UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Studies on the hydrated electron Kenney, Geraldine Anne 1970

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

Item Metadata

Download

Media
831-UBC_1970_A1 K45.pdf [ 10.71MB ]
Metadata
JSON: 831-1.0059867.json
JSON-LD: 831-1.0059867-ld.json
RDF/XML (Pretty): 831-1.0059867-rdf.xml
RDF/JSON: 831-1.0059867-rdf.json
Turtle: 831-1.0059867-turtle.txt
N-Triples: 831-1.0059867-rdf-ntriples.txt
Original Record: 831-1.0059867-source.json
Full Text
831-1.0059867-fulltext.txt
Citation
831-1.0059867.ris

Full Text

STUDIES ON THE HYDRATED ELECTRON BY GERALDINE ANNE KENNEY L.R.I.C. Licentiate, of the Royal I n s t i t u t e of Chemistry, London, 1965 M.Sc. University of B r i t i s h Columbia, 1968 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY i n the Department of Chemistry We accept t h i s thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA In presenting th i s thes is in pa r t i a l f u l f i lmen t o f the requirements fo r an advanced degree at the Un ivers i ty of B r i t i s h Columbia, I agree that the L ibrary sha l l make it f r ee l y ava i l ab le for reference and study. I fu r ther agree that permission for extensive copying o f th i s thes is fo r scho la r l y purposes may be granted by the Head of my Department or by h is representat ives . It is understood that copying or pub l i ca t ion o f th is thes is fo r f i nanc ia l gain sha l l not be allowed without my wr i t ten permiss ion. The Univers i ty of B r i t i s h Columbia Vancouver 8, Canada Department ABSTRACT This describes a study of some unusual features of the hydrated electron, ea(^> i - n particular the kinetics of i t s decay during a period of non-homogeneity lasting tens of nanoseconds, the formation 'a, 2_ and photodissociation of a hydrated dielectron species (e^ ) a , and the photoexcitation of e~ . r aq Nanosecond pulse radiolysis (p.r.) studies on the kinetic behaviour -4 of e (> 10 M) in pure deaerated water revealed a complicated interplay of mechanisms for the f i r s t half l i f e ^ 110 nsec. This is partly attributable to an i n i t i a l non-homogeneity in the distribution of reacting species within the system, because the spurs are essentially isolated for tens of nanoseconds. Calculations based on a qualitative model revealed that the times necessary for spur-overlap through diffusion (during which > 40% e~^ were lost to reaction) were in agreement with experimental observations. However the anomalous trends in K, a rate parameter describing - d [ — ]/dt within this period, led to the subsequent e "aq discovery of a process by which e &^ were formed after the electron pulse. The use of selective ion and radical scavengers strongly implied that the increase in [ea(^\ occurred via another radiolytic product, Xg. Three 2- * plausible mechanisms have been outlined in which X_ is (e_ )„,_, Ho0 or an D ^ <*q ^ entity [H.jO+.e~.OH]. Xfi undoubtedly affects the values of K but i t is not possible at this time to discard the notion of microscopic non-homogeneity within the spur i t s e l f as the trends in K might suggest. Four conclusions are drawn; (i) i n some p.r. studies we may not calculate meaningful second order rate constants with concentrations evaluated from o p t i c a l density data, ( i i ) the "instantaneous" y i e l d of e seen through nsec p.r. i s higher than that established through aq ysec p.r. or steady-state techniques because of the rapid i n i t i a l loss of e ( i i i ) but the t o t a l e~ y i e l d w i l l be less since the l a t t e r aq J aq 1 techniques cannot distinguish the source of e~^. (iv) there i s a c r i t i c a l need for a nanosecond p.r. y i e l d of e" to establish the true aq primary y i e l d , ^ ( e ^ ) . Some microsecond fl a s h photolysis (f.p.) experiments were performed on hydrogen saturated a l k a l i n e solutions. Hydrated electrons were produced following the u l t r a - v i o l e t photolysis of OH" and reacted bimolecularly to give a species which on subsequent infra-red f l a s h photolysis regenerated e" . This species i s postulated to be a 2-hydrated electron dimer (e_ ) the spin state of which i s unspecified The remaining purpose of t h i s work was to photoexcite e~ . The nature of the excited state of e and the o r i g i n of the o p t i c a l absorption band i s s t i l l open to speculation although Jortner and others have performed calculations i n which the t r a n s i t i o n at A i s assigned r max 6 to a 2p +• Is excitation. The photolysis of e~^ was attempted through both p.r. and f.p. techniques, neither of which yielded any conclusive 'a. 2-information because of the presence of Xg or ^ )  i n the system. - i v -Table of Contents Page Chapter I: The Interactions of Ionising Radiation i n a l i q u i d medium 1 (1) Electromagnetic Radiation 1 (2) P a r t i c l e Radiation 5 (3) The Range of the Electron 7 (4) Cerenkov Radiation 10 (5) The Radiolysis of Liquid Water 11 (a) physical stage 12 (b) physicochemical stage 17 (c) chemical stage 20 (6) Summary 22 Chapter I I : The Hydrated Electron 24 (1) A H i s t o r i c a l Perspective 26 (2) Some properties of the Hydrated Electron 29 (3) On How to Trap a Thermalised Electron 33 (4) The Formation of Hydrated Electrons i n Flash Photolysis 36 Chapter I I I : Structural Models for the Solvated Electron 44 (1) The Potential Well 46 (2) The Cavity i n a D i e l e c t r i c - Continuum Approach 50 (3) Atomic Models 58 Chapter IV: Pulse Radiolysis and Kinetic Laser Photometry 65 (1) Febetron Accelerator 65 (2) Experimental Solutions 67 (3) Optical equipment 71 - V -Page (4) Photomultipliers and Oscilloscope 74 (5) Grounding and Shielding 78 (6) Radiation c e l l s 80 (a) P l e x i c e l l 81 (b) Stainless Steel C e l l 84 (c) S c i n t i l l a t o r s 85 Chapter V: The Flash Photolysis Apparatus 87 (1) Double Flash Photolysis Unit 87 (2) Reaction Vessel and Experimental Solutions 92 (3) Optical and Detection Systems 93 Chapter VI: Interference Phenomena 98 (1) Chemical 99 (2) Electronic 1 0 0 (3) Optical 1 0 0 (4) Shock Phenomena 102 (5) F i e l d Effects 1° 5 Summary 107 - v i -Page Chapter VII: Preliminary Results and Observations on the Pulse _08 Radiolysis Experiments (1) Non C l a s s i c a l Kinetics - preliminary data 108 (2) Non-homogeneity and rate constants 1_7 (3) More k i n e t i c studies, using the Stainless Steel C e l l . (a) Origin of the curvature observed on the decay l - ^ l signal (b) On the nature of the species involved _27 (c) Summary 132 Chapter VIII: Data from the Remaining Pulse Radiolysis Experiments 133 (1) A Model for the Formation and Decay of e~_^  _37 (2) Experimental Data Correlated through the A, B Model 142 (a) Formation and decay of e &^ as a function of 142 dose and pathlength. (b) Formation and decay of e~ i n the presence of 150 aq various additives. (3) On the Contribution of B to the measured OD values 156 (a) Isolation of B from the Experimental data (b) Influence of solutes on the rate of change of 160 C0D) B (4) Photolysis Experiments with the S c i n t i l l a t o r s . 162 - V l l -Page Chapter IX: Discussion on Non-homogeneity 165 (1) The Time dependence of < and Non-Classical Behaviour 165 (2) On understanding Process B 172 (a) An attempt to describe B through the k i n e t i c s of 172 e" aq (b) Plausible mechanisms for the reaction X_ -*• e" 175 ^ 1 B aq ±10 (i) ( e ) _ _ a q 175 ( i i ) H20 179 .+ _ ~ n u l 180 ( i i i ) [H 30 .e".0H] (3) Implications of these Studies to Results i n the Current 183 Literature (4) An Alternative interpretation of B 187 Chapter X: Results from the Flash Photolysis Experiments 190 (1) Preliminary results from the fl a s h photolysis of KI 190 solutions (2) Results from the Flash Photolysis of hydrogen 194 saturated a l k a l i n e solutions Chapter XI: Discussion of the Photolysis Experiments 201 (1) The Double-flash photolysis of aqueous solutions 201 (a) I d e n t i f i c a t i o n of the absorbing species generated by 204 the two flashes (b) I d e n t i f i c a t i o n of the Intermediate Species 207 (c) The Absorption Spectrum of the Dielectron 214 (2) The Photolysis of e" 217 J J aq - v i i i -LIST OF FIGURES F Figure 1-1 Mass Absorption Coefficients for the Various Processes i n Water. Figure 1-2 The Range and Penetration of a Beam of Monoenergetic Electrons i n an Absorber. Figure 1-3 Energy Flow Diagram Figure 2-1 Optical Absorption Spectra of Solvated Electrons i n Various Media. Figure 3-1 Energy Levels Corresponding to Possible Absorption Processes. Figure 4-1 System Block Diagram for a Standard Model 706/2670 System. Showing Typical Waveforms from the Model 5515 Electron Tube as Measured i n t h i s Laboratory. Figure 4-2 The Accelerator Laboratory. Figure 4-3 (a) The P l e x i c e l l and Components. Figure 4-3 (b) Stainless Steel C e l l and Components. Figure 5-1 Trigger System for Flash Photolysis Unit. Figure 5-2 Showing the Double-Flash Photolysis Apparatus. Figure 6-1 Typical Interference Signals Figure 7-1 F i r s t and Second Order Treatments of Data i n Table 7-1. Figure 7-2 Showing the Time Required to Establish Homogeneity at a Given Instantaneous Dose for the Hydrated Electron. Figure 7-3 Oscilloscope Traces Showing the Effects of Different Solutes on the "Curvature". -XX-Figure 7-4 Oscilloscope Traces Showing the Effects of Different Solutes on the "Curvature". Figure 8-1 Dose-depth Curves. Figure 8-2 Graphical Representation of A, B Model. Figure 8-3 Second Order Treatment of Data from Stainless Steel C e l l . Figure 8-4 The Empirical Parameters t _ a x and t ^ plotted as a Function of pH. Figure 8-5 The Rate of Formation of e" (given by (0D) R) i n the Presence aq D of Various Solutes. Figure 9-1 Non-homogeneity as i l l u s t r a t e d by K. Figure 10-1 Some Oscilloscope Traces from the Double-Flash Photolysis Experiments. Figure 10-2 Second Order Plots for the Disappearance of e~^ Following the Main and A u x i l i a r y Flashes, Figure 10-3 Spectrum of the Hydrated Electron. -X-LIST OF TABLES Table 3-1 A Comparison Between Experimental and Predicted Values.Some Solvated 64 Electron Parameters. 7- 1 Calculations with Data from Preliminary Pulse Radiolysis Results. 113 8- 1 Dose-Depth Relationships for Water. 135 8-2 Percentage Curvature as a Function of I r r a d i a t i o n Pathlength. 143 8-3 Optical Density and Curvature Data as a Function of Pinhole P o s i t i o n , 144 (2.8 mm path). 8-4 Optical Density and Curvature Data as a Function of Pinhole P o s i t i o n , 145 (1.0 mm path). 8-5 Optical Density and Curvature Data as a Function of Pinhole P o s i t i o n , 146 (5.0 mm path). 8-6 Optical Density and Curvature Data as a Function of Pinhole P o s i t i o n , 147 (2.0 mm path). 8-7 Optical Density and Curvature Data as a Function of Pinhole P o s i t i o n , 148 (5.0 mm path). 8-8 Dependence of the Optical Density and Curvature Data on the Presence of 152 Solutes: NaOH and R-OH. 8-9 Dependence of the Optical Density and Curvature Data on the Presence of 153 Solutes: Na 2S0 4, H 2S0 4, Ba(0H) 2 8-10 Change of (0D) B as a Function of Time. 159 8-11 Formation and Decay of e i n the Presence of S c i n t i l l a t o r Emission. 164 aq - x i -Table of Reactions Reaction 1 2 3 4 5 6 7 e + e aq aq e" + H aq H 2 + 2 OH H 2 + OH e + OH -»• OH aq e" + H o0 o -> OH" + OH aq 2 2 e" + H_0 H + H o0 aq 3 2 H + H H H + OH H, H20 pH 7 - 1 4 10.5 10.5 7. 4.5 2 acid 7 Rate constant, M~* see"* (5.50 ± 0.75) 10" ao 2.5 x 10 3.0 x 10 1.23 x 10 10 10 2.32 x 10 ao 10 1.0 x 10 1.2 x 10 7.0 x 10" 10 8 OH + OH H2°2 9 e~ + H_0 ->- H + OH" aq 2 10 OH + H_ •+ H + H„0 7 8.4 7 13 5.0 x 10" 16.0 6.0 x 10' 1.6 x 10 8 11 12 13 14 15 16 17 H + OH" aq aq CO," OH + CO. e + CH-OH -»• aq 3 H + CH3OH CH2OH + H 2 OH + CH3OH •* CH20H + H20 OH + C2H5OH C2H4OH + H20 H + C2H5OH -> C2H4OH + H 2 > 11 11 7 7 7 6 - 1 0 7 2.3 x 10' 3.0 x 10 < 104 1.7 x 10f 5.1 x 10 1.1 x 10^ 1.6 x 10' 8 8 - X l l -Table of Reactions Reaction pH Rate constant, M-* sec"''' 18 OH + H-0- -*• - H-0 + H02 7 4.5 x 10 7 19 e •+ e -> 20 e 2 + hv . -> 2e 21 e" + e" th aq 22c e" + e" ->-th aq 2 23 e" + H„0+ -> H-0 th 2 2 24 e 2 •+ 2e 25 H+ + OH" + H20 26 (e2-) -»• H" H, ^ 2 'aq aq 2 27 (e2.") + H+ -> H" 2 'aq aq aq 28 H+ + H" ->• H2 29 (e?,") + Ho0 -> H" + OH" v 2 'aq 2 aq 30 H" + H-0 + H- + OH aq 2 2 - x i i i -ACKNOWLEDGEMENTS I would l i k e to o f f e r a deep fee l i n g of gratitude to Dr. David C. Walker who has continually found patience, encouragement and time to help me unravel these problems. I would also l i k e to thank Dr. Norman Basco, not only f o r providing the f a c i l i t i e s for the f l a s h photolysis experiments but also f o r the many invaluable discussions we had on the ensuing r e s u l t s . -xiv-"Nothing i n l i f e i s to be feared. It i s only to be understood." Marie Curie Chapter I THE INTERACTIONS OR IONISING RADIATION IN A LIQUID MEDIUM Some of the general characteristics and consequences of electro-magnetic and p a r t i c l e radiation are outlined as a basis fo r the discussion on the interaction of ioni s i n g radiation and l i q u i d water. Three d i s t i n c t stages are normally proposed to describe the i r r a d i a t i o n of aqueous solutions, and within these consecutive in t e r v a l s the changing spatial-temporal relationships that exist between the ra d i a t i o n products dictate t h e i r k i n e t i c behaviour. To a certain extent these relationships are also dependent on the i n i t i a l dose. The interactions of ioni s i n g r a d i a t i o n with matter have been treated extensively on both theoretical and experimental grounds ( 1 ) . The transfer of energy from the electromagnetic or p a r t i c l e radiation on the material through which the beam passes causes indiscriminate i o n i s a t i o n and excitation i n the molecular environment. The r e s u l t of these primary events i s a collage of r a d i c a l s , ions and excited species that i n turn react with each other or with unexcited molecules to give stable chemical products. The physical characteristics of both the r a d i a t i o n and i t s target influence the rate and manner i n which t h i s energy-transferred wholly or i n part to the medium - i s f i n a l l y dissipated within the medium. (1) Electromagnetic Radiation When a beam of photons arrives at the surface of an absorber the resultant energy transfer may lead to a change i n the energy and dir e c t i o n of some or a l l of the photons or simply to a reduction i n the transmitted i n t e n s i t y of the beam. For a monoenergetic beam of photons -2-incident on the mate r i a l , I = I. e " y l X + I , e- V2 X + I. e ^ j * 1 2 j J where (1^, y..) characterises a photon of energy E.. and x i s the thickness of the absorber and y i s the l i n e a r absorption c o e f f i c i e n t . When electromagnetic r a d i a t i o n penetrates an absorber one of three important i n t e r a c t i o n s may dominate the re s u l t a n t exchange of energy. These are: (a) the p h o t o e l e c t r i c e f f e c t (b) the Compton e f f e c t (c) p a i r production The p h o t o e l e c t r i c process i s most s i g n i f i c a n t f o r photons of inc i d e n t energies 1 KeV < E.. < 50.0 KeV according to the el e c t r o n density of the material and describes an absorption of the photon i n which i t s t o t a l incident energy E.. i s t r a n s f e r r e d to a bound e l e c t r o n within the medium. This e l e c t r o n i s subsequently ejected with an energy E _ g at an angle to the o r i g i n a l d i r e c t i o n of momentum which depends on E^. As the l a t t e r increases so do the number of photoelectrons with a large component of t h e i r v e l o c i t y i n the forward d i r e c t i o n , but, none are ever emitted along the o r i g i n a l a x i s . E = E. - <\> P e J where <j> i s the i o n i s a t i o n p o t e n t i a l of the medium. Unbound electrons may not p a r t i c i p a t e i n t h i s process because the momentum and energy r e l a t i o n s h i p s must be conserved through the r e c o i l of the nucleus. As the binding energy of the electron increases so does the p r o b a b i l i t y of ph o t o e l e c t r i c absorption. At incident photon energies greater than the binding energy of the K and L s h e l l s the involvement of outer s h e l l -3-electrons may be neglected and any vacancies that arise i n the inner shells as a consequence of th i s effect w i l l be f i l l e d by outer s h e l l electrons, energy being conserved by the emission of X-rays or low energy Auger electrons. In water the lower l i m i t of t h i s effect i s determined by the inner s h e l l binding energy of ^  0.5 eV. A weak coherent scattering of the incident photon by the atomic electrons of the medium becomes more pronounced as the electron density of the l a t t e r increases of E_. decreases, but by comparison with the photoelectric effect i t may be neglected. The Compton effect dominates those electromagnetic interactions involving photons of several MeV, p a r t i c u l a r l y i n materials of high electron density. Here an i n e l a s t i c c o l l i s i o n between a photon and an electron of the medium results i n the scattering of the photon with energy E < E.. and the angular r e c o i l of the electron whose energy E_ has been increased. E = E, - E r J When the r e c o i l angle 0 i s 180° E r has i t s maximum value calculable from (1 + E 0 / 2) (1- Cos 9) where E Q i s the i n i t i a l electron energy and m0 i t s rest mass. If the material i s of a lower atomic number then the Compton effect may be observed with photons of less k i n e t i c energy although seldom below 20 KeV; i n l i q u i d water such effects dominate for 1 MeV photons,, but t h i s process has been reported over a range of 30 KeV < Ej < 20 MeV. It i s not unusual for the Compton photon to undergo the same process again, and unless the absorber has a minimum thickness some of the second or t h i r d -4-1 1 ~ r 0-1 .1 10 Energy (Mev ) Figure 1-1. Mass Absorption Coefficients for the various Processes i n Water. -5-generation photons w i l l ultimately escape, leaving only a f r a c t i o n of the i n i t i a l Ej transferred to the medium. Photons whose energy exceeds 1.02 MeV may take part i n the process known as p a i r production. A positron and electron appear at the disappearance of the incident photon within the v i c i n i t y of an atomic nucleus i n accordance with the energy re l a t i o n s h i p : EP + Ee = E j " 2 m ° 2 The p a r t i c l e s spontaneously combine with t h e i r counterparts i n the medium and the simultaneous emission of a n n i h i l a t i o n radiation i n the form of two 0.51 MeV y-rays i s observed. This mechanism i s of major importance when high energy photons undergo energy exchange processes. The atomic cross section (for absorption c o e f f i c i e n t Uj) for each of these three processes increases as the electron density of the target material increases. Since these processes are independent of one another the l i n e a r summation of a l l three cross sections gives a good approximation to the t o t a l cross section. To a f i r s t approximation contributions from scattered photons, fluorescence and Bremsstrahlung emission may be included i n one correction factor, f. In the context of the r a d i a t i o n of l i q u i d water with ^ 1 MeV energy photons the Compton effect i s the most important. Figure 1-1 shows the mass absorption _ 3 c o e f f i c i e n t s ( y_ . f, where p = density i n grams cm , and f i s the cor-P rection factor) for these various processes i n water. (2) P a r t i c l e Radiation As a fast charged p a r t i c l e travels through to target material energy i s transferred non-selectively from the incident p a r t i c l e to i t s -6-surroundings within a very short period of time, < 10 seconds. This exchange can come about through direct c o l l i s i o n s or excita t i o n s , through scattering or by the emission of electromagnetic radiation along the track of the p a r t i c l e . The lower the energy of the p a r t i c l e the more s i g n i f i c a n t the extent of the e l a s t i c scattering and i n e l a s t i c c o l l i s i o n s , while the heavier the p a r t i c l e the more s p e c i f i c the i o n i s a t i o n that occurs. Since i n t h i s work we are concerned prima r i l y with a beam of high energy electrons the remaining comments are directed towards the behaviour of ^  0.5 MeV electrons. The s t a t i s t i c a l variance i n the number of c o l l i s i o n s experienced between any one incident electron and the electrons of the medium w i l l be r e f l e c t e d i n the spread of ranges of the monoenergetic incident electrons penetrating i n the same absorber. In one c o l l i s i o n the electron can lose a substantial f r a c t i o n of i t s k i n e t i c energy. The three most relevant processes are (a) e l a s t i c scattering or def l e c t i o n , (b) i n e l a s t i c scattering ( i o n i s a t i o n and excitation) and (c) the emission of Brernsstrahlung radi a t i o n . Once again the importance of each of these to the ove r a l l atomic cross section depends i n no small way on the electron density of the medium i t s e l f . E l a s t i c scattering arises from the deflection of the incident electron - without loss of energy - by the coulombic f i e l d s about the nuclei of the medium. The higher the electron density the greater the i n e l a s t i c scattering, p a r t i c u l a r l y for low energy electrons that experience the f i e l d of the medium electrons to a more considerable extent. (However any large angular scattering w i l l probably be of nuclear o r i g i n ) . Should the incident electron interact with these f i e l d s then a substantial f r a c t i o n of i t s k i n e t i c energy w i l l be conveyed to the -7-medium i n a single c o l l i s i o n ; the object of the c o l l i s i o n i t s e l f may eject electrons with s u f f i c i e n t energy to cause further ionisations and thus both primary and secondary electrons (possibly t e r t i a r y ) contribute to the t o t a l picture of ion i s a t i o n and exci t a t i o n . Although t h i s mode of behaviour dominates for incident electrons of < 1 MeV, the phenomenon of Bremsstrahlung should be mentioned. It i s e s s e n t i a l l y the emission of radiation within the medium i n order to conserve energy -momentum relationships when a high speed electron i s decelerated i n the f i e l d of an atomic nucleus. Of major importance when the incident beam comprises of electrons > 10 MeV, the rate of loss of energy due to Bremsstrahlung i s proportional to the electronic ( z ) and nuclear charges (Z) and the electronic mass, m. 2 2 -dE « z Z dX m2 (3) The Range of the Electron As a resu l t of scattering the track of the excess electron through the medium i s u n l i k e l y to be a straight one. The range i s defined as the t o t a l distance t r a v e l l e d before the p a r t i c l e i s stopped, whereas the penetration i s the distance t r a v e l l e d i n the d i r e c t i o n of the o r i g i n a l momentum only. The average range depends on the rate of energy loss of the incident p a r t i c l e , but i t i s found experimentally that the actual range i s less than would be predicted. This i s accounted f o r by the appreciable and d i v e r s i f i e d scattering effects we have already discussed. The rate of energy loss depends on chara c t e r i s t i c s of both the medium and the p a r t i c l e , and i s given as a loss per unit length i n the -8-100 °'1 °-2 0.3 0.4 Distance " Above: Electron Range i n an Absorber, Incident Beam Monoenergetic. Below: Electron Penetration i n Aluminum, 1/2 inch from Model 5510 and 5515. Tubes Figure 1-2. Aluminum 200 100 150 2 Thickness mg / cm The Range and Penetration of a Beam of Monoenergetic Electrons i n an Absorber. -9-expression below, formulated by Bethe. 2 , F 7 4 m v E i - „ " © = — 2 . N . Z [ l n - ^ j " ( 2 7 l - e " 1 + B )ln2 + V 2 1 C 1 _ B 3 2 / 2 2 -1 + 1 - 3 + U l - V l - B ) ] ergs cm 1 _ i 8 v = v e l o c i t y of incident p a r t i c l e , cm sec 8 = v, c i s the v e l o c i t y of l i g h t , cm s e c * c I = mean ex c i t a t i o n potential of atoms, ergs N = number of atoms per c.c. of stopping material e = charge on electron, esu mQ = rest mass of the electron, grams Z = atomic number of the stopping material The stopping power formula only treats the average case and ind i v i d u a l p a r t i c l e s of comparable k i n e t i c energy can lose varying amounts of energy within the same i n t e r v a l for s t a t i s t i c a l reasons. Thus there i s a pronounced d i s t r i b u t i o n i n both pathlength, the distance t r a v e l l e d while losing energy to E j , and i n the energies of the ind i v i d u a l electrons as they t r a v e l a fixed distance d^ to d j . By numerically integrating the stopping power formula a value for the range may be obtained; the lower energy l i m i t introduces a degree of uncertainty which i s added to by the assumption that the stopping power of the medium i s e s s e n t i a l l y an electronic mechanism, so the integration i s frequently cut-off at the lower energies. The time spent by the electron within the interacting f i e l d of the atomic electrons i s inversely proportional to i t s v e l o c i t y , and since any changes i n momentum or the transfer of energy must be accomplished within t h i s time, the lower the i n i t i a l v e l o c i t y of the incident electron the faster i t attains thermal energies. At n o n - r e l a t i v i s t i c v e l o c i t i e s (-dE) decreases as v increases, but at r e l a t i v i s t i c v e l o c i t i e s dx -10-2 the Un (1-B) term varies i n such a way as to predict an increase i n the s p e c i f i c i o n i s a t i o n . In Figure 1-2 the number of incident mono-energetic electrons penetrating a given thickness of material i s shown from a theo r e t i c a l and experimental basis. (4) Cerenkov Radiation -13 Within 10 seconds of the passing of charged p a r t i c l e t r a v e l l i n g at a r e l a t i v i s t i c v e l o c i t y through a medium, electromagnetic radiation of u.v., v i s i b l e and i . r . frequencies w i l l be emitted along i t s track. This luminescence i s c l a s s i f i e d as Cerenkov Radiation (2), i t s i n t e n s i t y and duration depend on the v e l o c i t y of the p a r t i c l e arid the distance t r a v e l l e d before i t slows down to below-threshold conditions. The l a t t e r may be calculated for any p a r t i c l e of known v e l o c i t y i n a given medium. I f the v e l o c i t y v of the charged p a r t i c l e moving through a medium of r e f r a c t i v e index n exceeds the phase v e l o c i t y of l i g h t (c) i n the same medium then an electromagnetic shock wave i s produced n at an angle to the track of the p a r t i c l e as a resu l t of the p o l a r i s a t i o n and subsequent relaxation of the molecules i n the immediate v i c i n i t y of the track. The angle 9 at which the l i g h t i s emitted r e l a t i v e to the track i s determined by cos 9 = c , where 0 < cos 9 < 1. v.n In the case of l i q u i d water, n = 1.332 therefore y_ w i l l be 0.751 corres-c ponding to 265 KeV electrons for the threshold p a r t i c l e energy. In designing experimental equipment for either absorption or emission studies of i r r a d i a t e d l i q u i d s , Cerenkov emission must be taken into account as both weak or short l i v e d signals can be e a s i l y d i s t o r t e d , i f not masked completely. (5) The Radiolysis of Liquid Water We have said that a fast charged p a r t i c l e moving through a medium w i l l transfer energy non-selectively to the molecules along i t s track u n t i l i t i s f i n a l l y stopped. The energy i s imparted to the medium within 10 ^ seconds as a number of events or spurs which at t h i s stage are clusters of e l e c t r o n i c a l l y excited species. The spur i s considered to be an average deposition of about 100 eV, and the distance between the spurs i s a function of the Linear Energy Transfer (L.E.T.), ch a r a c t e r i s t i c s of the p a r t i c l e . Excited and ionised species lose energy for example by c o l l i s i o n , d i s s o c i a t i o n or solvent quenching, -12 within 10 seconds and on average one ion pa i r i s produced for every 30 eV of energy absorbed i n the medium. Thus a t y p i c a l spur i s supposed to consist of three such ion pairs or s i x r a d i c a l s . Although i t cannot be too heavily emphasised that t h i s is_ an average s i t u a t i o n , some radiation chemical theories, i n which the outcome of quite detailed, l o c a l i s e d physical events has been extrapolated to predict o v e r a l l chemical behaviour, have been successful i n persuading many experimentalists that they understand the fundamental processes involved. That there i s a relationship between the energy absorbed and the chemical y i e l d i s knowledge based on experience, but to define t h i s proportionality i s a herculean task. Nevertheless, some of these theories w i l l be outlined from the point of view of an incident 0.5 MeV electron t r a v e l l i n g haphazardly through l i q u i d water u n t i l i t i s f i n a l l y thermalised. The environment seen by t h i s electron changes by vi r t u e of c o l l i s i o n s , scattering and, because -12-of the time scale involved, the environment i t s e l f w i l l be concurrently r e f l e c t i n g molecular rotations and thermal vibrations. The interplay of events can be approached from three levels of complexity: ( i ) A process that occurs continuously throughout the whole ir r a d i a t e d volume where the outcome of a l l reactions i s expressed as a bulk y i e l d for a single dose. ( i i ) A process that occurs continuously along the tracks of i n d i v i d u a l i o n i s i n g p a r t i c l e s , where the s i t u a t i o n i s described through the lin e a r energy transfer c h a r a c t e r i s t i c s of each p a r t i c l e , d i f f e r e n t tracks leading to d i f f e r e n t reactions. ( i i i ) A process that i s an ensemble of i n d i v i d u a l elementary acts of energy transfer whose correlations i n space and time govern the subsequent reactions. Extrapolation from chemical experience has i n t h i s context li m i t e d physical meaning and can only perhaps touch some of the ideas i n ( i i ) . The investigations along the lines of ( i i ) and ( i i i ) have been mostly the work of two schools of thought (section 1 (b)) but recently the o p t i c a l approximation previously used by Platzman (8 (c)) has been further elab-orated i n ( i i i ) to give a p r i o r i information on r a d i o l y t i c y i e l d s (1 (b)). Following the primary impact of the p a r t i c l e the events can be usefully divided into three stages, physical, physicochemical and chemical as i l l u s t r a t e d i n Figure 1-3, and described here from the second le v e l of complexity, (a) Physical The primary electron w i l l be deflected from i t s l i n e a r track as i t begins to lose energy. Secondary ionisations and excitations are 2 produced along the track giving r i s e to both low energy (< 10 eV) and -13-2 high energy ( > 10 eV) electrons. The l a t t e r 6-electrons can branch o f f to form another shorter track where the energy w i l l be f i n a l l y dissipated. The low energy electrons however w i l l tend to suffer further deflections causing i o n i s a t i o n and exc i t a t i o n i n a more l o c a l i s e d area to the main track. About 60% of the i n i t i a l electron energy w i l l be transferred to the medium i n t h i s way as a number of spurs. The distance between the spurs depends on the nature of the primary p a r t i c l e and the rate at which energy i s l o s t , or the Linear Energy Transfer (LET) of that p a r t i c l e formally expressed i n eV per X, The more rapidly the p a r t i c l e moves the lower the LET (^  0.02 eV o per A for 0.5 MeV electrons) and the more iso l a t e d the spurs, while a Po^l ^ a p a r t i c l e w i l l lose energy faster by several orders of o magnitude ^ 14 eV per A and the spurs w i l l eventually overlap to form a continuous c y l i n d r i c a l track i n the medium. Even at the end of the track of a high energy p a r t i c l e secondary effects r e s u l t i n g i n an increase i n LET may give r i s e to overlapping spurs, and the LET of 6-electrons w i l l be different again from the primary p a r t i c l e . Thus the generic background of each area of i o n i s i n g a c t i v i t y w i l l dictate the l o c a l chemistry and i t becomes very complicated to have any one model to describe a l l the events following the impact of-a primary p a r t i c l e of spe c i f i e d energy. Before pursuing t h i s point we w i l l discuss i n more d e t a i l mechanisms through which the energy may be lost up to the point where the electron i s very close to thermal energy, -13 that i s up to ^  10 seconds following the time of impact. An energy -flow scheme may be seen i n Figure 1-3. Although there i s no a p r i o r i rule that says i n a c o l l i s i o n any single energy loss greater than the io n i s a t i o n potential of the molecule IMPACT e e high energy —»• sub excited *• solvated fluorescence t phosphoresc spur reactions i stable spec y y EXOTHERMIC REACTIONS ENDO-THERMIC REACTIONS GROSS MOLECULAR MOTIONS HEAT -15-must lead to the appearance of an ion pair - for the molecule may dissociate into neutral fragments - reference to the cross sections for i o n i s a t i o n , e x c i t a t i o n and d i s s o c i a t i o n show that t h i s i s the most e f f e c t i v e way of d i s s i p a t i n g the excess load, whether on the main track or not. The unrestricted loss of energy through Cerenkov radiation only accounts for less than one percent while as much as s i x t y per cent of the energy loss can occur i n 5-rays. Local heating effects could also contribute to the energy loss spectrum but i t has been estimated that for water such a thermal pulse would be no greater than a l o c a l -12 30°K increase which would be equilibrated within about 6 x 10 seconds and thus be of l i t t l e d i rect consequence on the system as far as the ultimate y i e l d s of r a d i o l y s i s are involved; a three-fold increase i n t h i s temperature over 10 seconds would s t i l l not affect the very fast reactions of the primary chemical species. (3) The excitation of singlet electronic states f a r outweighs that of t r i p l e t states as an e f f i c i e n t mechanism of energy transfer, however such excitation may also stimulate the i o n i s a t i o n of impurities present i n the system i n small quantities. This would probably lead to the observation of small yi e l d s of a given r a d i o l y s i s product when absolutely none was expected. We have referred to the main track of the primary p a r t i c l e 0.5 MeV and now add to t h i s the idea of a branch track created by an energetic (> 5 KeV) secondary electron curving away from the o r i g i n a l d i r e c t i o n of momentum. The spurs w i l l s t i l l be f a i r l y well spaced on the 3 0 average 10 A apart but since the charged p a r t i c l e steadily loses energy the spurs soon overlap into a short track. The theory of track formation i s based on the electronic stopping power of the medium. (3,4) . -16-As f r a c t i o n a l energies are approached, v i z . 500 to 100 eV, neither the S-electron nor the l o c a l i s e d products of i t s i o n i s a t i o n power are s u f f i c i e n t l y energetic to move far from t h e i r o r i g i n . In consequence there arises a densely packed area of i o n i s a t i o n and ex c i t a t i o n often referred to as a blob. Any electrons of < 100 eV w i l l form an i s o l a t e d spur somewhere near one of these tracks, and these may be r e l a t i v e l y small ^ 40 eV or so. The next generation of electrons w i l l have sub-excitational energies and w i l l a t t a i n thermal energies through two l i k e l y mechanisms. I f the slow electrons were to have a r e l a t i v e l y long t r a n s i t time i n an area of high p o l a r i s a b i l i t y then they might stimulate molecular rotati o n as well as v i b r a t i o n which would quickly reduce t h e i r excess energy. The maximum feasible LET through o v i b r a t i o n a l loss has been computed to be 0.3 eV per A and the rate of loss of energy through any dipolar interactions appears to be constant 1 3 - 1 at ^ 10 eV sec . Since t h i s sub-excitational energy i s i n s u f f i c i e n t to be transferred to excited electronic states of the water molecules, the stretching and bending modes of the hydrogen bonds become the means of e f f i c i e n t energy quenching for slow electrons ^ .5 eV. The number of energetic electrons i n the medium cascades dramatically - for every 0.5 MeV electron impinging on water there are estimated to be > 30,000 slow electrons available, either ejected from late i o n i s a t i o n events or at the end of a series of c o l l i s i o n s from primary events. These slow electrons may react chemically or become attached to a solvent -13 molecule within t h i s period, 10 seconds on the scale i n Figure 1-3; they do not have to be thermalised. In high d i e l e c t r i c l i q u i d s such as water energy losses through these kind of molecular vibrations are thought -17-to remain s i g n i f i c a n t within the 2 to 10 eV l i m i t s and so the slow electrons generally w i l l be thermalised. The electrons have now been reduced to a d i s t r i b u t i o n of energies whose mean i s ^ 0.5 eV and the f i n a l slowing down of the electron i n the polar medium may permit i t to move randomly over quite a long distance. This concept of a thermalisation t a i l involves mechanisms that strongly depend on the properties of the l i q u i d , but we w i l l confine our remarks to water. I f the d i e l e c t r i c relaxation time i s less than the time required to a t t a i n thermal energy then the electron w i l l remain free from the influence of i t s concommitant p o s i t i v e ion, an entity that of course has always been simultaneously created i n the medium with the electron but up u n t i l now has been neglected, (b) Physicochemical Stage The thermalisation t a i l i s defined as a distance t r a v e l l e d by the p a r t i c l e i n order to incur an energy loss of 0.5 to 0.04 eV. Although we have so far considered the energy transfer as re s u l t i n g i n a net loss with respect to the mobile electrons, the Maxwell-Boltzman d i s t r i b u t i o n of energies (for a l l encounters i n which energy i s exchanged) becomes much more s i g n i f i c a n t f o r the thermal electrons, and should be included i n calculations of the t o t a l thermalisation length. Even i f t h i s distance i s not very long e ^  may survive recapture i f the medium i s s u f f i c i e n t l y polar, and the d i e l e c t r i c relaxation time i s long i n comparison to the period of thermalisation. I f on the other hand the energy of e ^ does not exceed the a t t r a c t i v e potential energy from the coulombic f i e l d of the parent p o s i t i v e ion, the electron w i l l be eventually recaptured. -18-It i s precisely these thermalisation t a i l s that cause the dis p a r i t y between models of electron escape and recapture. Two long opposing views are based on a mean free path estimate of ^  20 A and one of 150 °A. The smaller of the two values was i n i t i a l l y proposed by Samuel and Magee (5) who extrapolated vapour phase data (describing the f r a c t i o n of energy lost per encounter i n the high energy radiation of water vapour) to the condensed medium and suggested e" ^  was recaptured. It i s argued that despite t h i s questionable basis, the energy loss mechanisms are so s i m i l a r i n either phase that the value i s a very good approximation. Later modifications to the S-M model (6) including a notable contribution from Mozumder on track theory and prescribed d i f f u s i o n (7), calculated the i n i t i a l s p a t i a l d i s t r i b u t i o n s of the primary chemical species (see next paragraph), that i s the spur, and thus predicted ultimate yields for r a d i c a l and molecular products. The other estimate f o r the mean free path of e ^  was put forward by Lea Gray and Platzman (8,9) and naturally'predicts a much broader s p a t i a l d i s t r i b u t i o n of reacting species leading to di f f e r e n t y i e l d s . The difference may be i l l u s t r a t e d by returning to the general spur picture and the events taking place within i t s sphere. As a cluster of excited and io n i s i n g species the solvent molecules i n the spur may lose energy by c o l l i s i o n and di s s o c i a t i o n into r a d i c a l s , - 1 2 or quenching by other solvent molecules within 10 seconds. Any meta-stable states are thought to return to the ground state by non-radiative processes. In an average spur of 100 eV there are considered to be s i x radicals or three ion p a i r s . This evaluation i s made on the basis of cloud-chamber studies where i t has been shown that on average about 30 eV i s expended for every ion p a i r produced i n a wide range of -19-gases (10, 11). As unsatisfactory as extrapolations from the gas phase to the l i q u i d phase are, we have no way of making di r e c t measurements. In l i q u i d water, then, the i n i t i a l events i n an average spur are represented H20 .