenaphthenediol Dinitrate m EPA at 77°K. 134 37. ESR Spectra of Irradiated .trans-1,2-Acenaphthenediol Dinitrate (0.1 M) (/) and N0 (0.2 M) (B) in Benzane Solution 136 38. ESR Spectrum and Components of Irradiated trans-1,2-Ace-naphthenediol Dinitrate 138 39. Rates of Generation of Components of ESR Spectrum of Irradiated Nitrate Ester 140 40. ESR Spectrum of Acenaphthene It-•It Triplet State m EPA 143 41. Proposed Mechanism of Nitrate Ester Photolysis in Solution 144 42. Infrared Spectra of Ethylcarbonates of 1,2-Diphenylethane Derivatives 157 43. Flow Sheet of Separation of Nitration Products from trans-1,2-Acenaphthenediol 159 44. Reported Power Output and Observed Spectral Distribution for G.E.-^ H85- A3 Medium Pressure Mercury-Arc Lamp 170 45. Reported Power Output for G.E.-A-H6 High Pressure Mercury-Arc Lamp ............... 170 46. Light Energy Emitted by Han#!ovia 100 watt High Pressure Mercury Arc Lamp 171 47. Photoreactor 173 48. TLC of trans-1t2-Acenaphthenediol Dinitrate Photoiysed in Benzene Solution 175 49. ESR Tube 175 50. Solid State Infrared Spectra of,trans-1,2-Acenaphthenediol Dinitrate and one of its Photolysis Products 176 51. TLC of trans—l»2-^Acenaphthenediol Dinitrate after Photolysis in Benzene Solution 178 52. Spectrum of Acetaldehyde from Photolysis of meso-Hydrobenzom Dinitrate in Ethanol. 183 (x) LIST OP TABLES I. Second I o n i z a t i o n P o t e n t i a l , E l e c t r o n A f f i n i t y and E l e c t r o n e g a t i v i t y of 9 I I . Estimated Ionic Character of X-N Bonds i n (R 1) XNOg .... 9 I I I . Molecular Parameters of the ONO2 Group 15 IV. Reaction Mechanisms 22 V. E l i m i n a t i o n Reactions with Nucleophilic Reagent OH i n 9Uf° Aqueous Ethanol S o l u t i o n 33 VI. Isotope E f f e c t s i n the Reaction of Benzylnitrate with Sodium Ethoxide i n Abeolute Ethanol at 60.2 34 V I I . Log - Frequency Factors and A c t i v a t i o n Energies f o r Thermal Decomposition of N i t r a t e Esters 39 V I I I . The E f f e c t of Deuterium S u b s t i t u t i o n on Burning Rates of N i t r a t e E s t e r s . 40 IX. Strvdwral'-Prqarties of Nitrogen-Oxygen Compounds 50 X. n i f * ' ^ * T ransitions of (R^XNO Compounds 51 X I . Fundamental Infrared Frequencies (cm"1) of (R^^XNC^ Compounds 51 X I I . N i t r a t i o n Products from c i s - and trans-1,2-Acenaphthenediqls 65 X I I I . Relationship of R^ Values and Structure i n Polynitroxy Compounds 70 XIV. Chromatographic Constants f o r N i t r a t e Esters ........... 73 XV. Combustion Analyses of P o l y n i t r a t e s 75 XVI. Group E l e c t r o n e g a t i v i t i e s and T -values f o r Oj-Hydrogens a n Ni t r a t e Esters and Related Compounds * 77 XVII. Infrared Frequencies of Nitroxy Groups from Condensed State Spectra 82 XVII I . Infrared Frequencies of Nitrnxy Groups from S o l u t i o n Spectra 83 XIX. Solvent E f f e c t s i n D i f f e r e n t i a l Spectra of Benzyl N i t r a t e and Benzyl Alcohol , 90 (xi) XX. Photolysis of Nitrate Esters in Benzene Solutions: Preliminary Experiments 94 XXI. Photolysis of Nitrate Esters in Oxygen-free Absolute Benzene Solutions 95 \ XXII. Apparent First-Order Rate Constants for the Photolysis of Aromatic Nitrate Esters at 24.2 C 119 XXIII. Calculated Ratios of Charge-Transfer Equilibrium Constants for Benzyl Nitrate m Solution 125 XXIV. Calculated Values of for a-Fhenyl Substituted Nitrate Esters in Benzene and in Ethanol and Ether Solutions ... 128 XXV. Calculated Quantum yields for the Photolysis of a-Phenyl Substituted Nitrate Esters in Three Different Solvents at 24.2°C 129 XXVI. Observed Components of ESR Spectra of Irradiated Nitrate Esters 138 XXVII. Melting Points and NMR Spectra of 2,4-Dimtrophenyl-hydrazones 149 XXVIII. Melting Points and NMR Spectra of Nitrophenols 151 XXLX. Reduction Products of Benzil 154 XXX. Attempted Syntheses of Aromatic Nitrate Esters from Cyclic and Ethyl Carbonates 164 XXXI. Adsorbents and Supporting Materials for Chromatography.. 167= XXXII. Solvents for Chromatography 168 XXXIII. Spray Reagents for Chromatography 168 GENERAL INTRODUCTION - 2 -Synthetic photochemistry u t i l i z e s l i g h t energy to bring about specific chemical reactions. C l a s s i c a l examples are the hydrogen chloride synthesis i n the gas phase and the vitamin synthesis i n the l i q u i d phase. There are, m addition, numerous other photosynthetic reactions which may be used for the preparation of a variety of compounds in excellent yields whose syntheses by conventional chemical techniques would involve lengthy and tedious routes from available starting materials ( l ) . During the i r r a d i a t i o n of a chemical system particles of matter suffer " c o l l i s i o n s " with the photons of l i g h t and i n the annihilation of the photons the molecules absorb the energy of the li g h t quanta (€»V»v ) prooessecanjbecclassifa&d^asochenucal. In p\ photochemical reactions the primary fragments of the decomposition are free radicals. Most of the free radical intermediates possess very short lifetimes and their structure and r e a c t i v i t y determine the further steps i n the reactions. The properties of the e l e c t r o n i c a l l y excited molecule usually d i f f e r from those of the same molecule i n the ground state and thus the excited molecule may undergo reactions with chemical reagents to which - 3 -i t would be resistant i n the ground state. Therefore detailed study of the nature of photochemical reactions gives information not only about the nature of the free radicals formed i n the primary reaction but also furnishes valuable information about the chemical bond, Photosynthetic processes occur d a i l y i n every green l i v i n g organism m the presence of lig h t and raw materials. Much ef f o r t has been directed toward achieving understanding of these processes which occur i n the green chloroplasts of plants (3). Recently an I t a l i a n school reported investigations (4) (5) of the i n v i t r o photosynthesis of amino acids based on the photolysis of nitrate (N0^~) and n i t r i t e (NC^-) ions i n aqueous solutions of r e l a t i v e l y simple organic molecules e.g. glucose. Consequently there i s a demand for a better understanding of the photolytic processes of nitrogen-oxygen compounds from both theoretical and pr a c t i c a l points of view. This thesis i s concerned with the synthesis and photolysis of a certain type of aromatic nitrate esters. The aim of this research work was twofold. F i r s t l y i t was hoped that radicals liberated by the photolysis of the aromatic nitrate esters i n solution might have lifetimes s u f f i c i e n t l y long that t h e i r characteristic properties could be studied. Secondly i t was hoped that the fragmentation pattern of the nitroxy group during photo-decomposition might be determined which i n turn would throw l i g h t on the chemical and physiological properties of the several classes of organic compounds which contain oxygen-nitrogen bonds. HISTORICAL INTRODUCTION The Chemistry of the Nitrate Esters - 6 -I . (R 1) XNCV, Compounds, n £ The group as a substituent plays a pa r t i c u l a r l y important role i n chemistry. The compounds in which i t occurs (II) may be considered as derivatives of hydrides (I) i n which a hydrogen attached to a particular atom (X), has been substituted by NC^ ( l ) . (R 1) XH * (R 1) XNCL (l) n n 2 — I II where = I, II, III, ... .. represents the type of the substituents and n = 0 , 1 , 2 , or 3 represents the number of the substituents. Thus, most conveniently, compound II may be termed an X-nitro derivative. This c l a s s i f i c a t i o n may be extended i n principle to the whole periodic system. As examples the alk y l hydrides and the corresponding X-nitro derivatives for the second period elements are as follows: LiH RBeH R V B H R ' R V C H R'R"NH ROH EH* LiN0 2 RBeN02 R'R'BN0 2 R 'R'V'CNC^ R'R"NN0 2 R0N02 FN0 2 It i s interesting to note that at the beginning of the period NC"2 bears a negative charge while at the end the positive charge i s on the N02 group and that there are intermediate distributions of charge over the X - N bond within the period. The formation of the X - N bond may be considered as a combination of free radicals which may proceed by one of three different routes: (R 1) X'+^N0o (R 1) X + + :NC~ ( 2 ) (R 1) X- + -NCL > (R 1) X:N0o ( 3 ) n 2 n 2 W (R 1) X.*+"'N0o • (R 1) X' + Not (4) n 2 'n 2 N — Although one may distinguish these three reactions i t should be - 7 -emphasized thai: 3. i s "the general case resulting i n a covalent bond* The bond formed, however, may s t i l l possess a certain degree of ionic character. The extreme cases are 2^ and 4_. The amount of ionic character i n a given bond may be estimated (6) from Pauling*s equation (5) for the reaction A« + -B *- A-B Amount of ionic character = l - e ~ 4 V , " r D ) (5) where i- - if- i s the difference between the electronegativities of the A a radicals A« and B* expressed i n the units of Pauling's scale (kcal/mole). The amount of ionic character of the X-N bonds m (R 1) XN0„ n 4 compounds may be estimated i f one takes i n the f i r s t approximation '/•^ as the electronegativity of X* rather than of (R 1) X«• The electronegativity of the N02 r a d i c a l i n Mulliken's scale (ev/particle) i s the sum of the ionization potential and electronaffmity of NC^ and i t s equivalent i n Pauling's scale (kcal/mole) may be calculated (7) from equation j5: i&i ji? = 0.168 ( iM - 1.23) ^ where the subscripts P and M refer to Pauling's and Mulliken*s scales respectively. D i f f i c u l t y arises however in the selection of the value of the ionization potential of nitrogen dioxide among the several which have been an a I acfron in reported so far« Each value corresponds to the ionization of^a particular molecular or b i t a l (8). The f i r s t ionization potential has been re-determined recently (9) as 9*80 + 0.05 ev while the published second ionization potentials range between 11 and 12.3 ev; thus the reported value of 11.62 ev (10) may be accepted as reasonable. In the calculation of the electro -negativity of N0 2 the f i r s t ionization potential should be used since i t has been assigned as the ionization of the unpaired electron. For some - 8 -unknown reason, however, the second ionization potential provides more reasonable values for the estimated p a r t i a l ionic character of X-N bonds (Table I I ) . For example, the f i r s t ionization potential gives about 8CfS ionic character for the F-NO^ bond whereas the second ionization potential gives the value of 6O9S which i s i n better agreement with the chemical properties of the compound. The selected constants for NO^ are given i n Table I. - 9" -Table I Second Ionization Potential, Electron A f f i n i t y and Electronegativity of NO. Experimental Theoretical Theoretical after (10) (11) (12) zero adjustment I p (ev) 11.62 12.89 11.62 E a (ev) - 4.12 2.85 I + E (ev) p a - - 14.47 *P (kcal/mole) - - 2.23 The corresponding values i n Pauling's scale (13) and the values of ^ - /^ NO^ together with the estimated ionic character of the X-N bonds are given i n Table I I . Table II Estimated Ionic Character of X-N Bonds i n (R 1) XNO, Atomic Number X * (kcal/mole) " ^N02 (kcal/mole) Ionic Character of X-N02 3 L i 0.55 -1.68 0.50 4 Be 1.10 -1.13 0.27 5 B 1,85 -0.38 0.04 6 C 2.50 +0.27 0.01 7 N 3.15 +0.92 0.19 8 0 3.60 +1.37 0.37 9 F 4.15 +1.92 0.60 From Reference (13). - 10 -If Pauling 1 s theoretical curve for ionic character (5_) i s extended over the negative side of the scale, one may estimate the magni-tude of the positive or negative charge existing on the nitrogen atom i n the (R 1) XNO, compounds as shown i n Figure 1. The two extremes are singular cases i . e . there i s only one lithium n i t r i t e and only one n i t r y l fluoride, however as the middle of the period i s approached the number of p o s s i b i l i t i e s are increased. At the geometrical mid-point the Cj-nitro derivatives usually called "nitro compounds" have the smallest amount of ionic character i n the C-NO2 bond and are the most thoroughly studied N0 2 derivatives. Because of the possible variation of the substituents R', R", R1" the number of possible C-nitro compounds i s v i r t u a l l y unlimited. On the l e f t hand side of the period only L i N 0 2 is known so f a r , which i s a simple inorganic compound. The beryllium and boron compounds have not been reported as yet but there seems to be no obvious reason why they should not e x i s t . It might be possible that the known mixed alkyl or aryl beryllium (14) (15) and boron halides (16) (17) would undergo a metathetical reaction with s i l v e r n i t r i t e as i n 7 and 8_s R-BeHal + AgN02 ^R-Be - N 0 2 + AgHal (7) 6+ 5-R'RMBHal + AgN02 *• R1 R"Bf- N02 + AgHal (8) - 12 -The NCv, i n AgNC^ would be a nucleophilic reagent so that the reactions may be considered as nucleophilic substitutions on Be or B (18). These compounds containing elements from the l e f t side of the periodic system probably would constitute "chemical mirror images" of their counterparts from the right hand side. In other words the s t a b i l i t y of the above hypothetical beryllium and boron nitro compounds would be expected to depend on the successful creation of a p a r t i a l positive charge on the metalloid element (especially i n the case of beryllium) and t h i s would be a function of the nature of the R1, groups. Prom this reasoning i t follows that groups with higher electron withdrawing power (such as aryl or per-fluoroalkyl) would tend to s t a b i l i z e these compounds. Two pieces of indirect evidence seem to support this idea. ^ J - N O , F 3 C x B / N 0 2 F a C ^ N O g 5 _ ^ > 0 2 • RON02 + HT ( l l ) Method (l) i s a favoured technique because the corresponding halogen compounds are readily available (35) i n many cases and the by-product s i l v e r bromide can be Separated easi l y * The reaction may be carried out i n either homogeneous (in g l a c i a l acetic acid or acetonitrile) or i n heterogeneous (in ether, benzene, nitrobenzene or nitromethane) medium. In the case of synthesis of d i n i t r a t e s , d i f f i c u l t i e s arise i n the exchange of the second v i c i n a l halogen atom which was indicated c l e a r l y by Fishbein (36) i n a study of the reaction of racemic- and meso-dibromobutane (VIII) with s i l v e r - 1 7 -FIGURE 3. Resonance S t ruc tu res of the N i t roxy Group - 18 -nitrate i n a c e t o n i t n l e . The reaction occurred i n a stepwise manner and the exchange of the f i r s t bromine atom for nitroxy group was considerably faster than that of the second one. Furthermore the f i r s t step proceeded with retention of configuration and the second with inversion. The f i r s t step was explained by a push-pull mechanism where "the neighbouring bromine atom participates i n a back side internal displacing action while the C-Br bond of the carbon undergoing substitution i s being weakened by an e l e c t r o p h i l i c attack on halogen by s i l v e r " (IX) (12). The c y c l i c bromonium ion (X) i s attacked by the nitrate ion and with overall retention of configuration produces threo-2-bromo-3-nitroxy butane (XI) (13). Since the nitrate ester group does not participate i n the elimination of the second bromine atom v i a a c y c l i c ester therefore the next step proceeds by simple S^2 mechanism Method ( i i ) was introduced f i r s t by Boschan (32) m 1959» This technique has the advantage that i t "does not involve rupture of the bond on the carbon atom adjacent to the nitrate ester group" therefore the nitrate ester obtained retains the configuration of the parent alcohol. The r e -action is considered to proceed v i a an intermediate which readily decomposes to the f i n a l product (l_5). Further development of this reaction i s now i n progress (38), The chloroformate starting material is usually prepared from the corresponding alcohol and phosgene. For this reason dinitrates may not be prepared where the geometrical location of the two hydroxyl groups favors the formation of cyOlic carbonates. with p r a c t i c a l l y complete inversion of configuration (14). - 1 9 -Br H i I H U C — C — C — C H T H Br VIM (dl ) Ag NO3 • Br:-) H i ~> I f-UC-C- •CH-I Ag IX NO-H 3 C B r / \ C — C l I H H CH- NO3 + A g B r (12) C H . / ,C — B r Nc-r \ NO; CH-Br H,C / H H / • V C H 3 (13) X XI C ? 