J. Chem. SOC., Faraday Trans. I, 1984, 80, 1745-1756 Photophysics of the Excited Uranyl Ion in Aqueous Solutions Part 3.-Effects of Temperature and Deuterated Water: Mechanisms of Solvent Exchange and Hydrogen Abstraction from Water in Excited States BY SEBASTIAO J. FORMOSINHO* AND MARIA DA GRACA M. MIGUEL Chemistry Department, University of Coimbra, 3000 Coimbra, Portugal Receiued 25th July, 1983 The effects of temperature and D20 on excited uranyl decay in aqueous solution have been studied at several pH values and uranyl concentrations. The data are interpreted in terms of reversible crossing between two excited states of uranyl with different electronic distributions. The highest state, called U*, has a 71; 4; configuration whereas the lowest state, called X*, is nf, 8; (n = 1).The reversible decay is interpreted in terms of a solvent-exchange process. The values of AH+ and AS+ for this process reveal that solvent exchange proceeds by different mechanisms for the U* and X* states. The activation parameters are pH dependent, but the U* state always has more dissociative character for solvent exchange, whereas the X* state has more associative character. A weak effect on the solvent-exchange rates was found in D20. The rates of the irreversible decays are attributed mainly to a hydrogen-abstraction reaction from water molecules by excited uranyl ions. The activation energies and the frequency factors are pH dependent, revealing that hydrogen abstraction from the coordinated equatorial H 2 0 and from the more loosely bound water in axial positions can occur.The temperature and the deuterium isotope effects on the decays were analysed in terms of the current theories of radiationless transitions for the involvement of the 0-H and U=O stretching modes in the abstraction process. The autoquenching processes and the effect of the exciting intensity on the decays are in agreement with the proposed mechanisms. The photophysics of the excited uranyl ion in aqueous solution depends on many factors, including the acidity, uranyl concentration, nature and concentration of anions, intensity of excitation, temperature and deuterium isotope effect.l. * In spite of the complexity of such a system, the dynamic and stationary behaviour of uranyl fluorescence can reasonably be interpreted in terms of a reversible-crossing process between two almost isoenergetic electronic states, U* and X*, of [UOit,,,]*.Each of the excited states also decays by an unimolecular irreversible-decay pr0cess.l Decays of both kinds of excited states seem to be of a chemical nature. The reversible crossing is attributed to a solvent-exchange mechanism1 and the irreversible decays to a hydrogen-abstraction reaction from H,01-5 with rates that are pH dependent. A better understanding of the nature of such a process requires study of the effect of temperature and of D,O on the photophysics of excited uranyl at different pH values and at different uranyl concentrations. This paper presents the results of such studies and reveals that although in terms of the rates of decay both states U* and X* behave in a similar manner, under certain experimental conditions they can be involved in different types of mechanisms for solvent-exchange and hydrogen-abstraction reactions.17451746 PHOTOPHYSICS OF URANYL ION IN AQUEOUS SOLUTIONS EXPERIMENTAL Excited uranyl decays were studied on a nanosecond flash-photolysis apparatus with an N, laser (Aex = 337.1 nm) at constant laser intensity. For temperature studies the solutions were thermostatted in a water bath giving temperatures which were constant to f0.5 "C. Solutions in H,O were prepared with triply distilled water; solutions in D,O were prepared with deuterium oxide (Sigma, 99.8% D). Uranyl nitrate was the purest grade commercially available. Further experimental details have been given elsewhere.RESULTS The fluorescence lifetime of uranyl nitrate in water was measured as a function of temperature and uranyl concentration at various pH values between 1 and 4. Decay of the uranyl excited state in all cases was biexponential, but could be fitted to a model involving reversible crossing between the two lowest excited states of the uranyl ion1 ki u*ex*. kr Increasing temperature increased both decay rates, which followed good