-*• H20* e", H20* followed by H 2 0 + + H2° "** H 3 0 + + 0 H H20* •> H + OH within the d i e l e c t r i c relaxation time of water, IO"** seconds, any "free" thermialised electrons w i l l become hydrated. e -> g thermal hydrated According therefore to the estimate of the length of the thermalisation t a i l or mean free path of e" ^  the hydrated electrons o o w i l l be found within a spur of radius % 20 A or<v 150 A, either close to or removed from the po s i t i v e ions and other r a d i c a l s . Yields of ra d i c a l and molecular products can be determined experimentally and fed back into these two models as support for or evidence against the theory. The S-M model has enjoyed more success i n as much as radi a t i o n chemists tend to think i n terms of small spurs, but s t r i c t l y speaking since the cross-sections for energy loss mechanisms i n the c r i t i c a l 10 eV to 0.04 eV region are not yet available (3) the models represent l i m i t s on the pr o b a b i l i t y of certain s p a t i a l d i s t r i b u t i o n s for e . The spur picture appears greatly over s i m p l i f i e d . Certainly -20-k i n e t i c data from recent very fast pulse r a d i o l y s i s studies (12), and product analysis using high scavenger concentrations (13) would indicate that whatever the i n i t i a l a c t i v i t y i s within the spur, i t i s not f u l l y anticipated by these models. We s h a l l however continue to use the concept of a spur as a basis f o r the remaining discussion. The d i f f u s i o n of the radicals begins i n the physicochemical stage and the spurs increase i n size u n t i l the overlap of adjacent spurs establishes a true homogeneity i n the system. Again a geometry has to be assumed for the mathematical treatment of t h i s period of d i f f u s i o n and so the spur i s regarded as an expanding sphere i n which radicals and ions w i l l react with each other or having excaped from t h e i r immediate k i n , with solvent or solute molecules or species from another spur. In a high LET s i t u a t i o n the p o s s i b i l i t y of intertrack as well as intraspur and interspur effects must be considered. These reactions frequently take place at d i f f u s i o n controlled rates and take us into the f i n a l chemical stage of r a d i o l y s i s . A t h e o r e t i c a l treatment of a one radical-one solute problem has been given by Kupperman (14) and extended to more complex systems. (c) Chemical Stage In accounting for the r a d i o l y t i c products radiation chemists invoke many mechanisms including rearrangements, H-atom migration, bimolecular ion-molecule reactions, unimolecular reactions from rupture of the o r i g i n a l molecules, d i s s o c i a t i v e and non-dissociative charge transfer and a l l the possible permutations of reaction between ions, radicals and neutral species. The currently accepted balance sheet for the r a d i o l y s i s of pure l i q u i d water i s given below (1(a)): the number i n brackets i s the -21-y i e l d of species per 100 eV deposited i n the system, or the G-value. H2° eaq ( 2' 8 )> H (°-7)> 0 H t 2 - 2 ) H 2 (0.42), H 20 2 (0.71) and H.OH, H20* (?) The following intraspur reactions i n the previous stage are postulated to give r i s e to these products: e" + e" •* H_ + 2 OH" (1) aq aq 2 aq e" + H -> H- + OH" (2) aq 2 e" + OH + OH" (3) aq aq e" + H o0 o -»• OH + OH" (4) aq aq e" + H T0 + -> H + Ho0. (5) aq 3 2 H + H -> H 2 (6) H + OH -> H20 (7) OH + OH -> H 20 2 (8) The same reactions can occur i n interspur kinetics once the spurs have overlapped, and as well any species may react with solute or solvent molecules present. eaq' H> 0 H + s products -22-The r o l e of excited water, s t i l l a controversial moiety (15), i s t e n t a t i v e l y omitted from t h i s d i s c u s s i o n but probably would r e s u l t i n the formation of fragments. H-0* ->• H ,0H, H-, 0 ? Fluorescence from e l e c t r o n i c a l l y excited states may occur within t h i s period and i n the event that slow electrons give r i s e to s i n g l e t t r i p l e t t r a n s i t i o n s there w i l l also be phosphorescence. However, no such emission i n the v i s i b l e has been detected from pulse i r r a d i a t e d water (2 (b)) and i t was concluded that i f luminescence other than Cerenkov were present, the number of photons per eV would be < 10 ^ and thus mechanistically unimportant. (6) Summary Both the i n t e n s i t y and o r i g i n of the i o n i s i n g r a d i a t i o n must be considered i n a study of the r a d i o l y s i s of a given system. The LET effects of the incident p a r t i c l e and the electronic stopping power of the medium determine the s p a t i a l d i s t r i b u t i o n of the radicals and ions within the concepts of track and spur theory. This d i s t r i b u t i o n - incorporated into d i f f u s i o n theory - governs the p r o b a b i l i t y of interspur and intraspur reactions between the primary chemical species, the predominance of either kind of reaction being re f l e c t e d by the r a t i o s of molecular to r a d i c a l yields i n the f i n a l chemical analysis. Since technology now permits us to explore the physicochemical stage with sub-nanosecond pulses, we may presume to gain further insight into spur effects and possibly refine the Samuel-Magee or Lea-Gray-Platzman models to give a more consistent picture -23-of the immediate fate of slow secondary electrons. This thesis w i l l be concerned with one of the primary species ©aq> i t s structure and i n i t i a l mechanistic behaviour within the spur. -24-Chapter II THE HYDRATED ELECTRON The hydrated electron, e , was f i r s t d i r e c t l y observed i n 1 9 6 2 climaxing almost a decade of speculation and experiment, Although e may be described as the primitive anion i n solution chemistry, i t s remarkably broad, assymmetric and apparently structureless absorption band has to-date defied any d e f i n i t i v e i n t e r p r e t a t i o n ; nor i s there much agreement as to how the electron i s f i r s t trapped i n a solvent sheath. As a chemical e n t i t y , however, e i s better understood, and i t s ' ' aq hitherto unrecognised presence i n f i e l d s other than radiation chemistry continues to widen the old concept of primary reduction mechanisms. With almost explosive cha r a c t e r i s t i c s the l i t e r a t u r e already contains over 700 rate constants* measured for the very fast reactions of e J aq with inorganic, organic and b i o l o g i c a l molecules; solvated electrons have been since subjected to a dozen reviews some comprehensive, some confined to the p e c u l i a r i t i e s of the electrons i n aqueous media, but a l l communicating an upsurge of interest i n th i s fascinating and novel species that l i v e s for a r e l a t i v e l y short time on the chemical scale. In t h i s chapter a short descriptive biography of the hydrated electron w i l l be given as a prelude to discussing the quantum mechanical models currently proposed for e g i n the next chapter. One of the purposes of t h i s thesis was to examine the possible origins of the absorption * A l l rate constants quoted i n the text are taken from tables compiled by M. Anbar and P. Neta, Int. J . Appl. Rad. Isotopes 18, 493-523 (1967). -25-spectrum of e through variations in its kinetic behaviour in the r aq 6 presence of a photolysing flash, but the long term results had far broader implications. -26-(1) A H i s t o r i c a l Perspective Several years after an unusual phenomenon or novel species has been discovered, we may look back upon the t r a i n of events p r i o r to t h i s finding and recognise a point at which discovery would now seem inevitable. The hydrated electron i s an int e r e s t i n g example of a speculation that spasmodically gathered momentum i n a decade where technology was making rapid advances and ultimately permitted i t to be observed. We now know that you can as e a s i l y shine a u.v. lamp on the surface of an aqueous solution i n a beaker and e w i l l be formed a l b e i t for an exceedingly short l i f e t i m e . Certainly the success of Weiss' free r a d i c a l hypothesis (16) i n apparently accounting f o r the r a d i o l y t i c products i n l i q u i d water forced an inconspicuous r o l e on the slow secondary electrons, one which was f i n a l l y questioned by Platzman (8 (a)) who dissented i n the following words: "Having attained thermal energy the electron finds i t impossible to carry out the chemical reaction written i n a l l the a r t i c l e s on rad-i a t i o n chemistry, just because of the d i s p a r i t y between the time the actual reaction takes and the time which would be required to u t i l i s e the hydration energy which makes i t possible. For t h i s reason the electron becomes hydrated ... I mean the electron polarises the d i e l e c t r i c and i s bound i n a stable quantum state to i t . . . Then ... the chemical reaction can proceed ... In between there i s the time for hydration to take place which must, as Onsager says, be a minimum of the relaxation time 10 seconds." The fate of the thermalised electron was no longer immediate capture e" + H-0 -*• H + OH ethermal 2 aq aq but instead conversion into a short-lived but stable species, the hydrated electron. e -> Q thermal aq The hydrated electron, and not the H-atom, was the primary reducing species The implications of his remarks were not to be f u l l y r e a l i s e d for almost ten years, but the incentive to b u i l d a new conceptual framework for the theory of molecular yi e l d s i n r a d i o l y s i s studies was r e f l e c t e d i n the experimental and t h e o r e t i c a l work of that time and indeed the discovery of e~ merely increased the momentum of the research on fundamental problems. The f i r s t reported attempt to observe the absorption of e (assumed to be i n the v i s i b l e ) was by Linschitz i n 1952 who i r r a d i a t e d water with 1 ysec X-ray pulses. (17) We r e a l i s e now his f a i l u r e to see the spectrum was not due to the resolution time as he supposed, but the s e n s i t i v i t y of the detection equipment. In retrospect t h i s was a c r i t i c a l point for the subsequent discovery of e - i n a sense the experiment had been done because we believe hydrated electrons were generated i n that system, but the negative results might have deterred further studies were i t not for an increasing amount of evidence from experiments using steady-state radiation sources. So the idea germinated and by the late f i f t i e s there was enough evidence from d i f f e r e n t laboratories to j u s t i f y the existence of two d i f f e r e n t kinds of reducing species i n i r r a d i a t e d pure water (18 - 24). By 1962 the dominant reducing species was reported to have a unit negative charge (25, 26, 27) and the experimental emphasis had moved to more -28-sophisticated pulsed radiation sources as a technique for i d e n t i f y i n g t h i s obviously short l i v e d species. A transient absorption i n the v i s i b l e was observed during and af t e r the delivery of a 1 ysec electron-pulse to an aqueous solution; the absorption disappeared i n the presence of electron scavengers but was not i d e n t i f i e d . (28) It was at t h i s time that Hart and Boag with t h e i r combination of two sensitive techniques, pulse r a d i o l y s i s and f l a s h spectroscopy, published f u l l d e t a i l s of t h e i r spectral and chemical analysis of i r r a d i a t e d aqueous solutions i n which the hydrated electron had been p o s i t i v e l y characterised. (29) In Hart and Boag's now c l a s s i c experiment a 2 ysec pulse of 1.8 MeV electrons - giving a mean dose of between 4 and 12 kilorads per pulse -was used to generate transient species whose absorption c h a r a c t e r i s t i c s were recorded photographically during a 4 ysec underwater spark between two uranium electrodes. The spark which gave a good continuum from a ^ 400 nm to ^  900 nm, could be triggered simultaneously or at a given delay after the pulse. The solutions they investigated were prepared from t r i p l y d i s t i l l e d water and were deaerated before use (the 0 2 and C0 2 content were determined to be ^  0.1 yM and ^  0.4 yM respectively). Absorption of the e was observed i n pure neutral water, i n 0.05 M solutions of a l k a l i metal carbonates, i n the presence of electron scavengers such as 0 2, C0 2, N20 and at spe c i f i e d time intervals a f t e r the electron pulse. And so i t was that the hydrated electron was discovered,an evolutionary species that belongs as much to the primitive beginnings of chemistry as to the refined understanding of radiation studies. -29-(2) Some Properties of the Hydrated Electron When i t was recognised that e was a most powerful reducing agent, and probably was formed as the precursor to hydrogen when ever pure water was reduced, many other systems were re-investigated to ascertain the nature of the effec t i v e reducing species. As a re s u l t i t i s now believed that e~ can be generated electrochemically i n which the e l e c t r i c a l aq & potential i s supplied externally, photochemically (which includes photo-emission,photoionisation and photoinduced electron t r a n s f e r ) , chemically i n which the potential i s spontaneously supplied from the in t e r n a l energy of the reactants, and i n radiation chemical systems. (30, 31) The production of e through f l a s h photolysis i s elaborated i n the next section. In fact many of the reactions previously attributed to the H-atom have been i n i t i a t e d by ©aq, i t s conjugate base. The hydrated electron can be conveniently pictured as an excess electron surrounded by many orientated water molecules, whose charge i s considerably delocalised over the sheath. The d i f f u s i o n constant of e has been calculated to be aq -S 2 -1 4.5 x 10 cm sec from the measured io n i c mobility of -3 2 -1 -1 1.77 x 10 cm sec Volt . (32) The act i v a t i o n energy of i t s very fast reactions i s therefore expected to be small and i n most cases corresponds to the ac t i v a t i o n energy of s e l f - d i f f u s i o n of water, 3.7 kcal mole that i s the energy of activation i s independent of the substrate, and may be the energy required to create a solvent hole. The hydration energy, calculated from the differences between the heats of formation i n the gas phase reaction (electron + proton) and the l i q u i d analogy, i s -40 kcal mole *. (33) This has been compared to that of the iodide i n whichit i s -57 kcal m o l e . I n fact e" i s often likened to I" -30-i n terms of a unit negative charge diffused over a large volume. The reaction radius of e calculable from k i n e t i c data i s 2.9 X (34) aq K J o whereas the I" radius i s ^  2.0 A. A further promising corre l a t i o n that has been recently observed i s between the E X of e" i n various ' max s media of widely ranging d i e l e c t r i c constants and E X of I" i n the J 6 6 max same solvent (35, 36) whose absorption i s a C.T.T.S. band. Several thermodynamic predictions concerning have been based on a model that extrapolates the observed values for halide ions to an ion of radius that i s compatible with the experimental free energy data, A of e aq. from t h i s equation. (37) In the standard state of a gaseous electron a l l thermodynamic properties are taken to be zero, the standard state for Vi^O i s the pure l i q u i d at 25°C and 1 atmosphere, and for the other species i t i s an idea l one molal solution e" + Ho0 i H + OH" k n = 16 M _ 1 sec" 1 k n = 2 x 10 7 M _ 1 sec" 1 (9) aq 2 aq aq 9 -9 * J There i s s t i l l doubt associated with the very slow forward d i r e c t i o n , ^ 16 M 1 sec 1. (31) Nevertheless i t remains to be seen how the predicted A H^, S° and V° appear i n the l i g h t of appropriate experimental o evidence. In p a r t i c u l a r the S value i s expected to be p o s i t i v e -i f t h i s proves to be true then the charge of e must be considerably delocalised over i t s solvent neighbours because a charged species usually . causes an decrease i n l o c a l entropy, that i s at least a short range ordering of the l i q u i d molecules. Solvated electrons exhibit s i m i l a r absorption spectra i n several respects. Apparently lacking any discernible f i n e structure they are -31-intense asymmetric broad bands ( ^  1 eV width at h a l f height) i n the v i s i b l e region. The spectrum of the ammoniated electron (38) whose X occurs i n the near i . r . s t i l l has an appreciable t a i l i n the max r r v i s i b l e ; the deep blue colour of stable metal-ammonia solutions i s well known (39). The hydrated electron absorbs strongly from ^ 300 nm to 900 nm with X at 720 nm; recent work (40) reports an increase max ' J r i n the int e n s i t y due to the presence of another band as wavelengths about 200 nm are approached presumably with a second maximum well i n the u.v. A more rigorous u.v. analysis of the spectral c h a r a c t e r i s t i c s i s r e a l l y necessary before r a t i o n a l i s i n g the appearance of a new band compared by the authors to a |3 band i n the spectrum of an F centre. A comparison of such spectra i s made i n Figure 2-1. The absorption spectrum maximum of e s h i f t s to the red as the temperature increases, and also fl a t t e n s out; omitting any theory concerning the traps for the time being, such a s h i f t would be anticipated by the v a r i a t i o n i n the d i e l e c t r i c constant of the medium, -LT a parameter c r i t i c a l to the trap, D = D0e . Interestingly enough the value of the empirical constant L (with dimensions of recipr o c a l temperature) i s ind i c a t i v e of the io n i s i n g a b i l i t y of the solvent. I f LT > 1 where D was measured at T°K, the solvent i s an io n i s i n g medium. (41) The red s h i f t i s also consistent with an increase i n the cavity s i z e , just as the blue s h i f t observed i n p.r. studies of e under aq pressures up to 1000 atmospheres i s in d i c a t i v e of a decrease i n these dimensions, (42).. In contrast to the e /NH_ system, reactions of e~ i n * amm' 3 J ' aq water are pressure - independent which infers a vanishingly small ac t i v a t i o n volume for these reactions. The Y - r a d i o l y s i s y i e l d of'e i s 1 3 3 aq also unaffected by pressure (43) and i n conjunction with the p.r. to Figure 2-1, 3.0 Optical absorption spectra of solvated electrons i n various media. A = l i q u i d NH^ ( a l k a l i metal solution (39), B = l i q u i d water (transient i n pulse r a d i o l y s i s (34) and C = F-centre (KC1 crystals (131)). -33-experiments suggests that the extinction c o e f f i c i e n t i s also pressure -independent. Trapped electrons i n ice exhibit the same features with EX r r max between 600 to 680 nm (44, 45, 46) but the y i e l d of electrons i s very -4 small; e.g. G(e t) = 2 x 10 i n pure c r y s t a l l i n e i c e . (47) It seems highly probable that the methods of introducing the electrons into the matrices or the freezing of solutions influences the environment seen by the electron, and thus not only the pos i t i o n of X 3 ' 1 Y m a x but the y i e l d of e as we l l . Not unexpectedly the absorption bands are narrowed somewhat i n comparison to e ^ , but the basic s i m i l a r i t y between the spectra observed i n l i q u i d and frozen samples where d i e l e c t r i c constants and d i e l e c t r i c relaxation times are so vastl y d i f f e r e n t must provide a clue as to the nature and a v a i l a b i l i t y of the trapping s i t e s i n both systems. 4 The extinction coefficient of e i n the region of X i s ^  10 aq 6 max l i t r e mole 1 cm 1 which suggests that the t r a n s i t i o n corresponding to EX i s "allowed". But to what l e v e l the e i s excited i s another max consideration, and so the next chapter w i l l be concerned prim a r i l y with the possible origins of th i s e x t r a o r d i n a r i l y uninformative absorption spectrum. The procedure involved i n the actual trapping of the electron i s another unresolved issue which we w i l l b r i e f l y examine. (3) On How to Trap a Thermalised Electron The cause and effect arguments put forward to explain how the electron becomes solvated and s e t t l e s i n i t s stable quantum state seldom c l a r i f y t h i s complex s i t u a t i o n . This i s not the r e s p o n s i b i l i t y of the electron so much as of the continually changing environmental structure of which we -34-know very l i t t l e - especially as far as water i s concerned. F i r s t , whereas the excess electron moving through the medium with thermal energies momentarily sees a l o c a l environment whose physical parameters such as molecular motions, p o l a r i s a b i l i t y , d i e l e c t r i c values and so on are a l l at l o c a l values, we i n our reconstruction of the same s i t u a t i o n are forced to use microscopic or bulk values which are average numbers associated with many thousands of molecules over a comparatively long period of time. Secondly, since we r e a l l y know very l i t t l e about the structure of water i t i s d i f f i c u l t to envisage any p a r t i c u l a r trapping s i t e as being favoured by the long or short range l i q u i d order. X-ray studies on water have produced a pattern of electron density that would indicate some short range tetrahedral order (48) i n the l i q u i d , and at room temperature water i s approximately 85% hydrogen bonded (49). There i s some disagreement as to whether the appearance of isobestic points i n the spectra of water i s evidence for an equilibrium between two molecular species ( d i f f e r i n g perhaps i n t h e i r degree of association) or merely a common feature of a l l OH stretching bands i n the l i q u i d and therefore are not s p e c i f i c evidence at a l l (50). It i s therefore not surprising to f i n d experimentalists looking for frozen and glassy matrices i n which to study the trapping procedure. Given that the extrapolation to l i q u i d systems w i l l s t i l l be quan t i t a t i v e l y awkward the controversy w i l l probably continue u n t i l the importance of each trap parameter has been correctly weighted i n keeping with interpretations of the absorption band. S u p e r f i c i a l l y we imagine e ^  t r a v e l l i n g through the water r e l a t i v e l y quickly, affecting a degree of p o l a r i s a t i o n i n the water dipoles that are randomly orientated along i t s track, but moving too fast to be trapped -35-i n the resultant f i e l d . I f , however, there existed already a region of accidental p o l a r i s a t i o n a r i s i n g from the thermal motions of the molecules themselves the l a t t e r would be polarised to a more s i g n i f i c a n t degree by the excess electron. Consequently e ^ would f i n d i t s e l f i n a potential well and once trapped could "dig i n " by causing further p o l a r i s a t i o n into the next layer of solvent molecules. The parameters considered most important i n assessing the pro b a b i l i t y of trap formation are the d i e l e c t r i c constant of the medium D, the d i e l e c t r i c relaxation time T , the s t r u c t u r a l order of the solvent, and the v a r i a t i o n of these parameters as a function of time. Two models describing the immediate fate of e~^ w i l l i l l u s t r a t e the type of correlation sought between these parameters. The f i r s t model developed by S c h i l l e r (51) sees the d i e l e c t r i c relaxation time as the decisive variable i n an approach that stems from track and spur theory. He considers the time dependent d i e l e c t r i c properties of the medium for a non-conservative e l e c t r i c f i e l d i n which the pr o b a b i l i t y of e ^ being trapped or captured increases as the relaxation time x increases. Experimentally he measured the degree of recapture of e~^ with the use of scavengers i n three systems, liquid-water, super-cooled water and ice < 0°C. The s t a t i c d i e l e c t r i c constants Dg are a l l s i m i l a r , and so are the physical and chemical1/ properties, but x ranges from 10 ^ seconds i n ice to 10 1 1 seconds i n water. As predicted by t h i s theory, he observed t o t a l recapture i n ice and considerable recapture i n water. The opposite basis was taken by Freeman and Fayadh (52) who propose that the e ^ become trapped i n the ca v i t i e s present i n the l i q u i d structure, the mobility of e being equated to the migration of these c a v i t i e s . -36-The differences between excess electrons i n polar and non-polar solvents becomes one of degree rather than kind a r i s i n g from variations i n Dg. Empirically formulated these conclusions were derived from an approximate proportionality between the y i e l d of e escaping the parent spur aq C"_free ^ Q n) and Dg of the l i q u i d , over a range 2 for cyclohexane to 79 for water at 22°C. Recent studies i n t h i s laboratory on formamide, Dg = 109, evaluate a free ion y i e l d compatible with t h i s model (53). -11 I f the e ^ forms i t s own trap the process w i l l take ^ 10 seconds, but no technical d i s t i n c t i o n s can be made at t h i s time. Calculations pertaining to e^ i n pure ice (54) have shown that the E ^ m a x °f the absorption spectrum cannot be explained as the type of t r a n s i t i o n commonly attributed to EX for e because the true D i s too low K 3.0 compared max aq s r to ^ 75, t h e o r e t i c a l l y necessary). I f the same t r a n s i t i o n does accurately describe the excitation i n both e. and e~ , and i t seems reasonable to ice aq assume t h i s because of the respective EX values, then we must conclude r max ' that i n ice at least e ^ seek pre-existing traps i . e . l a t t i c e defects or anion vacancies whose inherent p o l a r i s a t i o n contributes to the depth of the trap. No single factor governs electron trapping p r o b a b i l i t i e s and to what extent the conclusions of the l a s t paragraph relate to the s i t u a t i o n i n water i s s t i l l open to speculation. (4) The Formation of Hydrated Electrons i n Flash Photolysis In contrast to the r a d i o l y t i c systems where both e x c i t a t i o n and -37-io n i s a t i o n occur extensively without regard to the i n i t i a l state of the molecules, the photochemical generation of e &^ i s very se l e c t i v e . The u.v. photolysis of l i q u i d water does not follow the same mechanistic pattern as we observe i n r a d i o l y s i s studies i n which the dominant reducing species i s e , but there i s some c o n f l i c t i n the l i t e r a t u r e as to ^ aq' whether e~ may be detected at a l l by photolysis experiments because of the f a i l u r e to observe an i o n i s a t i o n continuum i n water as low as 105 nm. I f suitable aqueous solutions are used then e~^ may be generated through a dir e c t photoionisation or a charge-transfer-to-solvent (C.T.T.S) excitation after which i t s reactions are rea d i l y observed spectrographically or spectrophotometrically. Competitive scavenging techniques provide a means of cal c u l a t i n g l i m i t i n g quantum yields for e (and the other aq photolytic by-products) i n any system, and c o l l e c t i n g information on i t s chemical r e a c t i v i t y . Below ^ 200 nm water absorbs s i g n i f i c a n t l y . The extrapolated gas-phase value for the i o n i s a t i o n potential of water i s 12.6 eV, but i t has been proposed (55) that since molecules i n the condensed phase often have subs t a n t i a l l y lower i o n i s a t i o n potentials compared to the same discrete molecules i n the gas phase, water might be ionised at about* 7 eV. Nevertheless, various attempts to photoionise pure water with X ^ 147 nm have been ambiguously successful (56), and severe technical d i f f i c u l t i e s with experimentation i n the vacuum u.v. eclipsed any possible absorption signals i n the only reported photolysis of water with X as low as 58.4 nm (57). The electronic properties of l i q u i d water i n the vacuum u.v. described recently (58) do not confirm the hypothesis that the diffuse nature and overlap of the water molecules i n the condensed phase f a c i l i t a t e photoionisation because no sign of an -38-i o n i s a t i o n continuum was observed as low as 105 nm; however, i n order to explain two bands at 9.6 eV and 8.3 eV i t was speculatively proposed (59) that either the same processes were ocurring i n the l i q u i d as i n the vapour phase but at di f f e r e n t energies, or that additional processes might be involved i n the l i q u i d state. The gas phase u.v. photolysis of water vapour (60) i n the range 147 to 185 nm (8.4 to 6.6 eV) leads mostly to radicals hv H-0 -> H + OH <f> g 5 = 0.45 *147 = °'7 A very small i o n i s i n g effect was noticed at 185 nm (56 ( b ) ) ; i f e" aq were produced the y i e l d would be < 1% that observed for the H-atom. In ra d i o l y s i s studies over 50% of the H atoms i n pure water combine to give molecular hydrogen whereas i n photolysis studies only ^ 15% have a sim i l a r fate. In an attempt to observe a y i e l d of io n i s a t i o n products l i q u i d water has been photolysed at 124 nm and a <j> ( e a^) of 0.12 ± 0.056 reported (61) but th i s value also includes the r e a c t i v i t y of any excited water molecules. Later work from the same laboratory gave values at 147 nm of <j> (e~ , H20*) = 0.075 ± 0.056 and cf) (H,0H) = 0.72 ± .02,the l a t t e r agreeing with previously accepted data, and at 185 nm, <|> (e" , H90*) = 0.04 ± .028, and cj) (H, OH) = 0.33 ± .01 which i s less than other quantum y i e l d s . Whereas i n t h i s work the observation of e &^ was claimed through conductivity measurements i n solution during u.v. i r r a d i a t i o n , a more recent claim to observe e 6 ' aq i n u.v. photolysed water i s based on direct observation through fl a s h photolysis studies at 185 nm (62). The wavelength l i m i t that established -39-t h i s i o n i s a t i o n was < 192 nm which corresponds to photons of > 6.5 eV. This i s close to the speculative value of 7 eV for the "condensed" water molecule. The quantum y i e l d however was only 0.004, but i t i s probably more meaningful to say <f>(e )was < 20% of the y i e l d measured aq for H and OH r a d i c a l s , since many of the l i t e r a t u r e texts use d i f f e r e n t reference <j) and some dif f e r e n t actinometry for the same data. Experiments i n our laboratory have not detected any e following the u.v. photolysis of water with 185 nm < X < 220 nm. Even i f e &^ are being formed i n such a photolysis i t i s obvious that they are present i n exceeding small numbers and are not an e f f i c i e n t supply for further reaction. The alternative method i s to photoionise a solute i n water at X > 200 nm, within a C.T.T.S. absorption band. This approach f i r s t characterised e photochemically (63, 64) through aq f l a s h photolysing d i l u t e halide and K^Fe(CN)^ solutions, and has since been employed extensively, p a r t i c u l a r l y with the halide ions: I" + hv -»• I + e" <J>oc. 'v. 0.25 aq aq aq T254 Unfortunately some of the systems also generate species that i n t e r f e r e with the spectroscopic determination of the k i n e t i c behaviour of e , and the usefulness of the halide ions i s considerably aq' J r e s t r i c t e d by t h e i r "masking" tendencies. The following series of sulphur s a l t s has also been investigated (65, 66), sulphate, sulphite, b i s u l p h i t e , thiosulphate and thiocyanate, and i n each case the t r a n s i t i o n complex i s thought to be a spectroscopically excited state of the -4 -2 solute ions; t y p i c a l l y the f l a s h photolysis of 10 to 10 M aqueous -40-solutions gave S0_ 2" . Ho0 t (S0_ 2 _ . Ho0*) t SO! + e" 3 2 5 1 5 aq CNS" . Ho0 t (CNS~ . Ho0*) t CNS + e" 2 2 aq It seems most probable that the e" becomes solvated i n the solvation sphere of the parent species immediately a f t e r photodetachment has occurred, and evidence for a photochemical cage effect has been reported following scavenging experiments i n the halide system (67). The fla s h photolysis of several phosphate anions i n the vacuum u.v. also gives r i s e to a transient o p t i c a l spectrum p o s i t i v e l y assigned to 5aq - 4-© a (68), and the photoreduction of complex ions such as Mo(CN)g , 4- 3- -Ru(CN)^ and I r C l ^ to give e i s well documented. (69) I f i t i s proposed that the sh o r t - l i v e d intermediate i s the hydrated electron, presumably the absorption bands assigned to e i n the di f f e r e n t systems should have comparable c h a r a c t e r i s t i c s . At f i r s t glance the X and width at half-height of the absorption bands are max not quite as s i m i l a r as anticipated. Perhaps the r e a l l o c a t i o n of the molecules i n forming a solvation sheath about the electron within the sphere of influence of the parent species contributes appreciably to the energetics; thus ^ m a x and the broadness vary according to the nature and size of the ions involved. Certainly the degree of solvation of the ground and excited state has to be taken into consideration when evaluating the threshold photon energy for photodetachment processes i n any polar medium. It has been observed these quantum y i e l d s are generally temperature dependent and less than unity, which implies the -41-p a r t i c i p a t i o n of excited states i n the primary photolytic act, I" ~t r l * . H0G") -»• I + e" aq aq 2 J aq aq In t h i s respect the spectrum of a photoejected electron that has become solvated should possibly resemble that of an F-centre rather than a free e~^. The most recent data published concerning t h i s point was obtained through a modified spectrophotometric technique (70) which measured the absorption spectrum i n the stationary state a f t e r photo-excitation with a modulated l i g h t source. The transmitted l i g h t was monitored by a phase sensitive detector. The u.v. i r r a d i a t i o n of some of the negative species mentioned i n t h i s section gives the same spectrum i d e n t i c a l with that established for e~^ i n equivalent radiolysed systems. The photolysis of ferrocyanide at 254 nm produces e with a l i m i t i n g quantum y i e l d higher than most other processes, 0.67 < $ 2 5 4 < 1.00 (71) Fe(CN)?" + hv -* Fe(CN)^" + e D O aq The p a r a l l e l reduction can be chemically induced. When the solvated ferrous cation i s photoreduced the electron i s transferred from the solvation s h e l l and the l i m i t i n g quantum y i e l d i s low, $254 = 0.06. (72) ^ 2+ , x, 3+ Fe + hv •* Fe + e aq aq aq There i s , however, some controversy as to whether the intermediate i n this reduction i s e or an H-atom, since recent k i n e t i c s a l t effects aq . support the l a t t e r mechanism. -42-The f l a s h photolysis of aqueous N_ and NO, ions produces e , but i n the l a t t e r instance not as the primary photolytic act. (73) Organic substrates have been photolysed to produce an e which i s subsequently hydrated i n the aqueous medium. (74 (a)) For example, Ph - NH- + hv -*• Ph - NH* + e~ , * 254 = °-°25 (71) Electrons photoejected from aromatic amino acids i n frozen aqueous solutions of metal (II) s a l t s have participated i n a t r i p l e t - t r i p l e t absorption process p r i o r to the photoionisation (74 (b)) and become trapped at both chemical centres (H +, M II) and physical s i t e s within the matrices from which they may be photo-bleached. Probably the least chemically complex photochemical system i n which e~ may be generated i s the photolysis of H 2 - saturated a l k a l i n e solutions. Here OH absorbs l i g h t of X < 210 nm OH" + hv -> OH + e" , fJ) l o r. ^ 0.2 aq' T185 and the OH r a d i c a l i s converted v i a reactions (10) and (11) OH + H2 ' -»• H + H 20 (10) H + OH" -»• e" (11) aq aq ' to a second hydrated electron. Although somewhat i n e f f i c i e n t i n r e l a t i o n to some other systems (e.g. halide ions) none of the reaction products in t e r f e r e with o p t i c a l measurements and the predominant decay mechanism should be e" + e" -* (e2,") -> H- + 20H" (1) aq aq 2 • aq .2 aq v ' - 4 3 -Consequently t h i s reaction sequence results i n no o v e r a l l chemical change. In t h i s system the electron l a s t s for several hundred microseconds a f t e r a 25 microsecond f l a s h and preliminary results 2-i n t h i s thesis project indicated that the (e 2 ) a q intermediate i n the decay mechanism i s stable possibly f o r milliseconds, (75). However, the l i f e t i m e of the electron i s extremely sensitive to submicromolar concentrations of oxygen or other impurities. We have used t h i s system to explore the relationship between e and the intermediate 2_ ( e2 )aq5 t n e techniques w i l l be discussed i n greater d e t a i l i n chapter V and further t h e o r e t i c a l considerations l a t e r i n the thesis. - 4 4 -Chapter I I I STRUCTURAL MODELS FOR THE SOLVATED. ELECTRON The elucidation of the structure of the hydrated electron through the interpretation of i t s absorption band i s a d i f f i c u l t task when the evidence consists of one single structureless band covering the entire v i s i b l e spectrum. Despite the growing the o r e t i c a l interest i n the t r a n s i t i o n that gives r i s e to EA the f i r s t excited state of 6 max e~^ i s for the main part assumed to be a 2p and therefore a bound — * state. However the p o s s i b i l i t y that e i s not a bound state i s s t i l l open to experimental judgement. The context of th i s issue i s as follows, (57) .The intense and very broad band corresponding to the absorption band of e has several possible electronic origins of which the three aq most l i k e l y w i l l be discussed. These energies may be modified by the fact that variations i n the immediate dipolar environment are l i k e l y to perturb the t r a n s i t i o n . By photolysing e at an appropriate aq energy and following the subsequent k i n e t i c behaviour of the species at d i f f e r e n t wavelengths, variations i n the in t e n s i t y and po s i t i o n of E A m a x would permit a d i s t i n c t i o n to be made between these proposals. They are: (a) The observed X max may be regarded as the r e s u l t of a single electronic t r a n s i t i o n to an excited state within the potential w e l l ; the broadness and assymmetry of the spectrum suggesting the presence of di f f e r e n t well shapes and possibly thermal v i b r a t i o n a l effects. (b) The spectrum may on the other hand comprise of an envelope of bands ind i c a t i n g the exc i t a t i o n of the electron from a single depth well into an io n i s a t i o n continuum. -45-(c) The spectrum may arise from a s t a t i s t i c a l d i s t r i b u t i o n of solvation wells of various depths i n which excitation i s s p e c i f i c for one p a r t i c u l a r trap whether i t be to an upper state or to a continuum. A l l the models that have been developed so far consider (a) to be the case and the excited state to be a bound state. In creating a conceptual model for a solvated electron the nature of the o p t i c a l t r a n s i t i o n i s not the only assumption that must be made. The potential contours of the trap may be looked at i n at least two ways. F i r s t the excess electron i s i n a natural physical vacancy i n the l i q u i d , which means that i t resides largely i n a hole of uniform p o t e n t i a l , and may be treated as a f i n i t e case of electron-in-the-box problem. A l t e r n a t i v e l y , having polarised the medium during i t s presence the electron then f a l l s into an a r t i f i c i a l cavity the depth the of which may or may not be so l e l y due to the effect o f A e l e c t r o n . In this case the potential f i e l d f a l l s o f f coulombically towards the perimeter of the cavity and the system i s analogous to the H-atom. In polar solvents the effect of the electron on the neighbouring molecules must be included i n any model and generally one of two assumptions i s made; either the medium i s treated as a d i e l e c t r i c continuum or as layers of specified atomic structure most compatible with the structure of water as we know i t (or do not know i t ) . Nearly a l l the models considered i n t h i s chapter have been developed i n the l a s t two years, and most of them i n response to the inconsistencies of e a r l i e r calculations based on the concept of a polaron. The r e a l usefulness of the l a t e r calculations l i e s i n t h e i r variety of approach, i n the s i m p l i s t i c arguments as often successful as the complex, and t h e i r -46-incl u s i o n of terms necessary to account for a l l the new - and anomalous -experimental data. The underlying theory of solvated electron traps i s s t i l l b a s i c a l l y that o r i g i n a l l y developed by Jortner for excess electrons i n l i q u i d s (76, 77, 78). The reference medium for the s t r u c t u r a l ideas that follow i s water unless stated otherwise, for the i m p l i c i t assumption that a l l solvated electrons are p h y s i c a l l y the same i s not one to which radiation chemists would agree at the present time. (1) The Potential Well The electron we are considering i s solvated i n a polar medium and as such i s trapped by a vacancy, a cavity or perhaps i n a potential well that i s a combination of the two. The c r i t i c a l feature of the well i s the potential f i e l d experienced by the electron, which varies according to the physical aspects of the trap. (a) Within the vacancy created by the geometry of neighbouring molecules i n the l i q u i d there i s a uniform zero potential f i e l d . The walls are of i n f i n i t e l y high p o t e n t i a l , but for a r e a l s i t u a t i o n the electron i n the vacancy would see f i n i t e p o t e ntial walls. This r e s t r i c t i o n s l i g h t l y d i s t o r t s the wave functions used thus allowing a small p r o b a b i l i t y that the electron w i l l be found outside the box at any quantum l e v e l . As an electron-in-a-box problem the spacing between the quantum levels w i l l be determined by the dimensions of the box, and i t i s conceivable that the f i r s t absorption (n = 1 to n = 2) could correspond to the release of the electron from the vacancy. The presence of d i f f e r e n t sized vacancies ( i . e . boxes) i n the l i q u i d must be considered the most reasonable cause for the broadening of the spectrum since the separation between quantum levels increases as n increases thus making -47-several t r a n s i t i o n s within the box u n l i k e l y . The shape of the spectrum also r e f l e c t s the additional uncertainty i n the location of the electron due to "tunnelling" and perturbations of the vacancy by the v i b r a t i n g molecular boundary. An increase i n volume i s observed when sodium metal i s added to l i q u i d ammonia (39) and t h i s expansion i s acknowledged as a sign that c a v i t i e s are formed within the solvent molecules and that the electron i n o l i q u i d ammonia i s s t a b i l i s e d i n a large hole ^ 6 A i n diameter; the s i t u a t i o n i n water i s as yet unknown. It was on the vacancy theme that the e a r l i e s t model for an electron solvated i n ammonia was formulated by Ogg (79) who envisaged the electron trapped i n a spherical vacancy with an i n f i n i t e potential wall and surrounded by solvent molecules. The solvation energy and vacancy size derived from t h i s model were not compatible with experimental measurements and so c e r t a i n parameters such as surface tension and electronic s t r i c t i o n were incorporated by other workers. (80) Jortner subsequently achieved s i g n i f i c a n t agreement between the predicted and experimental o p t i c a l and thermodynamic properties of t h i s species, (77). (b) I f the electron finds i t s e l f attracted to an area of favourable potential which i s then further polarised by i t s continued presence a cavity would be created as a result of short range repulsive forces. The potential of the a r t i f i c i a l cavity and associated electron now more resemble the H-atom case which j u s t i f i e s the use of hydrogenic wave functions ( i . e . s,p...) i n assigning the spectrum to an o p t i c a l l y allowed t r a n s i t i o n from the Is ground state to the higher p l e v e l s . The f i r s t t r a n s i t i o n A would be due to a (2p *• Is) e x c i t a t i o n max (presumably at higher energies the t r a n s i t i o n corresponding to electron -48-escape i s given by ( 0 0 p «- Is)) and would extend to 3/4 of the io n i s a t i o n potential Ip. Because energy (Ep) i s required to polarise the medium during the formation of the cavity, Ip w i l l be greater than the heat of solvation AHg by the same amount. It transpires that the AHg estimated by Baxendale has a value very s i m i l a r to the observed EA , (33). J J max' v J (c) A combination of a vacancy and a cavity to give a single depth potential well i n which t r a n s i t i o n s occur into a continuum i s the basis for a t h i r d postulate a t t r i b u t i n g the dominant photo