3 B r 0 2 N O \ NO: O 2 N O -° 2 N O \H C K , ^ 3 H 0 2 N 0 ^ C / + Br ° 2 N O \ H , C -+H (14) XI XII XIII JZH-O CI A g N Q 3 <1> O 1 .c. > H - O 0 Co R 2 . / \ AgC I C H - 0 + C O , N - O * /' " O (15) •O- + N 0 2 Y + / R - O : N O ; H / i o 2 R—O: + HY (16) - 20 -Method ( i i i ) is the most widely used to synthesize organic nitrate esters* The reacting species is N02 as in aromatic nitration (C-nitration) "where the attack by the substituting agent is on the un-saturation electrons, i.e. conjugated carbon 2p electrons. Since non-bonding electrons of nitrogen and oxygen oan participate deeply in such conjugation we should expect them to share many properties with the unsaturation 2p electrons of carbon including general vulnerability to electrophilio substituting agents" (39). Since the mechanism (16) is closely related to that of C_-nitration the: originally suggested (40) O-nitration term is retained (41). There are a number of reagents which possess the structure NOg I" required for O-nitration and consequently a great diversity of preparative procedures. The reaction is usually carried out at or below 0°C with one of the following reagents (a) HNO^ f (b) HNOj/HOClj| (c) HN03/H2S04f (d) HN03/Ac20) (e) HNO^ AOjO + AcOHj (f) N^/fcCC!^ (g) gaseous ^ 0,.} (h) NOgCl (with or without catalyst) j (i) NOgP (with or without catalyst). These and other special synthetic routes were extensively reviewed recently (27) (24) (42), A more detailed examination of the various nitration mechanisms i s made i n the ohapter on "Results and Discussion." V. Reactions of Nitrate Esters. Since the reactions of organic nitrates depend on the structure of the nitrate ester as well as on the reagent and reaction conditions and often two or more types of reaction take place simultaneously only a few mechanistic studies have so far been reported (43). It has been pointed out recently (44) that a l l of the chemical transformations on - 21 -record may be considered as involving one or more of five possible modes of scission of the ester group. Reactions which cause such bond cleavages range from solvolysis and hydrolysis through ca t a l y t i c hydrogenolysis to photochemical decomposition and explosions. The f i r s t three of the five modes of scission (Figure 4) are heterolytic while the la s t two are claimed to be homolytic. In other words a l l reactions which proceed v i a an ionic mechanism seem to cause bond cleavage i n the nitrate moiety by one of modes 1, 2 or 3, while r e -actions occurring by free radical mechanism seem to involve mode 4 or 5. The f i r s t four modes are well established, while mode 5 i s tentative as yet and i s based on indirect evidence. Reagents causing ionic decomposition (modes 1, 2 and 3) may be either e l e c t r o p h i l i c or nucleophilic and the reactions which occur may be substitutions or eliminations. The substitution may be el e c t r o p h i l i c (Sg) or nucleophilic (S^.) according to the nature of the reagent. The elimination reactions may be c l a s s i f i e d as carbonyl elimination (E p n) and o l e f i n i c elimination (E„ _) depending on whether carbonyl or o l e f i n i c compounds are the f i n a l products (43). Ionic reactions of the above types are summarized i n Table IV. It i s interesting to note that e l e c t r o p h i l i c reagents i n sub-s t i t u t i o n ( A, which may be charged e.g. H + or uncharged e.g. Lewis acids) always attack the electron r i c h oxygen atom while nucleophilic reagents m substitution ( B; , which may be charged e.g. H0~ or E'tO or uncharged e.g. pyridine) always attack the carbon neighbouring to the NO^ or the nitrogen of the nitroxy group which indicates that they must be more electron deficient than oxygen. - 22 -1 2 3 4 5 ? F I G U R E 4 M o d e s o f S c i s s i o n o f t h e N i t r o x y G r o u p T A B L E IV R e a c t i o n M e c h a n i s m s R e a c t i o n M o d e o f M e c h a n i s m s S c i s s i o n 1 E l e c t r o p h i l i c s u b s t i t u t i o n o n o x y g e n - . O O A O Y A~^ \ + N ° 2 :C> R • • X R 2 N u c l e o p h i l i c s u b s t i t u t i o n o n c a r b o n : B : ^ * > C:-ON0 2 B ; C + NO3 S N C 3 N u c l e o p h i l i c s u b s t i t u t i o n o n n i t r o g e n : O v O B./—XN ^ B : N 0 2 + R O b N N J O R 4 ^ - H y d r o g e n e l i m i n a t i o n B : + H - C — C - O N C ^ — - B : H * + > C = C < + NO3 E C = C 1 5 o c - H y d r o g e n e l i m i n a t i o n : B : + H - C — O N 0 2 * B : H * + > C = 0 + N O ^ ^-Q=0 ^ - 2 3 -A. E l e c t r o p h i l i c Substitution on Oxygen (S^Q ) E l e c t r o p h i l i c substitution by proton (protonation) of nitrate esters may be considered as the reverse reaction of_0-nitration (17)* The reaction occurs when a nitrate ester is dissolved i n concentrated s u l f u r i c acid and the equilibrium i s very l i k e l y shifted far to the r i g h t (the alcohol may react further with the concentrated acid) since the o r i g i n a l nitrate ester may not be recovered by addition of water. Furthermore u l t r a v i o l e t spectroscopy indicated the presence of NO* and the absence of the original ester i n the aqueous solution ( 4 2 ) . Because of this phenomenon nitrate esters may be used as n i t r a t i n g agents ( 4 5 ) which i s an obvious extension of the principle of n i t r a t i o n by NOgY since the more frequently used n i t r i c acid may be considered as the zero member of the nitrate ester family* It i s worthy of mention that certain nitrate esters with high mobility of NOg can exhibit mild n i t r a t i n g a c t i v i t y through s e l f ionization. Among these compounds are the nitrate esters of cyanohydrins (XIV, XV, XVI) which can be used-for n i t r a t i o n of sensitive compounds as was discussed recently m a review by Topchiev ( 4 6 ) . Nitrate esters also undergo el e c t r o p h i l i c substitution on oxygen ) by other agents (18)« A number of Lewis acids may be used as effective n i t r a t i o n catalyst with nitrate esters. Thus benzene and toluene are nitrated by ethyl nitrate i n the presence of aluminium chloride ( 4 7 ) . Just as protonation or other e l e c t r o p h i l i c substitution occurred on the oxygen atom of a nitrate ester i t should also occur on the element (X) of every (R 1) XML compound i n which the element X possesses the JX £ necessary electron density. In other words this should be true for - 24 -R — Q NOV •°;H N O , R — O ' NO. ( 1 7 ) H 3 C \ / C H 3 C OgNO \ ; N 0 2 N O CN O 2NO' X C N XIV XV XVI R 0 ^ + AICI 3 NOo AICI-R - o ( ^ N 0 2 + RO CI \ / A l / \ CI CI ( 1 8 ) (R') NX — N 0 2 * H — (R' ) N X H • NO* ( 1 9 ) - 25 -elements other than oxygen from the right hand side of the period (N, 0, P). Consequently the o r i g i n a l l y introduced S^ (electrophilic substitution on oxygen) i s a special form of £L^ (electrophilic substitution on element X, i.e.s E g N5 E g Q j E g p) (19). Since the NO* formed i n this reaction i s the ni t r a t i n g species i n every n i t r a t i o n therefore the a v a i l a b i l i t y of N0 + determines the n i t r a t i n g power of (R 1) XN0~ or more precisely that of the corresponding conjugate acid, (R 1) X"*^^^ n H It has been known since 1905 that n i t r y l fluoride (FNO,,) i s a powerful n i t r a t i n g agent (48) (49). This fact was confirmed i n a more recent publication (50) and the efficiency of n i t r y l fluoride as an _0-nitrating agent was compared (51) with that of n i t r y l chloride (CINO^) (52). It was pointed out that "generally there i s a greater y i e l d of nitrated material with n i t r y l fluoride than with the chloride, even when the l a t t e r i s used with a catalyst. This suggests that n i t r y l fluoride stands higher than n i t r y l chloride i n the series of ni t r a t i n g agents of the type XN02 i n which the n i t r a t i n g power increases with the electron accepting quality of X." ( 5 l ) , On t h i s basis one may assemble a preliminary series of n i t r a t i n g agents from the electronegativity estimations i n Figure 1 as shown i n equation 20 0 R3C-N02 -< R2N-N02 R0-N02 c< an S ^ reaction should take place. Under special circumstances ( « — t) both reactions may occur simultaneously. In practice, however, the number of reported S^. reactions i s much larger than the number of S^, reactions, indicating that the polarization of the 0-N bond i s much larger than that of the C-0 bond. "Alkylation with nitrate esters" i . e . the S x i n reaction, has been recently reviewed i n d e t a i l by Boschan and co-workers (42). One of the - 27 -c l a s s i c a l examples (54) of the S reaction i s the decomposition of benzhydryl nitrate ester with primary or secondary amines such as piperidme. On the basis of the reaction products isolated one may interpret the reaction according to equation 22. However this i s not the only reaction which occurs under these conditions, the carbonyl elimination proceeds simultaneously with the formation of benzophenone but to considerably smaller extent (3:l)« Weaker bases (aniline, benzylamme) on the other hand gave almost ex-c l u s i v e l y the corresponding product with N - C bond formation according to 22. In certain reactions such as basic and neutral hydrolyses and solvolyses of nitrate esters (55) (56) i t i s not obvious whether the S or mechanism i s operative since the products would be the same i n either case (23^ 24). For this reason Toffe and co-workers (57) studied the mechanism 18 of hydrolysis of several nitrate esters by means of the 0 isotope technique. According to their findings the mechanism of hydrolytic cleav-age may proceed by either mechanism 25_ or 26_ depending on the structure of the nitrate ester i n question. A change i n the hydrocarbon portion may cause a change i n the order of the reaction (S^2 -^ S^l) and furthermore such a change may also a l t e r the point of attack i n the alkaline hydrolysis of nitrate esters from S,^ . » S._T. These changes are summarized i n the series 27: N C N N — Fh- , n-alkyl, sec. a l k y l , t e r t , a l k y l , Ph^ C - , H- , V — V — S N C 2 — S N C 1 <2Z) Extensive investigation has been made of sugar and related poly-nitrates by treating them with nitrogen bases (such as pyridine, piperdme, -28-H Ph Ph H ONO-H Ph Ph -N-----C ONOu r i • H ( 2 2 ) H I N -Ph Ph H N O ; C 5 H | Q N H C 5 H | 0 N H 2 N O 3 + Ph / H .Ph PhCH 2ON0 2 + O H P h C H 2 O H + N 0 3 (23) t - B u O N 0 2 + 2 H 2 0 ^ t - B u O H + H 3 0 * + N0 3 ( 2 4 ) 1 8 O H + -HC . - O - N O 2 H 1 8 Q - C - + N 0 3 ( S N C ) ( 2 5 ) ^ C - 0 - N 0 2 + L 8 O H " • ^ C - O " + H L 8 O N O , •2 ( S N N } O R O O \ / N -L O R - - O R ( 2 8 ) - 2,9 -hydroxylamine> etc.) (24) (62). In a l l cases f u l l or p a r t i a l demtration produced the parent alcohol with retention of configuration. This proved that the C-0 bond was not ruptured and therefore the nucleophilic sub-s t i t u t i o n took place on the nitrogen (Sj^.). In general such a reaction involves the formation of the alcoholate anion as an intermediate (28) while the other component of the ion pair contains the N-nitro base cation. During the customary working up procedure i n aqueous solution the alcoho-late anion produced the corresponding alcohol. In certain cases special stereoselective demtration took place. Pyridine denitrated mannitol hexanitrate (58) sel e c t i v e l y at or i t s equivalent C^, at 25° (29). D u l c i t o l hexanitrate gave the pentanitrate by selective demtration (59) on and equivalent with pyridine at 50° (30). Cellulose 2, 3, 6-trimtrate denitrated at se l e c t i v e l y (60) with hydroxylamine i n pyridine ( 3 l ) . Similar results were obtained with methyl(3 —D-glucoside tetranitrate and hydroxylamine i n pyridine ( 6 l ) . In t h i s case (32) the 4-^-nitro group, not present m c e l l u l o s e , was also replaced by the hydroxyl group and to a greater extent than the 2-O-^iitro group. Isohexide ( i . e . isoidides 1, 4j3,6-di-0-anhydro-L-iditol (XVIII); isosorbidet l,4j3,6-di-0-anhydro-D-glucitol (XLX); and isomannide: 1,4;3,6-d 1-0—anhydro-D-mannito1 (XX)) dinitrates (62) and 1,2- cyclohexane-d i o l d i nitrates (XXI) (XXII) (63) reacted with boi l i n g anhydrous pyridine. Kinetic evidence (62) excluded the p o s s i b i l i t y of an S^, mechanism and instead of the parent d i o l s , as might be expected for reactions, poly-meric material was obtained as the chief carbon-containing reaction product. - 3 0 -ONOo I ^ H - C - H I 0 0 N O — C — H 2 I 0 2 N O - C - H H—C—ONOo t 2 H - C - O N 0 0 H - C - H C 5 H 5 N i 25°C O N 0 2 H - C - H I O o N O - C - H I H O - C - H I H - C — O N O -H - C - O H—C—H ONGv (29) ONO. ONO. ONC2 H-C-H H-C-ONO? I 0 0 N O - C - H * I CuNO-C-H 2 I H-C-ON0 2 H-C—H I ONOg C 5 H 5 N;t 50°C ONOg H-C-H H - C - O N O -H O - C - H 0 2 N O - C - H H-C-ONOg H - C - H ONOo O N 0 2 H-C-H I CuNO-C-H 2 I H-C —OH I H-C—ONOg O^NO-C-H H-C-H l O N 0 2 (30) (32) - 3 1 -- 33 -C. Olefin (EQ_Q) and Carbonyl (E^_Q) Elimination Reactions. The difference between these elimination reactions _33_ and _34, i s that the double bond formed i s between carbon atoms i n one case, and between carbon and oxygen atoms i n the other. This variation i n product formation r e f l e c t s a difference i n the detailed mechanism. In the case of o l e f i n elimination the hydrogen atom at the (b—position to the nitroxy group i s attacked by the nucleophile (33) and the reaction i s thus frequently called "(3-elimination" (64). Carbonyl elimination represents a nucleophilic attack (34) on the hydrogen atom at the a—position to the nitroxy group and consequently i s called " (X -e 1 lminat I on" « These elimination reactions may occur as side reactions to nucleo-p h i l i c substitution on nitrogen (Sj^) and on carbon (Sj^,), thus they caused considerable confusion i n the older l i t e r a t u r e . Nucleophilic reagents (like HO ) attack the p a r t i a l l y positive nitrogen or (X—carbon i n nitrate esters. I f , how-ever, by any means hydrogen atoms elsewhere i n the molecule become acidic to such an extent that their e l e c t r o p h i l i c i t y becomes comparable to that of the nitrogen or ex,-carbon atom then i n addition to nucleophilic substitution e l i m i -nation reactions also w i l l take place. In the aliphatic series the percentage of o l e f i n i c elimination i n -creases with the branching of the chain; i . e . with the s t a b i l i t y of the i n t e r -mediate carbonium ion. Both second and f i r s t order reactions (cf. 33) may take place i n o l e f i n elimination while carbonyl elimination (34) seems to be always bimolecular. Comparable figures for the two reactions are given i n Table V from the data of Baker and coworkers (65) (66). 