" II." I , :III‘I' “. I- : . .. “' II ““1““... :{f’Ii,L.““““"I:-‘.‘:‘~"I“I; “I“IIIII“II “ I III “ _. . «I» . I... .I .I " IIIII'II IIIIII :2 III “I" II II“. 5“,“ .I I“ INT“: “ I' III. II “ ““6. “““ ““ ..“““I“ II III“ ‘i‘ IL“ I“ “II“ “I'I III I”: I. ’“ ”II-w“ III“. I H“ .I'lI‘ .. - .. .IIIIIIm. I -I- v 3:. II I “ II h I“ I“ “ I“ I“ “KIM.“ “ I“ “ I“ ““0““ FF)“ :f‘“ ,.II.. I. HUI...“ I I..--‘“- m ‘I .1“ . . 'I-a IIII I“ I] I. I .I‘Ii‘lhfw‘. I... I II. II I III I'I .I. ....I. . I. III I . I I II . *3?" I II 'I.“I“ II. III"-I"II ' III I II' .I. In 1 " ‘IIIII :. II,:~.I . I III... .IIIIII'... I'I.‘II. “i. , III“... IIII' IIIIIII II “II““I“““ .‘IIIIII “III' ‘ II. I I?“““.I-'“ “I W “I'.“':“‘II'I“I III] I I. In...“ - I“II“II II:“II I I“.“l.“ “ ”II” “I " J .4: 'I‘ y 3 I n. .I.I I... II III. .III. I I' ' I“I.'I“.““ I'I'II “I.“ II “III “i“ “I“ “I ““““ I“ III“r ““ I. "I“ m“ I “IIII. VII “““ I I :“I II I...II ."L:. .l. ... '...;_if..“‘.....Hd“ “1““ I’IIl I :I . III .N 1‘“. '. I I...I!.. VIII“. “ “ .I“ ““II. S w... ” IIIIIIHI ;; ‘t' I'.I' I.I‘....;“.I'“3 II ’I M “II“I.iI..I.I'“ I'I INHIII “I“IIII' "II"II. IIII“U I..I “II I I I‘““““II III I «III N . III . I. .I...; I II .I III. IIII . ‘IIIIII.II'IIII I.“ ’ .I III. .III .III. III I“ I I .III III I “III .I “. '.I“'. I. I I I;.I.I I‘I “ I . I I III III“ III' . 2 . .. I. I .II I III‘“‘I “I“ “I“ III III“““““I““ “III “““I I.I.“III“II..II .I' .II “III .II‘ |“I: I...” IIIIII III".l III.II INN. Wig...“ II I ;'I III‘ I I '. II I .“'II IIII ““““ I “flII“ “III“. II I II “““““ “II “““II“ I.I 'II“II“'““ I “If ' I II ‘III.IIII".I I” III.“ .II III" I... II. If “““II “““““I“ I “ 'II' “ " “ “I. "““I '“I LII“. “ “ “““IIIII IIII““I“I““I“I “III. II“ 3‘“ “I. f”. “ {III .. (fj‘II'I. III III II'IIIIIIIII IIIII IIIIII.I.“ I.“ W. “I“ 3?; ,. . ' I III" III... .II III. .. II'IIII ““““““I“II I?” III .I'.‘ I I .III .. III“ 'IIIII'I I II I III“ II In? I “““III “M ..I..“I.II I'IEI; “I“ III. II I.IIII“II“ II““III.II IIIII III.) "\0 ““I“ I. I. II. I.“ . II ' “““I'm‘ I I “H““I ‘I...I. “I “'II I I ”.IIIIH. .“““I“I“II - I- .III; 'II’I . . II',I“II.II “HI“IIII'. . . Mr. “I“.- . . '“II' “'1'“. ‘ 'I,‘ III “III IIIIIIII. IIIIIIIIIIII IIIIII' I“I IITII'II'III Izfi-‘i c:;m . b "' 3' -. .‘ - “u. .«o a h... l .0}.- —4‘.¢ ‘I v u n. N.'«.“ u g 'l.‘—I A I.II'.I “I'I 211:..._:L va ”$.14” ‘ g" .. . I 'IIIII‘II‘I" II“ III. IIII'II .IIII .. IIII ' III II III .I II II III I“ ‘I'I‘III'I I“ III II‘II’II‘““I“““I'“ I.III“““III“ .II IIIIIIII .I. .l....I.. IIIIII IIIIII .I.... I .I.IIII ...IIII .rIIIfi.“ __:- . II.I' I I “III .I‘Ii‘.I~ It. “.“I I“ .I.'IIIII...- I” ' “““U I'“‘“ “ “HI“ “““ I'M” I ““II“ III “““““““ “““ III“I “I“ INK II?!“ "II "I“I.“III. “I“.“III “III '“IHI “IN“ “.0.“ “ “II. “'I “III“I“ .I M”... II“ II I.I““‘“““;““ If] . II I .II. I III'II““IIII“ ‘1. ‘III .“ ' I 'I‘I“ “I'I“ “ “ “ ““ I I““““ “ “““l ‘“'II““’“"..II“'“'“.“7'I. ' .. ' I'.. I...“ “II... III... I , I.;.I.II.I“““. “.‘5 I“ “'! I I.IIII 'II.1 “I. “MUM“? II'II . 'III'III “ ..I.. .I. “II. “II.“I IIJII. “I“! “III III"... ;. III“ “III I“.' I“II“ I'. .I. IIIII‘ w... . I III I'll. . “ I M .. I.“ I“ 9...”. ..... I... .' . . ..I.'_ .. . I! ‘, . .H ...I . .II'I . H I” “.II M ”I... I .. I'II..‘.I‘ “LIT"... ‘I I. ”I... In“; ‘I. I in: I I.‘ . . ”IL.-.‘_ I.I....I;I.I_.“TIII - I III III; .III'.I_ “I“I‘I " ‘I' .I'IIIIIIIIII III allmllrlzlllltllllulllullwill W This is to certify that the thesis entitled Investigation of the Perturbation of the Excited State Processes of Naphthalene Crown Ether Derivatives by Complexation of the Cations of Alkali Metal Chlorides, Barium Bromide, and Silver Triflate; Investigation of the External Hea Atom Pert r tion . vy presenttleggy James M. Larson has been accepted towards fulfillment of the requirements for Ph . D . degree in Chemis try U V Major professor Date 3“ %/?78 0-7639 OVERDUE FINES ARE 25¢ PER DAY PER ITEM 0 remove ord. Return to book drop t this checkout from your rec l INVESTIGATION OF THE PERTURBATION OF THE EXCITED STATE PROCESSES OF NAPHTHALENE CROWN ETHER DERIVATIVES BY COMPLEXATION OF THE CATIONS OF ALKALI METAL CHLORIDES, BARIUM BROMIDE, AND SILVER TRIFLATE; INVESTIGATION OF THE EXTERNAL HEAVY ATOM PERTURBATION OF THE EXCITED STATE PROCESSES OF NAPHTHALENE CROWN ETHER DERIVATIVES BY ETHYL BROMIDE AND BY MEANS OF COMPLEXED HALOALKYAMMDNIUM CATIONS BY James M. Larson A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Chemistry 1979 ABSTRACT INVESTIGATION OF THE PERTURBATION OF THE EXCITED STATE PROCESSES OF NAPHTHALENE CROWN ETHER DERIVATIVES BY COMPLEXATION OF THE CATIONS OF ALKALI METAL CHLORIDES, BARIUM BROMIDE, AND SILVER TRIFLATE; INVESTIGATION OF THE EXTERNAL HEAVY ATOM PERTURBATION OF THE EXCITED STATE PROCESSES OF NAPHTHALENE CROWN ETHER DERIVATIVES BY ETHYL BROMIDE AND BY MEANS OF COMPLEXED HALOALKYAMMONIUM CATIONS BY James M. Larson The excited state behavior of naphthalene, naphthalene crown ether derivatives 2,3—naphtho-20-crown-6, 1,8-naphtho-21-crown-6, and 1,5-naphtho-22-crown-6; 2,3-, 1,8-, and 1,5-bis(methoxymethyl) and dimethyl naphthalene derivatives; and alkali metal chloride, barium bromide, silver triflate, ammonium, and haloalkylammonium chloride complexes of the above named naphthalene crown ether derivatives was investigated in alcohol glasses at 77 K. Ultraviolet absorption, fluorescence, and phosphorescence spectra were determined at 77 K. Fluorescence and phosphorescence quantum yields (relative to naph- thalene) and lifetimes were also measured. The dissertation considers complexation as a tool to study a phenomenon - naphthalene's excited state behavior - by characterizing James M. Larson the phenomenon in terms of the way in which it is perturbed by a variety of complexed species (perturbers). The approach is to con- sider changes in rate constants of excited state processes as a function of perturber and of crown by which it is complexed. The relative advantages and disadvantages of the proposed per- turbational method are compared to those of other perturbational methods which have been used to study excited states. The major advantage of the proposed method is that it allows the investigation of effects of perturbers which aren't chemically affixed to the chromophore yet whose orientation relative to the chromophore is reasonably well defined. A complicating feature of the proposed method is the perturbation due to chemically affixing the complexing agent to the chromophore. Methylene groups were interposed between the oxygens of the crown ether and the naphthalene nucleus to elec- trically insulate the chromophore from the oxygen atoms. The per- turbation due to attachment of the complexing agent is considered by comparing 2.3-, 1,8-, and 1,5-dimethy1 andflg—(methoxymethyl) substituent and complexed perturber induced changes in spectra and rate constants to those induced by complexation of metal cations. It is concluded that changes in rate constants due to complexed metal cation perturbers are different from those due to substituents and are characteristic of the properties of the metal cation and of where and how it is held by the crown ether complexer relative to the chromophore. Perturbation of excited state behavior by complexed haloalkyl- ammonium chloride salts is considered, and, from a comparison to James M. Larson perturbation by alkylammonium chloride salts, it is concluded that the perturbation due to complexation of the ammonium group can be "separated" from those due to perturbation by the alkyl halide func- tion. The results provide new information on the external heavy atom effect. The relative susceptibilities of rate constants for excited state processes to perturbation by complexed haloalkylamr monium salts are different for each crown. Changes in quantum yields and triplet lifetimes for naphthalene, naphthalene crown ether derivatives, and various alkali metal come plexes of these crowns due to perturbation by ethyl bromide containing glass (20% by volume) were also determined. The major conclusions of the investigation are: 1) there is a directional dependence for external excited state perturbation (from results of ethyl bromide and haloalkylammonium salt perturbation); 2) there is either a strong distance and/or directional dependence (from metal cation perturbation); 3) the positive charge of cationic perturbers in itself induces little change in the rates of excited state processes of naphthalene derivatives; 4) the external heavy atom effect due to alkyl halides does not require the mediation of charge transfer states; 5) the effects of two external perturbers are not additive and that the perturbation due to one external per- turber {ngL, ethyl bromide) decreases as the perturbation due to a complexed cation increases; and 6) there is no general order for the susceptibility of rate constants for excited state processes; the relative susceptibility is found to depend upon both the per- turber and where it is held relative to the chromophore. DEDICATION I dedicate this dissertation to my parents, family, and friends, for their economic and moral support; and to my teachers, both from The College of St. Thomas and Michigan State University, for their en— couragement and for their excellence in teaching. I would especially like to remember here my research director, Dr. Lynn R. Sousa, for his patience, encouragement, warmth as a human being, and his excellence in directing my development as a scientist. Also, I would like to remember Dr. William D. Larson (College of St. Thomas) for removing the terrors of introductory chemistry through his excellence in teach- ing and his kindly manner, Dr. Richard J. Connell, Dr. Thomas D. Sul- livan, and Rev. Vincent E. Rush (College of St. Thomas, Department of Philosophy) for their excellence in teaching and for providing me with a general intellectual orientation and help in training my mind to think; and to Dr. Hubert R. walczak, for his excellence in the teaching of mathematics. I would like to remember here the members of the Lynn R. Sousa group, with whom.I have had the privilege ofassociating both as friends and colleagues: Houston 8. Brown Joseph G. Bucher, III John E. Emswiler Mark R. Johnson Sandra Klassen Jerry L. Meinzer Also, I would like to remember here Barb A. Duhl-Emswiler and Kim.S. Chamberlin colleagues who have become dear to me as friends. 11 ACKNOWLEDGMENTS It would not have been possible to obtain the large quantity of data contained in this dissertation without the help of many people. I am deeply indebted to my colleague Houston S. Brown for the many useful computer programs which he wrote and especially for his work in computer interfacing the Hitachi/Perkin-Elmer spectrophotofluromoter used. The large number of precisely determined quantum yields reported herein would not have been possible without the use of a computer interfaced instrument. I am also indebted to Dr. Lynn R. Sousa, my research director, John E. Emswiler Mark R. Johnson, Joseph G. Bucher, III, and Sandra Klassen for their help in manually digitizing spectra before the computer interface was available. I would like to thank Dr. Peter J. Wagner for use of his Hitachi/Perkin-Elmer spectrophotofluoromoter, without which the high quality corrected emis- sion spectra, precise spectroscopic energies, and accurate quantum yields would not have been possible. Also, I would like to thank Dr. Wagner for use of his Aminco-Bowman spectrophotofluromoter, which was used for the early experimental work and used throughout for phosphores- cence lifetime determinations. I thank Dr. Michael Thomas, a post- doctoral with Dr. Wagner at the time, for technical assistance in the early part of the project in the determination of quantum yields and lifetimes and for many useful theoretical discussions. Dr. Thomas modified the Aminco-Bowman spectrophotofluromoter with a solenoid operated shutter, which greatly facilitated phosphorescence lifetime determinations. I would like to thank Dr. Ashraf El-Bayoumi for use iii of his single photon counting apparatus and David B. Carr for techni- cal assistance in obtaining fluorescence lifetimes. I am very grate- ful to Dr. Lynn R. Sousa for his help with fluorescence lifetime determinations, digitization and computer analysis of both phosphor- escence and fluorescence lifetimes, and help with synthesis of the crown ether derivatives. I consider it an honor and privilege to have had the opportunity to have worked on this project for him. I am also indebted to him for his help in pushing through the difficult theoretical problems which arose in interpretation of the results. I also would like to thank Dr. Donald G. Farnum for making his group meetings open to members of other groups. His group meetings provided an excellent opportunity for discussion of topics in organic chemistry and for practice in problem solving. I consider myself to have been privlege to observe the Farnum mind at work on solving a problem. Finally, I would like to thank the National Science Foundation, the Petroleum Research Fund, and Michigan State University for financial support of this work. iv TABLE OF CONTENTS Chapter Page LI ST OF TABLES o o o o o o o o o o o o o o o o o o o o o o o o 0 v1 1 LIST OF FIGURES . . . . O 0 . . C . O O . . . . C . C . . . . xiv INTRODUCTION 0 O O C . . O O C . . . . . . O . O . O C O O C O 1 General Methodology. . . . . . . . . . . . . . . . . . . Excited State Processes. . . . . . . . . . . . . . . . Perturbational Methods . . . . . . . . . . . . . . . . . Proposed Perturbational Methods. . . . . . . . . . . . . 13 Proposed Experiments . . . . . . . . . . . . . . . . . . 20 Background of Heavy Atom Effects . . . . . . . . . . . . 24 RESULTS 0 O O O O O O O O O O O O O O O O O O O O O O O O O O 40 synth8818 o o o o o o o I o o o o o o o o o o o o o o o o 40 Ultraviolet Absorption Spectra . . . . . . . . . . . . . 41 Low Temperature (77 K) Fluorescence and Phosphorescence Spectra. . . . . . . . . . . . . . . . . 85 Spectroscopic Energies.. . . . . . . . . . . . . . . . . 139 Fluorescence and Phosphorescence Quantum Yields at 77 K O O O I O O O O I O O O O O O O O O O O O 148 Fluorescence and Phosphorescence Lifetimes at 77 K. O O O O O I O O O O O O O O O O O O O O I O O O 184 Tests for Anion Effects on Emission Properties at 77 K . . . . . . . . . . . . . . . . . . . 198 DISCUSSION. 0 o o o o o o o o o o o o o o o o o o o o o o o o 199 Preamble O I O O O O 0 O O O 0 O O O O O O O O O O O O O 199 Praliminaries O O O O O O O O O O O O O O O O O O O O O I 200 Validity of the Perturbational Method Used . . . . . . . 203 Conclusions for Validity of Studying Excited State Perturbation via Complexed Perturbers O O O O 0 0 O O O O O O O O O O O O O O O O O 241 Chapter Page Validity of Method Used for Study of Perturbation Due to Alkyl Halides'gig Complexed Haloalkylammonium Chloride Salts. . . . . . . . . . . . . . . . . . . . . . . . . . 242 Method for Consideration of Perturbation by Ethyl Bromide Containing Glass. . . . . . . . . . . . 252 Directional Dependence of External Heavy Atom Effect. . . . . . . . . . . . . . . . . . . . 256 Distance Dependence of External Heavy Atm Effect. 0 O O O O O O O O O O O O O O O O O O 259 Effects of the Positive Charge of the Perturber Necessity of Charge Transfer States for the External Heavy Atom Effect. . . . . . . . 263 Effectiveness of External Perturba- tion as a Function of Perturbation Already Present. . . . . . . . . . . . . . . . . . . . . 269 Relative Susceptibility of Rate Constants for Excited State Processes. . . . . . . . . . 271 mum“ O O O O O O O O O O O O O O O O I O O O O O O 0 O 27 6 General Procedures . . . . . . . . . . . . . . . . . . . 276 77 K UV Spectra. . . . . . . . . . . . . . . . . . . . . 276 Reagent Puri ty 0 O O O O O O O O O O O O O O 0 O O O 0 O 280 Preparation of Samples for Emission Spectroscopy and Quantum Yield Dateminationo O O O O I O O O O O O O O O O O O O O O O 286 Emission Spectra . . . . . . . . . . . . . . . . . . . . 287 Spectral Energy Levels . . . . . . . . . . . . . . . . . 288 Quantum Yields . . . . . . . . . . . . . . . . . . . . . 289 Titrations Monitored by Following Relative Integrated Emission Intensity . . . . . . . . . 294 Complexation Competition Studies . . . . . . . . . . . . 299 Variables Affecting Quantum Yield and Integrated Intensity Determinations. . . . . . . . . . . 300 Fluorescence and Phosphorescence Lifetimes. O O O I I O O O O O O O O O O O O O O O O O O 305 synthesis 0 O O O O O O O O O O O O O 0 O O O O O O O O I 315 REWCES O O O O O O O O O O O O O O O O O O O O O O O O O O 3 2 4 vi Table LIST OF TABLES Absorption and Emission Frequencies and Changes in Energy Level Separations (all in cmfl) for 2,3-, 1.8-, and 1,5-Disubstituted Naphthalenes in Alcohol Glasses at 77 K . . . . . . . . . . . . . Absorption and Emission Frequencies and Changes in Energy Level Separations (Relative to Free Crown 1) for Alkali Metal Chloride, Barium Bromide, and Silver Triflate Complexes of 2,3-Naphtho-20- Crown-6 (1) at 77 K in Alcohol Glass. All Numbers 1 in cm. 0 O O O O I O O O O O O O O O O O O O O O O 0 Absorption and Emission Frequencies and Changes in Energy Level Separations (Relative to Free Crown 2) for Alkali Metal Chloride, Barium Bromdde, and Silver Triflate Complexes of 1,8-Naphtho-21-Crown- 6 (§) in Alcohol Glass at 77 R. All Numbers in cm-ooooosooooooosooooooooo Absorption and Emission Frequencies and Changes in Energy Level Separations (Relative to Free Crown 3) for Alkali Metal Chloride, Barium Bromide, and Silver Triflate Complexes of 1,5-Naphtho-22-Crown- 6 (g) at 77 K in Alcohol Glass. A11 Numbers in cm-oooooooooooooooooooooooo Absorption and Emission Frequencies and Changes in Energy Level Separations (Relative to Free Crown 1) for Alkyl and Haloalkylammonium Chloride Complexes of 2,3-Naphtho-20-Crown-6 ) at 77 R in Alcohol Class. All Numbers in cm. . . . . . . . . Absorption and Emission Frequencies and Changes .in Energy Level Separations (Relative to Free Crown‘z) for Alkyl and Haloalkylammonium Chloride Complexes of 1,8—Naphtho-21-Crown-6 (g) at 77 R vii Page .140 .141 . .142 .143 Table 10 ll 12 13 in Alcohol Glass. All Numbers in cm-1 . . . Absorption and Emission Frequencies and Changes in Energy Level Separations (Relative to Free Crown‘é) for Alkyl and Haloalkylammonium Chloride Complexes of 1,5-Naphtho-22-Crown-6 (a) at 77 K in Alcohol Class. All Numbers in cmfl. . . . Emission Quantum Yields (¢f and op) and Lifetimes (If and IP) for Naphthalene and 2.3-, 1,8-, and 1,5-Disubstituted Naphthalenes in Alcohol Class at 77 K. O O C O O I C C C O O C C C O O C . Titrations of 2,3-Naphtho-20-Crown-6 (1) with Ammonium and Alkylammonium Chloride Salts in Un- cracked 95% Ethanol Glass at 77 K Followed by Monitoring Relative Fluorescence Peak intensities. Titration of 2.00 x 10-4 tensities O I O O O I O O O O O O O O O O O O Titration of 1,5—Naphtho-22-Crown-6 (a) with Ammonium and anropylammonium Chlorides in Un- cracked 95% Ethanol Class at 77 R Followed by Monitoring Relative Integrated Fluorescence M 1 , 8-Naphtho-21-Crown- 6 (a) with Ammonium and Alkyl Ammonium Chlorides in Uncracked 95% Ethanol Class at 77 R Followed by Monitoring Relative Phosphorescence Peak In- Intangities O O O I O O O O O O O O O O O I O O O O Titration of 2,3—Naphtho-20-Crown-6 (I) with Halo- alkylammonium Chlorides in Uncracked 95% Ethanol Class at 77 K Followed by Monitoring Relative Fluorescence Intensities . . . . . . . . . . Titration of 1,8-Naphtho-21-Crown-6 (2) with Halo- alkylammonium Chlorides in Uncracked 95% Ethanol Glass at 77 K Followed by Monitoring Relative Integrate Emission Intensities . . . . . . . viii Page .145 .146 .150 .161 .162 .163 .164 .165 Table 14 15 16 17 18 19 20 21 Page Titration of 1,5-Naphtho-22-Crown-6 (a) with Halo- alkylammonium Chlorides in Uncracked 95% Ethanol Class at 77 K Followed by Monitoring Relative Integrated Fluorescence Intensity. . . . . . . . . . .166 Titrations of 2,3-Naphtho-20-Crown-6 (I) with Barium Bromide and Silver Triflate in Uncracked 95% Ethanol Glass at 77 K Followed by Monitoring Relative Integrated Fluorescence Intensities . . . . .167 Titration of 1,8-Naphtho-21-Crown-6 (2) with Silver Triflate in Uncracked 95% Ethanol Class at 77 K.Followed by Monitoring Relative Integrated Phosphorescence Intensity. . . . . . . . . . . . . . .168 Titration of 1,5-Naphtho-22-Crown-6 (a) with Barium Bromide and Silver Triflate in Uncracked 95% Ethanol Glass at 77 K Followed by Monitoring Relative Integrated Emission Intensities . . . . . . .169 Competition for 1.00 x 10-4 M 1,8-Naphtho-21- Crown-6 (2) in Uncracked 95% Ethanol Glass at 77 K by Sodium and Rubidium Chlorides Followed by Monitoring Arbitrary Relative Integrated In- tensity. . . . . . . . . . . . . . . . . . . . . . . .171 Competition for 1.00 x 10-4'§_l,8-Naphtho-21- Crown-6 (a) in Uncracked 95% Ethanol Class at 77 R by Barium Bromide and Rubidium Chloride Followed by Monitoring Arbitrary Relative Integrated In- tensity. . . . . . . . . . . . . . . . . . . . . . . .173 Emission Quantum Yields (¢f and ¢p) and Lifetimes (Ti and 1p) for 2,3-Naphtho-20-Crown-6 (I) and Alkali Metal Chloride, Barium Bromide, and Silver Triflate Complexes of Crownll in Alcohol Class at 77 K. . . . . . . . . . . . . . . . . . . . . . . .175 Emission Quantum Yields (¢f and ¢p) and Lifetimes (Ti and Tp) for 1,8-Naphtho—21-Crown-6 (2) and ix Table 22 23 24 25 26 27 28 Page Alkali Metal Chloride, Barium Bromide, and Silver Triflate Complexes of Crown 2,1n Alcohol Glass at 77 K. . . . . . . . . . . . . . . . . . . . . . . .176 Emission Quantum Yields (¢f and ¢p) and Lifetimes (If and IP) for 1,5-Naphtho-22-Crown-6 and Alkali Metal Chloride, Barium Bromide, and Silver Tri- flate Complexes of Crown 3,1n Alcohol Class at 77 K . . . . . . . . . . . . . . . . . . . . . . . . .177 Emission Quantum Yields (¢f and op) and Lifetimes (Ti and 1p) for 2,3-Naphtho-20-Crown-6 (l) and Ammonium and Alkylammonium Chloride Complexes of Crown‘l in Alcohol Class at 77 K . . . . . . . . . . .178 Emission Quantum Yields (¢f and op) and Lifetimes (If and Tp) for 1,8-Naphtho-21-Crown-6 (2) and Ammonium and Alkylammonium Chloride Complexes of Crown‘g in Alcohol Class at 77 R . . . . . . . . . . .179 Emission Quantum Yields (¢f and op) and Lifetimes (1f and 1p) for 1,5-Naphtho-22-Crown-6 (3) and Ammonium and ngropylammonium Complexes of Crown 3 in 95% Ethanol Glass at 77 R . . . . . . . . . . . .180 Emission Quantum Yields (¢f and ¢p) for 2,3- Naphtho-ZO-Crown-6 (1) and Haloalkylammonium Chloride Complexes of Crown 1 in Uncracked 95% Ethanol Class at 77 K. . . . . . . . . . . . . . . . .181 Emission Quantum Yields (¢f and op) for 1,8- Naphtho-Zl-Crown-G (2) and Haloalkylammonium Chloride Complexes of Crown z'in Uncracked 95% Ethanol Glass at 77 R. . . . . . . . . . . . . . . . .182 Emission Quantum Yields (¢f and op) for 1,5- Naphtho-ZZ-Crown—6 (a) and Haloalkylammonium Chloride Complexes of Crown 3,1n Uncracked 95% Ethanol Class at 77 K. . . . . . . . . . . . . . . . .183 Tables 29 3O 31 32 33 34 Page Emission Quantum Yields ( and Lifetimes (Tp,EtBr) in Ethyl Bromide- Ethanol-Methanol (1:4:1, v/v) Glass at 77 K. . . . . .255 xii Table 41 42 43 44 45 Relative Susceptibility of Rate Constants of Excited State Processes of 2,3-Naphtho-20-Crown-6 (l), 1,8-Naphtho-21-Crown-6 (Z), and 1,5—Naphtho- 22-Crown-6 (3) to Perturbation by Complexed Alkali Metal, Barium and Silver Cations in Alcohol Glass at 77 K. O O O O O O O O O O I O I O O O I O O O O 0 Relative Susceptibility of Rate Constants of Excited State Processes of 2,3—Naphtho-20—Crown-6 (l), 1,8—Naphtho-21-Crown-6 (2), and 1,5-Naphtho- 22-Crown—6 (3) to Perturbation by Complexed Alkyl and Haloalkylammonium Cations in Alcohol Class at 77 K. O O O O O O O O O O O O O O O O O O O 0 O 0 Comparison of Room Temperature UV Absorption Spectra of 2,3-Naphtho-20-Crown-6 (1) in Square Cells and in Round Dewar . . . . . . . . . . . . . . wavelengths (nm) at which Quantum Yield Comparisons Relative to Naphthalene Were Made and Number of Determinations Made for Disubstituted Naphtha- lenes O O I O O O O O I O O I O O O O O O O O O O O O Wavelengths at Which Quantum Yield Comparisons Relative to Free Crowns Were Made. . . . . . . xiii Page .272 .273 .278 .292 .293 Figure LIST OF FIGURES Page 77 K UV spectra of 2,3-disubstitutednaph- thalenes in 95% ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absorbancies for shorter wavelength regions are given by the scale on the left; for longer wavelength region, by scale on the right. Concentrations (at room temperature) were 4.00 x 10-4 M for longer wavelength region and 1.20 x 10-4 M for shorter wavelength region (except for naphthalene, 1.00 x 10-4 M). The scales have been adjusted so that the absorbancies of both sections, after multiplication if neces- sary by the factors given in the figure legend are those of a 8.00 x 10-51M_solution. Curve A, 2,3-naphtho-20-crown-6 (1); curve B, 2,3- big:(methoxymethyl)naphthalene (4); curve C, 2,3-dimethylnaphthalene (5); curve D, naphthalene (19). . . . . . . . . . . . . . . . . . 43 77 K UV spectra of l,8-disubstitutednaphthalenes in 95% ethanol-methanol (4:1, v/v) glass. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absorbancies for shorter wavelength region are given by the scale on the left; for longer wavelength region, by the scale on the right. Concentrations (at room tempera- ture) were 4.00 x 10-4 M for longer wavelength region and 8.00 x 10'.5 M for shorter wavelength xiv Figure Page region (except for naphthalene, 1.00 x 10-4 MD. The scales have been adjusted so that the ab- sorbancies of both sections, after multiplica- tion if necessary by the'factors given in the figure legend, are those of a 8.00 x 10-.5 _1_4_ solution. Curve A, 1,8-naphtho-21-crown—6 (2); curve B, 1,8fibigf(methoxymethyl)naphthalene (6); curve C, 1,8-dimethylnaphthalene (7); curve D, naphthalene ($0) . . . . . . . . . . . . . 45 77 K UV spectra of l,5-disubstitutednaphthalenes in ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explana- tion). Relative absorbancies for shorter wavelength region are given by the scale on the left; for the longer wavelength region, by the scale on the right. Concentrations (at 4 M for the longer wavelength region and 6.00 x 10".5 M room temperature) were 2.00 x 10- for the shorter wavelength region. The scales have been adjusted so that the absor- bancies of both sections are those of a 6.00 x 10-5 crown-6 (a); curve B, l,5fibi§7(methoxymethyl) naphthalene (8); curve C, 1,5-dimethylnaphthalene (9); curve D, naphthalene (10) . . . . . . . . . . 47 77 K UV spectra of 2,3-naphtho-20-crown-6 M solution. Curve A, 1,5-naphtho-22- (I) alone and with 5:1 added molar excesses of alkali metal chlorides in 95% ethanol- methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absorbancies for shorter wavelength Figure Page region are given by the scale on the left; for longer wavelength region, by scale on right. Crown concentrations (at room tempera- ture) were 4.00 x 10-4 M for longer wavelength region and 1.20 x 10-4 M for shorter wavelength region. The scales have been adjusted so that the absorbancies of both sections are those of a 1.20 x 10-4Msolution. . . . . . . . . . 49 77 K UV spectra of 1,8-naphtho-21-crown-6 (2) alone and with 5:1 added molar excesses of alkali metal chlorides in 95% ethanol- methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absorbancies for shorter wavelength region are given by the scale on the left; for longer wavelength region, by scale on right. Crown concentrations (at room temperature) were 4.00 x 10"4 M for longer wavelength region and 8.00 x 10-5 M for shorter wavelength region. The scales have been adjusted so that the absorbancies of both sections are those of a8.00xlO-5Msolution.............. 51 77 R UV spectra of 1,5-naphtho-22-crown-6 (3) alone and with 5:1 added molar excesses of alkali metal chlorides in 95% ethanol-methanol (4:1, v/v) glass. See figure legend. Inten- sities are given in terms of relative absor- bancies (see text for explanation). Relative absorbancies for shorter wavelength region are given by the scale on the left; for longer wavelength region, by scale on right. Crown concentrations were 2.00 x 10-4 M for longer Figure Page wavelength region and 6.00 x 10"5 M for shorter wavelength region. The scales have been adjusted so that the absorbancies of both 5 M solution. . . 53 77 K UV spectra of 2,3—naphtho-20-crown-6 (1) alone and with 5:1 added molar excess of sections are those of a 6.00 x 10- ammonium and alkylammonium chlorides in 95% ethanoldmethanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explana- tion). Relative absorbancies for shorter wavelength region are given by the scale on the left; for longer wavelength region, by scale on right. Crown concentrations (at room temperature) were 4.00 x 10-4 M_for the longer wavelength region and 1.20 x 10-4 M for the shorter wavelength region. The scales have been adjusted so that the absorbancies of both sections are those of a 1.20 x 10-4 M solution.. . . . . . . . . . . . . . . . . . . . . 55 77 K UV spectra of 1,8-naphtho-21-crown-6 (2) alone and with 5:1 added molar excess of ammonium and alkylammonium chlorides in 95% ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explana- tion). Relative absorbancies for shorter wavelength region are given by scale on left; for longer wavelength region, by scale on right. Crown concentrations (at room temperature) were 4.00 x 10-4 M for the longer wavelength region and 8.00 x 10-5 M for the shorter wavelength region. The scales have been xvii Figure 10 Page adjusted so that the absorbancies of both sections are those of a 8.00 x 10-5 Mgsolu- tion. . . . . . . . . . . . . . . . . . . . . . . . 57 77 R UV spectra of 1,5-naphtho-22-crown-6 (3) alone and with 100:1 molar excesses of ammonium and gfpropylammonium chlorides in ethanol—methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absorbancies for shorter wavelength region are given by the scale on the left; for longer wavelength region, by scale on right. Crown concentrations (at room temperature) were 1.00 x 10-4 region and 5.00 x 10-5‘M_for the shorter wave- M for the longer wavelength length region. The scales have been adjusted so that the absorbancies of both sections are those of a 5.00 x 10.5 M solution. - . . . - - . . 59 77 R UV spectra of 2,3-naphtho-20-crown-6 (1) alone and with 5:1 added molar excess of bromoalkylammonium chlorides in 95% ethanol- methanol (4:1, v/v) glass and with 12.5:1 added molar excess of iodoalkylammonium chlorides in ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explana- tion). Relative absorbancies for the shorter wavelength region are given by the scale on the left; for longer wavelength region, by scale on right. The concentration (at room temperature) of crown was 4.00 x 10.4 length region and 8.00 x 10'5 M for the shorter M for the longer wave- wavelength region. The scales have been xviii Figure 11 12 Page adjusted so that the absorbancies of both sections are those of 8.00 x 10'.5 M solutions. Baselines: curve F, B-iodoethylammonium chloride; curve C, y-iodopropylammonium chloride; curve H, 95% ethanol—methanol (4:1, v/v) glass. . . . . . . . . . . . . . . . . . . . . . . 61 77 K UV spectra of 1,8-naphtho-21-crown-6 (2) alone and with 5:1 added molar excess of bromoalkylammonium chlorides in 95% ethanol- methanol (4:1, v/v) glass and with 12.5:1 added molar excess of iodoalkylammonium chlorides in ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absorbancies for the shorter wavelength region are given by the scale on the left; for the longer wave- length region, by the scale on the right. The concentration (at room temperature) of 4 crown was 4.00 x 10- M for the longer wave- length region and 8.00 x 10-5 M for the shorter wavelength region. The scales have been adjusted so that the absorbancies of both sections are those of a 8.00 x 10"5 M solution. Baselines: curve F, B-iodoethylammonium chloride; curve C, y-iodopropylammonium chloride; curve H, 95% ethanol-methanol (4:1, v/v) glass . . . . . . . . . . . . . . . . . 63 77 R UV spectra of 1,5-naphtho-22-crown-6 (3) alone and with 50:1 added molar excess of bromoalkylammonium and iodoalkylammonium chlorides in ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are xix Figure 13 Page given in terms of relative absorbancies (see text for explanation). Relative absorbancies for the shorter wavelength region are given by the scale on the left; for the longer wavelength region, by the scale on the right. The concentration (at room temperature) of crown was 1.00 x 10'4 M for the longer wave- length region and 5.00 x 10-4 M for the shorter wavelength region. The scales have been adjusted so that the absorbancies of both sections are those of a 5.00 x 10.5 M solution. 3 g B-iodoethyl— ammonium chloride; curve G, 2.50 x 10'.3 M Baselines: curve F, 2.50 x 10- Y—iodopropylammonium chloride; curve H, ethanol- methanol (4:1, v/v) glass. . . . . . . . . . . . . 65 77 K UV spectra of 2,3-naphtho-20-crown-6 (1) alone and with 5:1 added molar excess of cesium chloride in 95% ethanol-methanol (4:1, v/v) glass and with 12:1 added molar excess of barium bromide in ethanol-methanol (4:1, v/v) glass. See figure legend. Inten- sities are given in terms of relative absorban- cies (see text for explanation). Relative absorbancies for the shorter wavelength region are given by the scale on the left; for the longer wavelength region, by the scale on the right. Crown concentrations (at room tempera- ture) were 4.00 x 10-4 M for the longer wave- length region and 8.00 x 10.5 M for the shorter wavelength region. The scales have been adjusted so that the absorbancies of both sections are those of a 8.00 x 10-5‘M_Solu- tion . . . . . . . . . . . . . . . . . . . . . . . 67 Figure 14 15 Page 77 K UV spectra of 1,8-naphtho-21-crown-6 (2) alone and with 5:1 added molar excess of cesium chloride in 95% ethanol-methanol (4:1, v/v) glass and with 12:1 added molar excess of barium bromide in ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absor- bancies for the shorter wavelength region are given by the scale on the left; for the longer wavelength region by the scale on the right. Crown concentrations (at room temperature) were 4.00 x 10-4 region and 8.00 x 10.5 M_for the shorter wave- M for the longer wavelength length region. The scales have been adjusted so that the absorbancies of both sections are those of a 8.00 x 10.5 M solution. . . . . . . . . 69 77 K UV spectra of 1,5-naphtho-22-crown-6 (3) alone and with 5:1 added molar excess of cesium chloride in 95% ethanol-methanol (4:1, v/v) glass and with 50:1 added molar excess of barium bromide in ethanol-methanol (4:1, v/v) glass. See figure legend. In- tensities are given in terms of relative absorbancies (see text for explanation). Relative absorbancies for the shorter wave- length region are given by the scale on the left; for the longer wavelength region by the scale on the right. Crown concentrations (at room temperature) were 4.00 x 10-4 M_for the longer wavelength region and 8.00 x 10-5 M for the shorter wavelength region. The scales have been adjusted so that the absorbancies xxi Figure 16 17 Page of both sections are those of a 8.00 x 10-5 Msolution.................... 71 77 K UV spectra of 2,3-naphtho-20-crown-6 (1) alone and with 50:1 added molar excess of silver triflate in ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative ab- sorbancies for the shorter wavelength region are given by the scale on the left; for the longer wavelength region by the scale on the right. Crown concentrations (at room tempera- ture) were 4.00 x 10-4 M for the longer wave- length region and 8.00 x 10-5 M for the shorter wavelength region. The scales have been adjusted so that the absorbancies of both sections are those of a 8.00 x 10"5 M solu- tion . . . . . . . . . . . . . . . . . . . . . . . 73 77 K UV spectra of 1,8-naphtho-21-crown-6 (2) alone and with 50:1 added molar excess of silver triflate in ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative ab- sorbancies for the shorter wavelength region are given by the scale on the left; for the longer wavelength region by the scale on the right. Crown concentrations (at room tempera- 4 ture) were 4.00 x 10- ‘M_for the longer wave- length region and 8.00 x 10-5 M for the shorter wavelength region. The scales have been adjusted so that the absorbancies of both 5 sections are those of 8.00 x 10- M solu- tion 0 O O O O O O O O 0 O O O O O O O O O O O O O 75 xxii Figure 18 19 20 21 Page 77 K UV spectra of 1,5-naphtho-22-crown-6 (3) alone and with 200:1 added molar excess of silver triflate in ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absor- bancies for the shorter wavelength region are given by the scale on the left; for the longer wavelength region by the scale on the right. Crown concentrations at room temperature were 1.00 x 10-4 M for the longer wavelength region and 5.00 x 10"5 M for the shorter wavelength region. The scales have been adjusted so that the absorbancies of both sections are those of a 5.00 x 10-5 M solution . . . . . . . . . . . . . 77 Corrected fluorescence spectra of 1.00 x 10-4 M;2,3-disubstituted naphthalenes (2,3—naphtho- 20-crown-6 (1), curve A; 2,3fi§i§7(methoxy- methyl)naphtha1ene (4), curve B; 2,3—dimethyl- naphthalene (5), curve C; and naphthalene (IQ), curve D) in uncracked 95% ethanol glass at 77 K. O O O O O O O O O O O I I O O O O O O O 0 86 Corrected phosphorescence spectra of 1.00 x -4 10 M 2,3-disubstitutednaphthalenes (2,3— naptho-ZO-crown-6 (1)), curve A; 2,3-l_)_i_s_- (methoxymethyl)naphtha1ene (4), curve B; 2,3—dimethylnaphthalene (a), curve C; and naphthalene (lg) curve D) in uncracked 95% ethanol. . . . . . . . . . . . . . . . . . . . . . 88 Corrected fluorescence spectra of 1,8-di- substitutednaphthalenes (1,8-naphtho-21-crown— 6 (2), curve A; 1,8fi2igf(methoxymethyl)naph- thalene (6), curve B; 1,8-dimethylnaphthalene xxiii Figure ' Page (Z), curve C; and naphthalene (IQ), curve D) in uncracked 95% ethanol at 77 K. Curve D should be multiplied by 3.0. . . . . . . . . . . . 89 22 Corrected phosphorescence spectra of 1,8- disubstitutednaphthalenes (1,8-naphtho-21- crown-6 (2), curve A; 1,8§§$§7(methoxymethyl)- naphthalene (6), curve B; 1,8-dimethylnaphtha- lene (1), curve C; and naphthalene (L8), curve D) in uncracked 95% ethanol glass at 77 K. . . . . 90 23 Corrected fluorescence spectra of 1,5-disubsti- tuted naphthalenes (1,5-naphtho-22-crown-6 (3), curve A; 1.572337(methoxymethyl)naphthalene (8) curve B; 1,5-dimethylnaphthalene (9), curve C; and naphthalene (19), curve D) in uncracked 95% ethanol glass at 77 K. . . . . . . . 91 24 Corrected phosphorescence spectra of 1,5- disubstitutednaphthalenes (1,5-naphtho-22- crown-6 (a), curve A; 1,572i§(methoxymethyl)- naphthalene (8), curve B; 1,5-dimethylnaphthalene (9), curve C; and naphthalene (lg), curve D) in uncracked 95% ethanol glass at 77 K . . . . . . 92 25 Corrected fluorescence spectra of 2.00 x 10-4 M 2,3-naphtho-20-crown—6 (1,) alone and with 5:1 molar excess of alkali metal chlorides added in uncracked 95% ethanol glass at 77 K . . . . . . . . . . . . . . . . . . . . . . . 93 26 Corrected phosphorescence spectra of 2.00 x 10"4 M 2,3-naphtho-20-crown-6 (1) alone and with 5:1 molar excess of added alkali metal chlorides in uncracked 95% ethanol glass at 77 K . . . . . . 94 27 Corrected fluorescence spectra of 2.00 x 10.4 M 1,8-naphtho-21-crown-6 (,2) alone and with 5:1 molar excess of added alkali metal chlorides in uncracked 95% ethanol glass at 77 K. . . . . . . . 95 xxiv Figure Page 28 Corrected phosphorescence spectra of 2.00 x 10-4 M 1,8-naphtho-21-crown-6 (2) alone and with 5:1 molar excess of added alkali metal chlorides in uncracked 95% ethanol glass at 77 K . . . . . . . . . . . . . . . . . . . . . . . 