Arrhenius plots independent of the pH of the solution. However, at certain pH values a limiting rate was observed at temperatures > 40 "C (fig. 1). The decays were analysed in terms of absolute rate theory for a first-order rate, assuming that the decay arises from exchange of all coordinated solvent molecules, using the expression nkT h k,,, = -exp (ASTIR) exp (- A H t / R T ) where n is the number of coordinated water molecules in the excited uranyl ion.Entropies and enthalpies of activation calculated assuming a hydration number of 4, identical to that suggested for the ground state,6 are presented in table 1 for pH values between 1 and 4. Since all of the exchange rates are considered as first-order processes, the AS? values in table 1 are strictly applicable only to a dissociative-exchange reaction ( S , 1). However, if exchange occurs by a second-order process ASt will only decrease by 17 J K-l m01-l.~ The effect of added nitrate ion (0-1 mol dmP3) on the decay of the uranyl excited state was studied at pH 3 and 20 "C. The biexponential decay was analysed in terms of the reversible-crossing mechanism.The results presented in fig. 2 reveal that NO; increases ki and decreases k,. The decay of laser-excited uranyl ion luminescence was also studied at pH 3 in D,O solutions. The decay was biexponential at all temperatures over the range studied (3.5-56 "C) and was analysed in terms of a reversible-crossing mechanism. The rates of reversible crossing are smaller in D,O than in H,O, but show a temperature- dependent isotope effect. At 10 "C k,,,(H,O)/k,,,(D,O) = 1.0-1.1; the ratio then attains a maximum of 1.3 at 30 "C and decreases at higher temperature. Study of the effect of temperature on ki and k, (fig. 3) allowed an estimation of AH? and ASt as in H,O (table 1). The effect of temperature on the rates of the irreversible decays in the lowest excited states of uranyl u * e x * was also studied in H,O at different pH values and in D,O at pH 3.In H,O, the rates k, and k , increase with increasing temperature and follow good Arrhenius plots. AtS. J. FORMOSINHO AND M. DA G. M. MIGUEL 1747 13 h - I v) \ -Y, c I 12 1' I I I I I I .o 3.2 3.3 3 . 4 1 0 3 KIT Fig. 1. Arrhenius plots for the rates of reversible crossing ki (A) and k , (A) at (a) pH 1 and (b) pH 4; [UOi+] = 0.02 mol dm-3 and [NO;] = 0.15 mol drnp3. low [NO;] (0.04 mol dm-3) the rates are independent of temperature for temperatures Z 20 OC.' In D20, k, shows good Arrhenius behaviour but k , does not follow an Arrhenius plot. The solvent isotope effect for k, and k , at pH 3 is presented as a function of temperature in fig. 4. Nitrate ion concentration (fig.2) does not affect the rate k , but slightly increases the rate constant k,. The activation energies, E,, and the pre-exponential factors, A , for k , and k , in H,O are presented in table 2 for different pH values; k, in D20 has E, = 18.0 kJ mol-1 and A = 1.5 x lo8 s-l.1748 PHOTOPHYSICS OF URANYL ION IN AQUEOUS SOLUTIONS Table 1. Enthalpies and enthropies of activation for the rates of reversible crossing for the excited uranyl ion in H,O 1 .o 62 58 52 29 2.0 46 8.5 33.5 - 37.5 3 .O 67 75 50 17 3.0 (D20b solution) 46 2 27 - 58 4.0 42 - 17 29 - 58.5 a Estimated for the exchange of four solvent molecules; for six molecules A S is lower by Value obtained from glass electrode but not corrected for the change in the 3.5 J mo1-l K-l. autoprotolysis constant on going from H 2 0 to D,O.10 8 7 6 v) 0 - --- * 4 2 0 4 - I v) " 3 2 . * 2 I I I I I 0.2 0.4 0.6 0.8 1.0 [NO;l,/mol dm-3 ( b ) 0.2 0.4 0.6 0.8 1.0 [NOJ,/rnol dm-3 Fig. 2. Effect of NO; on the decay rates of excited uranyl at 20 "C and at pH 3.0; [UOi+] = 0.02 mol dm-3: A, ki; A, k,; a, k , ; 0, k,.S. J. FORMOSINHO AND M. DA G. M. MIGUEL 1749 14 13 h " I m --. v Y E - 12 11 lo3 KIT Fig. 3. Arrhenius plot for the rates of reversible crossing ki (A) and k, (a) in D,O at pH 3 ; [UOi+] = 0.02 mol drnM3. 