4 - 33 -Table V. Elimination Reactions with Nucleophilic Reagent OH i n 90$ Aqueous Ethanol Solution Ec=c 2 w * k x 10 5 k x 10 5 Nitrate Ester 20° 60° 20° ' 60° 20° 60° 20° 60° CH3CH2ON02 2 0.08 4.8 0.21 (CH,) CH0N0o 2 < L 14.5 0.09 13.7 0.09 (CH3)3CONQ2 23 1.54** 260* - - - -PhCH20N02 - - 87 - 9.24 847 RiCHgCHgONGg 90 8.1 5 0.33 PhCH(CH3)0N02 1 0.016 3.0 0.127 (Ph)2CH0N02 - - - 1.0 31.7 ) Extrapolated from values observed at 0 , 20 and 30 $ kxlO = 0.06; 1.54 and 5o5 respectively. Because of deviations the true value may l i e between 130 and 460. ^ F i r s t order reaction: E n _1. Nucleophilic reagents other than OH also cause carbonyl elimination reactions. Aralkyl nitrate esters yielded almost exclusively the corresponding carbonyl compounds with anhydrous pyridine (35) (67) (36) (68). - 34 -A detailed study of the elimination mechanism i n benzyl nitrate has been carried out recently by means of kinetic isotope techniques (69). Both the 5.04 deuterium isotope effect and the 1.02 nitrogen-15 isotope effect ("which i s one of the largest observed for nitrogen i n a rate process") favoured a concerted mechanism (37) over a two step carbanion mechanism (38) for carbonyl elimination (Table VI). A value of 1.16 for the secondary deuterium isotope was indicated m the nucleophilic substitution. No significance, however, was attached to the difference between t h i s value and unity because of the rather large percentage error i n the determination of the small amount of nitrate ion. Table VI. Isotope Effects i n the Reaction of Benzylnitrate with Sodium Ethoxide i n Absolute Ethanol at 60.2 . PhCH20N02 PhCD20N02 ?|E C 02 88.7 64.4 foS^ 11.3 35.6 k t o t a l x 10 3 13.9 3.79 ^ ^ 2 x 10 3 12.3 2.44 x 10 3 1.57 1.35 ( k ^ k ^ E ^ 5.04 ± 0.25 ( V V S f l 2 ^ (k 1 4/k^ 5)E C ( )2 1.0196 ± 0.0007 - 36 -D„ Homolytic Decomposition of Nitrate Esters Presumably both thermal decomposition (70) and explosion (T3J of nitrate esters proceed v i a a free r a d i c a l mechanism. The principal cleavage occurs between the ester oxygen and nitrogen atoms with the formation of alkoxyl radicals (39)» R0N02 - RO' + N02 (39) This conclusion was confirmed by several authors working either on slow ther-molysis with the aid of reaction kinetics or on flames (combustion) u t i l i z i n g infrared technique. Equation _39 thus indicates that homolytic cleavage takes place according to mode of scission 4 (Figure 4). A number of simple primary aliph a t i c nitrate esters have been studied by both of the above techniques and the results from the thermolysis of n—propyl nitrate (72) support a degradation scheme essentially similar to that postulated for ethyl nitrate (40). CEjC^CI^aTK^ - N02+CH3CH2CH20. CH^H^+CILjO (40) However, i t i s not certain whether the i n i t i a t i o n proceeds according to 40 or by intramolecular rearrangement of the b i r a d i c a l formed i n 41. CH3CH2CH20N02 - H0N0+(.CH2CH2CH20.) - C H ^ ^ + C ^ O (41) Further reaction may occur v i a attack of the origi n a l ester by N0 2 or HONO (but not C3H^.0»), i n agreement with a "chain thermal" process suggested by Gray and Xoffe (73) for the ignitions of methyl and ethyl nitrates (42 and 43). N02+GH3CH2CH20N02 HONO+tCILjCH^^ONO^ CH^B^+R^CO+NOg (42) N02+CH3GH2GH2ON02 HONQ+^CHCHtjONOg) ~ CH3CH0+H2C0+N0 (43) -37 -The decomposition of the secondary isopropyl nitrate ester showed close analogy to the pattern observed i n the case of the normal isomer (44). CH, H CH, H H 3N / 3x / / .C - NO- + C v C H « + CH - C . (44) CH3 X0N0 2 CH 3 X 0. J 0 It was suggested furthermore (72) that the NC^ attacks the nitrate esters as i n 45 and 46 » A remarkable difference has been observed (74), how-ever, i n the flame decomposition of butanediol dinitrates with respect to that of monoesters« "The results indicate that these esters {2, 3-butanediol dinitrate and 1, 4-butanediol dinitrate) i n contrast to those of mononitrates so far studied, break down i n a unimolecular fashion" according to 47 and 48. Only traces of the suspected intermediates acetom (XXIII) and diacetyl (XXIV) were found among the products. Nitrate esters with more complex structure were also investigated. Glycerol t r i n i t r a t e (75) showed a close analogy to the decomposition of v i c i n a l dinitrates (47). The f i r s t step i n the thermal decomposition i s probably the scission of an 0~N bond (49). There are, however, two p o s s i b i l i t i e s since nitroglycerine has both primary and secondary nitroxy groups. Both of these alkoxyl radicals (49) w i l l y i e l d essentially the same products according to (50). Recent publications of Wolfrom and co-workers (76) (77) (78) pro-vide some interpretation of the cellulose nitrate thermal decomposition. In-teresting results were obtained by means of ^ C - labelled cellulose (mostly at position 2 and 5) which gave an i n i t i a l pattern for the thermal degradation of cellulose nitrate as shown in 51. In spite of the early (1901) pioneering work of W i l l (79) on nitrate s t a b i l i t y , kinetic studies of thermal decomposition have been carried out i n a - 3 8 -* 3 C N / H CH3-NC2 N 0 2 + • o r + CH3-CHO + N 0 2 (45 ) HC3 O N 0 2 C K j - O - N O H 3 C V / H ^ C V N 0 2 + C=0 + H O N O + N 0 2 ( 46) H3C D N 0 2 HjC O N Q 2 C H j - C H - C H — C H j - 2 C K j - C H O + 2 N 0 2 ( 47 ) O N 0 2 O N O , C C H ^ C H 2— C H 2 2 C H g O + C H ^ C H g * 2 N 0 2 ( 4 8 ) O N O 0 OH O CH3- C H — C-CH3 O O 1 n C H 3 - C — C • -CH-, XXIII XXIV C H . -CH- C H . O N 0 2 O N 0 2 O N 0 2 N 0 2 + C H 2 — C H — C h ^ ONC2 O ' O N 0 2 O N 0 2 O N 0 2 0 « (49) C H 2 — CH C H 2 ONC2 ONC2 A." CHgO + C H 2 — C H . O N O , ONOg CH^-CHO + N 0 2 ONCg (50 ) C H 2 - C H O O N O , C H N O , + T 2 o-2-CHO < H C O + C H 2 0 H« + O H C — C H O 6 ^ O N 0 2 H / o H 6 N O 2 6 H 2 C O C 0 2 + H C O O H 1 2 O H C — C H O (51) - 39 -Table VII. Log Frequency Factors and Activation Energies for Thermal Decomposition of Nitrate Esters. log A(sec~"'') E a (kcal/mol) Year Ref. CH3CH2ON02 16.85 41,23 1954 81 CH3CH2CH20N02 14.7 36.86 1949 82 cf42 02N0GH2CH20N02 15.9 39.0 1947 83 C^NOCILjCH (0N02 JCl^ONC^ 13.6 35.0 1959 84 17.1 40.3 1947 83 Nitrocellulose 18.95 43.7 1955 85 - 32.8-35.8 1961 86 Isosorbide di n i t r a t e 15 37 1960 62 C10N02 14.18 30.0 1961 62 FON02 13.76 29.7 1958 88 - 40 -more elaborate way only i n the last 15 or 20 years. The unimolecular decompo-s i t i o n of nitrate esters followed Arrhenius' rate equation (52). -Ea k = A.e R T (52) The apparent activation energies, E (kcal/mol), and log frequency factors, log A (sec "*"), for representative esters are summarized i n Table VII. Although one would expect some sort of correlation between structure and r e a c t i v i t y ^ one should treat these rate constants cautiously because two or more types of decomposition may take place simultaneously. Furthermore, the measurements were carried out by different experimental techniques m different laboratories and because of the large variations i n the values no generalization seems permissable. Steinberg and co-workers (80) reported differences i n the burning rates, k(cm/sec)$ of ordinary and deuterated nitrate esters. As the degree of deuteration increased the burning rates were decreased extensively. For example, perdeutero—isopropyl nitrate did not even burn under the experimental conditions. The isotope effects obtained are summarized i n Table VIII. It has not been determined as yet whether these effects are comparable with those kinetic isotope effects which might have been observed i n slow thermal decomposition* Table VIII The Effect of Deuterium Substitution on Burning Rates of Nitrate Esters (80). Compound V 1 ^ CD3CD20N02 1.4 (ca3)2cjiom2 1.26 (CD 3) 2CH0N0 2 1.54 (CD 3) 2CD0N0 2 CO The Photochemistry of the Nitrate Esters and Related Compounds. - 42 -Photolysis of the ^C-O-X group where X may be halogen. -OR, -NO, or -NO^ i s of current interest, (89)(90). Photochemical reactions are considered i n general as transformations v i a free r a d i c a l intermediates and thus photolysis provides an alternative route to thermolysis for the decomposition of n i t r i t e and nitrate esters by free r a d i c a l mechanisms. A recent review (9l) on the photolysis of n i t r i t e esters, RONO, sum-marized the synthetic potential of the technique. The photolytic decomposition i s considered to proceed v i a a homolytic 0—NO bond cleavage which provides n i t r i c oxide and alkoxyl r a d i c a l . The alkoxyl r a d i c a l thus formed has a r e l a -t i v e l y short lifetime and may undergo further transformation by one of several possible routes. The products isolated thus e n t i r e l y depend on the route "selected" by the r a d i c a l and the "selected" mechanism i s a function of the intra—and intermolecular chemical environment. Alk y l n i t r i t e s (XXV) exhibit a number of p o s s i b i l i t i e s f or electronic excitation (92). The lowest energy (longest wavelength) t r a n s i t i o n (^3600 X ) seems to be (93) an n_T —+• <\t* excitation since the non-bonded electrons of trans- cis -XXV XXVI nitrogen (x) are the most loosely bound and i n this excitation one nonbondmg electron of nitrogen i s transferred to the lowest empty antibondmg TC ( i . e . TT *) o r b i t a l . The nonbonded electrons of oxygen ( • ) are more t i g h t l y bound because of the greater electronegativity of oxygen and thus their electronic - 43 -excitation n ^ — » T£* requires higher energy (i.e. shorter wavelengths) seemingly around 2700 £ (Figure 5). At even shorter wavelengths is the very intense TT •» Tt * band representing the excitation from a low energy level Tt orbital (o) to the antibondmg Tu orbital ( TT *)» Conehtiifaly Apparently i t has not beenAdetermined as yet which one of these ex-citations brings about the suggested 0-N0 bond cleavage in solution photolysis. It is possible that one of the two low energy excitations, TT * or n,Q — TC*^ or both of them are responsible for the photolysis since the n^ •» TL* excited nitrite ester readily decomposed in the gaseous phase (94). In nitrate esters, however, the nitrogen atom does not possess non-bondmg electrons, therefore, the corresponding long wavelength absorption is absent from the spectrum (Fig. 6 B). On the other hand there are twice as many Tt and XIQ electrons in the nitroxy group (XXVI) as in the oxynitroso group (XXV). - 44 -Figure 5. A: Energy Levels of Molecular Orbitals and Possible Electronic Transitions for N i t r i t e Esters. B: Typical Electronic Spectrum of a N i t r i t e Ester (2-butyl n i t r i t e i n ether (95)). The dotted lines represent the estimated separation of the various transitions. Figure 6. A: Energy Levels of Molecular Orbitals and Possible Electronic Transitions for Nitrate Esters. B: Typical Electronic spectrum of a Nitrate Ester (2-butyl nitrate i n ethanol (95)). The dotted lines represent the estimated separation of the various transitions. - 45 -No systematic study of the electronic spectra of nitrate esters has been published as yet but Rao (96) suggested that the shoulder at 2700 & i s r e a l l y due to an n — TU* transition, while the high intensity band,at the shorter wavelength would represent the TC ——*- 1T * excitation, A ty p i c a l nitrate ester spectrum together with an i l l u s t r a t i v e energy l e v e l diagram* i s shown i n Figure 6* Very l i t t l e i s known about the photochemical behaviour of nitrate esters. Photolysis of ethyl nitrate i n the gas phase with the 2537 A* lines and 2650 1 of the mercury arc led to the conclusion that the oxy-nitro bond scission i s the predominant reaction of the excited nitrate ester molecule* Equations J53_ to J58 seemed to explain the observations (99); (1.00) C ^ O N O ^ * C2H50.+N02 (53). (0.53) 0 ^ 0 * CRy-K!H 20 (54) (0.47) C 2H 50. B>+CH3-CH0 (55) (0.485) CRj* +N02 ~ CH 3N0 2 ( 56) (0.045) CBy+C^OM^ • C 2H 50CH 3+N0 2 (57) (0.37) H*KJ 2H 50N0 2 C2H50H+N02 ^ The gaseous phase photolysis of n i t r i t e and nitrate esters by sunlight i s also a current problem of a i r pollution (100). A recent Japanese patent (101) seemed to confirm the free r a d i c a l nature of the photo-decomposition fragments of nitrate esters i n solution since the esters (generated i n s i t u from alcohol, inorganic n i t r a t e , and acid) were claimed to function as polymerization accelerators when irradiated during the preparation of cr y s t a l l i n e polymers. A detailed study some years ago (102) *) These energy l e v e l diagrams (Figure 5 A and 6 A) were constructed by analogy to those of N 0 2 ( l l ) , $(£(97) and Na ~CH2N02 (98). 