absorption processes i n the v i s i b l e region to the photorelease of the electron from i t s trap. (57) The broadness of the spectrum would be thus accounted f o r , the continuum levels analogous to excitpn conduction bands. The absorption spectrum then represents the envelope of the photoionisation e f f i c i e n c y p r o f i l e ; i t i s f a i r l y common for the photoionisation cross-section of vapour phase molecules to exhibit t h i s sort of p r o f i l e . (d) I f there i s a d i s t r i b u t i o n of wells of di f f e r e n t depths within the solvent then a s p e c i f i c t r a n s i t i o n - be i t electron-in-a-box, a (2p Is) or to an ion i s a t i o n continuum - w i l l take place at di f f e r e n t energies. The re s u l t would be a broad structureless absorption spectrum which was an envelope c h a r a c t e r i s t i c of the solvent. A specified single electronic t r a n s i t i o n w i l l vary i n energy according to the well depths, although factors l i k e interconversion through thermal equilibrium and other solvent temperature effects of necessity complicate the picture. Trapped electrons have been photo bleached i n a s o l i d matrix and the re s u l t i n g change of spectrum i s consistent with t h i s l a s t interpretation,. (71) -Figure 3-1 i l l u s t r a t e s these four intepretations. Although not a l l of the models for e profess the same conceptual approach, a d) Figure 3-1. Energy Levels Corresponding to Possible Absorption Processes (See text for d e t a i l s ) -50-preference for the chara c t e r i s t i c s of an a r t i f i c a l cavity i s generally found which leads to the use of hydrogenic type wave functions. Noticeably absent i s any consideration of the nature of the upper state which i s t a c i t l y assumed to be a 2p l e v e l i n a l l of these calculations. (2) The Cavity i n a D i e l e c t r i c Continuum Approach A model of a polaron i n a continuous d i e l e c t r i c medium has been developed by several workers (81, 82, 83) and the early concepts of the hydrated electron were formulated i n t h i s way with varying degrees of success. The essence of t h i s approach l i e s i n the difference between the r o l e of the s t a t i c (D g) and o p t i c a l (D ) d i e l e c t r i c constants of the solvent i n the presence of rapid electronic motion. Dg i s considered to be the dominant parameter. The in t e r a c t i o n energy between an excess electron and the p o l a r i s a t i o n f i e l d of the medium (represented by Dg and D ) can be estimated using either the electron adiabatic (EA) or self-consistent f i e l d (SCF) approximation. Jortner (77) employs the SCF approximation to derive an expression for the energy of the system E^ containing the medium electrons and the excess p a r t i c l e . E. l The energy levels and charge d i s t r i b u t i o n (r) of e can be obtained aq from t h i s expression using a v a r i a t i o n method with one parameter hydrogenic wave functions for both the ground and f i r s t excited states. -51-The energies of these states having been minimised with respect to the parameters chosen (essentially variations i n the cavity radius R) the opt i c a l e x c i t a t i o n energy hv representing the (2p Is) t r a n s i t i o n i s determined. The potential f i e l d i s continuous on the cavity boundary and constant within, the reference state the polarised medium. The t r a n s i t i o n energy hv = E hv 2p E. , which by substitution i s Is' 3 * l s ( < r - V ^ * l s d T 8TT m J 8T m / 3 - ! _ ) / * f l s dx - e ( l - 1 ) f D D / P 1 S D~ D~ / op s J op s J K f, dx Is Is e d 2 D / op J ^ f dx - e ( l 2p 2p D / op J f, dx Is Is A k i n e t i c energy A orientational p o l a r i s a t i o n = A electronic p o l a r i s a t i o n The upper 2p state cannot be an equilibrium state according to the Franck-Condon p r i n c i p l e , as the nuclei w i l l s t i l l be i n the same r e l a t i v e p o s i t i o n a f t e r the electronic t r a n s i t i o n ; the orientational p o l a r i s a t i o n terms are therefore both governed by the ground state charge d i s t r i b u t i o n . The inclu s i o n of l a s t term, the contribution of electronic p o l a r i s a t i o n energy of the medium electrons to the binding energy of the excess electron i n either s t a t e , i s the basic difference between the SCF and EA treatments; the l a t t e r i s r e a l l y only v a l i d when the binding energy of the excess electron i s much less than that of the medium electrons. -52-In Jortner's picture the solvent molecules i n the immediate v i c i n i t y are repelled by short-range interactions due to the overlap of the o r b i t a l of the excess electron with those of the medium electrons. The two most important relationships governing the eventual size of cavity formed are the magnitude of these interactions i n comparison to the energy required to form the cavity (surface tension s p e c i f i e d ) , and the decrease i n self-energy of the trapped electron with respect to an increase i n cavity radius. Even so the electron i s not s t a b i l i s e d by the actual formation of the cavity, rather by long range interactions i n the d i e l e c t r i c . The o p t i c a l t r a n s i t i o n energy was determined to be 1.35 eV but t h i s corresponded to a cavity size of zero. ( F u l l t h e o r e t i c a l and experimental values are compared i n Table 3-1.) Although t h i s conclusion is_ e n t i r e l y consistent with the trend as anticipated from the e" } r a mm treatment the interpretation of t h i s ultra-diminished cavity i s of great concern. The simplest explanation i s that the locus of the excess electron i s not r e s t r i c t e d to the mathematically formal notion of a cavity and i s delocalised over at least the inner solvent sheath. It i s therefore more meaningful to consider the charge d i s t r i b u t i o n r of the electron instead of R, as the c r i t i c a l parameter, p a r t i c u l a r l y since r i s amenable to experimental confirmation. In the ground state 2.5 < r. < 3.0 A and i n the excited state r„ ^ 4.9 A. Recent k i n e t i c Is 2p experimental work has determined the e f f e c t i v e radius of e i n i t s r aq o ground electronic state to be 2.9 A^ (34). The o s c i l l a t o r strength f^, calculated from the t r a n s i t i o n dipole moment J\>is £ ^2p ^ T ^ o r £ i v e n »^ -53-was 1.1 which implies that any higher t r a n s i t i o n s such as (3p •«- Is) must be very weak i f they occur at a l l as th i s i s a one electron system for which s f. = 1, Furthermore i f the upper state i s a 2p bound state then we can make predictions about the outcome of photo-bleaching and photo-conductivity experiments. In the case of e a m m n ° photo-conductivity has been observed (84) from which i t might be concluded that the f i r s t excited state i s a bound state rather than a l e v e l i n a continuum but the circumstance of the measurements makes them rather i n s e n s i t i v e . No appropriate evidence has been available f o r ea(^> however the recent l i t e r a t u r e does contain many references to photo-bleaching and photo-conductivity exhibited by electrons trapped i n glassy polar matrices, and a l i t t l e for e. , but the evidence i s not consistently i n one i c e ' • J d i r e c t i o n . The temperature dependence of X m & x has been measured i n steady-state and pulse radiolysed systems. Curiously enough the dependence observed i n the former experiments has never been observed i n p.r. work (34). The values of d(hv)/dT for e and e f i t t e d reasonably well v aq amm 3 into the Jortner picture, but those measured since for other solvated electrons do not always f a l l into the framework set by these two e g using Ds and DQp parameters to define the trap. In t h i s context both Dorfman (85) and Holyroyd (86) have found a q u a l i t a t i v e c o r r e l a t i o n between D and X for electrons solvated i n an homologous series of s max 6 alcohols. -54-Jortner more recently demonstrated how a semi-empirical approach could be equally informative as f a r as understanding the properties of e i s concerned (78). aq The potential V(r) acting on the excess electron i n a cavity radius R was calculated following Landau's work on electrons trapped i n i o n i c c rystals (81) l i e 2 V(r) = - ( i — - - i - ) — ; r > R ^ J D^op Ds i * The energy of the l e v e l i s the sum of the orientational and electronic p o l a r i s a t i o n a l contribution. The wave functions are determined v a r i a t i o n a l l y by s e t t i n g 8W^ /3<JK = 0. I2 2 where W. = / ^ ( - ^ v + V(r)) 4^ d x J 1 2n S - e 2 2 r i Dop o when the cavity radius i s set at 1.45 A, a value compatible with the intrepretation of X-ray analysis of water (60 b) and implying a hole about the size of one solvent molecule, t h i s much s i m p l i f i e d approach leads to closer agreement than the previous model; but semi-empiricism should not be taken too seriously as Jortner himself remarks, and even with a judicious selection of R, t h i s model s t i l l cannot explain the s o l -vent s h i f t s observed for other solvated electrons, nor contribute any physical understanding of the structure. Nevertheless t h i s type of potential (V(r)) has been independently used i n models where the l o c a l atomic structure i s an intimate part of theory, and we w i l l look at the outcome i n the next section. Weiss, an original proponent of the polaron model, later proposed quite a different conceptual approach inspired by the similarity between the absorption spectra of e and e. ., (87) . Given D as the cLQ I C G S dominant parameter in determining EA m a x o n a cavity model, a red shift would be anticipated for ^ m a x i - n moving from liquid to ice systems, in which Dg is very small. A blue shift is observed. This discrepancy could be avoided by using a low frequency value for D but by doing so the degree of orientational polarisation is by implication very strong. Since the dielectric relaxation time is 10 ** sec for ice, six orders of magnitude slower than water, such a manoeuvre is not logical. The presence of pre-existing, partially polarised trapping sites must be anticipated, and calculations have shown (54) this quantitatively. Weiss on the other hand proposes a bound electron-hole pair (EL^O* ... H^ O ) which he refers to as an exciton of small radius. This entity can be cleared by selective electron or hole scavengers prior to the appearance of the normal properties of irradiated systems. To explain the relatively long-lived species in scavenger free systems he entertains the idea of triplet excitons. The absorption spectrum of the pair is a charge-transfer process and preliminary calculations using hydrogenic wave functions (really only applicable for excitons of large radius), D and the reduced effective mass of the exciton as the critical parameters, give EA =1.6 eV, a not unreasonable value for ice or water. This 6 max ' electron-hole pair is quite a compact entity compared to the diffuse e as seen through the low free energy of hydration and the large aq effective radius experimentally measured. Nor are the current spectroscopic and e.s.r. data easily assimulated into this framework. Electrons trapped in lattice defects would seem a simpler starting point,, -56-although the electron-hole concept has been invoked i n the studies of electrons trapped i n glassy matrices^ (88). A further series of calculations on a cavity-continuum model published by Noda, Fueki and Kuri (NFK) employ the electron adiabatic approximation to calculate the energy levels of electrons solvated i n alcohols, (89). Contributions from electronic p o l a r i s a t i o n to the Is and 2p states are added i n the f i n a l stages of the calculations once the energy of the levels has been determined using the Landau potential V(r) acting back on the electron. The v a r i a t i o n a l method was used i n conjunction with one parameter hydrogenic wave functions, and the results of the calculations are tabulated i n Table 3-1. These authors also estimated the temperature dependence of the Is and 2p energy levels i n terms of the d i e l e c t r i c constants and the cavity radius, presenting a very interesting contour map of o p t i c a l t r a n s i t i o n energy i n 0.1 eV steps as a function of both R and Dg (D i s considered to be constant at a value of 2) for those solvents i n which the binding energy of e g did not invalidate the EA approximation they had used. In matching the observed EA and D to obtain a value of R over a range „ fflclX S of alcohols, i t was found that R decreased almost l i n e a r l y with increasing D , with the exception of n-butanol. It also appeared that o _ i d R/^j. A deg was correlated with the percentage of hydroxylic bonds to the t o t a l number of bands i n each alcohol, i n f e r r i n g that there i s at least a q u a l i t i a t i v e relationship between the cavity dimensions and the strength of inter-molecular bonds such as H-bonds within the solvent sheath. -57-Th e experimental temperature dependence of E ^ m a x for electrons solvated i n s t r u c t u r a l l y different alcohols was s t i l l too high for a simple cavity.model, that i s an unusually large c o e f f i c i e n t of thermal expansion had to be assigned to the cavity to reproduce the data. Nevertheless as was stated e a r l i e r the increase i n the width of the absorption band may be ind i c a t i v e of an increase i n R as the temperature i s raised. NFK's contention that the .asymmetry on the low energy side of the band i s due to the presence of larger c a v i t i e s would be supported by photo-bleaching e g with selective wavelengths and observing a blue s h i f t i n X . A s yet such predictions remain unsubstantiated, max J * In the d i e l e c t r i c continuum models discussed so far a one parameter hydrogenic wave function was used for both ground and excited states. By choosing a three parameter wavefunction that includes about 47% 2s character i n the ground state, and a one parameter 2p, hydrogen-like wave function for the excited state Fueki, Feng and Kevan (FFK) perform the same calculations as Jortner and obtain a t r a n s i t i o n energy EX 6 / max that when R i s set to zero i s greater than the observed value, (90). This energy w i l l decrease as the cavity dimensions increase, demonstrating that the inherently more appropriate SCF approximation does not necess-a r i l y terminate i n an abstract concept of R. The best f i t for EX } r max o i n a revised FFK cal c u l a t i o n would appear to be for R ^  1.5 A and although these computations have not been made, i n the authors' own words thi s should make the experimentalises a l i t t l e happier with theory". The influence of l o c a l d i e l e c t r i c saturation, the lowering Ds below the bulk value i n the e l e c t r o s t a t i c p o tential gradient of the polar solvent, was investigated by Fueki who treated the problem within -58-the EA framework on a cavity-continuum model, (91). R was assumed to be zero and D a function of distance from the "centre" of the c a v i t y . s ' The values for the o p t i c a l t r a n s i t i o n energy and Is, 2p charge -d i s t r i b u t i o n of the e l e c t r o n were c a l c u l a t e d i n c l u d i n g a c o r r e c t i o n f o r d i e l e c t r i c s a turation. Fueki concluded that f o r a small c a v i t y of o R 4 1 A the e f f e c t on the ground state Is energy l e v e l was appreciable while the upper states remained r e l a t i v e l y unaffected. I f R > 1.5 X and the p o t e n t i a l energy was constant within the c a v i t y then d i e l e c t r i c s a t u r a t i o n becomes unimportant. The e f f e c t on the charge d i s t r i b u t i o n i n the Is state was markedly diminished i n the more d i f f u s e 2p state which suggests that the solvated electrons i n the excited state ( i f i t i s a 2p state) w i l l be found i n p o t e n t i a l l y s i m i l a r environments regardless of the type of c a l c u l a t i o n s used. (3) Atomic Models The c r i t i c a l l o c a l geometry of the l i q u i d i n which the e l e c t r o n i s trapped has been assigned t e t r a h e d r a l , s p h e r i c a l and planar symmetry i n an attempt to explore the f e a s i b i l i t y of the s t r u c t u r a l approach. Natori and Watanabe (NW) have evaluated the energy l e v e l s of an excess e l e c t r o n caught i n a tetrahedral arrangement of water molecules analogous to an e l e c t r o n i n a l a t t i c e defect of a c r y s t a l , (92). The p o t e n t i a l acting on the e l e c t r o n due to the four i n s i d e hydrogen atoms, the four oxygen atoms that form the tetrahedron, and the four next nearest hydrogen neighbours i s computed f o r each case and the t o t a l p o t e n t i a l V(r) experienced by the electron i s written: 12 V(r) = V ( S ) + V o s = 1 -59-Th e origin of the co-ordinate system is the centre of the tetrahedron, r is the position vector of the electron with respect to the position of the s-nucleus, V(s) the potential due to s nuclei and V q the induced potential from molecules outside the immediate tetrahedral environment. Spherical symmetry is assumed for V(s) which depends only on the value of r : V S T <\,S 1 + 6 Ks^i 2 dx Hydrogenic wave functions are used, and 6 is the fraction of the electron which occupies the Is orbital estimated from the dipole moment of water and the hydroxylic bond to be 0.674 atomic units. Substituting -1/2 *ls- ty = 1 7 e x p t 0 g i v e V s ( rs^ Vs(r s) = -0:326/ - 0.674 (1 + 1/r ) exp (-2^) s the hamiltonian for the one electron system, neglecting'V , is H = - A + V Vs ( r J 2 s = 1 In the variational calculations performed NW are basically stretching the bond distances and bond angles in the water molecules to vary the oxygen-oxygen distance for a best f i t energy. The results are included in Table 3-1. To what extent the low theoretical value of E\ Amax arises through neglect of V , or perhaps the experimentally observed value is raised through participation of upper excited states is at the present time an open question. The NW oscillator strength was -60-determined to be 0.26 and t h i s i s more consistent with hidden, higher contributions i n EA fobs") than the Jortner value of f. = 1.1. max v J I Experimental measurements give a value of 0.65. Further evaluation of the tetrahedral model becomes questionable i n the l i g h t of recent ESR experiments that detected hyperfine interactions for electrons trapped i n i c e ^ (93) . Whereas the NW water-ice analogy would predict interactions between four protons, the ESR data d e f i n i t e l y indicates that s i x equivalent protons are involved, thus the tetrahedral order they propose i s incompatible with e g i n ice structure. I t remains to be seen whether calculations along these lines are extended to a symmetry involving, for example, s i x protons i n either an octahedral f i e l d of three orientated water molecules or c e n t r a l l y located i n a puckered hexagon of s i x oxygen atoms, with hydrogen atoms situated along the lines between adjacent oxygen atoms. The ESR data had been obtained from ice that had c r y s t a l l i s e d i n cubic form at 160°K. (X-ray analysis of the various temperature dependent phase structures i n the ice was performed on s i m i l a r l y prepared samples). Although the l i n e width was narrow i n ice ind i c a t i n g a f a i r degree of uniformity between the trapping s i t e s , s i m i l a r experiments on electrons trapped i n alcohols at 77°K produced r e l a t i v e l y broad lines lacking i n hyperfine structure and leading to the opposite conclusion about the d i s t r i b u t i o n of traps. The g value for e g i n ice 2.0008° ± .0005 was lower than the standard free spin value g = 2.0023, which could mean than that the o r b i t a l of the excess electron does not possess f u l l spherical symmetry and overlaps with the molecular o r b i t a l s of i t s immediate environment. The s i t u a t i o n for e i n s o l i d s alcohols i s complicated by the appearance of ESR signals from other -61-r a d i c a l s . However the g-values are higher than those i n ice but s t i l l marginally less than the free spin value. This negative s h i f t c h a r a c t e r i s t i c of e and F-centres was also recently reported f o r amm 1 . * e > (94). The signal had a su r p r i s i n g l y narrow l i n e width < 0.5 gauss, aq and a g-value of 2.0000 ± .0002 once again implying a degree of d e l o c a l i s a t i o n over a few solvent molecules with s t r u c t u r a l l y quite s i m i l a r c a v i t i e s . I f i n l i q u i d water these trapping s i t e s must (for energetic reasons by analogy with ice) ex i s t p r i o r to the a r r i v a l of the excess electron then i t i s reasonable to suppose that certain transient molecular arrangements w i l l be favoured by the electron as s a t i s f y i n g the minimal energy pre r e q u i s i t i e s for a trap. These consequently become s t a b i l i s e d by i t s presence, and i f the dipole arrangement i s not appreciably distorted by the electron the traps may appear f a i r l y uniform at the levels of s e n s i t i v i t y employed. A second atomic model has been put forward by Iguchi,(95) . The molecular dipoles are orientated i n spherical geometry about the excess electron, and hydrogenic wave functions are fed into the by now f a m i l i a r one electron problem. Since the o r i g i n a l purpose of these calculations was to incorporate a thermal dependency for ^ m a x for the electron solvated i n a series of alcohols, t h i s parameter becomes a function of both the angle (y) between the permanent dipole ( u ). of the solvent molecule and the s t a t i c e l e c t r i c f i e l d (E) due to a l l the solvent molecules, and of the thermal expansion properties of the l i q u i d . The potential f i e l d seen by the electron i s computed as a sum of the permanent dipole contributions and induced p o l a r i s a t i o n for dipoles within the locus of the electron i t s e l f . The co-ordinate o r i g i n of the system i s taken as the equilibrium -62-point i n the 'trap' about which the electron moves with radius r and towards which solvent dipoles are orientated by t h e i r own s t a t i c e l e c t r i c f i e l d . The density n^ of molecular dipoles U q with orientational energy z at T°K i s calculated from a Boltzman type d i s t r i b u t i o n . This density n i s included i n the derivation of the potential V(r) whence calculations take t h e i r normal v a r i a t i o n a l course. Encouraged by the improvement i n agreement between his predicted values and those observed, Iguchi l a t e r imposed a further condition i n which the electron sees a cut-off for V ( r ) ; i n other words R becomes a v a r i a t i o n a l parameter i n keeping with i t s r o l e i n other models. Additional experimental data on electrons solvated i n alcohols i s now required for any further evaluation of t h i s model. This chapter concludes by mentioning some molecular o r b i t a l treatments on the basis of a simple structure for e a q - Several years ago Raff and Pohl (96) published o p t i c a l t r a n s i t i o n energies calculated for a H 2 + system perturbed by two OH ions; the excess electron was envisaged caught within the f i e l d of the protons of two water molecules i n planar configuration. Spin resonance studies on e g i n ice l a t e r cast doubts on t h i s model. Recently McAloon and Webster revived interest i n such a dimer as a simple model for the purpose of testing the m.o. description of the excess electron i n a polar medium (97). They also considered a non-planar dimer based on a fragment of the l a t t i c e structure for ice i n which the planes containing the hydrogen and oxygen atoms from each of two water molecules are adjacent and the angle between them i s a variable parameter. Both cavity and polaron features were alternately retained for the trap by allowing the electron to either remain l o c a l i s e d between protons from di f f e r e n t molecules or d e l o c a l i s i n g -63-. over the whole dimer v i a 2p o r b i t a l s . The crux of the approach l i e s with the hydrogen o r b i t a l s which i n t h e i r model are not those of free atom but are shrunken with respect to an increased eff e c t i v e nuclear charge. (cf. ', and the i n c l u s i o n of 2s character i n the wave function i n FFK c a l c u l a t i o n s ) . The outcome of the calculations are i n Table 3-1. Although la b e l l e d with the authors' self-accusation of "blatant a r t i f i c a l i t y " , t h i s exercise provides quite credible dimensions f o r the cavity size of e - and an acceptable value for EA , The ' aq * max comparative behaviour of e and e under t h i s treatment does add to r amm aq the impression that electrons solvated i n water are not the same as excess electrons i n ammonia, but a more r e a l i s t i c perspective may be available a f t e r m.o. calculations are extended to more complex s t r u c t u r a l models. In the meantime, the o r i g i n of the t r a n s i t i o n or t r a n s i t i o n s giving r i s e to the unusual absorption spectrum of e s t i l l remains unknown. One of the purposes of t h i s research was to t r y and distinguish between the three possible origins outlined at the beginning of the chapter, (a) a single electronic t r a n s i t i o n to an upper bound state, (b) the excitation of the electron into an i o n i s a t i o n continuum, or (c) a s p e c i f i c t r a n s i t i o n occurring i n p o t e n t i a l wells of various depths. By photolysing e at selected wavelengths i t was hoped to at least aq narrow the choice to (a) or (b) and (c). Experimental Fueki, Feng and Kevan Values Water o R A A H (eV) hv (eV) - (A) Is (A) -d_(hv) x 1(T dT f . 1 1.7 1.72 2.9 •2.9 0.65 (SCF) 0.0 1.81 2.18 2.4 4.6 JORTNER (SCF) 0.0 1.32 1.35 2.5 to 3.0 4.9 •3.3 to 2.2 1.1 FUEKI (SCF) > 0.0 1.81 2.18 2.4 4.6 JORTNER (EA) 1.45 1.85 1.65 2.8 Natori Watanabe Webster and McAloon planar non planar 0.75 0.90 0.80 1.72 0.26 1.71 i R A A H eV hv r A s -d hv dT Ethanol' <2.0 1.77 -3.4 Noda et a l (SCF) 1.32 * 1.77 2.88 -2.99 Iguchi 3.61 1.69 -1.63 f . l 0.87 Table 3-1. A Comparison Between Experimental and Predicted Values: Some Solvated Electron Parameters. -65-Chapter IV PULSE RADIOLYSIS AND KINETIC LASER PHOTOMETRY The events during and after the pulse r a d i o l y s i s of aqueous solutions were followed through the technique of k i n e t i c laser spectro-photometry. A laser beam acting as an ideal monochromatic coherent and continuous l i g h t source was directed through the radiation c e l l . Variations i n the transmitted i n t e n s i t y of the beam during and a f t e r the i r r a d i a t i o n of the sample could be detected by a photomultiplier A.C. coupled to an oscilloscope. Narrow band pass f i l t e r s and lead shielding were used to cut out any d i s t o r t i o n of the signal due to Cerenkov emission: and X-ray pulses respectively. Kinetic data could be obtained from the analysis of the decay curves of the absorbing species, photographed from the C.R.T. screen as a time-dependent voltage, and traced out on 10,000 A.S.A. polaroid f i l m . (1) Febetron Accelerator A pulsed electron accelerator known as a Febetron and operating on the Marx Surge C i r c u i t p r i n c i p l e (manufactured by F i e l d Emission Corporation,' Oregon) was employed as the source of high energy i o n i s i n g radiation. Our prototype machine model 2660 has been modified through the additions of a pulse shortener (model 2770) and with the use of suitable electron tubes (models no. 5510 or 5515) delivers a 3 nanosecond pulse of ^0.5 MeV electrons with an integrated beam current of 5000 amperes. This corresponds to an output power of 10 Joules per pulse. The beam characteristics were measured with one of the two apertured Faraday cups, s p e c i f i c a l l y designed for t h i s task, positioned -66-i n front of the electron tube window i n such a way as to simulate the geometry of the radiation c e l l s . One cup contained a T § M Research Products Model GR - 1 - 05 current viewing r e s i s t o r (C.V.R.) with an impedance of 0.0507 ohm. The voltage pulse across the C.R.R., trans-mitted v i a RG 58 doubly shielded 50 fi cable, was monitored on a very fast (2.4 n sec) r i s e time oscilloscope, (Tektronix 454). The observed magnitude of the pulse of 217 V represented an integrated beam current of 4350 amperes from the 5510 electron tube. The width at half-height of the pulses from t h i s tube measured on a Tektronix 519 oscilloscope (rise time < 0.35 n sec) was determined to be 6 n sees. The second Faraday cup (also a F i e l d Emission design) functioned on a s l i g h t l y d i f f e r e n t p r i n c i p l e . There was no r e s i s t o r within the cup; the beam was collimated i n a logarithmically tapered aluminum c o l l e c t o r and the current a r r i v i n g at the end plate merely saw the cable as a 50 ft impedance. Consequently the wave form of the resultant voltage pulse followed that of the beam current giving a width at h a l f height of 3 n sees for the 5515 electron tube, but the t o t a l current had to be determined by other means, namely calorimetry. The majority of experiments were performed under conditions of maximum output from the accelerator; that i s , with a beam of 0.5 19 MeV electrons depositing a t o t a l energy of 10 eV at a dose rate of 26 — 1 — 1 5 x 10 eV gm . sec . I f necessary the average k i n e t i c energy of the beam electrons and the beam current could be decreased by working at a lower D.C. charging voltage. The impedance of the electron tube remains reasonably independent of the f i n a l charging voltage. -67-Although the r e p r o d u c i b i l i t y of both the pulse shape and beam output was determined to be within ± 4% (r.m.s.), in d i v i d u a l pulses were s t i l l s ensitive to the rate at which the capacitors were charged and so at a l l times a uniform charging procedure was followed i n addition to purging both the switch and module gas chambers of the accelerator every 100 pulses. Figure 4-1 shows t y p i c a l pulse shapes, and a block diagram of the accelerator. (2) Experimental Solutions Laboratory d i s t i l l e d water was r e d i s t i l l e d from either a c i d i f i e d sodium dichromate or a c i d i f i e d potassium permanganate solutions. This water w i l l be referred to as doubly d i s t i l l e d or 2D water. When a further degree of p u r i f i c a t i o n was deemed necessary the 2D water was y-irradiated i n a C o ^ Gammacell f a c i l i t y for several hours and then refluxed over a l k a l i n e potassium permanganate solution for a minimum of 10 hours. The water collected i n t h i s way w i l l be referred to as t r i p l y d i s t i l l e d or 3D water. The u.v., v i s i b l e and near i . r . spectra of 2D and 3D samples were compared against the spectrum of laboratory d i s t i l l e d water and there was no evidence for any impurities i n the r e d i s t i l l e d l i q u i d s , nor any discernible difference introduced by the chemistry of the d i s t i l l a t i o n substrate. Once prepared the p u r i f i e d water was collected and transferred to 2 - l i t r e resevoir flasks b u i l t into the flow systems of the various pieces of equipment and kept under an atmosphere of spectroscopic grade (Matheson) helium gas. Any solutions of other s a l t s were always made up from the supply of 2D or 3D water i n s p e c i a l l y cleaned f l a s k s . ' -68-HIGH VOLTAGE POWER SUPPLY ISOLATION RESISTOR AIR REGULATOR c v o I o < to cx, o 00 I o ! CHARGING [ VOLTAGE 1 I PULSE TRANSFORMER r i i i I J I TRIGGER DELAYED TRIGGER AMPLIFIER I I | | PULSE I I FREON REGULATOR pi i—i < MODEL 706 PULSER i i ! • U . L i PUMP Transformer O i l ASSEMBLY I 5- E. siG FREON - 12 SHORT PULSE ADAPTER 8 TUBE FARADAY CUP-X 50 Attenuation s s hf 1 / \ / V 5 V 20 KV , 7\ <5 KV X 5 Attenuation 10 jnY 5 nsec per d i v i s i o n 5 nsec per d i v i s i o n Figure 4-1. System Block Diagram for a standard Model 706/2670 System. Showing t y p i c a l waveforms from the Model 5515 Electron Tube as measured i n th i s laboratory. -69-Sulphuric acid used i n these experiments was d i l u t e d from a concentrated analar BDH supply, the sodium sulphate solutions from BDH anhydrous sodium sulphate; the other chemicals, namely sodium hydroxide, potassium hydroxide and barium hydroxide were analar grade and used without further p u r i f i c a t i o n . The methanol and ethanol were Eastman Kodak grade l i q u i d s . The flow system, from the resevoir supply into the radiation c e l l and out to the draining l i n e s , was constructed e n t i r e l y from glass. The 2 - l i t r e three-necked resevoir flasks each had an i n d i v i d u a l flow l i n e to the rad i a t i o n c e l l that could be sealed by stopcocks when not i n use. The c e l l s were f i l l e d by pressurising the system and forcing the solutions through the l i n e s . The helium prepurified gas passed through a safety trap (to prevent the accidental f i l l i n g of the gas lines due to a sudden back pressure from the resevoirs) into the experimental solution v i a a fine f r i t t e d glass cone. . The solutions were thus bubbled vigorously and deoxygenated - to less than 2 ppm - over several hours before actual use and continuously during any experimental day. The over-pressure of gas escaped from the resevoir through a stopcock on the side neck of the fla s k which led to another bubbler system i n a second f l a s k . In t h i s way the water-saturated He gas could be used to degas a second aqueous solution whose contaminants might be d i f f i c u l t to remove from the pure water system. In any case the radiation c e l l s could be f i l l e d and replenished f o r each pulse from working behind the lead shielding screens which increased the experimental e f f i c i e n c y compared to e a r l i e r practice. Discarded l i q u i d s drained d i r e c t l y into a covered sink. When the flow lines were not i n commission they remained -70-A Accelerator B Control Console C Radiation C e l l D Laser E Lead Screens F Flow Systems G Photomultiplier H Lenses, I r i s I Power Supply J Junction Box K Oscilloscope L Ground Stake Figure 4-2. The Accelerator Laboratory -71-f u l l of 2D water and closed to a i r to minimise any contamination. A l l experiments were done at room temperature 19° C ± 3°C and fresh solutions introduced into the c e l l s p r i o r to each radiation pulse. The entire layout of the laboratory can be seen i n Figure 4-2. (3) Optical Equipment A continuous helium-neon gas laser provided an intense source of appropriate wavelength for the monitoring of the k i n e t i c behaviour of e . Several models, each of varying output, were employed over aq the research period but that used predominantly was a Spectraphysics model 120. This D.C. excited laser has an extremely stable output over long periods of time, and a r e l a t i v e l y small divergence of 1.7 m i l l i -radians. The output at 632.8 nm i s 5 mW and the beam diameter 0.65 mm within which over 85% of the output power was contained. In f a c t , i t was seldom necessary to monitor the f u l l beam on the photomultiplier -which would have given r i s e to far too high a current i n the dynode chain even at very low operating voltages - because the transmitted beam was collimated to a very small cross section through pinholes on the radiation c e l l window. In those cases where the events occuring across the whole beam were under investigation, a group of neutral density f i l t e r s between the c e l l and the photomultiplier could reduce the 3 int e n s i t y of the beam by 10 i f necessary. The laser was mounted on a stand machined from aluminum and brass (14" x 12" x 30" long) s p e c i f i c a l l y designed to permit very precise l a t e r a l , v e r t i c a l and angular control over the di r e c t i o n of the beam. The stand was about 1 metre i n distance from the radiation c e l l , and -72-any small degree of d i f f r a c t i o n i n the incident beam caused for example by marginal off-axi s optics within the laser cavity or i r r e g u l a r i t i e s along the inner surface of the plasma tube could be removed by ins e r t i n g a non-reflecting variable aperture immediately i n front of the window of the radiation c e l l . The beam emerging from the opposite window of the' c e l l contained not only p a r a l l e l l i g h t but r e f l e c t e d , d i f f r a c t e d and scattered frequencies as we l l . Out-of-phase components cause a severe deterioration i n terms of noise and the introduction of pinholes into the alignment enhances t h i s undesirable ef f e c t . It therefore was necessary to remove the detection equipment as far away as possible -about 3 metres - from the source of the problem and once again remove the distorted periphery or outer d i f f r a c t i o n rings from the p a r a l l e l beam before focussing the l a t t e r onto the photocathode. Despite a l l these precautions taken to reduce the noise accompanying the laser signal there remained an ultra-high frequency r i p p l e which could not be eliminated. This was not caused by any inherent i n s t a b i l i t y i n the output of the laser whose long term power d r i f t was < 5% nor by 60 cycle pick-up from the main power l i n e s . These very high frequency > 200 MHz reproducible signals superimposed on the trace w i l l be referred to as laser r i p p l e and are thought to arise i n any laser that does not operate on a single-mode p r i n c i p l e . That i s , they are additional resonance frequencies present i n the normal o s c i l l a t i o n s , spaced f = C, 2C, 3£ ... where C = speed of l i g h t , 2L 2L 2L and L = o p t i c a l cavity length for the laser (L = 39 cms for t h i s model). Further perturbation of these p a r t i c u l a r out-of-phase components occurs when the cavity dimensions vary - imperceptibly to us - as a resu l t of thermal effects within the laser head. Then these cavity resonances - 7 3 -become symmetrical about the main neon resonance l i n e they go to zero i n a s e l f - l o c k i n g mechanism. This event may be observed on the oscilloscope as a KHz "beat" i n the displayed s i g n a l . Fortunately the laser r i p p l e did not normally i n t e r f e r e to any appreciable extent with the transient absorption signals but decreased the s e n s i t i v i t y of the detection equipment f o r input signals < 5 mV and enforced a l i m i t on the signal-to-noise r a t i o . The laser beam was collimated either externally from an 0.7 to 0.07 2 mm cross sectional area or within the radiation c e l l (on the trans-2 mitting side) to 0.02 mm . The l a t e r a l p o s i t i o n of both sets of pinholes could be f i n e l y adjusted, that on the c e l l to within 0.1 mm by means of a b u i l t - i n micrometer. A narrow band pass f i l t e r transmitting wavelengths of 632.8 nm ± 0.5 nm was positioned i n front of the photomultiplier to prevent any Cerenkov photons from being detected since the photocathode was much more sensitive i n the lower wavelength regions where Cerenkov i n t e n s i t y reached i t s maximum. This f i l t e r , Baird-Atomic Inc. Interference type B 11 which transmitted 60% of the incident l i g h t at 632.8 nm and less that 1% outside the band pass, was mounted on an adjustable o p t i c a l bench stand manufactured to allow both angular and l a t e r a l adjustment. The neutral density f i l t e r s used were of two types; those of O.D. 1 and 2 were interference f i l t e r s made by Baird-Atomic which were li n e a r to within 1% over the v i s i b l e and near i . r . region, and the remainder were p a r t i a l l y transmitting s i l i c a plates vacuum coated with Chromel A according to a published technique, (98). The l a t t e r were of varying density, and often were more useful than the interference type as -74-combinations of the Baird-Atomic f i l t e r s usually led to destructive interference patterns which increased the noise l e v e l . The l i n e a r i t y of the detection system i t s e l f was p e r i o d i c a l l y checked using neutral density f i l t e r s . To improve the signal-to-noise r a t i o , of p a r t i c u l a r importance i n those experiments u t i l i s i n g the pinholes, the non-parallel periphery of the laser beam was screened from the photocathode and the true beam focused through a rotatable polished quartz lens (focal length 8.5 cm) onto the gr i d some 10 cms distant. (4) Photomultipliers and Oscilloscope Two photomultipliers were used i n th i s research, one RCA 1P28 with a S5 spectral response and a Hamamatsu (HTV)R213 considerably more sensitive i n the red with a spectral response close to S20. This increase i n sensitivity,however, was accompanied by a decrease i n s t a b i l i t y when operated continuously for several hours i n r e l a t i o n to the performance of the 1P28. With s i m i l a r nine-stage dynode chains they both have fast time resolution c h a r a c t e r i s t i c s ; t y p i c a l l y f o r the -9 1P28 the anode pulse r i s e time was 2 x 10 seconds under normal operating conditions. Each phototube, plugged into a compact metal base holding the anode and .dynode r e s i s t o r s and any u l t r a high-frequency electronic f i l t e r s , was housed i n a blackened metal cylinder containing a s l i t 3 mm x 10 mm that exposed part of the photocathode. The entire assembly was then placed i n a copper box lined with 1/4" thick lead which shielded the tube from extraneous radio-frequency signals, X-radiation and electromagnetic f i e l d pick-up r e s u l t i n g from the f i r i n g -75-of the accelerator. The output and input coaxial cables were grounded to t h i s box which had been blackened both inside and out. Reasonable v e n t i l a t i o n was provided by d r i l l i n g several small holes i n the l i d of the box and i f necessary blowing.dry a i r or nitrogen through to maintain a cool temperature. A small hole of 1/8" diameter was d r i l l e d through the front face of the box to allow l i g h t to f a l l onto the cathode; the photomultiplier at a l l times was automatically protected by a square of thick black velvet draped over t h i s front face u n t i l experiments began. The power supply used i n both the r a d i o l y s i s and photolysis experiments was a Fluke model 412 B supplying a D.C. high voltage s t a b i l i s e d to within 0.005% per hour or 0.02% over a day's a c t i v i t i e s . Operating voltages were generally between 500 to 750 v o l t s , normally at 650 volts where the l i n e a r i t y of the photomultiplier response was good and the dark current and noise pulses from the low impedance anode load r e s i s t o r s were s t i l l minimal. Below 550 volts the l i n e a r i t y i n the response was not very good. Doubly shielded coaxial transmission cables connected the power supply to the photomultiplier. Having outlined the input features of the electronic system, we w i l l b r i e f l y discuss some of the cause-and-effect aspects of the output signals from the photomultiplier on a nanosecond time scale. Often i t was a question of p r i o r i t i e s between the r i s e time and the