2 0 20 40 60 TIo C Fig. 4. Temperature dependence of the isotope effect ratio in H,O and D,O for the rates of irreversible decay k, (0) and k, (0) at pH 3.1750 PHOTOPHYSICS OF URANYL ION IN AQUEOUS SOLUTIONS Table 2. Activation energies and pre-exponential factors for the rates k , and k , in H,O 1 .o 31 7 x 10'0 13.5 7.5 x 107 2.0 44 3 x 1013 42 4 x 1013 3 .O 24 2 x 109 24 2 x 109 4.0 30.5 4.5 x 1Olo 19 4 x lo8 DISCUSSION SOLVENT-EXCHANGE REACTIONS SOLVENT EXCHANGE versus TUNNELLING Analysis of the biexponential decay of the uranyl ion in water gave activation parameters for reversible excited-state crossing between U* and X*.The activation enthalpies determined for such processes are too large for a purely physical radiationless transition from U* to X* since U* is ca. 300 cm-l above X*,l and consequently such a physical transition to an almost isoenergetic state should have almost negligible AH+ (< 15 kJ mol-l). Furthermore, for a physical process k, would be expected to have AH! higher than AH] by ca.4 kJ mol-l, contrary to experimental observations (table 1). Ikeda et aZ.6 have studied by n.m.r. the kinetics of the water-exchange process for the equatorial positions of ground state of UO;t,,,. The exchange rate for all the coordinated molecules at 25 "C is estimated to be 4 x lo6 s-l with AH+ = 42 kJ mol-l. Both the rate constant and the activation enthalpy are comparable to the values found for ki and k,, supporting the idea that the reversible-crossing process may be due to a solvent-exchange mechanism. Further support comes from the effect of [NO;] on the exchange rates at pH 3 (fig. 2). Nitrate ion increases ki and decreases k,. Because solvent exchange is a special case of a nucleophilic substitution reaction in metal complexes, for a dissociative solvent-exchange mechanism the negative charge of the NO; ligand and its larger size, when compared with H,O or OH-, are predicted to increase the rate of solvent exchange.8 However, for an associative ( S , 2) mechanism NO, can decrease the rate of exchange.As table 1 shows, at pH 3 AS1 is very positive, suggesting that the solvent exchange in the U* state proceeds via a dissociative mechanism. This can be a D-type mechanism since AH1 is also high.g In contrast, the solvent exchange mechanism for the X* state has a positive AS: value and the AH: is also smaller than AH!. This suggests a weakly dissociative mechanism of the type In any case, the X* state has a more associative or assisted character and the U* state a more strongly dissociative character for solvent exchange.NATURE OF THE ELECTRONIC STATES OF THE EXCITED URANYL ION At all the pH values used the states U* and X* appear to exchange coordinated water by different mechanisms. For transition-metal complexes solvent exchange is currently interpreted in terms of crystal-field effects.*' Reasoning by analogy we can expect that for UO$+ and (UOE+)* crystal-field effects on the U atomic orbitals would play a role in the solvent-exchange processes. The small differences in the rates ki andS. J. FORMOSINHO AND M. DA G. M. MIGUEL 1751 k, suggest that the atomic orbitals involved in these processes have anf nature; d orbitals have crystal-field effects larger thanforbitals and affect solvent-exchange rates of transition-metal ions by several orders of magnitude.The involvement off orbitals is also in agreement with the studies of JerrgensenlO and Denning et aZ.,ll which consider the lowest vacant orbitals in UOi+ to be the Sforbitals 4, (f,,) and 6, (f,,). The fo andf,, orbitals are considered to be strongly involved in the U=O bonds. Whilst controversy exists over the actual hydration number of the uranyl ion in aqueous solution, and between 4 and 6 coordinated waters have been l3 it is generally agreed that the linear UOi+ ions are arranged at right angles to the water channels, the H,O molecules being in equatorial positions with respect to the linear In an excited hydrated (UOi+)* ion the approach of an H 2 0 molecule in an associative mechanism can occur on the plane in which the low-energy atomic orbitals are empty.'* The 4,(5f) orbitals have lobes pointing in an equatorial direction whereas the 6,(5f) orbitals have lobes pointing in directions making an angle of 45" with the equatorial plane.Consequently if in the excited state the 4, orbital is empty an associative mechanism for solvent exchange can be expected. However, the same cannot be said when the 6, orbital is empty but the 4, orbital is occupied by an un- paired electron, because repulsion between the lone-pair electrons of H 2 0 and the unpaired electron in 4; would hinder the approach of H,O molecules in the equatorial plane. This argument suggests that the state X*, with the more associative character, has the excited electron in a SL(5f) orbital and the state U* has the excited electron in a 4;(5f) orbital.In (UOi+)aq solvent-exchange rates are ca. 10 times higher than in (UOi+)&.6 This would not support the electronic transition 0, -+ 6, proposed by Denning et a1.l1 for the two lowest excited states of the uranyl ion. In fact, the 0, orbital has an electron density in an axial direction [p,(O) and fo(U) atomic orbitals] and consequently the 0, .