46 -proved that thermolysis of alkylnitrate also accelerated additional poly-merization of methyl methacrylate v i a a free r a d i c a l mechanism,, The ef f i c i e n c y of alkylnitrates as "ant1—knock" additives i n gasolines points to a similar mechanism. The photo-decomposition of cellulose nitrate has been recently studied by Claesson and co-=woikers (103) (104). F u l l y and p a r t i a l l y nitrated cellulose samples (13.87 and 12,12$ N respectively) were photolysed with 99*5$ mono-chromatic l i g h t of 2537 %» The extent of depolymerization was followed v i s -cometncally. I t was found that the quantum yi e l d s for the depolymerization did not d i f f e r greatly for the two samples being 0.02 for the p a r t i a l l y nitrated cellulose and 0*01 for the f u l l y nitrated polymer. This might indicate that the presence of an active group (—OH) i n the molecule aided the depolymerization. On the other hand the possible occurrence of reactions other than depolymerization was not excluded and the low quantum y i e l d of depolymerization might mean that most of the l i g h t quanta were u t i l i z e d for other processes. A wavelength dependence study (104) indicated that while the e f f e c t i v e -ness of the l i g h t quanta was roughly the same at 2537 and 3020 jL mono chr omatlc l i g h t at 3340 and 3650 A* was p r a c t i c a l l y inactive. On the other hand the u t i -l i z a t i o n of (o -*aphtbylamme as photosensitizer caused photo-x * (kcal/mole) (R1) XNO, n * (i) c 2.50 (CH3)3C-NO 6650 s 2045 (CH3)3CS-NO 5988 CI 3.10 Cl-NO 4600 N 3.15 (CH3)2N^NO 3610 0 3.60 CH3(CH2)30-NO 3560 F 4.15 F-NO 3110 *) Values taken from (13). Table XI Fundamental Infrared Frequencies (cm"1) of (R1) XN0O n 2 Compounds Type of Vibration CH3-N02 HO-N02 F-NO, ^ s t r e t . as (N02) 1562 1540 1675 1779 s t r e t . sym (N02) 1377 1379 1300 1306 deform (ONO) 657 709 680 466 s t r e t . (x-^o 2) 919 1043 925 821 _ Hg* (62) Hg* + NO NO* + Hg (63) NO* + NO - (NO)* (64) (NO) 2 »» N 2 + oxygen (65_) (NO)* + NO ~ N 20 + N02 (66) Photolysis of NO i n benzene solution produced unexpected products as recently reported by Kemula and Grabowska (113). The formation of .o-nitro-phenol and 2,4-dinitrophenol and the absence (or questionable presence) of _p_— or m- nitrophenol could not be explained by a simple reaction mechanism. The primary process was explained i n terms of the forbidden smglet-t r i p l e t (T S q) excitation* of benzene since the mixture was irradiated with a wavelength range of 2900-3600 X and from previous experiments (117) i t was clear that the regularly forbidden s m g l e t - t r i p l e t absorption would appear i n this range of wavelengths in the presence of paramagnetic species l i k e NO or O^* It was proposed that the excited t r i p l e t state of benzene (which acts as a b i r a d i c a l ) , being in contact with the paramagmetic NO molecule, produced the nitrophenols i n some unknown manner (67)• Control experiments were carried out with oxygen (118) and i n addition to minor amounts of o-qumone, phenol was isolated as the major product (68). — This electronic t r a n s i t i o n i s forbidden by the "spin momentum conser-vation" rule of quantum mechanics (114) (115) (116), - 54 -Photolysis of NO, The gas phase photolysis of nitrogen dioxide was investigated by Norrish (119) (120) as early as 1929. Recent reviews ( l 2 l ) (122) proposed seven d i s t i n c t i v e mechansims (depending on the reaction conditions) i n terms of f i f t e e n equations. " I t now appears that below 3700 A* atomic oxygen i s an important product of the primary photochemical process" (120) according to 69. NO, ^ » N0 2 — NO + 0 (69) The reaction of nitrogen dioxide with aromatic compounds was r e -viewed by Riebsomer (123). "Reaction may be brought about by heating to 80° in sealed tubes, by the action of li g h t (4000-7000 £) at 55-60°, or by a glow discharge i n a Siemens tube" (124). Photolysis of HN03 The decomposition of n i t r i c acid by li g h t was described i n the la s t century by Berthelot (125). A more detailed study of the reaction as reported by Reynolds and Taylor (126) i s summarized by equation 70. li g h t (fast) 4 HN03 - 2H 20 + 2N 20 4 + 0 2 (70) dark (slow) According to the authors "the decomposition may possibly take place i n stages, the f i r s t product being nitrous acid and oxygen, the former and the excess n i t r i c acid then producing water and nitrogen peroxide" as represented i n equation 71, 72, 73. H0-N02+ - HO-NO + 0 (71) H0-N02+0 HO-NO + 0 2 (72) H0N0„+H0N0 H„0 + N„0„ (73) 2 2 2 4 — This reaction would thus support the suggested f i f t h mode of scission of nitrate esters (Figure 4)« - 5 5 -- 56 -The results of Co -radioly s i s of HNO^ suggested (127) that both the fourth and f i f t h modes of scission occurred simultaneously under high energy i r r a d i a t i o n (74) (75). H0N02 — HON02 *- HO* + N02 (74) HON02 » HON02 HN02 + ^ (75) Photolysis of N0~ Nitrate ions undergo photochemical reduction i n aqueous solution to N0 2 which i n turn i s transformed to hydroxylamine. The photoreduction occurs readily i n the presence of simple organic substances at the cost of t h e i r simultaneous oxidation (4). Inorganic nitrates were photolysed i n the presence of diphenylamme by Coldwell and McLean (128) (129). Nitrodiphenylamines were isolated as i n the case of the photolysis of covalent nitrates (59)(105) but no evidence of nitroso derivatives was found. Photolysis of CBLjN02 Nitromethane i n argon matrix at 20°K has been photolysed and the nature of the products studied by infrared spectroscopy (130). The homolytic cleavage according to equation 76 has been ruled out by the absence of N0 2, CH^ and C 2Hg« CHj-NOg ^ c n y + NO2 (76) The photolysis was performed on s o l i d CH^NC^ and CD^NOg with the aid of a high pressure (A-H6) (Figure 45) or medium pressure (A-H4) mercury arc spectrum. The stepwise transformation i s described by 77 where the major pro-ducts were CR^O. CO, C0 2, N20, NO, H 20, HOCN, and HNO. CH^-N02 ^ - CE^O-NO ^ a - products. (77) The fa c t that methyl n i t r i t e was formed exclusively as the trans-i s omer at 20°K provided interesting evidence for the mechanism. Although more - 57 -than one tentative mechanism has been suggested, i t seems rational that the methyl group participated m a stereo-specific manner m the rearrangement by association with one of the oxygen atoms of the excited NO, group. Photolysis of Pyridine N-oxides Pyridine N—oxides are somewhat analogous to —NO, groups since one doubly bonded oxygen of —NO, may be considered as replaced by the aromatic double bond. No n^ *• II excitation i s possible but the group exhibits both n —«-7t and H>—»• tc transitions (131). o Irradiating pyridine N-oxide at two different wavelengths (132) produced i n both cases (78, 79) pyridine and atomic oxygen (which i n turn caused some destructive oxidation) i n spite of the fact that 3261 (pyrex Cd resonance lamp) caused an n — v TC* tr a n s i t i o n while the line at 2537 £ (quartz Hg resonance lamp) produced a TC—•-TiT* excitation. On the other hand when the Ov- picoline N-oxide was irradiated at the same two wavelengths, the products of the photolysis varied according to the wavelength applied. In the * case the c\-methyl group did not effect the course of decomposition (80), while i n the n 3»tC* excited state the CX -methyl group acted as an oxygen acceptor (81)analogous to the formyl group i n the o»»nitrobenzaldehyde photorearrangement (60). These results were clear demonstrations of the chemical differences of the two types of excited states ( 'HT —»-TC* and n —*-'VU*)# even so, this d i s t i n c t i o n was not always reflected i n the product formation. RESULTS AND DISCUSSION - 59 -I* Synthesis of Aromatic Nitrate Esters. It has long been known (53) that i n n i t r i c acid-sulfuric acid n i t r a -t i o n nitronium ion, NO^, i s actually the n i t r a t i n g species. Although this mixed acid i s a powerful and widely used n i t r a t i n g agent i t has been established that for the n i t r a t i o n of sensitive polyols n i t r i c acid-acetic anhydride mixture i s more desirable (24). The question of the identity of the n i t r a t i n g agent i n the l a t t e r mixture, however, remained unsettled u n t i l recently. Both N ,0^ (134) and CH^COONO, (135) were proposed as the n i t r a t i n g agent i n agreement with Raman spectroscopic evidence (136), however, NO," also has been detected by infrared spectroscopy (137) i n concentrated solutions of n i t r i c acid i n acetic anhydride. Recent kinetic investigations confirmed nitronium ion as the nit r a t i n g entity. Equations 82, 83, 84 were proposed (138) to f i t the experimental results, HONO + H + J>quil., - NO, (82) I T 2 \ + e q u i l . J^O-JTO, + A c 2 0 * 2 Ac OH + N0+ (83) H C 6H 6 + NOj - S K * - C 6 H ^ H + (84) NO2 Under the experimental conditions the rate of n i t r a t i o n was f i r s t -order with respect to the n i t r a t i n g substance (84), but was second-order with respect to n i t r i c acid concentration i n agreement With the s e l f — i o n i z a t i o n shown i n 82_ and 85. JJ ' HN03 + H N 0 3 e q m l . „ • ^ Q-NO,, + NO^ (85) H The fact that the addition of 0.001 M N a N 0 3 slowed the rate of ni t r a t i o n confirmed that the se l f - i o n i z a t i o n (85) was an important component of the experimentally observed rate constant. On the other hand the addition - 60 -of 0,01 M s u l f u r i c acid introduced a large amount of protons into the system which dismissed the rate determining character of the s e l f i o n i z a t i o n (85) and the observed rate then became fi r s t - o r d e r with respect to n i t r i c acid concen-tr a t i o n (82). The observation (139) that acetoxylation occurred as a side reaction i n aromatic n i t r a t i o n (86), (87), (§&) could also be taken as chemical evidence for the nitronium ion mechanism. Brown has pointed out (140) that a l l e l e c t r o p h i l i c substitutions (including nitration) on aromatic nuclei proceed by the same general mechanism v i a the formation of charge-transfer complexes. According to this model when the electrophile NO^ approaches the aromatic nucleus, negative charge i s trans-ferred from the aromatic " TT. -electron cloud" to the positive nitronium ion (XXVII). Consequently, the c l a s s i c a l intermediate or t r a n s i t i o n state (84) (XXVTIl) has to be replaced by a series of e q u i l i b r i a as shown i n 90, 91 and 92. N0_ i » H 3 C .H \ (86) N O o ( 87) X X V I I A r N 0 2 [ X X V I I I M o l e c u l e R e l a t i v e r e a c t i v i t y t o w a r d N 0 2 * 1 0 1 4 2 4 3 0 - 62 -complex intermediates to the formation of nitrate esters. In the l a t t e r case the oxygen of the alcohol, with loosely held unshared electrons, played the role of electron donor i n the place of the aromatic nucleus i n formation of the charge-transfer complex. Hughes ( l 4 l ) created reaction conditions (high substrate concentration with respect to NO, concentration by d i l u t i o n with water) i n which the rates observed were measures of the r e l a t i v e r e a c t i v i t i e s of C, N and 0 atoms located i n various chemical environments. In this manner a scale of r e l a t i v e r e a c t i v i t i e s (89) was established. In t h i s research aromatic nitrate esters were required for photochemical study and for several reasons (as discussed i n the introduction) direct e s t e r i f i — cation of the corresponding alcohols seemed to be the most desirable synthetic route. The structures of the selected aromatic alcohols were such that the r e l a t i v e location of the OH group with respect to the aromatic nucleus was analogous to that i n benzyl alcohol. The synthesis seemed to be feasible since Hughes' figures (89) indicated the higher efficiency of O-nitration compared to ri n g n i t r a t i o n of benzene. On the other hand the increased rate of n i t r a t i o n on an activated aromatic nucleus, as i n toluene, pointed toward a close competi-t i o n between C*- and O-nitration i n the case of the aromatic alcohols. The experimental results showed higher yields of nitrate esters than would be predicted from the foregoing. In the case of meso-hydrobenzoin (XXDC) and dl-hydrobenzoin (XKXl) the amounts of aromatic nitro by-products were r e l a -t i v e l y small and did not cause any d i f f i c u l t y i n the i s o l a t i o n of the pure d i -nitrates ( y i e l d : 57$). In the n i t r a t i o n of benzoin (XXXIIl), under similar con-ditions, the formation of yellow G-nitro compounds was somewhat more pronounced and small amounts of benzil were also isolated as the result of unwanted oxida-tion; however, the pure nitrate was s t i l l isolated i n reasonably good y i e l d (35$). - 63 -In the case of the 1,2-acenaphthenediols (XXXV) (XXXLX) the r e a c t i v i t y of the naphthalene portions of the molecules was much higher than that of the benzene moiety of the hydrobenzoms as predicted from the lower ionization po-t e n t i a l s * and chemical r e a c t i v i t i e s of the corresponding hydrocarbons. Conse-co^ be ex pectel t° quently, a number of ring nitro by-products^originate^i from both the c i s - and trans-diols (96, 97). Three different compounds (A, B, and C) were isolated by chromato-graphy from the crude n i t r a t i o n mixtures from both isomers. Compounds A and'A, were the pure d i n i t r a t e s , while B , B. , C and C, were trans c i s trans c i s trans suspected to be ring n i t r o by-products. According to TLC analysis (Figure 8) the compounds isolated were chromatographically pure with the exception of B, which contained minor amounts of a second substance, trans Elomontary analyses indicated that ring mononitro din i t r a t e esters (Table XII) (XXXYII)(XLl) were among the products and this was confirmed by the infrared spectra (Figure 9). The nature of the C fractions remained undetermined although the slow running (low R^) isomer which originated from the trans-diol ( C - k r a n s ) was suspected to be nitroacenaphthom (XLIl) from the elementary analysis and infrared spectrum* Since acenaphthene i s rea d i l y substitutable by e l e c t r o p h i l i c r e -agents almost exclusively at the peri-positions (143), there could be l i t t l e doubt about the position of the ring-substituted NO^ , group. A systematic li t e r a t u r e survey (27) indicated that the present research (144) was the f i r s t attempted synthesis of aromatic nitrate esters by direct e s t e r i f i c a t i o n . * The ionization potential for benzene i s 9.25 ev and for naphthalene 8.12 ev (142). - 64 -X X X V X X X V I ( A c s ) X X X V I I ( B C | S ) X X X V I I I N 0 2 N O 2 X X X I X X L ( A t r a n s ) X L l ( B t r a n s ) X L I I ( C t r a n s ) ( 9 7 ) - 6 5 -o O O O i i ! i _ Q . . Q _ J \ i s B c i s ^ c i s A t r a n s B t r a n s ^-trans F f G U R E 8 Thin-layer C h r o m a t o g r a p h y of N i t r a t i o n P r o d u c t s f r o m a s - and trans-1,2-Acenaphthenediols ( M - 3 , S - 1 , R - 1 ) T A B L E XII N i t r a t i o n P r o d u c t s f r o m a s - a n d t r q n s - 1 , 2 -A c e n a p h t h e n e d i o l s . P r o d u c t Fo r rnu I a R f N ° /o Symbol Isomer Number Emp i r i c a l ( M-3 S-1 R-1 ) M P C C ) Obt Calc A CIS - XXXVI /— M S~\ K | 0 69 129 5-132 5 10 03 10.15 t rans - XL C 1 2 H e ° 6 N 2 0 . 7 9 9 8 - 1 0 0 10.07 B C I S - XXXVII / - * t_J /"\ k| 0 32 9 5 - 9 8 12.50 13.10 t rans - XLI C 1 2 H 7 ° 8 N 3 0 .47 o i l 11 81 C CIS — XXXVIII C 1 2 H 7 0 4 N 0 02 8 8 9 6.10 t r a n s - XLII 0 . 0 5 210 - 215 6 . 