noise on a very small s i g n a l ; i n practice we used the shortest cable lengths possible and varied the load r e s i s t o r from 50 ^ to 10 k ^ . In those experiments where the fluorescence of a s c i n t i l l a t o r was collimated to photolyse e , the i n i t i a l 10 to 20 nanoseconds after the aq electron pulse were the most c r i t i c a l . Therefore the voltage pulse -76-across a 50 load r e s i s t o r was fed into a 12" long 50 fi (RG58) impedance matched transmission cable terminating i n 50 fi at the oscilloscope, and the operating voltage of the photomultiplier adjusted to gain the 2.4 nsec r i s e time of a s p e c i f i c l e v e l of attenuation on the oscilloscope. The maximum current that the dynode chain could sustain was estimated by R.C.A. to be 10 m amps but to avoid fatigue, thermal t a i l s and non-l i n e a r i t y i n the response, the maximum current at a l l times was between 1 and 2 m amps. With too low a current i t was found that the signal became sensitive to fluctuations i n the i n d i v i d u a l dynode voltages. At currents over the self-imposed l i m i t the degree of amplified laser r i p p l e feedback effects s l i g h t l y distorted the l i n e a r i t y . Thus for very small signals rather than increasing the operating voltage the 50 a load r e s i s t o r was replaced by a 93 low noise r e s i s t o r , the lines' replaced with R G 71-B-U shielded and impedance matched coaxial cable terminating i n 93 fi at the oscilloscope. Whereas i n the previous case the r i s e time of the signal was on average < 4 nsec, t h i s substitution increased i t to < 10 nsec which was quite acceptable for the k i n e t i c experiments following the very slow decay of e i n low dose regions (an i n i t i a l aq O.D. of about 0.1) of the c e l l . For even smoother signals we replaced t h i s load r e s i s t o r with 500 and i n the microsecond flash photolysis studies used as high as 10k ^ r e s i s t o r s . O r i g i n a l l y a variable load r e s i s t o r grouping was inserted actually i n the photomultiplier base but experience showed that t h i s produced switch noise and ringing signals observable i n the 50 and 100 cases as a result of an induced capacitance as the signal was received; thereafter each c i r c u i t was i n d i v i d u a l l y made. -77-In order to observe pulses of a few nanoseconds i n duration, an oscilloscope of suitable bandwidth was necessary and the 150 MHz Tektronix model 454 f u l f i l l e d t h i s requirement except on occasions when we wished to accurately examine the electron pulse shape and for t h i s a Tektronix 519 with a r i s e time of ^ 0.35 nsec was used. The 454 oscilloscope had a r i s e time of 2.4 nsec on 20 m Vcm 1 s e n s i t i v i t y or less on the v e r t i c a l amplifier but t h i s f e l l to about 5 nsec when the most sensitive scale 5 m Vcm 1 was used. The additional gain i n time resolution of the 519 i s because the input signal i s not amplified but i s fed d i r e c t l y into the CRT. The minimum input voltage pulse required i s therefore 9.5 V and the maximum acceptable signal 15 V; with the incorporation of appropriate alternators or special pre-amplification units p r i o r to the input connector t h i s very fast r i s e time can be achieved for signals of varying strength. When the . electron beam pulse was measured using the Faraday cup technique; mentioned e a r l i e r , or i n any other experiment, a l l attenuators and terminators employed were Tektronix products. Although the power l i n e to the 454 from the 110 V mains supply was f i r s t diverted through a radiofrequency f i l t e r attached to the back of the oscilloscope, the l a t t e r s t i l l triggered through pick-up from the other transmission lines and i t was imperative to keep the doubly shielded cable from the power supply to the oscilloscope well out of the area as otherwise random series of "ringing modes" superimposed them-selves on the expected sig n a l . Normally the oscilloscope was i n t e r n a l l y triggered from a p o s i t i v e slope i n the AC coupling mode, a l l -78-low frequencies < 10 KHz being rejected. The 100% lev e l of transmitted l i g h t was recorded immediately before each experimental pulse by manually stopping the laser beam at about 1 KHz, with low frequency signals accepted. When a very low signal was anticipated the oscilloscope was triggered externally by a small pick-up voltage induced i n a co-axial cable, clipped to the radiation c e l l , by the passage of the electron beam. A Tektronix type C-40 polaroid camera f i t t e d to the screen of the oscilloscope recorded the signals on the CRT on 10,000 ASA Polaroid type 410 f i l m . This high speed f i l m i s necessary to u t i l i s e the c a p a b i l i t i e s of the fast lens on the camera and provides a very accurate photograph of u l t r a - f a s t low int e n s i t y traces. (5) Grounding and Shielding P a r t i c u l a r shielding problems arise with the use of a Febetron, including the intense electro-magnetic f i e l d generated during the electron pulse, the production of X-rays and continuous r . f . pick-up from the pulse transformer and other unspecified sources. The following pre-cautions have been taken permanently i n the laboratory, and more s p e c i f i c a l l y on the equipment i n t h i s research. (a) Lead shields, 1/4" thick and eight feet high separate the area i n which the accelerator i s f i r e d from the remaining laboratory space, i n which electronic detection equipment i s normally placed. The shields eliminate most of the X-rays, but since several holes (y 2" i n diameter) have been d r i l l e d through the lead to permit l i g h t from the laser beam or other sources to pass onto the lenses, f i l t e r s and -79-detection system, anything electronic on the o p t i c a l bench appeared to be subjected to an X-ray pulse. A second small 1/4" lead sh i e l d was mounted on the o p t i c a l bench immediately i n front of the photo-m u l t i p l i e r , and the l i g h t was focused into the centre of a 1/16" diameter hole i n the lead, from which the l i g h t beam then diverged into the hole i n the copper box. The l a t t e r gave the additional r . f . screening to the photomultiplier. (b) A l l the co-axial cable was doubly shielded and served as a ground since they connected the otherwise f l o a t i n g oscilloscope, power supply and copper box to a common ground. The Febetron was attached with thick 2 " grounding tape to a ground stake buried through the wall and deep into the earth outside. On the rare occasions the earth became too dry l i t r e s of saturated copper sulphate solution were poured around the stake area. The radiation c e l l s were always attached d i r e c t l y to the front flange on the Febetron and by vir t u e of contact should be at the same potential as the machine. The best way of reducing noise from the 'nose cone' area was to wrap the entire large pulse-shortmer adapter and the c e l l i n aluminum f o i l . The radiation c e l l i t s e l f was iso l a t e d from the f i e l d s i n a machined thick aluminum cap (allowing for entry and exit ports to jut out) that f i t t e d exactly over the c e l l onto the front flange of the machine. Two very small windows were cut out to allow the passage of the laser beam through the sample within the c e l l . With a l l these precautions, the inherent noise l e v e l of the system during operation appeared to be t r u l y minimised; the laser r i p p l e was the l i m i t i n g factor. Sometimes pick-up from other laboratories and as -80-yet unspecified r . f . sources added to th i s an u l t r a high frequency noise of immense proportions which generally c u r t a i l e d experimental a c t i v i t i e s ; r . f . noise from the pulse transformer on the Febetron had been success-f u l l y controlled by the use of a special cross-spun aluminum container for these components attached to the back of the accelerator using r . f . gaskets. (Field Emission Products) (6) Radiation Cel l s Two radiation c e l l s were used i n the pulse r a d i o l y s i s studies, a t h i n 2 mm diameter glass c a p i l l a r y containing a very small volume of l i q u i d and a 10 mm diameter stainless steel c e l l holding a much larger volume. Both c e l l s were designed to allow the maximum possible penetration of the electron beam into the solution and had f a c i l i t i e s to vary the path length of the subsequent interactions. Either could be placed into position or removed from the flow lines without much d i f f i c u l t y . The o r i g i n a l glass c e l l i n i t s plexiglass framework - and henceforth referred to as the p l e x i c e l l - had i n i t i a l l y yielded data whose further exploration required a radiation c e l l of extended v e r s a t i l i t y . The stainless s t e e l c e l l was then designed s p e c i f i c a l l y to investigate the temporal-spatial behaviour of e immediately a f t e r the electron pulse. Since the beam electrons seldom penetrated further than 2 mm into the water i t was possible to observe both very high dose and low dose regions i n which the k i n e t i c patterns were suspected to be quite d i f f e r e n t . This c e l l also had the capacity to act as a photolysis c e l l , but ultimately the photolysis experiments attempted i n th i s work -81-used instead the p l e x i c e l l modified from i t s o r i g i n a l form for t h i s purpose. Both c e l l s were almost f u l l y enclosed i n a "noise" s h i e l d , a machined aluminum cap that f i t t e d onto t h e i r supporting flanges tightened to the front of the Febetron. The c e l l s are shown i n exploded view i n Figure 4-3; photographs of them are above showing t h e i r r e l a t i v e s i z e s . Ca) P l e x i c e l l The p l e x i c e l l i s a h a l f - c i r c l e framework of polished 25 mm thick plexigl a s s out of which a block has been removed corresponding to the cross sectional area of the electron tube window. The t h i n walled pyrex i r r a d i a t i o n c e l l (100 mm long, 1.6 mm i . d . with walls ^ 0.15 mm thick) was inserted into t h i s block and held f i r m l y i n p o s i t i o n by a set of 0 rings and aluminum plates. At either end of the tube there was a recess i n the p l e x i g l a s s into which f i t t e d o p t i c a l windows, and a port protruding from the top. of the p l e x i c e l l through which the glass tube could be f i l l e d and emptied. The tubes were replaced a f t e r they became discoloured following excessive doses of r a d i a t i o n . The ports were BIO glass cones snugly f i t t e d into holes d r i l l e d into the c e l l , and sealed with a "plexiglass glue" ( f i l i n g s of p l e x i g l a s s dissolved i n dichloroethane and evaporated to a viscous l i q u i d made a very suitable water proof g l a s s - t o - p l a s t i c glue i n t h i s work). The flow lines f i t t e d d i r e c t l y onto these cones which were not covered by the aluminum cap. In chapter VI we discuss some of the problems of "shock wave" F i g u r e 4 - 3 ( a ) . Figure 4-3 (b). Stainless Steel C e l l and Components. (Photograph: c e l l , flange, A£ shielding cap) -83-interference from a semi-theoretical standpoint. Described below are the two anti-shock devices that were found necessary f o r d i f f e r e n t experiments. A block of aluminum 9.6 mm thick that just f i t t e d into the 4 x 4 cm cavity i n the p l e x i c e l l contained a 20 mm long x 1 mm wide s l i t that aligned with the'glass tube. A pair of adjustable 0.8 mm thick aluminum shutters s l i d along a groove i n the block and provided a way of defining the length of tube exposed to the electron beam. Since t h i s i s also that area i n which e w i l l be formed, t h i s length w i l l be the aq ' 6 o p t i c a l path length for k i n e t i c calculations. Needless to say the narrower the s l i t s the less well-defined the o p t i c a l path because of appreciable scattering effects. As the s l i t s were opened wide the ra d i c a l i n t e n s i t y d i s t r i b u t i o n of the electron beam i t s e l f added to thi s uncertainty,although scattering became r e l a t i v e l y less s i g n i f i c a n t at 5 mm for example compared to a 1 mm o p t i c a l path. The block also performed the essential purpose of reducing the shock-wave interference and delaying the appearance of the wavefront for about 250 nanoseconds after the pulse. The second anti-shock device was to f u l f i l the same require-ments but i n addition permit the i r r a d i a t i o n of s c i n t i l l a t o r s positioned behind the glass tube (see Figure 4-3). This time a bar i n the form of a l e t t e r 'H' was inserted into the cavity of the p l e x i c e l l . The H-bar was machined from aluminum, 9 mm thick and eollimated part of the electron beam through a s l i t 3 mm long by 0.7 mm wide i n the centre of the 'H'. Covering the. H-bar completely and i n contact with the c e l l flange was a 4 x 4 cm gr i d constructed from wide copper mesh, two strands having been removed to allow free entry into the s l i t for the electrons. It was estimated that the gr i d only removed about 10% of the t o t a l number -84-of electrons entering the cavity of the p l e x i c e l l to s t r i k e either the H-bar or the s c i n t i l l a t o r s behind the glass tube, while the i n t e r -ference signals were considerably reduced. (b) Stainless Steel C e l l A 10 mm diameter and 20 mm long c y l i n d r i c a l hole was machined from a block of stainless s t e e l (20 x 20 x 15) mm i n dimension. 12 mm quartz o p t i c a l windows f i t t e d into the recesses lined with t h i n neoprene gaskets at either end of the cavity and were held firmly i n place by an aluminum plate containing a s l i t across the centre, 10 mm long and 2 mm wide. These s l i t s defined the cross-sectional area to be scanned by the laser beam. A t h i r d 10 mm quartz window was sealed with A r a l d i t e into a hole i n the ste e l block immediately opposite the area of entry of the electron beam into the c e l l ; t h i s was incorporated for photolysis experiments and we had intended to focus the mega-watt output of a giant pulsed (14 nsec) ruby laser into the c e l l i n an attempt to photolyse e . These experiments have yet to be accomplished, postponement being aq primarily due to the technical problems of triggeri n g a 3 nsec electron pulse within the 14 nsec laser pulse when both exhibit j i t t e r of at least hundreds of nanoseconds. O r i g i n a l l y the electron beam passed through an area on the surface of the block milled to 3 ± 1 thousandths of an inch. The actual i r r a d i a t i o n area was defined by the posi t i o n of two thick stainless s t e e l f l a t s that moved across the entire front of the block on a calibrated gear mechanism. In the f i n a l stages of this work a section of t h i s side of the block was removed and replaced with a 1 thousandth of an inch thick stainless s t e e l f o i l , sealed to the main body and kept taut by the pressure of the ste e l shutters. -85-Th e radiation c e l l was f i l l e d and emptied on the same p r i n c i p l e as the p l e x i c e l l through two 1/4" stainless s t e e l tubes brazed on to the block and machined to f i t a BIO socket. For both easier handling and protection of the l i q u i d from the shock-waves discussed e a r l i e r , the block was fastened to a aluminum flange that r e s t r i c t e d the electron beam to 10 mm diameter before i t even reached the metal window of the c e l l , 20 mm away from the electron tube window. The aluminum shielding cap slipped over the c e l l and fastened d i r e c t l y onto the flange support. A micrometer gauge was attached to the outside of the cap and controlled the l a t e r a l movement of a 0.15 mm pinhole across the exit o p t i c a l window of the c e l l . By having the pinhole d r i l l e d at the point of a coned recess we were moderately successful i n obtaining a clean d i f f r a c t i o n pattern from which the outer rings could be e a s i l y eliminated. In t h i s way i t was possible to scan the gradient of i o n i s a t i o n and e x c i t a t i o n events across the c e l l with 0.05 mm reprod-ucible precision. ' (c) S c i n t i l l a t o r s The aluminum shielding cap for the p l e x i c e l l was modified to support either one of three s c i n t i l l a t o r s that were the l i g h t sources for the photolysis experiments. Each s c i n t i l l a t o r was commercially ground into a v i s u a l l y perfect hemisphere 3/2" or 1" i n diameter and polished u n t i l the surface was highly r e f l e c t i n g . We cut and polished a 2.5 mm wide shallow groove across the f l a t base which was to press against the glass i r r a d i a t i o n tube, and then deposited under vacuum a very thi n layer of aluminum over the whole hemisphere, c a r e f u l l y removing a 3 mm portion i n the centre of the groove with d i l u t e sodium hydroxide -86-solution on a small brush. The hemispheres were placed i n aluminum supports custom machined for each one and held i n by four c l i p s . This assembly s l i d into the back of the shielding cap over the p l e x i c e l l and could be removed without disturbing any of the alignment. The idea behind t h i s design was to create an intense f l a s h within the s c i n t i l l a t o r following the penetration of the electron beam i n a l l areas on the face of the hemisphere except the groove which was shielded by the glass i r r a d i a t i o n tube; the photons would have to r e f l e c t i n t e r n a l l y back and forth u n t i l they escaped through the 3 mm path i n the groove d i r e c t l y into the tube where a high concentration of e had been generated a few nanoseconds aq previously. To ensure as e f f i c i e n t a photon transfer as possible a few drops of water were squeezed onto the centre groove to make good contact with the glass through surface tension. The s c i n t i l l a t o r s were o r i g i n a l l y manufactured by Nuclear Enterprises Inc. ,type NE 111 with peak emission at 375 nm, NE 110 at 490 nm and NE 103 emitting close to the red at 575 nm maximum. Each emission band covered about 100 nm of the spectrum. It was anticipated that the fla s h from the s c i n t i l l a t o r would l a s t about 10 nsec following e x c i t a t i o n by the 3 nsec electron pulse. I f the G value for photoemission were ^ 1 then, for example, when hv = 4 eV (= 310 nm) there would be about (10 x 4)/100 joules of l i g h t emitted 2 within 10 nsec over an area of a few mm , with a spectral range of 100 nm. -87-Chapter V THE FLASH PHOTOLYSIS APPARATUS Experiments were designed i n which an aqueous solution would be subjected to two consecutive l i g h t flashes, and the chemical events studied through k i n e t i c laser photometry or conventional f l a s h spectroscopy. A double-flash photolysis equipment (99) was used to perform these experiments i n which the intense emission from each lamp lasted about 25 ysec and the i n t e r v a l between the flashes could be varied a few micro-seconds to tenths of a second. Following these experiments a new photolysis equipment of extended v e r s a t i l i t y was designed and constructed s p e c i f i c a l l y to pursue hydrated electron studies. In p a r t i c u l a r one of the lamps could be f i r e d a second time within the same delay range by means of a second charging c i r c u i t , thus making t h i s a t r i p l e - f l a s h photolysis unit. This apparatus w i l l be described below, together with the detection techniques u t i l i s e d throughout t h i s work. The Flash Photolysis Unit The new photolysis arrangement i s shown schematically i n Fig. 5-1 and amongst the rest of the equipment i n Fig. 5-2 (the general layout i s the same for both the double or t r i p l e - f l a s h u n i t s ) . The two flas h lamps and reaction vessel a l l rested i n a long, brass 5.4 cm diameter cylinder supported by 1/2" diameter rods that f i t t e d into conventional o p t i c a l bench stands. The upper h a l f of the 6.0 cm long cylinder was attached by hinges to the lower h a l f and when opened up revealed three separate compartments, the lamps and vessel l y i n g across . a l l three. The central and longest compartment (42 cms) was the fla s h -88-MANUAL TRIGGER OSCILLOSCOPE MAIN GENERATOR — i —>~j-1P28 O MAIN LAMP '-\\' spec lamp CD AUXILLARY 1 GENERATOR - AUX LAMP H« DELAY UNIT 1 ->1 AUXILLARY 2 GENERATOR DELAY UNIT 2 SPECTROFLASH 5C22 THYRATRON DELAY UNIT 3 Figure 5-1. Trigger System for Flash Photolysis Unit. (Sequence optional: i l l u s t r a t e d i s Main: A l : A2 : spectroflash) Figure 5-2. Showing the Double-Flash Photolysis Apparatus. -90-photolysis unit proper while the two much smaller sections at either end housed the extremities of the glass lamps containing the high voltage electrodes and the metal extensions to the reaction vessel. The whole brass cylinder could be adjusted l a t e r a l l y or v e r t i c a l l y by means of a fine thread mechanism functioning independently of the o p t i c a l bench stands. The f l a s h lamps lay either side of the vessel; i n the spaces between them f i t t e d an aluminium frame containing glass f i l t e r s of specified transmission. The frames were the length of the central compartment and could hold either 2 mm or two 1 mm thick f i l t e r s . Two long s l i t s , equivalent to the length and width of the frames, were cut out of the hinged top to the cylinder, thus enabling the frames to be inserted and removed at w i l l during an experiment without disturbing the lamps or reaction vessel. The f i l t e r s were custom-made to f i t the frame dimensions by Corning Glass corporation, and permitted selective transmission of wavelengths from 300 nm to 1200 nm. When they were i n place and the top of the cylinder closed the central compartment was l i g h t - t i g h t and the flash-emitted l i g h t reaching the reaction vessel had to pass through the f i l t e r s . The f l a s h lamps were conventional, 50 cm lengths of s p e c t r o s i l quartz tubing, 1.1 cm i n diameter and 1 mm thick, transmitting wavelengths down to ^  185 nm. Both ends of the lamp terminated i n a B 10 socket into which a machined brass cone had been sealed. Each lamp had one hollowed cone and one s o l i d cone f i t t i n g , into which tungsten pins were soldered to act as electrodes; they were f i l l e d and evacuated throughout the hollow cones, extensions of which were -91-sealed into a simple vacuum l i n e . This l i n e could pump the system down -3 and maintain a reasonable vacuum (10 torr) over several days i f desired. Spectroscopically pure Argon (Matheson) was used as the discharge gas i n the lamps, and the pressure nominally kept at 10 mm Hg. The actual pressure used for each experiment varied according to the desired breakdown voltage of the lamps, and to a certain extent the age of the lamps. When new lamps were f i r s t being used i t was necessary to discharge them several times and then evacuate them, re-pressurise and repeat the cycle, otherwise the inte r n a l pressure slowly increased due to "degassing" of the walls of the lamps and a b u i l d up of oxygen, water vapour and so on. The voltage cables to the lamps were shielded and grounded to the brass container. The c i r c u i t for the lamps and spectroflash i s in Fig. 5-1. Each lamp could be f i r e d separately or i n a predetermined sequence. They were charged i n d i v i d u a l l y up to a maximum of 10 kV. The main fl a s h i n our terminology i s that lamp f i r i n g f i r s t and causing the primary photoionisation to take place; the a u x i l i a r y lamp i s that lamp f i r i n g second and the spectroflash i s the monitoring f l a s h . The main flash could be f i r e d e l e c t r o n i c a l l y or manually and a pick-up voltage pulse fed into the a u x i l i a r y or spectroflash v i a a delay unit which acted as the trigger for the second lamp to f i r e . S i m i l a r l y another voltage pulse fed through a second delay unit could be made to tri g g e r the a u x i l i a r y lamp a second time and the spectroflash could be f i r e d 'during or after any of the three photolytic flashes. The delay units allowed a pre-selected delay time between the f i r i n g of the lamps from 2 microseconds to 50 milliseconds, or a l l the lamps could be f i r e d -92-simultaneously i f desired. The 5 ysec spectroflash was always f i r e d at 9 KV, the main and a u x i l i a r y lamps anywhere from 3 to 8 KV. Typically the photoflashes were ^ 20 ysec at h a l f - i n t e n s i t y and output was ^ 250 joules. (2) Reaction Vessel and Experimental Solutions *" The reaction vessel was also b u i l t from a s p e c t r o s i l quartz tube, 1.2 cm diameter with 1 mm thick walls. A polished 12 mm diameter quartz disc was c a r e f u l l y sealed to each end of the glass tube with A r a l d i t e . Metal tubular extensions that had already been chemically blackened were then glued c o a x i a l l y to these windows to form one 50 cm long vessel. Metal discs, containing pinholes of d i f f e r e n t sizes and i n various locations, could be held p a r a l l e l to the windows at either end of the reaction vessel with a screw-cap fitment. In t h i s way the events occurring i n d i f f e r e n t regions pf the c e l l could be s e l e c t i v e l y monitored. Solutions that were to be photolysed were forced into the vessel under a back pressure of 40 mm Hg through a side arm at the end of the s p e c t r o s i l c e l l and drained from the vessel the same way through another exit i n the three-way stop-cock. A second arm at 180° to the f i r s t and at the opposite end of the glass provided a port for the con-tinuous flow of solution or the escape of hydrogen flushing gas. The flow system was b u i l t e n t i r e l y from glass and a l l attachments were sealed from the atmosphere. The solutions were a l l prepared from fresh, t r i p l y d i s t i l l e d p re-irradiated water as described i n chapter IV. Reagents NaOH, Ba (OH) - 9 3 -and methanol were made up from Analar grade supply, KOH and glucose as a n a l y t i c a l grade. Hydrogen gas, (Matheson pre-purified) passed through a c a t a l y t i c deoxygenator to reduce the 0^ content to < lppm, was bubbled vigorously through a " f r i t t e d glass oval i n the experimental solutions for several hours p r i o r to t h e i r use. The hydrogen pressure was maintained at ^ 40 mm Hg above atmospheric pressure corresponding to -3 [^ 2]= 0.7 x 10 M i n solution. This was an important fact as regards the net chemical behaviour of the photolysed solutions. In the glucose and alcohol experiments the deoxygenating gas was nitrogen (Matheson) with an 0 2 content of < 5 ppm. (3) Optical and Detection Systems In the majority of experiments the analysing l i g h t was a 9 mW CW helium neon laser, optics Technology high output model 233. The laser, whose beam diameter was 1.25 mm with a divergence of 1.0 m i l l i r a d i a n s , was mounted on an aluminum table capable of l a t e r a l and v e r t i c a l precision adjustment. The l i n e a r l y polarised beam was directed along the length of the reaction vessel and passed through a series of narrow band pass and neutral density f i l t e r s before s t r i k i n g the photocathode of a well shielded 1P28 photomultiplier. While s i m i l a r i n other respects to that used i n the p.r. experiments, t h i s photomultiplier contained a 1 k ^ anode load r e s i s t o r primarily to increase the signal-to-noise r a t i o and exploit the advantages of a microsecond time scale compared to the nanosecond work. Variations i n the transmitted beam int e n s i t y a r i s i n g from the presence of one or more absorbing species were recorded on the 454 Oscilloscope as ' previously described. The actual coupling to the oscilloscope, AC or DC, -94-appeared to be unimportant as both gave experimental curves with the same gradient, but the AC coupling mode was preferred for s t a b i l i t y and also the capacity to cut out low frequency noise. However, i n some cases the decay of a very slow reacting species coincided with the AC coupling time constant, ^  15 msecs, disallowing the use of some sweeprates on the oscilloscope. Fortunately t h i s was more inconvenient than l i m i t i n g and i n such cases the curves could be followed i n the D.C. mode. , When the spectrum of a transient was to be taken a spectroflash lamp substituted for the laser. This v e r t i c a l lamp, housed i n a brass cylinder and pressurised with 30 mm argon (spectroscopic grade), was discharged at 9 KV and emitted a 5 microsecond flash of v i s i b l e l i g h t . The external structure of the lamp was such that the fla s h simulated 'point source emission which could be converted into an almost p a r a l l e l beam by the suitable placing of a lens. A small 1 cm diameter polished convex lens was mounted 4.8 cm - i t s focal length - from the lamp and directed the resultant p a r a l l e l beam along the desired region of the reaction vessel. Emerging from the other end of the photolysis c e l l the beam was either r e f l e c t e d from a front s i l v e r e d mirror angled to send the second r e f l e c t i o n , s t i l l 1 cm i n diameter, onto the s l i t s of a low resolution spectrograph, or focussed immediately onto the s l i t s of a high resolution spectrograph. Occasionally i t was found expedient to insert another lens into the o p t i c a l axis to defocus the beam to a smaller diameter than 1 cm on the s l i t s . -95-The spectra were recorded on a variety of f i l m and plates while a l l the time-dependent decay curves were obtained by photographing the oscilloscope screen on polaroid f i l m as outlined i n Chapter IV. Some spectra i n the region 600 nm to 900 nm were taken on Kodak Infra Red 35 mm f i l m , both slow and fast which necessitated t o t a l room darkness. The plates used i n the spectrograph were chosen for dif f e r e n t levels of s e n s i t i v i t y i n specified regions of the spectrum, namely I l f o r d HP3 plates for v i s i b l e up to 650 nm, Kodak. Spectroscopic plates type 1-N for regions above 600 nm and up to 900 nm and type l-'Z for regions up to 1100 nm. In order to prevent either the plates or f i l m from being exposed to second or higher order r e f l e c t i o n s when normally f i r s t order re-fl e c t i o n s were required, or from harmonics i n the laser beam and scattered l i g h t or induced fluorescence i n f i l t e r s and lenses, a defensive system of s e l e c t i v e l y transmitting f i l t e r s was devised for each experimental assignment. These w i l l be described i n the text where appropriate. The characteristics of the f i l t e r s , and combinations of f i l t e r s , were a l l recorded on the Cary 14 to ensure they f u l f i l l e d the requirements for which they had been.chosen. Three spectrographs were used during the course of th i s work. For high resolution spectra the Jar r e l l A s h (3.4 m) recorded a band of 125 nm on a single plate giving a resolution of 0.5 nm per mm; two plates could be used simultaneously for a t o t a l spectral width of 250 nm. Since the hydrated electron has an absorption band extending over the whole v i s i b l e region, 250 nm i s not always adequate t o t a l coverage and the density i s somewhat diffuse i f the s l i t s are too narrow. A - 9 6 -medium Hilger solved t h i s problem and band spectra could be readily recorded. One l i m i t i n g factor here was that the c r i t i c a l part of the spectrum approaching A max (680 to 720 nm) for e was confined to aq about 4 mm or so. The Bass-Kessler proved to be an ideal portable grating spectrograph with a resolution of 60 nm per mm for f i r s t order r e f l e c t i o n s . In contrast to the Hilger the dispersion i s l i n e a r , and the spectral wavelengths to ;.be analysed having entered the B-K through a telescopic lens are ref l e c t e d into the focussing lens of a Konica model FP camera loaded with 35 mm recording f i l m . The near i n f r a red work was carried out with t h i s p a r t i c u l a r arrangement. Further d e t a i l s of t h i s small spectrograph can be obtained i n reference,(100). The plates and f i l m were a l l developed according to manufacturers instructions with allowance for the ambient temperature of the solutions to be used. Standardisation of developing techniques was extremely important for the subsequent scanning and comparison of the plate of f i l m density with a dual beam Joyce Loebel Mark 2 microdensitometer, capable of X50 magnification and excellent r e p r o d u c i b i l i t y . Wavelength calibrations were automatically included i n the plates because there are several strong s i l i c o n lines emitted i n the argon spectroflash. This in t e r n a l c a l i b r a t i o n became weak i n the near i n f r a red region so the emission from a neon lamp was superimposed on the already exposed plate of f i l m p r i o r to i t s removal from the spectrograph. The r e l a t i v e spectral s e n s i t i v i t i e s of the f i l m or plate were recorded using fixed s l i t s and neutral density f i l t e r s or through variable s l i t s for each set of experiments and developing conditions. -97-Non-linear exposure - density relationships were studied with the use of a calibrated quartz-iodine lamp, and the camera shutter speeds measured independently by incorporating i t into the laser-photomultiplier - oscilloscope arrangement. (* Note added i n proof. The experiments reported i n t h i s thesis were performed on the double-flash photolysis appartus developed by Basco (99(a)) from e a r l i e r designs of single f l a s h units (99(b) and ( c ) ) . Use of the t r i p l e - f l a s h photolysis apparatus was delayed as a resul t of technical problems associated with the f i r i n g of the a u x i l i a r y twice. The electronic design of t h i s apparatus arose from experiences with the double-flash c i r c u i t e r y , and the only mechanical modifications to the photolysis unit were i n the b u i l t - i n f i l t e r frames, the i n t e r n a l mounting of the pinholes (previously hand mounted) and the use of hollow electrodes. References 99(a) N. Basco, unpublished. S. Dogra, Ph.D. thesis (1970) University of B r i t i s h Columbia. 99(b) R.G.W. Norrish, G. Porter, B.A. Thrush, Proc. Roy. Soc. A216, 165 (1953). 99(c) N. Basco and R.G.W. Norrish, Proc. Roy. Soc. A260, 293 (1961).) -98-Chapter VI INTERFERENCE PHENOMENA The decay of the electron-pulse-induced absorption i n pure water (attributed to e a q ) i s interrupted by what appears to be a second very strong "absorption" after a few hundred nanoseconds. It transpired that t h i s signal could not be assigned to any other absorbing species present i n the system but instead was a manifestation of a shock phenomena observed whenever liq u i d s were i r r a d i a t e d under these conditions. Although these fluctuations i n i t i a l l y r e s u l t i n the loss of 100% of the laser beam in t e n s i t y the effect eventually disappears, but the i r r e g u l a r l y shaped signal i s quite reproducible under each experimental arrangement. A systematic study was made of t h i s phenomenon, somewhat loosely termed shock waves since t h i s best described t h e i r i r r e g u l a r undulating form, and the results w i l l hopefully f i n d some relevance i n p.r. techniques and the future design of radiation c e l l s . The same effects have been subsequently observed i n t h i s laboratory i n a d i f f e r e n t c e l l , i n other radiation research groups (101), and i n a l l instances enforce a time l i m i t on any k i n e t i c studies of short l i v e d transients. There are several possible origins for these shock waves, and each was investigated independently (not always feasible i n a system where most of the causes are simultaneously present); the causes were thought to be (1) chemical (2) electronic (3) o p t i c a l (4) shock (pressure and/or acoustical) (5) electromagnetic f i e l d e f f e c t s. -99-(1) Chemical The p o s s i b i l i t y that a secondary absorbing species was formed as a res u l t of the reactions of e was ruled out by the fact these shock aq 1 waves were generated i n a highly a c i d i c solution - where no e aq survived longer than 10 nanoseconds - and t h e i r amplitude and duration were exactly the same as observed i n pure water. There were no other species present that could absorb l i g h t of 632.8 nm other than CO^ which would be there i n very small quantities as an impurity. In the presence of IM ethanol which e f f i c i e n t l y scavenged the OH r a d i c a l s , precursors to CO^ through reaction with CO^ , shock waves were unaffected. One i n t e r e s t i n g proposal sees t h i s phenomenon as an example of microactivation, of pocket explosions i n the l i q u i d due to the sudden increase i n l o c a l volume when the electrons become solvated, (102). Doubtless there i s an increase i n volume at t h i s point but the fact that t h i s interference has been observed by us i n the known absence of e i n non-aprotic solvents and the long time scale required for the interference to appear both argue against microactivation as a primary cause. F i n a l l y there i s always the p o s s i b i l i t y that the i n t e n s i t y of laser beam i s s u f f i c i e n t to photolyse a s i g n i f i c a n t f r a c t i o n of e and that the second absorption represents the return of this f r a c t i o n to t h e i r o r i g i n a l status. Again both the time scale of the events and the exper-iments i n a c i d i c media rule out t h i s l i n e of approach; furthermore, a simple c a l c u l a t i o n shows that for every available photon there are more -100-than 10 e present - not a very favourable r a t i o , aq (2) Electronic The remarkable r e p r o d u c i b i l i t y of these signals did not seem compatible with an electronic o r i g i n , nevertheless the following effects were held plausible: ringing or space-charges within the envelope of the photomulitplier, r e f l e c t i o n s i n the grounding system and mismatch of transmission lines or other impedance d i f f i c u l t i e s . The photomultiplier, whose l i n e a r i t y had already been checked under th-e range of operating conditions, was shown to be free of saturation, overshoot and fatigue effects. The waves neither reduced nor increased i n t h e i r r e l a t i v e i n t e n s i t y to the i n i t i a l absorption of e~ over a 1 v aq considerable v a r i a t i o n i n operating voltage, nor were they affected by any grounding or shielding of parts of the photomultiplier assembly. When the deliberate incorporation of external ground loops, mis-matched cables and d i f f e r e n t anode load r e s i s t o r s altered neither the shape nor frequency of the waves, i t seemed very doubtful that they were of electronic o r i g i n . In the absence of the radiation c e l l but otherwise i d e n t i c a l conditions the photomultiplier was triggered only with d i f f i c u l t y on the most sensitive setting of the oscilloscope, i n d i c a t i n g that the phenomenon arose from the events i n the radiation c e l l . (3) Optical In t h i s c l a s s i f i c a t i o n are included the malfunctioning of the laser i t s e l f , misalignment of the lenses, f i l t e r s and pinholes along the o p t i c a l axis and d i f f e r e n t i a l absorption of l i g h t by the species i n solution, including laser photolysis of e -101-I f the laser were switched o f f but remained i n the mains c i r c u i t , no interference was observed upon i r r a d i a t i n g the solution. In other words the laser beam was monitoring the shock waves, not contributing to them. Laser photolysis has already been discarded as a l i k e l y cause. The lenses, f i l t e r s and pinholes were c o l l e c t i v e l y exonerated through a series of experiments i n which the optics were either misaligned with respect to the o p t i c a l axis or eliminated from the arrangements completely. In no way could a set of equivalent interferences signals be reproduced or the genuine signals removed. It was observed that the presence of a l i q u i d was necessary i n order to propogate these waves i n a way v i s i b l e to the detection equipment. The a i r i n an empty c e l l apparently displayed no such behaviour and when a t h i n glass rod with polished ends was substituted for the l i q u i d i n the plexiglass c e l l no interference effects were discernible before, or superimposed upon, the permanent change i n transmittance of the rod due to the presence of trapped electrons. The waves therefore were a function of the l i q u i d but the enormous fluctuations observed could not sensibly r e s u l t from scattering of the laser beam by dust p a r t i c l e s or, for example, microsplinters of glass or metal thrust into solution by the impact of the electron beam on the t h i n windows i n the i r r a d i a t i o n c e l l s . The s i t u a t i o n could not be improved by replenishing solutions and constantly changing the windows. It was therefore most probable that waves i n part represented the delayed effects of sudden l o c a l i s e d changes i n the -102-r e f r a c t i v e index of the solutions along the path length of the beam as a r e s u l t of a v i r t u a l l y instantaneous high concentration of absorbing - -3 species, v i z e at ^ 10 M. In addition a concentration gradient existed across the c e l l following the depth-dose ch a r a c t e r i s t i c s and r a d i a l d i s t r i b u t i o n features of the electron beam. The l a t t e r , although detectable along the o p t i c a l path length I, at the most caused a s l i g h t deviation from the l i n e a r i t y of the OD = zCl relationship and was n e g l i g i b l e below I = 5 mm. The i n i t i a l non-uniformity i n the system appeared to be smoothed out by the time the interference arrived and moved across the laser beam, thus any time-dependent v a r i a t i o n i n r e f r a c t i v e index and so on should be over. Therefore, although t h i s explanation did not appear to be f u l l y compatible with the period and i n t e n s i t y of the shock waves observed, i t did lead to the idea of a wave front; that i s , a r e a l shock wave, moving out r a d i a l l y from the area of i n i t i a l beam impact on the water molecules and seen by the laser beam as an opaque front whose time and i n t e n s i t y dependence appeared as a function of the laser scanning p o s i t i o n . (4) Shock Phenomena The physical r e s u l t of impact of t h i s electron beam on the glass radiation c e l l has been d i r e c t l y observed (103) while a comparable beam of higher energy electrons (1.8 MeV) and 30 nanoseconds i n duration i s reported to shatter any glass c e l l s thus r e s t r i c t i n g designs to a l l s tainless s t e e l (101). It would not be surprising therefore i f a pressure wave were to r e f l e c t back and forth across the ra d i a t i o n c e l l , . - 1 0 3 -propagated through the material of the c e l l and the l i q u i d u n t i l the impact energy was dissipated. There are b a s i c a l l y two sources for t h i s shock wave; f i r s t , that generated by the beam electrons t r a v e l l i n g through the a i r space between the electron tube face and the radiation c e l l window, and secondly that a r i s i n g from the impact of the electron beam on the c e l l windows. Such impact energy could then be transferred to the l i q u i d which transmits the shock as a physical compression wave. A t h i r d p o s s i b i l i t y , ultimately indistinguishable from the previous two, i s an acoustical wave. Whenever there i s a non-uniform absorption of energy i n a contained medium then a wave showing acoustical properties can be propagated along the d i r e c t i o n of the energy gradient. In t h i s case the energy could be transferred to the medium as a sudden increase i n heat content, whereby a temperature gradient results across the c e l l , or the energy may be u t i l i s e d to create large number densities of species causing a gradient i n the value of the r e f r a c t i v e index of the solution ( or a concentration gradient) across the c e l l . Such non-uniformity observed i n the l i q u i d could be shown i n the general case to be of much greater magnitude than non-uniformity i n a gas, since the pattern of absorption of the i n i t i a l electron energy would be quite d i f f e r e n t . This would explain why the waves could only be observed after pulse i r r a d i a t i n g a l i q u i d . A series of experiments designed to measure the v e l o c i t y of the -104-the wave front i n l i q u i d s , the int e n s i t y of sound as the spark gaps in the Febetron were f i r e d , the rate at which the wave dissipated across the c e l l and the effects of b a f f l i n g the radiation c e l l both i n t e r n a l l y and externally produced the following information. The v e l o c i t y of the wave front was measured to be (1.52 ± . 