+ 6, transition brings electron density from an axial position towards the equatorial plane. This will increase the repulsion between the lone pairs of H 2 0 and the U atom. The effect is equivalent to a decrease in the charge of the U atom, an effect which is known to increase the rate of solvent exchange.8v9 Such an effect was not expected by Jerrgensen1O for a n, + 6, or 4, transition because the n, orbital has a large electron density in the equatorial plane, mainly from the p + , atomic orbitals of the 0 atoms.However, the ground-state values were obtained b p H n.m.r., a technique which leads to considerable difficulties in the extraction of absolute solvent-exchange rates.g So the possibility that the electronic configurations of the lowest excited electronic states of (UO;') are n3,& (X*) and nt 4; (U*) needs to be treated with some care. The total angular momentum (0) components can be found for the possible electronic configurations of the excited states:l5$ l6 n3, 6; ,lI(i2 = 0 , l and 2) llI(i2 = 1 ) n3, 4; 3A(R = 1,2 and 3) lA(R = 2) ,r(R = 3,4 and 5 ) T(C2 = 4) 0; 6; ,A(Q = 1,2 and 3) lA(Q = 2) a; 4; "(0 = 2,3 and 4) l4(R = 3). For the first excited state spectroscopic data reveal R = 1 .ll In UO;+, because of large spin-orbit coupling, spin is not a good quantum number. If the spin states are combined, conserving the symmetry and the total angular momentum, the relevant states are x* 7Cn3,a; n(n = 1 ) U* nn3,4; A(i2 = 2) or r(Q =4).1752 PHOTOPHYSICS OF URANYL ION IN AQUEOUS SOLUTIONS ACIDITY EFFECTS The effect of pH on the fluorescence quantum yield and decay of excited uranyl has been interpreted in terms of the presence of different hydroxo and polymer ionic species that vary in concentration with pH., Each of these species has different rates of solvent exchange and the variation of the observed rates ki and k , with pH reflect such complexity.As table 1 shows, in spite of uncertainty in the coordination number of the water molecules, for all pH values the U* state always has a more dissociative solvent-exchange character than X* since AS! > AS$.The difference in the activation entropies varies between 60 and 29 J K-l mol-l, but is sufficiently large to be attributed to a difference in mechanism and not to a difference in the number of coordinated waters around the two states. Even for the more extreme situation between [UO,(H,O):+]* and [UO,(H,O)?+]* the difference in the entropies of activation would be only of the order of 14.5 J K-l mol-l. Consequently we will discuss the results of table 1 in terms of a mechanism of solvent exchange and consider that the pH dependence of AHt and AS? reflects the solvent-exchange mechanism of the dominant ionic species at each pH.Those species are considered to be at pH 1 [U02(H,0),2+]*, pH 2 [UO,(OH)(H,O)i]*, pH 3 [UO,(OH),(H,O),]* and pH 4 [(u02)2(0H),1**2 The guidance used to establish the mechanism of solvent exchange comes mainly from the activation parameters. Although the activation enthalpy of one particular cation has no diagnostic value, individual activation entropies may have. A broad distinction is generally made between markedly positive AS? for a dissociative process and markedly negative values, indicating an associative proces~.