96 29 Corrected fluorescence spectra of 1.00 x 10-4 M 1,5-naphtho-22-crown-6 (3) alone and with 5:1 molar excess of added alkali metal chlorides in uncracked 95% ethanol glass at 77 K . . . . . . 97 30 Corrected phosphorescence spectra of 1.00 -4 x 10 M 1,5-naphtho-22-crown-6 (3) alone and with 5:1 molar excess of added alkali metal chlorides in uncracked 95% ethanol glass at 77 K. . . . . . . . . . . . . . . . . . . 98 31 Corrected fluorescence spectra of 2.00 x 10-4 M 2,3-naphtho-20-crown—6 (1) alone and with 5:1 molar excess of added ammonium and alkyl- ammonium chlorides in uncracked 95% ethanol glass at 77 K... . . . . . . . . . . . . . . . . . 99 32 Corrected phosphorescence spectra of 2.00 x -4 10 with 5:1 molar excess of added ammonium and M 2,3—naphtho-20-crown-6 (‘1’) alone and alkylammonium chlorides in uncracked 95% ethanol glass at 77 K. . . . . . . . . . . . . . . 100 33 Corrected fluorescence spectra of 2.00 x 10..4 M 1,8-naphtho-21-crown-6 (2‘) alone and with 5:1 molar excess of added ammonium and alkyl- ammonium chlorides in uncracked 95% ethanol glass at 77 K. . . . . . . . . . . . . . . . . . . 101 34 Corrected phosphorescence of 2.00 x 10-4 M 1,8-naphtho-2l-crown-6 (2) alone and with 5:1 molar excess of added ammonium and alkyl- ammonium chlorides in uncracked 95% ethanol 818.88 at 77 K. o a o o o o o s o a a o o o o o o o 102 XXV Figure 35 36 37 38 39 Page Corrected fluorescence spectra of 1.00 x 10-4 M 1,5-naphtho-22-crown—6 Q) alone and with 50:1 and 10:1 added molar excesses of amr monium and grpropylammonium chlorides, respectively, in uncracked 95% ethanol glass at 77 K. . . . . . . . . . . . . . . . . . . . . . 103 Corrected phosphorescence spectra of 1.00 x 10-4 and with 50:1 and 10:1 added molar excesses of ammonium and Efpropylammonium chlorides, M.l,5-naphtho-22-crown-6 (3) alone respectively, in uncracked 95% ethanol glass at 77 K. . . . . . . . . . . . . . . . . . . . . . 104 Corrected fluorescence spectra of 2,3—naphtho— 20-crown-6 (1) alone and with 5:1 added molar excess of bromoalkylammonium chlorides (2.00 x 10—4 M crown) and 20:1 added molar excess of iodoalkylammonium chlorides (1.00 x 10-4 M crown) in uncracked 95% ethanol glass at 77 K . . . . . . . . . . . . . . . . . . . . . . . 105 Corrected phosphorescence spectra of 2,3- naphtho-ZO-crown-G (1) alone and 5:1 molar excess of bromoalkylammonium chlorides (2.00 x 10" of iodoalkylammonium chlorides (1.00 x 10"4 M crown) in uncracked 95% ethanol glass at 77 K . . . . . . . . . . . . . . . . . . . . . . . 106 Corrected fluorescence spectra of 1,8—naphtho- 21-crown-6 (2) alone and with 5:1 added molar excess of bromoalkylammonium chlorides (2.00 x 10'4 a; crown) and with 20:1 added molar excess of iodoalkylammonium chlorides (1.00 M crown) and 20:1 molar excess x 10-4 M crown) in uncracked 95% ethanol glass at 77 K. I O O O O O O I O O O O O O O O O 9 107 xxvi Figure 40 41 42 43 44 45 Page Corrected phosphorescence spectra of 1,8- naphtho-Zl-crown-6 (2) alone and with 5:1 added molar excess of bromoalkylammonium chlorides (2.00 x 10-4 M crown) and with 20:1 added molar excess of iodoalkylammonium chlorides (1.00 x 10-4 M_crown) in 95% ethanol glass at 77 R. . . . . . . . . . . . . . . . . . . 108 Corrected fluorescence spectra of 1.00 x 10-4 M 1,5-naphtho-22-crown-6 (3‘) alone and with 20:1 added molar excess of bromo- and iodo- alkylammonium chlorides in uncracked 95% ethanol glass at 77 K. . . . . . . . . . . . . . . 109 Corrected phosphorescence spectra of 1.00 x 10-4 M 1,5-naphtho-22-crown-6 (3) alone and with 20:1 added molar excess of bromo- and iodoalkylammonium chlorides in uncracked 95% ethanol glass at 77 K. . . . . . . . . . . . . 110 Corrected fluorescence spectra of 2,3-naphtho- 20—crown-6 (1) alone and with 5:1 added molar “ a crown) and 10:1 added molar excess of barium bromide (1.00 x 10-4 95% ethanol glass at 77 K. . . . . . . . . . . . . 111 excess of cesium chloride (2.00 x 10- M crown) in uncracked Corrected phosphorescence spectra of 2,3- naphtho-ZO-crown-G (1) alone and with 5:1 added molar excess of cesium chloride (2.00 x 10-4 of barium bromide (1.00 x 10-4 M crown) in uncracked 95% ethanol glass at 77 R. . . . . . . . 112 M crown) and 10:1 added molar excess Corrected fluorescence spectra of 1,8-naphtho- 21-crown-6 (2) alone and with 5:1 added molar excess of cesium chloride (2.00 x 10-4 M crown) and 10:1 added molar excess of barium xxvii Figure Page bromide (1.00 x 10"4 M crown) in uncracked 95% ethanol glass at 77 K. . . . . . . . . . . . . 113 46 Corrected phosphorescence spectra of 1,8- naphtho-Zl-crown—6 (2) alone and with 5:1 added molar excess of cesium chloride (2.00 x 10". M crown) and 10:1 added molar excess of barium bromide (1.00 x 10'-4 uncracked 95% ethanol glass at 77 K. . . . . . . . 114 M crown) in 47 Corrected fluorescence spectra of 1,5—naphtho- 22-crown-6 (3) alone and with 5:1 added molar 4M crown) and 50:1 added molar excess of barium bromide (1.00 x 10-4 95% ethanol glass at 77 K. . . . . . . . . . . . . 115 48 Corrected phosphorescence spectra of 1,5- naphtho-22-crown-6 (3) alone and with 5:1 added molar excess of cesium chloride (2.00 excess of cesium chloride (2.00 x 10- M crown) in uncracked x 10.4 M crown) and 50:1 added molar excess of barium bromide (1.00 x 10—4 M crown) in uncracked 95% ethanol glass at 77 K. . . . . . . . 116 49 Corrected fluorescence spectra of 1.00 x 10.4 M 2,3—naphtho-20-crown-6 (,1) alone and with 50:1 added molar excess of silver triflate in uncracked 95% ethanol glass at 77 R. . . . . . . . 117 50 Corrected phosphorescence spectra of 1.00 x 10-4 M_2,3-naphtho-20—crown-6 (1) alone and with 50:1 added molar excess of silver triflate in uncracked 95% ethanol glass at 77 K . . . . . . 118 51 Corrected fluorescence spectra of 1.00 x 10-4 M 1,8-naphtho-21—crown-6 (2) alone and with 50:1 added molar excess of silver triflate in uncracked 95% ethanol at 77 R. . . . . . . . . . . 119 xxviii Figure 52 53 54 55 56 57 58 59 Corrected phosphorescence spectra of 1.00 x -4 10 M 1,8-naphtho-21-crown-6 (4%) alone and with 50:1 added molar excess of silver tri- flate in uncracked 95% ethanol glass at 77 K O O O O O O O O O O O O O O O O O O O 0 O Corrected fluorescence spectra of 1.00 x 10- M 1,5-naphtho-22-crown-6 ('3’) alone and with 200:1 added molar excess of silver triflate in uncracked 95% ethanol glass at 77 R. . . . Corrected phosphorescence spectra of 1.00 x 10.4 M 1,5-naphtho-22-crown-6 (’3’) alone and with 200:1 added molar excess of silver triflate in uncracked 95% ethanol glass at 77 K . . . . . . . . . . . . . . . . . . . Titration of 2.00 x 10"“ g 2,3—naphtho-20- crown-6 (1) with alkali metal chlorides in 95% ethanol at 77 K followed by monitoring integrated fluorescence intensity. . . . . Titration of 2.00 x 10-4 M 2,3-naphtho-20- crown-6 (1) with alkali metal chlorides in 95% ethanol at 77 K followed by monitoring integrated phosphorescence intensity . . . Titration of 2.00 x 10-4 M 1,8-naphtho-21- crown-6 (a) with alkali metal chlorides in 95% ethanol at 77 R followed by monitoring inte- grated fluorescence intensity. . . . . . . Titration of 2.00 x 10’“ E 1,8-naphtho—21- crown-6 (2) with alkali metal chlorides in 95% ethanol at 77 K followed by monitoring integrated phosphorescence intensity . . . Titration of 1.00 x 10" g; 1,5-naphtho-22- crown-6 (a) with alkali metal chlorides in 95% ethanol at 77 K followed by monitoring integrated fluorescence intensity. . . . . . xxix Page 120 121 122 153 154 155 156 157 Figure Page 60 Titration of 1.00 x 10'“ g 1,5-naphtho-22— crown-6 (a) with alkali metal chlorides in 95% ethanol at 77 K followed by monitoring integrated phosphorescence intensity . . . . . . . 158 61 Wavelength positions of peaks and shoulders (half-height lines) for fluorescence spectra of naphthalene and 2,3-, 1,8-, and 1,5-di- substitutednaphthalenes in 95% ethanol glass at 77 K. . . . . . . . . . . . . . . . . . . . . . 207 62 wavelength positions of peaks and shoulders (half-height lines) for UV absorption spectra of naphthalene and 2.3-, 1,8-, and 1,5- disubstitutednaphthalenes in 95% ethanol glass at 77 K. . . . . . . . . . . . . . . . . . . 210 63 Changes in estimated rate constants caused by 2,3-disubstitution of naphthalene and by alkali metal cation perturbation of the 2,3- crown (1). . . . . . . . . . . . . . . . . . . . . 222 64 Changes in estimated rate constants caused by 1,8-disubstitution of naphthalene and by alkali metal cation perturbation of the 1,8-crown (2). . . . . . . . . . . . . . . . . . . . . . . . 223 65 Changes in estimates of rate constants caused by 1,5-disubstitution of naphthalene and by alkali metal cation perturbation of the 1,5- crown (a). . . . . . . . . . . . . . . . . . . . . 224 66 Plot of log kdt versus the change in the T -S 1 1 0 caused by alkali energy separation in cm- metal, barium, and silver cation perturbation of crowns & and 2. . . . . . . . . . . . . . . . . 230 67 Plot of the change in the $1 - T1 energy separation caused by alkali metal cation per— turbation of crowns l and % versus the polarizing strength of the alkali metal cations . . . . . . . 233 Figure Page 68 Plot of the change in the T1 - so energy separation caused by alkali metal cation perturbation of crowns l and ,2 versus the polarizing strength of the alkali metal cations. . . . . . . . . . . . . . . . . . . . . . 234 69 Changes in rate constants of excited state processes of crowns l, %y and 3 caused by K+ ( a light cation) and Cs+ (a heavy cation) . . . . 239 70 Plot of log kdt versus the change in the T1 - So energy separation in cm...1 caused by ammonium, alkylammonium, and haloalkylammonium cation perturbation of crowns l and 2 . . . . . . . . . . 253 xxxi PREFACE This investigation of excited state processes makes use of a new perturbational method, so the introduction will first make some general comments on the nature of perturbational methodology. An analytical consideration of how excited state processes might be studied by perturbational methods will then be presented. Methods which have been used and their relative advantages and disadvantages are noted and compared to those of the method used in this investi— gation. The problem will then be recast in terms of the methodology presented and the general philosophy of the proposed method of analysis given. A brief background section on heavy atom effects will be presented, since some of the perturbing species used are known to cause what has come to be called "heavy atom effects". After the introduction, results will be presented in a fashion suitable to the proposed method of analysis. The discussion will analyze the results in terms of the questions and problems raised in the introduction. It should be pointed out at the onset that this investigation is very exploratory in nature. Reasonable sets of experiments in light of what was known and what was learned in the course of the investi- gation were proposed and carried out. While the results at hand undoubtedly answer some of the questions posed, they may also raise more questions than they answer. But this is the nature of explora- tory research. INTRODUCTION General Methodology One of the most general ways to study a phenomenon is to charac- terize it in terms of its behavior. This can be done not only by observing its normal behavior but also by observing how it behaves in response to a variety of stimuli. The Latins had a phrase which summarizes the philosophy of this methodology: ‘égere sequitur ease (to act follows to be). That is, a thing's behavior is a function of its nature. A problem inherent to all perturbational studies is the possibility that the perturbation applied might change the system to the point that the observed behavior is no longer characteristic of the unperturbed system but characteristic rather of a new system produced by the perturbation. The phenomenon with which this investigation is concerned is the behavior of electronically excited states produced by electronic excitation in condensed phases. Since the purpose of this investi- gation is in part to investigate the validity of a new perturbation- al method and since a general investigation of excited state pro— cesses is beyond the scope of any single study, the excited state behavior of a single compound will be investigated using this method. Naphthalene, a member of the aromatic hydrocarbon series, is the model compound selected for the purposes of this investigation. Reasons for its selection will be given later. It might be hoped that what is learned from this investigation based on naphthalene, provided that the perturbational method used can be justified, might be true for aromatic hydrocarbons in general. The generality of conclu- sions reached. however, will not be established by this investi- gation. Excited State Processes1 Before considering the variety of ways which might be used to study excited state behavior by perturbational methods, a discussion of what excited states normally do will be given. What excited states do is to dissipate excess energy. What is of interest is how they dissipate energy. The ways in which excited states dissipate energy fall into two classes: radiative (energy dissipated in the form of light) and nonradiative. Nonradiative energy dissipation may occur by energy transfer in the form of heat to the physical surroundings, by energy transfer to another molecule, or‘gig chemical reaction. This investigation is limited to a consideration of photophysical processes of photoexcited molecules in condensed phases, 3:3;3 electronic excited state in the absence of chemical reactions but under the possible physical perturbation of a solvent. The systems studied will be dilute (10-4 g range) solutions in glassy media at 77 K, so nonradiative energy transfer will most likely involve energy transfer to the physical surroundings. A Jablonski diagram depicting the photophysical processes of a typical aromatic hydrocarbon in condensed medium is given in Scheme 1.2 Radiative processes are indicated by straight lines, nonradiative by wavy lines. The main processes of interest are: fluorescence (radiative decay of lowest electronically excited singlet state, 81), Scheme 1. sn 5: C .2 .- Iss ‘3‘ 3 U lmernal conversion (is) kisc intersystern crossing TX intersystem crossing g . '3 0 Rd: 3 5 '- 8 E a O "p 2 CA > u l- 0 q- .5 Tu phosphorescence (radiative decay of the lowest electronically excited triplet state, T1), intersystem crossing (radiationless decay of S1 involving crossing to the triplet manifold, radiationless decay of 81 to So, radiationless decay involving intersystem crossing from T1 to S Each of these are unimolecular processes. The rate constants 0. of each of these processes are labeled as follows: kf (rate constant for fluorescence), kp (rate constant for phosphorescence), k c (rate is constant for intersystem crossing from 81 to T1), kds (rate constant for radiationless decay from S1 to So).kdt (rate constant for radia- tionless decay from T1 to 80). Note that when a process is indicated to occur from one state to another nothing more is intended than an indication of the initial and final states; no implication about how this process occurs is intended. In condensed media, rates of radia- tionless decay of higher excited singlets to S1 or higher excited triplets to T1 (referred to as "internal conversion" in both cases) are much faster than either radiative or radiationless decay of either 81 or T1, so fluorescence and phosphorescence are normally from S1 and T1, respectively.3 The rates of radiative and radiationless decays from S1 and T1 can be calculated from the following experimentally determined quanti- ties: ¢f (quantum yield of fluorescence), ¢ (quantum yield of phos- P phorescence), Tf (lifetime of fluorescence decay), Tp (lifetime of phosphorescence decay), and ¢isc (quantum yield of intersystem cross- ing. These quantities are defined by the rates in Scheme 1 as given 'by the following equations: ¢ I: 3kT f f f kf + kds + kisc ¢isc g kisc Tf =¢ __1:p_._ P i3Ck+k p dt ¢ ¢isc kp Tp ¢i sc is readily determined in solution, but in glassy media much more complex experiments are required.5 For many aromatic hydrocarbons, including naphthalene, however, it has been demonstrated (in solution and in glassy media) by the direct determination of the quantum yield of triplet formation, that ¢isc - l — ¢f-5 This means that all decay of S1 is either through fluorescence or through intersystem crossing. Alternatively, it means that kds is negligibly small compared to (Rf + kisc)' Thus, the above set of equations reduce to four equa- tions in four unknowns. Solutions for the rate constants in terms of measurable quantities are: 1 k --- -k dt 1p 9 Determination of these rate constants for naphthalene and for naph- thalene perturbed in a variety of ways will be one way of investigating how excited state processes are affected as a function of the way in which naphthalene is perturbed. Perturbational Methods An analytical consideration of how naphthalene excited states might be studied by perturbational methods will now be given. Ideally a perturbational study would involve taking a single electronically ex- cited naphthalene molecule, bringing a variety of different type perturbers up to it, holding them at a variety of distances and orientations, and characterizing changes in the rates of its excited state processes in terms of these parameters. Several possible orien- tations are indicated below: 00s 3‘ Such a study would reveal the following kinds of information: direction of approach of the perturber which is most effective; the minimum distance between naphthalene and perturber for perturbation to occur; and which perturbers are most effective at perturbing a given process or group of processes. In the absence of molecular forceps this method is obviously not capable of experimental realization. What are needed, then, are methods which will induce the perturber and perturbee to spend time near each other. Various methods have been used to do this and their relative advantages and disadvantages as perturbational methods will be given below. Finally, the method used in this in- vestigation will be discussed in terms of its relative advantages and disadvantages as a perturbational method. One of the surest ways of inducing two species to spend time together in a relatively fixed fashion is to hold them together via a covalent bond. Thus, one might investigate changes in naphthalene's excited state processes as a function of position, number and nature of substituents. This has been done for naphthalene using deuterium6 and halogen7 substituents in place of protium. The advantage of using deuterium as a perturber is that substitution of deuterium for pro- tium is the smallest possible chemical perturbation that can be made. One advantage of using the halogens as perturbers is that they form a larger series, though different in nature, than the limited isotopes of hydrogen. The disadvantage is that substitution of halogen for hydrogen is a large chemical perturbation. It is well known that the interaction of unshared electron pairs on halogen with adjacent n -systems has been used to explain the greater strength of aryl- halogen and vinyl-halogen bonds as compared to alkyl halogen bonds. Thus, results of such studies have to be treated with the caution that changes in the rates of excited state processes for different perturbers might not be comparable. The changes, rather than reflect- ing a change in some fundamental periodic property of the halogens, might instead reflect changes due to chemically altering the sub- stituted nucleus to such an extent that changes observed for different substitutents are no longer comparable. The alteration due to dif— ferent substituents might be due to changes in fundamental periodic properties. The problem is separation of effects, which might be dif- ficult for covalently bonded perturbers. Both deuterium and halogen substitution have provided useful probes of excited state behavior. The smaller amplitude of a vibrational wavefunction for a deuterium-carbon bond compared to that of a protium— carbon bond of the same energy has proven to be useful in the under- standing of radiationless decay processes.8 The halogens have pro- vided a useful series for investigation of the effects of variation of the atomic number of the perturber on excited state processes. Their effects as perturbers has led to much of what is known about so called "heavy atom effects" (gigg infra). McClure, for example, found that the triplet lifetimes of a- and of B-halonaphthalenes correlated reasonably well with a single atomic parameter of the halogens, the spin-orbital coupling parameter.9 Another method of inducing two things to spend time near each other is to simply mix them together. Barring any attractive inter- actions, statistically they will spend some time in each other's Vicinity. The chances of the two species spending time in each other's Vicinity can be increased by increasing the concentrations of one or both species. Results of studies done on compounds in the gas phase and in a condensed phase are often different. In the gas phase, fluorescence from higher electronic states than $1 and from vibrational levels of electronically excited states has been observed.10 As was previously noted, however, fluorescence is usually from S when the 1 compound is in the condensed phase. Thus, comparison of excited state behavior in the gas phase and in a condensed phase is one perturba- tional method of the intermolecular variety.11 One can also investi- gate the perturbing effects of one condensed phase relative to another.12 The advantages of this perturbation method are that 1) it avoids the possible pitfalls of making use of covalent bonding to ensure associa- tion, and 2) it allows investigation of perturbations due to species which can't be covalently bonded to a hydrocarbon framework. For this type of study, it is desirable that there be no strong attractive forces between the perturber and the perturbee, 3:24, that their assoc- iation be governed by statistics. Strong attractions between the per- turbee and perturber will change both of them and invalidate the per- turbational method. This is probably not a problem for cases which require high concentrations of perturber. A disadvantage of this method is that one has no idea of the spatial relationships between perturber and perturbee which give rise to the observed effects. In the absence of attractive interactions there should be a random dis- tribution of perturber around the perturbed species, thus giving rise to an infinite number of interactions. Also, for a given direction of approach, the distance between the perturber and perturbee will vary in a statistical fashion. Since one would not expect all these 10 interactions to give the same effect, the observed results will be a mixture of effects unless there is only one effective perturbational interaction. Analysis of a mixture of effects is less informative than analysis of isolated effects. One of the most common types of condensed phases used to study ex— cited state behavior has been alkyl halides, either neat or mixed with some other solvent. Solutions in neat alkyl halides, however, fre- quently give cracked glasses,13 which makes accurate quantum yield determinations impossible, and mixed first order triplet decay is 13’14 Thus, actual rate constant calculations invariably observed. can't be done, since the observed quantities are composites. Never- theless, trends in ¢f, ¢p’ and Tp have been noted for these studies and interpreted as heavy atom effects (see introduction section on heavy atom effects below). Another problem is that alkyl iodides absorb strongly at wavelengths shorter than 400 nm (see results section for ethyl bromide). Its use for studying a chromophore like naphthalene is impossible except in a very rough way, since naphthalene fluoresces in the 315 to 400 nm region.15 Condensed phases of rare gases at 4.2 OK have also been used as perturbing media. Condensed phases of this sort are probably less likely to interact chemically with a dis- solved species than are any others, which makes them ideal perturbers. Johnson has shown that the triplet lifetimes and phosphorescence spectra of benzene in an argon matrix differ according to the symmetry of the hole benzene is in and he was able to identify the symmetry 17 <>f one of the holes responsible for a given lifetime and spectrum. A variety of other types of external perturbers have been used: alkali ll 18 cesium chloride (perturba— 20,21 halide salts (perturbation due to halide), tion due to cesium cation),19 transition metals, and other para- magnetic species such as oxygen.22 The kinds of perturbers noted above have a wide range of properties and each kind of perturber provides limited information about the nature of excited state pro- cesses. The wide variety of perturbers which can be used is one of the advantages of the intermolecular perturbational method. The main disadvantage of this method is the random distribution of perturber, which gives rise to many different interactions. Johnson's work for benzene (vide infra) in rare gas hosts was able to partially overcome this difficulty by being able to control the symmetry of the hole. Another method which can be used to get two species to spend time together makes use of species which form complexes. Provided that the site of complexation is known, this method has the advantage of being less random than the intermolecular method. These studies depend upon attractive polar interactions involving atoms which are part of the chromophore as the basis for complexation. While such studies may be instructive in their own right, it does not seem to be desirable to base a perturbational method upon this type of inter- action. Some examples of this kind of study follow. Yuster and Weiss- man23 determined relative fluorescence and phosphorescence intensities and phosphorescence lifetimes for dibenzoylmethane complexes of tri- valent ions of aluminum, scandium, ytrrium, lanthanum, gadolinium, and lutetium as well as phosphorescence lifetimes for complexes of alkali and alkaline earth metals. It was found that Tp decreased with increasing atomic number for complexes of ions in the same column 12 of the periodic table. Fluorescence to phosphorescence intensities decreased as Tp decreased. Effects were observed to depend not only on atomic number but also on electronic configuration. The largest effects were observed for paramagnetic gadolinium. Marzzaccoza determined 77 °K.absorption, phosphorescence, and phosphorescence excitation spectra of pyrazine in ethanol with various amounts of lithium, sodium, potassium, and zinc (II) halide salts. Effects were shown to be anion independent, and shown to be a composite of uncomr plexed pyrazine, and 1:1 and 2:1 metal:pyrazine complexes (complexa- tion is between metal and nitrogen). Phosphorescence quantum yields were about the same for the 1:1 sodium and potassium complexes (0.9 1.2) and smaller for lithium, zinc, and water 1:1 complexes. Phos— phorescence quantum yields were about the same for sodium, potassium, and water 2:1 complexes (0.61.3) and smaller for lithium and zinc com- plexes (<0.15 and <0.l, respectively). This study did not attempt any interpretation of the results in terms of the properties of the per- turbers. Song25 observed an increase in the fluorescence quantum yield for a complex between retinal and sodium cation relative to the quantum yield for uncomplexed retinal (a similar effect was not ob- served for retinol). The increase in fluorescence quantum yield was interpreted as an anomalous heavy atom effect. Luminescence properties of complexes of metal ions with porphyrins26 and of bipyridyl ruthenium complexes have been extensively investigated.27 Another method which has been used to induce two species to spend time in the same vicinity is to bond the perturber to a part of the molecule which is not part of the chromophore yet holds the perturber 13 29 made use of the rigidity in the vicinity of the chromophore. Turro of the norbornene system fused to the 2,3- positions of naphthalene in an attempt to systematically study the interaction between bromine. substituents on the norbornene framework and the naphthalene chromo- phore. No systematic relationship between the position of bromine and changes in excited state rate constants could be discerned (both mono- and dibromonaphthonorbornenes were investigated in ether-iso pentane—ethanol (5:5:2) glass at 77 °K). The puzzling results of this study may be a function of an inadequate understanding of the impor- tance of throughbond interactions, even though the bromine is insulated by several carbon-carbon bonds from the chromophore. Proposed Perturbational Method This investigation has also relied upon complexation to induce two species to spend time in the same vicinity, but it has done so in a different fashion. The perturber itself is not directly bonded to the chromophore of interest but is held near to it by means of a complexing agent which has been attached. To do this, advantage was taken of the complexing abilities of crown ethers (cyclic poly- 29’30 The following crown ether-equipped naphthalene derivatives ethers). were synthesized: 2,3-naphtho-20-crown-6 (l), 1,8-naphtho-21-crown-6 (g), and 1,5-naphtho-22-crown—6 (a). 14 The versatile complexing abilities of crown ethers provide for the study of a wide range of perturbers. Crown ethers are known to complex 29a,b alkaline earth,293’b am- 32 a variety of cations: alkali metal, 31 31 33 monium, primary alkylammonium, diazonium, silver (I), lead (II),33 mercury (II),33 thallium (I),33 and zinc (II)34 and lanth- anides.35 In nonpolar solvents hydrogen bonding compounds such as alcohols36 37 and carboxylic acids are complexed by crown ethers. Crown ethers are reasonably inert chemically. Also, they do not absorb in the near ultraviolet region, so their absorption will not overlap absorption by the chromophore of interest. This proposed perturbational method has some of the advantages and some of the disadvantages inherent to the intramolecular, inter- molecular, and complexation perturbational methods noted above. Like the intramolecular and complexation methods, it has the advantage that complexed perturbers are held in reasonably well-defined positions 15 relative to the naphthalene nucleus. The position of a complexed perturber is less well-defined, however, than that of a covalently bonded perturber or a perturber which complexes with a given site in a chromophore. This is because the position of a complexed perturber can't be deduced from the structure of the free crown. Also, different perturbers may complex differently. Thus, the distance and position of different perturbers from the naphthalene nucleus may differ. While the positions of complexed perturbers will not be known precisely, at least the direction of approach of a perturber to the naphthalene nucleus is defined by the positions at which the naphthalene nucleus is incorporated into the cycle. Like the intramolecular perturbation method, this method perturbs naphthalene by making use of covalent bonding. On the other hand, crown z’affords a direction of approach to the naphthalene chromophore not practically possible gig_covalent bonding, i;g;, it allows positioning of a perturber in the vicinity of the n—face of the chromophore. In this case, however, it is the complexing agent, not the perturber, which is covalently bonded. The attachment of the complexing agent itself, of course, is a per- turbation. The attachment of the complexing agent, however, was done in such a way as to minimize the perturbation. The purpose of the methylene units inserted between the naphthalene nucleus and the oxygen atoms is to electrically insulate the naphthalene fl-system from inter- action with unshared electron pairs on oxygen and to minimize differ- ences in inductive effects felt by the naphthalene nucleus from complexed and from uncomplexed oxygen. The problem of perturbing naphthalene by attachment of the complexing agent will be considered 16 in greater detail below. Like the intermolecular and complexation perturbational methods noted previously, the proposed method has the advantage of making it possible to investigate the perturbing effects of species which can't be covalently bonded to a hydrocarbon framework. Unlike intermolecular perturbation of naphthalene, however, two of our models, crowns l and 2, have groups which block the approach of the perturber to the side of the chromophore. Thus, for these two crowns, our results will perhaps be more characteristic of his:(methoxymethyl)-naphthalenes than of naphthalene itself. For crown 3, approach to the w-face is not blocked by the methylene units, since they lie in the same plane as the naphthalene nucleus. Depending upon how different ions complex, however, the distance between the n-face and the complexed ion may vary. There are some practical advantages to the proposed perturbational approach, which are meaningful if the method turns out to be valid. One advantage is that it is easier to make complexes than compounds. Some survey studies have been hampered by the large amount of work involved in preparation and purification of a large group of related derivatives.6 While synthesis and purification of chromophores equipped with complexing agents may not be trivial, once they have been made and purified, one need only use pure salts and solvents to prepare pure complexes. A wide variety of salts of high purity are com- mercially available, and impurities in salts are less likely to be troublesome than organic impurities, which may emit or quench excited states. Also, provided that stability constants are reasonably large, 17 lower concentrations than those required for the intermolecular method can be used. This is especially important for ionic perturbers, since many salts have very low solubilities in organic solvents. Alternatively, crown ethers may increase the solubility of attached hydrocarbons in aqueous media. Also, perturbers which absorb light in the region that absorption or emission occurs may interfere when used at high concentrations but may not interfere at low concentrations. Thus, this investigation was able to determine the intermolecular effect of alkyl iodides zig_iodoalkylammonium chlorides on excited state behavior using concentrations at which absorption by the iodides did not interfere (vide infra). The perturbation due to disubstituting naphthalene with crown- methylene groups is potentially serious. A possible consequence of this perturbation would be that rates of excited state processes for crown complexes could not be compared to those of naphthalene. In this case, results for complexes of crowns l, 2, and 3 could not be compared, although results for different complexes of the same crown might validly be compared. The problem will in part be addressed by compar- ing rates of excited state processes of 2,3-, 1,8-, and 1,5- dimethyl-, l2i§f(methoxymethyl)-, and crown-methylene disubstituted naphthalenes to those of naphthalene and to those of crown complexes. The compounds indicated in the above series are 2,3-dimethylnaphthalene (4), 2,3- 233:(methoxymethyl)naphthalene,(a), crown l, l,8—dimethylnaphthalene (Q), 1,81bi§7(methoxymethyl) naphthalene (Z), crown a, 1,5-dimethyl— naphthalene (Q), 1,572igf(methoxymethyl)naphthalene (a), crown 3, and naphthalene (IQ). CH3\ CH3 CH CH3 ©© CH3 3 ©© CH3 ©© CH3 The problem will also be addressed by comparing 77 K UV absorption and emission spectra. The following sets of spectral comparisons will be made: naphthalene compared to disubstituted naphthalenes, free crown compared to complexes of same crown, and complexes of one crown com- pared to complexes of another crown. The comparisons will be quali- tative in nature. The extent of the comparisons will be to note how similar or dissimilar spectral shapes are. The assumption underlying such comparisons will be that if spectral shapes are the same, so also are the states giving rise to them. When spectral shapes are 19 different, there is less that can be said here, since this writer is not a theoretical spectroscopist. Full analysis of spectral shapes that are different might reveal how different the states involved are and the nature of the perturbation responsible for their difference. The difficulty of analysis in such cases may also be a function of the chromophore chosen for this model study. While naphthalene is a good candidate for study in many respects (vide infra). interpretation of results may be complicated because of the nature of its absorption (S0 + $1) and fluorescence (S1 + So), since the orbitally allowed com- ponent of these transitions is weak and much of the observed intensity is induced by vibronic coupling with higher electronic states.38 Perturbation due to introduction of a substituent may enhance mixing of electronic states resulting in enhanced transition intensities. Naphthalenes phosphorescence, however, is orbitally allowed (though, 9 of course, spin forbidden).3 Perturbation due to introduction of a substituent may only introduce symmetry restrictions which affect the allowedness of various transitions.40 There are several reasons which make naphthalene a good candidate for study. Naphthalene's excited state behavior has been extensively studied, which makes comparison of the proposed perturbational method to previously used perturbational methods much easier than if a little studied system had been chosen. Also, it provides for comparison of results obtained by the proposed method to results previously obtained by other methods. Naphthalene has acceptable photophysical properties, i.e., it both fluoresces and phosphoresces in easily accessible spec- tral regions. Naphthalene has less symmetry than, say, benzene, and so 20 provides a better model for probing for directional dependence of external perturbation. Naphthalene derivatives are readily available, thus simplifying the synthetic task, and there was a reasonable chance that the crown derivatives would be crystalline solids. Crystalline solids can be effectively purified by recrystallization, whereas oils are difficult to purify. Provided that the problems alluded to are not overpowering, crowns like 1, 2, and 3 offer a new method for the study of excited state processes by nonbonded perturbers. At worst, the excited state behavior of a large variety of complexes of a new variety will have been in- vestigated. Proposed Experiments Previous to this investigation, there were no reports in the litera- ture of perturbation of excited states of aromatic hydrocarbons by alkali, alkaline earth, or by ammonium cations, save one study of the 19 quenching of nine aromatic hydrocarbons by cesium cation. In fact, previous studies using alkali metal halides found that effects were due to the halide, not to the cation.41 Thus, the alkali metal chlorides should be an interesting series to investigate. Since the cation perturbation studies are exploratory, it is not desirable to prejudice the results by saying that the purpose is to investigate this or that theory or effect. There was very little way to know what to expect. 