09).xl0"' cm sec * compared to 1.49 x 10^ cm sec * the l i t e r a t u r e value for the vel o c i t y of sound i n water at 25°C (41). The design of the stainless s t e e l c e l l permitted t h i s v e l o c i t y to be determined by scanning a cross section of the radiation c e l l , into regions where no e~ would be ' 6 aq formed but a propagated wave front might reach. The 0.15 mm diameter pinhole could be moved l a t e r a l l y with 0.01 mm precision across the transmitting window of the c e l l . The measurements were taken over a distance of 2 mm with delays of up to 2 psec. A wave "picture" thus obtained i s shown as an oscilloscope trace i n figure 6-1, where (a) i s a t y p i c a l delayed wave form, and (b) i l l u s t r a t e s the r e p r o d u c i b i l i t y of the interference phenomenon i n pure water and acid solution, described i n section (1). No refl e c t e d Waves could be detected, but since the int e n s i t y of the signal had decreased by 50% at a distance of ^  5 mm from the c e l l window, i t would be very weak by the time i t reached Figure 6-1. Typical Interference Signals. 5 mV/div / r . - l t \ / — t 100 nsec/div electron pulse (a) 5 mV/div \ • f t / 1 / 200 nsec/div upper trace, H^ O lower trace, H^SO^ (b) -105-the f a r wall of the c e l l . The larger the pinhole the more diffuse the wave appeared. B a f f l i n g the a i r space between the c e l l and the accelerator with aluminum rings soldered concentrically onto the support flange of the c e l l did not have any appreciable effect on the interference signals. When a copper g r i d b a f f l e was slipped inside the c e l l the effect was negative. The delay of ^ 400 nanoseconds p r i o r to the appearance of the shock Wave i n the zero pinhole p o s i t i o n immediately behind the c e l l window represents a distance of 0.6 mm, comparable to the estimated distance between the i n i t i a l point of impact i n the c e l l and the pinhole, and the 0.2 mm uncertainty i n the zero p o s i t i o n of the l a t e r a l c a l i b r a t i o n . Unfortunately experiments intended to reproduce a shock wave t r a v e l l i n g i n the opposite d i r e c t i o n across the c e l l (towards the c e l l window) and thus measure the wave cha r a c t e r i s t i c s i n the absence of the electron beam were unsuccessful. I f indeed mechanical shock as devised f a i l e d to produce detectable shock waves then either the waves originate at_ the window by virtue of an energy gradient discussed e a r l i e r or the momentum behind the impact t r a v e l l i n g through the whole c e l l i s extraordinarily high. (5) F i e l d effects Here we could speculate on the effect of the intense e l e c t r i c and magnetic f i e l d s generated during the electron pulse on the r e f r a c t i v e indices of the materials, the p o l a r i s a b i l i t y of the materials, the polarised laser beam and the charged absorbing species, e aq However, the duration of these strong f i e l d s i s only 3 nanosecond's and -106-unless some long term secondary effect induced by the f i e l d a l t e r s the properties of the system to give t h i s interference, such consider-ations are u n r e a l i s t i c . Within the time of the pulse i t would be feasible to imagine the e rotated with a s p i r a l l i n g motion along the di r e c t i o n of the magnetic f i e l d (that i s along the o p t i c a l axis) perpendicular to the dir e c t i o n of momentum of the electron beam penetrating the solution. As w e l l , i t i s known that pure l i q u i d s i n a transverse magnetic f i e l d exhibit a strong double r e f r a c t i o n as a resu l t of the orientation of the magnetically and o p t i c a l l y anisotropic molecules i n the d i r e c t i o n of the f i e l d , regardless of the permanent or temporary nature of the magnetic dipole moments. This, the Cotton-Mouton e f f e c t , i s the magnetic analogue of the Kerr Electro-optic e f f e c t , and i s proportional to the square of the f i e l d strength. This l a t t e r phenomenon i s observed when materials are placed i n a strong e l e c t r i c f i e l d ; they too show double r e f r a c t i o n due to the l i n i n g up of the molecules i n the f i e l d d i r e c t i o n . Both these e l e c t r i c and magnetic f i e l d effects are temperature dependent, (104). To assess the contribution of such induced or natural anisotropy i n the system i s presently impossible. That there was no long term p o l a r i s a t i o n effect could be shown with the selective use of p o l a r i s i n g f i l t e r s on the incident and transmitted laser beam. When a copper mesh grid was inserted between the radiation c e l l and the electron tube window (the mesh cut away from the window area on the c e l l ) the amplitude of the shock wave was considerably reduced. Grounded to both c e l l and accelerator, the grid must have acted as a "net" for the electrons, thus reducing the f i e l d i n t e n s i t y , but at the same time reducing the net impact of the beam on the c e l l by 10 to 15%. -107-A further p o s s i b i l i t y of looped r i n g currents on the surface of the very t h i n stainless s t e e l window receiving such a large current could not be discounted, but was considered inevitable with the specified c e l l design and hopefully small i n magnitude. The potential space charge set up within the solution i t s e l f by the "instantaneous" i r r a d i a t i o n was studied by using high concentrations of Na2S0^ i n solution (0.2 M),but the presence of an active charge c a r r i e r made no apparent difference to the frequency or amplitude of the waves. Summary This interference phenomenon, which was observed i n the form of reproducible i r r e g u l a r l y shaped "absorption" signals, possessed char a c t e r i s t i c s that inferred i t was acoustical i n o r i g i n . Neverthe-less other contributions to the amplitude of the signals were not completely eliminated, and some of the possible causes discussed would have given r i s e to eventually the same acoustical pattern of behaviour. The c e l l designs were modified several times to a t t a i n optimum conditions for k i n e t i c analyses of the hydrated electron weighing the concentration of e desired against the i n t e r v a l before the f i r s t shock wave appeared. The various b a f f l e s and beam collimators u t i l i s e d have been described i n Chapter IV. -108-Chapter VII PRELIMINARY RESULTS AND OBSERVATIONS ON THE PULSE RADIOLYSIS EXPERIMENTS A study of the bimolecular reactions of e with e" , H_0+ and OH ' aq aq 3 produced i n approximately the same numbers during the pulse and at -4 r e l a t i v e l y high concentrations O 10 M) was intended to establish a firm k i n e t i c basis for s i m i l a r experiments i n which e would be r aq simultaneously photo-excited into i t s unspecified upper state. Preliminary results indicated that the i n i t i a l rate of disappearance of e s i g n i f i c a n t l y altered after about one h a l f - l i f e (y 110 nsec). aq Following a t r a n s i t i o n period l a s t i n g about 10 to 20 nsec the system appeared to obey a second-order k i n e t i c law. This unusual behaviour was t e n t a t i v e l y attributed to an i n i t i a l non-homogeneity i n the d i s t r i b u t i o n of reacting species i n view of the high dose rates employed, and further experiments were planned to assess the v a l i d i t y of t h i s hypothesis. (1) Non-classical k i n e t i c s - Preliminary Data The t r a n s i t i o n from one k i n e t i c behaviour to another had been observed while monitoring the f u l l laser beam as i t emerged from the radiation c e l l . In order to investigate t h i s effect i n regions of 2 dif f e r e n t electron penetration i n the c e l l , a pinhole 0.07 mm i n cross-section was used to scan across the transmitted beam so only a t h i n pencil of l i g h t reached the photomultiplier. Neutral 2D water was i r r a d i a t e d i n the p l e x i c e l l with pathlengths ranging from 1 mm to 5 mm. It was shown that over t h i s range O.D. was approximately proportional to the pathlength i n accordance with Beer's Law. The formation and -109-decay of the absorbing species was recorded as a time-dependent voltage s i g n a l superimposed on the c a l i b r a t e d g r a t i c u l e of the C.R.T. by photographing the o s c i l l o s c o p e trace on p o l a r o i d f i l m . Measurements of the height of the s i g n a l at a given time were made on these photographs using a Microscale divided i n t o 0.1mm d i v i s i o n s . A reasonable estimate could be put on the second decimal place. T y p i c a l l y , an absorption s i g n a l would be 14.80 ± .05 mm corresponding to nearly two-thirds displacement of the trace across the useful C.R.T. screen when the zero percent transmission was equivalent to 20.00 mm. These readings were converted i n t o o p t i c a l density data and other time-dependent functions of O.d. necessary f o r k i n e t i c analysis by * means of a computer programme. Attempts were made to f i t the data to one of the following i n view of the mechanisms believed to take place i n a pure water system. (a) One s i g n i f i c a n t process occurring a f t e r the pulse. (b) Two s i g n i f i c a n t simultaneous processes occurring a f t e r the pulse. (c) Two s i g n i f i c a n t consecutive processes occurring a f t e r the pulse. From the early data i t was evident that f i r s t - or second - order processes or both might be involved. When more than one r e a c t i o n i s responsible f o r the t o t a l loss of e & ^ i n the system as i n case (b) the following equation may be rearranged to give a graphical separation of the rate constants. (In the text k^ represents any f i r s t , and k^ any second - order rate constant.) * The author g r a t e f u l l y acknowledges the assistance of S.C. Wallace i n compiling the programme. -110-d (e~ ) = k [e~ ] + k'fl [e" ] 2 ( or k Q [e" ][x]) aq' a L aq J 3 aq ^ g *• aq J L i J dt 1 d [e" ] = k + k Q [e" ] L aq a 3 aq [e~ ] dt aq J d In [e~ ] = k + k„ [e" ] aq a g L aq J dt A plot of either -1 . d[e ] or d(£n[e~ 1) versus [e" ] should r aq u aq J L aq [ ea q J dt dt give a straight l i n e of slope -k^ and intercept k^. In the event that only one process i s s i g n i f i c a n t the appropriate k w i l l equate to 0. Conventional f i r s t and second - order relationships were used for (a) and (c). On the assumption that o p t i c a l density was l i n e a r l y related to concentration of the absorbing species O.D. was substituted f o r t e~q] i n the calculations and the slopes l a t e r converted to k and k. . The r a g least-squares-fit programme f i r s t treated the data f o r case (a) and (b) and then systematically eliminated data points up to a s p e c i f i e d time after the electron pulse recording variations i n the slope to provide information for case (c). The absorption signal was analysed over the f i r s t 200 to 350 nsec at which time the a r r i v a l of the "shock-wave" disrupted the signal and further measurements were impossible. Nevertheless t h i s period corresponded to two or three h a l f - l i v e s of the decay. Three pin-hole positions A, B, and C were selected to follow the decay of e , where A was near the 1 aq' front face of the c e l l and thus saw the highest [e ]. B was a central 5 L aq J' i position a few tenths of a millimetre away and C a s i m i l a r distance from -111-B where the lowest [e - ] would be detected. The mean L.E.T. i n the aq region C i s s i g n i f i c a n t l y higher than i n A. The reaction zones that 2 were scanned through the pin-hole were 0.07 mm i n cross section and of varying path length, although the results reported below apply to a 5 mm path. Gross fluctuations i n the average concentration of e within this t h i n pencil ( i . e . due to non uniformity across or down the c e l l ) would not be s i g n i f i c a n t (105). The loss of e i n a l l three regions could be best described by case (c), where the decay moved from our i n i t i a l f i r s t - order f i t through a b r i e f t r a n s i t i o n period into second - order behaviour; the t r a n s i t i o n period commenced at different times and for di f f e r e n t percentage losses of e . The changes i n slope of either f i r s t or second - order plots were quite d i s t i n c t , and occurred at 80, 100 and 150 nsec for positions A, B and C respectively. Figure 7-1 i l l u s t r a t e s these features which were characterised for a considerable number of experiments using d i f f e r e n t path lengths, and D20 (99.7 % pure) as wel l . The data was plotted i n terms of 0D .rather than concentration i n view of the problem under examination. Despite the concentration gradient across the i r r a d i a t e d volume of l i q u i d the i n i t i a l rate constant for the e decay p r i o r to the t r a n s i t i o n period was s i m i l a r for the positions A, B and C i n keeping with the f i r s t - o r d e r data f i t . The t o t a l % loss of e incurred by t h i s time was aq J however dependent on the length of i n t e r v a l before the t r a n s i t i o n and increased as the delay became longer. A set of resu l t s calculated from the data i n Figure 7-1 i s -112-0 40 80 120 Time (nsec) . 200 Figure 7-1. F i r s t and Second Order Treatments of Data i n Table 7-1. -113-given i n the table below. Table 7-1 Transition Loss of (M) (nsec) (sec" 1) (M^sec- 1) (%) Pinhole [e~ ] time, t k k. e~ at t. L aq J ' a S aq A 1 x 10" 4 80 6.8 x 10 6 7.2 x 1 0 1 0 43 B (7 x 10" 5) 100 7.4 x 10 6 (1.1 x 10 1 1) 50 C (5 x 10" 5) 150 5.2 x 10 6 (1.6 x 10 1 1) 55 Some uncertainty i s attached to the c a l c u l a t i o n of [e~ ] and therefore kg i n B and C because of the uncertainty i n the value of the path length at a distance from the immediate front face of the c e l l where i t had been set at 5 mm. In the l i g h t of these results i t was concluded that the system exhibited a d e f i n i t e change i n k i n e t i c behaviour after a time somehow related to the i n i t i a l numbers of the reacting species e~^ following the electron pulse. The observed t r a n s i t i o n point was interpreted as indicating that the system had attained a uniform d i s t r i b u t i o n of these species, and "concentration" as a bulk parameter was now a meaningful term. Prior to t h i s time the reacting species were non-randomly d i s t r i b u t e d and any c l a s s i c a l k i n e t i c analysis of the reactions taking place would be i n v a l i d . -114-Th e t r a n s i t i o n times that were recorded i n positions A, B and C were amenable to theoretical confirmation i n at least a q u a l i t a t i v e sense. The reference state for the calculation i s the i n i t i a l d i s t r i b u t i o n of energy i n the form of average 100 eV spurs throughout the i r r a d i a t e d volume. About 60% of the t o t a l energy of a primary incident p a r t i c l e i s deposited i n the system i n the form of spurs, the remainder i n short tracks, branch tracks and blobs (see chapter 1). We neglect track e f f e c t s , which ultimately must be considered i n order to establish true microscopic homogeneity, and calculate the time taken for spur-overlap to occur through d i f f u s i o n i n the regions A, B and C. 13 -1 For an instantaneous radiation dose R rad (6.2 x 10 eV g ) there w i l l be R x 6.2 x 10 1 1 spurs per gram. In a l i q u i d of density P 1 spur 11 3 there w i l l be Qn average Aper (p /R.x 6.2 x 10 ) cm . I f the spur i s considered to have spherical geometry i t w i l l occupy a volume of 3 4irr 0 /3 where r Q i s the root-mean-square radius of the spur. As th i s volume expands by virtue of d i f f u s i o n , the spurs w i l l overlap when a distance of [(p/nR x 6.2 x l O 1 1 ) 1 ^ - r 0 ] has been covered. By regarding the t o t a l volume of the l i q u i d as close-packed touching spheres about 1/3 of the re a l volume has been neglected, therefore the volume occupied by the d i f f u s i n g spur has been mu l t i p l i e d by a factor of 4/3 as a compensating, i f approximate, move. The mean-square distance t r a v e l l e d i n any of s i x co-ordinates of space i n time T i s D/r , where D i s the d i f f u s i o n constant for the species involved. The minimum time x^ taken to achieve spur-overlap i s therefore x H = [( P/TTR x 6.2 x 10 1 1) 1 / 3 - r D ] 2 D"1. ( i ) -115-When the appropriate values for e i n water D = 4.7 x 10" 5 cm2 sec" 1 (32), r Q = 30 A and R = 6.8 x 10 4, 5.7 x 10 4 and 3.9 x 10 4 for A, B and C respectively are substituted i n equation (i) values can be determined for these experiments. They were x^  = 60 nsec for A compared to 80 nsec observed, 80 nsec for B compared to 100 nsec observed and 100 nsec for C compared to 150 nsec observed. In calculating R from the i n i t i a l O.D. measurements, G(e ) = 2.5 was assumed which i s a conservative estimate i n the l i g h t of aq s recent scavenger studies (106); a higher value would increase T^. The calculated and observed t r a n s i t i o n times compare remarkably well considering the approximations that were made and support the i n t e r -pretation of the change i n slope and k i n e t i c order of the e a decay as indi c a t i v e of nonhomogeneity i n the system. Figure 7-2 i s a graphical representation of the time taken to achieve homogeneity following an instantaneous dose, which we have calculated f o r the hydrated electron from equation ( i ) , (107). A s i m i l a r curve can be determined for other species that are present i n a pulse radiolysed system. The immediate significance of th i s interpretation i s that i n studying reactions that occur i n tens of nanoseconds we cannot j u s t i f i a b l y convert o p t i c a l densities nor G-value and dose data into concentrations on which bimolecular rate constants depend. The same substantive conclusion may be reached concerning medium or low dose-rate studies by re f e r r i n g to Fig. 7-2, where for example the system takes ^ 50 ysec to achieve homogeneity following a 10 rad pulse. Figure 7-2 Showing the time required to establish homogeneity at a given "instantaneous" dose for the hydrated electron fsee textT. -117-(2) Nonhomogeneity and Rate Constants The non-classical k i n e t i c behaviour of e i n the pulse aq r radiolysed system has been attributed to an i n i t i a l non-homogeneity i n the d i s t r i b u t i o n of the reacting species. The duration of t h i s period depends on the dose-rate. What exactly does th i s non-homogeneity mean? We w i l l diverge b r i e f l y from further experimental results to answer t h i s question. The primary yields of e , HjO + a n d ^ H i - n t n e pure water system are approximately the same and therefore the i n i t i a l rate of disappearance of e through reaction with i t s e l f , H^0+ or OH should obey a second order r e l a t i o n s h i p , v i z - d [e" ] = k Q [e" ] 2 or k 0 [ C ] 2  L_aq J B L aq J 3 dt where kg i s the sum of (k^ + k^ + k,.) f o r the appropriate reactions. The v a r i a t i o n of OD with time (eA/OD v.t) should y i e l d a l i n e a r plot whose slope i s k^. This l i n e a r i t y w i l l not be evident i f one of the basic assumptions i n the rate equation i n incorrect. The concentration C i s a macroscopic parameter and refers to the number of e per bulk volume. I f the d i s t r i b u t i o n of e" i s non-aq r aq random, then C i s no longer a meaningful parameter. Within the volume there w i l l be l o c a l regions of widely varying concentration where the number density of reacting species may bear l i t t l e resemblance to the experimental value (E&/OD). Consequently the rate constant calculated with t h i s averaged "C" value w i l l not be the true k^. From the experimental plots i t w i l l appear as though the rate constant i s -118-time-dependent whereas i n a c t u a l i t y i t i s the lo c a l concentration which i s changing by virtue of d i f f u s i o n . A rate constant cannot be time-dependent i n the l i t e r a l sense. U n t i l the system achieves a t r u l y random d i s t r i b u t i o n of reacting species and the slope of the el/OD vs t plot i s independent of time we cannot r e a l l y discuss second - order rate constants. Yet there i s s t i l l a need to describe the mechanisms taking place i n terms of some parameter,particularly i n th i s work where the i n i t i a l period of non-homogeneity dominates the time available for k i n e t i c analysis. We s h a l l thus define and use K as the rate of disappear-ance of e within a non-homogeneous i n t e r v a l , where K i s K = d(e£) OD dt When the system has achieved homogeneity, K becomes a true second -order rate constant. I f the l o c a l concentrations of species are very high then the average (or observed) concentration w i l l be lower and the K value higher for a given rate of loss of electrons. As the species diffuse and c (the observed concentration calculated from OD data) approaches c (the r e a l concentration i n solution) K should become smaller u n t i l eventually i t approximates to k^ . In other words K w i l l decrease with time u n t i l i t reaches k Q . P When species capable of p a r t i c i p a t i n g i n both f i r s t and second -order reactions are distri b u t e d homogeneously i n a system i t i s not unusual for the system to move from second - order to f i r s t - order -119-behaviour as the concentration of reactant i s depleted. The observation that our system under went a sequential conversion from f i r s t to second -order c h a r a c t e r i s t i c s , contrary to the c l a s s i c a l sense, could be s a t i s -f a c t o r i l y interpreted i n terms of non-homogeneity. But the further observation (see Figure 7-1) that K increased with time was not f u l l y consistent with t h i s view of non-homogeneity unless the observed con-centration of species was higher than the r e a l concentration. This paradox, cautiously described as"non-classical non-classical" k i n e t i c behaviour, could be t e n t a t i v e l y explained i n terms of the i n i t i a l non-random d i s t r i b u t i o n of radicals and ions within the spur i t s e l f . However, we do not propose to continue t h i s discussion on non-homogeneity at t h i s stage, but to return to the results of further experiments on t h i s topic i n which this further manifestation of non-classical behaviour was investigated while laying the k i n e t i c ground work for the photolysis of e aq (3) More detailed Kinetic Studies, using the Stainless Steel C e l l It was to be anticipated that the k i n e t i c analysis of a new series of experiments i n which the stainless s t e e l c e l l featured would lead to a more detailed picture of these i n i t i a l events. The equipment had been de-signed to reduce the noise l e v e l on the signal and signal-to-noise r a t i o was vastly improved. The increase i n s e n s i t i v i t y , p a r t i c u l a r l y noticeable over the f i r s t 50 nsec or so of the trace, and the new precision adjustment of the in t e r n a l 0.15 mm diameter pinhole enabled us to closely 2 examine the decay of e within a cross-sectional area of 0.018 mm 3 aq over the i n i t i a l 20 nanoseconds af t e r the pulse. This portion of the decay was most important as f a r as the photolysis experiments were concern--120-_ * ed, as the photoexcitation of e -> e would occur within t h i s time r aq aq (p. 86 ). A conspicuous and quite unpredicted increase i n the absorption signal was observed immediately following the electron pulse. This was very surprising as the formation of e was expected to be complete after the electron pulse was over, that i s a f t e r 3 nsec. The increase i n the signal which appeared as an appreciable curvature on which had been presumed to be a negative slope continued for possibly 10 or 20 nsec and having reached a maximum o p t i c a l density decreased i n the manner previously described. In the e a r l i e r work an accurate analysis of the f i r s t 20 nsec or so was frequently hampered by noise on the s i g n a l . This i n t e r v a l was usually represented by a single datum because of uncertainty i n the measurement (103, p. 52) or neglected i n as much as the f i r s t measurement for time zero had been sh i f t e d to + 20 nsec for the same reason. Repetition of these experiments on the p l e x i c e l l , s p e c i f i c a l l y modified to reduce the noise for the photolysis experiments, did indeed show the same effect. Thus the observation of curvature was not confined to a p a r t i c u l a r radiation c e l l . Confronted with the p o s s i b i l i t y that t h i s effect might be part of the k i n e t i c complexity of the e decay and important to the general understanding of spur behaviour an experimental programme was carried out to answer the following questions: (i ) Was the observation of an i n i t i a l increase i n the s i g n a l , the so-called curvature e f f e c t , p r i o r to the decay a real event or an a r t i f a c t ? -121-( i i ) I f i t were r e a l was the increase due to additional numbers of e or some other species absorbing at 632.8 nm? ( i i i ) I f the curvature could be assigned to eaCj» then what sort of dose and solute dependence could be established as a clue to the reactions involved? The results and discussion that follow are based on experiments with the stainless s t e e l c e l l using an i r r a d i a t i o n path length of 2.8 mm unless stated otherwise. (a) The o r i g i n of the curvature observed on the decay si g n a l . T y p i c a l l y the absorption signal rose to a maximum value ^ 40 nsec after the delivery of the electron pulse. The increase i n o p t i c a l density was ^ 30% of the value immediately after the pulse. The curvature could arise from an external cause, namely electronic or o p t i c a l , or from an in t e r n a l cause within the radiation c e l l , through mechanical, o p t i c a l or chemical means. This included the p o s s i b i l i t y that interference phenomena such as described i n the previous chapter were responsible for d i s t o r t i n g the sig n a l . Although i t was not l i k e l y that the r i s e time of the photo-m u l t i p l i e r was responsible for the curvature, nevertheless the 50 S3 load r e s i s t o r was replaced by successively higher resistances. These radiation experiments were carried out on 2D water and ^^SO^ solution. In each case the measured r i s e time was very close to that estimated for the c i r c u i t . In the presence of 5 x 10 2 M [H +], there was no curvature at a l l and the rate of decay of the signal corresponded exactly to the disappearance of e &^ i n pseudo f i r s t - order k i n e t i c s with a proton. The oscilloscope saw these very fast signals s a t i s f a c t o r i l y on a l l levels of s e n s i t i v i t y . -122-Different photomultiplier tubes were substituted for the IP28 normally used, the operating voltage on the photomultiplier was taken up to 800 V i n 25 V stages and the signals fed into the oscilloscope vi a d i f f e r e n t co-axial cables with and without terminators. Although extraneous capacitances are a problem i n high frequency transmission the long duration of the curvature speaks against t h i s as a primary cause. In no instance was the detection system shown to be responsible for the curvature. The oscilloscope was checked independently i n another arrangement designed to monitor the Cerenkov emission from pulse radiolysed l i q u i d s . (35)- At the same time i t was observed that there was a t a i l on the Cerenkov emission from pure water that lasted for 7 nanoseconds, but t h i s could not account for the curvature as observed because the emission was both very weak and too short-lived. When the t o t a l output power of the electron beam was monitored c a l o r i m e t r i c a l l y by determining the temperature r i s e i n a disc of t h i n aluminum previously calibrated on a microvoltmeter, the output was 9.55 joules compared to 10.0 joules o r i g i n a l l y specified. At a l a t e r date t h i s tube was replaced by a new model whose beam out-put was on average s l i g h t l y higher than the previous one but the curvature s t i l l remained very much i n evidence. It seemed clear that the cause was not to be found i n the external electronic equipment. The effect of H + on the curvature and decay implied instead that i t was chemical i n o r i g i n . Figure 7-3 reproduces oscilloscope traces for the pure water and acid systems. The optics i n the experimental arrangement were then tested for fault y alignment by deliberately misaligning or removing lens and f i l t e r -123-components from the o p t i c a l axis. The shape of the curvature always remained the same. The use of a narrow band pass f i l t e r to eliminate wavelengths outside 632 ± 0.5 nm prevented other than a small f r a c t i o n of the Cerenkov emission from reaching the detector. The s e l f -absorption of Cerenkov radiation by e i s considerably reduced by high aq concentrations of acid (35) thus we might have expected to see an increase i n the curvature i n acid solution i f Cerenkov emission were the cause. This was not observed and i n fact the base l i n e on the oscilloscope trace was quite f l a t u n t i l the onset of the interference some 400 nsec l a t e r . The 5 mW He-Ne gas laser supplied monitoring l i g h t that was l i n e a r l y polarised, so i n order to establish the importance ( i f any) of these p o l a r i s a t i o n c h a r a c t e r i s t i c s on the shape of the signal a 3 mW non-polarised He-Ne gas laser source was substituted i n some of the runs. The shape and duration of the curvature were.unaffected, and the rate constant for the eventual decay of e was the same, ^  l O 1 ^ M 1 sec ^ . 1 aq ' The p o l a r i s a t i o n of the beam transmitted from the c e l l was examined next with a p o l a r i s i n g f i l t e r . The l a t t e r only transmitted l i g h t i n one plane. The signals from the 3 mW laser were unaffected by rotating the f i l t e r i n d i f f e r e n t experiments while those from the 5 mW laser were either unaffected,or the laser l i g h t was not transmitted as would be predicted. It was concluded that neither the o p t i c a l arrangement, nor Cerenkov emission during the electron pulse, nor the polarised nature of the monitoring l i g h t was the cause of the curvature we had observed. Although the radiation c e l l was lined up on the o p t i c a l bench,and the laser beam and lenses adjusted with meticulous care u n t i l the pinhole attached to the c e l l was on the o p t i c a l axis, there - 1 2 4 -nevertheless remained the p o s s i b i l i t y that the position of the pinhole "moved" following the electron pulse. This would come about through mechanical movement of the pinhole (or the radi a t i o n c e l l ) with respect to the fixed laser beam, or through micro-movement of the laser beam i n the l i q u i d (from a temperature effect perhaps) that simulated an increase i n i n t e n s i t y as seen by the photomultiplier through the pinhole. It had already been recorded that the extent of the curvature did depend on the posit i o n of the pinhole as i t scanned across the radiation c e l l ; t h i s point w i l l be elaborated l a t e r i n t h i s chapter. The simplest way of confirming the part played by the pinhole was to remove i t e n t i r e l y from the radiation c e l l assembly. This resulted i n small changes i n the shape and duration of the curvature which could be explained i n terms of an averaging effect since a p o s i t i o n a l dependence for the curvature had already been noticed, but the pinhole was neverthe-less s a t i s f a c t o r i l y collimating the beam and not d i s t o r t i n g i t . F i e l d e f f e c t s , interactions between photons and the free electrons (108) (before hydration) and space charges created within the l i q u i d volume were other proposals but, with the exception of the la s t one, seemed inappropriate i n r e l a t i o n to the length of time over which the curvature could be observed. Additional chemical e/idence l a t e r confirmed t h i s viewpoint. The features of the curvature were compared for both 2D deaerated _2 water and 10 M Na2S0^ solution i n which any space charges should have been reduced, i f not eliminated. Both the shape and duration of the curvature and subsequent bimolecular decay of e were i d e n t i c a l for pure water and a solution of quite d i f f e r e n t i o n i c strength, disposing -125-o£ any space charge effects. The stainless s t e e l window on the radiation c e l l was quite t h i n (2 to 3 thousandths of an inch) and presented the p o s s i b i l i t y of "window flapping" i n the system under the impact of the electron pulse. Unfortunately there was nothing that could be d i r e c t l y investigated along t h i s l i n e of reasoning so i t was decided to cut away that part of the st e e l c e l l and reseal i t with 1 thousandth stainless s t e e l f o i l . When t h i s was accomplished the t o t a l dose deposited i n the c e l l was expected to be a l i t t l e higher because of the higher average k i n e t i c energy of the beam electrons penetrating the solution. Dose depth studies confirmed t h i s fact - and also showed that the curvature increased for the same experimental conditions. The ambiguous conclusion was that the curvature increased as a re s u l t of the higher dose or the greater f l e x i b i l i t y of the window. This ambiguity was removed by two further observations. F i r s t , that the curvature was present on those traces taken during experiments on the modified p l e x i c e l l , and secondly the observation of a strong dependence of the i n t e n s i t y and duration of the curvature on the presence of certain chemical species. The decay of e i n the presence of small concentrations of 0 2 or H + was d e f i n i t e l y faster as would be expected but the curvature was not f u l l y eliminated u n t i l a concentration of [H +] = 1.6 x 10 3 M was used. Reaction (5) i s very fast with a rate constant of 2.32 x l O 1 ^ M 1 s e" + H 70 + -*• H + H-0 (5) aq 3 2 The value calculated for pseudo f i r s t - order k i n e t i c s was 2.05 x l O 1 ^ M 1 sec Electron scavengers thus reduced or removed -126-• i — r 1 n •r T — l l l l "l i | -%T %T -20 20 -60 • f 60 -100 J 100 -« 1- « i i J i J... .J i i (a) H^ O, 12.5 nsec per d i v i s i o n . %T 20 60 100 (b) %T 20 60 100 (d)^10" 4 M NaOH — i — r "' "T r | |- | - ' — T " ' l 1 T 1 1 I - - %T -20 ^ -• K 60 100 I 1 1 1 1 1 1 1 1 1 1 i upper trace, H20 (e) 'UO-3 M NaOH lower _3 trace, ^ 10 H 2S0 4 — i — i I — i — i — I — Tl 1 — " i — i — T - T — r j — i - %T -- f 20 - c —* ;J 60 1 100 -i i i i i 1 _ 1 .. 1 I .1 1 , 1 (c) upper trace, 0.1 M, methanol (f) 5 x 10" z M NaOH lower trace, H20 A l l traces except (a) are scaled at 62.5 nsec per horizontal d i v i s i o n . Figure 7-3. Oscilloscope Traces Showing the Effects of Different Solutes on the "Curvature". -127-the curvature i n accordance with known reaction rates. The evidence was thus strongly i n favour of a chemical o r i g i n for the i n i t i a l curvature on the absorption signal and experiments were designed to i d e n t i f y the species involved, (b) On the nature of the species involved. The effects of the 0^ and H + electron scavengers suggested that the posi t i v e slope on the signal was due to the delayed formation of some additional e during the usual decay. However, the p o s s i b i l i t y that a second absorbing species, was present as an impurity had to be considered. The only feasible candidate was CO^- whose absorption i s comparatively weak i n th i s region of the spectrum. Various solutes were therefore added to the pure water system to see what effect the presence of di f f e r e n t ions might have on the curvature. To avoid confusion i n the scavenger terminology, those solutes deliberately added to the otherwise pure system w i l l be referred to as "scavengers" while those species produced r a d i o l y t i c a l l y with e and subsequently aq reacting with i t w i l l be referred to as "electron predators". The l a t t e r include e" i t s e l f , H T0 +, OH, H and H o0 o. aq 3 ' • 2 2 The addition of 0.1 M methanol to the pure water system increased the duration of the curvature by a factor of 2, the signal taking some 100 nsec to reach i t s maximum value (see Fig. 7-3 (c)). In the presence _2 of 5 x 10 M NaOH the maximum o p t i c a l density was recorded after 140 nsec and the percentage increase i n the signal was subst a n t i a l l y higher. (See Fig. 7-3 (d), (e) and ( f ) ) . The same trends were observed when -2 an aqueous deaerated solution of 10 M BafOH^ was pulse i r r a d i a t e d , the absorption signal reaching i t s maximum i n 100 nsec (see Fig. 7-4 (c). When both 1 M ethanol and 5 x 10 M NaOH were present the signal continued to r i s e for about 350 nsec a f t e r which time the oncoming interference -128-signal o b l i t e r a t e d the trace as shown i n Fig. 7-4 (b). These results were interpreted i n the following way. The reactions taking place i n the pure water system are: e + H + H 0 + OH" aq 2 e" + OH -+ OH" e" + H_0+ -+ H + H-0 aq 3 2 e" + H o0 o -> OH + OH" aq 2 2 aq OH + OH ^ H 20 2 2.5 x 1 0 1 0 M"1 sec" 1 (2) 3.0 x 1 0 1 0 M"1 sec" 1 (3) 2.32 x 1 0 1 0 M _ 1 sec" 1 (5) 1.23 x 1 0 1 0 M 1 sec" 1 (4) 5 x 10 9 M"1 sec" 1 (8) and i f CO^ i s present as an impurity, OH + CO, CO, '12 3.0 x 108 M"1 sec 1 (12) The additional of 0.1 M methanol, a hydroxyl and hydrogen atom scavenger, should reduce the loss of e &^ through reaction with these predators. The methanol reacts competitively with the r a d i c a l s , the reaction with e being r e l a t i v e l y slow. aq 6 J e + CH,OH aq 3 13 H OH + CH30H .-*• CH20H + H 2 ' k 14 m 4 M - 1 " I < 10 M sec 1.7 x 10 6 M 1 s e c " 1 + CH3OH CH2OH + H20 k 5 = 5.1 x 10 8 M"1 sec" 1 (13) (14) (15) -129-%T 20 60 100 20 60 100 %T 20 60 100 •'--r- r — i i i i — i l l i i i (a) H 2 6 isT i—«—i—i—r _ l I 1 I I 1 J _ 125 nsec per div horizontal 125 nsec per div horizontal (b) 1 M ethanol + 5 x 10" 2 M NaOH i i i T — i — I — r • ' l i i i_ 125 nsec per div horizontal (c) upper trace, Ba(OH). lower trace, H^ O Figure 7-4. Oscilloscope Traces Showing the effect of Different Solutes on the "Curvature". -130-Reactions (2), (3), (8) and (12) would be eliminated, and reaction (4) i n as much as the main source of ^2^2 ^ a s D e e n removed. Some molecular ^2^2 a^rea^y exists from i n t r a spur reactions _9 within 10 sec but i t s y i e l d i s small,'v 0.8. The increase i n the duration of the curvature i s thus consistent with the removal of electron predators, and not CO^ . The OH radicals would survive less than 20 nsec at that solute concentration and therefore cannot p a r t i c i p a t e i n reaction (12). 3 - 1 - 1 The extinction c o e f f i c i e n t of CO^ i s 1.6 x 10 M cm at 630 nm 4 - 1 - 1 compared to 1.23 x 10 M cm for e ; t h i s difference i n conjunction with at least two orders of magnitude difference between t h e i r concentrations suggests that any contributions from CO" would be i n s u f f i c i e n t to account for the extent of curvature observed. Neverthe-less i t was desirable to obtain direct proof of the non-involvement of CO^ , p a r t i c u l a r l y i n the alkalin e solutions where CO^- was known to be present as a natural impurity. The observation of an extended curvature effect i n BafOH^ solutions demonstrated that CO^- could not be the -9 absorbing species involved. The s o l u b i l i t y product of BaCO^ i s 7.5 x 10 at 20°C. In a solution of 10 2 M B a + + ions the amount of carbonate remaining i n solution would be less than 10 ^  M, and yet there was > 50% increase over the i n i t i a l o p t i c a l density within 100 nsec. The l i k e l i h o o d of C0.j as a major contributor was discarded while the evidence f o r e strengthened, aq 5 The decay of e i n i r r a d i a t e d solutions of NaOH at dif f e r e n t aq concentrations showed a marked dependence on the [OH ]. As th i s was -4 -2 raised from 10 M to 5 x 10 M the time taken for the curvature to -131-reach i t s maximum value steadily increased. The OH competitively removes H^0+ and reduces the loss of e through reaction (5) but the increase i n the o p t i c a l density maximum was far i n excess of that expected by eliminating t h i s reaction. Obviously the scavenging of H,0+ from the system not only extended the l i f e t i m e of e but 3 . aq enhanced the effectiveness of the mechanisms that led to the i n i t i a l increase i n o p t i c a l density. The dramatic increase i n both the in t e n s i t y and duration of the _ 2 curvature i n the presence of 1 M ethanol and 5 x 10 M NaOH confirmed the role of H, OH and H_0+ i n suppressing e through reactions (2), (3) and (5). Ethanol i s more eff e c t i v e i n scavenging H and OH by a factor of 10 compared to methanol, but reacts with .e < 10^ M - 1 sec OH + C2H5OH -»• C 2H 40H + H20 - lc = 1.1 x 10 9 M _ 1 sec" 1 (16) H + C 2H 50H -> C 2H 40H +• H 2 k 1 ?