3 6 - 6 6 -F IGURE 9 ' In f ra red Spec t r a of N i t r a t i o n P r o d u c t s of c j s - and t r a n s -1, 2 — A c e n a p h t h e n c d i o l . - 67 -IIo Chromatography of Nitrate Esters. It has been recognized for more than a decade (l45)(l46) that hydrogen bonding plays an important role i n adsorption on s i l i c i c acid and that the ad-sorption sites (147) are the weakly acidic hydroxyl groups of the adsorbent. Recently i t was shown (148)(149)(150)(l5l) that the infrared absorption band of the surface s i l a n o l group i s shifted toward lower frequencies upon in t e r -action with absorbate moieties. This frequency s h i f t (A~J ) i n the adsorption A of diethylamme was as large as 73 cm-"'' (from 3743 to 3670 cm-''") (152). Hydrogen bonding i s a special case of charge—transfer interactions, and i t has been found that the strength of hydrogen bonding ( l . e . A ^ c m "*"), which i s i n fact a measure of adsorption, varies with the ionization potential of the adsorbate (152). This i was experimental evidence that the adsorbate molecule i s the electron donor ( i . e . proton acceptor) and the hydrogen atom of the s i l a n o l group acts as a hydrogen bridge between adsorbent and adsorbate. It has been calculated by Sporer and Trueblood (153) that the 0.....0 distances of hydroxyl groups on adjacent s i l i c o n atoms l i e between 4«3 and 5.8 %. It was found experimentally by the same authors that the most favorable "inter-adsorbing atom distance" (0...0 or 0....N) i n an adsorbed molecule was about 6.1 - 6«2 A" (or a multiple of that value) which was close to the 5.8 i calculated value. This favoured inter-nuclear distance was shown to be available i n meta- and para- substituted benzene derivatives. The equilibrium constant K for the adsorption process (98) of the chromatogr'aphed substrate (S) was determined experimentally (99) and was termed the "adsorption a f f i n i t y " of the substrate. solution "* adsorbed (98) - 68 -From equation (lOO) the standard free-energy change for the adsorption process was calculated and shown to be dependent on the structure of the chromato-graphed substrate* A F ° = - RTlnK (100) Furthermore i t was pointed out (153) that there was a characteristic "adsorption a f f i n i t y " not only for molecules (K) but also for substituent groups (K ). Thus the value of K for a particular molecule which carried 1 different substituents was the product of the separate IL values ( l O l ) . K T T K (101) i 1 and the resultant standard free energy change for a molecule ( A F ° ) was the sum of the values of the individual substituting groups ( AF^°) according to equation (102)* A F° = £ F ° (102) i 1 A sequence of group adsorption a f f i n i t i e s (K ) was compiled for i n -dividual substituents on aromatic nuclei i n decreasing order of adsorption a f f i n i t i e s (103). -CH2NH2, -COGS, -CB^OH, 4& 2, -COCEj, -OH, -CHO, -N02, -OCHy H (103) 7000 370 260 80 70 27 23 3.1 This sequence explained why the r i n g nitro by-products of 1,2-Ace-naphthenediol dinitrates possessed lower R^ values than the unsubstituted d i -nitrates (Figure 8» Table XII). "In chromatography the experimentally measurable quantities are the distances that the zone of solute and the front of solvent have travelled i n the same period of time" (153). The ra t i o of these distances, the'Vlevelop-ment rate", designated R°, may be determined in column chromatography when the - 69 -concentration i s low enough that the process i s i n the linear region of the Langmuir adsorption isotherm. In this case K may be calculated from the ex-perimental R° values according to (104). K = 1 - R° (104) R° In the more recently developed microadsorption thin-layer chroma-tography (TLC) the value R° would be replaced by R^ and K i n equation 105 may be taken as a measure of the adsorption a f f i n i t y . 1 - R „ K = £_ = ( 1 - 1 ) (105) B f R f Similarly the value R^ (106) which is also used i n paper chromatography (154) would be proportional to the standard free energy change of adsorption according to equation 107. f A F ° = -2.303 RTRj^ (107) On t h i s basis the TLC technique should be capable of distinguishing compounds according to th e i r functional groups.* It was suggested some time ago i n this laboratory (155) that elucidation of the molecular structure of unknown compounds (particularly that of nitrate esters) might be aided by TLC through a comparison of R^ values determined under standardized conditions and this principle has been successfully applied i n a qualitative manner (105), (144). In a quantitative study a number of carbohydrate nitrates and their parent alcohols were chromatographed on chromatoplates (M—3, S—10. R-2) and the experimental values together with the calculated R^ and R^-NO,) * Phenol protons may contribute s l i g h t l y to the t o t a l hydrogen bonding process (153)» - 70 -values for representative compounds are summarized i n Table XIII. Table XIII Relationship of R^ Values and Structure i n Polynitroxy Compounds Compound A n a ) R f K-1 mannitol hexanitrate mannitol pentanitrate 1 0.747 0.215 0.339 3.647 -0.469 0.562 -1.031 -1.031 d u l c i t o l hexanitrate d u l c i t o l pentanitrate 1 0.706 0.190 0.417 4.272 -0.380 0.631 -1.011 -1.011 methyl-]3i-D-glucoside tetranitrate me thy 1 -JS-D-gluc o s ide— 2,4-dinitrate 2 0.775 0.044 0.291 21.727 -0.536 1.337 -1.873 -0.936 average ARj^ per nitroxy group -0.993 a) Difference i n number of nitroxy groups between corresponding pair of c ompounds. b) Average of four determinations. c) from (106). d) Difference of R^ values between corresponding compounds. e) ARy per nitroxy group. The nearly constant value of AR^ (variation about + 5$) for the ^l-NO, group over the range of compounds with varied structures indicated an independent contribution to the adsorption by each nitrate ester group i n thin-layer chromatography. The order of group "adsorption a f f i n i t i e s " calculated according to equation 108 from AR^ values. K = 10~ A RM (108) - 71 -for O-substrtuted (RQX) polyhydroxy compounds (where -X may be NC^ or other substituent) determined for a series of compounds (155) are given i n 109> where the numerical figures for K apply for TLC (M-3, S-10, R-2) at 300°K. -C0CH3 > -N02 > -CH3 (109) 0.121 0.102 0.076 The remarkable fact that i n changing from column chromatography to TLC, and from aromatic C_-substitution to aliphatic O-substitution the sequence of adsorption a f f i n i t i e s of the substituents was v i r t u a l l y unchanged (103) from those determined by Sporer and Trueblood (153) made i t evident that TLC results would be just as important i n structural elucidations.-ss^^^ffi^p fvrim nib 1 in iruil im llinili . Figure 10 shows a ty p i c a l chromatographic pattern (TLC, M—3, S - l , R-2) of aromatic and representative non-aromatic nitrate esters and Table XIV summarizes the observed R^ values (300°K) together with the calculated constants: R^, A F ° (the standard free energy change for adsorption) and K (the adsorption a f f i n i t y ) . It was clear from these data that i f other active substituents were also present i n the molecule, such as the two c y c l i c ether oxygen atoms i n (A) or the ketone oxygen i n benzoin nitrate (B) the compounds showed increased a f f i n i t y toward the adsorbent (K = 5.54 and 1.98 respectively). A similar ef-fect resulted from an accumulated number of nitroxy groups i n a molecule as i l l u s t r a t e d by mannitol hexanitrate (K = 9.00) (C). In contrast, cholesteryl nitrate (D) having p r a c t i c a l l y no other group to adsorb with other than the one O-NO,, group had the lowest adsorption a f f i n i t y : K = 0.258. Benzylnitrate (E), also a mononitrate, with the benzene ri n g as a second functional group for hydrogen bonding, showed an a f f i n i t y more than twice as large (K = 0.647) -72-08-06^ 04 021 B D H F I G U R E 10 Chromatograph i c P a t t e r n of R e p r e s e n t a t i v e N i t r a t e E s t e r s T L C ; M-3, S-1, R - 2 . The c o m p o u n d s and c a l c u l a t e d & F ° and K va lues a r c l i s t e d in Table X IV i - 73 -Table XIV Chromatographic Constants for Nitrate Esters Compound Name Formula R f \ cal^mole K A Isosorbide Dinitrate XLX 0.151 0.743 -rl020 5.54 B Benzoin Nitrate XXXIV 0.336 0.296 1 -463 1.92 C Mannitol Hexanitrate 0.100 0.954 -1310 9.00 D Cholesteryl Nitrate 0.258 -0.588 -807 0.258 E Benzyl Nitrate 0.607 -0.189 +259 0.647 F me s o-Hydr obenz oin Dinitrate XXX 0.578 -0.137 +188 0.730 G dl-Hydrobenz oin Dinitrate XXXII 0.560 -0.105 +144 0.786 H cis-l,2~Acenaphthene-d i o l Dinitrate XXXVI 0.477 0.040 -55 1.10 I trans-1,2~Acenaphthene— d i o l Dinitrate XL 0.596 -0.169 +232 0.678 J trans-1,2-Cyclohexane-d i o l Dinitrate XXII 0.536 -0.062 +85 0.866 See Figure 10, - 74 -as that of cholesteryl n i t r a t e . meso- and dl-Hydrobenzom dinitrates (P)(G) although dimers of benzylnitrate did not exhibit doubled adsorption a f f i n i t i e s (Table XIV") indicating that rotation on the C^-C, bond, s t a t i s t i c a l l y did not on the average bring much more than one nitroxy group per molecule into contact with the adsorbent. The trans-isomer of 1,2-acenaphthenediol dinitrate ( i ) , having a fixed conformation, apparently could be adsorbed by one nitrate ester per molecule, thus the value of K (0.678) was very close to that of benzyl n i -trate (E). In contrast the cis-isomer had almost-twice as large adsorption a f f i n i t y (K = l o096) since both of the nitroxy groups were situated on the same side of the acenaphthene moiety. It was of interest also that the trans-1,2-cyc1ohexandiol din i t r a t e (J)(K = 0.866) f e l l i n between trans- and cis-l,2-acenaphthenediol dinitrates, which indicated that the two equatorial nitroxy groups were stereochemically not as favourably oriented for adsorption as were those of the c i s - d i n i t r a t e . This probably meant that the separation of the two nitroxy groups i n the trans-1,2-cyclohexandiol dinitrate was larger than that of the two neighbouring active sites (5o8 A*) i n the s i l i c i c acid. They were, however, more suitably oriented for absorption than those of the trans-1,2-Acenaphthenediol d i n i t r a t e s . In addition to providing structural information on the pure sub-stances TIC was used i n the analysis of n i t r a t i o n and other reaction mixtures and as a guide for developing appropriate column chromatographic techniques (156) for the p u r i f i c a t i o n of crude products. Furthermore TLC was also i n -valuable as a technique (157) for preparative i s o l a t i o n of compounds on a micro scale. - 75 -I I I , Analysis of Aromatic Nitrate Esters. Elementary analysis carried out by the conventional combustion techniques does not provide satisfactory results with aliph a t i c nitrate esters (158, 159) due to i n s u f f i c i e n t combustion and too rapid gas evolution. The d i f f i c u l t has been largely overcome i n Dumas nitrogen analyses by d i l u t i n g the sample with glucose to aid combustion and decrease the rate of gas f o r -mation. Table XV Combustion Analyses of Polynitrates me so-Hvdr obenz om Dinitrate Element Content ($) Reproducibility Calcd. — Obs. Calcd. Obsd. Abs. Per. Abs. Per. c 55.26 55.43 -0.03 ^0.05$ -0.17 -0.4$ H 3.28 3.87 -0.06 -1.5$ -0.59 -19.7$ N 9.21 9.11 -0.39 -4.3$ -0.10 -1.1$ Mannitol Hexanitrate N 18.59 18.06 -0.46 -2.6 -0.53 -2.9$ In the present work experiments were carried out to assess the ac-curacy of combustion analyses of the c r y s t a l l i n e aromatic nitrate esters since these had not previously been studied. A chromatographically pure sample of meso-hydrobenzoih din i t r a t e was selected as model compound and three p a r a l l e l micro Liebig-Pregl combustions and six p a r a l l e l micro Dumas analyses were carried out without added glucose. The average values from these analyses and the calculated reproducibility as well as the differences between the - 76 -analytical values and the theoretical values are summarized i n Table XV. For comparison the nitrogen content of mannitol hexanitrate, obtained as an average from twelve analyses (160) i s also given. It was apparent from these results that the C$ obtained was very consistant and also close to the calculated value, while there was a noticeable discrepancy i n the H$ values. For the N analysis the absolute values of both reproducibility and difference seemed to be better for the aromatic nitrate ester than for mannitol hexanitrate. Although the percentage reproducibility was somewhat worse for the aromatic nitrate (because of low N content), the percentage difference between the average of the analyses and the calculated value was almost three times larger for mannitol hexanitrate. These results were consistent with the high carbon content of the hydrobenzoin nitrate i n which the aromatic moiety acted as a " b u i l t i n carbon diluent" for the Dumas nitrogen combustion. IV. Spectra of Aromatic Nitrate Esters. A. Nuclear Magnetic Resonances Spectra (NMR). Methyl nitrate (161) ethyl nitrate (162) were the only nitrate esters which had previously been investigated by proton magnetic resonance spectro-scopy. In aromatic nitrate esters, i f the aromatic nucleus can rotate f r e e l y i n the molecule the resonance of the aromatic hydrogens average out resulting i n a single "benzene peak". This was found to be the case for benzyl n i t r a t e . On the other hand i f the aromatic portion constitutes a r i g i d system the coupling between ri n g protons may be observable. This was the situation i n the case of the l«2-acenaphthenediol d i n i t r a t e s . The analysis of these and related r i n g proton spectra as ABC systems by the use of ABX approximation (163) i s s t i l l i n progress (144) (164) and i s not included here. - 77 -Prom the spectra the greatest effect of the nitroxy substituents was revealed i n the resonance of the protons i n the d--positions. Table XVI Group Electronegativities and f Values for c< -Hydrogens m Nitrate Esters and Related Compounds. Substituent. Group -H -OH -OAe -ONO, Group Elecfcionegati v i t y 3.81°,3.91 b 2 . 