'Thus, although the alkali metal cations are isoelectronic with the and lifetime (Tf,TD) Ineasurements. Investigations of HAE on these rate-constants based 28 on both quantum yields and lifetimes are few. Conclusions are often drawn on the basis of increases in ¢pl¢f or solely on the basis of p' f 50 and, since 51 kds is assumed to be negligible for aromatic hydrocarbons, an T Since k is assumed not to be affected by the HAE, increase in ¢pl¢f is interpreted to be due to an increase in kisc' Since both radiationless and radiative decay from T1 are spin for- bidden, the operation of a HAE can be concluded solely on the basis of a decrease in Tp. The relative susceptibilities of kp and kdt to the HAE are not known then; however, their relative susceptibili- ties have been determined in some cases without recourse to quantum yield determinations by determining relative HAE on protiated and deuterated analogs and making use of the supposition that kdt is small for the deuterated analog.52 Other studies have determined their relative susceptibilities by making use of electron spin resonance to directly monitor triplet state concentrations.53 The relative susceptibility of kp and k to the HAE is of interest dt because phosphorescence involves coupling of the electronic field with the photon field while radiationless decay involves coupling of the electronic field with the phonon field. These processes are inherently different and might be expected to be affected differ- ently.54 There seems to be general agreement that kp 13 generally 55 dt' than demonstrated, however. The susceptibility of k18c relative more susceptible to the HAE than k This is often assumed rather to kp and kdt is not as well known, since its unambiguous determina- tion requires determination of ¢f, ¢isc’ and Tf. As noted above, tnvo simplifying assumptions are frequently made, and an increase in 29 k18c is presumed based solely upon a decrease in ¢f or increase in ¢p/¢f. These observed changes in quantum yields could be accounted for by a decrease in kf. Relatively little is known about how kf is affected, since techniques for determining lifetimes in the nano- second range are relatively recent. Some authors have proposed "in- 28b verse HAE" to account for apparent decreases in kisc' But these evaluations of kisc were based on less than complete data and a de- crease in kisc could also be accounted for in terms of an increase in the energy separation between the singlet and triplet states involved. It has been shown, for instance, that for some aromatic hydrocarbons 02123 perylene) ¢f is decreased by neither internal nor external HAE.56 It was shown that if kisc < 102 kf, there is no increase in kisc due to presence of a heavy atom. It was concluded that if the energy gap between the singlet and triplet states is larger than 10 kK, that there will be no quenching of fluorescence by either internal or external heavy atoms. For naphthalene, T2 is close in energy to S1 and it is thought that intersystem crossing is Xifl T2, not directly to T1.57 Therefore, I: c for naphthalene may be is sensitive to changes in energy levels due to substitution and con- sequent changes in salvation. This investigation may add to what is known about the relative susceptibilities Of kf, k , kp’ and kdt to external perturbers isc since both quantum yields and lifetimes have been determined with time exception Of ¢isc (see introduction section on excited state processes). It is possible, however, that there is no general order of relative susceptibilities to external perturbers. The 30 relative susceptibilities may depend upon both the chromophore and the perturber. I The interpretation of enhanced transition probabilities between states of different multiplicity in the presence of a heavy atom is universally accepted as being due to enhanced SO coupling. Thus "HAE" has come to be used as a synonym for "S0 coupling". There has, . however, been considerable controversy over the mechanism of SO coupling for inter— and intramolecular HA perturbation. One issue is the identity of the perturbing singlet state, iLEL’ whether it is one or more singlet states of the chromophore, whether it is an ex- 13 or whether it is a singlet cited singlet state of the heavy atom, charge transfer state of a molecule-perturber complex.68 On the basis of complete depolarization of phosphorescence in heavy atom containing glass (vide infra), Kearnsl4 concluded that the perturb- ing singlet state could not be one of the chromophore and, therefore, must be a singlet state of the heavy atom.S7 Another issue is whether the interaction is best described theo- retically in charge transfer (CT) terms or in exchange terms. Mc- Glynn58 points out that it is difficult to make a choice between these descriptions on theoretical grounds, but prefers a CT descrip- tion, based on what are thought to be various, experimental indica- tions of CT complexes. There is disagreement, however, based on the :insensitivity of T +* S0 frequencies to solvent effects.59 This 1 .firiding is countered by McGlynn on the basis of his own results. McGlynn had found that the 0-0 phosphorescence band of naphthalene in propyl halide glasses was red shifted relative to the frequency of the 31 0-0 band in EPA (ether-isopentane-alcohol (5:5:2, v/v) by the followb ing amounts: 55 cm.-1 (propyl chloride), 155 cm-1 (propyl bromide), 13 noted that the difference and 325 cm.-1 (prOpyl iodide). McGlynn in cm.-1 in going from propyl bromide to propyl iodide was larger than difference in going from propyl chloride to propyl bromide. He also noted that the decrease in Tp (or increase in S0 +-T ab- sorptivity) was larger in going from propyl chloride to propyl bromide than in going from propyl bromide to propyl iodide. Mc- Glynn concludes that the spin—orbital coupling interaction between naphthalene and the propyl halide medium cannot be primarily respon- sible for the observed red shifts, since, if it was, one would expect to see a correspondence between the cases where the largest increase in red shift is produced and where there is the largest increase in $0 coupling (as evidenced by largest decrease in Tp or increase in S0 '* T absorptivity). Since 80 coupling can't be primarily responsible for the red shifts, something else must be. This something else could be any perturbing singlet state higher in energy than the triplet state. McGlynn notes that, while it could be a CT singlet state, the results warrant only that it is some singlet state higher in energy than the triplet. The observed red shifts are used to favor CT interactions as the dominant effect because of the presumed insensitivity of the T1 +rso frequency to solvent effects by an exchange mechanism. While McGlynn is of the opinion that CT complexes are required 14 counters with evidence (vide infra) for an external HAE, Kearns that CT complexes aren't essential at all. The results of the present investigation may have some bearing on the importance of CT 32 complexes to the external HAE. The issue concerning the external HAE on which this investiga- tion will have the largest bearing is the directional dependence of the effect. That there is such a dependence and that an aromatic hydrocarbon is most susceptible to external perturbation in the vicinity of the n-face is suggested but not experimentally proven by previous investigations. By monitoring the optical density for $0 + T absorption and the refractive index of mixtures of a-chloro- 60 was able to show evidence naphthalene and ethyl iodide, McGlynn for at least weak complexes (stabilization energy of about 0.6 kcal, equilibrium constant approximately 0). The stoichiometry of the complex was 1:1, though the existence of other stoichiometries were evident but much less preponderant. McGlynn13 found, however, that phosphorescence decay curves for solutes in alkyl halide glasses could only be reproduced analytically by a sum of several first order single exponential decays. This fact was interpreted as an indication that the conformation of the complex could vary widely about some most probable configuration. Evidence for a 1:1 complex was used to favor a CT mechanism rather than exchange, since, it was contended, an exchange mechanism would not require such specific 58 It does not seem, however, that McGlynn's interpre- interaction. tations of either the room temperature data or of the multiexponen— tial decay character in a glass are interpretations required by the data. The seeming preponderance of a 1:1 complex at room tem— perature could also be accounted for by many different interaction geometries which all give rise to about the same effect. A heavy 33 atom anywhere in the vicinity of the neface might have a large effect. Furthermore, interpretation of the multiexponential decay character in terms of conformational variation around some most probable con- figuration is not consistent with Kearns'la finding that there is complete depolarization of the phosphorescence of aromatic hydro- carbons (including naphthalene) in ethyl iodide-EPA glasses at 77 K, since, according to Kearns, complete depolarization requires a completely random distribution of perturbers. Also, while Kearns14 observed multiexponential phosphorescence decay character from solutes in heavy atom glasses, he observed single exponential decay for naphthalene in crystals of polyhalobenzenes. Furthermore, Kearns14 observed that naphthalene was more perturbed in crystals of pr dibromo- and'gymrtetrabomobenzene (T1) 3 33 msec), compounds which don't form strong CT complexes with naphthalene, than it was in a strong CT complex with tetrabromophthalic anhydride (Tp z 390 msec). While the shorter naphthalene lifetime in the crystals is probably in part due to the higher number of bromines around the naphthalene, the comparison is sufficiently accurate to show that the presence of lowblying CT states in the naphthalene-tetrabromophthalic anhydride complex has not enhanced the magnitude of the external HAE relative to external heavy atom perturbers which contain the same heavy atoms but which do not form CT complexes. Other arguments by McGlynn in favor of the mediation of CT states in the external HAE are based on the similarities in (bf/d)p arui Tp behavior for CT complexes to those produced by external heavy atoms.61 For the gfl-trinitrobenzene complex of anthracene, for 34 example, Tp decreases and ¢p increases relative to uncomplexed an- thracene. Examples of a wide variety of CT complexes for which Tp decreases by an order of magnitude are cited. It is also noted that ¢pl¢f increases for some of these CT complexes which don't contain any heavy atoms. Enhancement of So + T absorptivity due to oxygen is also noted and mediation of CT complexes invoked. While it may very well be true that CT states can enhance SO coupling in a molecule, it cannot be concluded from this that the external HAE also involves CT states. HAEs and effects produced by CT com- plexation may be similar because they both promote S0 coupling, and, therefore, lead to enhanced transition probabilities for spin-for- bidden processes. However, the basis for SO coupling induced by a heavy atom is the high nuclear charge of the heavy atom. While the effect of this high nuclear charge must in some way be mediated to the perturbed species, it is not at all clear that an interaction as strong as a CT interaction need be invoked to accomplish the mediation. While the external HAE certainly involves some distance dependence,62 it is not known experimentally what this dependence is. But n orbitals extend further than the van der Waals radius of a molecule (theoretically, indefinitely), so an exchange mechanism is reasonable. It cannot be assumed that because heavy atoms pro- duce effects similar to those produced by other kinds of perturba- tions that, therefore, the mechanism of the interaction is the same in all cases. Oxygen is paramagnetic, and formation of a CT complex has to be admitted to be a rather gross perturbation of a cflrromophore. A whole new set of states is produced in a CT complex. 35 These are evidenced by the appearance of broad characterless absorp- 63 Thus, this investigator finds it dif- tion and emission bonds. ficult to accept McGlynn's arguments for mediation of the external HAE by CT states based on the similarity of effects observed for heavy atoms, paramagnetic species, and CT complexes. Kearnslé attempted an investigation of the distance dependence of the external HAE by making use of known crystal structure data. He was not able to find any correlation between the distance of the heavy atom to an arbitrary reference point and the effectiveness of the crystal as a perturber. It was concluded that relative orienta- tions must be important for the external HAE. Berlman's42 work also suggests that relative orientations are important for the external HAE. From measurements of ¢f in benzene and in bromobenzene, it was found that phenyl substituents on naphthalene and anthracene reduced the efficiency of quenching by bromobenzene relative to the unsubstituted system. The reduced efficiency was explained in terms of reduced perturbation of energy transfer to bromobenzene (due to red shift induced by phenyl groups) and in terms of steric hindrance of approach to the external heavy atom due to steric hindrance by the noncoplanar phenyl groups. Also, it was noted that planar aromatic hydrocarbons which are susceptible to heavy atom quenching are also susceptible to concentration quench- ing (anthracene) or excimer formation (naphthalene). The quenching ‘was interpreted as collisional in nature and the steric hindrance as evidence that the collisional quenching was short range in nature and involved a close encounter between the fl-system and the quencher. ‘The close encounter was viewed as involving overlap of a p orbital 36 of the halogen with a H orbital of the chromophore. While the inter- pretation of reduced quenching efficiency as partially due to steric hindrance seems to be warranted, the more detailed analysis of the data seems not to be warranted. As McGlynn13 has pointed out, the external HAE does not require collisions, since his work has demonstrated that the external HAE at room temperature (evidence from increased SO + T absorptivity) is the same as the external HAE in heavy atom glasses at 77 K (evidence from 1p) (EASE ingga). Furthermore, no evidence beyond the steric inhibition of quenching was advanced to warrant the analysis of the close encounter thought to be necessary. The indications of steric inhibition do seem to indicate that aromatic hydrocarbons are most susceptible to external HAE in the vicinity of the n-face. The strength of the evidence alone, however, would not preclude the possibility that aromatic hydrocarbons are most susceptible to the external HAE in the nodal plane. Investigation of the external HAE of ethyl bromide on naph- thalene and crowns l, 2, and 3 will provide a means of systematically blocking approach of an external heavy atom to the naphthalene chromo- phore from a variety of directions. Internal and external HAE have been compared in a variety of ways. Based on the following experiments, it was concluded by McGlynn13 that they are similar, though judgment on the generality of the con- clusion was reserved. Tp and phosphorescence spectra were determined .for each of the Ohhalonaphthalenes (-F, -Cl, -Br, -I) in each of the :following solvents: EPA, propyl chloride, propyl bromide, and propyl iodide. The triplet decays were all non-single exponential, so 37 "first half-lifes" were reported. It was found that (51 + 5E)2rp was roughly constant within a range of 102. E1 is the SO coupling parameter for the internal halogen and 5E is the SO coupling parameter for the external halogen. It was noted, however, that there were regular variations in (512 + EEZ)Tp' (SI + EE)2Tp increases as 5E increases (51 constant) and (£1 + gE)21p decreases as SI increases (5E constant). It was suggested that these variations might be accounted for in terms of l) a larger effect of 5E on kp than kdt (would explain increase in (£1 + EE)ZTp as 5E increases) and/or 2) lack of constancy of overlap factors in the integral (¢0THSO¢OSP), where ¢0T is the zeroth order triplet wavefunction, H50 is the SO coupling Hamiltonian, and ¢08p is the zeroth order perturbing singlet wavefunction. An alternative explanation is that k and/or kp dt are less susceptible to 5E than £1. Increases in So + T absorptivity for the o-halonaphthalenes at room temperature due to propyl iodide perturbation were also de- termined.13 A remarkable parallel was found between the increases in S0 + T1 absorptivity and the decreases in Tp due to propyl iodide perturbation. In both cases, as 51 increases, the ratio of the external to the internal effect decreases. Based on the results of alkyl halide perturbation of the S + T 13 0 absorptivity of naphthalene, McGlynn also suggested that external SO coupling will be greatest when the internal SO coupling is greatest . Patterson19 tested this suggestion by determining the irate constant for quenching (kq) of nine aromatic hydrocarbons by cesium chloride in methanol. A log-log plot of kq 19; kisc showed 38 a reasonably good linear correlation between increases in log kq and log k (k provides a measure of the inherent S0 coupling isc isc between S and the triplet to which intersystem crossing occurs). 1 (Linear log-log plots of rate constants for spin-forbidden processes !§_€2 for internal and external HAE have been reported.)64 The investigation reported herein will provide a somewhat dif- ferent test for the ratio of external to internal perturbation as a function of increasing gI. Perturbation of crowns l, 2, and 3 containing light and heavy cations by ethyl bromide was investigated. Since complexed cations are really external perturbers, the investi- gation is more appropriately described as a test of the effect of an external perturber as a function of the SO perturbation already present due to a complexed external perturber. Internal and external HAE have been investigated spectroscopically by Coulson and Cash.40 The phosphorescence and T1 absorption spectra of naphthalene in p-dihalogenated benzene host crystals at 4.2 K were determined and found to be composed of only totally symmetric vibrational modes. The phosphorescence spectrum of naphthalene in perdeutereonaphthalene is composed of non-totally symmetric as well as totally symmetric modes. From the fact that only totally sym- metric modes are enhanced by externally induced SO coupling, it is concluded that the external HAE is purely electronic in nature. The internal HAE, however, is shown to enhance the totally sym- metric modes of the parent naphthalene and to vibronically enhance out-of—plane halogen modes. This is viewed as being reasonable, since out-of—plane modes probably enhance the totally symmetric modes 39 of the parent naphthalene and to vibronically enhance out-of—plane halogen modes. The out-of—plane modes for naphthalene are of only moderate intensity. The same conclusion was reached by El-Sayed and Pavlopoulous65 based on phosphorescence polarization measurements on naphthalene and naphthalene derivatives at 77 K. RESULTS66 Synthesis 2,3-Naphtho-20—crown-6 (I) and 1,8-naphtho-21-crown-6(Z) were synthe- sized by room temperature reaction of the appropriate bis-(hydroxy- methyl)-naphthalene and pentaethyleneglycol ditosylate in tetrahydro- furan dimethylformamide (9:1) with potassium Efbutoxide used as 30“. INCH — " 0H OMF cw base. Isolated yields of l and 2 following careful alumina chromatog- raphy and crystallization from ether-pentane averaged 16% and 8%, respectively. 1,5-Naphtho-22-crown-6 (3)'was synthesized by reaction of l,Sfibig(bromomethyl)naphthalene with pentaethylene glycol in tetra- hydrofuran with potassium tfbutoxide present as base.1 a was iso- lated in 14% yield after careful alumina chromatography and repeated recrystallization from ether-pentane and then cyclohexane. The 2,3-, 1,8-, and 1,512557(methoxymethyl)naphthalenes (4,‘6, and 8, respec- tzively) were also prepared by a Williamson ether synthesis using the laiérihydroxymethyl)naphthalenes and methyl iodide in the case of 40 41 4 and,é and the his:(bromomethylnaphthalene and sodium.methoxide in the case of 8. The alkyl ammonium chloride salts were obtained by passing gas- eous hydrochloric acid through an ethereal solution of the parent amine. The iodo- and bromoalkylammonium Chloride salts were pre- pared from the corresponding iodide and bromide salts via anion ex- change. 3-Iodopropylamine hydrochloride was obtained from treatment of 3-bromopr0pylamine hydrobromide with excess sodium iodide in acetone at room temperature. Ultraviolet Absorption Spectra The following general comments apply to Figures 1 through 18. The 77 K ultraviolet (UV) absorption spectra were recorded on samples in either 95% ethanol-methanol (4:1, v/v) or ethanol-methanol (4:1, v/v) uncracked glasses (methanol added to prevent cracking). For spectra of crown complexes, sufficient salt (as determined from quantum yield titrations, vide infra) is present to assure that absorption is essentially from only complexed crown. The spectral shapes and intensities of Figures 1 through 18 are reproducible with a given tube and dewar (see Experimental), and the relative intensi- ties at a given wavelength are meaningful, but the relative intensi- ties at two different wavelengths are somewhat distorted from what the true relative intensities would be, due to the use of annon- identical reference sample (a nitrogen purged atmosphere was used as the reference). Salt and substituent induced changes in the energies and intensities of absorption peaks are evident and will 42 Figure l. 77 K UV spectra of 2,3-disubstitutednaphthalenes in 95% ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absorbancies for shorter wavelength region are given by the scale on the left; for longer wavelength region, by scale on the right. Concentrations (at room temperature) were 4.00 x 10'“ M for shorter wavelength region (except for naphthalene, 1.00 x 10-4 M). The scales have been adjusted so that M for longer wavelength region and 1.20 x 10'-4 the absorbancies, of both sections,after multiplication if necessary by the factors given in the figure legend, are those of a 8.00 x 10—5 E solution. Curve A, 2,3- naphtho-ZO-crown-6 (1); curve B, 2,3fibi§f(methoxymethyl) naphthalene (4); curve C, 2,3-dimethy1naphthalene (a); curve D, naphthalene (lg). 43 330 _ .4 _ m m 5 o. . o. O o 2 IIIII \ rue” _ A IIIIIIIIIIIIIII tt tt . Bat: :tl ....... O . _ t...” Cunnhvuhd'l'll'l llllll .fls . _ _ I ltl.l..:.:. _ _ 4 I . H . I‘ \oV \— C . . _ H \.I..tt tit... \. m 3 n \\ .- WH navy C i (tut-KW /. . C C M A ...... I]. v. O O / Bath“. it......tlt|Lt..t\ . (at. 2 3 H Dfl..Hl.|..l|. It .tI ts}. .\.\ m R RH H H m l I..l..lt... .rfli. 4 C C C H E C t\. . D.\..V _ ... = _. = W.“ A film-WW tt..J _ © D.n R R R 9 At... 8 \JHUJHUWuTtHP..iI. _ © .. .. c\f..nl It! .9. . ...... t... n _ l I. . I lit . . aflihhtrtuvtl u m. m . II. \ wot...» o .1 CA...- .1..I. . - ‘eo‘l-unII-oylu |\o w - 1“- . o 0 o u 0" JVFAHW k. M E . ......... .‘u‘l \ . - .. a u N . ~~ 3:03.034. o>=o_om 310 Mnm) 300 290 240 Figure l Figure 2. 44 77 K UV spectra of l,8—disubstitutednaphthalenes in 95% ethanol-methanol (4:1, v/v) glass. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absorbancies for shorter wavelength region are given by the scale on the left; for longer wavelength region, by the scale on the right. Concentrations (at room temperature) were 4.00 x 10.4 M for longer wavelength region and 8.00 x 10"5 M for shorter wavelength region (except for naphthalene, 1.00 x 10"4 M). The scales have been adjusted so that the absorbancies of both sec- tions, after multiplication if necessary by the factors given in the figure legend, are those of a 8.00 x 10"5 .M solution. Curve A, 1,8-naphtho-21—crown-6 (2); curve B, 1,81§i§f(methoxymethyl)naphthalene (6); curve C, 1,8-dimethylnaphthalene (7); curve D, naphthalene (18). Relative Absorbonce 45 : —R-= CH2(OCH2CH2)SOCH2 '— . R =CH20CH3---- : R =CH.3 ----- ‘03 R =H _...._..— .- 0.2 Mnm) % Figure 2 Figure 3. 46 77 K UV spectra of l,S-disubstitutednaphthalenes in ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absorbancies for shorter wavelength region are given by the scale on the left; for the longer wavelength region, by the scale on the right. Concentrations (at room temperature) were 2.00 x 10"4 6.00 x 10.5 M for the shorter wavelength region. The M_for the longer wavelength region and scales have been adjusted so that the absorbancies of both sections are those of a 6.00 x 10'5‘M_solution. Curve A, 1,5-naphtho-22-crown-6 (3); curve B, 1,5jbigf (methoxymethyl)naphthalene (8); curve C, 1,5—dimethyl- naphthalene (2); curve D, naphthalene (lg). 47 Onn m ouswfim “Sc: con 0mm CON Ohm OmN OnN N it. .t . q a . ...o. N.°1 . ..o ... .... .... ......» ..... . . .......... ”a ..Ia ..n pa. .\.\.o s‘Ix\..I Col: lob”: % m .w M w um \ :t. m 1. -3 .. .. ... .. . . m. ... . ... ... ...... t t t ...... O M ....m ... .... sent t... 0.0. O .. .- ... ... U‘ v .1 .— .s. .. .s a. so .. 8 .... . ... o.(. ... . .. .. S a .. . .... .. .. 0.. .. ..(. . w . . . . v . . . an o I a m “0 . m. ...“. ...... .../... .. ..... e W ”Io . m .o ...... a. . .... ....... 3. m foofo . m .m ..u. a .... .. .... ...... . 3 tuxoomaxofoomo . tan ”4 ...... . .... .1 ... w. t¢.O m ..... a. .. .. .. < O... @@ .. .. ... ..1. . m ... . N... Ind .. p b M bf - P p P OGn Figure 4. 48 77 K UV spectra of 2,3—naphtho-20-crown-6 (I) alone and with 5:1 added molar excesses of alkali metal chlorides in 95% ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absorb- ancies for shorter wavelength region are given by the scale on the left; for longer wavelength region, by scale on right. Crown concentrations (at room temperature) were 4.00 x 10-4,M_for longer wavelength region and 1.20 x 10.4 M_for shorter wavelength region. The scales have been adjusted so that the absorbancies of both sections are those of a 1.20 x 10.4 M solution. 49 ...... p m 0 O. O mug . . 1.. m . _ .xt— .. . a a d _ _ a _ nu . MW e H u . _ m N. ha. n26 - . . - A |||||| this n _ m i W .. -- tional... 3 . _ _ _ mm . ._ - .. - . .. ..., . CICC% ic P m 0C b 3 .\¢ 0 5 A\. n ”we...” .. 3......“ A B C D E F .2 t...t. . . m 0 c t4 .5 EOBA \\ . o A EthlfNflVflH-t\. ..Ov ... . ADV .. ... _m a‘.‘\ 1’3 m r . p » 5.3.43 _ . p . .53 p b 5 A” ..tvutut [9 Acw - t . .. : . 4. 2 CAD”. ......WM . . . . ttt t 0 *1 co .... ..u. .. ....... . 1 3.”! tttttt .t..ta.. . ...xlt Cal-rm?! .\ cs.-u-.._ \...\.H b p p b r3 “1&7. - — p t“... p w 0. 5. 2 t nu 3:03.034. o>=o_om Mnm) Figure 4 50 Figure 5. 77 K.UV spectra of 1,8-naphtho—21-crown-6 (2) alone and with 5:1 added molar excesses of alkali metal chlorides in 95% ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absorbancies for shorter wavelength region are given by the scale on the left; for longer wavelength region, by scale on right. Crown concentrations (at room temperature) were 4.00 x 10-4 M for longer wavelength region and 8.00 x 10"5 M for shorter wavelength region. The scales have been adjusted so that the absorbancies of both sections are those of a 8.00 x 10"5 a solution. Relative Absarbance 51 P = none -—-—-- =IWaCl """ = KC] " ''''' : RbC' — ooooo - = CsCl --'--' = 95% EtOH/MeOHM/l) "" B Figure 5 0.10 0.05 Figure 6. 52 77 K UV spectra of 1,5-naphtho-22-crawn-6 (3) alone and with 5:1 added molar excesses of alkali metal chlorides in 95% ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absorbancies for shorter wavelength region are given by the scale on the left; for longer wavelength region, by scale on right. Crown concentrations were 2.00 x 10-4 M for longer wave- length region and 6.00 x 10-5 region. The scales have been adjusted so that the absor- bancies of both sections are those of a 6.00 x 10.5 M M for shorter wavelength solution. 53 33.10 0.3% 02.. 0.! — 330 320 3|O p A > I ! | t \> ' ' i ’ t: I ! . I \ I l I | q» I I o — — - o c U _ Q U 0 0— o C> -’:—"$‘:C " --- ‘ "‘."..-;:‘-'o“" ‘ '3’.:',-‘ 05" o . 7- - n ."f‘.'fl .6 0' _J"..O‘ C’Qfl‘fl‘- ‘. -'.n‘ .” (OUII. " .‘ \\ I— 0“ o ‘0 _o o-.~ 'OJ,‘:" ./ 0"" o ' 3:": ‘0‘..' ,a'.o: ‘5'“ .‘, “lest", O ‘0 ’r ' s h‘ ‘s ' . ‘. . I , . a ' ': ’ I, I" ‘u I». [U ,‘i ..' .§ " I. p—- o( ‘ .‘ .’ I. -1 . . N.. o , fly.“ ~.. I a V "O o \ g a I ‘ . '.\ I. .- .. y .0 ‘ ' 1‘ 1’ O . ', '. . o o “ \ it .I ' _- \\‘ _l I! w. ‘ \. '0' I ._ _ (us A - o -4 . a. “ I I. .‘s‘ .0 a O . s ‘ o‘ a ' s. .0 , s '5 " ' I ¢ 0 " ‘ 0" I“ ' I ‘. .‘ I .l ' ' o ' ‘ " i '5“ o ‘\ I. .y ' ’0 II O O1 o ‘ ' "\ '. p _o ' ‘ h ‘ '0 v ‘ ' ' 1' ’3' .| “a, O ' ‘ “ ' I D,’ , . \‘ ‘ . n h ‘ '0 [a 'a ',' ': o " 0' ‘ O a, ' ' I. on ' ,1 I‘ - (Q ‘1 n no . e 0 .' . o o v v t ' “ " ' ' "' __ l 1_ l ' l. A l l L l ‘.'_1 W BONVBHOSBV BAILVWBB 330 320 255 265 275 285 295 305 3IO Mnm) 245 Figure 11 Figure 12. 64 77 K UV spectra of 1,5—naphtho-22-crown-6 (a) alone and with 50:1 added molar excess of bromoalkylammonium and iodoalkylammonium chlorides in ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explana- tion). Relative absorbancies for the shorter wavelength region are given by the scale on the left; for the longer wavelength region, by the scale on the right. The con- centration (at room temperature) of crown was 1.00 x 10-4 g for the longer wavelength region and 5.00 x 10-4 31 for the shorter wavelength region. The scales have been adjusted so that the absorbancies of both sections are those of a 5.00 x 10-5 g solution. Baselines: curve F, 2.50 x 10"3 2.50 x 10.3 y Y-iodopropylammonium chloride; curve H, M B-iodoethylammonium chloride; curve G, ethanol-methanol (4:1, v/v) glass. 65 can NH munmfim AIIII ea: oon 2m com 03 oz filll, 9‘...- F r)! q . d a - ...u a : u-.-u..ni. ‘1... ¢ 0.. o on”? also . .1. U. \ I. o ...a ‘ ... .0 Into 100.0 .J .0... . .. ,. :10..." . u 7.30 . too. . W 31...... . u...< ... _. .. “mean _. .n. .. H < _ .... .. .. B ro_.o II. €3\>:\vuIOms_\Io.m Ho m .. iii: Gnu “u ... iii utmom nm. .0 .oud mcoc ”4 a "..u . @ ... .. .... - a ... ..r .36 ...l. o... ’0‘ _ . . _ L _ _ onn com com Figure 16. 72 77 K UV spectra of 2,3-naphtho—20-crown—6 (l) alone and with 50:1 added molar excess of silver triflate in ethanol- methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absorbancies for the shorter wave- length region are given by the scale on the left; for the longer wavelength region by the scale on the right. Crown concentrations (at room temperature) were 4.00 x 10.4 g for the longer wavelength region and 8.00 x 10-5 )1 for the shorter wavelength region. The scales have been adjusted so that the absorbancies of both sections 5 are those of a 8.00 x 10- y_solution. 73 I 08 I 9.0 0H manage AEc: 0N 0mm OmN Ohm CON onm O¢N A33 _\v «1002\IOE £0883 mcoc n. ©© BDNVBBOSBV BALLVWBB awn olvw Figure 17. 74 77 K UV spectra of 1,8-naphtho-21-crown-6 (a) alone and with 50:1 added molar excess of silver triflate in ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absorbancies for the shorter wavelength region are given by the scale on the left; for the longer wavelength region by the scale on the right. Crown concentrations (at room temperature) were 4.00 x 10"-4 8.00 x 10-5‘§_for the shorter wavelength region. The ELfor the longer wavelength region and scales have been adjusted so that the absorbancies of both sections are those of 8.00 x:lO-5‘g_solution. 75 0mm [00 l0_.0 ... 9.0 l0N.0 fl 0_n NH unamfim AEcZ nOn nmm now Ohm cam now new E3 («icogxem .i.... ”.6683 mcoc a ©© . a a . . . . . . . . . a . a . . . . . . . . . . . . o . . . o o a . . o a . . g A men N0 v.0 0.0 l 0.01 BDNVBBOSBV BAIlV'BH QT. ova Figure 18. 76 77 K UV spectra of 1,5-naphtho-22-crown-6 (g) alone and with 200:1 added molar excess of silver triflate in ethanol-methanol (4:1, v/v) glass. See figure legend. Intensities are given in terms of relative absorbancies (see text for explanation). Relative absorbancies for the shorter wavelength region are given by the scale on the left; for the longer wavelength region by the scale on the right. Crown concentrations at room temperature were 1.00 x 10"4 5.00 x 10.5 g for the shorter wavelength region. The E for the longer wavelength region and scales have been adjusted so that the absorbancies of both sections are those of a 5.00 x 10'5 g solution. 77 ma ouswfim BDNVBEOSBV BALLV'IBH A? “FE: Cfln CNM 05 con CmN CCN CNN CCN CnN C‘VN iod rCfiC E3 .xqioszom _. IIII 1.. o womomog CCCC n. rCNAv .A"””'_ (\u p — Ch” 0 n OQN 78 be qualitatively noted. Precise energies will be given below in the results section on energy shifts. The lowest energy peak (longest wavelength) which is apparent will be referred to as the "0-0 band". Discussion of whether or not these are true O-O bands will be given later. The lower energy absorption band (from approxi- 67 mately 330 to 300 nm) will be referred to as the "8 band". Here 1 "band" is used to refer to a whole series of peaks belonging to the same electronic level. The 31 band, for instance, is comprised of transitions from the ground state to various levels of the first electronically excited singlet. The higher energy absorption band (from approximately 240 to 300 nm) will be referred to as the "$2 band". Absorption bands higher in energy than the S2 band were not easily experimentally accessible due to significant absorption by the alcoholic solvent below 240 nm. Figures 1, 2, and 3 show, respectively, the 77 K UV spectra for 2,3—, 1,8-, and 1,5-disubstituted naphthalenes and naphthalene itself. The substituents in each case are methyl, methoxymethyl, and crown- methyl. The solvent is 95% ethanol-methanol glass (4:1, v/v) for the 2,3- and 1,8- derivatives and ethanol-methanol glass (4:1, v/v) for the 1,5- derivatives. The absorbancies shown (after correction by the multiplicative factors given in the figures in some cases) are those of 8.00 x 10-5 31 (at room temperature) solutions. For ethanol-methanol (4:1, v/v) glass the contraction factor is 0.802 (Volume at 77QK/volume at 293010.69 For the 2,3- disubstituted naphthalenes, the 82 bands of the derivatives are quite similar to naphthalene in fine structure, 79 energies, and intensities, but the S bands of the derivatives appear 1 to be quite dissimilar relative to naphthalene. For both bands, however, note that the crown and methoxymethyl derivatives have very similar fine structure and energies. The 0-0 bands are all red shifted relative to naphthalene. The shifts are of similar magni- tude, but largest for the crown and smallest for the dimethyl deriva- tive. For the 32 bands of these derivatives, the entire band is red shifted for the 2,3-dimethy1 derivative, whereas the entire band is blue shifted for the crown and methoxymethyl derivatives. The situa- tion is not quite so simple for the 81 bands, however, since all of the derivatives have less fine structure than naphthalene. As is seen for the 2,3—derivatives, the 1,8-crown and methoxy- methyl derivatives have absorption spectra which are similar in intensity, fine structure, and energy (Figure 2). The 82 bands of the 1,8-derivatives have fine structure which is similar to that of naphthalene but all these bands are considerably red shifted rela- 1) tive to naphthalene (by about 1000 cm. . As is seen for the 2,3- derivatives, the S bands of the 1,8-derivatives show much less fine 1 structure than does naphthalene. The 0-0 bands are all red shifted, but the 0-0 band of the 1,8-dimethyl derivative is red shifted most, whereas the crown was red shifted most for the 2,3-derivatives. The red shift of the $2 band is also largest for the 1,8-dimethyl derivative. Comparison of the 77 K UV spectra of the 1,5- and 1,8-disub- stituted naphthalenes shows that they are quite similar in most respects: fine structure, energy shifts, and, for the $2 band, 80 relative intensities. The following differences for the S1 region are noted by com- parison of Figures 2 and 3. The $1 band of the 1,8-dimethy1 deriva- tive is about three times more intense than for the 1,5-dimethyl derivative. The 1,5-dimethyl derivative has a peak at 303.8, whereas there is only a shoulder in this region for the 1,8-deriva- tive. The 1,5-crown and methoxymethyl derivatives absorb approxi- mately two times more intensely in the S region than do the correspond- 1 ing 1,8-derivatives. These differences aside, the two sets of ab- sorption spectra are quite similar. Figures 4, 5, and 6 show, respectively, the effects of 5:1 molar excesses of alkali metal chloride salts on the 77 K UV absorp- tion spectra of crowns l, 2, and g. The absorption spectra of the complexes are very much like those of the parent crown in many respects. For crown l, the complexes and free crown have similar fine struc- ture, energies, and intensities. The S1 bands of the complexes, however, have three small peaks not observed for the free crown. The 0-0 band and whole spectrum of each complex are slightly blue shifted relative to free crown. For crown 2, the 82 bands of the complexes and of free crown g, as for crown l, are very similar in fine structure, intensity, and energy (complexes are slightly red shifted relative to free crown). For the potassium, rubidium, and cesium chloride complexes, there is some broadening of the 82 band in the 305 nm region. Thus, the peaks at approximately 308 nm for crown % and for its sodium complex 81 in the S1 region, appear only as shoulders for the other complexes (see note in results section on quantum yield titrations for sodium complex of crown g)' This difference aside, the 81 bands of free crown,% and of its complexes are quite similar. The 0-0 bands and whole 81 band of each complex are slightly red shifted relative to free crown. For crown 2’ the $2 and 51 bands of complexes and free crown shown in Figure 6 have very similar fine structure and the 0-0 band and whole spectrum of each complex are slightly blue shifted relative to free crown. Both the $2 and S bands of the complexes 1 are relatively more intense than those of the free crown. For the $2 band, all peaks of a given complex are increased in intensity by about the same amount relative to free crown. The increase is largest for the sodium and potassium complexes and smallest for the cesium complex. For the s1 band, the intensities of various peaks are affected differently for different complexes. Many of the comments made above concerning changes in fine struc- ture and the direction of energy shifts for complexes of crowns l, 2, and 3 relative to the respective free crowns are also true for other complexes of these crowns which were investigated: alkyl and haloalkylammonium chlorides, barium bromide, and silver triflate complexes. In relating the results for the 77 K UV spectra of these other complexes, only differences between the results for the alkali metal chloride complexes will be noted. Figures 7 and 8 show the effects of 5:1 molar excesses of ammonium and alkylammonium chloride salts on the 77 K UV absorption spectra 82 of crowns,l andlé in 95% ethanol-methanol (4:1, v/v) glass. The effects of 100:1 molar excesses of ammonium and grpropylammonium chlorides on the 77 K UV absorption spectrum of crown.%, in ethanol— methanol glass (4:1, v/v), are shown in Figure 9. For the alkylammonium complexes of crownll, the energy shifts are all small blue shifts, as is typical for complexes of this crown, except for the Brpropylammonium complex, for which the whole 82 band is significantly red shifted (by approximately 500 cm-l). Also, an extra peak appears for this complex in the 280 nm region, although the dip is small enough that it may be due to curvature in the base- line. The alkylammonium complexes of crown 2 require no special com- ments. The ammonium and Erpropyl complexes of crown.