~ At pH 1 and 3 the solvent exchange seems to be of a dissociative character in both states although, as previously stated, with stronger character in the U* state. Support for the dissociative character of these processes comes also from the activation-enthalpy values which are quite high, revealing that in the breaking of the H,O-UOi+ bond there is virtually no contribution from the incoming solvent molecule to such breaking.The limiting rate attained at high temperatures (30-40 "C) also supports this view and should correspond to a situation where the breaking of the H,O-UOi+ bond is no longer the rate-determining step. At pH 2 and 4 AH+ values are smallest and AS+ are negative, suggesting an associative mechanism, at least for the X* state. For U* and at least at pH 2 for solvent exchange with a dissociative character, a strong interaction with the incoming ligand exists (I, mechanism). The changes in mechanism can be attributed to the nature of the ligands in the different ionic species.At pH 1 in [UO,(H,O)i+]* the presence of the electron-supplying ligand H,O favours a dissociation mechanism.* At pH 2 the effect of the negative charge of OH- in [UO,(OH)(H,O)$]* seems to be the dominant factor and favours a more associative mechanism. At pH 3 in the neutral species UO,(OH),(H,O), the effect of charge is less significant and dominates the electron-withdrawing character of the two OH- groups; such an effect favours a dissociative mechanism. Finally, at pH 4 the dominant species is a polynuclear cation where the strong bonding of the ligands, particularly those in the bridges between the U atoms, makes dissociation of the ligand molecules difficult and therefore favours an associative mechanism. EFFECT OF D20 The decay of the excited uranyl ion was studied in D,O at pH 3 and analysed in terms of reversible crossing between U* and X*.For both states there is much stronger interaction of the incoming D20 molecules in the transition state since the enthalpies of activation are ca. 20 kJ mol-1 lower and the AS? are ca. 75 J mol-1 K-' lower inS. J. FORMOSINHO AND M. DA G. M. MIGUEL 1753 D,O than in H,O. Even for cations in the ground state very little is known about D,O effects. In V02+ at pH 7 D,O slightly increases the dissociative character of the exchange process.17 However, V has a much smaller radius than U and consequently for the latter an associative mechanism is favoured. D,O is an enhancer of the struc- ture-making or -breaking of the different cations and a similar effect may also be caused in UOi+ through enhancement of the associative character for solvent exchange.lo Kemp et a1.ls have studied the decay of the uranyl excited state at various temperatures in both H,O and D,O, and although they noted a strong isotope effect, the results are not strictly comparable, as under their conditions only single-exponential decay was observed. This may have been a result of either increased excitation intensity or a different excitation wavelength. AUTOQUENCHING AND LASER-INTENSITY EFFECTS Uranyl concentration has also a very marked effect on the characteristics of the fluorescence decay of urany1.l As far as the solvent-exchange rates are concerned an increase in [UOi+] increases the quenching rate k, but leaves k, virtually unaffected at all pH values studied, with the exception of pH 1 where there is no influence on ki., In the U* state with a n3,#; configuration the half-filled fk3 orbital has all the lobes pointing in the equatorial direction and can strongly overlap the same type of empty orbitals of a ground-state UOi+ molecule.This one-electron bond is the main bond for the excimer formed with the two O=U=O axes parallel, since no significant binding can occur with the nu orbitals owing to the fact that they are completely filled in the UO:+ ground-state species. However, in the n3, S; configuration of the X* state, the half-filled SL(f+,) orbital makes an angle of 45" with the equatorial ~ 1 a n e . l ~ The overlap of these orbitals for the formation of the uranyl excimer is much weaker than in the U* state and consequently the possible excimer would have very low binding energy.It is consequently no surprise that uranyl affects only the rates of decay for the U* state (n3,4L) and the solvent-exchange rate ki. At low pH (ca. 1 .O) the high charge on excited and non-excited uranyl ions hinders the formation of the excimer. The temperature studies of the autoquenching rate for the exchange rate ki at pH ca. 3 gives AH? = 46 kJ mol-1 and AS? = 21 J mol-l K-l. When such values are compared with those in table 1 it can be seen that uranyl increases ki by increasing the associative character of the solvent exchange. This effect can be attributed to an increase in the radii of the central atom (bicentre) in