2 i a 3.51b 3.83b 4.18d Hydrocarbon Group Isomer 1 w e H C 1 S - 4.47f 3.87 3.27 1,2-Acenaphthenyl" 6.71 trans- 4.60 4.00 3.46 m- 6.04 4.02 3.97 1,2-Diphenylethyl- 7.60 d l - 5.87 3.93 4.05 Benzyl- 9 • • 7.70 5.35 4.90 4.45 1,2-Cyc1ohexany1- trans- 8.57 6.17 5.27 4.77 a Ref. (7), b Ref. (162), ° Ref. (161), d This work In acetone solution 0.2 to 1.0 M, tetramethylsilane = 10.00. In g l a c i a l acetic acid. Changing the substituents i n the order -H 5 -OH , -OAc , -ONO, caused est decrease i n "the MicroaBO of the aoi d i t i o c ( l o w o r " values^ of the oc-hydrogens (Table XVI). A plot of the T—values of the ©{-hydrogens against the reported group - 7 8 -C y c l o h e x a n e - , ( B ) B e n z y l - . ( O t r a n s - a n d ( D ) c i s - 1 2 - A c e n a p h t h e n y l - D e r i v a t i v e s - 7'9 -electronegativities (7)(162)(Figure 11) revealed that the nitroxy group was apparently much more electronegative than was hitherto reported (I6l)(l62) and i t appeared to be even more electronegative than fluorine (3,93 kcal/mole). From the straight line plots (Figure 11) of the values for the hydrocarbon^alcohols and acetates the extrapolated value of the electronega-tivity of the nitroxy group was 4.18 kcal/mole (Table XVII). The meso- and dl-hydrobenzoin derivatives did not give a linear cor-relation and this was attributed to the f l e x i b i l i t y of the molecular confor-mations which could therefore be different for each type of derivative. The relative contribution to the shielding of the o. -protons by the hydrocarbon portion of the molecule, was assessed from the intercepts on the-axon of Hie plol uud was in the order cyclohexyl > benzyl > aoenaphthenyl which corresponded exactly to the known order of the positive inductive effect of these groups. The twidity of the a(N02) N)s(N02) V(0N) V N 0 2 > 3 (N02) Benzyl Nitrate 1626 1278 860 756 697 me s o-Hydr obenz o i n Dinitrate 1625 1645 1652 1279 1295 1310 842 860 770 (697)* 709 dl-Hydrobenzom Dinitrate 1636 1660 1271 852 867 761 (696) Benzoin Nitrate 1643 1672 1265 1285 840 844 745 (675) (690) (705) trans-1 p2-Acenaphthenediol Dinitrate 1649 1621 1283 860 754 708 cis-1,2-Ac enaphthe ne d1o1 Dinitrate 1648 1627 1294 1278 868 758 (728) 1269 Values i n parentheses represent bands overlapped by others i n the spectra of the parent alcohols. Five principal infrared absorption bands were defined for the nitroxy group by Guthrie and Speddmg (166) as shown i n Table XVII. The infrared ' -spectra of the aromatic nitrate esters synthesized i n the present work were recorded i n the condensed state. - 83 -Table XVIII Infrared Frequencies of Nitroxy Groups from Solution Spectra** Band I II I l l IV V Assignment S! (NO,) a 5(NO2) Benzyl Nitrate 1637 1276 842 750 6"94 trans-1,2-Acenaphthene— d i o l Dinitrate 1646 1276 843 750 705 cis-1,2-Acenaphthene- 1655 1285. 844 (779)* (726)* d i o l Dinitrate ** _2 10 M i n cyclohexane. as neat liquids^or as potassium bromide windows (Table XVII) and also for some of them i n 10 2M cyclohexane solutions (Table XVIII). It was clear from the tabulated data that the m u l t i p l i c i t y of bands i n the s o l i d state was a function of the crystal structure rather than of intramolecular interaction of v i c i n a l nitroxy groups since only singlet absorptions were observed i n the solution spectra. The s i g n i f i c a n t l y higher frequencies (9 to 18 cm ^ ) of the asym-metric and symmetric stretching frequencies i n the solution spectrum of cis-l,2-acenaphthenediol dinitrate compared to those of the trans-isomer and benzyl nitrate has been attributed to steric interaction of the con-tiguous nitroxy groups i n the c i s - d i n i t r a t e (144). - 84 -C. U l t r a v i o l e t Spectra (UV) The UV- spectrum of 1so-amylnitrate i n ethanol solution was. similar to that of other aliphatic mononitrates (95). It was suspected from the spec-trum that the t a i l on the high wavelength side hid weak band(s) of unknown origin, as shown i n Figure 13. It was reported recently (105) that the shoulder at 2700 R. was not observable i n dmitroxy compounds such as isosorbide d i n i t r a t e (XLX) indicating that upon disubstitution the "Tr —• r^r band became twice as intense as i n the mononitrates and overshadowed the weak n-. b rown ; * : c o l o u r l e s s be fo r e s p r a y i n g , w h i t e spo t cn pink b a c k -g round a f te r s p r a y i n g w i th R-3 - 97 -The same yellow photo-products were detected i n the case of benzoin nitrate but i n addition other colourless spots, not present among the products originating from the other nitrate esters showed up after spraying with reagent R-3. Other differences i n the products from benzoin nitrate (an c*- -keto nitrate ester) and the other nitrate esters are shown i n Table XXI» B. Id e n t i f i c a t i o n of Photolysis Products. The preliminary experiments showed that jwiitrophenol and probably other nitrophenols were formed from the solvent benzene. This consecutive (or simultaneous) oxidation and n i t r a t i o n of the benzene ri n g could have taken place only at the expense of the nitrate esters since no other oxygen and nitrogen source was available i n the system. This observation made i t evident that care had to be taken to distinguish between products originating from the solvent and those originating from the nitrate ester i n the photolysis, meso-Hydrobenzoin din i t r a t e was therefore irradiated i n thirteen different solvents* on a micro scale to obtain preliminary information and then i n three solvents} benzene, ether, and alcohol for more detailed inves-t i g a t i o n . ( 1 ) Products from Solvents Irradiation i n Benzene Solutions Benzene solutions of nieso—hydrobenzoin d i n i t r a t e (0,01 M) waSh" i r -radiated m the photoreactor (Figure 47) with Corex f i l t e r (Figure 46) for various lengths of time, ji-Nitrophenol was isolated from the i r r a d i a t i o n mixtures by vacuum and steam d i s t i l l a t i o n s and i t s identity was confirmed Diethylamine, ethanol, ethylmercaptan, dioxane, cyclohexane, cyclo-hexane, benzene, carboi (titBulf ide, acetic acid, ethyl.*|cetate, diethyl ether, acetal, and phenol. -98 -by i t s nitrogen content and by thin layer chromatography. For i s o l a t i o n and i d e n t i f i c a t i o n of the other yellow photolysis products, the unreacted nitrate ester and the j3-nitrophenol were removed by column chromatography and the r e -sulting mixture was analyzed by TLC as shown i n Figure 21 (A-P) for various times of i r r a d i a t i o n . It was evident that not a l l of the coloured spots originated from the primary interaction of nitrate esters and benzene since the pattern changed with the time of i r r a d i a t i o n . Certain spots (G and J) appeared early (0.5 hrs.) and their concentration after reaching a maximum value gradually decreased with.time* Other spots (D, H, L, M, and 0) were not detectable i n the early stages of the i r r a d i a t i o n and some of them started to appear after 5 or even 10 hours, indicating that they either accumulated by a very slow process which did not dominate over the main course of the reaction or that they were secondary products originating from the ni t r o compounds formed i n the primary process. Because of the s i m i l a r i t i e s i n R^ values, column chromatographic separation under similar conditions f a i l e d completely (156) and thick-layer microadsorption chromatography (105) ( 208) also did not provide satisfactory results. Repeated TLC was used successfully for separating these products but because of the micro scale of the method and the large number of com-ponents, only the compounds which occurred i n r e l a t i v e l y high concentration were isolated on a milligram scale. Figure 22 shows a chromatogram of ten pure compounds isolated from the photoreaction products. Comparison with known compounds indicated that E might be 2,4-dinitrophenol, while G agreed very well with 2,6-dmitrophenol and F was similar to 2,6-dinitro-4=phenylphenol. The infrared spectra of C and D (Figure 23) indicated the presence of nitroxy groups as well as aromatic nit r o groups i n a structure similar to that of meso-hydrobenzoin d i n i t r a t e . - 99 -F IGURE 21 C h r o m a t o g r a p h i c S e p a r a t i o n of P r o d u c t s f r o m P h o t o l y s i s of m e s o — H y d r c b e n z o i n D i n i t r a t e in Benzene S o l u t i o n ( M - 3 j S-5 ) t f. f l u o r e s c e n c e (UV ) y y e l l o w , o o r a n g e , o-br orange b r o w n ; b r o w n = b r . — 100 — f f l uo rescence ( U V ) ; y y e l l o w ; o orange , o - b r orange b rown , b b rown - 101 -Since no unreacted meso-hydrobenzom dinitrate was present, these compounds were i d e n t i f i e d as ring-nitrated meso-hydrobenzoin derivatives (Figure 23). Although C showed a broad OH band at 3400 cm 1 this was possibly due to moisture smoo the parent meso-hydr obenzo i n exhibited a sharp OH peak. On the basis that R ^ ( C ) ^ R^(D) and that D showed up later than C i n the photoreaction (Figure 21) i t was suspected that C was the 4-mononitro- and D the 4,4' - d i -ni t r o meso-hydrobenzom d i n i t r a t e . These compounds would be the products of intermolecular n i t r a t i o n rather than of intramolecular re-arrangement. The infrared spectrum of E, the petroleum ether soluble portion of E', (Figure 24) matched that of 2,4-dinitrophenol even i n the fingerprint region so that there was no doubt that E was 2,4-dinitrophenol i n agreement with TLC r e s u l t s , however, there were two additional peaks i n the spectrum of E at 2900 and 1726 cm 1 indicating that some E 1 contaminated the sample. E' and E" gave d i s t i n c t i v e although quite similar spectra (Figure 24) and were probably isomeric dinitrophenols. The infrared spectra of F, K, and N were compared to that of 2,6-dmitr 0—4—phenyl phenol (Figure 25) but no firm i d e n t i f i c a t i o n was pos-sible i n these cases* However, the f a c t that orange spots similar to F occurred also among the photolysis products from dl-hydrobenzoin d i n i t r a t e and from benzyl nitrate but were completely absent i n the i r r a d i a t i o n mix-tures from c i s — and trans-1,2-acenaphthenediol dinitrates suggested that this compound originated i n the nitrate esters rather than i n the solvent. On the other hand i f F were r e a l l y 2,6-dinitro-4-phenylphenol that would mean that phenyl radicals were generated with three out of fiv e nitrate esters and at-tacked the simultaneously formed nitrophenols as represented i n equation 110. The suspected reaction would be somewhat analogous to one ( i l l ) reported by Price and Convery (175) for meta-dinitrobenzene. F IGURE 23 Infrared Spec t ra of mcso-Hydrobenzo in D in i t ra te (A) N i t robenzene (B ) and P h o t o l y s i s P r o d u c t s of A ( C and D) — 103 -1 1 1 1— A ^ / i i r 1 1 r~ 1 p-^C-NO \ A 1 AA A i . i i ' i i i • t 2 1 r 1 1 ' 1 1 1 1 1 1_ 1 1 1 1 1 1 , , t E ' i i i i 1 i i 1 , i 1 1 1 1 1 1 i 1 1 1 1 1 1 1 1 ..... 1 1 l 1 1 1 - ' I I I 1 1 , 4 0 0 0 3000 2000 1800 1600 1400 1200 Tooo"" 8 0 0 cm" 1 FIGURE 24 Spec t ra ( IR) of 2 , 4 - D i n i t r o p h e n o l (A) and Pho to l y s i s P r o d u c t s f rom m e s o - H y d r o b e n z o i n D i n i t r a t e ( E , E ' a n d E" ) - 104 — 3 5 0 0 2 5 0 0 1900 1700 1500 1300 1100 90O 700 cm-1 F IGURE 25 Infrared S p e c t r a of 2 , 6-D .mt ro-4-Pheny lpheno l ( A ) and Pho to l y t i c P r o d u c t s f r o m r n e s o - r i y d r o b c n z o i n D,n, trat< (F , K and N ) - 105 -Irradiation i n Ether Solution, No product which originated solely from the solvent was isolated i n the i r r a d i a t i o n of meso-hydrobenzoin dinitrate i n ether solution, A major spot in the chromatogram was located ahove the unreacted nitrate ester which was quite unusual since i n most cases the nitrate ester possessed a higher R^ value than any of the photolysis products. The characteristic feature of this compound was i t s bright fluorescence under u l t r a v i o l e t l i g h t and upon spraying the chromatogram with concentrated n i t r i c acid-sulfuric acid mixture (R-l) the spot turned green immediately. Only solvents which pos-sessed the structure X—0—Y, i . e . diethylether, ethyl acetate, acetal, and dioxane, produced this or similar compound. From the high R^ value the compound was suspected to be the ketal of the corresponding diketone (benzil), however, acid catalysed hydrolysis of the isolated c r y s t a l l i n e compound (m.p. 95-98°) f a i l e d to y i e l d benzil. The NMR spectrum of the sample was recorded and there seemed to be l i t t l e doubt that some sort of alkylation had taken place on the parent molecule (Figure 26), i n other words the s o l -vent molecule was incorporated (in part or i n f u l l ) into the meso-hydroben-zoin moiety, however, further work would be required to elucidate the struc-ture of this unknown product. Irradiation i n Alcohol Solutions Acetaldehyde was isolated as the 2,4-dmitrophenylhydrazone from i r r a d i a t i o n of ethanol solutions of the n i t r a t e s . The same product was ob-tained by i r r a d i a t i n g either me s o-hydr obenz oin dinitrate or benzyl, nitrate under similar conditions. The 2,4-dmitrophenylhydrazone was i d e n t i f i e d by i t s melting pointy elementary analysis, t h i n layer chromatography and nuclear magnetic resonance spectrum. This oxidation of alcohol to aldehyde during the photo—reduction - 1 0 6 -B H L - i L L. J , 1 1 , l _ V -1 L JLJL -1 _ J ! _ , l _ -\ L -J , l_ -J l_ 8 0 70 _ _ i 60 5.0 4 0 3 0 2 0 L 4 5 0 400 350 300 2 5 0 200 150 1 0 0 ppm (h) 10 0 50 0 cps F IGURE 26 NMR Spec t ra of 1, 2 - D i p h e n y l e t h a n e D e r i v a t i v e s A m e s o - H y d r o b c n z o m , B meso-Hydrobenzo in D i n i t r a t e , C Product f r om B I rradiated in Ether So lu t ion - 107 -of nitrate esters seemed to be analogous to the oxidation of benzene to phenol. ( 1 1 ) Products from Irradiated Nitrate Esters. The nitrate esters during photolysis s p l i t up to fragments, the nitrogen-oxygen portion reacted with the solvent or active solute but gaseous nitrogetvoxides were also detected. Semi-quantitative mass spectral analysis indicated a r a t i o of 40si for NOtNO, i n a gas mixture which was a blue s o l i d when trapped at 77°K. Skeletal fragments of the nitrate esters were also isolated and id e n t i f i e d i n most cases. Irradiation of 1,2-Acenaphthenediol Dinitrates. Both c i s — and trans-1,2—acenaphthenediol dinitrates irradiated i n benzene solution yielded, i n addition to some unidentified fluorescent spots of low values, naphthalene—1,8-dialdehyde which was isolated and i d e n t i f i e d both as the 4-nitrophenylhydrazone and i n the oxidized form as napfo.thalene-1, 8-dicarboxylic acid. Thus the main course of the photolysis seemed to proceed to equations 112 and 113. On the other hand i t was l i k e l y that _o-nitrophenol was not formed i n one simple step but that the n i t r a t i o n was preceded by the oxidation of the aromatic r i n g . This production of the intermediate phenol occurred at the cost of the simultaneous reduction of nitrate ester to n i t r i t e ester which i n turn photolysed to alkoxyl r a d i c a l and NO at the longer wavelengths. The s p l i t t i n g of the carbon-carbon bond between adjacent nitrate ester groups (112 and 113) has also been observed (74) m the thermolysis of 2,3-dinitroxy-butane (47) as discussed i n the introduction. The simi-l a r i t y of the reaction i n the photolysis of v i c i n a l dinitrates was experi-mental evidence that i n the solution photolysis of nitrate esters just as - 108 -- -109 -i n thermolytic reactions alkoxyl radicals play a v i t a l role as intermediates. Irradiation