% have absorption spectra which are practically identical to that of free crown. Note, however, that the 0-0 band is slightly red shifted for the n: propyl complex but slightly blue shifted for the ammonium chloride complex. The alkali metal chloride complexes were slightly blue shifted. In contrast to the alkali metal complexes of crown 3, these complexes show very little change in absorption intensity relative to the free crown. The effects of excess bromoalkyl and iodoalkylammonium chloride salts on the 77 K UV absorption spectra of crowns‘l (S to 12.5-fold molar excess of salt), 4% (5 to 12.5-fold molar excess of salt), and ,3 (SO-fold molar excesses of salt) are shown in Figures 10, 11, and 12, respectively. The spectra for the bromoalkylammonium compiexes of crowns % and ,2 are from 95% ethanol-methanol glass (4:1, v/v). fmhe remainder of spectra in these figures are from ethanol-methanol 83 glass (4:1, v/v). Absorption due to excess iodoalkylammonium chloride salts becomes appreciable at wavelengths shorter than 280 nm. Base- lines corresponding to the amount of excess salt present are shown in the figures. While absorption by excesses of these salts obscure the higher energy portion of the 82 bands, the spectra as presented are sufficiently undistorted by this extra absorption for most of the $2 band so that accurate comparison to absorption by free crown and other complexes can be made. For crown l, the entire 82 band of both y-halopropylammonium complexes is significantly red shifted relative to free crown (ap- proximately 500 cmfl), as is also the case for the Eypropylammonium complex. The haloalkylammonium complexes of crown 2 give spectra and energy shifts similar to those of the alkali metal cation com- plexes. For crown a, note that the complexes and free crown have very similar absorbancies, which is in contrast to the differences seen for the alkali metal chloride complexes. Figures l3, l4 and 15 show, respectively, the effects of excess cesium chloride and barium bromide on the 77 K UV absorption spectra of crowns l (S-fold molar excess cesium chloride, 12-fold molar excess barium bromide), % (S-fold molar excess cesium chloride, 12- fold excess barium bromide), and g (S-fold molar excess cesium chloride, 50-fold molar excess barium bromide). These spectra require no further comments beyond those made for the alkali metal chloride complexes, except to note that both the 31 and S2 bands of the barium complex of crown,é are more similar in energy and intensity to free crown than are those of the cesium complex. 84 The effects of excess silver triflate on the 77 K UV absorption spectra of crowns 1 (50:1 molar excess), 2 (50:1 molar excess), and a (200:1 molar excess) in ethanol-methanol glass (4:1, v/v) are shown in Figures 16, 17, and 18, respectively. These spectra require no further comment except for the spectrum of the silver complex of crown 3. This is the only complex of this crown for which the 82 l 1 and 81 bands are significantly red shifted (-150 cm? and ’50 cm. . respectively). All other complexes are slightly blue shifted rela- tive to free crown 3, except for the Efpropylammonium chloride complex, which is slightly red shifted. At room temperature, with concentrations similar to those of Figures 1 and 2 added alkali metal chlorides do not appreciably change the UV spectra of crowns l and g. The intensities change by less than 5% and energy shifts are not discernible. It should be noted, however, that the extent of complexation of crowns‘l and 2 under the given conditions (2.00 x 10-4‘§_crown, 5:1 molar excess of salt) is not known precisely but is estimated to be approximately 60 to 80%. For crown 3, however, added alkali metal salts do change the room temperature absorption spectra relative to free crown. The shapes remain the same, but the intensities of both the S1 and $2 absorption bands are increased by approximately the same factor in each case as for the 77 K spectrum. Also, the order of the intensity increases at room temperature are the same as for the $2 band of the low temperature spectra (KCl 2: NaCl > RbCl > CsCl). The intensity of the S3 absorption band for room temperature spectra is also increased by added alkali metal chloride salts by approximately 85 502 in each case (1.20 x 10"6 21 crown with S-fold molar excess of salt present). The room temperature complexation constants for crown % with alkali metal chlorides must be large (106), since the absorbance 5 to 1.20 x 10.6 is only a few percent after dilution from 6.00 x 10- less than it would be assuming no increase in dissociation upon dilution. Low Temperature (77 K) Fluorescence and Phosphorescence Spectra The following general comments apply to Figures 19 through 54. All spectra are from solutions in uncracked 95% ethanol and are corrected. The spectra were recorded with instrument settings ap- propriate for obtaining maximum resolution, but the excitation band passes were too broad to allow the resolved spectra to also accurately indicate relative quantum yields. Furthermore, differences in ab- sorption are not taken into account. Thus, the relative intensities of two different curves are only approximately correct. They do, however, in most cases, qualitatively indicate the trend of the cor- responding relative quantum yields. More importantly, they indicate the presence or absence of changes in spectral shapes and energies. The term "0-0 band" will be used to refer to the highest energy (lowest wavelength) band observed in a spectrum (see Discussion section). In the case of spectra of crown complexes, sufficient salt is present in each case to ensure that the emission is essentially chae only to complexed crown. The amount of excess salt required was determined from quantum yield titrations (vide infra). The unilar (room.temperature) concentrations of naphthalene and naphthalene 86 Fluorescence of 2,3’Disubslituled Naphthalenes A = -R-= CH2(OCH2CH2)5OC —— Figure 19. 8 R= -CH;OCH3 - .. .. C = R = CH3 ..... D' R = H ._ ..... _ '\ ‘\ -\. \\\ \.\D A \-_‘-;\h.\ ‘-e~h l I ‘3 ~- ~- 400 Mnm) ; Corrected fluorescence spectra of 1.00 x 10"4 )1 2,3- disubstituted naphthalenes (2, 3-Naphtho-20-Crown-6 (1), Curve A; 2 ,3—Bis-(methoxymethy1)naphthalene (4), Curve B; 2 ,3-dimethylnaphtha1ene ( ), Curve C, and” naphthalene (lg), Curve D) in ncracked 95% ethanol glass at 77 K. 87 Figure 20. Corrected phosphorescence spectra of 1.00 x 10'-4 ! 2,3—disubstitutednaphthalenes (2,3—naphtho-20-crown- 6 (1), curve A; 2,3j§i§f(methoxymethyl)naphthalene 0%), curve B; 2,3-dimethylnaphthalene (a), curve C; and naphthalene (*0) curve D) in uncracked 95% ethanol. 88 )50C H 2 A B \ , 5 / o \ 3 R = CH20CH 3 \ \ \ oo " R 'R- 3 CH2(OCH2C H2 \ H \\\.\. \ rCHH \nr3\ D\\ \\.H\ \nl\\ .\ \..\ R R \\“\ «\‘uo-‘oo‘. ‘ 0‘ J A \\|l.\ CD“'. I: B. ..ildfl...l A B C D .11.: .I/l 1.1. " O/ou " .- a \ -\. \J “\ ‘ \ coll ‘ Al' """ \‘0'0'I .\ \\ 11.1.. I\. |I|‘\‘I.IIIII|II Dale! \\“II‘ Ch“ l | ' 'u‘u."oonlv 'M‘ “H ..... ‘I A 3 fl IIIII Dfiulcnnl null“! I. ..l. .I I I..- 1:1/..: .l ......... ”111.! I ...... l .l l.. I, lo'c'o'o' '- I. \ 'Io'o'.‘ a “| ...‘h D «Hall: all. WhflfllflhlwuuuH .l .. . |.!. “Mfihlllu "-.:-will..- .- .-oo‘uo'no'oo'o- Phosphorescence of 2,3-Disubstituted Naphthalenes \\l I I 600 550 Mnm) Figure 20 500 450 89 Fluorescence of 1,8- Disubstituted Naphthalenes A H R R D u !c,: ©© I u .q, !!H A=-R—=CH2(OCH2CH2)SOCH2 — HI“: 8 R=CH20CH3_-_- ifl'; A C: R=CH3 _____ iii?! 0- R=H _ ..... _ i": i'“ i l Mun. D by 3.0 310 350 400 Mnm) Figure 21. Corrected fluorescence spectra of 1,8-disubstitutednaph- thalenes (1,8—naphtho-21-crown-6 (2), curve A; 1,8-Bigf (methoxymethyl)naphthalene (6), curve B; 1,8—dimethy1naph- thalene ( ), curve C; and naphthalene (lg), curve D) in un- cracked 9 Z ethanol at 77 K. Curve D should be multiplied by 3.0. 90 Phosphorescence of 1,8- Disubstituted Naphthalenes a ’5; R R i: 2 i! . H“: 1i i. :| : c .. 3 - -= H20CH CH OCH Ii ”i .. a: R =CHzoCH:-3)’._ 2 i; !i l C: R =CH3 ..... i! E!i \.| D: R =H _.._..... .i Hi 1 ' n "' " "‘ i: 'i ‘1 ‘\ ICN Hf ! ' -. ’ 'k .l ! 1 i *' ‘ l= ! ‘~ ' . ’ \ : l " ‘ I | l I ' \' ' x. 1.? i ‘ El Y j ‘ \\ ”I . -' " ‘° Ii ‘ I. ‘= ‘ ii ' “a " " \\ : . r. ‘ ‘ . . l i i!= " ‘~ “7 \ \ i i Ii= '\ I" \' : 3 l: \ \.\ c ! UH | I \-'\ 8"”\' i ‘l‘ : \I ‘- ‘~ "7’D\\ \ . “u! l: \ ft \ \‘ \ i w -.i \" \‘ \' i '1 L Mull. Dby 3.0 \°‘\‘ \\ I i ‘30. -.., ! \°\\ / J O? 450 500 5J50 J Mnm) 600 ¢ Figure 22. Corrected phosphorescence spectra me of 1 8-dis - Bightgzlfiges (1,8-naphtho-21-crown-6 (é), cuggztifu1eg ‘Ezzhylnaphtzgithyl)naphthalene (6), curve B; 1 8—di-’ - curve D) 1 ene (1), curve C; and naphthalene ( n uncracked 95% ethanol glass at 77 K 18), 91 Fluorescence of I5-Disubsri1uted Naphthalenes c F? @@ r. R 93 .1, A=~R~ =CH2(OCH2CH2)50CH2 ;-. B=R =CH20CH3 .9. CrR =CH3 -D' E, [3 IF? =|4 : . 'I - “q - -. ...é.".n ‘ 0 o a -- -_--.c_....._-.e .. it- - - J.-....-.-....-.-... JD 3‘. ...Oso-o-O-O-I-u-. ' iJga 4(N0 Mnm) Figure 23. Corrected fluorescence spectra of 1,5-disubstituted- naphthalenes (1,5-naphtho-22-crown-6 ), curve A; 1,51Bigf(methoxymethyl)naphthalene (8) curve B; 1,5- dimethylnaphthalene , curve C; and naphthalene (lg), curve D) in uncracke 95% ethanol glass at 77 K. 92 Phosphorescence of Ib-Disubstituted Naphthalenes A i R 00 R A =-R- = cr-gocraz CH2)5 OCHZ B=R =CH20CH3 CtR =c 02R gH MultC and D by 0.3 ‘P Figure 24. 500 550 600 630 Corrected phosphorescence spectra of 1,5-disubstituted- naphthalenes (1,5-naphtho-22-crown—6 (3), curve A; l,51§ig(methoxymethyl)naphthalene (8), curve B; 1,5- dimethylnaphthalene (2): curve C; and naphthalene (*2), curve D) in uncracked 95% ethanol glass at 77 K. 93 Perturbed Fluorescence of 2,3-Noph-20-Cr-6 .A H j) — “-I.- ----..- C :: I :91 .{t ’i: I; I .I 1.0. H Um .; u; Figure 25. Corrected fluorescence spectra of 2.00 x 10.4 M 2,3-naphtho-20-crown-6 ( ) alone and with 5:1 mblar excess of alkali metal chlorides added in uncracked 95% ethanol glass at 77 K. 94 Perturbed Phosphorescence of 2,3-Noph-20-Cr-6 E f 1 . coo : I ‘x i ‘. .E E '\ {‘g P ! '2 ‘\ (I \. A: none ,‘ ;",o ‘\ i \ B: NoCI ------ fig". \J \\ c: KCI - ----- III-m \ D: RbCl '_ """ '- im"\ \ E= CsCI ------ E ”ii \‘- \\ E p ”I \ \ :: 4 3r ‘ P ‘ ME '1 I i X \. / \. \ hf“. ,’ \‘ '1! '\ / B \ \\ 4311!; A1! A\. \_/ u“ \ C 'I' I ‘ L. I l‘: ' i i I"! ‘ I'/. \\ \. .\ 0*,4} 51' U ,7 x ,v,‘ x; \ -\ “Lida |' L (J, \. \ \ HM‘JI‘. N U. n. \\ =\ \ :h! . It‘ I, ‘ \\\\. " A .. . .\ "1 i: A :’ I “ \‘\.D \'\ 9:! ' " ' i \ ".\ '\ \. 1' I ‘ V %‘~~ \\ 3!; / \\ \., .\ 'l, \ \\, C \ \ a“ \ {IE '.\ .\ F I: V “x \ g A. Q, x\\ \ i! \\\, \\ I. “ \ \-.\\. r] . \.,\\ --.” \\ =" l l l ‘ 450 500 550 600 Mnm) > Figure 26. Corrected phosphorescence spectra of 2.00 x 10'“ M 2,3—naphtho-20—crown-6 (1) alone and with 5:1 molar excess of added alkali metal chlorides in uncracked 95% ethanol glass at 77 K. 95 Perturbed Fluorescence of 1,8-Noph—21-Cr-6 t—.——. “In. C D \ .. \______ w n I? l l J l 1 I l 320 350 400 Mnm) ——-——+ Figure 27. Corrected fluorescence Spectra of 2.00 x 10_4 M 1,8- naphtho—Zl-crown-6 (2) alone and with 5:1 molar excess of added alkali meta1 chlorides in uncracked 95% ethanol glass at 77 K. 96 Perturbed Phosphorescence of 1,8-Noph" Zl-Cr-b P A= none —'——' B =IVOCH‘ """ C: KCI "" ''''' D: RbCl " """ " E: CsCI ------ AB \ I K‘ ’ \ \ ‘E r .’ |‘\ / . ’/ ‘\\\ \ \ 9 \ \ .-/c \\ \\ ../ f‘\ ‘\ \ \J.’/ ‘\ \ \ \_’/ .\ 2‘ ‘\ A g E \\\- I ~ ‘C'x ' \ \.\\..\ D\\.\ C'\. \. ...\\ \-'\\ \} l l l 500 550 600 Mnm) 4 Figure 28. Corrected phosphorescence spectra of 2.00 x 10-4 M 1,8- naphtho-Zl-crown—6 ( ) alone and with 5:1 molar excess of added alkali meta chlorides in uncracked 95% ethanol glass at 77 K. 97 Perturbed Fluorescence of LES-Naphtho-ZZ-Crown-G i W999??? x Q g ! ! 3l0 Figure 29. 350 ‘ ' ‘ 4oo Mnm) > Corrected fluorescence spectra of 1.00 x 10'4 M 1,5- naphtho—22-crown-6 (3) alone and with 5:1 molar excess of added alkali meta1 chlorides in uncracked 95% ethanol glass at 77 K. 98 Penurbed Phosphorescence of I,5-Noph1ho-22-CrOWn-6 (9 R i\. ,! A: none ! 8: N00 ----- g' C: KCI - ----- i D= RbCl — —-— E= CsCl ----- i4 "-1 l 11 l J 450 500 550 600 630 Mnm) > Figure 30. Corrected phosphorescence spectra of 1.00 x 10-4 M 1,5-naphtho-22-crown-6 (3) alone and with 5:1 molar excess of added alkali metal chlorides in uncracked 95% ethanol glass at 77 K. Perlurbed .3-um 9.0m ear. .2: #:50133- I. - .~H ‘5 .-ufdwg s suns-r 99 Fluorescence of 2,3-Nophlho-20-Crown-6 ‘ o I: P l"! A= none —- g: D B: NH4CI ____ 35‘- C= n -PrNH3Cl ---~ .3 D: i-PrNH3CI —..—-- Egg E: 1-BuNH3Cl —----- 3:5; 5! ' ~; Figure 31. Mnm) —--—> Corrected fluorescence spectra of 2.00 x 10“4 M 2,3- naphtho-ZO-crown-é ( ) alone and with 5:1 molar excess of added ammonium an alkylammonium chlorides in un- cracked 95Z ethanol glass at 77 K. 100 Perturbed Phosphorescence of 2,3“N0phtho-20-Crown'6 ooo £ P ' ' A= none ' ‘ ' ' B: NHqCI c = n-PrNHSCI --—-- E i x ”’9 ‘° 0 ' P NH3CI ys' . \ :|- r -n-«- x :7 £\\‘ ‘. c . i'. I" .1. '9: \ \ ’--.\ E ' 1-BUNH§I --'-"' f, g a 3‘: 1‘. ‘ ‘- i a A ' .5." A?" ‘\ X I" ‘ g! f. 31'! A" t '{Lo‘a ‘\ 3 t ' ‘3’" ‘ [11.3 \ \ i=3!" 5’ \ \ "'1 \\\ u :f°\ W \\ ‘ =3 V" s s—‘A .3 xx x ‘3 in? Ii 3“ 1! A35 \ \ gt‘ : A \J',‘ ‘\ ..’ I “R \ \. I: \ ’3’ ‘ ,' \_ ._ \ \ 5' . ." ‘ I \ \ \ \. J I \ ‘ \ \o~ _ 3 .' I. '\ \ \ .3 g \ \ \ _ .g \ Q g ‘\ . "T.‘\ \ \ cc I ~' ".\I \ . a. I H" A \s} \‘ \. «5 . ‘s \- . . s \ A.” - ‘\.\~‘\ \' 3.. ‘§3'\ “H 2 L_, l l 1 ‘3] 450 500 550 600 630 Mnm) >— Figure 32. Corrected phosphorescence spectra of 2.00 x 10-4 g 2,3-naphtho-20-crown-6 (l) alone and with 5:1 molar excess of added ammonium and alkylammonium chlorides in uncracked 95% ethanol glass at 77 K. 101 Perturbed Fluorescence of l,8-Nophtho-2l—Crown-6 C I? H. D' g :3“. «b c ©© H ‘. --‘.’..-‘ :5 :0 .‘ '3 '9! ‘r "fll 3 ‘ .S F) I; “5“ 0" . : ‘ t ~D| V I‘,’ :‘ w P. 0‘ ‘ \‘f ‘..\ If“: . . ' “.i' H! A- none ‘b: X“ H". B= NHQCI ----- z‘ ‘ \ o ' ... a C: n-PrNH3CI «- E E \\\ ' E n {\\ [)3 f—{Drrqp¥§:|._u_mu. 3 c 33 3‘ a | ' . .‘ o 2 ‘g 5 '4‘. 3.! E- 1-BuNH3CI ----—- ‘ I f; ! 'J ‘ \, EIAt . \‘\ H 3 x u a .- ' “ o ' ‘ '\! ‘ \\\ ‘ '\\ \ | -.\ \ \ \\ . \ \3‘g ‘\ \ \§\ . \:\\° \ \‘\\.‘ ‘ \. Y;\ \ O ‘\ -.\‘ \. \3‘. \ ‘\\ \ \.O\.\ \ \\ ‘. §..\\. \ \ \°\‘.§..‘\- "\;\§-\ o ‘~ . \f.,\‘ .§§\,K‘ ‘ ~. 4 1 l J 1 4 J ’ - Mnm) -—-"' Figure 33. Corrected fluorescence spectra of 2.00 x 10-4 M 1,8- naphtho—Zl—crownré ) alone and with 5:1 molar excess of added ammonium an alkylammonium chlorides in un- cracked 951 ethanol glass at 77 K. 102 Perturbed Phosphorescence of l,8-Nophtho-2l—Crown-6 A E ‘x 1 l I : ©© ' . : i * 1 ‘ ‘ i P 5 ‘ | : i i ’ -'; : A= none —— 1 E . DI . 1 1 3 A! 8' ””40 ““““ 1 i . l\‘. a s 3 .3; C: n-Per-IBCI -«--- . ' i : ‘ . “ ; °_ .. ..... _ 3 3 , :5}; , D . PrNH3CI D: : l i ‘ ‘. . ___.____ . 'o' :’ 33} ' f I! ‘3 I : l " :I.‘ ‘3. .' 0 I 03 3“ ,5 A i'i '3. ' is“. 8 t ' ‘ z o. f- .‘ 'l.‘ ‘ ' '0 ‘3 3'! - ." i" c ‘t i..." ,. I. ’I’ -.* 0 \. c - ‘ 8' . ' II °‘t! “ ,",’ l' ‘ :4 ‘3" ‘3, ’13; s. ‘z i 'I ’3. ;§\.-"/ I; v3 3 ‘. ‘s \ c "" Corrected fluorescence spectra of 1,8-naphtho-21-crown- 6 (g) alone and with 5:1 added molar excess of bromoalkyl- ammonium chlorides (2.00x10‘4 ! crown) and with 20:1 added molar excess of iodoalkylammonium chlorides (1.00x 10‘4 11 crown) inguncracked 95% ethanol glass at 77 K. 108 Perlurbed Phosphorescence of l,8-Nophlho-2l—-Crown-6 A l‘ ©© P A= none B= BrlCHzlz NH3CI ---- c= Br(CH2)3 mam-«- D= IlCH2)2 NH3CI - ----- — E I (cl-:2)3 NH3CI ----- D A I . '. i ' X i I a' ‘ f- I Q... ‘ o. ' ‘. \ g. i. a 3" E i ’Q. .\ i I, \- ‘. ‘ . | \ ,‘\ A I ‘ \-\ ‘ U | ‘ s“. \ ‘ .\ .‘3‘33 2 : ‘. 4?? 137'“. n ‘x. \x :‘ “'\ E \' -\ i a - - ‘~.‘._\\ :3 ‘3' ~.\ 35 .=§\ 3;? ...,» .‘9 *- a - - - :9 . ....-. - I” [* ... ...... L 1 l J 450 500 550 600 630 Mnrn) Figure 40. Corrected phosphorescence spectra of 1,8-naphtho-21- crown-6 ) alone and with 5:1 added molar excess of bromoalky ammonium chlorides (2.00x10’4 g crown) and ‘with 20:1 added molar excess of iodoalkylammonium chlorides (1.00x10‘4 y crown) in 95% ethanol glass at 77 K. 109 Perlurbed Fluorescence of l,5-Noph1ho-22-Crown-6 £3 (.3 p =Iwone ‘-—‘—' = BrlCHz)2 NHBCI --—- - BrlCHzls NHSCI ---~-~ = IlCHzlz NHBCI -« ---- “1 C3 (3 GD I> = 1(CH2)3 NH3CI --.-- Figure 41. Mnm) —* Corrected fluorescence spectra of 1.00 x 10-4.§_1,5- naphtha-22-crown-6 (3) alone and with 20:1 added molar excess of bromo- and iodoalkylammonium chlorides in un- cracked 952 ethanol glass at 77 K. 110 Perlurbed Phosphorescence of l,5 -Nophtho-22-Crown-6 C 9 5 d ;i .33 F 33 63> a; I I 35 c P I ! rs A: “me _ '2 i l , 3‘, . . e- Br(CHz)2 Nl-l3Cl ---- 3 .' a I . ' 3.9;; i i C ' Br (CF12): NHSCI ’“""" hr J ' : ...m_ ' ‘ l ' : -...-- go. 2' 3 E I (CH2)3 NHBCI .H ! a 'I I - . I E E 35;: ' c i ‘3’, . ‘ 5“ .' .' c l i! ' D ,3 :1 '- it i A... °\ .3' ii ii 332‘ ’ 3 ' ‘ - . _ : . J\ ’. . . 5 ’l ‘ K . g 3 E -' I, " i' 2‘- ' - I l ' 3 ° . \ . 5. I E - ' 3n! ' i 5 i g I) l l I” I I 9‘ ' , \ I. . 5n: cc 2 , ‘- .' - \ a :é‘l" ! - bx! . . \‘x 3. 3A 1 “\ I . ‘ x " E i l"- \‘ f ' ‘\ ‘. ' ... i « I“ -\- .\ ‘. .. {’ ;; é \ \ I: ! ‘3: . '- 1.. ‘3"; \\ \‘ «J — \ i . 9.2!? \d ‘X .\'—’.’ é \\ i. K. \‘bh‘e. f.‘ \ .\ ", .I 3“?\\.\. v \\§\§ ' l l I .1 450 500 550 600 630 Mnm) : Figure 42. Corrected phosphorescence spectra of 1.00x10-4 M 1,5-naphtho-22-crown—6 ) alone and with 20:1 added molar excess of bromo- and iodoalkylammonium chlorides in uncracked 95% ethanol glass at 77 K. 111 Perlurbed Fluorescence of 2,3—Nophlho-20-Crown-6 a) ‘0 Figure 43. Corrected fluorescence spectra of 2,3-naphtho-20-crown- 6 ( ) alone and with 5:1 added molar excess of cesium chloride (2.00x10'4 91 crown) and 10:1 added molar excess of barium bromide (1.00x10'4‘§_crown) in uncracked 95% ethanol glass at 77 K. 112 Perturbed Phosphorescence of 2,3-Nophlho-20-Crown—6 I o"- i! l ‘. I I ' I , . . 1 . a, P , .‘ A none . o c t, . — ————— \ 3‘i . 3 c CsCl -—-—— .. 1 g I‘ a 5'- l l . I! I! '. P‘c .! II p I fb . ‘ .. J “I \ 9 . ‘I I. I ' ’ n ' - o ‘ I . II ;\ ' I '\ i! .‘i 3‘. -’ . ‘\ I . 3‘ .|\\./ P 3‘. I ' " I ! \‘ I 3 I II I ' '\ - ' fl I' ' ' I \ I 'ci 9 ' ! I i‘ 3n} {I ; I I \ A ...I l! 1 ! .I \\ : III '! ‘ 3 | ‘ ° 3 Ii ’ ' I ' 'I \\ I l I; ./ l I ‘ I ' c ! ‘ l° ’ ' ' \\. I I’ I \ I | s ‘I ‘u i . J - ‘ !. \.' I \. \\ i . ‘-~‘n - Q -U'UI-‘I '- fir..- ‘q'd'U’U’U'Jg “'3’ ‘: ' C l J I 450 500 550 600 630 Mnm) * Figure 44. Corrected phosphorescence spectra of 2,3-naphtho-20- crown-6 ( ) alone and with 5:1 added molar excess of cesium ch oride (2.00 x 10" 31 crown) and 10:1 added molar excess of barium bromide (1.00 x 10"4 ! crown) in uncracked 95% ethanol glass at 77 K. 113 Perlurbed Fluorescence of l,8-Nophtho-2l-Crown‘6 J l I l J 1 L ° 3l0 350 400 Mnm) -——> Figure 45. Corrected fluorescence spectra of 1,8-naphtho-21-crown- 6 ( ) alone and with 5:1 added molar excess of cesium chloride (2.00x10‘4 E crown) and 10:1 added molar excess of barium bromide (1.00x10'4 g crown) in uncracked 952 ethanol glass at 77 K. 114 Perlurbed Phosphorescence of l,8-Nophlho-2l-Crown-6 Figure 46. 1 I l 4 500 550 600 630 Mnm) = Corrected phosphorescence spectra of 1,8-naphtho-21- crown-6 ( ) alone and with 5:1 added molar excess of cesium ch oride (2.00x10‘4 y crown and 10:1 added molar excess of barium bromide (1.00x10' fl crown) in uncracked 95% ethanol glass at 77 K. 115 Perlurbed Fluorescence of l,5-Nophlho-22-Crown-6 9‘. § 0 ,‘-'--o—--’-‘~_. "2: s. - «vs-"r l I --'r-'----r-----r “rm-r =v BIO 350 400 Mnm) -———-> Figure 47. Corrected fluorescence spectra of 1,5-naphtho-22-crown- 6 ( ) alone and with 5:1 added molar excess of cesium chloride (2.00x10'4 5 crown) and 50:1 added molar excess of barium bromide (1.00x10‘4 )1 crown) in uncracked 951 ethanol glass at 77 K. 116 Perturbed Phosphorescence of l5-Noph1ho-22-Crown-6 C a E. P g: If“. A=none .3; ; ‘. B‘BoBr2 g: i I: C=CsCl __ 3i. i ': i i i '; 3P3. . i 3&3 I ' . 1 '3; . ‘ 5.50-0 . I K; '. ‘.“~-o-O-- 0--“ A... fi”‘0-‘IWIJ.U A"‘.‘-‘ '0': o" I I l J J 500 550 600 630 —— Mnm)———> 8. or Figure 48. Corrected phosphorescence spectra of 1,5-naphtho—22- crown-6 ) alone and with 5:1 added molar excess of cesium ch oride (2.00x10‘4 g crown) and 50:1 added molar excess of barium bromide (1.00x10‘41§_crown) in uncracked 952 ethanol glass at 77 K. 117 Perturbed Fluorescence of 25-Nophtho-ZO-Crown-6 “ eeo H P FKNTE Ag 0 SOZCF3 -—-— '- 3|O Mnm) -—-—-> Figure 49. Corrected fluorescence spectra of 1.00x10'4 1!. 2,3- naphtho-ZO-crown-6 ) alone and with 50:1 added molar excess of silver tri late in uncracked 95% ethanol glass at 77 K. 118 Perturbed Phosphorescence of 2,3-Nophtho-20—Crown-6 l i: P I .' . none —— g I "I ‘, Ag 0 SOZCF3 —'""' '- I I . \ P. . : II ' H I |‘ f\ I 1 ‘ ‘ I \ I | ' I ‘ I | ' “ I \ I I ‘ \ I ‘ I . I I ‘ I I : I |‘ : : ' ' ‘ l I : ‘\ " ‘I I I ‘ I ‘ ' I I ' I I l ' g I I \ ’I I I ‘ l I I . ' I | ' I ' I I ' I I - ' \ I \s "\ i ‘ x \ . \ I \\ i ‘ \ : \ I “ \ I \ . \ C 1 l 1 J 450 500 550 600 630 Mnm) ———"' Figure 50. Corrected phosphorescence spectra of 1.00x10'4 fl; 2,3-naphtho-20-crown-6 ( ) alone and with 50:1 added molar excess of silver triflate in uncracked 95% ethanol glass at 77 K. 119 Perturbed Fluorescence of l,8-Nophtho-2l-Crown-6 F r I I D ©© P t n none ——"" '. " Ag 0 SOZCF3 ——-- — . ’----‘-—"’- ‘- I BIO 350 400 Mnm) -—--'- Figure 51. Corrected fluorescence spectra of 1.00x10-4 g 1,8- naphtho—Zl-crown—G ( ) alone and with 50:1 added molar excess of silver tri late in uncracked 952 ethanol at 77 K. 120 Penurbed Phosphorescence of l,8-Nophtho-2l—Crown-6 P fl I I I I I I I I I I I I I I I I I I I l l I I I I I I I I I I .' EL I i J J 500 550 600 630 450 Mnm)-———-- Figure 52. Corrected phosphorescence spectra of l.00xlOm4 ! ) alone and with 50:1 added 1,8-naphtho-21-crown-6 (% molar excess of silver triflate in uncracked 95% ethanol glass at 77 K. 121 Perlurbed Fluorescence of l,5-Nophlho-22-Crown‘6 l I; P A= none ———- n e: AgOSOZC§---- c= Agosozcg ---——- (xlO) . o-------"°——----- I L; 1" l J J “-~:“"‘3“-‘-I------r- _l 350 400 Mnm) —-—-> Figure 53. Corrected fluorescence spectra of 1.00x10-4 _)_1_ 1,5- naphtho-ZZ-crown—6 ( ) alone and with 200:1 added molar excess of silver tri late in uncracked 95% ethanol glass at 77 K. 122 Perturbed Phosphorescence of l,5-Nophlho-22-Crown-6 C.) b.‘ none Ag 0 5020-}, ---- — l l l 450 500 550 600 630 Mnm) : Figure 54. Corrected phosphorescence spectra of 1.00x10-4 M 1,5-naphtho-22-crown-6 ca) alone and with 200:1—added molar excess of silver triflate in uncracked 95% ethanol glass at 77 K. 123 4 M. The molar concentrations of crowns 4 derivatives were 1.00 x 10- in the cases with added salts were either 2.00 x 10— 10.4 M. Solvent blanks gave essentially flat baselines under similar :!_or 1.00 x conditions. Figures 19, 21, and 23 show the low temperature (77 K) fluores- cence spectra for naphthalene disubstituted at the 2,3—, 1,8-, and 1,5-positions, respectively. In each case, substituents are methyl, methoxymethyl, and crown-methyl (naphthalene spectra are also included for comparison). The analogous set of phosphorescence spectra are shown in Figures 20, 22, and 24. For the fluorescence spectra of the 2,3—disubstituted naphthalenes note that the spectra for the crown and for the methoxymethyl sub- stituted naphthalenes are quite similar, with the latter being somewhat less intense. The fluorescence of 2,3-dimethylnaphthalene exhibits fine structure that is significantly more pronounced than that of the crown or of the methoxymethyl derivatives but includes all the peaks seen for methoxymethyl and crown substitution. The spectra are all red shifted compared to naphthalene by similar but slightly different amounts. Precise substituent and complexation induced energy shifts will be presented below and discussed later. The phosphorescence spectrum of 2,3-dimethylnaphthalene is quite similar to that of naphthalene. As is also the case for the fluores- cence spectra, the crown and methoxymethyl derivatives exhibit less pronounced fine structure. This is especially true for the crown, the highest energy band of which appears as a broad rounded hump. By comparison, naphthalene and its methyl and methoxymethyl 124 derivatives have two peaks in this region. The spectra are all blue shifted relative to naphthalene. The "0-0 band" for the 2,3- crown will receive special discussion later. For the fluorescence spectra of the 1,8-disubstituted naphthalenes, note that, as for the 2,3—series, the spectra for the crown and for the methoxymethyl substituted naphthalenes are quite similar both in intensity and in energy. As is also the case for 2,3-dimethyl- naphthalene, 1,8-dimethylnaphthalene exhibits fine structure which is significantly more pronounced than for the crown or methoxy- methyl derivatives. The spectra are all red shifted relative to naphthalene, as for the 2,3-series, but the red shift is much larger for dimethyl substitution in the 1,8- series. The phosphorescence spectra of the 1,8-disubstituted naphthalenes are all very naphthalene-like. They are all red shifted compared to naphthalene. The dimethyl derivative, as for the fluorescence spectrum, is red shifted much more than the crown or dimethyl derivatives. Also, as with the 2,3- series, the crown and methoxy- methyl derivatives have similar intensities and energies. Considering the fluorescence spectra of the 1,5-disubstituted naphthalenes, the crown and methoxymethyl derivatives have fine structure and energies which are quite similar, as is also the case for the analogous 2,3- and 1,8- derivatives. The fluorescence quantum yields for the 1,5-crown and methoxymethyl derivatives are more similar than the relative spectral intensities shown would indicate (vide infra). As is also the case for 2,3- and 1,8-dimethyl- naphthalene, the fluorescence of 1,5-dimethy1naphthalene exhibits 125 fine structure which is more pronounced but otherwise similar to that of the crown or methoxymethyl derivatives. The fluorescence spectra of the 1,5-derivatives are all red shifted relative to naphthalene, as is also the case for the 2,3— and 1,8-derivatives. The red shift is much larger for 1,5-dimethyl substitution as is also the case for the 1,8-series. The phosphorescence spectra of all the 1,5-derivatives are remarkably naphthalene like, which is also the case for the 1,8- and, to a lesser extent, for the 2,3- derivatives (zidg'ggpgg). The 1,5-crown and methoxymethyl derivatives have fine structure and energies which are quite similar, though the fine structure of the methoxymethyl derivative is slightly less pronounced than that of the crown. The 1,5-dimethy1 derivative has somewhat more pronounced fine structure and is red shifted (relative to naphthalene) much more than the 1,5-crown and methoxymethyl derivatives. A much larger red shift is also seen for 1,8-dimethylnaphthalene. The relative phosphorescence intensities shown for the 1,5-derivatives are in the right relative order, but the cruves for naphthalene and 1,5-dimethyl- naphthalene should be about three times less intense than shown. Figures 25, 27, and 29 show, respectively, the effects of S-fold molar excesses of alkali metal chloride salts on the fluorescence spectra of crowns l, g, and Q' The analogous set of phosphorescence spectra are presented in Figures 26, 28, and 30. The molar concen- 4 4 tration of crown is 2.00 x 10- g for crowns '1’, and ,2, and 1.00 x 10- y. for crown g. The fluorescence spectra for the alkali metal cation complexes 126 of crown l are all decreased in intensity relative to free crown, the decrease being smallest for the sodium complex and largest for the cesium complex. The spectra of these complexes are similar to each other but somewhat different from that of the free crown. A comparison of Figure 25 to Figures 31, 37, 43, and 49 will show that all other complexing species investigated have fine structure which is similar to that of the alkali metal cation complexes. The 0-0 fluorescence bands of the alkali metal complexes are slightly blue shifted relative to free crown. Inspection of Figures 31, 37, 43, and 49 or reference to the results section on energy shifts will show that all of the other complexes of this crown which were in- vestigated also have 0—0 fluorescence bands which are slightly blue shifted relative to free crown 1. Some bands in the spectra of these complexes are red shifted relative to what appear to be the cor- responding bands of the free crown, however, and one cannot say that the entire spectrum of a given complex is blue shifted. The phosphorescence spectra of the alkali metal cation complexes of crown 1 (Figure 26) are all increased in intensity relative to free crown, the increase being largest for cesium and smallest for sodium. The gross structures of the phosphorescence spectra of these complexes are similar to that of the free crown, but the com- plexes have somewhat different fine structure. The high energy band for the crown is a broad, rounded hump (as noted above), whereas there are two peaks in this region for each of the alkali metal complexes (two peaks are also seen for naphthalene, naphthalene derivatives 4 and a, and, as will be pointed out in further detail 127 below, for other naphthalene derivatives and salt-crown complexes). The relative intensities of these two peaks change in going from the sodium complex to the cesium complex. For the sodium complex, the higher energy peak is more intense than the lower energy peak, whereas, for the cesium complex, the higher energy peak is less intense. The 0-0 phosphorescence band and the entire spectrum of each of these complexes are blue shifted relative to free crown 1. Comparison of Figure 26 to Figures 38, 42, and 48 or reference to the results section on energy shifts will show that all other complexes of this crown which were investigated also have 0-0 phosphorescence bands which are blue shifted relative to free crown 1. Also, it is generally true that entire phosphorescence spectrum of each complex is blue shifted relative to free crown l. (Curves C, D, and E of Figure 42 are exceptions.) For crown 2, the potassium, rubidium, and cesium complexes have fluorescence spectra which all have similar fine structure but which appear to be somewhat different from free crown. There is a general muddling of fine structure and an increase in fluorescence intensity for these complexes. A comparison of Figure 27 to Figures 33, 39, 45, and 51 will show that most of the complexes investigated show a similar muddling of fine structure and increased fluorescence intensity relative to free crown. Note that crown 2 and the related methoxy- methyl derivative 3 also exhibit less well defined fine structure than either naphthalene or l,8—dimethylnaphthalene (See Figure 21). The spectrum of the sodium complex of crown 2, however, is similar to that of free crown. But see the note on complexation of crown z 128 with sodium in the results section on quantum yield titrations. The 0-0 fluorescence bands of these complexes are slightly red shifted relative to those of free crown 2. Comparison of Figure 27 to Figures 33, 39, 45, and 51 or reference to the results section on energy shifts will show that all other complexes of crown 2 which were investigated also have 0-0 fluorescence bands which are red shifted relative to free crown. Though the fine structure of the fluorescence spectra of these complexes is muddled, there appears to be a general red shift of the entire spectrum of each complex relative to free crown 2. Note that this general red shift contrasts with the general blue shift of at least the 0-0 band for complexes of crown l. The phosphorescence spectra of the alkali metal complexes of crown % show decreased intensities relative to free crown. As is the case for the fluorescence spectra, the fine structure of the complexes is less pronounced than that of the free crown, though the difference is qualitatively less than the difference between the fluorescence spectra of the complexes and free crown. The high energy region is most affected by complexation with alkali metal cations. The crown itself has two reasonably well defined peaks in this region (at approximately 475 and 487 nm) which become less well defined for the potassium, rubidium, and cesium complexes. This is the converse of what is observed in the phosphorescence spectra of the alkali metal cation complexes of crown ’1’: the high energy region for crown l (a broad, rounded hump) is less well defined than the high energy region of its complexes (two peaks are evident; see 129 Figure 26). Thus, the fine structure of the phosphorescence spectra of alkali metal cation complexes of crowns l and 2 end up having a fine structure pattern which is more similar than that for the crowns themselves. The 0-0 phosphorescence band and the whole spectrum of each alkali metal chloride complex of crown 2 are all red shifted relative to the free crown. Inspection of Figures 28, 34, 40, 46, and 52 or ref- erence to the results section on energy shifts will show that the same is true of all investigated complexes of crown'%. Note that the phosphorescence spectrum of the sodium complex of crown'g is very similar in intensity and energy to that of free crown. This is also the case for the fluorescence spectra (Kid; .1222)- The fluorescence spectra for the alkali metal cation complexes of crown'g are all decreased in intensity relative to the free crown, the decrease being about the same for the sodium and potas— sium complexes, greater for the rubidium complex, and very much greater for the cesium complex. See Figure 47 for an expanded fluores- cence spectrum of the cesium complex. The phosphorescence spectra are decreased in intensity for the sodium and potassium complexes, about the same for the rubidium complex, and increased for the cesium complex, relative to free crownte. In contrast to crowns l and g, the alkali metal cation complexes of crown % have fine structure 'which is very similar to that of the free crown. Also, energy shifts of the spectra of the complexes relative to free crown are sufficiently small so as not to be evident from the figures. 130 (See results section on energy shifts.) Comparison of Figures 29 and 30 to Figures 35, 36, 41, 42, 47, 48, 53 and 54 will show that all of the complexes of crown,3 investigated have fluorescence and phosphorescence spectra which have fine structure which is very similar to those of the free crown. Also, in general, energy shifts relative to free crown a are small. The effects of excess alkylammonium chloride salts (ammonium, Infpropyl,‘ifpropyl, and tybutyl) on the fluorescence spectra of crowns l (S-fold molar excesses of salts), % (S-fold molar excesses of salts) and 3 (100:1 molar excess of ammonium chloride, 20-fold molar excess of Erpropylammonium chloride) are given, respectively, in Figures 31, 33, and 35. The corresponding set of phosphorescence spectra are given in Figures 32, 34, and 36. The fluorescence spectra of the alkylammonium chloride complexes of crown 1 all are decreased in intensity relative to free crown. The decrease is largest for nfpropyl and smallest for tfbutyl. Note that the fluorescence spectra of these complexes fit the gen- eralizations about fine structure and energy shifts made for com- plexes of crown.l in relating the results for the alkali metal chloride complexes. The phosphorescence spectra of the alkylammonium complexes of crown l are all of higher intensity than the free crown, with the 27 propylammonium complex showing the largest increase. The fine structure of the phosphorescence spectra of these complexes is similar to that observed for the alkali metal cation complexes (See Figure 26). The difference in relative intensities for the 131 two highest energy peaks (at approximately 463 and 472 nm) is much larger than for the alkali metal cation complexes, but otherwise the fine structure is quite similar. The 0-0 phosphorescence bands of these complexes are all blue shifted relative to free crown 1. Also, the whole spectrum of a complex is blue shifted except for the gfpropyl complex, which has many bands-which are red shifted. The fluorescence spectra of the ammonium, nfpropylammonium, 'ifpropylammonium, and tfbutylammonium complexes of crown % are all increased in intensity relative to free crown. This contrasts with the decrease seen for complexes of crown l. The increase is largest for the nfpropyl complex and smallest for the tfbutyl complex. Note that the fluorescence spectra of these complexes fit the generaliza- tions made above for the alkali metal cation complexes of crown 2. The phosphorescence spectra of the alkylammonium complexes of crown 2 are all decreased in intensity relative to free crown. This contrasts with the increases seen for complexes of crown‘l. The decrease is greatest for the nrpropyl complex and smallest for the tfbutyl complex. All these complexes show fine structure in their phosphorescence spectra which is similar to that of free crown 2 and which is similar to that of the alkali metal complexes of crown % (See Figure 28). The fluorescence spectra of the ammonium and‘nfpropylammonium chloride complexes of crown % are only slightly diminished in intensity relative to free crown. The fine structure of the fluorescence spectra of these complexes is essentially the same as that of the 132 free crown. Small but discernible spectral energy shifts are ob- served for these complexes relative to free crown. The 0-0 band and whole spectrum of the ammonium complex are blue shifted slightly relative to free crown, as is the case for the alkali metal cation complexes of crown 2. The 0-0 band and whole spectrum of the £7 propylammonium chloride complex, however, are slightly red shifted. The converse is true for the phosphorescence spectra of these com— plexes: the 0-0 band and whole spectrum of the ammonium complex are slightly red shifted (as is the case for the alkali metal cation complexes), whereas the 0-0 band and whole spectrum is slightly blue shifted for the gfpropylammonium chloride complex. The differences in spectral shapes and intensities of the phosphorescence spectra of these complexes are very small. The effects of excess bromoalkyl and iodoalkylammonium chloride salts on the fluorescence spectra of crowns 2 (5-fold molar excesses of bromoalkyl and 20:1 molar excesses of iodoalkylammonium chloride salts), 2 (S-fold molar excesses of bromoalkyl and 20-fold molar excesses of iodoalkylammonium chloride salts), and‘e (20—fold molar excesses of bromoalkyl and iodoalkylammonium chloride salts) are shown, respectively, in Figures 37, 39, and 41. For crowns 2 and 2, the bromoalkyl work used 2.00 x 10-4.