= 1.6 x 10 7 M _ 1 sec" 1 (17) By reducing OH i n the system i t also follows that H 20 2 and H0 2 have been removed to a considerable extent through reactions (8) and (18) OH + OH -* H 20 2 kg = 5 x 10 9 M _ 1 sec" 1 (8) OH + H 20 2 -> H20 + H0 2 k 1 8 = 4.5 x 10 7 M - 1 sec" 1 (18) Conceivably OH and H 20 2 could be i n t e r f e r i n g with the chemical processes giving r i s e to the apparent increase i n t e a q ] a s w e H as preying on e &^ i t s e l f . -132-(c) Summary + - ++ The effects of adding different scavengers, H , 0^, OH , Ba , • CH^ OH and C2HC-OH, to the pure water system has evidently demonstrated that the curvature on the absorption signal was due to the delayed formation of e through' a mechanism as yet unknown. -133-Chapter VIII DATA FROM THE REMAINING PULSE RADIOLYSIS EXPERIMENTS The evidence so far led to the conclusion that e was the only aq } known absorbing species i n the system - other than an intermediate transient of which we have no present knowledge - and that the variations i n the i n t e n s i t y and duration of the curvature on the ab-sorption signal were the r e s u l t of variations i n the concentration and l i f e t i m e of or i t s parent species, or both. The experimental results further suggest that ,H.jO+, OH and possibly H are intimately involved i n whatever mechanism are taking place during the period of the curvature to create additional hydrated electrons. We now present data from experiments i n which the curvature effect was studied as a function of the dose across the radiation c e l l and as a function of the i r r a d i a t i o n pathlength. In view of the curvature theory scavenger experiments were also extended. In order to correlate a l l the data a model for the formation and decay of e w i l l be developed for t h i s system i n the next section. F i r s t a short description of the pattern of energy deposition measured i n the stainless s t e e l c e l l . The dose-depth ch a r a c t e r i s t i c s of the 0.5 MeV and 0.6 MeV electron beams (electron tubes Model 5510 and 5515 respectively) i n water were investigated and the dose-depth curves are shown i n Fig. 8-1. The i n i t i a l o p t i c a l densities seen through the pinhole are plotted as a function of displacement of the pinhole from the stainless s t e e l window through which the beam penetrated the solution. The 0.5 MeV electrons had to penetrate between 2 and 3 thou' of stainless s t e e l before reaching the water, and t h i s would have cut out some of the lower -134-energy electrons i n the beam (which loses i t s monoenergetic character as i t travels through the tungsten window of the electron tube, the a i r and steel media). The 0.6 MeV electrons only saw a 1 thou' stainless s t e e l b a r r i e r following the window modifications to the c e l l . Nevertheless the difference i n the average k i n e t i c energy of the beam electrons and divergence ch a r a c t e r i s t i c s of the beams gave r i s e to d i s s i m i l a r dose-depth curves, regardless of the v a r i a t i o n i n the thickness of the c e l l window. The immediate f a l l - o f f i n the o p t i c a l density (= t o t a l energy deposited) observed when the lower energy beam was employed occurred at a l l i r r a d i a t i o n pathlengths. Data recorded from 2.8 mm and 1 mm paths are shown and i n the text this dose-depth pattern w i l l be referred to as type I. In contrast the higher energy electron beam, which i s also more focused than the 0.5 MeV beam, gave r i s e to the maximum number excitations and ionisations at about 1 mm depth i n the water, a f t e r which there was a sharp decline i n the energy deposited. This dose-depth pattern w i l l be referred to as type I I . For comparison a published dose-depth curve (109) for the l a t t e r energy i s included i n Fig. 8-1, and i t may be seen that the curves match quite w e l l . The two patterns r e f l e c t the different L.E.T. and energy c h a r a c t e r i s t i c s from these two electron tubes. A study of Beer's Law (OD = ecil) i n t h i s system was unsatisfactory, p a r t i c u l a r l y with the 0.6 MeV electron beam because the r a d i a l d i s t r i b u t i o n was s t i l l confined to < 5 mm by the time i t reached the radi a t i o n c e l l window,2 cms away from the electron tube window. Consequently pathlengths were r e s t r i c t e d to 2 mm or less i n using t h i s beam. The 0.5 MeV beam had diverged s u f f i c i e n t l y to be able to work at s l i g h t l y longer Table 8-1 POSITION mm 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.8 1.0 1.3 1.5 1.6 1.7 pathlength Stainless Steel C e l l , model 5510 electron tube 1.0 mm 0.46 0.40 0.34 0.22 0.17 0.15 0.10 2.8 mm 0.63 0.53 0.53 0.53 0.50 0.51 0.42 0.30 0.12 5.0mm 0.43 0.46 0.48 0.48 0.27 0.10 Stainless Steel C e l l with modified window, model 5515 electron tube 2.0 mm 0.58 0.58 0.60 0.59 0.50 0.26 5.0 mm 0.53 0.59 0.58 0.70 0.51 0.30 0.26 on Depth-dose Relationships for water : op t i c a l density as a function of pinhole displacement •< ' ' I I I I I , . 0.2 0.4 0.6 0.8 pinhole displacement (mm) 1.6 -137-pathlengths but some uncertainty s t i l l existed because of scattering effects. The scattering resulted i n a s i g n i f i c a n t d i s p a r i t y between the s l i t width as calibrated at the radiation c e l l window and the r e a l pathlengths of absorbing species as seen by the laser beam i n di f f e r e n t regions of the i r r a d i a t e d l i q u i d . (1) A Model for the Formation and Decay of e The increase i n the numbers of hydrated electrons for times up to 40 nanoseconds i n pure water following a 3 nanosecond electron pulse i s a chemical phenomenon, but the exact reason has to be pin-pointed i n order to give meaning to the data. It i s impossible for the random di f f u s i o n of hydrated electrons from areas outside the immediate scanning zone to be adequately fast and e f f i c i e n t to account for t h i s increase i n [e - ]. The effect of a space charge set up i n the volume of l i q u i d might be to e l e c t r o s t a t i c a l l y "funnel" the d i f f u s i o n i n one d i r e c t i o n , but the chemical evidence does not support the idea of an intense space charge present i n the system. In any case the hydrated electrons could only move i n one d i r e c t i o n about 1.4 x 10 mm within the 40 nanoseconds period and the pinhole scans a diameter of 0.15 mm. It i s equally improbable that the increase i n [e~ ] arises from hydrated electrons i n i t i a l l y photolysed by the laser or a Cerenkov * photon to e and then returning to the ground state to absorb the 632.8 nm l i g h t . In the f i r s t place there i s a very unfavourable photon:electron r a t i o for laser or Cerenkov photolysis as we discussed e a r l i e r , and i n the second place the dramatic increase i n the effect i n the presence of varying scavengers suggests that t h i s i s not the cause. -138-_9 We therefore make the naive assumption that there are from 10 seconds onwards two types of hydrated electrons (physically indistinguishable of course) one of which i s p a r t i c i p a t i n g i n a bimolecular decay mechanism A, the other being generated (or regenerated from events -9 before 10 seconds) through an unknown mechanism B. The experimentally observed formation and decay of e i s therefore interpreted as a time-dependent summation of these two processes, as i l l u s t r a t e d i n Fig. 8-2. From inspection of the experimental curves both process would be extremely fast. The model i s based on the following assumptions. (a) That there i s a given s p a t i a l d i s t r i b u t i o n of the rad i c a l s and ions associated with the spur. I f i n an average case there are s i x species, then t h i s i n i t i a l d i s t r i b u t i o n w i l l f i n d the e some 100 A r ' aq away from the centre of the spur where the OH and H^0+ are clustered as shown i n Fig. 8-2. (b) As a consequence of t h i s s p a t i a l separation, the only viable decay mechanism for e during the f i r s t 50 nanoseconds i s reaction (1) represented by A. e" + e" -*• H 0 + 20H" (1) aq aq 2 aq Within t h i s period of time e could diffuse a t o t a l distance of r aq ^ 150 A. (c) That the decay of e during t h i s i n t e r v a l i s only through aq reaction (1) and up to 100 nanoseconds after the electron pulse may be represented by a sum of the reactions with OH and H^0+, excluding H and H 20 2. -139-(d) That the formation of e after the delivery of the pulse i s through one dominant process B. Some of these over s i m p l i f i c a t i o n s w i l l now be b r i e f l y j u s t i f i e d . E s s e n t i a l l y the thermalised electron has been allowed to escape ^ 100 A from the p o s i t i v e ion H^O before being hydrated. Within that time H 20 + reacted with another water molecule to give H^ O* and OH. (This assumption i s consistent with the Lea - Grey - Platzman model only). Since none of these primary species are immobile, the d i f f u s i o n constant for H,0+ exceeding that for e - and OH (9.10 5, 4.7 x 10 5 and 3 ° aq -5 2 - 1 -2.8 x 10 cm sec res p e c t i v e l y ) , the time lapse during which e are supposed only to react with each other must be c r i t i c a l l y evaluated. _ o + o Within 50 nanoseconds e may diffuse 150 A, H,0 some 210 A and the OH aq J ' 3 o r a d i c a l about 120 A from t h e i r o r i g i n a l positions. At the dose-rates employed i t i s more l i k e l y that the e w i l l meet a second e ^ perhaps from another spur than be caught by the other species i n t h i s time. The measured decay of e i n the pure water system after ^ 100 nanoseconds aq was 5.8 x 101(^ M _ 1 s e c - 1 and t h i s includes the reactions with HgO* and OH (k^ + kg + kj. = 6.2 x l O 1 ^ M 1 sec * ) . This imposes an upper l i m i t o _ of < 200 A on the average d i f f u s i v e plus escape distance for e on t h i s model. It should be emphasised however that t h i s i s a working model constructed to ass i s t i n the presentation of data. The rea l physical significance i s not necessarily meaningful at t h i s stage. Accepting that A represents a loss of e within the system and B a gain of e aq> then the following conclusions are drawn i n r e l a t i o n to the height and shape of the experimental traces. -140-50 100 Time (nsec) 200 Figure 8-2. Graphical Representation of the A,B Model. - 1 4 1 -The number of e o r i g i n a l l y generated i n the system i s given by the aq i n i t i a l o p t i c a l density value. This could be due to those e i n aq A alone or i n both A and B. We w i l l assume at time zero there i s no contribution form B. When the rate of formation dB greatly exceeds the rate of disappear-dt ance of the electron - dA there w i l l be a'net gain i n e and a po s i t i v e dt a q slope on the experimental absorption signals. When dB ^  -dA dt dt an equilibrium i s attained whereby there i s no observed net loss or gain of e , and. the slope approaches zero. At th i s stage the number of aq absorbing species p a r t i c i p a t i n g i n both A and B w i l l be given by the maximum o p t i c a l density. When the relationship between A and B i s such that there i s a net loss of e through reactions discussed e a r l i e r , then the slope w i l l be negative and once the o p t i c a l density has returned to i t s value at the end of the pulse i t i s reasonable to suppose that only an i n s i g n i f i c a n t number of e" are s t i l l being generated i n the system. aq The information required to evaluate the r e l a t i v e importance of these two processes A and B as a function of time, i n i t i a l concentration of e , radiation dose and the presence of other ions i n solution w i l l be tabulated using the following abbreviations: Position horizontal displacement of pinhole from radiation c e l l window OD^  i n i t i a l o p t i c a l density OD maximum o p t i c a l density max r ' t time delay for OD. to reach value of OD ' max 1 1 max t ^ time delay before absorption signal returned to i t s i n i t i a l value, OD. k^ - f i r s t - order rate constant for curve after t °r max -142-K- second - order "rate - constant" for curve a f t e r t p max AOD % increase i n absorption with respect to OD.. (0D- - OD.) x 100 r r i ' max 1  OD. l (2) Experimental data correlated through the A, B Model The tables i n t h i s section characterise the experimental data i n terms of the parameters-just defined. Each p a r t i c u l a r measurement i s an average determined from many separate experiments. The r.m.s. deviation calculated for parameters other than rate constants seldom exceeded 10%. The determination of K values has been discussed i n the la s t chapter, and the error involved i s d i f f i c u l t to assess. Variation on an i n d i v i d u a l basis was within 30%, but K that f a l l outside that range should not necessarily be ;taken as evidence for additional reactions because of the uncertainty interposed on any decay by the presence of process B. Definite trends i n the K values are probably more meaningful. Optical density data was calculated to f i v e decimal places and are given here as three s i g n i f i c a n t numbers. When quoted as two t h i s i s . because the data warranted such approximation. The % AOD have been chopped to integer numbers. Both t and t. could be measured to * r 6 max I ± 2 nsec unless bracketed; once again trends are probably more useful than in d i v i d u a l variations i n these empirical parameters. (a) Formation and decay of e as a function of dose and pathlength. The curvature was studied for both dose-depth patterns and d i f f e r e n t pathlengths, and the results are summarised below i n Table 8-2. The in d i v i d u a l sets of data for d i f f e r e n t pathlengths and dose are presented i n Tables 8-2 to 8-7. -143-Table 8-2 Percentage Curvature (% OD) as a Function of Ir r a d i a t i o n Pathlength POSITION (mm) 0.2 0.6 1.5 PATHLENGTH (JMJ) SS c e l l 1.0 9 0 0 type I 2.8 37 6 0 dose 5.0 68 12 0 SS c e l l 2.0 15 24 15 type II 5.0 33 30 12 F l e x i c e l l 3.0 *-16% over whole beam -> Percentage Curvature (%A0D) as a Function of Ir r a d i a t i o n Pathlength The degree of curvature depended on the i n i t i a l concentration of e~ and followed the dose-depth curve for a l l the pathlengths studied. As the Te" 1 decreased i n the depths of the i r r a d i a t e d volume so L aq J r the po s i t i v e slope became less marked u n t i l i t f i n a l l y disappeared leaving a plateau on the absorption signal p r i o r to the decay. These plateaux had been observed i n the e a r l i e r work (103 p. 72) and l a s t between 30 and 100 nsec depending on the i n i t i a l number of e~ formed i n the pulse. These plateaux or exceedingly slow decays are now attributed to the superposition of a small po s i t i v e slope on the normal decay. Table 8-3 POSITION mm I/) o W> C • H (/> O 0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.9 1.0 OD. l 0.625 0.53 0.53 0.53 0.504 0.51 0.419 0.35 0.126 0.123 0D max 0.822 0.72 0.73 0.70 0.625 0.60 0.444 0.40 AODi 31 36 37 24 24 15 6 12 t nsec max 37 29 24 20 30 15 10 10 0 0 t.nsec I 94 96 100 96 100 80 42 KgM s e c 3.32 x 10 10 4.97 x 10 1.96 x 10 10 10 i 1.96 x 10 10 In these experiments a 2.8 mm path of 2 D deaerated water was irr a d i a t e d i n the o r i g i n a l stainless steel c e l l . Optical Density and Curvature Data as a Function of Pinhole Position Table 8-4 POSITION mm OD. l OD max A0D% t nsec max t.nsec l K^M sec type I 0.0 0.465 0.530 14 13 37 1.44 x 1 0 1 0 type I 0.1 0.402 0.444 11 13 32 > i o 1 0 bo c 0.2 0.336 0.353 5 17 30 1.20 x 1 0 1 0 :eas: 0.3 0.219 0.240 10 15 30 1.13 x i o 1 0 dec] 0.4 0.166 0.178 7 10 > i o 1 0 dose 0.45 0.149 0 1.00 x i o 1 0 0.55 0.113 0 3.90 x i o 1 0 In these experiments a 1mm pathlength of 2 D deaerated water was ir r a d i a t e d i n the o r i g i n a l stainless steel c e l l . Optical Density and Curvature Data as a Function of Pinhole Position Table 8-5 POSITION OD. OD AOD% t nsec t.nsec K_M- 1sec- 1 mm 1 max max 1 /3 P, 0.0 0.430 0.721 72 42 130 3.77 x 1 0 1 0 ^ 10 0.2 0.458 0.770 68 41 150 3.00 x 10 - „ w 10 ° 0.4 0.484 0.649 34 44 152 . 2.57 X 10 c 0.6 0.475 0.530 12 20 50 2.12 x 1 0 1 0 w " if) 2 0.8 0.271 0 3.36 x 10 u o 10 •S 1.0 0.108 0 * 10 In these experiments a 5 mm pathlength of 2 D deaerated water was irr a d i a t e d i n the o r i g i n a l stainless s t e e l c e l l . Optical Density and Curvature Data as a Function of Pinhole Position Table 8-6 POSITION mm OD. OD max AOD% t nsec max t.nsec 1 KgM sec 60 c • H in 0 3 <U V i o CD CD o Q 0.1 0.4 0.8 1.0 1.3 1.5 0.580 0.575 0.600 0.594 0.495 0.258 0.665 0.754 0.724 0.722 0.546 0.297 15 31 23 22 10 15 20 20 17 15 15 25 45 50 52 40 40 33 5.05 x 10 3.60 x 10 2.02 x 10 1.85 x 10 1.43 x 10 10 10 10 10 10 4^  1.31 x 10 10 In these experiments a 2 mm pathlength of 2 D deaerated water was ir r a d i a t e d i n the o r i g i n a l stainless s t e e l c e l l . Optical Density and Curvature Data as a Function of Pinhole Position Table 8-7 POSITION. 0.2 0.4 0.6 1.0 1.3 1.6 1.7 mm OD. l 0.525 0.594 0.580 0.701 0.510 0.295 0.260 OD max 0.697 0.730 0.752 0.933 0.625 0.324 A0D% 33 33 30 33 23 10 t nsec max 20 15 15 25 23 10 t.nsec I 45 43 45 77 120 50 K ^ M sec 7.67 x 10 6.25 x 10 9.06 x 10 4.73 x 10 1.50 x 10 1.5 x 10 10 10 10 10 10 10 In these experiments a 5 mm pathlength of 2 D deaerated water was ir r a d i a t e d i n the o r i g i n a l stainless s t e e l c e l l . Optical Density and Curvature Data as a Function of Pinhole Position 2.6 7 0 D 2.0 & 1.0 o A o o A A o A © A • • o A ° © A T zr © 2mm path length (dose depth type II ) G o A e Pinhole p o s i t i o n 0.4 mm 0. 8 mm 1.0 mm 1.3 mm 0 50 Time (nsec) 100 150 Figure 8-3. Second Order Treatment of Data from Stainless Steel C e l l 200 -150-At longer pathlengths, a greater curvature effect was observed. The time taken to reach the maximum OD at any pathlength decreased as the percentage of curvature on the signal decreased over the cross-section of the c e l l , i n accordance with the dose-depth curve. These observations, together with the fact that the time taken to return to the i n i t i a l value of the o p t i c a l density also decreased as the curvature effect became less s i g n i f i c a n t , point towards the presence of a species Xg whose i n i t i a l concentration i s proportional to the i n i t i a l [ e a q ] • Xg generates extra e through mechanisms as yet unknown but does not appear to have an existence independent of t e a q ] i n t h i s system. The K values determined a f t e r t were > l O 1 ^ M 1 sec 1 max Of p a r t i c u l a r interest was the difference i n K values for the type II dose depth curves. For a few tenths of a millimeter from the c e l l window, K was s i g n i f i c a n t l y faster than i n the maximum dose pos i t i o n and further back s t i l l the K values became even slower. This must be an L.E.T. effect since i n terms of the number of e i n i t i a l l y formed aq ' we might have expected s i m i l a r K values either side of the maximum dose position. Data from a set of these decays, plotted as 1 v t , i t OD shown i n Fig. 8-3_ (b) Formation and decay of e i n the presence of various additives Because of the complex nature of the ki n e t i c s of formation and decay within the period of the curvature and non-homogeneity, concen-trations of solutes were selected i n order to simulate a given s p a t i a l d i s t r i b u t i o n of ions within the l i q u i d . Hopefully i t would be possible to relate the subsequent findings to the general spur picture discussed i n the f i r s t chapter and the simple model outlined a few pages back. -151-The radius of a sphere i n which a solute molecule or ion has a high p r o b a b i l i t y of being found may be calculated as follows: Assuming a random d i s t r i b u t i o n of molecules, each molecule w i l l -3 -1 require a spherical volume of (NoM 10 ) ml of which the radius r g i s (3/4 IT No.M. 10 " 3 ) 1 / 3 . 10 8 A j i n a 5 x 10" 2 M solution, r g = 20 A. On average, therefore, there w i l l be a solute molecule within every o spur i f the mean spur radius i s taken as 20 A. Mol a r i t i e s were chosen i n these experiments to give 15 A. < r g < 350 A. The results are summarised i n Tables 8-8 and 8-9. -2 The addition of 10 M Na2S0^ to the otherwise pure system had no effect at a l l on the formation and decay of e as seen through the curvature on the absorption s i g n a l . This demonstrated that the presence of Na + was not an influencing factor as f a r as process B was concerned. When the absorption signals were studied as a function of [OH ] using NaOH, the curvature i n t e n s i f i e d i n both duration and degree as [OH-] -2 increased. At the highest concentration, 5 x 10 M [OH ], the increase i n o p t i c a l density over the i n i t i a l l y recorded value was i n excess of 100%, and the time taken for the system to reach t h i s value t = 135 nsec 3 max was almost a factor of 7 longer than the time i n pure water. When the -3 source of [OH ] was BaOOH^ at 9.5 x 10 M the maximum o p t i c a l density was observed after 115 nsec and contributed a 55% increase to the absorption s i g n a l . A plot of t ^ and ^ m a x v log [OH-] was l i n e a r i n both cases as shown i n Fig. 8-4. Apparently OH was a s s i s t i n g process B i n a quantitative way and from the %A0D values at high [OH ] th i s assistance did not seem confined to removing the predator H_0+. Table 8-8 Solute OD. l OD max AOTJ% t nsec max t.nsec I K 2M" •1 -1 sec H 20 0.525 0.697 33 20 45 > 1 0 U NaOH, 1.2 x 10" 4M 0.525 0.758 44 50 115 1.2 x 1 0 U NaOH, 1.15 x 1 0 _ 3 M 0.618 0.975 57 83 188 5.6 i n 1 0 x 10 NaOH, 5.0 x 1 0 _ 2 M 0.572 1.232 117 135 377 2.4 x 10 CH3OH, 0.1 M 0.554 0.786 42 100 195 5.2 x 10 NaOH, 5.0 x 1 0 _ 2 M + C2H5OH, 1M 0.72 1.25 74 210 ? + 1/5 atmos. 0 2 NaOH, 5 x 10" 2M + C2H5OH, 1M, \ 0.70 1.26 >77 >300 ? deaerated Dependence of the Optical Density and Curvature Data on the Presence of Solutes: NaOH and R-OH. (Stainless s t e e l c e l l with a 5 mm path at pinhole position 0.2) Table 8-9 OD. OD %OD t nsec t.nsec K 0 M _ 1 s e c _ 1 i max max 1 2 H20 0.409 0.473 . 15 55 106 4.6 x 1 0 1 0 Na 2S0 4, 0.433 0.494 14 55 105 4.3 x 1 0 1 0 1.2 x 10" 2 M H + (H 2S0 4) 0.288 2.05 x 1 0 1 0 1.6 x 10" 3 M (These experiments were performed on the stainless steel c e l l with a 5 mm path at pinhole 0.2). H20 0.556 0.648 16 28 72 2.5 x 1 0 1 0 Ba(0H) 2, 0.594 0.926 55 115 ^310 1.0 x 1 0 1 0 9.5 x 10" 3 M (These two experiments were performed on the P l e x i c e l l with a 3 mm pathlength, monitoring the f u l l laser beam). Dependence of the Optical Density and Curvature Data on the Presence of Solutes: Na2SO/[, H^SO^, Ba(OH)2. I -154--155-The presence of 0.1 M methanol i n the i r r a d i a t e d solutions caused a 5-fold increase i n t and the 42% increase was about that observed max -4 -for 10 M [OH ]. This seems to indicate that either R-OH and OH are p a r t i c i p a t i n g i n the same o v e r a l l mechanism i n process B which i t s e l f contains more than one reacting species or that these scavengers affect unrelated mechanisms i n the whole A, B system. This point emerges again from the mixed solute experiments. _2 Following the i r r a d i a t i o n of a solut i o n of 5 x 10 M NaOH + 1 M ethanol the %A0D was > 77% (this i s a lower l i m i t because the absorption signal was disrupted by the interference front i n the region of the maximum absorption) although the time taken to reach 0 D m a x w a s > 300 nsec, A comparison with the %A0D from the a l k a l i n e solutions suggests that the i n - -4 methanol may be^competition with the OH . When oxygen ^ 4 x 10 M was present i n the 0H~"/R-0H system the %A0D was not very d i f f e r e n t from the lower l i m i t put on the deaerated solutions and t was only r max 7 210 nsec. The oxygen had removed e~ from both process A and B, hence the faster decay, but at that r e l a t i v e l y low concentration had not competitively int e r f e r e d with the reactions of 0H~ and ethanol. H + at 1.6 x 10 3 M from H~S0. solut i o n successfully removed e~ 2 4 ' aq within ^ 20 nsec i n accordance with known reaction rates. The measured pseudo f i r s t - order rate constant led to a second - order rate constant of 2.05 x 10*^ M 1 sec 1 i n close agreement to the accepted value of 2.3 x 1 0 ^ M 1 sec At t h i s concentration the H + ions o are within spheres, r g = 130 A. Clearly t h i s sort of d i s t r i b u t i o n i s not only e f f e c t i v e i n reducing e formed '"Instantaneously" during the pulse but i n eliminating the extra e generated from process B. -156-Perhaps H + even i n h i b i t s t h i s mechanism. (3) On the Contribution of B to the Measured P.P. Values The quantitative relationship between [OH ] and t ^ , *- m a x C*ne time parameters describing the curvature) shown i n Fig. 8-4 and the compatible effects of other scavengers i n the system motivated an attempt to mathematically extricate information concerning process B from the experimental data. The purpose was two-fold, to establish the dependence ( i f any) of the rate of formation of e~ •on these salutes and then to use t h i s information i n constructing mechanisms from which the extra e were being generated. The parent species of these "chemical" e~^ w i l l be c a l l e d Xg i n the remaining text. At the beginning of the chapter a model was developed which involved two processes A and B, and i t was proposed that the experimental data represented a time-dependent summation between A, i n which e aq was disappearing through reaction with predators and B i n which e was being formed. Hydrated electrons generated from Xg are immediately available for reaction and i n consequence any o p t i c a l density data we can extract for B w i l l s t i l l represent a r e l a t i v e rather than an absolute s i t u a t i o n . By setting l i m i t s on the rate at which e i s lost i n A, the contribution of e i n B to the experimental o p t i c a l density can be estimated. Given that e • does react, the slowest rate constant aq 9 - 1 - 1 for decay w i l l be with another e » at 5.5x 10 M sec , and the ' aq fastest reasonable rate for the duration of the curvature w i l l be 6 x l O 1 ^ M 1 sec 1 representing reaction with e , H^O+ and OH. (a) The i s o l a t i o n of B from experimental data. As a s t a r t i n g point the lower l i m i t on the loss of e w i l l be 6 r aq -157-used and we w i l l assume that at any time after the electron pulse the op t i c a l density data for process A which i s the bimolecular reaction (1), e" + e" -»• H 0 + 2 OH" (1) aq aq 2 aq can be calculated from the f a m i l i a r relationship 1_ - 1_ = k . t 0D OD — t o e£ 9 - 1 - 1 4 -1 -1 where k-^  i s known to be&Sx 10 M sec , eis 1.23 x 10 M cm and t , % (pathlength) are variables that take on specified values f o r each datum. A set of experimental data i s selected from which the OD^  value becomes the reference o p t i c a l density for the cal c u l a t i o n and i t s inverse y becomes a constant i n the equation below, a s i m p l i f i e d A = y + t (7.5 x 105)/£ form of the previous expression. A i s the inverse of the o p t i c a l density we wish to at t r i b u t e to the hydrated electrons reacting i n (1). Having calculated a set of values of A for the desired sequence of times, these are re a d i l y converted to o p t i c a l densities. The numerical difference between the value of (OD) and (OD) A exp represents the contribution from process B, (0D)g. This c a l c u l a t i o n i s only v a l i d as long as the bimolecular reaction (1) dominates the decay processes. According to the AB model, t h i s assumption i s v a l i d for 50 nanoseconds and not too unreasonable an approximation up to 100 nanoseconds (see section (3) i n the previous chapter). In the presence of H^0+, H and OH scavengers t h i s approximation w i l l be even better, since i n theory e cannot disappear i n t h i s system at a slower rate than dictated by reaction (1). (Non-uniform d i s t r i b u t i o n -158-of the species involved i n the r e a l system could well negate t h i s statement). The other i m p l i c i t assumption i s that at time zero for the measurements there are an i n s i g n i f i c a n t number of contributing to OD^  from the process B. The evidence so f a r suggests that accompanying the "instantaneous" formation of e during the electron pulse i s the formation of another as yet unidentified species Xg whose concentration i s somehow related to the numbers of e - produced. aq r The l i n e a r portion of the slope representing the rapid production of e from t h i s entity into the pool of absorbing species can be extrapolated back to time zero. In t h i s way i t i s estimated that even i f process B were active i n generating e within the f i r s t 5 nanoseconds, the contribution would be about 10% to the measured o p t i c a l density. Part of a t y p i c a l c a l c u l a t i o n i s shown i n Table 8-10. The data was taken from an experiment i n neutral water using a 2.8 mm pathlength. The dose-depth pattern was type I. Table 8-10 nsec Parameter 0 10 20 30 40 50 60 70 80 90 A (OD)A (OD) exp 1.8930 1.9198 1.9466 1.9734 2.0002 2.0270 2.0538 2.0806 2.1074 2.1342 0.5282 0.5210 0.5136 0.5068 0.500 0.4934 0.4869 0.4808 0.4747 0.4686 0.5282 0.6354 0.7324 0.7782 0.8030 0.7782 0.7324. 0.7010 0.6532 0.6102 (OD) B 0.0000 0.1144 0.2188 0.2714 0.3030 0.2848 0.2455 0.2192 0.1785 0.1416 * (0D) B max i s 57% of the value of 0D i Equation: A = y + 2.68 x 10 6 t Change i n (0D) D as a Function of Time o -160-(b) Influence of solutes on the rate of change of (0D) B Calculated (0D)g values are plotted as a function of time i n Fig. 8-5. The i n i t i a l slope of the curves did not vary with increasing [OH ] and had the same value as the pure water system, using the reaction (1) as the l i m i t on process A. The time taken to reach maximum (0D) B increased at higher [OH -]. In contrast the presence of 0.1 M methanol did not s i g n i f i c a n t l y a l t e r the maximum value of (OD)B r e l a t i v e to the pure water system, but noticeably affected the rate at which (OD)., was reached. ^ J Bmax If the loss of e - i n process A i s faster and the l i m i t of aq r 6 x 10* M 1 sec 1 i s imposed the (0D)g values increase i n pure water, the maximum of 0.30 at 40 nsec r i s i n g to 0.40. Similar calculations _2 on the 5 x 10 M NaOH data showed the same trend, (0D)g values increased. The difference between the maximum (0D) 's calculated from the lower and upper l i m i t s of process A did not account for the p a r t i c u l a r l y high %A0D value recorded for the highest.[OH ]. At -2 t for 5 x 10 M [OH 1, t h i s difference would be 0.22 o p t i c a l max L J' • r density units. Since the experimental OD i s 1.24 and 0.32 < (0D). max < 0.50, at least 0.74 can be attributed to (0D)g. In other words the dramatic increase i n the maximum value of (0D) D i n the D presence of strongly a l k a l i n e solutions could not be accounted f o r so l e l y by reducing the loss of hydrated electrons through reaction with predators i n A. Even though process B as we see i t i s a combination of the formation and loss of e , and addition of OH aq' could extend the l i f e t i m e of e formed i n t h i s way by eliminating reaction with H.jO+,the calculated (0D)g m a } C i s s t i l l too high. 50 Time (nsec) 150 200 Figure 8-5, The Rate of Formation of e" (given by (OD) ) i n the presence of various solutes. -162-The decay of e after t was determined for the dif f e r e n t J aq max solutes using (OD)^ values. Previous remarks concerning doubts over the rea l pathlength are even more applicable here, and the K values should be regarded as indicating trends rather than absolute differences. Pure Water 0.1 M methanol 1.2 x 10" 4 M NaOH 1.15 x IO" 3 M NaOH 5.0 x 10" 2 M NaOH (4) Photolysis Experiments with the S c i n t i l l a t o r s The p l e x i c e l l was designed f o r these experiments which u t i l i s e d the s c i n t i l l a t o r s described i n Chapter IV. 3D water previously deoxygenated with helium gas for several hours was i r r a d i a t e d with the 0.6 MeV electron beam collimated by the H bar,which r e s t r i c t e d the pathlength to 3 mm. The noise l e v e l was minimal. Careful scrutiny of the f i r s t 20 nsec of the absorption signal (from which readings could be taken every 1 nsec using the 5 nsec per d i v i s i o n oscilloscope sweep speed) could not detect any si g n i f i c a n t difference i n the i n i t i a l shape of the trace i n the presence of the s c i n t i l l a t o r s . The curvature on the signal was a d i s t i n c t nuisance i n t h i s respect because i t could have easily masked any small variations within t h i s i n t e r v a l . Another series of experiments i n which the decay of e~ K^ 2 = 1.8 x 10 M sec Ka = 1.8 x 10 1 1 M _ 1 sec A 5 K_ = 5.7 x 10 1 1 M 1 sec /» Kg = 3.4 x 1 0 1 1 M 1 sec . _ i r i10 --1 K „ = 4.7 x 10 M sec -163-was followed over the f i r s t 150 nsec yielded no d e f i n i t e conclusions. The data collected i s presented i n Table 8-11. The mean values for OD. and OD did not show any s i g n i f i c a n t I max 1 5 v a r i a t i o n and so differences between the %A0D values are not believed to be important. It i s to be noted however that when the photolysis l i g h t was 550 nm < hv < 590 nm the decay of e~ after t was slower, s 1 aq max ' and t h i s i s re f l e c t e d i n the 40% larger t ^ value. The presence of another process, B, through which e are being introduced into the aq system over a period of at least 30 nsec confuses any interpretation of t h i s fact. Conceivably the emission from the s c i n t i l l a t o r could be photoexciting the participants i n B as well as hydrated electrons. We believe, therefore, that although the data did not indicate that e were being photolysed during these experiments, the i n i t i a l aq positive slope on the absorption signal prevented any conclusive evidence from being obtained. The photolysis of e~^ remains an open issue. Table 8-11 OD. OD max AOD 3 t nsec max t.nsec K M sec l CONTROL 0.556 0.648 17 28 72 2.5 x 10 10 RED (575 nm) 0.531 0.639 20 27 100 1.7 x 10 10 GREEN (490 nm) 0.530 0.649 22 30 76 3.3 x 10 10 WHITE (375 nm) 0.544 0.636 17 24 70 2.2 x 10 10 Formation and Decay of e i n the Presence of S c i n t i l l a t o r Emission ' aq -165-Chapter IX Discussion of Non-homogeneity If the experimental results from nanosecond pulse r a d i o l y s i s are considered as a whole, three main issues emerge. Non-homogeneity, refle c t e d by the i n i t i a l d i s t r i b u t i o n of spurs i n the system, we have observed and accounted for. But further deviations within the system must be explained. 1. Can the time-dependence of K be attributed to a further microscopic non-homogeneity i n the spurs themselves - or i s i t due to B? 2. What i s process B? 3. How does the delayed formation of e i n B affect conclusions 1 aq drawn from microsecond pulse r a d i o l y s i s and scavenger studies? 4. Is B the simplest explanation? These questions w i l l be dealt with separately. (1) The Time-Dependence of K and Non-Classical Behaviour The sequential conversion from predominantly f i r s t to second -order decay char a c t e r i s t i c s i s , as we pointed out i n Chapter VII, quite contrary to c l a s s i c a l k i n e t i c i n t u i t i o n for species d i s t r i b u t e d randomly i n the medium and capable of taking part i n both types of reaction. The preliminary results on the p l e x i c e l l indicated that immediately after the delivery of a 3 nsec electron pulse the pure water system supported one or more reactions that moved from a f i r s t to second - order decay. This behaviour i n the system was attributed to an i n i t i a l non-homogeneity i n the d i s t r i b u t i o n of the reacting species, e s s e n t i a l l y the spurs, at these high dose-rates. Qualitative calculations estimated the time required for spur-overlap i n regions of -166-d i f f e r e n t dose within the same ir r a d i a t e d volume to be remarkably-close to the t r a n s i t i o n time observed for the change i n k i n e t i c order. It was concluded that the change i n the reaction-rate constant constituted a direct observation of the t r a n s i t i o n from non-homogeneity to homogeneity within the specified area. For th i s reason i t was proposed that a rate parameter K should be substituted for the rate constant u n t i l the homogeneous d i s t r i b u t i o n of species permitted the concentration of the same to be calculated from measured o p t i c a l densities. U n t i l t h i s time any second-order rate constants determined from the o p t i c a l density data would be i n v a l i d because concentration was not a meaningful term. The nature of the reaction giving r i s e to the f i r s t order slope p r i o r to the t r a n s i t i o n period was tent a t i v e l y attributed to the existence of correlated ion-pairs U^0+or an encounter p a i r (103). I f within t h i s b r i e f period of d i f f u s i o n those e~^ reacting were doomed to combine with species from the same spur, each e would have a pro b a b i l i t y of escape or reaction independent of the t o t a l number of spurs i n the system. The average number of species i n a spur i s governed by the radiation. The rate of decay i s therefore proportional to the frequency of pai r encounter i n the spur m u l t i p l i e d by the t o t a l number of pairs i n the system. This would give r i s e to an ov e r a l l f i r s t - order decay. These speculative mechanisms diminished i n importance when another feature of the decay curves was observed. This led us to consider the loss of e as second - order throughout t h i s period. -167-During straightforward non-homogeneity K , the second order rate parameter, should be larger than the true value because the average concentration of reacting species i s lower than the l o c a l number densities i n the reaction zones. I f the reactions are second - order on a microscopic scale, as the system attains true homogeneity the K values should decrease, and the shape of the 1 v t curve resemble (a) OD i n Fig. 9-1 below. In a c t u a l i t y , our experimental K values increased and resembled (b). -168-I f the system were t r u l y homogeneous from time zero, the concentration of species seen by e for second - order reaction w i l l be c. r ' aq — •When non-homogeneity d i s t o r t s the experimentally observed reaction rates, i t i s because the average concentration i s less than the true l o c a l concentration, and thus K i s i n i t i a l l y much bigger than k . I f the opposite i s the case, as i n our data, then i t implies that the average concentration i s higher than the l o c a l concentration. This l o g i c a l conclusion i s not meaningful so we interpret the s i t u a t i o n as follows. Within the spur the e f f e c t i v e number of reacting species seen by e i s lower than the actual value. This may lead to K being less than kg . This curious p o s i t i o n can be explained i f the spur i t s e l f has a non-random d i s t r i b u t i o n of species. We s h a l l refer to t h i s type of structure within the spur as"spur conformation." Let e react, for example, with a r a d i c a l R. I n i t i a l l y three R radicals produced i n the spur are i n such close proximity r e l a t i v e to the separation between e ^ and any one R, that the p r o b a b i l i t y of reaction between e and 3 R w i l l be the same as between e and one R because to aq aq e the eff e c t i v e [R] i s one-third of the r e a l value. As the species e~ aq u 1 r aq and R diffuse into a random pattern so there i s a f i n i t e and equal p r o b a b i l i t y of e reacting with any single R and the t o t a l p r o b a b i l i t y aq of losing e i s much greater. At this point spur conformation i s no aq longer superimposed on the non-homogeneity as a factor determining the value of K. The question i s , does spur conformation exist within t h i s system? The non-homogeneity exhibited on t h i s microscopic scale may be detected through the trend i n the value of K. -169-The detection and i s o l a t i o n of process B i n the early k i n e t i c interplay of t h i s system cast doubts on the v a l i d i t y of the interpretation of the change i n K with time. Certainly the i n i t i a l l y observed f i r s t -order process was no longer a feasible-explanation since that part of the curve was a summation of two processes, A and B. In order to distinguish the duration of non-homogeneity from the duration of curvature, calculations were performed as described i n Chapter VII (section 1) for each set of data. For each decay of e i n a given system, plotted as 1 v t , the was u u period of non-homogeneity^faster (on paper) but the ® ® m a x values were not a time zero parameter. In using OD^, we have been conservative i n estimating By separating the non-homogeneous from homogeneous sectors on the 1 v t graph at the appropriate time, inspection of the slope of the OD decay curve either side of the t r a n s i t i o n point would hopefully y i e l d unambiguous information about the trend i n K as i t approached k^. Only data for which an adequate number of points was available could be used. For example, i f x^ was i n excess of ^ 300 nsec, data was seldom forthcoming because of the interference wave. In general only those curves for which x.. < 150 nsec were useful. At high [e - ] T., was H 6 L aq H short, "^40 nsec and frequently was less than the time taken for the contribution of process B to the system to become i n s i g n i f i c a n t . As a res u l t when x^ < t ^ , the observation of a type (a) decay curve with IC increasing i n time (see Fig. 9-1) could be attributed to the remain-ing curvature effect. Certainly i t i s not possible to separate -170-contributions to the K from "chemical" electrons (those formed slowly i n B) and non-homogeneity, although we know they both are present up to specified times. When T . . >> t . , that i s the curvature i s over within the time H i ' taken for the system to achieve homogeneity, then a separation can be attempted. Unfortunately i n s u f f i c i e n t number of experiments were available that s a t i s f i e d t h i s condition. Coincidental or otherwise, much of the data f e l l into the x^  >' t ^ category and i t was not therefore j u s t i f i a b l e to draw any conclusions from i t . In the instances the condition x^  » t ^ was f u l f i l l e d , the < values looked as though they were decreasing as the curve became f l a t t e r towards the homogeneous sector of the graph, but there were exceptions. We ten t a t i v e l y propose that t h i s evidence even though sparse weakens the case for spur conformation superimposed on non-homogeneity. Whatever process B i s , the super-position of the formation and decay of these "chemical" electrons on the i n i t i a l bimolecular decay of "instantaneously" formed e has probably caused the apparent increase i n K observed p r i o r to the homogeneous stage of the reaction . U n t i l new data can corroborate this proposal or prove i t to be wrong, we do not intend to explain the time-dependence of < i n terms of the non-random d i s t r i b u t i o n of the spur e n t i t i e s as well as non-homogeneous reactions i n the medium before spur-overlap has taken place. Whether the spurs are i n fact s i t e s of true microscopic homogeneity i n which k i n e t i c order has meaning w i l l have to be assessed i n the l i g h t of further experimentation at these very high dose rates. It i s a f t e r a l l conceivable that this non-homogeneity exist through the conversion to e v i a B within the spur, aq r -171-In summary, the system moves from a period of non-homogeneity due to the presence of i s o l a t e d spurs to one i n which the spurs have f u l l y overlapped and concentrations may be used i n c a l c u l a t i n g second - order rate constants. The preliminary observation that the value of < increased i n time implied a further d e v i a t i o n from the c l a s s i c a l k i n e t i c understanding, one that might have been a t t r i b u t e d to spur conformation, the non-random d i s t r i b u t i o n of e with respect to the other species within i n d i v i d u a l spurs. More r e f i n e d experiments detected the delayed formation of e that l a s t e d f o r tens of nanoseconds and as such must aq have been c o n t r i b u t i n g to the measured K value. It i s t e n t a t i v e l y concluded that the trends i n K do not support spur conformation but a d e f i n i t i v e statement on the microscopic s i t u a t i o n must await new evidence, p a r t i c u l a r l y on the mechanisms i n B. t (2) On Understanding Process B (a) An attempt to des c r i b e B through the k i n e t i c s o f e . Let e be produced from a r e a c t i o n i n v o l v i n g an unknown species Xg, k n X D -> e" ' B aq which i s e v e n t u a l l y depleted t o the p o i n t where no more e are formed. / r r aq The r a t e o f appearance o f e i s the only observable i n our system. Knowledge o f the r a t e of disappearance o f Xg i s e s s e n t i a l to solv e k i n the r a t e equation: d J f a q l B = k ^ dt Since t h i s i n f o r m a t i o n i s i n p r i n c i p l e u n a v a i l a b l e from our experiments, l i t t l e progress can be made i n c a l c u l a t i n g any r a t e constants at a l l . Even i f the i n i t i a l c o n c e n t r a t i o n o f X n i s set at [X.J B B o and that o f the product e a f t e r a s p e c i f i e d i n t e r v a l t becomes ^aq^B t* 1 6 e c l u a t : ' - o n n e e d s boundary c o n d i t i o n s ; [Xg] i s not known. d [e" ] D = k ( [ X n ] n - n [e" ] ) L aq B V L B J° L aq JB n dt The q u a n t i t y [e ] cannot be measured at t = m because i n t h i s system aq g the hydrated e l e c t r o n s w i l l be r e a c t i n g once they are formed. At the time the o p t i c a l d e n s i t y of process B reaches a maximum value, the steady s t a t e r e l a t i o n s h i p must hold. Here k^ i s a known d [e" ]_ = k [ X j n - kQ [e~ ] 2 = 0 L aq J B L BJ 8 1 a q J R dt B -173-bimolecular rate constant, but k and (Xg] cannot be determined independ-ently. Since the i n i t i a l slope of the o p t i c a l density calculated for [e ]_, (OD) D jv. time plot for the conversion of X n into [e~ ]_ would at aq B' B' r B L aq B least be proportional to a rate of formation s t a r t i n g from the i n i t i a l -but unknown - concentration [Xg] Q, i t was decided to compare these slopes for d i f f e r e n t experimental conditions. In the presence of other ions, S, the rate of appearance of e can be described by, d J f a 3 l B = k ^ £sim. dt Variations i n the po s i t i v e slope of (OD)g v. t for OH" and R-OH compared to the pure water system would give information on the kind of involvement (m) between X D and the solute S. D The plot of (OD)g v. t i s shown i n Fig. 8-5. The i n i t i a l slopes did not vary s i g n i f i c a n t l y for the pure water and over a 100-fold increase i n [OH ]. This implies that the formation of e"^ from Xg i s zero order i n [OH-]. In contrast the presence of 0.1 M methanol appeared to slow down the rate of formation of e by a factor of 2 on th i s plot. It i s therefore aq 7 F feasible that OH or H radicals are associated with the conversion of X„ D to e , or at least are part of the ov e r a l l mechanisms. The value of m aq' r here i s not zero but undetermined. The time at which a l l the reactants s a t i s f y the equilibrium relationship was t . The interpretation of the graph i n Fig. 8-4 that i l l u s t r a t e s a max  o r & l i n e a r relationship between log [OH ] and t i s that the OH" ions i n some ^ max -174-way control the time at which equilibrium i s reached. The r e a l significance of the log function i s d i f f i c u l t to ascertain. The methanol also caused a s h i f t i n the t value for (0D) D. max ' B It would appear therefore that OH not only performs i t s normal function of scavenging the predator Hj0 + and extending the l i f e t i m e of e - seen i n B but also increases the e f f i c i e n c y of the conversion of aq ' Xg -> e a q * This could take place i f OH suppresses the depletion of X. by H_0+ (not normally a route that would y i e l d e - ). In solutions of B J 3 . ' 1 aq low pH ^ 3, there had been no evidence for any curvature even though OD^  i n acid had been reduced to less than h a l f i t s value i n pure water. In summary, high concentrations of OH increased the y i e l d of e beyond the degree to be expected by simple elimination of predator reactions. At high pH, H + OH" e" , k ' = 2.3 x 10 7 M ' 1 s e c " 1 . (11) aq 11 -2 -At 5 x 10 [OH] t h i s reaction has gone to half-completion by ^ 700 nsec, and may account for the observed increase i n [e 1. 