of Benzyl Nitrate Photolysis of benzyl nitrate in both benzene and alcohol solutions gave r i s e to the same products originating from the solvents, namely n i t r o — phenols from benzene and acetaldehyde from alcohol, as were found with the hydrobenzoin d i n i t r a t e s . Consequently, there can be no doubt that the frag-mentation pattern of the nitroxy group was the same for this primary mono-nitrate as for the secondary dinitrates, i n other words i t was a function of the nitroxy group and independent of the structure of the rest of the molecule. The intermediate alkoxyl r a d i c a l probably underwent one or more of the following reactions as described i n recent reviews (91) (70): 1. Intermolecular hydrogen abstraction (114) 2. Disproportionation (115) 3. Radical elimination, which could have occurred i n two different ways. (a) The elimination of the -hydrogen (116) (energetically unfavourable) (70) (176). (b) Carbon-carbon scission (117) The carbon-carbon cleavage i n benzyloxyl r a d i c a l has not been previously re-ported (70) (l77)» however, t e r t i a r y alkoxyl r a d i c a l transformation with simul-taneous generation of phenyl radical and the corresponding ketone(118) has been established (70), The generation of phenyl radicals would also be consistent with the formation of biphenyl derivatives (110). Irradiation of meso-Hydrobenzom Dinitrate. The two isomeric hydrobenzoin dinitrates behaved s i m i l a r l y to the acenaphthenediol dinitrates in that C-C bond scission occurred between the - 1 1 0 -- I l l -two v i c i n a l nitroxy groups to produce aldehydes. Since i n this case the o r i g i n a l molecule was s p l i t into halves i t gave r i s e to two moles of benzal-dehyde per mole of nitrate ester. This reaction (119) seemed to be preferred over aralkyl C~C bond scission (120), however, both could have taken place simultaneously. The benzaldehyde originated from meso-hydrobenzoin dinitrate upon ir r a d i a t i o n i n both benzene and ethanol solutions and was isolated and characterized as the 2,4-dmitrophenylhydrazone. A possible mechanism for the photo-decomposition of meso-hydrobenzom dinitrate i s given i n Figure 27. Since i n the preliminary investigation of solvent effects, phenol was found to be very reactive with excited nitrate esters (upon i r r a d i a t i o n the sample turned brown within minutes) i t was not surprising that the steady state concentration of phenol could not be detected. On the other hand i t was also possible that the phenol remained t i e d up to the decomposing nitrate ester molecule i n a complex u n t i l i t was substituted by the N02 group i n i t s orjfb-position. In the generation of dmitrophenols from _o-nitrophenol by the con-sumption of a second mole of nitrate ester, the preferential substitution for the entering N02 group was l i k e l y to be on carbon 4, however, substitution on position 6 also occurred but seemed to be somewhat less extensive as judged from the chromatograms. Other activated benzene rings i n unreacted meso-hydrobenzoin dinitrate may also have acted as N02 acceptors as was discussed above. By comparison with previous work on the photonitration of d i -phenylamme (105) i t seemed that ring photonitration occurred when the aro-matic system was activated by appropriate substituents, but with only non-activated benzene rings present oxidation seemed to be the preferred reaction. — 112 — 0 2 N 0 C-Ph Ph / \ ) N 0 2 hv, OH [»•] S , N 0 2 I 0 2 M O P n hv-Ph O-N \\ O [ H C ] H 2 0 OH [«•] NOn N 0 0 0 2 N O y Ph C-Ph / C-H V 'O vo hv. 0 2 N O H^ l C-/ Ph 0 , N O C-Ph V •H + NO O . C / Ph P h C H O Ph PhCHO 0 2 N O Ph H S c C Ph O . ; J 0 2 N Q I + MO I H. Ph + PhCHO •O Ph PhCHO FIGURE 27 Possib le Mechanism of Pho to l ys i s of me s o - H y d r o b e n z o i n Din i t ra te - 113 -The fact that i n the Kemula-Grabowska reaction (67) (113) the photonitration of benzene by means of NO also produced _o-nitrophenol and 2,4-dinitrophenol provided a clue for the elucidation of the mechanism of the present reaction which w i l l be discussed i n a later section. C. Kinetic Study of the Photolysis. Five aromatic nitrate esters were subjected to kinetic measure-ments. Three of the f i v e (benzyl n i t r a t e , meso- and dl-hydrobenzom d i -nitrates) contained benzene rings while the remaining two ( c i s — and trans-1,2-acenaphthenediol d i n i t r a t e ) had naphthalene nuclei i n their molecules. A Corex f i l t e r was used i n a l l kinetic experiments i n order to e l i m i -nate the short wavelength lines of the mercury arc. In these conditions only the t a i l of the nitrate ester spectrum was involved providing n -J^ excita-t i o n of the NO, group and only negligible numbers of quanta were supplied to the — " r \ C bands of both nitrate ester and benzene n u c l e i . This p r a c t i c a l l y selective excitation gave a reasonably clear picture of the photodecomposition of these nitrate esters bearing benzene rings i n their molecules. In the case of the dinitrates of c i s - and trans-1,2-acenaphthenediol this selective excitation was not possible because the naphthalene portion of the molecule has an extinction coefficient about 100 times larger than that of the two nitroxy substituents at the same wavelengths and i t over powered the weak (£.**= 25/nitroxy group) n —»ftC absorption. On this basis i t was expected that acenaphthene derivatives would photolyse considerably more slowly than the benzene derivatives. The rate measurements were based on the rate of disappearance of nitrate esters. Since not a l l of the nitrate ester molecules which reached the excited state decomposed (because a certain fraction of them became deactivated) the rate of decomposition should be slower than the rate of - 114 -l i g h t absorption. It i s generally accepted that t r i p l e t states are involved i n photo-chemical processes, and that t r i p l e t excited states always have longer l i f e -times than singlet excited states because of their lower oncrgioo. Figure 28 i l l u s t r a t e s these principles of primary photoreactions (178) applied to the nitrate esters (NE). Whether the radiative deactivation processes (fluores-cence and phosphorescence) labelled k ^ and k ^ * would be actually obser-vable experimentally i n the case of nitrate esters was not known but i t seemed reasonable to assume that they also occurred here as they do i n many other reactions* Reaction k^ was responsible for excitation and assuming 100$ ef-ficiency, the same number of molecules became excited as the number of photons absorbed by the nitroxy group. The t o t a l energy absorbed was then dissipated by three routes, two Of which (K ^ and k ^ ) were deactivations and thus rep-resented loss of energy as far as photolysis was concerned, only the t h i r d route ( l ^ ) led to the decomposition of nitrate ester molecules forming the reaction products. The quantum y i e l d ( l 2 l ) r e l a t e d the decomposition (k,) to the t o t a l excitation (k^) assuming that one photon excited one molecule. 0 = A [.NE] decomposed (121) A [NE]excited If there was no chain reaction and only one photon was required for causing the photoreaction of one molecule, 0 would have a value between 0 and 1, If the deactivation process was negligible with respect to decomposition then 0—»-l, i f the rate of decomposition (k^) was exceeded by the rate of the deactivation process, 0 »-0. It was found that at i n i t i a l concentrations of less than 0.1 mole of nitroxy group per l i t e r the early part of the photoreaction followed a 1 N E * f l uo re s cence rad i ationless t r ans i t i on <1 e x c i t a t i o n NE, 3 N E * * -13 phosphorescence p roduc t s F IGURE 2 8 P r i m a r y P h o t o c h e m i c a l R e a c t i o n s of N i t r a t e E s t e r s t N E ) - 116 -f i r s t - o r d e r rate law as shown i n Figures 29 and 31. Since the products (nitrophenols, etc.) also absorbed i n the effective wavelength region, on more extended i r r a d i a t i o n even with low i n i t i a l concentrations (0.02 M), the reaction slowed down as observed i n the rate plot (log concentration versus time) by the departure from the straight line (Figure 29). This effect was pa r t i c u l a r l y pronounced with the acenaphthene nitrates because the products containing the naphthalene nucleus absorbed strongly i n the active region. Equation 122 described the amount of l i g h t absorbed (I ) with A respect to the incident intensity ( l Q ) as a function of the optical density (O.D.). I. = I (l-e°' D') (122) A O If 0 . D . » 1 then the exponential term would be negligible and I —»• I , i f , however, 0.D.« 1 then 1 ^ — I x O.D. since the exponential term would be approximated by the f i r s t two members of a Taylor series. This second condition was s a t i s f i e d by the low i n i t i a l concentration of the nitrate ester. On the other hand at the low concentration the Lambert-Beer law was also v a l i d and O.D. could be replaced by [NE]to give equation 123. I. - I £ t [NE| (123) A o »• — ™ Since the t o t a l O.D. or t o t a l £ was the sum of the values for the several components of the system, equation 123 could be used to calculate the l i g h t absorption for a particular mode of excitation. Substituting for the n •IT band of the nitrate ester, the calculated 1^ (einstein/sec) represented the number of excitations, which i n turn was a measure of the number of n— - cuT excited states (moles/sec) generated during the i r r a d i a t i o n . Since the rate of decomposition was proportional to the number of quanta ab-sorbed per unit time, 1^, according to 123, the rate of decomposition was also p r o p o r t i o n a l to the concentration of the nitrate ester and this was the ° 2 4 G 8 10 12 14 I r r a d i a t i o n T ime in Hours F I G U R E 29 Ra tes of P h o t o r e a c t i o n of (A) Benzy l N i t r a t e , (B) dl- a n d ( C ) m e s o -Hydrobenzo in D i n i t r a t e s , ( D) t r a n s - and (E) cis-1 2-Accnaphthcnedio l D i -n i t r a t e s in Benzene S o l u t i o n at 2 5 ° - U S -origi n of the observed f i r s t - o r d e r rate law. In the photoreactor^ the solution thickness was 0.665 cm. ^ there-> fore 1^ for one particular wavelength ( /\ ) was given by 124. i j = 0.665 I 7 1 £ [NB] (124) A o L J In t h i s investigation not monochromatic l i g h t but a portion of the spectrum of the mercury arc (Figure 33) was employed and therefore the above equation was used for each of the active wavelengths and the values of 1^ were summed up for estimating the t o t a l number of quanta/sec absorbed (125). I * O T A L = 0.665 * ( I I V ) K [ N E ] (125) A ^ O The f i r s t - o r d e r rate constants were calculated from the rate plots (Figure 29) for benzy]Jnitrate (A) and meso-, and dl-hydrobenzom dinitrates (B and C respectively). For c i s - and trans-1,2-acenaphthenediol Awere c a l -culated by extrapolation to zero time. These data are summarized i n Table XXII. Table XXII Apparent First-order Rate Constants for the Photolysis of Aromatic Nitrate Esters at 24.20°C. In Benzene Solution Compound t i ( h r s ) 2 kxlO^(sec 5l In Ethanol Solution 4 1 b t ^ h r s ) kxlO (sec" ) k/ ^ o A b H A Benzyl nitrate B dl-Hydrobenzom dinitrate 13.0 0,14810.032 9.53 0.20210.023 1.0 1.4 -6,21 Q;31O?O-.053, l'.O ,2.1 C me s o -Hydr o b e nz o i n d i n i t r a t e D trans-1,2— Acenaphthenediol di n i t r a t e E cis-1,2-Acenaphthenediol dini t r a t e 6.88 0.28010.059 0.54 3.6 io.4 d 0.46 4.2 -ot2 d 1.9 3.11 0.619-0.061 2.0 2.2 24 28 I n i t i a l concentration 0.02 mole nitroxy group per l i t e r . For benzyl n i t r a t e . ° In 0.02 M ether solution, t ± = 0.79 hours, k = 2.42^0.37 x lO^sec"" 1, 2 d Extrapolated to zero time. - 120 -Instead of the expected slower decomposition of the acenaphthene nitrates because of competitive l i g h t absorption, the results showed that they photolysed 15-30 times faster than those containing phenyl groups and indicated that photosensitization took part i n these reactions (Figure 29, D and E ) . A possible explanation was that the naphthalene portion of the molecule absorbed most of the incident l i g h t , but the excited aromatic moeity did not deactivate by the usual processes (fluorescence, phosphores-cence, e t c ) but rather released i t s excitation energy v i a energy transfer to the unexcited nitroxy groups. In other words the naphthalene portion of the acenaphthene molecule acted as a " b u i l t i n " photosensitizer. A somewhat similar energy transfer but i n the opposite sense has been reported for a mixture of benzophenone and naphthalene. Benzophenone was s e l e c t i v e l y excited through the n-^SC t r a n s i t i o n and after the molecules had passed from the f i r s t excited singlet state to the f i r s t excited t r i p l e t state, energy was transferred to the unexcited naphthalene and brought i t to the t r i p l e t state ( t r i p l e t — " t r i p l e t energy transfer) (Figure 30). The mechanism of this process has been confirmed by phosphorescence spectroscopy (179), flash photolysis (180) reaction kinetic measurements (181) and ESR spectroscopy ( l 8 2 ) 0 In the present case energy ^ s transferred intramolecularly from the naphthalene portion to the nitroxy group. Energetically (Figure 30) t h i s process seemed to be^smglet —* singlet t r a n s i t i o n similar to those ob-served i n other intermolecular (183) (naphthalene —*- alkyliodide) and i n t r a -molecular (184) (naphthalene—anthracene XLIII, XLIV and XLV) processes. The rates of photolysis of benzyl nitrate (k^) and of meso-hydrobenzoin (k^) d i n i t r a t e were also determined i n ethanol solutions (Figure 31 and Table XXII). The fact that the r a t i o k^A-^ was about 2 i n both benzene and ethanol solutions indicated that the same type of-reaction - 121 -3 0 , 0 0 0 2 0 , 0 0 0 10,000 F i l te r cut off S T Benzophenone Naphtha lene N i t ra te E s t e r Acenaphthene TT— 1 < rir* <7r — fTr* FIGURE 30 T r i p l e t - T r i p l e t E n e r g y Transfer Be tween B e n z o p h e n o n e and Naphtha lene (180) and Si ng l e t -S ing l c t Energy Transfer w i th in 1,2 Acenaph thened io l D i n i t r a t e s XLIII XLIV XLV - 122 -occurred with the two nitrate esters even i n different solvents. On the other hand the observation that the rates of decomposition of the same nitrate ester, meso-hydrobenzom d i n i t r a t e , was different in three different solvents! ^Et 0 : ^EtOn" S kPhH = 8 , 6 S 2 * 2 : 1 ( F l g u r e s 2 9 a n d 3 1 a n d T a t , l e XXH) indicated that the solvent i n which the photodecomposition took place p a r t i c i -pated i n the reaction. This agreed with the evidence of solvent participation gained from the preliminary experiments and product analyses. Weller (185) summarized the characteristics of fast reactions of excited molecules into four classes: (a) Quenching of fluorescence (probably electron transfer) A* + B — (A(?) ' B ( i ) ] —«-A + B (126) (b) Complex formation A* + A—~A*A (127) Or A* + B — A*B (128) (c) Acid—base reaction A*H + B —«• A* + HB + (129) A* + HB — - A * & B" (130) (d) Isomerization A * — A * (131) Furthermore, i t was shown that the probable occurrence of any of these reactions in the excited state could be estimated on the basis of the UV spectra i f the reaction occurred, even to a minor extent, i n the ground state. A recent report (186) showed that the a c i d i t i e s of weak acids were enhanced i n t h e i r excited states with respect to their ground states. For example, phenol became some 20,000 times more acidic upon UV-irradiation ( i . e . pK =10.02, pK = 5.7) according to reactions 129 and 132. Si £L ' ' - 1 2 3 -2 0 . 