§_crown and the ammonium chloride work which was done later, used 1.00 x 10'4 g crown. The crown reference spectra were of approximately the same arbitrary intensity in both cases, however, so the comparison of the relative emission intensities of the bromoalkylammonium chloride complexes to those of the iodoalkylammonium chloride complexes is qualitatively 133 correct. The corresponding set of phosphorescence spectra are shown in Figures 38, 40, and 42. The fluorescence spectra of the B-bromo- and B-iodoethylammonium and y-bromo- and y-iodopropylammonium chloride complexes of crown 2 are all decreased in intensity relative to free crown. In each case, the decrease is larger for the y-halopropylammonium complex than for the B-bromoethylammonium complex, with the largest decrease being for the -iodopropylammonium complex. The fluorescence spectra of these complexes of crown.2 look remarkably like the fluorescence spectra for other complexes of this crown which are reported here (Figures 25, 31, 43, and 49). The 0-0 band of each of these come plexes is somewhat blue shifted relative to free crown. The shift is smallest for the y-bromoammonium complex and larger by approxi- mately the same amount for the other complexes. The whole fluores- cence spectrum of each of these other complexes is blue shifted relative to the fluorescence spectrum of the y-bromoammonium complex. For other complexes of crown 2 it is not possible to tell whether or not the whole fluorescence spectrum follows the blue shift of the 0-0 band, since the spectral shapes are all similar and the magnitudes of the 0-0 band shifts are all similar. The phosphorescence spectra of the haloalkylammonium chloride complexes of crown‘2 are all increased in intensity relative to free crown. In each case, the increase is larger for the y-halopropyl than for the y-haloethylammonium complex. At first sight, the fine structure of the phosphorescence spectra of these complexes appears to be much different than that of the other complexes of crown 2 134 referred to previously. However, comparison to Figures 20, 26, and 32, for example, suggests that the prominent peak at approximately 495 nm for the B—bromoethyl 480 nm for the Y-bromopropyl, 487 nm for the B-iodoethyl, and the shoulder at 480 nm for the y-iodopropyl- ammonium chloride complexes may correspond to the less prominent peaks or shoulders in this region of the other phosphorescence spectra. With this in mind, the high energy sections of the phos- phorescence spectra of these complexes may be considered to be similar to those of other complexes of both crowns 2 and 2. As in the phosphorescence spectra of other complexes of crowns 2 and 2, there is a marked change in the relative intensity of the two highest energy peaks. In these cases, the peaks in the 470 to 475 region are of about equal intensity for the y-bromo, B-iodo, and y-iodo- alkylammonium complexes, but the intensity of the 0-0 bands change. The 0-0 bands are least intense for the Y-halopropylammonium complexes, the 0-0 band for the y-iodo being less intense than that for the y-bromopropylammonium complex. The 0-0 bands for the B-haloethyl- ammonium complexes are more intense than for the y-analogs, the 0-0 band for the B-iodo complex again being less intense than for the B—bromo complex. All the phosphorescence 0-0 bands of these com- plexes of crown 2 are blue shifted relative to the free crown, as is also the case for other complexes of this crown (vide supra). The bands in the 510, 550, and 600 nm region for the y-halopropyl and B—iodoethylammonium complexes, however, are all red shifted rela- tive to free crown. The bands at approximately 540 and 585 nm for the B-bromoethylammonium complex, however, are blue shifted. The 135 relative intensities and energies of bands in the phosphorescence spectrum of the B-bromo complex are qualitatively different from those of the other haloalkylanmonium complexes of crown 2. The fluorescence spectra of the haloalkylammonium chloride com- plexes of crown'2 are all increased in intensity relative to free crown. The increase is largest for the Y-halopropylammonium com— plexes. The fluorescence spectra of these complexes have the same muddled fine structure as other complexes of crown 2, as was pre- viously noted to be the case in general. Also, note that the red shifts of the 0-0 bands and the general red shift of the whole spec- trum of each complex relative to free crown 2 fits the generaliza- tion made previously. The phosphorescence spectra of the haloalkylammonium chloride complexes of crown 2 are all decreased in intensity relative to free crown, the decrease being greatest for the y-iodopropyl complex and smallest for the B-iodopropylammonium complex. In contrast to the differences seen for the phosphorescence spectra of the haloalkyl- ammonium complexes of crown 2 compared to other complexes of crown 2, the phosphorescence spectra of the haloalkylammonium complexes of crown‘2 are quite similar to other complexes of crown'2. Again, the changes in relative intensities of various bands are largest for the two highest energy bands. The 0-0 band and whole spectrum of each of these complexes of crown'2 are red shifted, in accord with the generalization made previously. For crownle, the fluorescence spectra of the haloalkylammonium chloride complexes are all decreased in intensity, the decrease 136 being much larger for the iodo than for the bromoalkylammonium com- plexes. The fluorescence spectra of these complexes all have essen~ tially the same fine structure as for the free crown and spectral energy shifts are small. This is in keeping with the previously made generalizations for complexes of crown 2 (vide supra). Howh ever, as is also the case for theinfpropylammonium complex, there is a discernible red shift of the 0-0 band and of the whole spectrum for the y-bromo and y-iodoammonium complexes relative to crown 2. And, as is also the case for the ammonium chloride complex, there is a blue shift for the B-bromo and B-iodoammonium chloride com- plexes relative to free crown. The phosphorescence spectra of the haloalkylammonium chloride complexes of crown 2 are all approximately of the same intensity as the free crown, except for the spectrum of the B-bromopropyl complex, which is approximately twice as intense. In contrast to the fluores- cence spectral energies, the phosphorescence spectral energies are essentially unchanged. Also, there are only small changes in fine structure for the complexes relative to free crown 2. Figures 43, 45, and 47 compare, respectively, the effects of cesium chloride and barium bromide on the fluorescence of crowns 2 (5:1 molar excess cesium chloride, 10:1 molar excess barium bromide), 12 (5:1 molar excess cesium chloride, 10:1 molar excess barium bro- nude), and 2 (5:1 molar excess cesium chloride, 50:1 molar excess barium bromide. The corresponding set of phorphorescence spectra are found in Figures 44, 46, and 48. For crowns 2 and 2, the crown concentration was 2.00 x 10”4 y for the cesium chloride work 137 and 1.00 x 10-4 g for the barium bromide work. For crown ,3, the 4 M. The spectra for the cesium crown concentration was 1.00 x 10- complexes are also given elsewhere (Figures 25 through 30). The barium bromide and cesium chloride complexes of crown 2 both have much lower fluorescence intensities than the free crown. The relative fluorescence intensities of the barium and cesium com- plexes as shown in Figure 43 are significantly more different than their relative quantum yields, which are essentially the same (vide infra). This is primarily due to the different absorbancies of the complexes in the region in which they were excited. Note that the fine structure of the fluorescence spectra of these complexes is essentially the same. Spectral energy shifts are not easily discerned from Figure 43, but see the results section on energy shifts for precise values. The phosphorescence quantum yields for these two complexes are also essentially the same, a conclusion one would not arrive at from the relative phosphorescence intensities shown in Figure 44. The 0-0 bands and the whole spectrum of both complexes are blue shifted relative to free crown and the spectra of the com- plexes have fine structure which is very similar. As in the case of the barium bromide and cesium chloride complexes of crown 2, the fluorescence and phosphorescence quantum yields of the barium and cesium complex of crown.2 are more similar (approxi- ‘mately the same, vide infra) than the relative spectral intensities ‘would indicate. The fluorescence spectra of these two complexes ‘have the same muddled appearance as do other complexes of this crown. Also, as for other complexes of this crown, the 0-0 band 138 and whole spectrum of each complex are red shifted relative to free crown. The phosphorescence spectra of these complexes are also very similar, although the whole spectrum of the cesium complex is red shifted further than is the barium complex relative to free crown 2. Whereas the intensities of the fluorescence and phosphorescence spectra of the barium bromide and cesium chloride complexes of crowns 2 and‘2 are similar, the barium and cesium complexes of crown 2 are greatly different (the fluorescence spectrum of the cesium complex is approximately 50 times less intense than that of the barium com- plex; and the phosphorescence spectrum of the cesium complex is ap- proximately twice as intense as that of the barium complex). As is usual for this crown, both fluorescence and phosphorescence fine structures and spectral energies are similar for both the complexes and the free crown. Figures 49, 51, and 53 show, respectively, the effects of silver triflate on the fluorescence of crowns 2 (SO-fold molar excess of salt),‘2 (SO-fold molar excess of salt), and 2 (lOO-fold molar excess of salt). The corresponding set of phosphorescence spectra are found in Figures 50, 52, and 54. The fluorescence intensities of the silver triflate complexes of crowns 2, 2, and 2 relative to free crown are as follows: some- what less intense for the complex of crown 2; of approximately equal intensity for the complex of crown 2; but of greatly decreased in- tensity for the complex of crown‘2 (decrease is similar to that for the cesium complex of crown 2). And the phosphorescence intensities 139 relative to free crown: moderate increase for complex of crown 2; small decrease for complex of crown 2; and very large increase for complex of crown 2 (a much larger increase than for cesium complex of crown 2). The fluorescence 0-0 bands of the silver triflate com- plexes of crowns 2, 2, and 2 are shifted as follows relative to free crown: blue shift for comples of crown 2; red shift for complex of crown 2; and negligible blue shift for complex of crown 2, and the phosphorescence 0-0 bands relative to free crown: blue shift for complex of crown.2; red shift for complex of crown 2; and negligible blue shift for complex of crown 2. The complexes of crowns 2 and 2 show changes in the fine structure of their fluorescence and phos- phorescence spectra relative to free crown which are similar to changes seen for other complexes of crowns 2 and 2. As with other complexes of crown 2, the fine structure of the spectra of the silver complex is essentially the same as that of free crown 2. Spectroscopic Energies In the two preceding sections on absorption and emission spectra, qualitative comments concerning the direction and size of spectral energy shifts for naphthalene derivatives relative to naphthalene and crown complexes relative to parent free crown were made. Tables 1 through 7 gather together precise values (in cm-l) for frequencies of the 0-0 bands of absorption (S1 and $2) and emission (fluores- cence and phosphorescence) for naphthalene (10), naphthalene deriva- tives (compounds 2 through 2), crowns 2, 2, and 2, and complexes «of these crowns. Also, values for energy differences (AE, in cmfl) 140 .auau On “some couonaau you mu ousououuunu .mumaw A>\> .Hucv aoauSuoEIHoamnuu aouu ma cows: 0 '80 “OH” .Acowmmaomav «may umuoco one usuu saw no massage» as aw mash“ .mmmaw Hocozuu «ma voxuuuunz aouu mus nuav aoumauao Ha\> .Hucv Hocosuoanaoausuo . .au on"; .H.au o~um Nam Scum a“ sump coauauompo HHuuwuon wumaum v.mcoamnuam u.oco«unuomn< accuumuanom ~o>og zwuocm cw mmwcmnu .x up an mommuao Honoua< ca mocoawsunnmz vousuwuupaawntn.a van .Im.H .un.~ now Aauau Ca Macy mcouumumamm Ho>oa amuoam cw mownocu van nouocosvoum coauowau van cowunuooa< .H manna .Auxou owmv mwcmu m.¢loa may ca ma cowumuucoocou caouo .mmme Hocmsuoa IHocmnum Nmm Scum mum muwm aofiuauomnm nonuo Ham “macaw A>\> .Hucv Hocmaumalaosmsum Scum mum mmmenaoo oumawauu um>aam was ocaaoun spawns you mumv cowuouomn¢ .H wanna you a swaounu c can .A .m mmuoc moms omH- o- an as NNeHN semHm owNHm QHmsm mmomomow< + w on- o~H oh Ne mHmHN eeuHm moNHm oowqm Human + w cc- omH oHH om wenHN HHMHm NomHm Osman Home + a ow- ONH om oe mHmHN momHm eemHm ONwem Hone + w oh- ceH OOH OOH oemHH OOMHM eemHm Homqm Hoe + w 1 “m OMH omN 00H oHH «NoHH aOMHm mHmHm semen Homz + w o o o o ooeHN oONHm oNNHm mmeem vaounsouonm.N H -H o -H o -H o -N H a mmv< H m Hqu H m mqu H m mmva Ho.ov .moee Honey .posHm Ho-ouHmVpH HNmeH a H cacao ou m>fiumaox mcowumumaom Ho>m4 hmuoam ca mmwamno aowuauomn< m.H.au sH mumnasz HH< .mmmHo HosooH< :H x mm um A&V eunzououowuonunewz um.~ mo mmmeaaou mumHHHua um>HHm was .mcHaoum asHumm .meHuoHso Hess: HmeH< you H~.eaouo mmum ou m>HumHon maoaumummmm Ho>mq hwpmam :H mmwamgu can moaocosvoum cowmmfiam can sowuauomn< .N manna 142 .Auxmu momv owcmu m «IOH aw ma coaumuucmoaoo :Bouo .mmmaw A>\> .H»¢v Hoamnuoa IHosmnuo Nmm Scum mum sumo coaunuomnm posuo Ham “mmme A>\> .Huev HoamSumsnaoamnuo Eouw mum moxmaaaou .H manna aw : nwsounu v can .u .m mmuoa 00mm mumHMHHu um>afim was ovaaoun aswumn mnu you name cowuauomn< Om OHH- Oe- Oe- HOOOH OHOHO ONOHO OHOOO OOONOOOO< + w OO OOH- OOH- OH- OHOON NOHHO ONNHO HHOmm Human + w OO OOO- OsH- Os- ONOOH OQHHO «HmHm HOOOO HOOO + w OHH OHO- Oem- OH- OOOON OOHHO OHOHO OOOOO Hope + w OHH OOO- OON- Om- OHBON OOHHO BHOHO OOOOO HOM + w Om OO- Om- OO- “NOON OOOHO NHOHO NOOOO Homz + w O O O O OOOHH OOmHm HeeHm mOOmm HwVO-eaouO-O.H H -H O -H O -H O -N m H H OOOO H m HOOO H m OOOO A m OOOO HO-OO ..OOO HO-OO .uoaHm AO-OMHOOOH HHOO H N cacao ou m>wumamm mcoaumumamm cofimmwam coauauoma< Hm>oa hwumam aw mowcmnu m.H-so OH mumpasz HH< .x Be on mmmHO HO£OUH< cH va O-csouO-HN-osunamz Im.H mo mmxmaaaou mumHmHuH um>afim can .mvwaoum asaumm .mvauoanu Hmuoz Hamxa< How AW cacao moan ou mpaumammv maoauwumamm Hm>oq %wuwam ma mowamno cam mofioamsvmum scammwsm was coaumuomg< .m manna 143 .Auxmu mmmv owmmu_m.MIbH cw ma coaumuuaooaoo nacho .mmme A>\> .Huqv Hoamnuoa Iaosmsuo Nmm Eouw mum mumv coauauomnm Hosuo Ham “macaw A>\> .anV Hoawnumauaocmnum aoum mum mmxoamaoo oumHMHuu um>afim mam mpwfioua asaumn you mumv cofiuauomn< .H manna you : nwsousu v new .A .m mmuoa momm Om- ON- Om- OOH- HOOOH NBOHO ONOHO Nemmn mmONOmOm< + w OH- OH O OH- HNOHH NHOHO OOOHO HOOOO Human + w OH- Om ON Om OOOHN OOOHO OOOHO meemm HOOO + m OO ON OO Om OOOHN NOOHO ONOHO Henna Hone + w ON ON- Om Om OOOON ONOHO OmmHm NNNOO Hue + w O OO Os ON NOOHH OOOHO NHOHO NOHOO Homz + m O O O O NHOHN HHOHO ONQHO HOOmm Hva-esouO-O.H HHe-Hmmv< HOm-HOOOO HOm-HOOOO HOm-Hmmvc HO-OO moan HO-OO .uosHm HO-OMHOOOH Humva N c3ouu cu m>fiumamm wcofiumumaom Ho>m4 hwumcm a“ mmwcmcu cofimmfiam coauauoma< m.H-Eo a“ oneness HH< .mmmHO HOeOOH< aH x he as va O-csouO-NN Im.H «0 momeano oumamwua uo>awm was .mvwaoum Eswumm .okuoasu Houmz Hamxfl< How A ovum ou o>wumammv maowuwumamm Ho>oA hwumam a“ mmwcmso was mmaocmsvmum cowmmwam cam coauapomn< -onu:Omz m nacho .q manna 144 .Auxou momv owcmu_m.cIOH mnu cH mH cOHumuuamosoo cacao .mmme A>\> .Hu¢v HocmsuoalHosmsum Nmm Scum mum mumv :OHuauomnm nonuo HHm "mmmHm A>\> .Hqu HoamSHoalHoamnum Bonn mum H nacho mo mmmen Laoo mvHHOHco 55Hsoaam HthmovOH mnu how sump GOHuauomn< .H mHan you H nwsounu m van .9 .m mmuos momm OO- OOH OO Omn- OOOHN OONHO HOHHO HONOO HommzmHNOOOH+w OH- OHH OOH OHH OHOHH NOOHO OONHO HHOHO HummzHHNOOOH+H OOH- OOH Os OHn- OHOHN HONHO OOHHO OQHOO HOOmzmHNOOOuO+H OO- OOH OO Om OHOHN HONHO OHOHO OOOOO HOOOzNHNOOOuOHH OO- OHH OO Om OOOHH OONHO ONOHO OOqu Hummzsmnm.+ H OO- OOH OO Om HHOHN OOOHO OHOHO OOOsO HOmmzHOLH.+.w OOH- OOH Os OOO- OOOHH HONHO HONHO mHOeO Hummzumnm.+ H OsH- OHH OO OH HHOHN OOHHO HHOHO «NOQO HOHmz + H O O O O OOOHN OONHO OHNHO OOHOO quO-aaouO-m.~ N H O - mmv< H m mmOO H H o H o H H- mm: H O. OH: H m- HO-OO .OOOH HO-Ov .uOOHO HO-OHHOOOH HNOOmH & azouu ou m>HumHmm mGOHumumamm Ho>oH “whoam cH mowcmnu GOHmmHam aOHunuomA< m.HIEu CH mumnasz HH< .mmmHo HOOOOH< :H M HH um A&v olcsouolcmlonunamZIm.N mo mmwaano mnHHOH£o abHaoEEmHthmOHmm paw Hth< How a& cacao omum ou o>HumHomv maoHumummmm Ho>oH hwuoam :H mowcmso can moHoaosuoum GOHmmHEw can aOHuauomp< .m oHnma 145 .Auxwu momv mwcmu «IOH mnu cH mH aoHumuucmocoo naouo .mmem A>\> .Huc Hoamnuma IHoamnum Nmm Eoum mum mumv cOHuauomnm nonuo HHm “mmme A>\> .Huqv HocmnuoalHocmnum scum mum asouu mo mmwaaaoo mvHHOHno EHHHcofinm Hhu—HmovoH on“. now mummy COHunuomflw .H 3an .Hom s 595.23 u can 3 .m mono: moms ON- OON- ONN- OO- OOOON HHHHO HOOHO HNOOO HommzmHmmOOH+w OO- OO- OHH- OO- OOOON HNNHO HHOHO OHOOO HOmmzNHNOOOH+w OO+ OHN- OON- OO- HOHOH OOHHO OOOHO HNOOO HOmmzmHNOOOumaw OHH- OOH- OON- Om- OOOON OOHHO HOOHO OOOOO HOOmzHHHmOOuO+w OH- OH- Om- Om- OOOHN «OOHO HOOHO mmOmm HOOOzOOLm.+ w OH- OH- OHH- OO- OOOON OONHO OOOHO HOOOO HOmOzumLH.+ w OOH- ONH- ONH- Om- OOOOH HHHHO OHOHO HOOOO Hommzsmnm.+ H OO- OO- OOH- Oe- OOOON NOHHO HmmHm OOOOO HOOmz + w O O O O OOOHN OOOHO HOQHO OOOOO va O-naouo-O.H Hake: HOO-HHHOO HOm-HOEO HOO-Hmmz 8.8 .85 8.8 8:: HO-OmHm} mm} & aBoHo ou m>HuMHmm chHumumaom :OHmmHam QOHunuomn< Hm>oH hwuocm :H mmwcmno m.H-ao :H muonssz HH< .wmmHu HonoUH< aH M um um Hwy elaaouolHulonunamzuw.H mo mmwanaou ovHuoHnu asHaoaamH%meonm was Hka< now Aw nacho moum ou m>Humemv maOHumumoom Hw>mH >wumsm aH mmwcwno van moHoaosvmum aOHmmHsm can aoHuduomn< .w oHnms 146 A>\> 0Hu¢v HOfiQSHOBIHOd—NSUO BORN 0H0 mumv SOHHQHOmflm HH< .Auxou oomv omamu_quIOH mnu 6H mH :oHumuucmucoo nacho .mmme .H oHan you : cwsou5u v cam .A .m mouoa mmmm OH- OH Om- OH mmOHm OHmHm HHOHm HOOmm HOmOzmHmmOOH+~ OH OH Om Om mmOHm OmeHm OOOHm OHHmm HOmmzmHmmOOH+m Om- OH OO- OH mmOHm OmmHm OHOHm HOOmm HommzmHmmOOuetw Om Om- Om OH- OOOOH OmsHm OOOHm mmomm HOmmzmHmmOOum+~ OO- OH Om- OH mmOHm OHmHm mOOHm HmHmm HOmmzuOLm.+ m OO OH- Om Om OOOOH OmeHm OOOHm OHHmm Hoemz + w OO OO O O NHOHN OmOHm OHOHm mmOmm Hwo O-caopO-m.H HHH-HmmOO HOm-HmmOO HOm-HmmOO HOm-mmmOO HO-OO .OOOO HO-OO .uOOHO HO-OHHOOHH HmmOmH % nacho ou o>HumHom m:0Humumamm Hm>oH hwumcm :H mmwamao GOHmmHBm coHumuomA< m.H:au :H mumnasa HH< .OOOHO HonouH< aH e Hm um Awu ouczouonmuaonunnmz-n.H mo mmmeosoo ovHuoHso EchoaammeHmonm was meH< you AW cacao moan ou o>HumHomv maoHuouwaom Ho>mH hwuocm aH mowcmsu van moHuawsvoum :onmHEm was coHuauomg< .5 meme 147 between the 0-0 bands of naphthalene and those of naphthalene derivatives and between the 0-0 bands of crowns and those of their complexes are given. These differences are given for S2 0-0 ab- sorption bands (A(Es2 - 50)), 0-0 fluorescence bands (A(ES1 - so), and 0-0 phosphorescence bands (A(ET1 - so)). Differences in the singlet-triplet energy gap (E51 - Tl) between naphthalene and naph- thalene derivatives and between crowns and their complexes are also given. These differences (A(Esl - T1)) indicate whether the singlet- triplet gap gets larger or smaller for derivatives of naphthalene relative to naphthalene or for complexes of crowns relative to free crowns. A positive number indicates that the energy separation increases. See the general introductory comments made in the results section on UV spectra for definitions of the terms "band", "0—0 band", "81", and "82". Here we use the term "0-0 band" to refer to the lowest energy (longest wavelength) peak in the $2 band. The table headings also give the Platt68 notations for the S1 and 82 bands: "La", for the $2 band; and "Lb", for the S band. Discussion of 1 whether or not the values given are for true O-O bands and special discussion of the 0—0 phosphorescence band for crown.l will be given later. Table 1 gives the errors appropriate to each kind of frequency measurement and, also, errors appropriate to the changes in energy separations derived from them. While most of the absorption data are from 95% ethanol-methanol (4:1, v/v) glass, some of the later ommsu cu m>HuMku mum mpHmHh Educmsv ponuo HHd .Auxmu mmmv mosHm> vmsbmmHumHmu vmumEHummm .m£uwcmHm>m3 acmummqu woman an owns mcom IHumgaoo pHmH> asucmsu mo mommum>m you mGOHumH>mp pumpcmum mum mvHoH> asuemsu How poumoHpaH muouuo snap .mmme A>\> .Huqv HocmcumEIHocmSum Eouw mum :oHcs .W van Q How omosu uaooxm mmme Hocmsuo Nmm Eoum mum mmsHm> up HH Figure 57. Titration of 2.00x10-4'§_l,8-naphtho-21-crown-6 (z) with alkali metal chlorides in 95% ethanol at 77 K followed by monitoring integrated fluorescence intensity. 156 Perturbed l,8- NODIVZVCI"6 ‘I’phos. A 0.075 4 NoCl A ‘ 4 CsCl 0.050 KCI RbCI q"phos 0.025 '- 0'00 O I 2 3 4 5 —— MCI /Cr0wn ——§ Figure 58. Titration of 2.00x10-4 g 1,8-naphtho-21-crown—6 ( ) with alkali metal chlorides in 95% ethanol at 77 followed by monitoring integrated phosphorescence intensity. 157 0.10 \ ‘I -‘o fx,‘ \ 0.08 . 3““ ~A \\ “ H __~ - _. ‘e‘ A ----'+NOCI A ‘0~ - - ‘ ' “’ ”"---8-KC| 0.06 '- ¢f|uor °-°‘ " © 6’} 0.02;- 4 RbCl £0035 , .0021 ‘g .0020 0.00 - CsCl O 1 2 3 4 5 -—-—- MCVCrown —-—9 Figure 59. Titration of 1.00x10-4 fl 1,5-naphtho-22-crown-6 Q) with alkali metal chlorides in 95% ethanol at 77 K followed by monitoring integrated fluorescence in- tensity. 0.4 0.3 0.2 4’phos 0.l Figure 60. 158 + CsCl 9 © ’r—‘r--' F?tWC” 3 T NOCI % KCI J 41 .1 J .11 I0 20 3.0 4.0 5.0 —MCI/Cr0wn ——a- Titration of 1.00x10-4 )1 1,5-naphtho—22-crown-6 (,2) “with alkali metal chlorides in 952 ethanol at 77 K followed by monitoring integrated phosphorescence intensity. 159 or even due to complexation of rubidium. The relatively sharp breaks at the 1:1 salt to crown ratio are .important features of these figures (which are not due to reexpression of the integrated intensities in terms of quantum yields). These sharp breaks indicate the formation of 1:1 complexes with high (>10S complexation constants. The apparent degree of decrease (or increase) of the quantum yields of complexed crown relative to free crown is meaningful in Figures 55 through 59 only because the data have been reexpressed in terms of carefully determined quantum yields of completely free and com- pletely complexed crown. For example, the decreases in ¢f for the alkali metal complexes of crown 1 relative to free crown 1 shown in Figure 55 are less dramatic than the decreases in the correspond- ing integrated fluorescence intensities. This is because the com- plexes all absorb less strongly than free crown in the region in which they were excited (approximately 306 nm). The remainder of the titration data is presented in tabular form. Since, as remarked previously, the main purpose of these titrations is to establish the molar excess of salt required to achieve essen- tially complete complexation of the crown, it is only necessary that either the fluorescence or phosphorescence intensity be monitored as a function of added salt, whichever happens to be experimentally easier to do. For the ammonium, alkylammonium, and bromoalkylammonium com- plexes of crowns‘l and.%, sufficiently accurate integrated intensi- ties are not available (see the experimental section for the variety of ways integrated intensities were obtained). For these complexes, 160 relative peak intensities (complex intensity/free crown intensity) are tabulated. ‘See Tables 9, 10, 12, and 13. The data are somewhat more scant than for other cases (relative intensities are only available for 1:2, 1:1, and 5:1 salt to crown ratios), but in most cases the relative peak intensities are approximately the same for the 1:1 and for the 5:1 salt to crown ratio. In a few cases (ng;, ammonium chloride complex of crownll), there is a somewhat larger difference. But, at the crown concentration used (2.00 x 10-4 34), even if there were 20% dissociation for the 1:1 point, there would be only 2% dissociation at the 5:1 point. The tabulated relative peak intensities are referred to as "arbitrary relative peak intensi- ties" in the tables because they depend not only upon the excita- tion wavelength and band passes used but also on which peaks are compared. Table 11 gives relative integrated emission intensities for crown % as a function of the mole ratio of ammonium and gfprOpyl- ammonium chlorides to free crown. Tables 12, 13, and 14 give rela- tive integrated emission intensities for crowns 1,42, and‘e as a function of the mole ratio of iodoalkylammonium chlorides to free crown. Table 14 also gives relative integrated emission intensities for crown g as a function of the mole ratio of bromoalkylammonium chlorides to free crown. Tables 15, 16, and 17 give relative integrated emission intensities for crowns l, 2, and g as a function of mole ratio of barium bromide (not for crown 2) and silver tri- flate to free crown. The integrated intensities reported in these tables have been made proportional to the quantum yield of free 161 .mme mcomHumaBou mo Honese ozu mH mommnuamuma :H Hones: osu ”maoHumH>mv pumpemum mum am>Hw muouum 0:80 .xmma oneEoo mnu mo huHmemucH ocu kn pmpH>Hp xmwa cacao o:u mo huHmaoueH onu mo mauou cH so>Hw mH zuHmeouaH m>HumHou 0:9 .pmm: mm3 Home moeoomouosHm nuwcon>m3 umwuuonm one a .m.HumHom zumuanu< OHHOO-HO-O.HHHHOO .moHuHmaoueH xmmm oucoomouosHm o>HumHom wcHuouHeoz mp ooeoHHom M um um mmmHo Hoamnum Nmm poxomuu -OO OH OHHOO OOHHOHOO OOHOoOaOHHHHO OOO OOHOoaOO OHH; va O-OOOHo-Om-OOHOOOz-m.~ Ho OOHHOHHHH .O mHOOH 162 .uxmu mom .muuooom.>p M mm :H mowcmso kn omcmHHnmumo COHummeoEou mo mooumo .moma maomHuma I800 mo Hones: may mH memosuamuma :H Hogan: mnu “coHumH>oo oumocmum mcu mo mahou 6H cm>Hw mH uouum 0:90 .meoaoo msu mo kuHmcmucH man an oooH>Ho Home e3ouo mnu mo huHmaoueH mnu mo mfipou :H ao>Hw mH >uHmamucH 0>Hu0Hou 0:9 .voms was xmma moamommuosamoca numaon>m3 umouuonm 0:9 a HM «IOH x oo.N mH Amusumumasmu Boouv GOHumuuemucoo HmHoZm HNV ~o.o H wO.o Hmv Ho.o H mq.o HOV no.0 H Hm.o HOV ~o.o H Hm.o H\m HNV Ho.o H oo.H HNV Ho.o H me.o HNV mo.o H om.o Amo Ho.o H Om.o H\H HNV Ho.o H No.H Hmo Ho.o H ow.o Hmv no.0 H mo.o OHNV Hmo.o H Hn.o ~\H oo.H oo.H oo.H oo.H osoz OHOmszme HOmOZHOLH HommzHOLm Huqmz OHwoO-HO-O.H\HHOO nzuHmcouaH 0>HumHmm humuanu< .mmHuHmamuaH xmom momma Imouonomosm 0>HumHom waHuouHoo: an oo3oHHom M um um mmmHo Hoamsum Nmo uoxomuoco cH mooHHOHnu OOHOoOOO HHHHO OOO §HOO§O OHH: H8 O-§OHO-HH-OOHOOOz-O.H ..:- .H-OH H OO.~ Ho OOHHOHHHO .OH OHOOH Table 11. 163 Titration of 1,5-Naphtho-22-Crown-6 ) with Ammonium and n-Propylammonium Chlorides in Uncracked 95% Ethanol Class at 77 K Followed by Monitoring Relative Integrated Fluorescence Intensities. Salt/1,5-Cr—6 (3)3 NH4C1 ‘anrNH3Cl None 0.114 0.114 5/1 0.102 1 .002 g§)b 0.090 t .002 (2) 10/1 0.097 t .001 (9) 0.089 t .005 02) 20/1 0.089 t .002 (g) 0.091 1 .003 (2) 50/1 0.089 1 .002 (Q) ----- 100/1 0.091 t .003 (a) ---- a - Molar crown concentration (room temperature) is 1.00 x 10 4:5. b The error given is in terms of the standard deviation. The number in parentheses is the number of comparisons made. 164 .moHuHmcmuaH 0>HumHmu omumuwmueH Eoum mum muHSmom .m.qloH x oo.H mH Amusumumosmu Eoouv COHumuuamocou nacho .HmHozHo .moma mcomHumm 1800 m0 Hones: onu mH mommSHGmumo cH Manama 0:9 .cOHumH>oo unmoamum onu mo magma aH a0>Hw mH uouum 0:90 .mcowumcHEhmuoo mechn .eBOHo mmum man «0 xmma numemHm>ms ammuuonm msu mo huHmemucH ocu up omoH>Ho meanu mnu no #000 mofimommuosHm zuwcmHm>m3 umouuonm osu mo huHmamuaH mzu mo mahmu CH am>Hw mum muHsmmm .m.cIOH x oo.~ mH Amnsumumoamu Eoouv cOHumHuemoeou aeouo HmHozm HOV HOO. H Om0.0 HOV HOO. H Hm0.0 -- -- HHON HOV HOO. H Om0.0 HOV HOO. H OH0.0 -- -- HHOH Hmv HOO. H Om0.0 HOV HOO. H OH0.0 HOV H0.0 H O0.0 mH.O H\m HNO HOO. H Hm0.0 HOV HOO. H OH0.0 HOV H0.0 H ON.O mH.O H\~ HOV mOO. H mO0.0 HOV HOO. H mOO.O HOV HO.O H O~.O -- OHH HOV HOO. H OOH.O HOV HOO. H OOH.O onO H0.0 H Om.O OHO.O m\H O~.O mH.O OO.H OO.H OOOz OHOmmzmHmmovH OHOmOZNHOOOVH OHOmmzmHmmoon OHOmmzNHmmoon HHO O-HO-m.m\HHOO zuHmemucH 0>HumH0m humuanH< .mmHuHmcoueH oucmomouosHm 0>HumHmm wcHuouHcoz mp omsoHHom M um um mmmHo Hoemnum NmO OuxuOHOOO OH OOOHHOHOO OOHOOOEOHOHHOOHO: OHHs HHV O-OOOHo-Om-OOHOOOz-m.u Ho OOHHOHHHH .OH «Heme 165 .moms maomHumaBoo Ho nomads 0:» 0H monocucmuma :H Hoaasd 0:9 .:0HumH>0v onUGMHm msu mo msumu eH co>Hw mH Houum 0:90 .2 oH x oo.H 0H Amuaumumaamu Eoouv cOHHmHuaoocoo c30uo HMHoz H O .quIOH x oo.~ mH Amusumumaamu aoouv cOHumuucmocou nacho HmHozm HOV HOO. H OOH.O HOV HOO. Om0.0 -- -- HHOH HHO OOO. H NHH.O HOV HOO. H HO0.0 Hmo mO0.0 H Om0.0 Hmo HO0.0 H Hm0.0 HHO -- HOV OOO. H Om0.0 -- -- H\m.m HOV mOO. H HOH.O -- HOV OO.O H HO0.0 HNO HO0.0 H Hm0.0 HHO HOV OOO. H OOH.O Hmo HOO. H Om0.0 HHO H Hm0.0 Hmo HO0.0 H Hm0.0 HHH HHO HOO. H HmH.O HOV HOO. H HO0.0 HHO HO0.0 H OO0.0 OHNV HOO.O H mO0.0 OHH H.HOOHOV OOH.O H.OOOOo mO0.0 H.monO mO0.0 H.OOOOO mO0.0 OOoz OHOmOZmHNOOVH OHOmOzNHOOOOH OHOmmzmHmmOOHm OHOmmzmHmmOon Hwo O-HO-O.H\HHOO muHmcmuaH 0>HumHmm humuquu< .mmHuHmcmuaH GOHmmHEm omumuwouaH 0>HumHmm wcHHOuwcoz ma omaoHHom M um um mmMHu HoamSum HmO OOHOOHOOO OH OOOHHOHOO OOHOOOOO HHHHOOHOO OHHz HOV O-OsoHO-Hm-onuOOOz-O.H HO OOHHOHHHH .mH OHOOH 166 1500 mo “0:8:c 0:u mH m0mo:ua0uma CH H0:a:e 0:H .eoHumH>0o oumoamum 0:0 «0 mauou aH a0>Hw 0H uouu0 0:9 .0oms maomHumo : IoH x oo.H um m0uaumu0oa0u aoouv coHumuuc00c00 caouu HmHozm H HOV OOOO. H HHHO.O HHO HOO. H mHO.O HOV HOO. H OO0.0 HOV mOO. H HmO.O HHOH HmO OOOO. H OOHO.O H O mOO. H mHO.O HHO HOO. H mOO.O HHO HOO. H mOO.O HHOH HOV mOOO. H mmHO.O H v mOO. H mHO.O HHO HOO. H HOO.O -- H\O HHO mOO. H OH0.0 H V HOO. H HmO.O HOV HOO. H mO0.0 HHO mOO. H OO0.0 HHN HHO HOO. H mm0.0 Hmv HOO. H mOO.O HHO HOO. H OmO.O HOV HOO. H OHO.O HHH HOV HOO. H mO0.0 HOV mOO. H mHO.O HmO HOO. H mOO.O OHmV HOO. H HOO.O OHH HHH.O HHH.O HHH.O HHH.O HOoz HummzmHNOOvH HommzmHmmooH HOmHmmOoHO HOOOZHHNOOOHO kuHmamucH 0>H00H0M humuanu< Ova O-HO-m.H\HHOm .huHmaouaH 00:0000H05Hm v0umuw0ueH 0>HumH0M wcHuou :0: m: 0030HHom M um um wmmHo Hocm:um NOO OOHOOHOOO OH OOOHHOHOO OOHOOOaOHOHHOOHOm OHH3 H v O-OOOHO-mH-OOHOOOz-O.H Ho OOHHOHHHH .OH OHOOH Table 15. 167 Titration of 2,3-Naphtho-20-Crown-6 (l) with Barium Bromide and Silver Triflate in Uncracked 95% Ethanol Class at 77 K Followed by Monitoring Relative Integrated Fluorescence Intensities. Salt/2,3-Cr-6(1)a Arbitrary Relative Intensity BaBr AgOSOZCF 2 3 None 0.26 0.26 1/2 0.119 1 .004 (2)b ---- 1/1 0.063 1 .001 (2) 0.253 1 .001 (2) 2.5/1 0.064 1 .006 (2) —-—- 5/1 0.065 1 .002 (2) 0.217 1 .001 (2) 10/1 0.065 1 .003 (2) 0.180 1 .002 (2) 20/1 ---- 0.165 1 .003 (2) 50/1 ---- 0.165 1 .003 (2) 100/1 ---- 0.164 1 .001 (2) aMolar crown concentration (at room temperature) at 1.00 x 10-4 g, bError is given in terms of the standard deviation. The number in parentheses is the number of determinations made. 168 Table 16. Titration of l,8-Naphtho-21-Crown-6 (a) With Silver Tri- flate in Uncracked 95% Ethanol Class at 77 K Followed by Monitoring Relative Integrated Phosphorescence Intensity. AgOSOZCF3/1,8-Cr-6 (2)a Arbitrary Relative Intensity None 0.076 1/2 0.0748 1 .0005 (2)b l/l 0.0748 1 .0002 (2) 2.5/1 0.0742 1 .0006 (2) 5/1 0.0710 1 .001 (6) 10/1 0.0675 1 .002 (12) 20/1 0.0681 1 .003 (12) 50/1 0.0660 1 .008 (8) aMolar crown concentration (room temperature) is 1.00 x 10-4 M. bThe error is given in terms of the standard deviation. The number in parentheses is the number of comparisons made. Table 17. 169 Titration of 1,5-Naphtho-22-Crown-6 and Silver Triflate in Uncracked 95% ) with Barium Bromide thanol Class at 77 K Followed by Monitoring Relative Integrated Emission In- tensities. Salt/1,5-Cr-6 (3)3 Arbitrary Relative Intensity BaBr2 Ag0S020F3 None 0.114 (Fluor.) 0.16 (Phos.) 1/2 0.115 1 .003 (3)b 0.152 1 .004 (2) 1/1 0.110 1 .003 (3) 0.177 1 .008 (2) 2/1 0.108 1 .003 (2) ---- 5/1 0.104 1 .001 (2) 0.38 1 .01 (2) 10/1 0.0944 1 .005 (2) 0.64 1 .03 (2) 25/1 0.0837 1 .0002 (2) 0.86 1 .03 (2) 50/1 0.0837 1 .0002 (2) 1.02 1 .04 (3) 100/1 0.0846 1 .0003 (2) 1.02 1 .06 (3) 200/1 ---- 1.02 1 .03 (3) aMolar crown concentration (room temperature) is 1.00 x 10-4 fl, bThe error is given in terms of the standard deviation. in parentheses is the number of comparisons made. The number 170 crown but have not been modified in any other way. As tabulated, they provide information on the extent of complexation as a function of added salt. They are referred to as "arbitrary relative inte- grated intensities" in the tables because they depend upon the choice of experimental conditions (excitation wavelength and width of band- pass). The errors given in Tables 9 through 17 are standard deviations. The number of values averaged in each case is given in parentheses after the error. There are several special cases in which changes in both fluores- cence and phosphorescence intensities are too small to be reliable indicators of the extent of complexation. In two cases, the extent of complexation was established through competition experiments: a crown/salt mixture which has a quantum yield greatly different from that of the crown alone is titrated with the species which does not cause a very large change in intensity. Relative equilibrium constants can be calculated (see experimental section), thus allow- ing one to calculate how much salt would be necessary for complete complexation in the absence of competition, provided that there is an independent estimate for one of the equilibrium constants avail- able. The results of a competition study between sodium chloride and crown glrubidium chloride are given in Table 18. The integrated intensities are expressed relative to that of rubidium chloride/ crown 2 - 2/1, formal crown concentration at 1.00 x 10-4 M, The results from systems containing 100:2:1 and 50:2:1 mole ratios of 171 Table 18. Competition for 1.00 x 10-4 M l,8-Naphtho-21-Crown-6 (2) in Uncracked 95% Ethanol Class at 77 K by Sodium and Rubidium Chlorides Followed by Monitoring Arbitrary Relative Integrated Intensity. Arbitrary Rel. Equil. Rel. Integrated Constant System Intensity KRb+/KNa+ RbCl/g = 2/1 1.00 l,8-Cr-6 (2) 0.47 NaCl/g = 100/1 0.49 i .04 (3) NaCl/RbCl/je - 100/2/1 0.71 1 .05 (3) 50 NaCl/RbCl/g 8 50/2/1 0.8 i .1 (2) 60 NaCl/RbCl/2 8 10/2/1 0.96 i .04 (2) 100 8Errors are given in terms of the standard deviation. The number in parentheses after the error is the number of determinations. 172 sodium chloride to rubidium chloride to crown 2 show that rubidium complexes 50 to 60 times better with crown 2 than does sodium. The results from the 10:2:1 case are less reliable due to the small dif- ference in intensity between it and the rubidium chloride/crown 2 standard. The initial experiments on the sodium complex of crown 2 were only done using a five-fold excess of sodium chloride (2.00 x 10.4 M crown). The above results indicate that this would not be a suf- ficient excess for complete complexation of crown. Assuming an equilibrium constant of 106 for the rubidium complex of crown 2 (based on the essentially complete complexation with a two-fold excess of salt indicated by Figures 56 and 57), an equilibrium 4 is indicated for the sodium complex. constant of approximately 2 x 10 Thus, complexation would be only about 94% complete with a five-fold excess of salt. Quantum yield, energy, and emission spectra were redetermined with a hundred-fold excess of sodium chloride and the results were essentially the same as those for a five-fold excess. The overall changes are small, however, so this isn't surprising. The results of a similar competition study between barium bromide and rubidium chloride/crown 2 are given in Table 19. The results are somewhat rougher than for the sodium competition study, probably because of the smaller differences, but indicate that rubidium and barium have about the same complexing affinity for crown'2. The third type of quantum yield determination is similar in nature to the first. The fluorescence and phosphorescence quantum yields of fully complexed crowns were determined relative to those 173 Table 19. Competition for 1.00 x 10-4;fl l,8-Naphtho-21-Crown-6 ( ) in Uncracked 95% Ethanol Class at 77 K by Barium Bromi e and Rubidium Chloride Followed by Monitoring Arbitrary Relative Integrated Intensity. Arbitrary Rel. Rel/ Equil. Integrated Constant System Intensity KRb+/KBa++ BaBrz/z . 10/1 1.00 l,8-Cr-6 (2) 0.96 RbC1/2 - 2.5/1 0.68 i .01 (2) BaBr2/RbC1/2 = l/l/l 0.87 i .01 (2) 0.5 BaBrZ/Rb01/2 - 12.5/2.5/1 0.92 i .01 (2) 1.7 aErrors are given in terms of the standard deviation. The number in parentheses after the error is the number of determinations. 174 of the parent free crown, which, in turn, were determined relative to naphthalene (vide infra). Sufficient salt, as determined from the titrations described above, was added so that the crown would be essentially completely complexed. Emission quantum yields of the alkali metal chloride, barium bromide, and silver triflate com- plexes of crowns‘2, 2, and 2 are given in Tables 20, 21, and 22, respectively; those of the ammonium and alkyl ammonium complexes of crowns‘2,‘2, and 2 in Tables 23, 24 and 25, respectively; and those of the haloalkylammonium chloride complexes of crowns 2, 2, and 2 in Tables 26, 27, and 28, respectively. The values given in Tables 20 through 28 are the result of at least four determinations made using only one set of excitation wavelengths. (See experimental for wavelengths at which determina- tions were made.) The errors given are standard deviations and indicate the precision of the comparison of the quantum yield of the complex to that of the free crown. In many cases, they are better than the standard deviation indicated for the quantum yield of the parent crown. The absolute accuracy of the quantum yields of the complexes obviously can't be any better than the accuracy of the quantum yield to which they are referred (naphthalene, ultimately). But quantum yield comparisons between free crown and complexed crown can probably be made more accurately than quantum yield comparisons between free crown and naphthalene, since the 77 K UV spectra of the complexes are more similar to free crown than are the 77 K UV spectra of free crown to naphthalene. The fourth type of quantum yield experiment involved comparison 175 “NmH 0H0 muouuo 0>H00H0H pmumeHuwmo =.x0HoEoo moHSOH: asHuop How NoHH .nmmHm A>\> .HHHV Hocm:u081Ho:0:00 Eoum 0H :0H:3 .x0HoEO0 ooHaoH: asHum: 0:u How um:u 0000x0 mmme Hocm:u0 Nmo Baum 0H0 m0=Hm> we HH03 0HwaHm m on maomHumaaoo meHz spammed usom unmoH 00 mo 00w0H0>m How meoHumH>0o oumocmum 0H0 muouuo mounoHvaH 0:9 .% macho 00pm ou 0>H00H0H 00¢ Eu0u0o 0H03 00x0HoEo0 mo monH» saucepan .Am 0Hpma 000V 0c0Hm:u:ams ou 0>H00H0H moaHsu0u0o 0H03 cacao mo mpHlo anaconda H.m HOO.O H mOO.O -- HOO.O H mHH.O mOOHOmOOO + H H.m HOO.O H mmH.O Hm HOO.O H mH0.0 NHOOO +.H H.m HOO.O H HmH.O Om HOO.O H mHO.O HOOO + H H.m HOO.O H HOO.O Om mOO.O H HHH.O HOOO + H H.m mOO.O H HmO.O HO HOO.O H HmH.O Hoe +_H H.m HOO.O H HOO.O HHH mOO.O H OOH.O HOOz +.H O.H OHOO.O H OHO.O HO OHO.O H OO.O va O-HO-m.m 0.0Amv up 0.:oe 0.0Amev we 0.:me aounhm .H HH HO OOOHO HOOOOHO OH H OaOHO HO m8EHOOOO HHOHHHHH H6>HHO OOO .OOHOOHO OOHHOO .OOHHOHOO HOHmz HHOHH< OOO HHO a a olcaouolowlo:u:omzlm.~ How A p was mwv m0aHu0mHH one A 9 van mev mvHOHw asucmac 60HmmHam .om 0Hnma .NmH 0H0 muouuo 0>Hu0H0H omumBHummo .o0x00u0 mommme umoa .mowmem Hocm:u0 Nam cHo .mmmHm Hoc0:u0 Nmo o0M00u0 Ian eHo .:umc0H0>03 0chHm 0 um maomHumoaoo 0H0HM Bonanza Haom 0000H 00 Beam 00w0u0>0 pom meoHu 10H>0o onmocmum 0H0 muouuo woumqucH 0:H .w czouo 00Hm ou 0>HumH0H o0aHaH000o 0H0? mmonnaou mo moHoHA asucmso: .Am 0H:mH 000V 000H0:u:oma ou 0>HumH0n woeHsH0uov 0u03.w :30H0 mo moHon aauamsom mOOmOOOOO + w 176 OH.~ HOO.O H HOO.O - HOO.O H OOH.O HH.H HOO.O H HOO.O - HOO.O H mHH.O HHOOO + w mH.H HOO.O H HOO.O OH HOO.O H OHH.O HOOO + w HH.H HOO.O H OHO.O HO OOO.O H OHH.O HOOO + w Hm.H. HOO.O H OHO.O OO OOO.O H OHH.O Hoe + w m.~ HOO.O H mHO.O OH OOO.O H HOH.O HOOz + w H.m OOOO.O H OHO.O HH OOOO.O H OOH.O HMO O-HO-O.H 0.0A3 up some 0635 HO. some 339$ .O OH HO OOOHO HOOOOHO OH w nacho mo 00x0HoEoo 0H0HMHHH H0>HHm can .0oHEoum asHumm .0oHH0H:U H0002 HH0MH< one Awu O O-OOOHo-Hm-OOHOOOz-O.H H6O H H OOO OHH OOOHHOHHH OOO HOO OOO HOV mOHOHH OOHOOOO OOHOOHOO .HH 6HOO.O 177 .Nm»-. GHQ mHOHHU 0>HUNHOH flwugfi um00 .mmme A>\> .Huev Hoam:u0alHoem:u0 Nmm aHo .mmme Hoqm:u0 Nmo v0xomu0a: eH0 .:uwa0H0>m3 0HmaHm m um maomHumoaoo 0H0Hh azuemno Haom umm0H um aoum m0wmu0>m you 000H00H>0o vumocmum 0H0 muouu0 00000HocH 0:9 .m cacao 00am ou 0>H00H0H o0: au0u0o 0H03 m0N0Hoaou mo moH0Hm asuamso: .Aw 0H:0H 000V 0c0H0:u:ome ou 0>H00H0H o0aHau0u0u 0903 aeouu mo moH0Hz anaconda OOONOOOOO + m OH.O OH.O HOO.O H H0.0 - OOOO.O H OOOO.O H.~ HOO.O H OOH.O - HOO.O H OO0.0 HHmOm + w O0.0 HOO.O H mmm.O - HOOO0.0 H HmHOO.O HOOO + w H.H mOO.O H HOH.O - OOOO.O H OOH0.0 HOOO + m O.O HOO.O H OHH.O HH HOO.O H HOO.O H0O +.w H.H HOO.O H OHH.O OH HOO.O H OHO.O HOOz + m H.m OHO.O H OH.O Om OOOO.O H HHH.O va O-HO-O.H 0Amv up 0.:ae 0.oAm:V we 0.:me a0unhm .M um H0 000H0 Ho:00H< 0H m.:3ouu mo 00x0Hoaou 000HMHHH H0>HHm was .0oHaon EsHuom .0cHHoH:u H0002 HHmMH< cam a a. olnsou01-|o:u:mmZIm.H How A P van wpv 00SHH0MHH was A 6 one mev moH0H» Bayonne :onanm .NN 0Hnme .NmH 0C0 muouu0 0>HumH0C o0unfiHumm0 .mmme A>\> .HHHV H0C0:00EIH0C0:00 Nmo CHo .mmew HoCm:u0 Nmo 00:00C0 ICC CH0 .:uwC0H0>03 0HwCHm m 00 wCOmHummaoo oH0H9 ECqusv CCom umm0H um Eoum m0wmu0>0 Com mCOHu 10H>0v mumvaum 0C0 muouu0 o0umoHoCH 0:9 .