1 L aq J In comparison to the maximum (0D)g value calculated for water, 0.296, _2 that for 5 x 10 M NaOH was 0.738. The value for 0.1 M methanol solutions was 0.275 but i n view of the approximations i n the c a l c u l a t i o n s , i t i s not certain whether t h i s value i s s i g n i f i c a n t l y lower than water or not. The y i e l d of e from B has not been increased i n an obvious way even J aq J though the high concentration of methanol would eliminate loss of e 6 6 aq through predator reactions involving H and OH. In the absence of data from other concentrations of methanol t h i s point cannot be c l a r i f i e d . From these mechanistic clues and general knowledge of the processes -175-believed to occur i n pulse radiolysed aqueous solutions, we w i l l conclude th i s section with some speculations on the i d e n t i t y of Xg. (b) Plausible mechanisms for the reaction X n •> e J B aq The scavenger results would seem to imply that the decay of X D D to e~^ i s not the only fate of the unknown species. So i n the following mechanisms provision has been made for alternative fates of Xg where possible. We begin from the premise Xg i s a primary species i n the spur, 2 B aq -»• product and consider three types of Xg. (i) Xg = (eobaq About 60% of the energy deposited i n th i s system during r a d i o l y s i s w i l l be found as isolat e d spurs. On average, these w i l l contain three ion pairs as primary products. The pos i t i v e H 20 + fragments move r e l a t i v e l y slowly compared to the mobile free electron (e ) which may cover a large distance before being solvated. During the 3 nsec pulse these i s a continuous formation of these ion p a i r s , H-0 -> H„0+ e" ->• H_0+.0H + e" 2 2 3 m and the following events may occur i n the spur: e" e" (21) m aq em + 6aq * ( e2 _ )aq o r ^iPaq' ( 2 2 ) e" + H_0+ -»• Ho0 (23) m 2 2 v ' -176-Reaction (21) must take place within the d i e l e c t r i c relaxation time of water 10 1 1 sees but e m moves exceedingly quickly and randomly before i t s capture. It i s feasible that on encountering a large hole i n the solvent occupied by an aquated electron whose charge i s dispersed over the inner solvent molecules, the now thermalised electron pops i n , reaction (22). Any excess energy would be lost to the solvent molecules around i t . The alternative fate of the mobile electron i s recapture, a geminate recombination given by (23). 2-After the pulse, the number of (e^ ) a q may increase through i n t r a -spur or inter-track reactions of the type, - - 2-e + e -»• (e„ ) (1) aq aq v 2 'aq v ' The L.E.T. of the primary p a r t i c l e s i s r e l a t i v e l y low when they f i r s t penetrate the water and consequently the tracks of the p a r t i c l e s are close to one another and inter-track reactions are more probable than intra-track reactions. The average distance between spurs within an o i n d i v i d u a l track w i l l be ^ 170 A at the beginning. The formation of further dielectron species comes about through intra-spur or inter-spur reactions. The number of e available for reaction (1) w i l l be quickly reduced by predators such as H ^ 0 + , OH which are generated i n the same i n i t i a l y i e l d (from H,,0+) and H. e" + H + H" -> H_. + OH" (2) aq 2 K J e" + OH -y OH" (3) aq e" + H_0+ -»- H-0 -*• H + Ho0 (5) -177-Recent work has indicated that the spur i s a c i d i c i n th i s early period (110) because of the low y i e l d of OH , however the ne u t r a l i s a t i o n reaction rate constant between H.^ 0* and OH i s so fast > 10 1 1 M - 1 s e c - 1 i t may be that OH had disappeared from the reaction zone. _2 The addition of solutes > 10 M to the system would i n t e r f e r e with these reactions, alcohol eliminating reactions (2) and (3) while OH" ions would scavenge H,jO+ i n (5). The net resu l t would be that the l i f e -time of e was extended, hence i t s a v a i l a b i l i t y for reaction (1) 2_ and thus the numbers of (e„ )„_ 'enhanced. In order to observe an increase i n [e~ ] , V N has to dissociate i n aq pure water with a h a l f l i f e of ^  20 nsec. (e2~->aq 2 eaq ? H 1' tl/2 * 2 0 n s e c W However we have evidence for a pH dependence i n which the y i e l d of e i s increased i n alkalin e solution. I f the additional e - are not aq aq sole l y due to H + OH -> e" (11) aq aq v J then the diss o c i a t i o n of Xg must be assisted d i r e c t l y or i n d i r e c t l y by OH . Either the dielectron can be attacked by HjO + to give a lower y i e l d of e aq» which OH would prevent, 2 H 3 0 + (e- ) •* H-0 + e" v 2 'aq 3 aq or i f a second equilibrium exists that involves hydride ions, the 0H _ can influence t h i s too. -178-2-(e 0 ) »• 2 e (a) v 2 'aq aq v (b) HO | t OH H20 H H 2 An increase i n [OH ] would s h i f t the equilibrium (b) i n favour of the dielectron and thus a s s i s t d i s s o c i a t i o n through (a). The y i e l d of H 2 from the completion of the pathway (b) might be the "un-scavengable" y i e l d of H 2 within the spur currently attributed to no one single cause. In t h i s way the rate of formation of i s always determined by pathway (a) regardless of [0H~]. The main objection to t h i s otherwise compatible mechanism i s directed at the h a l f - l i f e of the species i n pure water. There i s no evidence i n r a d i o l y s i s studies for or against the rate dissociation of 2-(e 2 ) aq, but f l a s h photolysis studies at high pH reported i n t h i s thesis would not support that l i f e t i m e . The behaviour of mobile electrons has previously been discussed i n terms of "dry electrons" (88) i n a model for the r a d i o l y s i s of water. Perhaps the dielectron species i n i t i a l l y formed by e" + e" -* (e2")„n (22) m aq v 2 'aq »• > has a shorter l i f e t i m e than the dielectron produced i n a chemical reaction (1), already having formed i t s solvation sheath. There can be no normal b a r r i e r to reaction i n (22), no solvent hole to enlargen ( i f pre-existing traps e x i s t , (see Chapter II)) no energy involved i n reorganising a solvation sheath. The dry electrons are purported to t r a v e l at a thermal -179-veloc i t y of 10 cm sec which would give them a long mean free path before capture. ( i i ) X B = H20* In t h i s mechanism two excited water molecules react to give an ion-pair and neutral molecule within the spur. Ho0* + Ho0* •* [H-0. H o0 +.e] -> OH. H,0+. e" 2 2 L 2 2 J 3 a q The excited water molecule i s believed to have a l i f e t i m e of -9 ^ 10 sec and has been postulated as a precursor to molecular hydrogen i n order to explain anomalies i n the molecular yields as determined through scavenger studies at high concentration (15 (b)). As two excited water molecules interact through c o l l i s i o n , one molecule i s deactivated and the other ionised within the encounter "cage" This means that only a f r a c t i o n of the species w i l l escape geminate recombination,be solvated and diffuse away. The remainder of the species w i l l reform neutral water molecules ( any excess energy i s small and can be lost i n v i b r a t i o n a l modes) or a r a d i c a l pair [H^O.OH]. The l i f e t i m e of t h i s pair w i l l depend on whether the species diffuse rapidly away and react as radicals within the volume of the spur or i n inter-track reactions,(since there i s no coulombic f i e l d that can influence t h e i r behaviour unless H^ O ->• H 30 +.e ),or whether they experience germinate recombination to give water. We w i l l term recombination within the i n i t i "cage" as primary geminate recombination, and that taking place outside the cage after d i f f u s i o n as a secondary event. In the presence of OH , the secondary recombination of H,0+ and -180--2 -another species w i l l be reduced and at > 10 M [OH ] the y i e l d of e contains those electrons that have escaped primary recombination, i . e . the cage. The free electrons w i l l t r a v e l some distance before becoming solvated and i n the presence of high [OH ] and (R-OH] may only disappear e f f e c t i v e l y with another e a q - A t high pH any H^ O w i l l be converted over several hundred nsec to e a q , thus the product of primary recombination between H , 0 + and e w i l l ultimately y i e l d e anyway. The R-OH at 0 . 1 M concentration w i l l prevent the OH and H^ O from p a r t i c i p a t i n g i n further recombination reactions, but w i l l not release the f r a c t i o n of e~^ already lost to H^0 + within the "cage". At high [R-OH] therefore only the l i f e t i m e of e would be increased, not the t o t a l numbers, aq The behaviour of Xg i n generating e can be accounted for i n general terms by the reaction of two excited water molecules. A few years * ago i t was proposed that ^ 0 reacted with oxygen i n oxygenated solutions to explain the increase i n e y i e l d that was observed. In the absence aq * 7 6 of oxygen, ' ^ 0 " survived 10 to 10 sees before disappearing as H^ O and OH ( 1 1 1 ) . There i s s t i l l no direct evidence on the role of excited water i n aqueous systems to t h i s date. ( i i i ) Xg = [ H 3 0 + . e T 0 H ] Although at f i r s t glance the above i d e n t i t y resembles the caged species i n the previous discussion, t h i s time i t i s formed as a primary r a d i o l y t i c product i n water, not a secondary from the l a t e r reactions of H 2 0 * . It i s a "spur" of one i o n i s a t i o n , sometimes regarded as an is o l a t e d pair. H 2 0 + [ H 3 0 + . e " . 0 H ] -181-either geminate recombination takes place within the cage or the species diffuse away. At any given time during the d i f f u s i o n period there w i l l be a f i n i t e p r o b a b i l i t y of secondary geminate recombination according to (a) H_0+ e" OH or (b) H_ot e" OH 3 aq aq 3 aq aq H30 OH" t h e i r r e l a t i v e s p a t i a l d i s t r i b u t i o n . As i n the e a r l i e r mechanisms, high [OH ] i n h i b i t s primary geminate recombination within the cage and removes H^ O* from i t s predatory role outside the cage. In the presence of R-OH the radicals OH, H^ O are removed as and ^ 0 from the areas where both primary and secondary recombination are the most probable events. In s i t u a t i o n (a) e" that aq have already been l o s t to H.^ 0* w i l l not be recovered with alcohol as a scavenger but a high [OH-] can convert H^ O into e~ . Therefore we would expect to see an increase i n y i e l d of e" at high pH. In s i t u a t i o n (b) aq no increase i n [e ] should occur through reaction, only a decrease i n aq the numbers of e i n i t i a l l y l o s t . aq J I f n e u t r a l i s a t i o n occurs i n the primary "cage" i n the pure water system, H^ O could s t i l l ionise to produce e &^ i n the way that Magee o r i g i n a l l y envisaged (16). -182-The OH- would f a c i l i t a t e the production of e~ , the R-OH i n h i b i t the ionisa t i o n i f the concentration were high enough. The idea of a caged species such as (H^C^.e.OH] formed i n a primary r a d i o l y t i c act i s perhaps only another way of looking at the s p a t i a l d i s t r i b u t i o n of species within the spur. Nevertheless the solute effects that have been recorded i n t h i s work may be explained i n these terms, OH increasing not only the primary y i e l d but the l i f e t i m e of e once aq i t i s a secondary species, R-OH only influencing the l a t t e r by eliminating the scavenging of e~ by rad i c a l s . -183-(.3) Implications from these Studies and Results i n the Current Literature The significance of the observations i n th i s thesis w i l l be discussed b r i e f l y i n reference to some published work on pulse r a d i o l y s i s and steady-state scavenger studies. There are two immediate areas of int e r e s t , (a) the y i e l d of hydrated electrons i n r e l a t i o n to the techniques employed and (b) the c o n f l i c t i n g evidence on the existence of spurs and the duration of intra-spur reactions from experiments on a comparable time scale to.our own. (a) Comments on the reported yi e l d s of e The dose-rates i n our experiments are among the highest employed i n p.r. studies and accordingly the i n i t i a l [e ] i s several orders of aq -4 magnitude larger than commonly achieved, being .< 5 x 10 M. We observe some unusal features i n the i n i t i a l behaviour of the pulse radiolysed system namely an increase i n the [e - ] af t e r a 3 nsec pulse of aq 26 — 1 1 0.5 MeV electrons (5 x 10 eV gm sec ) followed by an appreciable loss of e within the next 150 nsec. These points raise three questions that are d i r e c t l y relevant to the measurements of hydrated electron. y i e l d s . F i r s t , are we aware of a l l the processes giving r i s e to e , aq secondly what affect does the [ e a^]g have on the yields measured through microsecond p.r. or steady state scavenger techniques, and l a s t l y how does the rapid i n i t i a l loss of e affect conclusions from the very same experiments? In t h i s work the y i e l d of e as a G-value was not determined J aq because the absolute value of the dose i n diff e r e n t regions of the c e l l had not been measured. Instead we have equated the i n i t i a l o p t i c a l density to a y i e l d of e for the purpose of discussion. -184-I f , as we have proposed, there i s another source of hydrated electrons formed from Xg a r a d i o l y t i c precursor whose y i e l d i s proportional to the number of e "instantaneously" formed these extra electrons w i l l be detected by microsecond p.r. as part of the primary y i e l d . On the other hand a nanosecond p.r. measurement would be completed before [©aq]g had reached i t s maximum value. I f t h i s were the case, the microsecond p.r. value for Gfe" ) should be the aq larger of the two. Likewise i n the steady-state r a d i o l y s i s experiment, with scavengers at any concentration the electrons from X_ w i l l be D removed together with those formed i n the usual sense unless X_ i t s e l f were attacked by the scavenger. I f however the formation of Xg were peculiar to the very high dose-rate s i t u a t i o n , these yi e l d s would not be suspect. Obviously there i s a need for a nanosecond y i e l d of e~ , so f a r not reported. Aside from these "chemical" electrons, the system s t i l l suffers an appreciable loss of e~^ within the i n i t i a l 100 to 150 nsec a f t e r the pulse, a loss that would not be observed by microsecond p.r. nor f u l l y _3 by scavenger studies < 10 M. This now implies that the nanosecond p.r. y i e l d of e~^ would be higher than i n the other systems. In view of these two opposing effects to what do the e yi e l d s that are reported aq actually refer? The s i t u a t i o n i s not simple, because i n the very high dose rates we presume the i n i t i a l loss of e~ to be partl y inter-spur and p a r t l y intra-track. Most of the information concerning yields of r a d i o l y t i c a l l y -185-produced species i n water comes from experiments using low L.E.T. radiation. As the mean L.E.T. increases so does the primary (100 eV) y i e l d for and to a lesser extent. The y i e l d of e~ and OH decreases, H marginally so. There i s a noticeable v a r i a t i o n i n the measured yields of e for both neutral and al k a l i n e solutions using the microsecond p.r. and steady-state scavenging techniques (15 (b) p. 27, (112)). G(e a^) varies between 2.3 and 2.9 at pH 7. A nanosecond y i e l d would hopefully throw some l i g h t on these discrepancies. Studies at high scavenger concentrations have already indicated that the y i e l d -9 of e with lifetimes i n excess of 2 x 10 sec may be much greater than the accepted value of 2.6, possibly 4.5, (106). (b) Spurs and non-homogeneity It i s our contention that the postulated existence of spurs i s a prima faci e case for non-homogeneity. The experimental evidence i s c o n f l i c t i n g . Calculations based on the "prescribed d i f f u s i o n " model of Samuel and Magee (5) have predicted that intra-spur reactions w i l l be almost complete after 2 nsec (14) p. 85, (113)). Following an attempt to look for spurs at very short times i n aqueous solution (114) i t was concluded that e was formed within 0.5 nsec at J aq 'v- 5 x 10 ^  M and that there was no ind i c a t i o n of s i g n i f i c a n t intra-spur reaction within t h i s period. In fact there was less than 2% loss of signal i n 50 nsec a f t e r the 3 nsec pulse. More recently experiments u t i l i s i n g a 12 nsec pulse with [e ] at ^  2 x 10 ^  M reported a very aq fast i n i t i a l loss (A-15%) of e over the f i r s t 50 nsec a f t e r which a aq much slower decay lasted for microseconds, (115). The rapid decay could -186-_2 be e n t i r e l y eliminated i n the presence of 1 M ethanol +10 M NaOH. When either of these solutes were used i n d i v i d u a l l y , the decay was reduced by only 50%. It was concluded that t h i s was direct evidence for intra-spur reactions of e with H^ O* and OH, and that the system did not a t t a i n homogeneity u n t i l the slow part of the decay became dominant. This does not appear consistent with the prescribed d i f f u s i o n model. Modifications to the same (116) however have successfully accounted for these intra-spur reactions which are probably not complete after 10 nsec, and at the same time elegantly included other anomalies previously countered against the spur d i f f u s i o n theory. Nevertheless, i t appears as though simple mechanisms of io n i s a t i o n do not adequately account for a l l that i s observed i n these systems and such a model s t i l l requires i t s defenders. Our own experiments -4 -at 2 x 10 M [e 1 see as much as 40 to 50% loss of e within L aq J aq 100 nsec ( d i f f i c u l t to establish a figure at 50 nsec because of the formation of e from X g ) through intra-spur and ultimately inter-track reactions. This loss i s greatly reduced i n the presence of the H^ O* . and OH scavengers as noted e a r l i e r . At high dose rates intra-spur reactions are not complete f o r tens of nanoseconds. The exceptionally high value of the second-order rate constant for reaction (1) e" + e" -»• H_ + 2 OH (1) aq aq 2 aq v ' reported as 3.2 x 1 0 1 1 M _ 1 s e c - 1 at high dose rates (117, 118) 24 -1 -1 (10 eV gm sec ) probably stems from the use of mean bulk concentrations derived from OD values during the period of non-homogeneity the consequences of which were discussed at the beginning of t h i s chapter. -187-Th e inter-spur distance at our very high dose rates i s about 200 A 3 and comparable to the separation observed along the track of a H £ p a r t i c l e o _ whose mean L.E.T. i s ^ 0.5 eV per A. The G(e ) i n the l a t t e r instance i s r aq ^ 1.8 (( 3 ) p. 121). The timescale and scavenger concentrations used i n determining t h i s value would not detect the i n i t i a l loss of e whereas nanosecond studies can. Possibly the v a r i a t i o n i n G(e~ ), and thus G(OH) too, with L.E.T. i s a consequence of the l i m i t s imposed by the techniques employed, not of the system i t s e l f . (4)' An Alternative Explanation of the Curvature In discussing the o r i g i n of the curvature on the absorption signal and the nature of the species involved we concluded that the evidence presented was consistent with the delayed formation of "chemical" electrons. The i d e n t i f i c a t i o n of the absorbing species as e i s convincing, but aq there could be an alternative interpretation f o r t h e i r observed behaviour. The proposals that follow are merely outlines of two independent issues that happen to be i n t e r r e l a t e d i n t h i s system. (a) The e f f e c t i v e extinction c o e f f i c i e n t of e may be a time-dependent parameter as a res u l t of l o c a l e l e c t r o s t a t i c interactions. (b) The period of non-homogeneity i n the system dictates the time over which these interactions are most important. For the case i n (a) we envisage two possible situations i n which the effective extinction c o e f f i c i e n t may be altered. F i r s t the e l e c t r o s t a t i c i nteraction between two e~ i n close proximity to one another may cause a d i s t o r t i o n i n the d i s t r i b u t i o n of charge of either species, hence affect the wavefunctions of the ground and upper states i n r e l a t i o n to these of the "free" e , according to the degree of perturbation. -188-These changes would be reflected i n the t r a n s i t i o n moment int e g r a l and thus the p r o b a b i l i t y of absorption, or the extinction c o e f f i c i e n t i n experimental terms. The second effect i s also a proximity issue, i n that the i n i t i a l separation of hydrated electrons within the spurs i s extremely small compared to the wavelength of the photon, and therefore they appear as a cluster rather than indi v i d u a l species. I f we refer to a, the cross-section of absorption as the e f f e c t i v e target area seen by the photon, a cluster of three e n w i l l have a smaller cross-section than the sum aR of three independent a values. The hydrated electrons are screening each other. A consideration of the e f f e c t i v e l i f e t i m e of these clusters takes us to case (b). The photons w i l l not see the true numbers of e~ u n t i l the clusters have broken up, that i s u n t i l homogeneity has been attained. As d i f f u s i o n gradually leads to a random d i s t r i b u t i o n of e ^ j . s o the t o t a l cross-section for absorption increases. This i s manifested as an increase i n OD during t h i s period. t In any case the p r o b a b i l i t y of an e absorbing a photon i s very aq 7 2 -1 small when the extinction c o e f f i c i e n t i s 1.2 x 10 cm mole . I f 50% > of the available l i g h t i n t e n s i t y at 632.8 nm i s absorbed i t can be -4 shown that at 2 x 10 M [e ] (a t y p i c a l value for t h i s system) only aq g 1 i n 10 molecules absorb a photon i n a nanosecond. During the period of non-homogeneity only 1 i n 10^ e w i l l absorb a photon. The effect of the scavengers i n t h i s explanation of the curvature would be to extend the l i f e t i m e of e by removing electron predators, and thus the i n t e r v a l of time i n which non-homogeneity would be important i s - i n -e f f e c t i v e l y increased. To date no one has published the absorption spectrum of e" at these high concentrations. I f these proposals are v a l i d then the spectrum w i l l either decrease i n i n t e n s i t y as a res u l t of screening eff e c t s , or the A s h i f t as a result of the modified t r a n s i t i o n moment ' max i n t e g r a l . Attempts to examine such p o s s i b i l i t i e s are currently under way, (119). -190-Chapter X RESULTS FROM THE FLASH PHOTOLYSIS EXPERIMENTS In these experiments solutions containing I or OH were f l a s h photolysed i n order to generate e f o r further studies. I" + hv I + e~ aq OH" + h\> -»• OH + e" aq (1) Preliminary Results from the Flash Photolysis of KI Solutions Deaerated mM KI solutions were subjected to an intense 25 ysec f l a s h , containing l i g h t of wavelengths from about 190 nm to the near-I-ft., from the main lamp discharged at 6 kV, and the spectrum of the transient species taken by d i r e c t i n g the 5 ysec v i s i b l e f l a s h from a spectroflash through the solu t i o n and onto the s l i t s of a medium Hilger Spectrograph. The spectroflash was always discharged at 9 kV, and could be triggered at any predetermined delay time from the main or a u x i l i a r y lamp. Spectra were recorded on HP3 plates. A broad absorption band was detected from 350 nm to 630 nm. Spectra taken at i n t e r v a l s up to 1 msec showed that the absorption at the longer wavelength region was rap i d l y disappearing while that i n the 350 to 450 nm range slowly increased. Since the photoionisation of I would produce I and e~q, the iodine would probably form ionic and molecular products and the build-up i n absorption over milliseconds was attr i b u t e d to t h i s . The concentration of KI was reduced and a scavenger added to eliminate I, since the spectrum of e would be p a r t i a l l y masked by the presence of absorption due to I-. -191-When 0.2 M methanol was added to 2.5 x 10 KI solutions and flashed the spectrum s t i l l indicated the presence of \^  spectro-graph^ analysis of the region above 500 nm on the plates (after correction for plate s e n s i t i v i t y ) revealed part of an absorption band that was similar to the known spectrum of e - . It was decided to r aq complete the preliminary studies using the technique of k i n e t i c laser photometry i n place of the spectrograph. The He-Ne laser l i n e at 632.8 nm was conveniently close to the absorption maximum of e but could not be absorbed by 1^- Lower fl a s h energies were to be t r i e d as w e l l . -4 A solution of 2.5 x 10 M KI containing 0.2 M methanol and deoxygenated for several hours p r i o r to use was subjected to a 3 kV fl a s h from the main lamp. A transient absorption signal peaking at ^ 20 ysec and with a h a l f - l i f e i \ j 2 > 180 ysec was recorded. On multiple flashing the h a l f - l i f e decreased considerably - after 2 flashes T l / 2 > 100 ysec. When pure deoxygenated water or highly a c i d i f i e d KI solution was flashed no signal could be detected on the highest s e n s i t i v i t y of the oscilloscope. On introducing 0^ or N20 into the experimental solution T2/2 m a r k e d l y decreased. From the known absorbing species present and the measured rate of decay of the signal the transient absorption at o -632.8 A was attributed to e aq The laser beam had been directed down the centre of the reaction vessel for those experiments. Now the beam was directed down the side of the vessel nearest the main lamp. An increase i n the absorption was observed during the 3.5 kV main fl a s h from which we estimated 'v 4 x 10 ^  M e were formed following the photoionisation of I . -192-The solutions were next subjected to a double f l a s h , the main lamp discharged at 3.5 kV and the a u x i l i a r y at 6.5 kV. Pyrex f i l t e r s , 2 mm thick, were placed along the side of the a u x i l i a r y lamp such that only l i g h t > 270 nm would reach the reaction vessel. The second lamp was f i r e d i n i t i a l l y at delay times from 2 ysec to 62 ysec a f t e r the main f l a s h . There appeared to be some delayed absorption p a r t l y hidden i n the main absorption signal when two lamps were f i r e d . The fl a s h from the a u x i l i a r y lamp alone did not give r i s e to any absorbing species. When an a c i d i f i e d Kl/methanol solution was double-flashed no signals were observed at a l l . A solution of KI was prepared without the methanol, but the absence of methanol merely altered the duration of the two absorption signals. When glucose was substituted for methanol s i m i l a r effects were noted, and on f i r i n g the a u x i l i a r y lamp some 100 ysec after the main lamp a d i s t i n c t shoulder appeared on the decay of the signal induced by the main f l a s h . Since there was the p o s s i b i l i t y of the a u x i l i a r y lamp i t s e l f photoionising I , or 1^, we decided to use another solute. At main flash energies > 5 kV we also had observed a plateau on the decay- of the signal a f t e r about 100 ysec which lasted for milliseconds. It was important to f i n d out i f both these effects were peculiar to the KI system or not. _3 A solution of 10 M KOH was made up from 3D water, deoxygenated and flowed into the reaction vessel. Since the ef f i c i e n c y of the photoionisation of OH was expected to be le s s , the solution was double-flashed with 5 kV and 7 kV main and a u x i l i a r y f l a s h energies respectively. Both effects were observed. When the second f l a s h was -193-delayed by 100 ysec there was no doubt about the appearance of a second absorption within the decay of the f i r s t . When methanol was added to the KOH solution, no difference was made to the height of the signal but T ^/l i - n c r e a s e d - Th e a u x i l i a r y f l a s h alone produced no signal - we detected neither absorption from the l i q u i d nor scattered l i g h t which might have been picked up by the photomultiplier. With due precautions scattered l i g h t was not a problem and we were s a t i s f i e d that the pyrex f i l t e r s eliminated the u.v. l i g h t from the a u x i l i a r y f l a s h entering the reaction vessel. The base l i n e on the oscilloscope trace was quite f l a t , the high frequency laser r i p p l e constituting a " l i n e width" of ^ 10 mV compared to the average absorption signal of 400 to 500 mV. Having observed what appeared to be the absorption of an intermediate product i n the pure KOH system i t was desirable to eliminate any other known species present as an impurity. The CO^ r a d i c a l - i o n has a much lower extinction c o e f f i c i e n t for radiation of 632.8 nm than does e aq to which we had attributed the f i r s t peak, but nevertheless i t remained a p o s s i b i l i t y . Solutions of deoxygenated mM Ba (0H) 2, K0H/Ba(0H)2 and KI/Ba(0H) 2, with and without methanol, were double-flashed. It was evident that CO^ was not contributing to the observed signals. In order to study the dependence, of the i n t e n s i t y and l i f e t i m e of the second peak on the f l a s h energies and solute concentration, i t was essential to have a system that was as chemically "clean" as possible. To achieve t h i s we moved to a OH /H2 system with which the main part of the fl a s h photolysis studies were, subsequently done. -194-(2) Results from the flash photolysis of hydrogen saturated a l k a l i n e  solutions As we anticipated there was a non-uniform absorption of the exciting l i g h t across the c e l l at high [OH ]. When the cross-section of the reaction vessel was scanned by the laser beam to investigate the gradient i n [e~ ], aq i t appeared as though the o p t i c a l density i n the centre of the vessel was approximately a factor of 4 less than that observed near the wall on the main lamp side. V i r t u a l l y no absorption could be detected along the wall on the a u x i l i a r y lamp side when the main lamp was f i r e d . The following optimum operating conditions were selected. The main and a u x i l i a r y lamps were to be discharged at 4.5 kV and 6.5 kV respectively and the back pressure of H 2 to be maintained at 800 mm Hg (equivalent to -4 - -4 7 x 10 M H 2 i n solution). The OH concentration was between 10 to _3 10 M for a l l experiments and the laser beam monitored those events occurring 2 to 3 mm from the wall of the reaction vessel on the main lamp side. Care was taken to eliminate any stray u.v. i r r a d i a t i o n from reaching the vessel i n the second f l a s h . Some t y p i c a l oscilloscope traces showing the absorption induced by the main f l a s h , the second absorption induced by the a u x i l i a r y f l a s h after the main lamp discharge and the same experiment with oxygen i n the system are seen i n Fig. 10-1 (a) and (b). The absorption signal following a 4.5 kV main fl a s h decreased by a factor of 2 within ^ 200 usee. With a 1 msec delay between the two lamps, the height of the second absorption signal was ^ 20% the height of the f i r s t . The discharge of the a u x i l i a r y lamp occurred at 6.5 kV. -195-In other words although the absorption signal attributed to e had disappeared within 600 ysec the a u x i l i a r y f l a s h was photodissociating a product of the reactions i n solution present i n appreciable amounts 1 msec l a t e r . The height of the second absorption peak was measured as a function of the delay time between the lamps. Even after several hundred milliseconds the second peak could be detected a l b e i t diminished i n i n t e n s i t y . The slopes of the f i r s t and second absorption signals on the oscilloscope traces were analysed. The i n i t i a l 20 to 30 ysec of the signal refers to a build-up of e and any data taken during the i n t e r v a l of the f l a s h i s not useful for the ca l c u l a t i o n of decay rate constants. The o p t i c a l densities were calculated therefore as a function of time for the period 40 ysec to 400 ysec after the f l a s h . The data was t r i e d for f i r s t and second - order treatments through adaption of the computer programme described i n chapter VIII. The k i n e t i c analysis was not simple over the whole range of data. The i n i t i a l 150 ysec of the decay of the absorption induced by the main f l a s h c l e a r l y gave second - order plots i n a l l instances with an average 9 - 1 -1 rate constant of 6 x 10 M sec . The data taken over longer inte r v a l s exhibited predominantly second - order c h a r a c t e r i s t i c s , but the rate constants varied 0.5 x l O 1 ^ M 1 sec 1 < k < 2 x l O 1 ^ M 1 sec A s i m i l a r k i n e t i c analysis of the decay curves was performed f o r the second peak on those traces taken when the delay on the a u x i l i a r y lamp had been greater than 600 ysec. At t h i s time no absorption from the f i r s t j -196-peak remained to interf e r e with the.second sign a l . The species decayed through a second - order reaction whose rate constant was determined 9 -1 -1 to be ^  5 x 10 M sec . The data f i t t e d reasonably well to the second -order treatments. Some of these k i n e t i c plots are shown i n Fig. 10-2. When the height of the second peak generated by an a u x i l i a r y flash energy of 6.5 kV was plotted as a function of delay time between the lamps, no conclusive k i n e t i c statement could be made regarding the rate of loss of the species photolysed i n the second f l a s h . The height of the peak decreased i n time and the h a l f - l i f e of th i s species appeared to be between 50 and 100 milliseconds at pH 11. During these experiments the photolysing f l a s h had been r e s t r i c t e d to wavelengths > 280 nm and only 1.0% l i g h t below 300 nm was i n fact transmitted. In order to investigate the spectral properties of t h i s intermediate species, f i l t e r s were selected to transmit l i g h t X > 310 nm (0-54), X > 500 nm (3-70) and 680 nm < X > 1250 (7-54). The normal precautions were taken to prevent stxqr l i g h t from reaching the reaction vessel. The second absorption signal was observed i n a l l instances but i t was reduced by a factor of 5 by the 7-54 f i l t e r . When a length of aluminum f o i l was substituted for t h i s f i l t e r placed between the a u x i l i a r y lamp and the reaction vessel, and both lamps were f i r e d , only the f i r s t absorption signal was seen. Since water absorbs strongly i n the near infra-red region, we could only conclude that the species absorbs l i g h t from ^ 680 nm possibly i n a broad band covering the near infra-red to at least 1250 nm. Even i f the species absorbed feebly at 632.8 nm we would not necessarily detect t h i s . -197-In the presence of ^  2 x 10 M 0 2, neither absorption peak was observed. When the 0 2 concentration was reduced to ^  5 x 10 ^  M the f i r s t absorption signal was detected following the main fl a s h but i t disappeared rapidly i n comparison to i t s l i f e t i m e i n the pure, hydrogenated solutions. The rate constant was determined from a (pseudo) f i r s t - o r d e r plot and the second - order rate constant calculated to have a value of 2.5 x l O 1 ^ M 1 s This i s i n agreement with the published rate constant of 2.32 x l O 1 ^ M 1 se for reaction between e and 0.,. The second peak however could not be aq observed i n the presence of 5 x 10~^ M 0 2 (see Fig. 10-i(b)) suggesting that the hydrated electrons formed i n the main fl a s h were the precursor to th i s intermediate species and i n reacting competitively with 0 2 had decayed through mechanisms that did not generate the intermediate i n s u f f i c i e n t quantity to be e f f i c i e n t l y photolysed. The dependence of the maximum o p t i c a l density of the second peak with the energy of the a u x i l i a r y f l a s h was next investigated. When the discharge energy of the main flash was below 5 kV, i t was not possible to detect the second peak. As the input energy into the system increased so the height of the second peak increased i n r e l a t i o n to the f i r s t . We preferred not f i r e the a u x i l i a r y lamp at energies > 8 kV since t h i s shortened the l i f e of the photolysis lamps but at this energy the second peak was ^  30% of the absorption peak produced by the 3.5 kV main f l a s h . We next proceeded to examine the spectrum of each absorption peak using a spectroflash and spectrograph as outlined i n chapter V. The spectrum of the species giving r i s e to the f i r s t peak was taken on HP3 and 1 N plates on the Ja r r e l l - A s h with a resolution of 0.5 mm per mm. F i l t e r s to prevent'wavelengths < 500 nm from entering the s l i t s were -198-positioned on the o p t i c a l axis d i r e c t l y i n front of the s l i t s . Test plates showed that the degree of scattered l i g h t from the lamps reaching the plates was less than 10% of the in t e n s i t y of the monitoring spectro-flash. Control spectra were taken a f t e r successive flashes to provide a standard of o p t i c a l density over the plate i n the absence of any absorbing species. The experimental solutions were replenished for each f l a s h or group of flashes, but since we suspected the presence of a long l i v e d intermediate absorbing i n the near infra-red region, care had to be taken over using preflashed solutions to take control spectra. In l a t e r attempts observe spectra of the species giving r i s e to the f i r s t and second absorption peak, and of the intermediate, pure 3 D hydrogen saturated water was used as a control solution. This was to avoid any loss of inte n s i t y of the scattered l i g h t from self-absorption by e aq« The spectrum from 500 to 840 nm of the species giving r i s e to the f i r s t absorption signal i n the k i n e t i c laser photometry experiments i s shown i n Fig. 10-3. The spectrum of e (34) i s included for comparison. aq The r a t i o of i n t e n s i t i e s at two wavelengths matched the value obtained by other workers for the r a t i o of (G ( e aq) x extinction c o e f f i c i e n t ) at the same wavelengths. Attempts to observe the spectrum of the species giving r i s e to the second peak were less successful, p a r t l y because of the spread of what l i t t l e density there was across the plates i n the Jarrell-Ash high resolution spectrograph. The other problem arose because of the -199-non-uniformity of absorption of the main (and thus a u x i l i a r y ) flash across the reaction vessel. The highest [e ] was at the wall of the b aq vessel, and e a r l i e r studies had indicated t h i s region was also where the highest peak could be observed for the second s i g n a l . The spectro-flash however monitored mainly those events occurring i n the centre of the vessel. A modified arrangement was constructed such that the p a r a l l e l beam from the spectroflash would scan the whole of the reaction vessel and f a l l onto the s l i t s of the Bass-Kessler spectrograph. The 0H~ -4 concentration was reduced to 5 x 10 M. In t h i s way the concentration gradient of e across the c e l l was reduced s l i g h t l y , the monitoring f l a s h would sweep through the maximum volume of experimental solution and the absorption spectrum of any species would be taken under low resolution conditions, that i s the t o t a l v a r i a t i o n i n density would be spread over a r e l a t i v e l y small area. In view of the probable spectral properties of the intermediate species, i t was decided to use infra-red f i l m and study the region 600 nm to 850 nm. Consequently p a r t i c u l a r attention had to be paid to screening the f i l m from radiation < 600 nm because of the r e l a t i v e high s e n s i t i v i t y of the f i l m to lower wavelengths. A small lens positioned i n front of another spectroflash, whose emission reasonably simulated a point source,converted the l i g h t into an almost p a r a l l e l beam. There was s t i l l the chance that l i g h t scattered or re f l e c t e d at the walls of the vessel would in t e r f e r e with measurements and so the outer perimeter of the windows on the reaction vessel were covered with black tape, reducing the eff e c t i v e diameter of the monitoring f l a s h from 10 mm to 8 mm. The spectra were recorded on 35 mm f i l m . -200-Spectra taken under these conditions revealed the same information concerning the i d e n t i t y of the f i r s t peak, but we were unable to s a t i s f a c t o r i l y interpret the variations i n density either for spectra taken after the disappearance of e and p r i o r to the second f l a s h , or spectra of the second absorbing species. Scattered l i g h t severely hampered the detection of variations i n f i l m density between the control (water) and experimental solutions i n the absence of a strong absorption. Since the OD of e i n the f i r s t peak determined through laser photometry was only ^0.7 and that observed for the second ^ 0.15 at the wall of the vessel, the second absorption spectrum might be too weak to see at the levels of s e n s i t i v i t y employed. One other effect had been noticed during the e a r l i e r experiments and t h i s imposed a l i m i t on the main fl a s h energy to be used while taking these spectra. When the-'main fl a s h was discharged at energies of 6 kV and above the absorption signal did not f a l l to zero after a few hundred usee but reached a plateau that lasted for milliseconds. On f i r i n g the a u x i l i a r y lamp during this period, the second peak appeared superimposed over the plateau. (See Fig. 10-1 (c)). Therefore i t was not feasible to raise the energy of the main fl a s h i n order to produce more e and consequently a higher second peak too because absorption from the plateau species would mask any small contribution from the second species generated a f t e r the a u x i l i a r y f l a s h . -201-"~5 "oTI 0T2 Time (msec) u.v. fl a s h 100-J T 1st f l a s h u.v. 0 6 1.2 I 2nd fl a s h (no u.v.) Time (msec) Figure 10- 1. 