0 15 0 L 1 0 0 -A \ 5 0 " O x i——i Q. ZD o o 1 , 0 -A TV B \ \ 0 . 5 -1 , , , , , 1 1 1 . 1 1 1— 2 4 6 8 1 0 1 2 1 4 I r r ad i a t i on Time in H o u r s F I G U R E 31 R a t e s of P h o t o r c a c t i o n s of m e s o - Hyd robenzom D i n i t r a t e (A) and Benzy l N i t r a t e ( B ) in E t h a n o l and c f m e s o -(C) H y d r o b e n z o m D i n i t r a t e in E t h e r at 24 2°C - 124 -PhOH + H 20 -vFhO - + H 30 + (132) If one considered compounds A and B reacting with each other i n the ground state according to equation 133 and K ° A (+B) „ ** A 1 (+B«) (133) / * \ / \ 0 * excited A ( i . e . A ) also reacting with B (134); K and K would represent the corresponding equilibrium constants. # A* (+B) „ K " A*' (+B«) (134) According to Weller (185) equation 135 would give the r a t i o of the two equilibrium constants. In K* = A E - A E ' =- he A ^ (135) K RT kT where "A~?is the frequency interval between the long wavelength absorption bands of A and A 1. Equation |^ "l35 J holds with the assumption of equal reac-ti o n entropies i n the fluorescent and ground states" ( i . e . AH— ^ H = - A E ' ) . In the present case Weller's reaction (b) (128) could be con-sidered as the formation of a charge-transfer complex and applied to the previously discussed (Figure 15 ) charge-transfer interaction between nitrate ester and solvent according to equation 136, and the energy level diagram i n .Figure 32. The K /K rat IOS were calculated from n—*Tc band (2700 A*) and the charge transfer bands (2900-300QA) for benzyl nitrate (Figure 16 Table XIX ) according to equation 135 and provided a sequence of values K° + NE -»- Solvent===5: NE Solvent (136) NE + Solvent,, - NE~ 'Solvent (Table XXIII) which was consistent with the order of the ionization poten-t i a l s of the solvents. The order of the individual calculated values - 125 -(137) did not agree with the order of experimental rate constants (139), * / o however, the r a t i o K /K | AH N E * Solvent* NE 4 bo-nd NE AE C - T boond NE Solverit AH Figure 32. Energy Level Diagram for Nitrate Ester -Solvent Complex Excitation. (Table XXIII) d e f i n i t e l y showed that charge-transfer interaction between excited nitrate ester and solvent was much more pronounced than the same interaction i n the ground state. Table XXIII Calculated Ratios of Charge-Transfer Equilibrium Constants for Benzyl Nitrate i n Solution Solvent 4 (eV) ( c m ) (cm ) (cm ) he A ^ he AN> K / O kT 2.303kT benzene ether ethanol 9.245 33,223 3,814 18,44 8.007 1.0x10' 9.53 37,037 33,898 3,139 15.18 6.590 3.9xl0 6 10.50 34,247 2,790 13.50 5.857 7.2xl0 5 8 a From reference (142) b £ | = 4.8350 x 10 at 24.2°C kT - 126 -The known r e l a t i v e hydrogen-donating a c t i v i t y of the solvents (138) was also inconsistent with the order of the experimental rate con-stants (139). Electron transfer (K*) PhR> t 2 0 > EtOH (137) Hydrogen transfer (kg) FhH < Et^O < EtOH (138) Experimental PhH< E t 2 0 > EtOH (139) This apparent anomaly disappeared i f one accepted the suggestion of Porter (187) that i n solution both electron transfer and hydrogen trans-fer may occur together* The experimental rate constants would then mcor-porate both the charge transfer (K ) and the hydrogen transfer (it,) process constants according to equation 140 where S-H represents the photolysis solvent NE+Vp?—»-NE (singlet) '+S-H ^ K - NE*~ &-H —^2* NE-H + S (140) ^ ( t r i p l e t ) Products The selective n-^Tf excitation of nitrate ester groups i n benzyl nit r a t e , and meso- and dl-hydrobenzom dinitrates (Q{ -phenyl substituted nitrates) permitted an estimation of the quantum y i e l d of the photoreaction. The calculation of the required I^ 0^^values was carried out according to equation 125 as i l l u s t r a t e d i n Figure 33. The values of j ^ 0 ^ 8 ^ " f o r meso-and dl-hydrobenzoin dinitrates were essentially the same and for benzyl nitrate half of this figure was used since the change from d i - to mono-substitution reduced C by one ha l f . Quantum yields ( 0 ) were then calcu-lated from equation 141 where k = CxXj I S . . 5i 0 0 = Rate of decomposition _ k exp 0^0 ^ e x p j ^ (141) Rate of excitation ~ t o t a l = k INE1 I A 1 I J 3 O O X 3 << a. ro 3 o c m OJ ^ CD 3 o " 2 m t" 3 « CP m ~ 3 Q- R a to ° a < *< R ro c 3 n N O R 3 ^ Q 3 a. R l/i O 3" o cr R D. O ro O p O bo IQ ( m i c r o e i n s t e i n / s e c ) ro CD _l ro O £ ( M o l a r Ex t inc t ion C o e f f i c i e n t ) O _ l _ ro O _ l _ OJ o o o o o ZD o ro a - 128 _ The numerical values for i n both benzene and non-absorbing solvents (alcohol and ether) are tabulated i n Table XXIV. Table XXIV Calculated Values of k^ for cx -Phenyl Substituted Nitrate Esters i n Benzene and i n Ethanol and Ether Solutions* A 9 I xlO 6 I x £ x l 0 6 1 1 0 ° (A) (mole cm ) (emstejn (sec cm ) .sec ) 10 6 x2 I x £. 0 in benzene i n alcohol and ether 3341 0.4 0.301 0.120 3130 2.8 2.033 5.692 3025 7.0 0.670 4.690 2967 12.0 0.441 5.292 2894 23.0 0.110 2.530 2804 42.0 0.072 3.024 2753 49.0 0.020 0.980 2700 50.0 0.014 0.520 2652 48.0 0.030 1.440 2571 35.0 0.002 0.070 21.35 24.36 1 0 6 x L x A l o x t = 1^ x 10 6(se<5 - 1) 14.20 16.20 * See equation 125 and Figures 28 and 33. The calculated quantum yields are l i s t e d i n Table XXV. They were reasonably consistent for the three nitrate esters i n one solvent and thus there was l i t t l e doubt that they a l l reacted by the same mechanism. Further-more i n benzene solution 0 was approximately 2 indicating that two moles of - 1 2 9 -of nitrate esters were decomposed per mole quanta (einstein) and this was m agreement with the proposed mechanism (Figure 27) which was based on the r e -sults of product analyses* It has been shown ( l 2 l ) that the maximum quantum y i e l d for the gas phase photolysis of NOg was also 2 and since the mechanism of NO^ photodecomposition proceeded by N-O bond cleavage (69) this lent some support to the idea that the f i f t h mode of scission (Figure 4) was predomi-nant i n the solution photolysis of nitrate esters rather than the homolytic fourth mode (Figure 4) as i n the case i n thermolysis and gas phase photolysis. Table XXV Calculated Quantum Yields for the Photolysis of <=*. -Phenyl Substituted Nitrate Esters i n Three Different Solvents at 24.2°. Solvent k-xlO 4 (sec ) k xlO^(sec exp - 1) 0 A B C A B C Benzene 0.1420 0o148 0.202 0.280 2.08 1.42 1.97 Ethanol 0.1620 0,310 0.619 3.83 - 3.82 Diethylether 0.1620 _ _ 2.42 _ 14.9 As benzyl nitrate B: dl-hydrobenzoin dinitrate C: me s o-hydr obenz oin d i n i t r a t e The fact that the quantum yields obtained i n ethanol and ether were considerably higher than those i n benzene pointed to a chain mechanism i n the former solvents, however, further work would be required to decide whether they were r e a l l y free r a d i c a l chain processes or whether the mechanism simply required a larger integral number of molecules of nitrate ester per quantum than i n the case of benzene solutions. - 130 -D. ESR Study of Nitrate Ester Photolysis. The photolysis of the nitrate esters was carried out i n the cavity of an electron spin resonance (ESR) spectrometer i n the hope that free r a d i c a l intermediates might build up a s u f f i c i e n t l y high steady state concentration to give a detectable signals This was found to occur and signals obtained from the v i c i n a l dinitrates irradiated in benzene solution at room temperature are shown i n Figure 34. The spectra of the four nitrate esters were very similar, however, the d i s t i n c t l y greater intensity of the spectrum of the 1,2-acenaphthenediol dinitrates compared to the hydrobenzoin dinitrates indicated a higher steady state concentration of ra d i c a l intermediates i n agreement with the kinetic data and the previously proposed energy transfer process. A more powerful lamp (G.E.-A-H6) was used for the i r r a d i a t i o n of the dl-hydrobenzoin di n i t r a t e (Figure 34, D), however, the intensity of ESR spectrum was only s l i g h t l y i n -creased over that of the meso-isomer (Figure 34, C) at the same microwave power (100 m W). The s t a b i l i t y of these free ra d i c a l species was demonstrated i n an experiment (Figure 35) where the irradiated trans-1,2-acenaphthenediol d i -nitrate was kept i n the dark at room temperature. The spectrum obtained after ten days with the same microwave power (10 m Watts) (Figure 35) was extremely weak compared to the original signal, however, a more intense microwave power (330 mW) revealed that the species responsible for the spec-trum had not decayed completely. This extraordinary long lifetime suggested that the free r a d i c a l present was probably some sort of r e l a t i v e l y stable complex with an odd electron. Recently Calvin and coworkers (l88) reported ESR spectra originating from interaction of chloranil (tetrachlorobenzoqumone) and N,N,Nf ,N* , - t e t r a -3450 3400 3350 3300 H J 1 i . , , 1 . . . . 1 , , . 1 1— _1 . 1 . 1 , 1 . 1 ,_ 196 198 200 202 2 04 g F IGURE 34 S t e a d y S ta te ESR S p e c t r a o f I r r a d i a t e d c is - (A) and t r a n s - ( B ) 1 , 2 - A c e n a p h t h c n e d i o l D i n i t r a t e s a n d m e s o - ( C ) and _d_l-(D) H y d r o b e n z o m D i n i t r a t e s in Benzene S o l u t i o n at R o o m T e m p e r a t u r e - 132 -3450 H 1 1- H I-H( gauss) 3 4 0 0 3 3 5 0 —I . , , , L _ 3 3 0 0 A 1 1 1_ 1 9 6 19 8 2 0 0 202 2 0 4 g-A ( 1 0 mW) B (10 mW) C ( 330mW) F I G U R E 3 5 ESR S i g n a l s of I r r ad i a t ed t rans-1 2-Acenaphthene d io l D in i t r a t e Obta ined I n i t i a l l y (A) a n d A f t e r Ten Days in the Da rk ( B and C ) - 133 -methyl-p-phenylenediamine which involved charge-transfer complex and semi-quinone r a d i c a l . The reaction occurred i n several stages but "the free radicals disappeared completely i n the course of one week" (188). No ESR signal was detected i n irradiated alcohol solutions of the nitrates. This r e s u l t could be rationalized i n at least three different ways: (1) The spectrum obtained i n benzene solution was not due to the transforming nitrate ester but rather to the nitrophenol formation and thus i t did not occur i n ethanol solution. The experimental observation that signals appeared instantaneously i n the benzene solution would not favour this explanation. (11) The spectra obtained were due to the transforming nitrate ester but the mechanisms were different i n the two solvents. Although the photo-products which originated from the v i c i n a l dinitrates were the same from benzene and alcohol solutions i t was s t i l l possible, but not very l i k e l y , that there was a difference i n the free radical intermediates. ( n i ) The spectra were due to the transforming nitrate ester but the concentration of the species with unpaired spins was too low for detec-t i o n . This t h i r d p o s s i b i l i t y agreed with the previously proposed charge-transfer complex formation (e.g. from kinetics) since K for alcohol was small, therefore, the complex concentration was low and a continuous fast removal of the complex from the system (large kg) would decrease the steady state concentration further. The situation was d i r e c t l y opposite to this i n benzene solutions and the proposed mechanism would therefore also ex-plain the difference i n ESR behaviour. The s p l i t t i n g i n the spectrum of NO^ was about 50 gauss while the s p l i t t i n g of the three main lines i n these spectra (Figure 34 and 35) was - 134 -F I G U R E 36 ESR S p e c t r u m of N 0 2 (A ) in So l id A r g o n ( 2 1 4 ) a n d ( B ) G e n e r a t e d f r o m trans-1 ( 2 - A c e n a p h t h c n c d i o l D i n i t r a t e in E P A a t 77 ° K . - 135 -about 30 gauss and therefore the solutions did not contain appreciable amounts of NO2. If the mechanism of the photodecomposition r e a l l y involved RO-NO2 scission as found i n gas phase photolyses and the liberated NC^ was responsible for the reaction with the solvent then, at the steady state concentration, NX^ should have been detectable. Since the spectrum of NO2 did not appear at room temperature the ESR results established the pre-viously suggested f i f t h mode of scission (Figure 4) as the primary cleavage i n solution photolysis of nitrate esters. Irradiation of nitrate ester i n EPA (ether-i-pentane-alcohdl 8s3:5 vol/vol) glass at 77°K, however, did give r i s e to NO2 as shown i n Figure 36. This phenomenon could be rationalized as follows: The time required for "reactive interaction" with surrounding solvent molecules at 77°K might be longer than the lifetime of the excited state. Thus the ex-cited state, not making contact with reactive molecules within i t s 1 l i f e -time, as in gas phase, underwent decomposition at the weakest point, i . e 0 the R0-N02 bond. In a control experiment NO gas was irradiated i n benzene solution at the same i n i t i a l nitrogen concentration (0.2 M) as used for the nitrate esters. The close s i m i l a r i t y of the spectrum obtained to that obtained with the aromatic nitrate esters (Figure 37) pointed to a similar intermediate i n the two cases. The products from these reactions were also similar as men-tioned i n a previous section. In a l l ESR spectra obtained from photolysis of nitrate esters three lines were distinguished as major components. These were attributed to the s p l i t t i n g of the electron resonance by a nitrogen atom (for ^N, 1=1) which meant that the unpaired spin was located on a nitrogen atom. X=Irradiation of single crystals of potassium nitrate (mounted - 136 -— H (gauss) 3 4 5 0 3 4 0 0 3 3 5 0 3300 . 1 1 1. 1 1_—I 1 1 1 1 1 1 1 1 1 1— _1 . I 1 I • \ 196 1 9 8 2 0 0 2 0 2 2 0 4 g A (10 mW) (100 mW) F I G U R E 37 ESR Spec t ra of I r r a d i a t e d t r a n s - 1 , 2 - Ace n a p h t h c n c -d io l D i n i t r a t e ( 0 1 M ) ( A ) and NO ( 0 2 M ) ( B ) in B enzene S o l u t i o n . - 137-p a r a l l e l or perpendicular to the major crystal axis) was reported to generate nitrate ion negative ion, (NO") which had. s p l i t t i n g s of 62 gauss and 32 p