* Cacao 00Cw 00 0>HumH0H o0CHEC0u0o 0C03 00x0Ho500 mo moH0Hm aCquso: .Am 0H:09 000v 0C0H0:0:C0C ou 0>H00H0u o0CHBH0u0o 0H0: & C30u0 mo moH0H> aCquCom 178 H.m HOO.O H OOO.O OOH HO.O H mH.O HOmOzOO-H+ H O.m HOO.O H HHO.O HO HOO.O H HOH.O HOmHHzHHHhH. + H H.m mOO.O H OH0.0 OO mOO.O H OHH.O HOmOzHO-m + H OH HO0.0 H HOO.O HO HOO.O H HmH.O HOHOz + .H . Om OHOO.O H OHO.O HO OHO.O H OH.O H3 O-HO-m.m C C m 0.0Amv 9 0.: e 0.0A0Co 9 0.:me a0unhm .OH HH um mmCHo Ho:00H¢ CH & Cacao mo m0KOHCEoo 0vHHOH:o ECHCOEECH9MH< va ECHCoEE< ow va o a IC3ouoIomlo:u:Q02Im.N now A 9 CCC m9v m0EHu0MHH 0C0 Ame ow new mVH0H9 ECuCCCO COHmmHEm .MN 0H:m9 179 .NmH 0H0 muouu0 0>HumH0u o0u0aHumm0 .mmmHm A>\> .Hucv H0C0:u0alH0Cm:u0 Nmo CHo .mmme HoCm:u0 Nmm v0xomu0 ICC CH0 .:uwC0H0>03 0HmCHm 0 00 0ComHHmCEou 0H0H9 aCuCCCv HCom 0000H 00 Scam m0wmu0>0 Com mCOHu 10H>0v oumoCmum 0C0 muouu0 00000H0CH 0:9 .w Caouo 00um cu 0>HumH0H 00C au0u0v 0u03.00x0HCEO0 mo moH0H9 850C090: .Aw 0H:09 000V 0COH0:0:COC ou 0>H00H0u o0CH8h0u0v 0H0? Cacao mo moH0H9 aCuCCCom H.H HOO.O H HHO.O HO HOO.O H HHH.O HOOOOOOLH.+ w O.H HOO.O H OOO.O HO OOO.O H HHH.O HOOOzHOLH_+.w H.H HOO.O H OHO.O HO HOO.O H HOH.O HOmOzHOLm.+_w. H.H OOO.O H OHO.O HO OOO.O H OHH.O HOHOz +.w H.m OOOO.O H OHO.O HH OOOO.O H OOH.O Hmu O-HO-O.H Amv C9 . Ce . AnCv my . we a0ummm 0 0 : 0 o 0 : .M Bx. um mmmHU HOSOUH< a.“ .W 55090 mo mOXmHQEOU 0wHHOH£U gncoE—CQHHAxHJH vcw Eggnog find Ag O-OOOHO-HH-OOHOOOz-O.H HOH HOH OOO OHH OOOHHOHHH OOO HOO OOO HOV OOHOHH OOHOOOO OOHOOHOO .HH OHOOH 180 .NmH 0C0 muouu0 0>H00H0u o0uCEHumu0 .mmme H0C0:u0 Nmo 00:00C0 CHo .mmmHm HoC0:u0 Nmo v0xumH0CC CH0 .:HMC0H0>03 0HwCHm 0 00 mCOmHumaaoo vH0H9 BCHCCCU CCom umo0H um Scum n0wmu0>0 Com mCOHu ICH>0o manoCmum 0H0 muouu0 00000HoCH 0:9 aw Csouu 00mm cu 0>H00H0H o0CHEH0u0v 0H03.00C0Hoaou mo an0H9 aCuCCCc: .Ao 0H:09 000V 0C0H0:0:C0C ou 0>H00H0C v0CHau0u0v 0C03.W Cacao mo moH0H9 aCuCCCOC H.H OOO.O H OOH.O - HOO.O H HHH.O HOmOzHOng+.w H.H mOO.O H OOH.O - HOO.O H HOH.O HOHOz +_~ H.H OHO.O H OH.O OO OOOO.O H HHH.O H~O O-HO-O.H a a m m 0.0A3 9 0.: e oAva 9 0.: e a0umhm .M 99 um mmmHo H0C0:um Nmo CH N Csouo mo 00X0Hnaoo ECHCanCthoumLm.va ECHCoaa< 0C0 Awu C O-OaOHO-HH-OOHOOOz-O.H HOH H H OOH OHH OHOHHHHHH OOO HOO OOO HOV OOHOHH OOHOOOO OOHOOHOO .OH OHOOH 181 Table 26. Emission Quantum Yields (0f and 0p) for 2,3-Naphtho-20- Crown-6 (2) and Haloalkylammonium Chloride Complexes of Crown 2 in Uncracked 95% Ethanol Glass at 77 K. a,b a,b System 0f 0p 2,3-Cr—6 (2) 0.26 i 0.01 0.049 t 0.002 + Br(CH ) NH Cl 0.165 i 0.004 0.083 1 0.002 2 2 3 2 + Br(CH2)3NH3Cl 0.108 i 0.003 0.107 t 0.001 2 + I(CH2)2NH3C1 0.030 t 0.003 0.104 i 0.004 2 + I(CH2)3NH3C1 0.056 i 0.006 0.073 i 0.002 aQuantum yields of crown 2 were determined relative to naphthalene. bQuantum yields of complexes were determined relative to free crown The indicated errors are standard deviations for averages from at least four quantum yield comparisons at a single wavelength. 182 Table 27. Emission Quantum Yields (¢f and ¢p) for 1,8-Naphtho-21- Crown-6 (e) and Haloalkylammonium Chloride Complexes of Crown % in Uncracked 95% Ethanol Class at 77 K. System ¢fa,b ¢p35b 1,8-Cr—6 (g) 0.106 i 0.006 0.075 t .006 z + Br Wavelength positions of peaks and shoulders (half- height lines) for fluorescence spectra of naphthalene and 2,3-, 1,8-, and 1,5-disubstitutednaphthalenes in 952 ethanol glass at 77 K. 208 emerges for 82 absorption bands. The $2 absorption bands of all crown complexes and naphthalene derivatives are similar in shape to the 82 band of naphthalene. 82 bands for the 1,8- and 1,5-series of naphthalene derivatives are red shifted by about 1000 cm.1. Complexation of crowns 2 and 3 generally induces further red shifts, but these are generally much smaller (100 cmfl) than those induced by substitution. The energies of the 82 bands for the 2,3-series of naphthalene derivatives are not as affected by substitution as for the 1,8— and 1,5-series. The intensity of the $2 absorption bands of complexes are not of greatly different intensity relative to free crown except for the alkali metal complexes of crown a, which show a 50 to 100% increase in intensity relative to free crown. The above generalization suggests that the 82 states of naphthalene, naphthalene derivatives l through 2, and complexes of crowns l, g, and g are all similar but of somewhat lower energy relative to naphthalene for complexes and derivatives in the l,8- and 1,5-series and of similar energy relative to naphthalene for the 2,3-series. Since 82 is not directly involved with any of the observed decay rates (it is 83 that is thought to be vibronically coupled to S1 for 74,75 naphthalene), this conclusion does not have any direct bearing on whether rates involving decay from 81 and T ought to be comparable 1 for different crown complexes. Considering now the S absorption bands, it turns out that com- 1 plexes of the same crown have 81 absorption bands which are very similar to each other. This again suggests that results for dif- ferent complexes of the same crown ought to be comparable. Also, 209 for crownslg and'e, the shape of the S1 band of the free crown is similar to those of its complexes. The S1 absorption band of crown % is also similar to those of its complexes, but the latter show three additional small peaks not evident for the former. The fact, however, that the intensities of peaks in the S1 bands of complexes of crown‘l are generally of roughly the same or lower intensities than those for free crown l suggests that complexation does not promote vibronic coupling with higher excited singlets, since this would be expected to further increase the intensity of this band76 (see introduction section on proposed perturbational method). These results suggest that results associated with the singlet manifold should be comparable for complexes of a given crown. The only comparisons which remain to be made are between the fluorescence spectra and S1 absorption bands of naphthalene deriva- tives l through 2 to each other and to naphthalene. To aid comparison of these spectra, stick spectra (only peak positions indicated as a function of wavelength) for fluorescence are given in Figure 61 and for 81 absorption bands in Figure 62. It is difficult to discern peaks in the fluorescence tails for many of these compounds, so comparisons will be based on roughly the same portion of the higher energy region for all compounds. Positions of shoulders are indi- cated by lines one-ha1f the height of those for well-defined peaks. Suggested correlations between different spectra are indicated by dashed lines. These correlations are based on the similarity of spacings between sets of sticks in both spectra. (The spacings are linear in wavelength, not cmfl, but over a small range of 210 .M um um mmwam Hosmsuo Nmm ca nonmamnunamavmuauaunnomavnm.H pom aw.H .Im.~ can mamHmnusaoc mo muuumam coauauomnw >3 now AmocHH unwamnlmamnv muwpaoonm was mxmwa mo mcofiuwmoa Suwamam>m3 .Nc shaman ‘IlllAEc? meq q a q a q . d «CW. 44 a; _ q . .Ofln. _ Ado 2%. x_ ..._ \._a_ _ x_ _ _ _ a. fo r... _--- _.-- _\._.- _.. .. .13.... ,,_ ,._ .._ .._ ,i E wacmfofuo?o 3 ._ _ _ _ ._ E are 0.. _.... _.._. _... _... _ G. nIuo~Iu 0.. ..._ ... _l.._ ...— .zsomo. m m ... m a. «roomfofoo?u m... _ _ _ _ . a. #5 n.~ _. _ _ _ _ _ .3 £836 n.~ _ _ ._ _ _ a _ _ .2265. a «zooflrufuomro am _ . _ , hzuathmmam owStkmg m¥8¥tmoa 211 wavelengths energy differences will be roughly proportional to wave- length differences.) The object of these comparisons is to see whether or not these spectra differ only by being displaced to lower or higher energy and to see how similar or dissimilar their vibra- tional structure is. If the vibrational structures are the same, it .will be assumed that the states are the same. Inspection of the stick representations of the fluorescence spectra given in Figure 61 bears out observations made in the results section without the aid of this figure: 1) the vibrational spacings for crown and methoxymethyl derivatives are similar within the 2,3-, l,8-, or 1,5-series; 2) the fluorescence spectrum of 2,3-dimethy1- naphthalene has more peaks than the 2,3-crown or methoxymethyl derivatives but includes all of the peaks and nearly the same vibra- tional spacings seen for the latter two; and 3) the fluorescence spectrum of the 1,5-dimethy1 derivative has most of the same vibra- tional spacings and peaks as the 1,5—crown and methoxymethyl deriva- tives. A generalization suggested by Figure 61 is that the vibra- tional structure of all 1,8- and 1,5-crown, methoxymethyl, and di- methyl derivatives are the same (only the position of one shoulder for the 1,5-dimethyl derivative does not quite correlate). Note for the 2,3-series that in going from methyl to methoxymethyl to crown the higher energy peaks shift to higher energy and the lower energy peaks to lower energy. A similar trend was noted above for complexes of the 2,3-crown relative to free crown. No correlations have been suggested between the 2,3- and the l,8-series, since the 'vibrational spacings are different. Also, no correlations between 212 naphthalene and any of its derivatives have been suggested. Since vibronic coupling with higher singlets affects which vibrational levels of SO are populated by fluorescence,76 the fact that these spacings are the same for the l,8- and 1,5- series suggests that vibronic coupling is the same throughout both series. Since com— plexes of the 1,8- and the 1,5-crowns have the same vibrational spacings as the free crowns, this conclusion applies to them as well. The differences in vibrational spacings for the 2,3-series do not necessarily mean that S1 for the 2,3-series is coupled dif- ferently. The differences may instead be due to differences in molecular symmetry, which also affects the allowedness of vibrational modes.77 This point will be discussed further in the discussion of intensities of absorption spectra. Inspection of Figure 62 bears out observations made in the results section for $1 absorption bands of naphthalene derivatives: 1) the vibrational spacings are similar within the 2,3—, 1,8-, or 1,5- series for methoxymethyl and crown derivatives; 2) many of the same vibrational spacings are observed throughout the 1.8- and the 1,5- series (there are a few shoulders apparent in some cases but not in others); and 3) vibrational spacings for the 1,8- and l,5-series are similar to those observed for naphthalene. Only one further correlation is suggested by Figure 62 that was not suggested in the results section and that is that some of the vibrational spacings for the 2,3-series are the same as those observed to be constant throughout the 1,8- and 1,5-series. The similarity of vibrational spacings throughout the 1,5- and l,8-series indicates that vibronic 213 coupling is the same throughout both series. Vibrational spacings within the 2,3—series are similar but correlations with the 1,8- or 1,5-series is tenuous. Thus, essentially the same conclusion is reached based on similarity of vibrational spacings in absorption spectra as was reached based on vibrational spacings in fluorence spectra. The preceding arguments have been based on similarities of vibrational spacings. But changes in molecular symmetry can also 67 so differences in vibrational struc- affect vibrational structure ture does not necessarily imply a difference in vibronic coupling. The following arguments will be based on relative peak intensities in an attempt to determine whether the difference in vibrational structure between the 2,3- and the l,8- or 1,5-series is due to a difference in the way electronic states are vibronically coupled thus giving rise to peaks associated with nontotally symmetric vibra- tions, or due to a difference in molecular symmetry, which might be expected to change the intensity of the 0-0 band. O-O electronic transitions are not greatly affected by vibronic coupling, changes in the intensities of 0-0 peaks due to chemical substitution are likely to be due to changes in molecular symmetry, not to substi- 78 Thus, a naphthalene's 0-0 tuent induced vibronic coupling. absorption peak is 10 to 20 times less intense than the rest of the $1 absorption band and its 0-0 emission peak is 6 times smaller than the most intense emission peak. For all the 2,3-disubstitutednaph- thalene derivatives (Figure l), the intensity of the entire S1 absorption band is increased relative to naphthalene but the 214 intensity of the 0-0 peak is increased the most. The same situation obtains for the fluorescence spectra (Figure 19): the 0-0 peak is the most intense peak for each member of the 2,3-series. For the 1,8- (Figures 2 and 21) and the 1,5-series (Figures 3 and 22), the 0-0 band is not the most intense peak in either the absorption or in the fluorescence spectra for crown or methoxymethyl derivatives. But for the l,8-crown and methoxymethyl derivatives, the S0 O-O absorption peak is about two times more intense than for naphthalene, whereas the second peak is two times less intense than the correspond- ing peak for naphthalene. For the fluorescence spectra of these two l,8-derivatives, the 0-0 peak is about three times more intense than for naphthalene, whereas the second peaks are about of the same intensity as for naphthalene. For the 1,5-crown and methoxymethyl derivatives, the 31 0-0 absorption peaks are about five times more intense than naphthalene, whereas the second highest energy peaks are only about two times more intense than for naphthalene. The fluorescence spectra for the 1,5-crown and methoxymethyl derivatives have about the same relative peak intensities as for naphthalene throughout their spectra. The 81 0-0 absorption peaks and 0—0 fluorescence peaks for the l,8- and 1,5-dimethyl derivatives are the most intense peaks in their spectra. Therefore, the main effect of 2,3—, 1,8-, and 1,5- disubstitution appears to be to change molecular symmetry without enhancing vibronic coupling between elec- tronic states. Since, as noted in the previous comparisons (gigs 32252), complexation does not greatly alter fluorescence or absorp- tion spectra of complexes relative to parent free crown, this 215 Table 33. Estimates for Rate Constants of Excited State Processes of Naphthalene and 2,3-, l,8-, and 1,5-Disubstituted- naphthalenes in Alcohol Glass at 77 K.8 Naphthalene Substitution kfxlo- knrxlo- kpxlo kdt None (19) 1.4 3. 1.7 0.38 2,3—Dimethyl (4) 2.4 4. 1.2 0.36 2,3fi§i§(methoxymethyl) (5) 2.5 8. 1.4 0.35 2,3-Crown-6 (l) 2.8 7. 2.4 0.33 l,8-Dimethy1 (Q) 4.8 7. 1.1 0.53 1,8fi§i§(methoxymethyl) (7) 2.0 19 3.3 0.39 l,8-Crown-6 (2) 2.4 20 3.5 0.38 1,5-Dimethyl (8) 4.1 5. 1.5 0.37 1,5fi§i§(methoxymethy1) (2) 1.7 11 3.7 0.36 1,5-Crown-6 (3) 3.3 25 8.2 0.37 8All rate constants in terms of sec- . 1 Ll 216 Table 34. Estimates for Rate Constants of Excited State Processes for 2,3-Naphtho-20-Crown-6 ( ) and Alkali Metal Chloride, Barium Bromide, and Silver Triflate Complexes of Crown 1 in Alcohol Class at 77 K.8 6 6 System kfxlo- knrxlo- kpxlO kdt 2,3-Cr-6 Ck) 2.8 7.9 2.4 0.33 ,1, + NaCl 1.7 7.2 2.6 0.27 &‘+ KCl 1.7 9.1 3.1 0.29 ,1 + RbCl 1.5 11 3.9 0 33 ,1, + 0301 0.56 11 6.5 0.39 l + BaBr2 0.92 18 5.4 0.32 4, + AgOSOZCF3 --- 20b 2.4 0.29 1 8All rate constants in terms of sec- bCalculated assuming that kf - 2.8x106. UV absorption spectra (see text for rate constant calculations for alkali metal chloride complexes of crown 3 and Figure 16). Assumption based on 77 K 217 Table 35. Estimates for Rate Constants of Excited State Processes for l,8-Naphtho-21-Crown—6 (2) and Alkali Metal Chloride, Barium Bromide, and Silver Tfiflate Complexes of Crown 2 in Alcohol Class at 77 K.3 6 System kfx10' knrxlo’ kpxlO kdt l,8—Cr—6 (g) 2.4 20 3.4 0.38 g + NaCl 2.3 19 3.6 0.40 2 + x01 3.2 15 4.0 0.67 2 + RbCl 3.4 16 4.3 0.73 2 + CSCI 3.0 19 6.0 0.77 2 + BaBr2 --- 21b 6.4 0.81 g + AgOSOZCF3 --- 17b 3.2 0.43 8All rate constants in terms of sec-1. bCalculated assuming kf - 3.0 x 106. Assumption is based on 77 K UV absorption spectra (compare Figures 14 and 17 to Figure 5 and see text for calculation of rate constants for alkali metal chloride complexes of crown é)' 218 Table 36. Estimates for Rate Constants of Excited State Processes for 1,5-Naphtho-22-Crown-6 ) and Alkali Metal Chloride, Barium Bromide, and Silver Triflate Complexes of Crown 3 in Alcohol Class at 77 K.8 System kfx10-6 knrxlO-6 kpxlO2 kdt 1,5-Cr-6 (,3) 3.3 25 8.2 0.37 ,2 + NaCl 2.6 32 5.7 0.36 ,3 + x01 2.3 35 6.3 0.44 ,3 + RbCl --- 52b 11 0.48 ,3 + 0501 -- 670b 93 1.85 3 + BaBr2 --- 31 10 0.37 g + Agosoch3 --- 500 450 0.82 a .- All rate constants in terms of sec 1. bCalculated assuming kf - 1.0 x 106 (see text). cCalculated assuming kf - 3.3 x 106. Assumption is based on 77 K UV spectra (compare Figure 15 and 18 to Figure 6 and see text for calculation of rate constants for alkali metal chloride complexes of crown 3). 219 conclusion applies to the effects of complexation as well. This conclusion suggests that effects due to perturbers complexed by 2,3-, l,8-, or 1,5-crown derivatives should reflect properties of the perturber and the position of the perturber relative to the chromophore without overpowering effects due to substituent or perturber induced vibronic coupling. The discussion will turn now to a consideration of the validity of the proposed perturbational method based on a consideration of changes in rate constants as a function of type of substitution and as a function of the type of alkali metal cation. Rate constants (kf, k k , and kdt) for naphthalene and naph- nr’ p thalene derivatives (1 through a) and for the alkali metal chloride complexes of crowns l, 2, and 3 are given in Tables 33 through 36, respectively. These rate constants have been calculated from quantum yield (¢f and ¢p) and lifetime (Tf and Tp) data given in Tables 8 and 20 through 22 according to the equations given in the introduc- tion section on excited state processes. These calculations assume (see introduction) that k is negligible compared to kf + kisc which ds can be expressed by assuming that l - ¢f = ¢isc°6 Since, if this assumption is wrong, the value of k will be affected, the rate isc constants tabulated will be referred to as nonradiative decay from the singlet, knr (knr s kd8 + k ). Thus, even if the assumption isc is wrong, the tabulated values of knr would still be correctly lab- elled, although they would be a composite of two rates, the ratio of which would be unknown. knr will be thought of and discussed, however, as dominated by k1 . sc 220 If for the Rb+ and Cs+ complexes of crown.g could not be de- termined due to experimental constraints (see results section on lifetime determinations), either because of was too low or because 1f was too short. So a value of kf based on 77 K UV spectra was guessed to allow calculation of knr - (1 - ¢f)kf¢f-1). There is a general relationship between kf and the oscillator strength of the S1 absorption band, i;g;, kf is expected to increase as the intensity of the $1 absorption band increases.79 Generally, good absorbers 79 Put somewhat more exactly, the inherent radia- are good emitters. tive lifetime, 1° (1° - kf-l), is inversely proportional to the integrated intensity of the 81 absorption band. This relationship doesn't always hold, however.79 From the results of this investi- gation, a rough correlation is found between kf and the intensity of the 0-0 absorption peak; 3;g;, kf increases as the intensity of the 0—0 absorption peak increases. Unfortunately, this correlation doesn't hold up very well for the Na+ and K+ complexes of crown é! kf decreases slightly as the intensity of the 0-0 absorption peak increases. (Compare Table 36 to Figure 6.) The rest of the 81 bands for these complexes are increased in intensity as well. This being the case, it is difficult to do anything else but guess at what kf might be. The 0-0 absorption peaks for the Rb+ and Cs+ complexes are about half as intense as that for the free crown, but the rest of the S1 bands of these complexes are of about the same intensity as for the Na+ and Rf complexes. It is expected that 4. kf is about the same for the Rb and Cs+ complexes as for the Na+ and K+ complexes, so it is thought that a value of 1 x 106 is 221 sufficiently low so that increases in knr are underestimated rather than overestimated. If this assumption is correct, then 1f 8 ¢flkf - 2 nsec for the Cs+ complex of crown 3, which would be too short to measure given the experimental problems discussed in the results section. (¢f - 0.0015 would also make determination of a longer lifetime difficult.) Relative errors for kf, k , and k1) are estimated to be about nr 10% and, for kdt’ about 52. kdt is determined mainly by Tp in most cases and Tp estimates are thought to be good to about 5% (see results). The absolute errors may be larger if the estimates used for 0f and ¢p of naphthalene are in error. This investigation puts limits at least on how large ¢p for naphthalene can be, since ¢p for the Ag+ complex of crown a was found to be 0.84:.08. Rate constants as a function of naphthalene substitution and complexed alkali metal ion perturbers have been presented graphically in Figures 63 through 65 for the 2,3-, the 1,8-, and the 1,5-series, respectively, to facilitate comparisons. Tables 33 through 36 can be referred to for numerical values if desired. Of interest will be: comparison of the effect that each kind of substituent (CH3, CH30CH2, or crown) in a given series has on changing a given rate constant; comparison of substituent induced changes of rate constants to changes induced by complexation of alkali metal cations; and comparisons of substituent and complexation induced changes of rate constants between the 2,3-, l,8-, and 1,5-series. Considering changes in kf for the 2,3-series, it is seen that CH3, CH30CH2, and crown substitution all increase kf by roughly I0"6 k, 222 -N 0‘ 6083-0 III I l C '0 I C) in r o -6 IO km. :02 k omamUU'VtD o Figure 63. 01.01.0104 0010740 I I I I I C . N V I . Changes in estimated rate constants caused by 2,3- disubstitution of naphthalene and by alkali metal cation perturbation of the 2,3-crown (l). 223 ’0‘ J ©©¢©$ grain" K"' mi“ 05"" I65 kf PU O“ J3 (fl I l -6 IO kn,” Io2 k bin'm'w'momamusl 1 [— Figure 64. Changes in estimated rate constants caused by 1,8- disubstitution of naphthalene and by alkali metal cation perturbation of the 1,8-crown (g). 224 was ’8‘ 166'“ 2.0 b - 600 -. 400 1061nr 200 80p . 60 - 102:.ID 20- 1.6— 1.4- 1.2— o [—l O Figure 65. Changes in estimates of rate constants caused by 1,5- disubstitution of naphthalene and by alkali metal cation perturbation of the 1,5-crown.(3). 225 the same amount. Complexation of cations, however, decreases kf relative to free crown 1. There is about a 5-fold decrease relative to free crown induced by Cs+ complexation, whereas the increase due to substituents is only a 2-fold increase. For knr’ CH3 causes a slight increase relative to naphthalene; CHBOCH2 and crown both cause about a 3-fold increase (oxygen effect). Complexa- tion of Na+ causes little change relative to free crown, of K+ a small increase, and of Rb+ and Cs+ larger increases of the same mag- nitude. These increases in knr caused by complexation appear to be in addition to whatever change is induced by crown substitution it- self. Note that the relative increase in knr due to complexation is less than the relative decrease in kf, so the observed decrease in ¢f is mostly due to the decrease in kf, not to the increase in knr' For kp, substituents cause only small changes relative to naphthalene in comparison to the increase in kp relative to crown caused by complexation of alkali metal cations. Note the increase in kp due to complexation of Rb+ and Cs+, for example, the increases become progressively larger as Na+: K+ , Rb+, and Cs+ are successively come plexed. For kdt’ there is a decrease introduced by each substituent in going from H to CH3 to CHZOCH3 (oxygen effect) to crown (crown effect). Complexation of Na+ causes a further decrease in kdt rela- tive to crown, but kdt then progressively increases relative to kdt +, and Cs+ are successively complexed. for the Na+ complex as K+, Rb The decrease of kdt relative to free crown followed by an increase suggests the operation of competing factors (see discussion below on correlations of rate constants with energy differences). Note 226 that CH30CH and crown substitution have about the same effect on 2 each rate constant (except kdt)’ indicating that there is little crown effect - little effect due to joining the 011300112 oxygens in a crown cycle. Turning now to the 1,8-series (Figure 64), it is seen that CH3 causes the largest change in kf relative to naphthalene. Note that a similar large increase in kf is caused by 1,5-dimethy1 substitution (Figure 65), so it is doubtful that peri interactions are responsi- ble. Note also that there is a corresponding large increase in the intensity of the S1 absorption band (Figure 2) complexation of Na+ causes little Change in kf relative to free crown; complexation of K+, Rb+, and Cs+ causes changes in kf relative to free crown which are as large as those caused by CH30CH2 and crown substitution rela- tive to naphthalene. For knr’ CH3 substitution causes a 3-fold increase relative to naphthalene; CH30CH2 and crown substitution both produce about a 6-fold increase in knr' The changes in knr induced by complexation of cations relative to free crown is smaller than the change induced by CH30CH2 or crown substitution relative to naphthalene. Complexation of Na+ causes little change in knr rela- tive to free crown, but complexation of K+ causes about a 25% de- crease which is followed by progressive increases in knr for suc- cessive complexation of Rb+ and Cs+. As for k t in the 2,3-series, d this suggests the operation of competing effects. R1) for the 1,8- series is not affected very much for CH3 substitution but is increased about 2-fold by CHBOCHZ and crown substitution. Successive complexa- + tion of Na+, K+, Rb , and Cs+ results in progressively larger 227 increases in kp relative to free crown. Except for a larger oxygen effect on kp for the l,8-series, the effects of substitution and complexation are similar to those for the 2,3-series. For kdt’ only CH3 substitution causes a marked increase, but successive com- plexation of Na+, K+, Rb+ , and Cs+ results in progressively larger increases in kdt relative to free crown. Figure 65 for the 1,5-series, graphically shows that, for this series, substituent induced changes in rate constants are negligible compared to the effects due to complexation of Cs+. Note that if kf was plotted on the same scales used for knr or k that there p9 would appear to be essentially no variation in kf. Note that the value of kf for the Rb+ and Cs+ complexes is an assumed value (vide supra). For this series, inspection of tabulated values is neces- sary to put knr’ kp, and kdt in perspective to those in the 2,3- and 1,8- series. Inspection of Table 33 shows that there is a sig- nificantly larger difference between the effects of CH OCH2 and 3 crown substitution in the 1,5-series than in the 2,3- or in the 1,8- series. For the 1,5-series, kf, knr’ and k.p for crown substitution are each about twice as large as the corresponding rate constants for CHBOCHZ substitution (crown effect). That there is such a dif- ference in the 1,5- and not in the 2,3- or l,8-series is not sur- prising in view of the fact that the crown ring passes over the face of the naphthalene w-system for the 1,5-series. kdt is essentially the same as for naphthalene throughout the 2,3-, 1,8-, and 1,5- series for CH3CH30CH2, and crown substitution except for CH 3 substitution in the 1,8-series. 228 Figure 65 also makes the effects of complexation of other cations appear to be negligible relative to the effects due to complexation of Cs+. Inspection of Table 36, however, shows that knr progres- + sively increases relative to free crown as Na+, K+, Rb , and Cs+ are successively complexed. kp and k t first decrease and then in- d crease relative to free crown as complexation successively proceeds from Na+ through Cs+. It might be argued that tying the CH3OCH2 oxygens into a crown cycle changes the naphthyl carbon-methylene carbon-oxygen angles in such a way that the oxygens nonbonded elec- trons do interact more with the n-system. If this interpretation is correct, then it might be supposed that this kind of interaction or inductive effect would be modified by complexation of cations. Changes in knr and kp due to cation complexation might then be a function of how this interaction is modified by complexed cations. The fact, however, that k nr’ RP, and kdt (for which there is no oxygen effect) generally tend to increase relative to free crown in +, and Cs+ are succes- the 2,3-, l,8-, and 1,5-series as Na+, K+, Rb sively complexed, indicates the likelihood of some common interac- tion. The three crowns involved, however, are sufficiently different that it is likely that possible interactions of their naphthylic oxygens with the n-system would produce the same common effect in each case. For instance, for crown 3 there may be one kind of interaction between the naphthylic oxygens and the n-system enforced by the polyether chain which passes over the n-face (activation energy of about 6 kcal per mole for rotation of crown ring around 80 naphthalene substrate), another kind of interaction enforced by 229 peri interactions for crownle (vide infra), and, perhaps, only conformationally enforced interactions for crown’l. The suggested enforced interaction for crownle is evidenced by a much larger crown effect for crown substitution in the 1,5-series than in the 2,3- series (some crown effect) or in the l,8-series (little crown effect), as noted previously. The peri interactions suggested for crown‘% would restrict conformational freedom and, perhaps, result in a more or less restricted range of possible interactions. The absence of an oxygen effect in the 2,3-series for kp but the presence of a crown effect (kp is increased by a factor of two for crown substi- tution relative to CH30CHZ substitution) suggests a set of somewhat different kinds of interactions. The increases in kp observed for complexation of alkali metal cations by the 2,3-, l,8-, and 1,5-crowns, except for complexation of Na+ and Rf by the 1,5-crown, are more significant than the in- creases in km. and kdt’ since, in general, radiative transitions are less dependent on energy differences than nonradiative transi- tions.81 Indeed, it turns out that there is a good linear correla- tion (correlation coefficient (R) - 0.998) between log k and the dt change in triplet energy relative to crown, (ET1 - so), for alkali metal complexes of crown.g, as shown in Figure 66. Note that Figure 66 includes points for metals not discussed yet, Ba++ and Ag+. The correlation appears to be less good for the alkali metal com- plexes of crown.l. There appears to be a correlation for three of the points (R . 0.992), but the point for the Cs+ complex definitely does not fit the correlation (R - 0.466 if this point is included). 230 0'0. a P N Cl ©© x KCI ©© .. . O O,Psnone O RbCl °.P' none A . .CSCI (1;) -02.. *AQOSOzCFs 5 _ 6 :8“ * -0A~ ‘ I I5 I i a X *- -0.6" ‘ -400 -200 0 +200 +400 Change in Triple: Energy Relative to Crown. Increasing (T. '50) Separation (cm") Figure 66. Plot of log kdt versus the change in the T1 - 80 energy separation in cm-1 caused by alkali metal, barium, and silver cation perturbation of crowns 1 and 2. 231 Note that it may be fortuitous that points for the free crowns appear to fit with the correlation for the complexes. Also, it may be fortuitous that points for complexes of both the 2,3— and the 1,8— crown appear to fall on roughly the same line. Note that A(ET1 - SO) is estimated to be good to $20 cm.-1 and kdt to be accurate within :52, so the position of the point for the Cs+ complex of the 2,3- crown is not merely due to experimental scatter. The larger value of kdt observed with Cs+ complexation by 1 can be rationalized by enhanced spin-orbital coupling which may not depend on the T1 - So energy separation. Note that the point for the Ag+ complex of the l,8-crown appears to correlate with the points for the alkali metal complexes, whereas the point for the Ba++ complex definitely does not. The points for the Ag+ and Ba++ complexes of crown 1 appear to correlate with the points for the Na+, K+, and Rb+ complexes. There is, however, no apparent correlation between log kp versus A(ET1 - $0) or between log knr versus A(ET1 - SO). one would not 81 So) ' kisc’ however, is known to depend on the energy separation between the coupled states. The lack of correlation between log knr and necessarily expect a correlation between log kp and A(ET1 - A(ES1 - T1) may be because intersystem crossing does not necessarily occur directly to T1. T2 for naphthalene is energetically very close to 81.74 Thus, the decrease relative to the l,8-crown in knr for complexation of K+ and the decrease in knr relative to the 2,3-crown due to complexation of Na+ may be due to an increase in the energy separation between the coupled states. The subsequent increase in knr as other alkali metal cations are successively complexed may be 232 due in part, anyway, to a subsequent decrease in the energy separa- tion between the states that are coupled. For the 1,5-crown, energy shifts induced by complexed metal cations are much smaller than those induced by complexed metal cations for the 2,3- and the l,8-crown (compare Table 4 to Tables 2 and 3), and no correlations between logarithms of rate constants and energy differences are found. Therefore, the larger increases in knr and kdt for alkali metal come plexes of crownlé are more meaningful than those for the 2,3- or l,8-crown, since kdt’ anyway, for the latter cases shows a marked correlation with A(ET1 - 51) for most complexes. For the 2,3- and the l,8-crawn, it appears that there are cor— relations between AGES1 - T1) and A(ET - so) with the polarizing strength of the alkali metal cations. "Polarizing strength" is defined to be the absolute charge of an ion divided by the square of its ionic radius (l/r2 for monovalent ions).83 Plots of A(ES1 - T1) versus l/r2 are shown in Figure 67 for the alkali metal cation for complexes of the 2,3-crown (R - 0.93) and of the l,8-crown (R - 0.97). Plots of A(ET1 - 50) versus l/r2 are shown in Figure 68 for alkali metal complexes of the 2,3-crown (R - 0.91) and of the l,8-crown (R - 0.993). Thus, changes in kdt relative to free crown due to complexation of alkali metals by the 2,3-crown and by the l,8-crown may be a function of the properties of the complexed cations but partially, anyway, vis-a-vis energy shifts induced by the complexed cation. Note (see Figure ) that only the Ba++ complex of the 1,8- crown and the Cs+ complex of the 2,3-crown would not fit a linear correlation between log k and A(ET - ). Energy differences dt 1 S0 233 - 0 c' 3 _ 8 “0’ -,- IOO— *- E x g 3 P=ANaCI g .8 xiéCl '5 ; 0 bCl 0; § 5° I050: 9 m ‘ m U) (2 M A '7' '1 an: 93 ’ .s E I (D ... § g-so ©©° O o X 5 E -00..— A l l J J i J , 0.2 0.4 05 0.8 ID l2 Polarizing Strength of Salt———->- 2 (IR ) Figure 67. Plot of the change in the S1 - '1‘1 energy separation caused by alkali metal cation perturbation of crowns l,and 2 versus the polarizing strength of the alkali metal cations. 234 *300— P=ANoCI xKCl é .RbCl . A g 4200— ICsCl. C3 .::‘ 'I )( ‘9 '5 +100 ’ GD ._, —- {- c 2 .9. a g o 35 '25 ‘ +— (I) '1<)C)"' 2 ' 0 a _ 'C '— '— V ..c. ,8 -2oo-—- ©© Q) (0 a E (J t) 5 E -3001— x ’400— I . --500 J l L I l 0 0.2 0.4 0.6 0.8 l.0 Polarizing Strength of Salt-———> (l/rz) Figure 68. Plot of the change in the T1 - S energy separation caused by alkali metal cation perturbation of crowns l'and 2 versus the polarizing strength of the alkali metal cations. 235 and the polarizing strength of Ba++ and Ag+ do not fit with the ap- parent correlations of energy differences with polarizing strength for complexes of the 2,3-crown and the l,8-crown. This is perhaps because Ba++ and Ag+ are members of different periodic series and because there is difficulty in deciding what the actual polarizing strength of Ag+ should be.82 It is not surprising that complexation of metal ions induces energy shifts. What is perhaps surprising is that energy shifts are not larger (see below under question about the effects of the positive charge of the perturber). It is not clear what factors are responsible for energy shifts due to complexation of metal cations. The above correlations suggest that differences in solvation may be responsible since salvation energies are also related to the radius of the cation and of the anion. Alternatively, energy shifts may be due to field induced n-polarization due to complexed cations. 13C nuclear magnetic resonance (CMR) studies83 at The results of room temperature showed that changes in the chemical shifts of the naphthalene carbons which occur upon complexation of alkali metal cations could be explained in terms of field induced n-polarization. While there were shifts in n electron densities due to complexed cations (as indicated by chemical shifts for the naphthalene carbons of the 2,3- and of the l,8-crown) there was little shift in n elec- tron densities due to complexed metal cations for the naphthalene carbons of the 1,5-crown, as indicated by absence of chemical shifts for the naphthalene carbons. Chemical shifts of the ipso carbons for the 2,3- and the l,8-crown were consistent with an 236 increase of W electron density, which is as would be predicted by field induced n-polarization. Thus, the fact that larger energy shifts are produced by complexation of metal cations by the 2,3- or the l,8-crown than by complexation of metal cations by the 1,5- crown is consistent with an explanation of these energy shifts in terms of greater shifts of n electron density for the former cases than for the latter. However, the n electron polarizations (as evidenced by 13C chemical shift changes) were similar with Na+, K+. Rb+, and C8+ and not a function of l/rz. This argues against a direct dependence of energy shifts and n-polarization. In addition to the general increase of kp (and, to a lesser extent, k and kdt except for the 1,5-series) by complexation of alkali nr 83 together with metal cations, results from the CMR work cited above results from this investigation suggest that the changes in excited state rate constants due to complexation of metal cations don't merely reflect modified interactions of non-bonded oxygen electrons with the naphthalene n-system. The observations which suggest this conclusion are: 1) that changes in knr and kp relative to free crown for complexation of alkali metal cations are smaller for the 1,8- crown (2) than for the 2.3-crown; while 2), the CMR results suggest a much larger adjustment of the C(2): C(1): C(11): 0(12) dihedral angle for the former case (2) than the C(1): C(2): C(11): 0(12) dihedral angle for the latter case (1). The larger change in di- hedral angle postulated for the former case is reasonable in view of possible peri interactions. Provided that ground state and S1 geometries are not too different, these observations from room 237 temperature CMR studies may validly be applied in this argument. Also, it does not seem possible to account for changes in rate constants due to complexation of cations in terms of variation of the inductive effect of the naphthylic oxygens by different metal cations, since, if this was the major mode of interaction, one would expect similar results regardless of which crown the metal cation is complexed. (The effects of substituents on carbons l and 2 of naphthalene are expected to be of the same sign, but perhaps of different magnitude).83 But there are large differences (in size and often sign) in changes produced by complexation of Cs+ by the 1,5-crown and changes produced;by complexation of Cs+ by either the 2,3- or the l,8—crown. There seems to be evidence for the effects of peri-interactions on knr for l,8- and for 1,5-dimethylnaphthalene. There also has been precedence for this in the literature.84 For a series of mono and dimethylated naphthalenes and 1,4,5,8-tetramethylnaphthalenes, it was noted that two compounds had exceptionally high k values, isc l,8—dimethylnaphthalene and 1,4,5,8-tetramethylnaphthalene. These exceptionally high values of k18c were thought to be the result of enhanced spin-orbital (SO) coupling brought about by steric distor- tion of the planirity of the naphthalene nucleus. This distortion from planarity was thought to induce rehybridization of the skeletal naphthalene carbons to include more 0* character and less n character. The increase in kisc’ then, was viewed as being due to enhanced mixing of nn* and nn* states. The results of this investigation are consistent with the results of this previous investigation. 