100 1st flash u.v. 2nd fl a s h (no u.v.) Some Oscilloscope Traces from the Double-Flash Photolysis Experiments (See text) -202-0 100 0 50 Figure 10-2. 200 300 Time (ysec) 400 500 Decay of e i n presence of ' aq ^ following main fl a s h Decay of e following main flash 100 Time (usee) 'lots for th the Main and A u x i l i a r y Flashes. 200 250 300 Second Order Plots the Disappearance of e Following r r aq 0.9 0.8 o Q O > 0) OS 0.4 0. 3 0.2 I-o G O This work Gottschall and Hart (34) O o 1.0 560 600 650 700 750 800 840 X (nm) Figure 10-3. Spectrum of the Hydrated Electron -20 4-Chapter XI DISCUSSION OF THE PHOTOLYSIS EXPERIMENTS The implications of the results of the photolysis experiments performed with the double-flash photolysis equipment and pulse r a d i o l y s i s c e l l w i l l be discussed i n the context of two issues, (1) the species and mechanisms involved i n the fl a s h photolysis of strongly a l k a l i n e solutions and (2) the photolysis of the hydrated electron to an excited state. These are not unrelated i n an experimental sense, but as became clear during the course of t h i s work the events i n (1) impose a l i m i t on the observations of (2). (1) The Double-flash photolysis of aqueous solutions The f i r s t system we investigated generated e through the photo-aq i o n i s a t i o n of I , I~ + hv -* I + e" <f>oc. % 0.24 (71, p. 191) aq aq aq Y254 ' r Unfortunately further reactions occur re s u l t i n g i n a build-up of mo lecular 1^ a n ^ io n i c products such as I + I" ' + l~ aq aq 2 aq I + l~ •+ I~ aq 2 aq 3 aq Since \^ has absorption maxima at 290 nm, 350 nm and 750 nm, and 1^ absorbs strongly at 287 nm and 353 nm, the presence of these speci not only p a r t i a l l y masked the absorption of e but also caused a steady decrease i n the int e n s i t y of the transmitted l i g h t . The addition of -205-CH^ OH to the system to scavenge I was not as e f f i c i e n t as expected and contributions from these transients were s t i l l observed i n the spectra. The implications from the spectra were that e was present. Laser photometry data confirmed t h i s . The reduction of the l i f e t i m e of the transient absorption i n the presence of known e scavengers such as H +, N^O and 0 2 was consistent with the anticipated reactions and corroberated the evidence for e a q - Th e observation of a second absorption peak induced by the a u x i l i a r y f l a s h after the main fl a s h suggested that the product of the e decay with an impurity or some other species i n solution was being photoionised i t s e l f . However the presence of these iodine by-products and the fact that alcohol:iodine complexes are known to absorb i n the v i s i b l e region (120) made this system too ambiguous to interpret s a t i s f a c t o r i l y . It had been noted however that the second peak also d i s -appeared i n the presence of electron scavengers, and could be detected whether the methanol was there or not. A review of the usefulness and hazards of the I system i s contained i n ref. (121). In view of these results a system was chosen i n which the net chemical reaction was zero. This was a hydrogen saturated solution of NaOH or KOH. The e are generated from the u.v. emission i n the main aq 6 flash (122,123). OH" + hv -* OH + e" <f>101- ^0.2 aq 185 The subsequent reactions i n the presence of H 2 remove the hydroxyl radicals and hydrogen atoms to regenerate e a q : OH + H 2 -»• H + H20 k 0 = 1.6 x 10 8 M _ 1 sec" 1 (10) H + OH" •+ e" k. = 2.3 x 10 7 M _ 1 sec" 1 (11) aq 11 v -206-Therefore i t i s expected that e w i l l decay through the aq q -1 bimolecular reaction (1) with a rate constant (5.5 ± 0.75) x 10 M sec e" + e" -> (e2,") -+ H 0 + 2 OH" aq aq 2 'aq 2 aq and there w i l l be no ove r a l l chemical change. In practice the [H 2] was lim i t e d to 7 x 10 fvl(its s o l u b i l i t y ) and other mechanisms existed for removing both the electrons e" + H -> H" k„ 2.5 x 1 0 1 0 M"1 sec" 1 aq aq 2 e" + OH OH- k_ 3.0 x 1 0 1 0 M"1 sec" 1 aq 3 and the hydroxyl r a d i c a l s . Once formed, hydrogen peroxide may react with er aq '2"2 ^8 OH + OH -»• Ho0„ k D = 5.0 x 109 M"1 sec 1 e" + H o0 o -+ OH- + OH" k. = 1.23 x 1 0 1 0 M"1 sec" 1 aq 2 2 4 _7 In addition, the presence of an impurity S i n the system even at 10 M would compete e f f e c t i v e l y with other decay mechanisms i f the reaction were almost d i f f u s i o n controlled. At high pH, hydroxyl radicals are converted to 0 (pK a = 11.9) The reactions of e with 0 or H02 w i l l not be ne g l i g i b l e and may interfere with the sequence (1), (10), (11), by removing e aq F i n a l l y H atoms can pa r t i c i p a t e i n reactions (6) and (7) as opposed to reaction (11), and so conditions must be judi c i o u s l y optimised to obt -207-the desired chain of reactions. H + H -> H 2 k 6 = 1.0 x 1 0 1 0 M"1 s e c - 1 (6) H + OH + H20 k ? = 1.2 x 1 0 1 0 M _ 1 sec" 1 (7) (a) I d e n t i f i c a t i o n of the absorbing species generated by the  two flashes The mM [OH ] was selected i n order to produce 10 ^ M [e ] (empirically a c l determined from o p t i c a l density measurements) which would favour reactions (1), (10) and (11). The mean quantum y i e l d for the photoionisation of OH i n our experiments i s unknown but evidently was high enough to give t h i s i n i t i a l [e ] with readily accessible f l a s h energies. aq When 0.1 M methanol was added to the system the l i f e t i m e of e 1 aq (as seen through the f i r s t absorption signal following the main flash) was s l i g h t l y extended i n keeping with the elimination of reactions (3), (4) and (8). The addition of glucose had the same anticipated e f f e c t , however at these high concentrations i t was quite possible that impurity _7 was introduced into the system ^10 M and that was the reason why the decay of e was not as slow as i t might have been i n the presence of an OH r a d i c a l scavenger. Similar features were observed i n the mM Ba(0H) 2 solutions containing methanol or glucose. The only known transient that could absorb s i g n i f i c a n t l y at 632.8 nm i n t h i s system was CO^, produced from carbonate impurity by reaction with OH. The presence of equally intense transient absorption signals with the same l i f e t i m e following the double fl a s h i n solutions where B a + + would remove CO, , and methanol the OH r a d i c a l s , demonstrated that the -208-signals could not be attributed to the CO^" ion. I f N 2 gas were used to deoxygenate the OH" solutions p r i o r to use the l i f e t i m e of the absorption signal induced by the main flash was less than that for the OH /H2 system. Presumably t h i s was because the OH radicals were not being removed by reaction (10), and survived to react i n (3), (7) and (8) which also depleted the [H]. When electron scavengers such as N20 and 0^ were added to the system, the signals decreased or were en t i r e l y eliminated. • From th i s evidence and the spectrum shown i n Fig. 10-4, the absorption signal induced by the main f l a s h i s attributed to e~^. The evidence also strongly implies that the second absorption signal generated after the a u x i l i a r y f l a s h i s due to a regeneration of e~^ from the system. This could occur i n two ways, either (i) e reacts i n the system to give aq a r e l a t i v e l y stable product which i s photodissociated to give e i n the a u x i l i a r y f l a s h , or ( i i ) some artefact or impurity within the system gives r i s e to an increase i n [e ]• (i) The chemical evidence gives good reason to i d e n t i f y the species giving r i s e to the second absorption signal as e" . The only d i r e c t aq evidence would, of course, be the absorption spectrum. Known electron scavengers remove the r e l a t i v e l y small second peak while considerably reducing the [e ] as seen i n the f i r s t peak. The reactions that take place i n the OH /H2 system do not contain nor produce any species known to 9 - 1 - 1 absorb at 632.8 nm, other than e . The rate constant 5 x 10 M sec aq measured for the decay of the second peak for delays > 600 ysec corresponded 9 -1 -1 to the published value for reaction (1), (5.5 ± 0.75) 10 M sec -209-Even at shorter delay times the decay of the second signal was slower than of the f i r s t , but i t was d i f f i c u l t to measure a rate constant accurately because of the s i g n i f i c a n t [ e ^ ] s t i l l present from the photoionisation during the main f l a s h . The faster decay i s consistent with the fact that during the main fl a s h the i n i t i a l y i e l d of [OH] and [e~ ] i s the same, aq ~10 M, and therefore a f r a c t i o n of e" are lost through reaction (3) with OH r a d i c a l s . Since reaction (1), (3) and (11) are i n competition for the same two species, from a consideration of k^, k^ and k ^ i t may be estimated that ^ 50% of the i n i t i a l l y produced e~ w i l l be fated to undergo reaction (3). In the second f l a s h the energy of the photons i s below that required to photoionise OH and only e are produced which accordingly decay aq v i a reaction (1). ( i i ) There i s no evidence to support t h i s case and many convincing arguments against i t . I f e were being formed af t e r the a u x i l i a r y f l a s h aq quite independently of the events during the main f l a s h , then the a u x i l i a r y flash alone should give r i s e to a sig n a l . This was not observed. Even with a l l the f i l t e r precautions taken, i f an estimated 2% of u.v. emission reached the reaction vessel from the a u x i l i a r y lamp to photoionise OH we might have expected to see an absorption - but none was detected. As i t was, the height of t h i s signal following a double f l a s h was at least 15% of the signal from e generated i n the main fl a s h and often aq higher, depending on the output energy of the lamps. The remote p o s s i b i l i t y that the second peak was the res u l t of a thermal wave moving across the c e l l was discounted i n view of the scavenger studies - i t i s impossible to remove thermal shocks by adding 5 x 10 ^ M 0^ to the system. In pure water, no signals could be detected at a l l , following either flash. The p o s s i b i l i t y that an impurity were present i n -210-reproducibly correct amounts for our reaction scheme (CO^ has already been eliminated) and was being photolysed to give e i s most u n l i k e l y aq i n view of the many di f f e r e n t systems and concentrations of solutes employed over the period of the research project. We have good reason to suppose, therefore, that both absorption signals may be attributed to ®aq-(b) I d e n t i f i c a t i o n of the intermediate Species It would appear that some r e l a t i v e l y stable intermediate product of the events occurring during and af t e r the main fl a s h i s being photolysed to give e aq« Hydrated electrons themselves are precursors to this intermediate. The requirement of high pH i s not c r u c i a l since we have observed the same behaviour i n the KI system. S i m i l a r l y the presence of H 2 i s not c r i t i c a l , but i n hydrogen saturated solutions the whole process appears more e f f i c i e n t , probably through the scavenging of OH v i a reaction (10). Yet OH i t s e l f i s not an essential intermediate i n the sequence of events because methanol and glucose also scavenge OH and OH or I systems containing these solutes have exhibited the same features. The selective use of f i l t e r s indicate that the wavelengths responsible for the photolysis are i n the red or infra-red region, but i t was not possible to obtain an absorption spectrum of this intermediate. We believe that the most plausible explanation of the regeneration of e from photolysis of an intermediate species - of which e i s also the precursor - l i e s i n the formation and photodissociation of a hydrated electron dimer. e" + e" (e2,"*) (19) aq aq 2 'aq ( e 2 - ) + hv (X > 680 nm) -> 2 e" (20) -211-Th e observations i n a l l the chemical systems are i n keeping with th i s mechanism. The f u l l implication i s that reaction (1) does not go immediately to completion but through a r e l a t i v e l y stable intermediate. The l i f e t i m e of the intermediate (s) i n the reaction sequence below e" + e" -* (e2,") -> H„ + 2 OH" aq aq 2 yaq 2 aq could be determined from the rate of loss of e i n reaction (1), and the aq J' rate of formation of H 2 which has not been reported for t h i s reaction at the present time. What has been demonstrated i s that H atoms are not precursors of the hydrogen (125). Reaction (1) proceeds at a d i f f u s i o n controlled rate despite the expected loss of entropy on the formation of the dielectron 2-( e2 •'aq activated complex from two like-charge species. The dielectron i s presumably protonated by the solvent and yields H 2 v i a a hydride ion. The l a t t e r may be thought of as i t s conjugate base. H2° ^ a q + Haq + H2 <26> There may be an alternative intermediate i n the OH~/H2 system which could absorb l i g h t to regenerate e i n d i r e c t l y : Hydrogen peroxide. aq After main fl a s h OH + OH + H 20 2 (8) Au x i l i a r y f l a s h H 20 2 + hv -> ZOH OH + H 2 -> H + H20 (10) H + OH" -* e" (11) aq v J -212-However there are three arguments against t h i s , (i) OH would have been rapidly scavenged i n 0.1 M methanol solutions iTL/2 ~ ^ n s e c ) thus preventing the formation of H 20 2, ( i i ) and therefore reaction (10) i s not essential to the regeneration of e , ( i i i ) H 90 9 does not absorb at 632.8 nm. At pH 11 about 20% of the H 20 2 w i l l be dissociated, but the l i g h t absorption by HO,, formed i n the main (u.v.) fl a s h i s r e l a t i v e l y feeble compared to OH by virtue of i t s low concentration (123). The same statement may be applied to 0 also present i n strongly alkal i n e solutions. These ions w i l l not be absorbing the l i g h t from the a u x i l i a r y f l a s h which i s above 680 nm when the 7.54 f i l t e r s are i n place. I t has also been suggested (124) that H 20 2 might be the long-lived species that i s photoionised to give e , since recent pulsed conductivity work on aq i r r a d i a t e d water reported a stable negative intermediate formed during - _4 the conversion of e to OH at pH 10 i n the presence of 2 x 10 M H 2Q 2 (110). In any case the [H 20 2] concentration i n our system would be very small. I f the photolysis of H?0_ were to regenerate e then surely the second absorption peak i n neutral and non-hydrogenated solutions should be higher? The exact opposite i s observed. Indeed i n methanolic solutions t h i s peak i s very much i n evidence under conditions where OH would not survive long enough to react to give H 20 2 (reaction (15) opposed to reaction (8)). We conclude therefore that there i s no rea l case for the production of e from a mechanism involving H„0_. aq 5 2 2 2-Transient (e 2 ) species have been postulated i n some other systems with l i q u i d , glassy or s o l i d matrices (127, 129, 130) and as -213-P - centres i n a l k a l i halides and other crystals (131). The evidence presented here appears however to be the f i r s t demonstration of short-l i v e d solvated dielectrons i n water or any p r o t i c l i q u i d (75). This i n i t s e l f raises several questions concerning the nature and r e a c t i v i t y of the dielectron i n water. In r e f e r r i n g to t h i s species as the "dielectron" a certain amount of bias i s already evident. Although there i s no information to the contrary we assume that this species i s not a dimer of two temporarily associated e but i s best represented by two electrons i n a single cavity - solvation sheath, an entity i n i t s own r i g h t . As such, what interpretations may be put on i t s measured l i f e - t i m e of XY/2  % ^  t o milliseconds at pH 11, the position of i t s absorption band and i t s reactions i n aqueous solution? The formation and decay of the dielectron w i l l be considered i n r e l a t i o n to the l i q u i d ammonia system. The equilibrium i n that solvent i s attained rapidly (subscript w i l l be omitted except for the species i n water). 2 e" t (e 2") The rate of disappearance of e i s almost at the d i f f u s i o n F^ aq controlled l i m i t , therefore for t h i s equilibrium to exist i n water the 2_ reverse reaction must be r e l a t i v e l y slow. The l i f e t i m e of the (e„ ) ' ^ 2 'aq indicates that the favourable position i s far to the r i g h t . The long h a l f - l i f e would also suggest that the dielectron i s not as i n c l i n e d to -214-8 -1 -1 react with i t s e l f (k < 10 M sec ) as i s the monomeric species 9 - 1 - 1 (k^ = 5 x 10 M sec ). Even impurities naturally present i n the -7 -9 system at 10 to 10 M concentrations do not remove the dielectron as rapidly as they do e a q - The fact that 5 x 10 M 0 2 removed the second absorption signal could be (a) that the e were not available to form s u f f i c i e n t 2-(e 9 ) v i a reaction (1), or (b) that C> also scavenged any dielectrons i n the system. It i s not possible to make more than q u a l i t a t i v e statements about the pH dependence of the dielectron because of the many reactions occurring i n the system that would influence the dependence i n d i r e c t l y through e . At pH 7 T j y 2 °^ t n e dielectron i s ^ 1 millisecond; while at pH 11 4 xL/2 % *^ t 0 milliseconds. Even with a 10 f o l d decrease i n [OH ] concentration, T j y 2 * s o n i y a n order of magnitude or so smaller. We conclude that the dielectron i s not especially susceptible to attack by a proton and that production of molecular hydrogen by the sequence below i s not very l i k e l y . (e 9") + H + H" (27) 2 aq aq aq H" + H + H 0 (28) aq aq •> 2 At pH 11 these reactions would have a h a l f - l i f e i n excess of 1 second. I f the dielectron reacts with undissociated water to give a ( e 2 _ ) + H„0 -> H" + OH" (29) 2 'aq 2 aq ' hydride ion, where x^/2 ^ Msec, which i n turn reacted with H + -215-or H^O to give molecular hydrogen, H + H_0 -* + OH" (30) aq 2 2 aq i t i s possible that the H - might survive long enough to absorb the l i g h t a c l from the a u x i l i a r y f l a s h and regenerate e - i f a suitable absorption band were available. The processes giving r i s e to molecular hydrogen i n this system are s t i l l not f u l l y understood but probably do involve a l l these species as shown below: e" + e" + (e2.-) t H~ + OH" t H 0 + 2 OH" aq aq 2 'aq ^ aq 2 aq The rate of evolution of H 2 from the Na/^O system i s considerably reduced at high pH (126), which would suggest that reactions (27) or (28) are involved. Certainly the increase i n [OH ] would push reaction (30) to the l e f t but since the reverse of t h i s reaction i s r e l a t i v e l y slow, the decrease i n hydrogen evolution at high pH cannot be s o l e l y attributed to t h i s . It was possible that the plateau observed following the fast i n i t i a l decay of e~ a f t e r the main fl a s h > 5 KV could be attributed ' aq to the f i r s t step i n the e q u i l i b r i a above, but the evidence does not support t h i s notion since the p o s i t i o n of that equilibrium must be f a r to the r i g h t . 2 e" t (e 2") aq 2 'aq An alternative interpretation of the plateau may be found i n further considerations of the spin state of the dielectron, so far unspecified. Perhaps species of both m u l t i p l i c i t y , s e 2 and t e can exist under -216-suitable conditions i n the same system. Although somewhat speculative t h i s proposal i s not a novel one for the equilibrium s„ Z t e2 e2 has been postulated i n the case of e 2 species formed during the f l a s h photolysis of ether solutions of potassium metal (128). In both the potassium/tetrahydrofuran and potassium/dimethoxyethane solutions a stable species i s formed which has an absorption maximum near 900 nm and 720 nm respectively. These broad absorption bands are assigned to e^. On flashing the solutions within these bands (I, see below) the i n t e n s i t y of the absorption band attributed to e^ f i r s t diminished (II) and then rapidly grew i n again over a period of microseconds to give an o v e r a l l increase i n intensity at the o r i g i n a l absorption maxima ( I I I ) . The l a s t spectrum then flattened over a period of a few seconds to return to the o r i g i n a l l y recorded shape ( I ) , through a d e f i n i t e f i r s t - o r d e r process. The variations i n i n t e n s i t y did not exceed 0.15 OD units and were more commonly of the order 0.05 to 0.1 with respect to the i n t e n s i t y of the i n i t i a l spectrum of e 2- The ov e r a l l mechanism was postulated to be: s P o (I) ^ V 2e (II) Data for K/THF k 2e (II) _> t e (III) T i / 2 ~ ^ usee, but rate uncertain k -1 t (III) + s e (I) k = 12.8 sec e2 e2 The o r i g i n a l solutions of K/THF and K/DME were believed to contain s i n g l e t e 2 species because no evidence to the contrary had been obtained from ESR and magnetic s u s c e p t i b i l i t y experiments. -217-Although we have no evidence for changes i n m u l t i p l i c i t y i n the OH /H2 system (the effects of the photolysis i n the a u x i l i a r y f l a s h appear to be the same for a l l the delay times tested (10 usee to tens of m i l l i -seconds) ) i t i s not unreasonable to suppose that at very high [e" ] a aq 2-r e l a t i v e l y large number of (e 2 ) are formed during the main f l a s h , some of which may be photolysed but not photodissociated by the main f l a s h 2-i t s e l f . (At lower [ ( e 9 ) ] t h i s f r a c t i o n might be t r i v i a l ) . The photolysis of an e 2 species by the higher wavelengths i n the main fl a s h might lead to a change i n m u l t i p l i c i t y , and t h i s i n turn allow the dielectron to absorb the 632.8 nm laser l i g h t . The extinction c o e f f i c i e n t of the absorbing 4 -1 -1 dielectron at t h i s wavelength would have to be 'v 10 M cm i n order to 2-see the plateau i n t e n s i t y we observe, since the maximum [ ( e 2 ) ] cannot exceed h a l f the i n i t i a l t e a q ] even i f a l l processes reached t h e i r optimum 2- • -7 e f f i c i e n c y . Therefore i f (e 2 )• ^ 10 M, and only 10% were photolysed i n the a u x i l i a r y f l a s h the concentration of dielectron absorbing -8 at 632.8 nm would be ^ 10 M. However t h i s i s rather an academic exercise i n the absence of d i r e c t evidence pertaining to the actual spin state of e 2~ (c) The Absorption Spectrum of the Dielectron Our evidence would suggest that the absorption band which occurs > 680 nm and into the near infra-red region i s associated with the photo-di s s o c i a t i o n process. By analogy to the spectra of other e 2 centres and e g i n general, we may expect the absorption to be a very broad band. I f t h i s were true then the spectral data we have may only apply to the t a i l of the band i n the v i s i b l e region. 2-However i t appears clear that the X of (e„ ) i s on the low ^ max v 2 'aq - 2 1 8 -energy side of the X established for e . Unfortunately the absorption & / max aq / r spectrum of water extends throughout the near i n f r a - r e d region and w i l l impose an e f f e c t i v e l i m i t to the study of the aquated dielectron through normal spectroscopic methods. The o p t i c a l absorption spectrum of centres i n ice was reported quite recently (130) 'and the spectral data i s consistent with our observation that the X i s red-shifted with respect to that for monomeric max r electrons. In t h i s case the e 2 w a s i d e n t i f i e d through the appearance of a new, broad absorption band i n the near i n f r a - r e d with X ' ^ max occurring about 1000 nm. Bleaching of the normal trapped electrons caused the i n t e n s i t y of the new band to increase; the explanation was given i n terms of: e~ + hv (X > 580 nm) •> (e2~\ but the re v e r s i b l e process - photobleaching with l i g h t ^ 1000 nm - was not s a t i s f a c t o r y established because of the overlap between the absorption spectra of the two species. A t h e o r e t i c a l description of trapped dielectrons has been published recently by Fueki (132) using the d i e l e c t r i c continuum model o u t l i n e d i n Chapter I I I . Although t h i s i s the only known set of calc u l a t i o n s for dielectrons i n polar media, studies have previously been made on F 1 centres (two electrons at an anion vacancy) ; and by analogy used to predict the o p t i c a l and thermodynamic properties of e 2 centres i n l i q u i d ammonia (133). The calculations on F and F' centres were made using an electron adiabatic approximation which i s more appropriate for electrons trapped i n a s o l i d matrix. -219-Th e formal part of Fueki's calculations i s s i m i l a r to the SCF treatment of the hydrated electron. The electrons are both located at the same s i t e and the cavity i s assumed to have zero radius, the problem becoming an extension of the one-electron system as envisaged by Jortner including the assignment of ^ m a x to a •«-'1s t r a n s i t i o n . The conclusions from t h i s c a l c u l a t i o n may be summarised as follows: I f the s t a b i l i t y condition i s E - 2 E, < 0. The dielectron i s energetically stable provided that, D > (D /(D - 1)) s v op' v op " This implies that there i s no s i g n i f i c a n t chemical reaction involving e~ before the dielectron i s formed. D must be greater than 2.28 i f aq s 6 &2 i s to be stable i n aqueous systems where D = 1.78. We believe 2_ on the other hand that there must be reaction i n order to form (e~ ) 2 'aq It i s not clear therefore how we evaluate the energetics i n t h i s case. The po s i t i o n of A was calculated to be i n the same region of the r max 6 spectrum for the observed glassy and ice data but predicted a * m a x on the high energy side of A (e ) for a dielectron i n water (2.02 eV 6 & } max aq' v as compared with 1.75 eV). This value depends for the most part on the difference i n charge d i s t r i b u t i o n due to the influence of Dg which i n water i s ^ 80 as opposed to 3 for ice (at 77°K). The experimental value" for A i n ice was 1.3 eV compared to the 0.84 eV predicted, max r r Recalling the approximations and uncertainties involved i n the models of e , i t i s probably wiser to await further calculations and more experimental data on the dielectron than to r a t i o n a l i s e what l i t t l e i s already available i n too quantitative a manner. One reason -220-for this stand has already been emphasised: what i s the nature of the tr a n s i t i o n giving r i s e to E ^ m a x ? We can pose t h i s question for both e and (e? ) aq 2 yaq (2) The Photolysis of e" . ^ J 3 aq When the a u x i l i a r y lamp was f i r e d whilst a high concentration of e~ were s t i l l present, that i s at a 15 to 150 jusec delay.after the main lamp, there was no evidence of photobleaching nor of a sudden change i n the decay kin e t i c s of e a q - The l a t t e r might occur i f an electron had been photoreleased from i t s solvent trap. These mobile electrons move very rapidly indeed and would be expected to show a high chemical r e a c t i v i t y . In the event of e m reacting with e~^ the decay of e should proceed at a rate faster than that observed i n the main fl a s h alone. I f photobleaching took place the immediate loss of an appreciable f r a c t i o n of e would appear on the absorption signal as a aq sudden dip i n the curve. However the presence of an species prevents any d e f i n i t e conclusions from being drawn on the photolysis on e . 2-The only way of generating e without (e 9 ) i s apparently i n the presence aq z> aq of an electron scavenger such as 0 2 (see Fig. 10-1 (b)) whose concentration i s c a r e f u l l y selected to f i t into the reaction scheme for the OH system discussed e a r l i e r i n th i s chapter. Perhaps then attempts to photolyse the hydrated electrons w i l l give experimental data that i s unambiguous. 2-As well the nature of the photolysis of the (e^ ) should be explored, for t h i s process could be likened to the examples of e tra n s i t i o n s i l l u s t r a t e d i n Fig. 3-1. Are the photo-released electrons free or solvated, i s t h i s a photoionisation process or does the long wavelength l i m i t correspond - 2 2 1 -to the effe c t i v e "bond energy" of the dimer? The results of the other photolysis experiments using the s c i n t i l l a t o r f lash to photoexcite e generated through pulse r a d i o l y s i s were also inconclusive. Any observations that might have been made were hidden by the curvature on the i n i t i a l absorption s i g n a l . Clearly i t i s desirable to delay the photolysis f l a s h u n t i l the period of curvature i s over. -222-References •1. (a) I.V. Vereshchinskii and A.K. Pikaev, Introduction to Radiation Chemistry. English Version through Israel Program cfor S c i e n t i f i c Translations Jerusalem (1964). (b) I. Santar and J. Bednar, Theory of Radiation Chemical Y i e l d , I-V. Part I Col l e c t i o n Czech. Chem. Commun., 32, 953 (1967). Part II 33, 1 (1968) Part I I I 34, 1 (1969) Part IV 34, 311 (1969) Part V Int. J . Radiat. Phys. Chem. 1, 133 (1969) (c) M. Burton and J.L. Magee (Editors) Advances i n Radiation Chemistry, volume 1, Wiley, Interscience (1969). (d) A.O. A l l e n , The Radiation Chemistry of Water and Aqueous Solutions, D. Van Nostrand, Inc., N.Y., (1961). (e) N. M i l l e r , Rev. Pure Appl. Chem. 7_, 123, (1957). 2. (a) P.A. Cerenkov, Dokl. Acad.. Nauk. SSR 1934, 2, 451. (b) E.A. Shaede and D.C. Walker, Int. J . Radiat. Phys. Chem. 1_, 307 (1969). 3. A. Mozumder (ed. Burton and Magee) Advances i n Radiation Chemistry, opus c i t . . 4. (a) A. Mozumder and J.L. Magee, Rad. Res., 28_, 203-214 (1964). (b) A.Mozumder and J.L. Magee, Rad. Res., 28, 215-231 (1964). 5. A.H. Samuel and J.L. Magee, J. Chem. Phys., 21_, 1080 (1953). 6. J.L. Magee, Rad. Res. Suppl. 4, 20 (1964). -223-7. (a) A. Mozumder, J. Chem. Phys., 48, 1659 (1968). (b) A. Mozumder, J. Chem. Phys., 50, 3153 (1969). J. Chem. Phys., 50, 3162 (1969). 8. (a) R.L. Platzman, Physical and Chemical Aspects of Basic Mechanisms i n Radiobiology, Publ. no 305, 22 (U.S. NRC) ,(1953) (c) R.L. Platzman, The Vortex, 2_3, 372 (1962). 9. L.H. Gray, J . Chim. Phys., 48, 172 (1951). 10. C. Wingate, W. Gross, G. F a i l l a , Rad. Res., 8_, 441, (1958). 11. R. Cooper and R. Mooring, Austral. J . Chem., 2j_, 2417 (1968). 12. M..J. B r o n s k i l l and J.W. Hunt, J. Phys. Chem. 72_, 3762 (1968). o 13. D.A. Head and D.C. Walker, Can. J. Chem., 45, 2051 (1967). 14. A. Kupperman, Actions Chemiques et Biologiques des Radiations, ed. M. Haissinsky, Masson et Cie, Paris (1961). 15. (a) T.J. Sworski, Adv. Chem. Ser. 5£, 163 (1965). (b) Proc. V. Inf. Conf. on the Rad. Chem. of Water. AEC no. C00-38-519. 16. J. J . Weiss, Nature 153, 748 (1944). 17. H. Lin s c h i t z , Physical and Chemical Aspects of Basic Mechanisms i n Radiobiology, opus c i t . 18. G. Stein, Disc. Farad. S o c , 12_, 227 (1952). 19. E.J. Hart, J.A.C.S., 76, 4312 (1954). 20. J.H. Baxendale and G. Hughes, Z. Phys. Chem. (Frankfurt) 14_, 306 (1958). 21. E. Hayon and J.J. Weiss, Proc. 2nd Int. Conf. of Peaceful Uses of Atomic Energy (1959). -224-22. N.F. Barr and A.O. A l l e n , J . Phys. Chem., 63, 628 (1959). 23. E. Hayon and A.O. A l l e n , J . Phys. Chem., 65_, 2181 (1961). 24. G. Czapski and A.O. A l l e n , J . Phys. Chem. 66, 262 (1962). 25. G. Czapski and H.A. Schwarz, J. Phys. Chem., 66, 471 (1962). 26. E. Collinson, F.S. Dainton, D.R. Smith and S. Tazuke, Proc. Chem. Soc. 140 (1962). 27. F.S. Dainton and W.S. Watt, Proc. Roy. Soc. 275A, 447 (1963). 28. J.P. Keene, Nature 197, 47 (1962). 29. (a) E.J. Hart and J. Boag, J.A.C.S. 84, 4090 (1962). (b) E.J. Hart and J. Boag, Nature, 197, 45 (1963). 30. D.C. Walker, Adv. Chem. Ser. 8J_, 49 (1968). 31. G.A. Kenney and D.C. Walker, i n press. 32. K.H. Schmidt and W.L. Buck, Science 151, 70 (1966). 33. J.H. Baxendale, Rad. Res. Suppl. 4_, 139 (1964). 34. W.C. Gottschall and E.J. Hart, J . Phys. Chem., 71, 2102 (1967). 35. (a) S.C. Wallace unpublished data. (b) D.C. Walker and S.C. Wallace, submitted Chem. Phys. Lett. 36. M. Anbar and E.J. Hart, J . Phys. Chem, 69_, 1244 (1965). 37. R.M. Noyes, Adv. Chem. Ser. 81_, 65 (1968). 38. U. Schindewolf, Angew. Chem. Internat. Edit. 7_, 190 (1968). 39. G. Lepoutre and M.J. Sienko (ed) Solutipns Metal-Ammoniac Benjamin, N.Y. (1964). -225-40. S.O. Nielsen, P. Pagsberg and E.J. Hart, J . Phys. Chem. 73_, 3171 (1969). 41. E.A. Moelwyn-Hughes, Physical Chemistry, Pengammon Press, Oxford (1964) 42. U. Schindewolf, H. Kohrmann and G. Lang, Angew. Chem. Internat. Edit. 8_, 512 (1969). 43. R.R. Hentz, Farhataziz and D.J. Milner, J . Chem. Phys., £7, 4865 (1967) 44. L. Kevan, Radiation Chemistry of Frozen Aq. Solutions, AEC no. COO-1529-18. 45. I.A. Taub and K. Eiben, J . Chem. Phys., 49, 2499 (1968). 46. B.G. Ershov and A.K. Pikaev, S t a b i l i s a t i o n of Electrons i n Low  Temperature Radiolysis of Polar Systems (Review) Adv. Chem. Ser. 81, (1968). 47. G. Nilsson, H.C. Christensen, J. Fenger, P. Pagsberg and S.O. Nielsen, Adv. Chem. Ser. 81, 71 (1968). 48. (a) J. Morgan and B.E. Warren, J. Chem. Phys., 6, 666 (1938). (b) G.W. Brady and W.J. Romanov, J. Chem. Phys., 3_2, 306 (1960). 49. D.Eisenberg and W. Kauzmann, The Structure and Properties of Water O.U.P., Oxford (1969). 50. M. Falk and H.R. Wyss, J. Chem'. Phys., 5_1, 5727 (1969). 51. (a) R. S c h i l l e r , J. Chem. Phys., 47, 2278 (1967). (b) R. S c h i l l e r , J . Chem. Phys., 47, 2281 (1967). 52. G.R. Freeman and J.M. Fayadh, J. Chem. Phys., 43, 86 (1965). 53. D.A. Head and D.C. Walker, i n press. 54. P.S. Julienne and L.P. Gary, Molecular Crystals, 5_, 135 (1968). 55. F. Gutman and L. Lyons, Organic Semiconductors, Wiley, N.Y'. , (1967). 56. (a) U. Solokov and G. Stein, J . Chem. Phys., 44, 3329 (1966). (b) U. Solokov and G. Stein, J. Chem. Phys., 44, 3329 (1966). -226-57. D.C. Walker, Quart. Rev. 21_, 79 (1967). 58. R. Onaka and T. Takahashi, J . Phys. Soc. Japan, 24, 548 (1968). 59. L. Robinson Painter, R.N. Hamm, E.T. Arakawa and R.D. Birkhoff, Phys. Rev. Lett. 21_, 282 (1968). 60. M. Cottin, J . Massanet, and C. Vermeil, J . Chim. Phys., 63, 959 (1966). 61. N. Getoff and G.O. Schenck, Proc. Int. Conf. on Photochemistry, Munich (1967). 62. J.W. Boyle, J.A. Ghormley, C.J. Hochanadel and J.F. Riley, J. Phys. Chem., 73, 2886 (1969). 63. (a) J. Jortner, M. Ottolenghi and G. Stein, J . Phys. Chem. 66, 2029 (1962). (b) Ibid, J. Phys. Chem. 66, 2037 (1962). 64. M. Shirom and G. Stein, Nature 204, 778 (1964). 65. J . Barret, M.F. Fox and A.L. Mansell, J . Chem. S o c , A., 483 (1967). 66. L. D o g l i o t t i and E. Hayon, J. Phys. Chem., 72_, 1800 (1968). 67. G. Czapski, J . Ogdan and M. Ottolenghi, Chem. Phys. Lett. 3_, 383 (1969). 68. J.R. Huber and E. Hayon, J. Phys. Chem. _72, 3820 (1968). 69. W.L. Waltz and A.W. Adamson, J. Phys. Chem. 73, 4250 (1969). 70. R. Devonshire and J.J. Weiss, J . Phys. Chem. 72_, 3815 (1968). 71. F.S. Dainton, Fast Reactions and Primary Processes i n Chemical Kinetics, ed. S. Claesson, Interscience (1967). 72. J . J . Weiss, The Chemistry of Excitation and Ionisation, ed. G.R.A. Johnson and G. Scholes; Taylor and Frances Ltd., London (1967). 73. F. Barat, B. Hickel and J. Sutton, Chem. Comm. (1969), 125. -227-74. (a) L.I. Grossweiner and H.I. Joschek, Adv. i n Chem., 50_, 279 (1965). (b) R. Santus, A. Helene, C. Ilelene and M. Ptak, J. Phys. Chem., 7£, 550 (197C 75. N. Basco, G.A. Kenney, D.C. Walker, Chem. Comm. (1969), 917. 76. J. Jortner, J . Chem. Phys. 30_, 839 (1959). 77. J. Jortner, Rad. Res. Supp. 4_, 24 (1964). " 7 8- Weiz£InR e§ci^ Solutions, ed., G. Stein, The 79. R.A. Ogg, Phys. Rev. 69, 688 (1946). 80. (a) W.N. Lipcomb, J. Chem. Phys., 2l_, 52 (1953). (b) R.A. S t a i r s , J. Chem. Phys., 27_, 1431 (1957). 81. L. Landau, Physik Z. Sowiet Union 3_, 664 (1933). 82. A.S. Davydov, Zh. Eksperim Theor, F i z . , 18_, 913 (1948). 83. J . J . Weiss, Nature, 186, 751 (1960). 84. R.L. Potter, R.G. Shares and J.L. Dye, J. Chem. Phys., 35, 1907 (1961). 85. L. Dorfman, Solvated Electrons, ed. R. Gould, Washington (1965). 86. R.A. Holyroyd,in "Fundamental Processes i n Radiation Chemistry" ed. P. Ausloos, Interscience 1968, p. 413. 87. J. J . Weiss, Nature, 215, 151 (1967). 88. W.H. Hamill, J. Chem. Phys, 49_, 2446 (1968). 89. S. Noda, K. Fueki, Z. Kuri, B u l l . Chem. Soc. Japan, 42, 16 (1969). 90. K. Fueki, D.F. Feng and L. Kevan, Chem. Phys. Lett. £, 313 (1969). 91. K. Fueki, J . Chem. Phys., 49, 765 (1968). 92. (a) M. Natori and T. Watanabe, J. Phys. Soc. Japan, 2_1, 1573 (1966). (b) M. Natori, J . Phys. Soc. Japan, 24, 913 (1968). 93. J.E. Bennet, B. Mile and A. Thomas, J. Chem. Soc. A, 1967, 1394. 94. E.C. Avery, J.R. Remko, and B. Smaller, J . Chem. Phys., 49, 951, (1968). -228-95. (a) K. Iguchi, J . Chem. Phys. 48_, 1735 (1968). (b) K. Iguchi, J . Chem. Phys. 51_, 3137 (1969). 96. L. Raff, and H. Pohl, Adv. Chem. Ser., 50, (1965). 97. B.J. McAloon, and B.C. Webster, Theoret. Chim. Acta (Berl), 15_, 385 (1969). 98. M. Banning, J. Opt. Soc. Am. 37, 686 (1947). 99. N. Basco, unpublished. 100. A.M. Bass, and K.G. Kessler, J . Opt. Soc. Am., 49_, 1223 (1959). 101. C. W i l l i s , A.W. Boyd, A.E. Rothwell and O.A. M i l l e r , Int. J . Radiat. Phys. Chem., _1, 373 (1969), p. 374. 102. L. Dorfman, personal communication. 103. G.A. Kenney, M.Sc. Thesis (1968). 104. F. Jenkins-and H. White, Fundamentals of Optics, McGraw-Hill (1957). 105. J.W. Boag, Trans. Farad. S o c , 64, 677 (1968). 106. J.C. Russel and G.R. Freeman, J. Chem. Phys. 48, 90 (1968). 107. G.A. Kenney and D.C. Walker, J . Chem. Phys., 50, 4074 (1969). 108. H. Schwarz and H. Hora, App. Physics. Letters, L5, 349 (1969). 109. J.G. Trump and R.J. Van de Graaff, J . App. Phys. _19, 599 (1948). 110. K.H. Schmidt and S.M. Ander, J. Phys. Chem. 73_, 2846 (1969). f t l . E. Hayon, Trans. Farad. S o c , 60, 1059 (1964). 112. See A.O. A l l e n ; E. Hayon; E.J. Hart, i n Proc. V. Inf. Int. Conf. on the Rad. Chem. of Water, opus c i t . 113. P.J. Dyne and J.M. Kennedy, Can. J. Chem., 36, 1518 (1958). 114. J.W. Hunt and J.K. Thomas, Rad. Res., 32, 149 (1967). 115. J.K. Thomas and R.V. Bensasson, J . Chem. Phys., 46, 4147 (1967). -229-116. H.A. Schwarz, J. Phys. Chem. 73, 1928 (1969). 117. N. Klein and J. Warner, Trans. Am. Nucl. Soc. 9, 83 (1966). 118. N. Klein, C.N. Trumbore, J.E. Fanning and J. Warner, J. Phys. Chem., 72, 880 (1968). 119. S.C. Wallace, t h i s laboratory. 120. J.S. Ham, J. Chem. Phys. 20, 1170 (1957). 121. G. Stein, Action Chimiques et Biologuiqes des Radiations, ed. M. Haissinskey, 1^ 3, 130 (1970). 122. M.S. Matheson, W.A. Mulac and J . Rabani, J . Phys. Chem. 6J7_, 2613 (1963). 123. K. Schmidt and E.J. Hart, Adv. Chem. Ser. 81_, 267 (1968). 124. L. Grossweiner, private communication to N. Basco. 125. L.M. Dorfman and I.A. Taub, J.A.C.S. 85, 2370 (1963). 126. E.A. Shaede and D.C. Walker, Chem. Soc. Spec. Publ. 22_, 277 (1967). 127. (a) F.S. Dainton, G.A. Salmon, J. Teply, Proc. Roy. Soc. (1965) 286A, 27. (b) Ibid, 319. 128. J. Elorantaand H. Li n s c h i t z , J. Chem. Phys. 38, 2214 (1963). 129. D. Schulte-Frohlinde and K. Eiben, Z. Naturforsch, 17a, 445 (1962). 130. L. Kevan, D.R. Renneke and R.J. Friauf, Solid State Comm. 6_, 469, (1968). 131. J.J. Markham, "F-Centres i n A l k a l i Halides", Academic Press (1966). 132. K. Fueki, J. Chem. Phys. 50, 5381 (1969). 133. R.H. Land, D.E. O'Reilly, J . Chem. Phys., 46, 4496 (1967). 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0059867/manifest

Comment

Related Items