238 The order of increase of knr (Table 33) relative to naphthalene for dimethylnaphthelenes is 2,3 < 1,5 < 1,8. It is not unreasonable in the 1,5-case that there may still be significant peri-interactions between the o—methyl and o-hydrogen. Thus, the decrease in knr relative to crown observed for complexation of Rf by the l,8-crown and the subsequent increases relative to the Rf complex for complexa- tion of Rb+ and Cs... could be explained in terms of variations in the amount of distortion produced by complexation of various cations. This view would be supported by the CMR evidence cited above,83 which suggests that there are larger changes in dihedral angles for theILB-crown than for the 2,3-crown. But these changes in knr could just as well be explained in terms of shifts in energies for the states involved in intersystem crossing. Other decreases in rate constants due to complexation followed by increases for com- plexation of other metal cations have been previously noted. The preceding discussion of effects due to complexation of cations by crowns l, 2, and g_is summarized in Figure 69. The ef- fects relative to free crown for complexation of K+ and Cs+ by crowns l, g, and g are indicated by arrows, the length of which indicates (to scale) how the ratio of the rate constants for perturbed and un- perturbed crown differ from unity (the larger rate constant was taken as the numerator in each case). An arrow pointing up indicates that the rate constant for perturbed crown increases. For perturba- tion by Rf, a light cation, real but not overpowering changes in all rate constants occurs. Each crown and each rate constant responds differently; i.e., no two rows and no two columns agree even as to 239 Cation Induced Changes in Rate Constants , (0).: 6% k, 1 1 8 light knr . . f atonn k (K‘) 9 4 i 7 ka: v ‘ :14 ~ X2 kf 1 heavy knr f ‘ x26 010W" UCS+) p 1r 1{ ‘1‘ x4 “01 ‘ Figure 69. Changes in rate constants of excited state processes of crowns l, 2, and 3 caused by K+ (a light cation) and Cs+ (a heavy cation). 240 the direction of the cation induced changes. However, for perturba- tion by Cs+, a heavy cation, each of the rate constants involving intersystem CTOSSiflS. knr’ k and kdt are increased by cation p: perturbation for each crown. The magnitudes of the increases are much larger for Cs+ perturbation of the 1,5-crown than for the 1,8- or 2,3-crown. The results for Cs+ perturbation warrant several important conclusions. First, it can be concluded that there is a HAE for Cs+ perturbation of all three crowns, since each of the rate constants for a spin-forbidden process are increased in each case, although by different amounts. The fact that there is a large dif— ference for Cs+ perturbation of the 1,5-crown and the 2,3- or l,8- crown implies that the magnitude of the change is not a function of substituent induced changes which obfuscate HAE's but are a function of whg£g_and Egg the cation is held relative to the naphthalene chromophore. This difference in magnitude of the effect, however, does 335 establish that there is a directional dependence for the external HAE, since the distance and position of the complexed cation may vary from cation to cation (see below). Also, as prefaced in the introduction, the methylene units inserted between the naphthalene nucleus and the crown ether oxygen for the purpose of electronic insulation prevent the perturber from approaching the edge of the chromophore as closely as it could if there were hydrogen substi- tuents instead. It can be concluded, however, that approach of a heavy atom perturber to the edge of a substituted naphthalene chromo- phore results in much less effective perturbation than if the perturber approaches the n-face. 241 Conclusions for validity of Studying Excited State Perturbation via Complexed Perturbers General conclusions concerning the validity of the prOposed method for studying excited state perturbation gig complexed metal ions are summarized below. These conclusions are drawn from both rate constant and from spectral comparisons. Spectral comparisions suggest that perturber induced changes in rate constants for all three crowns might be comparable, since 1) all phosphorescence spectra were seen to be similar; 2) the S1 absorption bands and fluorescence spectra for l,8- and 1,5-disubsti- tuted naphthalenes were found to be very similar; and 3) comparison of the intensities of 0-0 absorption and 0-0 fluorescence peaks to those of vibronically induced peaks suggest that the main effect of substitution is to change molecular symmetry without enhancing vibronic coupling. Spectral comparisons also suggested that even if perturber induced changes in rate constants aren't comparable for different crowns, that at least all results for a given crown should be com- parable, since all spectra of a given type are very similar for a given crown regardless of the perturber. Rate constant comparisons within and between the 2,3—, l,8-, and 1,5- series of naphthalene derivatives and complexes suggest that changes due to cation perturbation are not merely the result of modi- fication of substituent effects, since 1) changes introduced by cation perturbers are as large or larger than those due to substi- tuents; 2) there is a general tendency for most alkali metal cations to increase kn k , and kdt for all three crowns; although 4) the r’ p 242 changes in kdt for cation perturbation for the 2,3- and l,8-crown were shown to correlate with A(ET1 - 30) except for Cs+ perturbation of the 2,3-crown (and Ba++ perturbation of the l,8-crown); 5) changes in knr for the 2,3— and l,8-crown may also depend on A(Es1 - Tl)’ although no correlation of A(ES1 - T1) was found; 6) changes in knr for cation perturbation of the 1,8-crown may reflect variable peri-interactions; 7) energy shifts are smaller for cation perturba- tion of the 1,5-crown, and no correlation of kdt or any other rate constant with energy differences were found; 8) A(ES1 - T1) and A(ET1 - so) were found to correlate with the polarizing strength of alkali metal cation perturbers; 9) although an oxygen substituent effect is observed for knr and kp of all three crowns (except kp for the 2,3-crown), the importance of non-bonded oxygen interaction with the naphthalene n-system is discounted on the basis of common trends for cation perturbation on rate constants for all three crowns and on the basis of indications from room temperature CMR studies; 9) there is an increase in knr’ kp’ and k for 03+ per- dt turbation of all three crowns; and 10) the difference in magnitude between Cs+ perturbation of the 1,5-crown and the 2,3- and l,8-crown implies that the effectiveness of the perturbation depends on where and how the cation is held. Validity_of Method Used for Study of Perturbation Due to Alkyl Halides via Complexed Haloalkylammonium Chloride Salts The discussion now turns to a consideration of the perturbing effects of complexed haloalkylammonium chlorides in terms of the 243 perturbing effects due to complexation of the ammonium function. The effects of these two different kinds of perturbations have to in some way be separable in order for any conclusions about the perturbation due to the alkyl halide portion of the molecule to be drawn. The proposed procedure is to compare the effects due to alkylammonium perturbation to those due to haloalkylammonium perturbation. Ef- fects due to alkylammonium perturbation are of interest in themselves but will assume here only an ancillary role in considering the ef- fects due haloalkylammonium perturbation. Estimates of kf, knr’ kp, and kdt for alkylammonium and halo— alkylammonium complexes of crowns l, g, and 3 are presented in Tables 37, 38, and 39, respectively. If estimates are available (Tables 37 and 38) for all alkylammonium complexes of crowns l and 2, so calculation of rate constants for these complexes requires no assumptions other than k is negligible (see introduction section ds on excited state processes). Tf is not known for the NH4C1 and 27 propylammonium complexes of crown %, but changes in the intensities of the S1 bands of the complexes relative to free crown g are small, so it is not likely that kf for the complexes is much different than kf for the free crown (vide infra). This assumption is also reason- able in view of the fact that ¢f, ¢p’ and T of these complexes are P not very different from those of free crown (see Table 25). There- fore, knr for these complexes was calculated assuming that kf - 3.3 x 106 (same estimate as for free crown). Comparison of Table 39 with Tables 37 and 38 will show that crown a is the crown that is least perturbed by complexation with NH4C1 or BTPINHBCI, a sur- prising result. 244 Table 37. Estimates for Rate Constants of Excited State Processes of 2,3-Naphtho-22-Crown-6 ) and Ammonium, Alkylammonium, and Haloalkylammonium Chlor de Complexes of Crown 1 in Alcohol Class at 77 K.8 System kfxlO“6 knrxlO-6 kpxlO2 kdt 2,3-Cr-6 (l) 2.8 7.9 2.4 0.33 l + NHACl 1.8 9.7 3.6 0.32 l + n-PrNHBCl 1.9 15 3.7 0.38 l + i-PrNH3Cl 1.9 8.4 3.0 0.30 l + t-BuNH3Cl 2.2 7.3 2.9 0.29 i + Br(CH2)2NH3C1 —- 10b 14c 1.3C l + Br(CHZ)3NH3Cl --- 17 12 0.88 l + I(CH2)2NH3C1 -- 65 51 4.3 i + 1(0112) 3NH3C1 --- 34 23 2.8 a . All rate constants in terms of sec 1. knr for haloalkylammonium chloride complexes calculated assuming kf - 2.0 x 10'6 (see text). ckp and kdt for haloalkylammonium chloride complexes calculated using shorter phosphorescence lifetime from double exponential analysis. 245 Table 38. Estimates for Rate Constants of Excited State Processes of l,8-Naphtho—21-Crown-6 ) and Ammonium, Alkylammonium, and Haloalkylammonium Chlor de Complexes of Crown,% in Alcohol Class at 77 K.8 6 6 2 System kfxlO knrxlO kpxlO kdt 1,8-Cr-6 (2) 2.4 20 3.5 0.38 g + NH401 3.2 15 3.5 0.55 g + n-PrNH301 3.5 16 3.3 0.56 g + i-PrNH3C1 3.2 15 3.4 0.49 ,2 + t-BuNH3Cl 2.2 17 3.3 0.38 b c c g + Br(CH2)2NH3Cl 3.6 15 5.3 0.82 g + Br(CH2)3NH3Cl 3.5 15 4.9 0.82 g + I(CH2)2NH3C1 —- 19 7.1 0.93 g + 1(0112) 33111301 -- 18 4.6 0.79 éAll rate constants in terms of sec-1. bknr for haloalkylammonium chloride complexes calculated assuming kf a 3.5 x 106 (see text). ckp and kdt for haloalkylammonium chloride complexes calculated using shorter phosphorescence lifetime from double exponential analysis. 246 Table 39. Estimates for Rate Constants of Excited State Processes of 1,5-Naphtho-22-Crown-6 ) and Ammonium, Alkylammonium, and Haloalkylammonium Chlor de Complexes of Crown,3 in Alcohol Class at 77 K.8 '6 -6 2 System kfxlo “an10 kpxlo kdt 1,5-Cr-6 (3) 3.3 25 8.2 0.37 ,3 + 3111401 --- 28b 8.6 0.37 3 + n-PrNH3Cl -- 26 10 0.37 ,3 + 131-(0112) 21411301 --- 50 22c 1.00° 3 + Br(CH2) 33114301 -—-- 42 27 0.73 ,3 + I(CH2)2NH3C1 --- 290 82 4.7 3 + 1(032) 31111301 --- 220 56 2.9 gAll rate constants in terms of sec-1. bknr for all complexes calculated assuming kf a 3.3 x 106 (see text). ckK and kdt for haloalkylammonium chloride complexes calculated using 3 orter phosphorescence lifetime from double exponential analysis. 247 The assumptions used to arrive at the rate constant estimates given for the haloalkylammonium complexes requires somewhat more comment. As noted in the results section on lifetimes, none of the fluorescence or phosphorescence decay curves for haloalkylammonium complexes (with the possible exception of the fluorescence decay from the bromoalkylammonium complexes of crown,%) are of good single exponential decay charaCter. This is not surprising, since one would expect single exponential decay only if one kind of perturbed species were present. Given the conformational freedom of the alkyl chain, one would not expect the halogen of the alkyl halide portion of the salt to always be in the same position relative to the chromophore unless there were sufficient attractive interaction between the halogen and the naphthalene n-system to compensate for loss of entropy. Thus, one would expect that the observed decay to be a composite of several decays, since each position of the halo- gen relative to the chromophore might give a different decay, unless there were no directional or distance dependence for the external HAE. Given the limitations of multi-parameter fits and the quality of the data, no more than double exponential decay fits were possible (see results section). But the estimates from such fits for the decay curves from haloalkylammonium complexes are reasonable. Com- parisons of triplet lifetimes for alkylammonium complexes (all from good single exponential fits, Tables 23, 24, 25) to the longer life- time estimates from the double exponential fits for the haloalkyl- ammonium complexes (Table 31) shows that the latter are about what 248 one would expect for an alkylammonium complex. For crown 3, Tp for alkylammonium complexes ranges from 2.4 to 3.1 sec.; the longer Tp estimate for the haloalkylammonium complexes range from 2.3 to 3.0 sec. For crown a, Tp for alkylammonium complexes ranges from 1.7 to 2.4 sec.; the longer 1 estimates for the haloalkylamr P monium complexes range from 2.3 to 2.8 sec. For crown 3, Tp is 2.1 and 2.4 sec for the NH4C1 and n-PrNHBCl complexes, respectively; the longer Tp estimates for haloalkylammonium chloride complexes range from 1.2 to 2.3 sec. Therefore, a reasonable interpretation of the data is that the halogen atom is sometimes too far away and/or in the wrong position relative to the chromophore for effective per- turbation. Thus, it appears reasonable to assume that the species responsible for the long I is similar to an alkylammonium complex. P It does not seem possible, however, to put a reasonable limit on the number of short decays present. However, the shorter Tp esti- mates given by the double exponential analyses are reasonable, hows ever. For haloalkylammonium complexes of crowns 3 and 3, the shorter Tp estimates for the iodoalkylammonium complexes are three to five times shorter than for the bromoalkylammonium complexes. This is as should be, since alkyl iodides are better heavy atom perturbers than alkyl bromides. Also, the estimates for the bromo- and iodoalkylammonium complexes (1.0 and 0.2 sec., respec- tively) are similar to the first "half-life" estimates for perturba- tion of naphthalene by propyl bromide and iodide glasses at 77 K 3 (0.14 and 0.076 sec., respectively).1 (First "half-life" is defined as the time required for the intensity to drop to one-half its 249 initial intensity). While this method of analysis neglects the multi- exponential character of the decay, it was found that the lifetimes obtained were in good agreement with estimates from the initial slope of a plot of the logarithm of intensity versus time.) The estimates from this investigation may be somewhat longer because there is essentially only one heavy atom in the vicinity of the chromophore, whereas in a heavy atom glass the chromophore is surrounded by heavy atom containing molecules. The estimates for the shorter lifetimes for the haloalkylammonium complexes of crown £ are about the same for both bromo and iodo analogs and for both chain lengths. This must mean that the halogen is oriented in such a way relative to the chromOphore such that its interaction is less effective. This result does not mean that 1,8- disubstitution interferes with the effect of the heavy atom, since ethyl bromide is found to perturb both crown % and alkali metal com— plexes of crown 2’. If the decay curves could accurately be analyzed accurately for several decay rate constants and for the fraction of the total decay associated with each constant, quantum yields for each decay could be found assuming that all species have the same extinction coef- ficient. The estimated fraction of the total decay due to the faster decays ranges from 0.5 to 0.8. These estimates are probably good enough to warrant the conclusion that most of the decay is from species perturbed by the alkyl halide portion of the haloalkylamr monium cations but probably not good enough to warrant adjustment of quantum yields before rate constant calculations. The proposed 250 procedure for calculation of rate constants is to use the shorter lifetime estimates and quantum yields uncorrected for contributions from the longer decay. This procedure will tend to underestimate rather than overestimate changes in rate constants relative to un- perturbed crown. This approximation will be best for the haloalkylammonium com- plexes of crown‘z, since ¢f and ¢p don't change very much relative to what they are for the alkylammonium complexes (compare Tables 24 and 27), yet the shorter decay is estimated to account for most of the decay. In general, but especially applicable to crowns k andl3, decreases in ¢f due to heavy atom perturbation will be underesti- mated, so (assuming kf doesn't change much, see below) increases in knr will be underestimated. Increases in ¢p will be underesti- mated, so increases in kp will be underestimated. kdt is generally about 102 larger than kp’ so kdt won't be affected very much (kdt - Tp-l - kp). Under the usual assumption that the HAE affects the rate constants of spin forbidden processes the most, it will be assumed that kf is not affected very much by the external heavy atom. (Note, however, that kf is the rate constant most affected by Cs+ perturbation of crown 3). For crown %,and 3, however, it is known that kf is changed by complexation of alkylammonium cations. Tables 37 and 38 show that kf is practically invariant for NHACl, n—PrNH3C1, and i-PrNH3Cl complexes of crown l (kf about 2.0 x 106). The fluorescence decays of the haloalkylammonium complexes of crown 3 appears to be double exponential, but the estimates are rough (see results section). 251 It was preferred to calculate knr for this series of complexes using the same value of kf in all cases rather than to have the values for the bromo analogs reflect experimental and analytical uncertainty and to have the values for the iodo complexes reflect the uncertainty of a guess. So knr for these complexes was calculated assuming kf - 2.0 x 106, a typical value for an alkylammonium complex. kf is also practically constant for the alkylammonium complexes 6 to 3.5 x 106; see Table 38). of crown % (kf ranges from 3.2 x 10 The fluorescence decay for the B- and y-bromo analogs gave a better single exponential analysis than double exponential analysis (see results). kf calculated using the If estimate from single exponen- tial analysis gives an estimate in either case of about 3.5 x 106 sec-1. Which is about the same as kf for an alkylammonium complex. So k c for the iodoalkylammonium complexes of crown g'was calculated is assuming kf - 3.5 x 106. Note that it is possible that fluores- cence and phosphorescence decay might not both be multi-exponential dt is affected by the perturbation. The results for the bromoalkylammonium com- in character, depending upon how knr vis-a-vis kp or k plexes 0f crown % suggests that knr is not greatly affected by the external heavy atom but kp and/or kdt are, since the fluorescence decay appears to be single exponential but the phosphorescence decay multi-exponential. Haloalkylammonium cations complexed by crown g'do seem to have some effect (perhaps due to the presence of the external heavy atom) which is different than would be predicted on the basis of changes in (ET1 - so). It appears that there is a good linear correlation 252 (R - 0.97) between log kdt and A(ET1 - so) for alkylammonium complexes, but that log k is either independent of A(ET1 - so) for the halo- dt alkylammonium complexes or that two of the points for the haloalkyl- ammonium complexes fit the same correlation as for the alkylammonium complexes but that two of the points don't (see Figure 70). For the alkylammonium complexes of crown,%, kdt appears to 227 £££é§2.33 the energy separation A(ET1 - 30) increases. The excep- tion in the case of n-PrNH3C1 may be explained by noting that, al- though the 0-0 phosphorescence band is blue shifted, that all other peaks are red shifted (Figure 32). Thus, the larger value of kdt than would be expected based on A(ET1 ' so) estimated from the 0-0 band may be because higher vibrational states of So are brought closer to T1 by red shifts. For the haloalkylammonium complexes of crown &, log kdt is either independent of A(ET1 - so) or decreases much more rapidly as a function of increasing A(ET1 - so) than for other plots of log kdt versus A(ET1 - so) (R B 0.97; see Figure 66 and 70). Note that the 0—0 peak (used to determine AE(T1 - SO) in Figure 70) is blue shifted by haloalkylammonium perturbation but that the rest of the peaks are red shifted (Figure 38). Since this is also true for n-PrNH3Cl perturbation, the halogen itself does not appear to be the cause of the shifts. Method for Consideration of Perturbation by Ethyl Bromide Containing_ Glass Calculation of rate constants from observed values of ¢f, ¢p’ and Tp for perturbation by ethyl bromide-ethanolmethanol (1:4:1, v/v) 253 ' e PsA NH cu ©© a - 4 = 0.6 - ©© + n-PrNH3CI 0 x .P none __ __ <> .P=none 0 i-PrNH3CI _ I t-BuNH3CI a 0.4 -— A Br(CH2)2 NH3CI "‘ * Br(CH2)3 NH3CI _ ' o 1(CH2)2NH3CI 0.2 — u 1(CH2)3NH3CI _ A a? 0.0 — u o .. * A _ l: 72:02 — - xv _. + A . .. 8‘ " “Q4 " I x -l- "l r— : A - * -O.6 - _. I l l l ‘300 ‘200 -I00 I00 200 300 Change in Triplet Energy Relative to Crown. . -1 ‘—'—"4’ Increasing (T|-So) Separatnon (crn ) Figure 70. Plot of log kdt versus the change in the T1 — 80 energy separation in cm-1 caused by ammonium, alkylammonium, and haloalkylammonium cation perturbation of crowns l and 7%- 254 glass would not be meaningful, since the observed quantities are probably complicated composites arising from different values due to different relative orientations and distances between the perturber and the perturbed chromophore. As is usually done for this kind of study, conclusions will be drawn based on the way in which ratios of quantum yields and ratios of lifetimes (Tp) change. Ratios in Table 40 are given so that all ratios increase as a function of per- turbation in order to make comparisons of relative decreases and increases easier. The ratio of ¢f in alcohol glass to ¢f in ethyl bromide (EtBr) containing glass (¢f,o/¢f,EtBr)’ 0p in EtBr containing glass to ¢p in alcohol glass (¢p,EtBr/¢p,o) and T in alcohol glass P to Tp in EtBr containing glass (1 ) are tabulated in Table p,o/Tp,EtBr 40 for naphthalene, crowns l, g, and 3, and alkali metal complexes of these crowns. The Tp,EtBr values used are the shorter lifetime estimates from a double exponential analysis (see Table 32). Longer lifetime estimates are in the neighborhood of 1 sec except for complexes of crown 2 and the Cs+ complex of crown 3. In the latter case, the decay is better represented by a single exponential analysis than double. The longer lifetime estimate for naphthalene perturbed by EtBr (0.72 sec) is similar to a literature estimate for naphthalene (1.14 sec) in ethanol-ethyl bromide (4:1, v/v) glass at 77 K which was based on the "first half-life" of the decay (vide infra).53 The shorter lifetime estimate for naphthalene (0.042 sec) is similar 13 to a literature estimate (0.14 sec) in propyl bromide glass at 77 K which was also based on the "first half-life". Thus the analytically 255 Table 40. Estimates for Ratios of Fluorescence (¢f o) and Phosphor- ’ escence (¢p 0) Quantum Yields and Phosphorescence Life- 9 times (TP 0) in Alcohol Glass to the Corresponding Quantum , Yields (¢f,EtBr and ¢p,EtBr) and Lifetimes (Tp,EtBr) in Ethyl Bromide-EthanoléMethanol (1:4:1, v/v) Glass at 77 K. ¢f,o ¢p,EtBr Tp,o a System d>f,EtBr ¢p,o Tp,EtBr Naphthalene 4.0 12.7 60 2,3—Cr—6 (%) 1.5 3.8 20 l + KCl 1.8 3.4 15 l + CsCl 1.5 2.4 14 l,8-Cr-6 (2) 2.0 2.3 46 g + KCl 1.7 2.8 110 g + CsCl 1.8 2.2 46 1,5-Cr-6 (3) 1.6 1.9 14 3 + KCl 1.4 1.8 13 g + RbCl 1.2 1.33 7 g + CsCl 1.0 1.00 1.3 an EtBr is shorter lifetime estimate from double exponential analysis. 9 256 obtained estimates of lifetimes for naphthalene as perturbed by EtBr are in good agreement with literature precedent. The discussion now turns to a consideration of the questions posed in the preamble to the discussion. Directional Dependence of External Heavy Atom Effect It had initially been hoped that differences in metal cation perturbation of crowns 3,33, and 3 would provide an experimental demonstration of the directional dependence or lack thereof for external perturbation. As pointed out above, there are at least two factors which cloud such a demonstration by these model systems: 1) the steric blocking due to the naphthylic methylene units in crowns 1 and 2, and 2) the indeterminacy of the position of cations complexed by crowns %, z, or 3. The results of perturbation by Cs+ and Ag+, however, shows that there is either a strong directional and/or distance dependence for external perturbation by metal cations, since the effects of perturbation by Cs+ or Ag+ on crown 3 are much larger than those for crowns 3 or g. The strongest piece of evidence for a directional dependence for the external HAE comes from perturbation of naphthalene and crowns 3, £3 and 3 by ethyl bromide containing glass (ethyl bromide-ethanol- methanol (l:4:l)) at 77 K. Table 40 shows that naphthalene is more perturbed (iygy, quantum yield and lifetime ratios are changed the most) by EtBr than any of the crown substituted naphthalenes. Crowns,3 and‘% are perturbed by EtBr to roughly the same extent, but crown 3, for which the crown ring passes over the fl-face, is least 257 perturbed. This is the best direct experimental demonstration to date to this writer's knowledge that an aromatic hydrocarbon like naphthalene is most susceptible to external perturbation in the vicinity of the n-face. As indicated in the background section on the HAE (see introduction), other authors have thought this to be the case, based on various experimental indications, but have not been able to directly demonstrate it. The results of perturbation of crown 3 by complexed haloalkyl- ammonium salts also provides evidence that the w-face is most sus- ceptible to a heavy atom perturber. Table 39 shows that knr’ kp’ and kdt are larger in each case for the B-haloalkylammonium complexes than for the y-haloalkylammonium complexes (kp is approximately the same for the y- and B—bromoalkylammonium complexes. This difference in behavior is consistent with the conclusion that the n-face is most susceptible to the external HAE, since, given the geometric con- straints imposed by crown 3, the halogen on the three carbon chain can't attain the same degree of proximity to the n-face as can the halogen on the two carbon chain without the alkyl chain assuming a much more strained conformation. For the haloalkylammonium complexes of crown 3, k nr’ kp, and kdt are generally larger for the B-haloalkylammonium complexes than for the y-haloalkylammonium complexes (kp is about the same for the B- and y-bromo cases and knr is larger for the y-bromo than for the B-bromo). This is not what one might expect for this crown, since in this case the halogen on the longer chain should be able to reach out over the fl-face better than the halogen on the shorter chain. 258 Corey-Pauling-Katum (CPK) models show that for crown.3, the halogen on the two carbon chain can just reach the edge of the n-system, whereas the halogen on the longer chain can reach out further over the n—face. The fact that larger effects are not observed for the n-halo analogs does not contradict the conclusion reached for the haloalkylammonium complexes of crown.3. The spatial constraints imposed by crown,3 for a complexed haloalkylammonium salt are dif- ferent from those imposed by crown 3. For crownl3, the halogen is more likely to be in the vicinity of the n-face for either the long or the short chain, since the site of complexation of the ammonium group is over the face, although the longer chain tends to keep the halogen closer to the outer edge of the n-system. An indication of this greater likelihood is that kp and kn: for a haloalkylammonium complex of crown 3 are larger than for the corresponding complex of crown 3. For crown 3, the site of complexation of the ammonium group is on the side of the naphthalene nucleus. Barring attrac- tive interactions between the halogen and the n-system, the likeli- hood of the halogen being in the vicinity of the n-face will be governed mainly by statistics. As the chain length increases, the number of conformational degrees of freedom increase, and the probability of the halogen being in the vicinity of the n-face decreases. In fact, the smaller effects observed for the fi-halo analogs suggests that any enthalpy gain due to attractive interactions between the halogen and the n system has to be less than the entropy loss which would be required. For haloalkylammonium complexes of crown 3, knr’ kp, and kdt 259 are not much larger than for the alkylammonium complexes (Table 38). This is an indication of either a strong distance or directional dependence for the external HAE. (This also suggests that there is very little tendency for the halogen atoms of the haloalkylammonium salts to be near the naphthalene nucleus.) Distance Dependence of the External Heavy Atom Effect Both exchange and CT mechanisms for the external HAE are pre- 4 distance dependence (where S is an inter- dicted to show an S molecular molecular orbital overlap62). Therefore, one would expect the external HAE to be subject to a very sensitive distance depend- ence. Other studies have experimentally indicated that there is a distance dependence (see background section of introduction) but weren't able to specify the degree of sensitivity. Results from this investigation also indicate a distance dependence. Although no precise formulation for how the magnitude of the effects vary as a function of the distance between the perturber and perturbed chromOphore will be possible, the fact that relative spatial orienta- tions are better known than for other investigations will allow some qualitative observations on the sensitivity of the effect as a func- tion of distance to be made. Results from the CMR work previously referred toe3 indicate that, for a given crown, the position of the center of positive charge is about the same for all alkali metal complexes. The evidence for this is that all alkali metal cations induce very nearly the same change in the chemical shifts of the naphthalene carbons for a given 260 crown (the shifts are somewhat more variable for l,8-Cr-6 (g)). The same is true for the alkaline earth metal cations, but the magni- tudes of the changes in CMR shifts are larger than those induced by the alkali metal cations. Also, for 2,3-naphtho-l7-crown-5 (2,3- Cr-S) and for l,8—naphtho—18—crown-5 (l,8—Cr—5), the magnitude of the changes induced by alkali metal cations are larger than for 2,3- Cr-6 (l) or l,8-Cr-6 (a). These results show that the chemical shifts of the naphthalene carbons are a function of the magnitude and position of the center of positive charge. Changes in the chemical shifts of the ether carbons induced by complexation of metal ions are consistent with either small conformational changes or with small adjustments of the center of positive charge. On the basis of the evidence from the changes in the chemical shifts of the naphthalene carbons, it is thought that the former interpretation is correct. Thus, if the center of positive charge is in roughly the same place for all alkali metal complexes of a given crown, then the smaller cations will be further away from the chromophore. The smaller effects due to perturbation of a given crown by light cations, then, might in part be because the higher cations (Na+ and K+) are smaller than the heavier cations (Rb+ and Cs+). Perhaps the best indication that the less effective perturbation by light cations is due in part to having their regions of electron density (but perhaps not their nuclei) further away from the chromo- phore comes from alkali metal cation perturbation of crown‘e. For typical internal or external HAE for a series of heavy atom perturbers, it has been found that there is a linear relationship between the 261 logarithm of the perturbed rate constant and the logarithm of the 5 It was thought square of the spin-orbital coupling parameter (E).8 that the rate constants for alkali metal cation perturbation might show such a correlation. But no such linear relationship was ob- served for log kn: or log kp versus log 52. The plots show a dis- proportionately large increase in log k as function of log g2 for Cs+ perturbation of crown,%. This suggests that the other alkali metal cations are less effective perturbers than would be expected. The most reasonable explanation for this seems to be that smaller cations are held with their van der waals radius further away from the chromophore. If this explanation is correct, then one would expect, say, Rb+ to be a more effective perturber of 1,5-naphtho-l9-crown-S than 1,5- naphtho-ZZ-crown-6 (a). This smaller 1,5-crown has been synthesized, but the effects of alkali metal cation perturbation have not yet been investigated. Alkali metal cation perturbation of smaller 2,3- and 1,8-crowns (2,3-naphtho-l7-crown-5 and l,8-naphtho-l8-crown-5) has been investigated. Cs+ was found to be a more effective perturber of 2,3-Cr-5 than of 2,3-Cr-6 (l) (i.e., knr’ k , and kdt all increased P more in former case than in latter). (Note, however, that Cs+ forms complexes with Cs+ :crown - 1:2 stoichiometry, since the break in the titration curve is at the Cs+:crown - 1:2 point. It is not known whether or not a 1:1 complex is formed at higher Cs+:crown ratios, since no further break in the titration curve is observed. The more effective perturbation of the smaller 2,3-crown may be because the Cs+ cation might be complexed in such a way that it is positioned 262 above the plane of the naphthalene nucleus and in closer proximity to the fl—system.) CS+, however, is a less effective perturber of l,8-Cr-S than of l,8-Cr-6 (%)’ Given what has been learned, one would expect different size 1,5-crowns to be more useful in discern- ing distance effects than different size 2,3- or l,8-crowns, since the latter have the steric interference due to the naphthylic methylene groups. A rough estimate of the degree of sensitivity of external heavy atom perturbation as a function of distance can be made from the deviation from linearity of the log k versus log 52 plots for k, k , and k p dt Of crown‘e 85 Perturbed by complexed alkali metal cations. Assuming that the center of charge is in the same place (which could be the case based on CMR work with crowns l and 2), then Rb+ would be 0.21 A further from the n-system than Cs+. (Because of the rigidly and widely spaced naphthalic oxygens of the 1,5-crown, it is also possible that the cations would tend to associate with the oxygens in the more flexible part of the ring.) Assuming that the perturbation due to Na+ would be small in any case (like an alkyl fluoride) and connecting the points for Na+ and Cs+ on the log-log plot by a straight line, then knr and kp for K+ (0.36 A smaller than Cs+) and Rb+ perturbation are three to five times smaller than would be predicted by the straight line. This suggests that even a few tenths of an angstrom difference in distance can greatly decrease the effectiveness of an external heavy atom perturber. 263 Effects of the Positive Charge of the Perturber 83 show that all alkali As has already been indicated, CMR studies metal cations complexed by 2,3—Cr-6 or l,8-Cr-6 cause shifts in n-electron density that are similar for a given crown, but that cations complexed by 1,5-Cr-6 cause smaller shift in electron density. This shows that the positive charge, independent of any other proper- ties of the cation, can perturb the chromophore, but that these shifts in electron density are unrelated to the observed changes in rate constants, since different cations have different effects on rate constants of excited state processes, but all ions of the same charge cause similar shifts in w-electron density (in the ground electronic state, anyway). The fact that light cations (Na+ and K+) induce only small changes in rate constants argues against any large perturbation due to the charge itself. The results of alkylammonium perturbation provide another indication that the mere presence of a center of positive charge has little perturbing effect on the rates of excited state processes. What is perhaps most surprising is that complexation of positively charged species in the vicinity of the n-face gig Crown 3 has little perturbing effect. It had initially been thought that a comparison of perturbation by 03+ and Ba++ might provide an indication of the role (if any) + and Ba++ have similar of the positive charge of the perturber. Cs nuclear charges, so it is thought that they might both be capable of inducing similar enhancements of spin-orbital coupling. All other things being equal, any differences in their perturbations 264 would be ascribable to their difference in ionic charge. Unfor- tunately, other things are not equal, since Ba++ is smaller than and less polarizable than Cs+. The fact that Ba++ is a much less effective perturber of crownle than Cs+ is a striking result (see Table 36). But this may be due to the fact that Ba++ is smaller than Cs+ (more similar in size to K+) and, therefore, it may be complexed in such a way that it is much further away from the n-face than is Cs+ (see above under distance effects). Alternatively, Ba++ is not as polarizable as Cs+, and it is possible that the degree of polariz- ability as well as the magnitude of the nuclear charge determine an atom's effectiveness as an external heavy atom perturber. Thus, it is thought that HAE due to rare gas matrices are a function of their physical polarizability, since chemical interaction is thought to be unlikely.86 Perturbation of a smaller 1,5-crown by Ba++ would perhaps indicate whether its ineffective perturbation of crown,e is a function of its smaller size vis-a-vis distance or of its lower polarizability. Comparison of perturbation of crown % by Cs+ and by Ba++ (Table 34) shows that, for this crown, Ba++ is as effective as 03+. kf is perturbed somewhat less by Ba++ than by Cs+ (3-fold decrease versus S-fold decrease), knr is perturbed more by Ba++ than by Cs+ (2-fold increase versus 0.5-fold increase), and kp and kdt are perturbed to approximately the same extent. Ba++ has about the same effect on crownlg as does Cs+, but note that (vide supra) k as dt perturbed by Ba++ does not correlate with A(ET1 - 30), whereas kdt as perturbed by Cs+ does (see Figure 66). These results for crowns 265 ,l and,% show that Ba++ and Cs+ are similar and yet different per- turbers, but it is not possible to say from these results which dif- ferences in properties of Ba++ and Cs+ account for the differences in perturbation. The greater similarity between Ba++ and Cs+ per- turbation of either crowns'l orig as compared to the great dissimilarity between Ba++ and Cs+ perturbation of crown 3 may be a function of the fact that the effects of Cs+ perturbation in the former cases is much smaller than in the latter case. Necessity of Charge Transfer States for the External HeavygAtom Effect As indicated in the background section on the heavy atom effect (HAE). McGlynnSB concludes that it is theoretically difficult to choose between an exchange and a charge transfer (CT) mechanism for the external HAE. He prefers a CT mechanism based on various experi- mental indications; however, his interpretations and arguments were criticized in the background section both on the basis of observations by other investigators and on logical grounds. The results of this investigation show no spectral evidence for charge transfer character except perhaps in the case of the Ag+ complex of crown‘e (vide infra). One of McGlynn's arguments for the involvement of CT character in the interaction between perturbing alkyl halide and the naphthalene chromophore was based on red shifts of naphthalene's O-O phosphores- cence peak induced by propyl halide glasses at 77 K (see background 266 section of introduction for fuller description of observations and arguments made from them). While the observed red shifts cannot be disputed, the observations of this investigation show that perturba- tion of crowns'l,’%, andle by haloalkylammonium chlorides do not induce similar red shifts but do enhance spin-orbital coupling, as evidenced by increases in knr’ kp, and k This shows that CT, dt' as evidenced by red shifts anyway, is not necessarily involved in 80 enhancement by external alkyl halides. For crown'l, complexation of y-haloalkylammonium salts induces blue shifts of the T1 O-O peak relative to free crown similar in magnitude to those induced by alkali metal and alkylammonium cations. Also, note that the magnitude of the blue shift for y-bromopropyl- ammonium perturbation is very similar to that due to Efpropylammonium perturbation. This indicates that the properties of the bromine have nothing to do with the shift. It might be argued that the failure to observe a red shift in the T1 O-O peak is because the blue shift induced by complexation obscures a red shift due to CT interaction with the n-system by the halogen. Indeed, smaller blue shifts are observed for perturbation of crown.l by B-haloalkylam— monium perturbation, which are more effective perturbers of crown % than the y-analogs, as has been previously noted. But the blue shifts are smaller by about the same amount (about 60 cmfl) for per- turbation by the B-bromo and the B-iodo isomers. A smaller blue shift should have been observed for perturbation by the B-iodo isomer if 88 CT was involved. McGlynn noted a red shift of 325 cm"1 for propyl iodide perturbation and of 155 cm.1 for propyl bromide 267 perturbation of naphthalene. Also, as has already been noted, while haloalkylammonium perturbation of crown‘l red shifts other peaks in the phosphorescence spectra, so also does EfPrNH3C1 perturbation. In a similar vein, B-halo perturbation of crown‘l red shifts the 82 0-0 absorption peaks by 500 cm-1, so also does ngrNH301 perturba- tion. For crown £9 complexation of all perturbers induces red shifts. The y-haloalkylammonium are somewhat more effective perturbers of crown 2 than the.y-halo analogs (see Table 38), but they induce smaller red shifts. It might be argued that this is because the halogen is not in the right position to form a CT complex with the fl-system, which may be true, but these perturbers do increase knr’ kp, and kdt’ though less so than for perturbation of crowns I or Q. But this is another indication that 80 coupling does not require a CT complex. For crown 3’ which is the crown most effectively perturbed by complexed haloalkylammonium cations I compare Table 39 to Table 37 and 38), only small blue shifts of the T1 O-O peaks are induced by complexation of these perturbers, except for a small red shift (30 cm-l) induced by perturbation by the B-bromo analog. But perturba- tion by NH4C1 also produces a small red shift (10 cm-l). Most per- turbers induce only small blue shifts (10 to 40 cmfl). These blue shifts are small enough so that they wouldn't obscure red shifts of the magnitude observed by ficGlynn. It might be argued that the failure to observe red shifts due to the interaction of the halogen and the w-system is because the 268 emission spectra are composites of emission from perturbed and un- perturbed species. But the results of double exponential analysis of the phosphorescence decay curves (Table 31) indicate that shorter decay accounts for 50 to 80% of the total decay. Furthermore, for crown %’ for which energies change least and perturbation is most effective, spectral shapes are unchanged by haloalkylammonium cation perturbation. It is for this case that observation of a red shift due to CT interaction of the halogen would be least likely to be obscured by opposing energy shifts due to perturbation due to complexation or changes in spectral shapes. But red shifts aren't observed. Therefore, it is concluded that there is no evidence for CT involvement in mediating the eternal HAE due to complexed halo- alkylammonium perturbers. There is a general tendency for complexation of crown % to induce small blue shifts, but in the case of Ag+ complex of crown‘% every spectral O-O peak is red shifted relative to free crown (see Table 4). This might be taken as an indication that n-molecular orbitals of naphthalene interact rather strongly, and with net stabilization, with atomic orbitals of Ag+. This is not unreasonable, since Ag+ is known to form n-complexes.73 Note that Ag+ perturbs knr’ kp, and kdt of crown,é differently than Cs+. knr appears to be perturbed to about the same extent by both Ag+ and Cs+, k.p is increased about 60—fold relative to crown versus a lO-fold increase due to Cs+, and k is increased about 2-fold by Ag+ but about 5—fold by Cs+ dt (see Table 36). Evidently, depending upon the kind of interaction, different rate constants may be affected differently. Thus, while 269 some external perturbers may form weak CT complexes, it does not appear that CT character is necessary to the external HAE, since Ag+ and Cs+ both have external HAE and it is doubtful that much CT occurs to Cs+ (unfortunately, the reduction potential of Cs+ in 95% ethanol is not known). .Effectiveness of External Perturbation as Function of Perturbation Already Present The results of EtBr perturbation (EtBr-ethanol-methanol (1:4:1, v/v) glass) of alkali metal complexes of crowns l,‘2, and a nicely fit with McGlynn's observations for the effectiveness of a given external heavy atom perturber as a function of the spin-orbital perturbation already present due to an internal heavy atom. McGlynn13 observed (see background section in introduction for more complete description) that as the internal spin-orbital coupling increased, that the ratio of the external effect to internal effect decreases. Table 40 shows that for alkali metal complexes of crowns l and e, /T ¢p EtBr/¢p 0’ or TP 0 both decrease as the spin-orbital ! 9 9 coupling parameter of the complexed alkali metal cation increases. p,EtBr Also, note that the ratio of the external to the internal effect is smallest for the "internally" most perturbed crown, the Cs+ complex of crown‘g. ¢f,o/¢f,EtBr (all ratios are given so as to be greater than 1) decreases less than ¢p,EtBr/¢p,o- This indicates that most of the change in ¢ 13 due to changes in RP’ not to increases p,EtBr/¢p,o in ¢isc (assuming that kf isn't affected and that kd8 is negligible; see introduction). Note that the only apparent anomoly is for the 270 K+ complex of crown %. It is perturbed more by EtBr than are free crown,% or the Cs+ complex of 2. This anomoly may in some way relate to the decrease in knr caused by Rf perturbation of crown'g (see Table 35). McGlynn's observations were for covalently bonded per- turbers. A complexed perturber is most like an external perturber, but as long as the observations are thought of in terms of the ef- fectiveness of the external perturber (i;g;, EtBr) as a function of the spin-orbital perturbation already present, be it due to a covalently bonded perturber, a complexed perturber, or another ex- ternal perturber, there is really not much difference between the present study and McGlynn's. The results in Table 40 (especially the results for crown.§) are not rationalized by McGlynn's suggestion that external heavy atom perturbers (EtBr in this case) are more effective when a sub- stantial amount of SO coupling is already present. If it is assumed that external 80 coupling, like that caused by Cs+ complexation by crown Q, is similar in effect to internal SO coupling, the effect of EtBr on a plus complexed Cs+ should be larger than the effect of EtBr on a plus complexed Rb+. The opposite is observed. There may be a sharp difference between internally and externally promoted SO coupling, or the suggested synergistic effect of internal SO coupling may not be general. 271 Relative Susceptibility of Rate Constants for Excited State Processes As indicated in the introduction, various investigators have been interested in the relative susceptibility of rate constants for excited state processes and different conclusions about which rate constants are most susceptible have been drawn,89 This isn't surprising, since this writer sees no reason to expect that conclu- sions drawn from CT complexes containing heavy atoms,13 for instance, should have anything to do with alkyl halide perturbation. Indeed, a major conclusion of this investigation is that the relative sus- ceptibility of rate constants is a function of where the perturber is and the nature of the perturber. Since the only chromophores studied are based on naphthalene, any conclusions reached will not be assumed to be true for other aromatic hydrocarbons. Tables 41 and 42 give relative rate constant susceptibilities as a function of crown and of perturber type. These orders of relative susceptibilities were derived from Tables 34 through 39 by selecting the rate constant which was changed most by a perturber of a given type. Cs+ was the most effective perturber in all cases for the alkali metal cations except for kf (Rb+) and knr (K+) of crown'%. ‘ngrNH3C1 was usually the most effective alkylammonium perturber. The most effective haloalkylammonium perturber depends upon the crown perturbed. Their relative effectiveness was discussed for each crown in the preceding discussion. Note that when the value of kf isn't derived from Tf measurements, that knr is not included unless it is thought that the relative value of Rat would not be much in error relative to the other rate constants due to the error in the 272 Table 41. Relative Susceptibility of Rate Constants of Excited State Processes of 2,3-Naphtho-ZO-Crown-6 (l), l,8-Naphtho—21- Crown-6 (a), and 1,5-Naphtho-22-Crown-6 ) to Perturbation by Complexed Alkali Metal, Barium and Silver Cations in Alcohol Glass at 77 K. Crown Perturber Type Relative Susceptibility of Rate Constants 2,3-Cr—6 Q) Alkali Metal kf > up > knr > k dt l,8-Cr-6 (g) Alkali Metal k dt> kp > kf > km. 1,5-Cr-6 Q) Alkali Metal knr > kp > kdt 2,3-Cr—6 q.) BaH kf z kp ~ km > kdt l,8-Cr-6 (3,) Ba++ kp ~ kdt 1,5-Cr-6 (3) Ba++l kp ~ kdt 2,3—Cr-6 q.) Ag+ kp 3 k dt l,8-Cr-6 (g) Ag+ kp z k dt 1,5-Cr-6 (,3) Ag+ kp > km, >> k dt 273 Table 42. Relative Susceptibility of Rate Constants of Excited State Processes of 2,3-Naphtho-20-Crown-6 (l), l,8-Naphtho-Zl- Crown-6 (2), and 1,5-Naphtho-22-Crownr6 ) to Perturbation by Complexed Alkyl and Haloalkylammonium ations in Alcohol Glass at 77 K. Crown Perturber Type Relative Susceptibility of Rate Constants 2,3-Cr-6 (l) Alkylammonium knr > kf z kp > kdt l,8-Cr-6 (’2) Alkylamonium kf z kdt > knr > kp 1,5-Cr-6 (3) Alkylammonium knr z kp ~ kdt 2,3-Cr-6 (l) Haloalkylammonium kp > kdt > knr 1,8—Cr—6 (a) Haloalkylammonium kdt > kp > knr 1,5-Cr-6 (3) Haloalkylammonium knr z kdt > kp 274 "guess" for kf. While one's initial impression from inspection of Tables 41 and 42 might be that there are no regularities, there are perhaps some generalities which can be extracted. For the alkali metals, note that kp is the second most affected rate constant in each case. kdt is usually least affected, except for perturbed crown g. In— spection of the rest of Tables 41 and 42 will show that, for crown 2, kdt is generally either the most affected rate constant, next to it, or, when only kp and k are known well, affected to approxi- dt mately the same extent as kp. Perhaps it is the out of plane dis- tortions due to peri-interactions in l,8-disubstituted naphthalene which are responsible for this enhancement of radiationless decay (kp is usually thought to be affected more by the HAE).55 Not enough is known about perturbation by Ba++ to allow much to be said about its effects relative to Cs+. For crown l, the curious result that emerges is that kf z kp 3 knr' Also, it is -H- curious that kp z kdt for Ba perturbation of both crown l and %. Perturbation of crowns l and 2 by Ag+ also affects kp and kdt to approximately the same extent. For haloalkylammonium perturbation of crowns l, g, and 3, it is curious that the relative order of knr’ kp, and kd for each crown. For crown a, the crown that is most affected by t is different haloalkylammonium cation perturbation, the nonradiative modes, knr and kdt’ are enhanced more than the radiative modes, kp. For crown 2 kdt is observed to be perturbed most, as was also noted to be the case with alkali metal perturbation. For crown l, kp (radiative 275 mode) is affected more than either of the nonradiative modes by halo- alkylammonium perturbation. EXPERIMENTAL General Procedures Melting points were determined using a Thomas-Hoover apparatus (capillary tube) and are uncorrected. Mass determinations were done on a Mettler HZOT analytical balance (1.01 mg). Infrared spectra were recorded on a Perkin-Elmer 237B infrared grating spectrophotom- eter. 1H NMR spectra were determined on a Varian T-60 spectrometer using tetramethylsilane as an internal standard. Ultraviolet absorp- tion spectra at room temperature and at 77 K were recorded on a Cary 17 spectrophotometer. Mass spectra were determined on a Perkin- Elmer Model RMU 6 mass spectrometer. Elemental analyses were per- formed by Instranal Laboratory, Inc., Rensselaer, NY; or by Chemaly- tics, Tempe, Arizona. 77 K UV Spectra Ultraviolet absorption spectra at 77 K were determined using a Cary Model 17 spectrophotometer. A round commercial grade quartz dewar and a round commercial grade quartz sample tube (:14 mm, i.d.) were used. The dewar was centered in the light path by means of a tip on its bottom and a hole in a plate which fitted on top of the sample compartment. The sample tube was centered in the dewar by means of a collar at the top of the tube and a notched teflon ring which fitted about the sample tube and into the dewar. Absorbancies from dup- licate runs were reproducable to within 112. The glassy solution 276 277 at 77 K was compared to a nitrogen atmosphere at room temperature. The base line obtained was relatively flat up to about 260 nm, where the alcoholic solvent and quartz used start absorbing. Relative extinction coefficients at a given wavelength can be determined quite well by this method, but the relative extinction coefficients for different wavelengthsixna given spectrum are somewhat distorted due to the fact that the change in refractive index with wavelength of the liquid nitrogen, the quartz, and the glassy solvent are not compensated for by comparison to a similar blank. The results are quite adequate for our purposes, however, since 1) only relative extinction coefficients are necessary for quantum yield determina- tions and 2) the spectral shapes are qualitatively correct and the spectral intensities semi-quantitatively correct. Comparison of the extinction coefficients (6) of 2,3-naphtho-ZO-crown-6 (l) at room temperature given by 1.000 cm cells with those given by the round low temperature apparatus at room temperature (calculated assuming a 1 cm path (V 77 K/V 298 K 2 0.8) length and neglecting increased concentration due to contraction of the glass; actual tube diameter was approximately 1.4 cm) showed the latter to be larger by a factor of approximately 1.5 for the 8 band (approximately 6% 1 variation within this band) and larger by a factor of approximately 1.2 for the $2 band (approximately 5% variation within this band). (See Table 43.) These factors should be >1, due to the longer path length, but they should all be the same if there weren't any other factors involved. This precludes the possibility of calculating an effective path length for the low temperature apparatus. But extinction 278 Table 43. Comparison of Room Temperature UV Absorption Spectra of 2,3- Naphtho-ZO-Crowne6 (l) in Square Cells and in Round Dewar. In Square Cells 8 In Round Dewara 8 320.5 320.5 s C c b round 1(nm) A E 6:: 1(nm) A 6 ex esquare 320.5 0.0462 274 1.00 320.5 0.0490 408.3 1.00 1.49 313.5 0.0463 274 1.00 313.5 0.0515 429.2 1.05 1.57 306.5 0.0612 363 1.33 306.5 0.0670 558.3 1.37 1.54 285(sh 0.54 3205 11.7 285(sh) 0.465 3875 9.49 1.21 275.5 0.842 4997 18.2 295.5 0.685 5708 13.98 1.14 266.5 0.823 4884 17.8 266.5 0.680 5667 13.88 1.16 aDewar filled with water to cut down on scattered light. bCalculated assuming a 1.4 cm path length. cAbsorbance (A). 279 coefficients roughly indicating the strength of an absorption band can be calculated from the data in the figures using either the effective path lengths indicated by the above comparisons (1.5 for the S1 band, 1.2 for the S2 band) or the actual diameter of the tube, 1.4 cm. Finally, it should be pointed out that in order to obtain a room temperature spectrum in the low temperature apparatus it was necessary to fill the dewar with some liquid to minimize changes in refractive indicies at interfaces. Water (nD25 - 1.33287 190 versus n - for liquid nitrogen) was used, but it is recognized d that the above comparison may not be valid if the refractive indices of water and liquid nitrogen are greatly different or change dif- ferently in the wavelength region with which we are dealing. All UV absorption spectra at 77 K were determined on solutions prepared from 95% ethanol-methanol (4:1, v/v) or from ethanol- methanol (4:1, v/v) solvent mixtures. With the exception of the determinations made in ethyl bromide containing solutions, all emis- sion, phosphorescence lifetimes, and many of the fluorescence life- time determinations were made on solutions in 95% ethanol glass. It was necessary to add at least 20% by volume of methanol to obtain uncracked glasses in the larger sample tubes used for determination of 77 K UV absorption spectra. Addition of 20% by volume of methanol to 95% ethanol did not always result in uncracked glasses for fluores- cence lifetime determinations, however. It did seem that addition of 20% by volume of methanol to absolute alcohol (200 Proof Gold Shield Alcohol) did result in fewer cracked glasses (see section on fluorescence lifetime determinations below). It is not thought 280 that addition of a small amount of another hydroxylic solvent or removal of most of a small amount of water (ethanol and methanol were dry but not serupously so) should affect spectral or photo- physical properties.9 UV absorption spectra of 2,3-naphtho-20- crown-6 (l) were the same for solutions in 95% ethanol-methanol (4:1, v/v) and in 95% ethanol-methanol (3:2, v/v), and UV spectra of 1,5-naphtho-22-crown-6 (a) were the same for solutions in 95% ethanol/methanol (4:1, v/v) and in ethanol-methanol (4:1, v/v). Reagent Purity Sodium, potassium, rubidium, and cesium chlorides were all ultra- pure salts from.Ventron, as were cesium bromide and ammonium chloride. Sodium fluoride (Baker), sodium bromide (Matheson-Coleman-Bell), sodium iodide (Baker), barium bromide (Fisher), and cesium nitrate (Fisher) were all reagent grade salts. All of the above salts were used without further purification. Solutions of silver triflate (from Ventron, 99%) were filtered through medium sintered glass before use to remove particulate matter. Solutions of this salt remained clear and colorless for several months provided that they were stored in the dark. All solutions of the above salts showed neither UV absorption nor fluorescent or phosphorescent emission. The alkylammonium chloride salts (prepared by passing technical grade gaseous hydrochloric acid (Matheson) through a solution of the liquid amine in diethyl ether) were all purified by recrystalliza- tion. 'grPropylammonium chloride (somewhat hygrosc0pic) was re- crystallized four times from certified A.C.S. spectranalyzed acetone 281 (Fisher), mp (162-163°. and was dried for 6 h at 120°C and 0.1 m. ‘inropylammonium chloride (very hygroscopic) was recrystallized twice from acetone, m.p. 157-159°C, and dried for 18 h at 78°C and 0.1 mm. ‘thutylammonium chloride (not very hygroscopic) was recrystallized four times from acetone and ethanol (200 Proof Gold Shield Alcohol), m.p. > 300°C and was dried for 12 h at 120°C and 0.1 mm. B-Bromoethylammonium chloride (not very hygroscopic, see below for preparation) was purified by recrystallization (three times) from benzene-ethanol (reagent grade benzene, 200 Proof Gold Shield Alcohol), m.p. 130-132°C, and was dried for 12 h at 78°C and 0.05 mm. 'y-Bromoethylammonium chloride (not very hygroscopic, see below for preparation) was purified by recrystallization from benzene- ethanol), m.p. 152-153°C, and was dried for 12 h at 80°C and 0.1 mm. The stability of these salts in 95% ethanol was investigated by monitoring the nuclear magnetic resonance (NMR) spectra as a function of time. The NMR spectrum of salt that had been dissolved in 95% ethanol at room temperature for one week was unchanged compared to that of the pure salt. Also, the NMR of salt that had been dissolved in refluxing 952 ethanol was unchanged. At the concentrations used for emission and absorption experiments, these salts showed neither fluorescent nor phosphorescent emission, nor did they show absorption discernibly different from that of the alcoholic solvent in which they were dissolved. B—Iodoethylammonium chloride (not very hygroscopic, see below for preparation) was purified by recrystallization (twice) from 282 chloroform-ethanol, m.p. 151.5-154.0. y-Iodopropylammonium chloride (not very hygroscopic, see below for preparation) was also purified by recrystallization from chloroform-ethanol, m.p. loo—101°C. 95% ethanol solutions of these salts which were left exposed to room light gradually turned yellow. Solutions of these salts stored in the dark, however, remained colorless for several weeks. The sta- bility of the B-iodoethylammonium chloride in 95% ethanol was checked by monitoring its NMR spectrum (yide infra) as a function of time. The NMR spectrum of salt that had been dissolved in 95% ethanol in the dark at room temperature for 6 days was the same as that for the pure salt. After 24 h at reflux, however, the NMR spectrum of re- covered material was different. Neither of these salts showed any emission under the same conditions used for emission determinations of their crown complexes. These salts do absorb in the UV spectral region, as shown in Figures 10, 11, and 12, but, at the concentrations used, absorption by the salts is negligible in the region in which the complexed crowns were excited (296 nm or longer). Ethyl bromide (Aldrich) was purified by first stirring with BA molecular sieves for 24 h and then distilling through a 1/2" x 15" column packed with stainless steel Heli Pak (yide infra). A large middle fraction (approximately 80%) with a constant boiling point of 37°C was collected. This sample showed no UV absorption at wave- lengths longer than 300 nm (absorbance at 290 nm for 1 cm path- length was, 0.03; at 280 nm, 0.62; and at 268 nm, 0.90). Under the conditions of the emission experiments in which ethyl bromide was used, no emission from the ethyl bromide containing glass was ob- served. 95% ethanol was purified by first refluxing for 24 hours 283 and then distilling from 100 ml of 12 E sulfuric acid, followed by distillation from excess distilled water to remove diethyl ether formed in the first distillation and to produce the 952 azeatrope. 18 mm by 100 cm vacuum jacketed columns were used for these distilla- tions. The column used for the distillation from sulfuric acid was packed with 1/16" glass helcies (estimated 15-20 theoretical plates); the column used for the second distillation was packed with 100 x 100 mm stainless steel Heli-Pak (Podbielniak, estimated 90-100 theoretical plates). The take-off rate for both distillations was approximately 1 2 per 24 hours. The alcohol before purification showed a maximum in its UV spectrum at 275 nm (absorbance of 0.020 in a 1 cm cell path length). The alcohol after purification showed no absorbance (0.1 A slidewire, 1 cm path length) at wavelengths longer than 265 nm in the UV region. Furthermore, no emission from the alcohol was observed under conditions used to determine spectra or quantum yields. At very high sensitivities (approximately 100 times those used for naphthalene phosphorescence measurements), a phos- phorescent impurity with a lifetime of approximately 0.1 sec was observed. Methanol (Fisher, reagent grade) was purified in the same manner, with the exceptions that the distillations were done under a nitrogen atmosphere and that the second distillation was used to remove water. A 95% ethanol-methanol (4:1, v/v) glass gave a fluorescence decay with a lifetime of about 8 nsec, but this decay is probably due to scattered light from the decay of the lamp flash (vide infra). Pentane used for recrystallizations was purified by stirring 284 practical grade pentane (MathesoneColeman-Bell) with concentrated sulfuric acid for four hours, washing with 10% sodium carbonate, then with water, drying over calcium chloride and distilling from sodium through the column packed with Heli-Pak described above. Naphthalene (Mallinckrodt) was recrystallized eight times from ethanol/water (200 Proof Gold Shield Alcohol). The emission spectra for the stock naphthalene and naphthalene at various stages of puri- fication showed no change in fine structure or relative intensity. Both fluorescent and phosphorescent decay curves gave good single exponential fits (see experimental and results sections lifetime dis- cussion) and were close to literature values.90 These same criteria (unchanged emission fine structure and relative intensities and good single exponential fluorescent and phosphorescent decay) were used to establish the purity of all compounds investigated. 2,3-Naphtha- 20—crown-6 (1), already recrystallized to constant m.p., was re- crystallized eight times from diethyl ether-pentane (ether, Matheson- Coleman-Bell; pentane, purified as indicated above). No difference was found between the purity of the crown after recrystallization to a constant m.p. (58.5-59.00) and after the eight additional re- crystallizations. In general, compounds were recrystallized four additional times after a constant m.p. had been achieved. 2,3-Bis- methoxymethyl)naphthalene (4) (gigguigggg) was obtained as an oil. It was purified by chromatography on neutral silica gel with di- chloromethane elution, followed by molecular distillation at 55° and 0.05 mm. Its fluorescent and phosphorescent decay gave good single exponential fits and no emission outside the emission range for other naphthalene derivatives was observed. The oil quickly 285 becomes colored if left standing exposed to air, but remains a clear, colorless oil if stored under nitrogen in the dark and in the re- frigerator. 2,3-Dimethy1naphthalene (5) (Aldrich, 992 +, m.p. 102-103°) was recrystallized from ethanol/water (200 Proof Gold Shield Alcohol) to a constant m.p. of 104-1050 and sublimed (0.04 mm, 40°). There was evidence of a small amount of phosphorescent impurity in the stock sample. Its phosphorescence occurred at shorter wavelengths and overlapped the main phosphorescence spectrum. Its intensity decreased as the number of times the sample was re- crystallized increased, while the shape of the main phosphorescence spectrum was unchanged. There was a residual amount of this imr purity in the final sample used for determinations (as indicated by the slightly higher baseline before the onset of the phosphorescence), but this residual amount would have been too weak to significantly affect the determinations, since 1) the fluorescence was much more intense (no change in fluorescence was observed as a function of purification), 2) the overlap with the phosphorescence spectrum in the final sample was negligible and, as noted above, the shape of the phosphorescence spectrum did not change as the impurity phos- phorescence decreased with the extent of purification, and 3) both fluorescent and phosphorescent decay gave good single exponential decay (though the lifetime of one species can be obtained indepen- dently of that of another provided that they don't emit in the same region). All of the l,8-disubstituted naphthalene derivatives were crystalline and satisfied the above criteria for purity. l,8-Naphtho- 286 21-crown-6 (a) was recrystallized four times from ether-pentane, m.p. 54.0-55.00. l,8figi§7(methoxymethy1)naphthalene (6) was recrystal- lized from purified pentane four times. l,8-Dimethy1naphthalene (7) ‘was recrystallized four times from alcohol/water (200 Proof Gold Shield Alcohol), followed by sublimation (m.p. 63.4-64.00). All of the 1,5-disubstituted naphthalene derivatives were crystal- line and satisfied the above criteria for purity. 1,5-Naphtho-22- crown—6 (3) was recrystallized several times from ether-pentane followed by two recrystallizations from spectrograde cyclohexane (Matheson-Coleman-Bell), m.p. 55.5-57.00C. 1,51§i§7(methoxy- methyl)naphtha1ene (8) was recrystallized three times from purified pentane (vide aapra), m.p. 62—6300. 1,5-Dimethy1naphtha1ene (3) (Pfaltz and Bauer) was recrystallized four times from ethanol-water (200 Proof Gold Shield Alcohol), m.p. 80.0-81.5 Rzaparation of Samples far Emission Saectroacaay aad Qaaatum Yield Determinations All emission spectra were from 95% EtOH solutions. Pyrex NMR tubes (Willmad, 505 PS, 4.26:.13 mm i.d., 5 mm o.d., highly polished) were used for samples excited at wavelengths longer than 296 nm (the pyrex in these tubes starts absorbing at about 310 nm). Quartz NMR tubes (Wilmad, 701 PO, 3.431.013 mm i.d., 5 mm o.d., highly polished) were used for all samples excited at wavelengths shorter than 296 nm. These tubes were joined to 10/30 male joints to allow for degassing and sealing under vacuum. Due to the large number of compounds and complexes investigated 287 and due to the fact that, in general, aromatic hydrocarbons are sus- ceptible to oxygen quenching, all samples were routinely subjected to six freeze-thaw degassing cycles on a system capable of obtaining a vacuum of less than 10-6 mm. In the course of degassing, the 5 mm by the last cycle. Further- pressure in the system dropped to 10- more, removal of oxygen should improve sample stability. For all cases checked, spectral shapes from samples stored at 00 in the dark remained unchanged even after several months. Spectra and lifetimes were generally determined within a week after preparation. Solutions containing crown and salt were prepared by diluting the appropriate volumes of stock solutions. Emission Spectra The corrected emission spectra shown in Figures 1 through 54 were determined using a Hitachi/Perkin-Elmer Spectrophotofluorometer Model MPF-44A with computerized corrected spectra accessory. The spectra show the full resolution obtainable at 77 K in 95% EtOH uncracked glass. A 1 nm emission bandpass was sufficient to resolve the fluorescence spectra and a 2 nm bandpass was sufficient for the phosphorescence spectra. An excitation bandpass sufficiently large so as to be consistent with good resolution and low noise level was used. The excitation bandpasses used to obtain well-resolved spectra were too large to allow for accurate quantum yield determinations from the same spectra. The relative spectral intensities qualitatively correlate with relative quantum yields, however. The phosphorescence spectra were determined without use of a phosphoroscOpe, since use 288 of the phosphoroscope introduced roughly a ten-fold attenuation of the signal and tailing fluorescence did not interfere. Concentra- tions were in the 10-4 M range. Spgetral Energy Levels In order to obtain precise frequencies for emission bands it was necessary to use a greatly expanded wavelength scale, usually either 1 nm/l6 mm or 1 nm/8 mm. For symmetrical bands, the position of the maximum was located using the symmetry of the band to locate the center. For unsymmetrical bands, the maximum was either visually sighted and/or located making use of the symmetry at the top of the band. The probable error in the band positions reported are :20 cmfl for the highest energy fluorescent band, and :10 cm.-1 for the highest energy phosphorescent band. The accuracy of these values should be within 1.2 nm ( 20 cm-1 at 320.0 nm and :10 cm.1 at 470.0 nm). Both the emission and excitation monochrometers were calibrated against the 450.1 and 467.1 nm bands in the emission spectrum of the xenon lamp. Also, the emission monochrometer calibration was checked by determining the absorption spectrum of holmium oxide. All band positions were within i.2 nm of the well known holmium oxide absorp- tion spectrum. The triplet energy for 2,3-naphtho-20-crown-6 (1) requires some comment as to how the value used was derived. It is almost certain that the highest energy band in the phosphorescence spectrum of 1 (472.3 nm) is unresolved (see Figures 20 and 26), as indicated by its width and by comparison to other naphthalene derivatives 289 and salt/crown complexes. Therefore, using the center of this band would give a low estimate of the triplet energy of this compound. The following scheme was used to get a higher and hopefully more cor— rect estimate of the triplet energy. The unresolved band in the crown spectrum and the "doublet" in the salt/crown spectra were assumed to have the same bandwidth in cmfl. The centers of the broad band in the crown spectrum and center of the "doublet" were found making use of their symmetry. The difference, in cm-l, between the center of the "doublet" and the position of the higher energy band in the "doublet" was found to be relatively constant for all alkali metal chloride/crown complexes (226:18 cmfl). This av- erage energy difference (in cmfl) was added to the position of the center of the broad band in the crown spectrum to obtain an estimate of the position of the "true" 0-0 band. 77 K UV absorption spectra were also determined using a suf- ficiently expanded wavelength scale (either 3 nm/inch or 6 nm/inch) so that precise frequencies of absorption bands could be obtained. The calibration of the Cary 17 monochrometer was checked by Gary personnel before the beginning and after the end of this investigation and was found to have been unchanged. Quantum Yields The preparation of samples for quantum yields was the same as for emission spectra. The concentrations were sufficiently low so that the optical density of the solution at the depth at which it 290 is viewed was §_0.02 A, or so that elcl - 8202 (where 1 is the relative extinction coefficient of solution one and C1 is its con- centration). 77 K UV spectra were the basis for selecting appropriate excitation wavelengths, and the relative extinction coefficients from them were used to either adjust concentrations before making quantum yield comparisons or to correct the apparent relative quantum yields. There is some difference of opinion as to how low the optical densities should be for the emission output to be a linear function of concentration. One source91 indicates that it should be 0.02 or less, otherwise a 4% error is incurred, while another source92 indicates it should be 0.1 or less, otherwise a 1% error is incurred. Except for the quantum yield determinations for the alkali metal complexes of crown 2 relative to free crown 2 (vide infra), all of the quantum yield determinations reported here satisfy the lower estimate. The optical density for these complexes in the region in which they were excited (approximately 297 nm) is estimated to have been 0.2. The relative emission outputs were corrected making use of the relative extinction coefficients, but these only differ by about 10%, so the difference in optical gradients was probably not very large. Furthermore, a check set of experiments done subse- quently in which the complexes and free crown g were excited in the 313 nm region (the optical density was approximately 0.02 A, and the relative emission outputs were corrected making use of relative extinction coefficients) gave quantum yield estimates which were the same within 3 to 5% to those obtained using higher optical densities. Quantum yields of naphthalene derivatives 1 through.2 were 291 determined relative to naphthalene (¢f - 0.30, ¢p - 0.030). Excita- tion bandpasses of 1.5 to 2.0 nm were used. The agreement between quantum yields obtained from determinations at different wavelengths (see results section and Table 8) shows that the exciting light was sufficiently monochromatic. Table 44 shows the wavelengths at which the comparisons were made and, in the first set of parentheses, the number of fluorescent quantum yield determinations, and, in the second set of parentheses, the number of phosphorescent quantum yield de- terminations. Since tailing fluorescence did not interfere, phos- phorescent quantum yields were determined without use of a phos- phoroscope, since the phosphoroscope introduces approximately a lO-fold attenuation of the signal. Quantum yields for the crowns in the presence of salts were determined by comparison to the free crown. Table 44 gives the exci- tation wavelengths used to make quantum yield comparisons between "fully" complexed crowns and the parent free crown. To show that the excitation bandpass was sufficiently monochromatic, comparisons were made using progressively smaller excitation bandpasses until the quantum yields became constant. A 2 nm bandpass was sufficiently monochromatic. The quantum yields for all naphthalene derivatives and the salt/ crown - 5/1 cases were determined using a Hitachi/Perkin-Elmer MPF 44A spectrophotofluorometer with computerized corrected spectra accessory, which was interfaced to a Digital PDP-8 computer. A 1 V corrected signal from the corrected spectrum accessory was collected as a function of time and stored. The monochrometer speeds of the 292 um oouauxo unmannunams cu o>wumHoMo .Es ¢.mom um oMuaoxo osoHQSunaoa on o>wumaom .moowumswsuouoo 9 mo noossz .ac w.mom . on A ocowumcwauouov e we Hooaszm Aevxev.ee~ Aoavaeve.na~ AeVAnVo.ea~ a.ea~ Amvauvo.m- Anvamvm.nhm ANVANvo.eNN o.¢- Auvaevo.ee~ Amvxeva.me~ Auvaevo.ee~ m.ae~ Amvaevo.nw~ AmVAaVo.am~ Aavaevm.em~ o.mmN AqVANvm.wunawm msooaumaaou vaoww azusmso nous: um Aacv msumnoao>m3 .eq nanny 293 Table 45. 'Wavelengths at Which Quantum Yield Comparisons Relative to Free Crowns Were Made. Salt 2,3-Cr-6 (1) l,8-Cr-6 (2) 1,5-Cr-6 (3) None 305.6 296.9 296.9 NaCl 304.5 297.3 296.7 KCl 304.6 297.1 296.6 RbCl 305.0 297.1 296.4 CsCl 304.9 297.3 296.3 NHACl 305.2 297.2 296.9 ndPrNHBCl 305.6 297.3 296.9 i-PrNHBCl 305.3 297.2 ----- t-BuNH3Cl 305.2 297.3 ----- Br