the excimers and of the charge of the central atoms; these two factors favour a stronger interaction of the incoming solvent molecule in the transition state.8p9 For pH values of 3.0 and 4.0 an increase in the laser intensity decreases both rates of solvent exchange, ki and kr.lY2 At pH 1 these rates pass through a minimum., (UOi+)* absorbs at 337 nm, similar to the ground state.This may be because of promotion of a second electron nu -+ 4,, S,, as these orbitals are degenerate. Such a transition will strengthen the binding H,O-U, since the empty bonding orbital nu can accept the 0 lone pairs from H,O. Other energy transitions that have electron-transfer character from U to 0 in uranyl will have a similar effect. Absorption of the excited uranyl ion within the laser pulse promotes (UOi+)* to a higher excited state (UOi+)** where the solvation shell acquires a configuration with a slower exchange rate.Although (UOg+)** rapidly relaxes to the lowest electronic states (UOg+)*, the same might not happen with a hydration shell that remembers the higher excited electronic configuration during the (UOE+)* fluorescence decay. At pH 1 the slight increase in ki and k , at high laser intensities could be because of some more significant excitation of a double excited state involving the promotion of a ou electron.1754 PHOTOPHYSICS OF URANYL ION IN AQUEOUS SOLUTIONS HYDROGEN ABSTRACTION FROM WATER EFFECT OF TEMPERATURE AND DEUTERIUM ISOTOPE The decay process characterized by the rates k, and k , at pH 3 can clearly be analysed in terms of the current tunnelling theories of radiationless transitions.According to the model of Jortner,20 when the physical process is characterized in terms of single-mode formalism the temperature dependence of a non-radiative transition has an activationless region at low temperatures and an activation region at high temperatures. The transition temperature, T,, between the two regions is given by P-Pl 4s kTa = where cu is the frequency of the modes involved in the radiationless process, p = A E / h and S = A2/2; AE is the difference in energy of the vibrational levels involved in the transition and A is the horizontal reduced displacement of their minima. The activation energy is given by ?iu)(S-p)2 E, = 4s and the ratio of the frequency factor at high temperatures and the rate in the activationless region is A , exp (- s- YP) with y = In (p/S) - 1 .Good agreement between theory and experiment is obtained for k, and k, (pH 3) throughout the temperature range studied with Tim = 0.1 eV, S = 9.0 and p = 1.2. These are reasonable parameters.20 The most striking feature of excited uranyl decay is the fact that, at low [NO;], k , and k , have a very high transition temperature, T, z 298 K, whereas in organic molecules this temperature is ca. 100 K. In an organic molecule normally only one XH local mode is effective in the radiationless process, but for hydrated [UO,(OH),(H,O),]* w x 5000 cm-l. This shows that in UO:+ the high-frequency local OH modes (v z 3800 cm-l) and also the UO stretching frequency ( v = 745 cm-l) are the most relevant vibrations in these radiationless transitions.Support for the involvement of these modes comes from the heavy-water effect in k,. According to eqn (1) and (2) a decrease in the frequency of the OD mode (v = 2700 cm-l) should decrease T, and E,. The decrease in Ea of 1.33 would correspond to u) = 3770 cm-l if S and p remain the same. Such a value agrees well with the involvement of v(OD)+ v(U0) x 3500 cm-l. T, should also decrease to 225 K and consequently it is not observed under the present experimental conditions. However, for k , there is no clear definition between activation and activationless regions, a situation that is possible for X-D vibrations.20 Further support for the involvement of 0-H and U=O modes also comes from the deuterium isotope effect, which at 50 "C is ca. 2.0 for k, and k,.In contrast for aquo-ions such as Sm1I1 and EulI1, which are de-excited by OH vibrations, D,O decreases the fluorescence lifetimes by 20-50 times4 Theories of radiationless transitions5 show that deuterium isotope effects are quite small if the radiationless reaction coordinate includes vibrations of heavy atoms together with the UO modes.S. J. FORMOSINHO AND M. DA G . M. MIGUEL 1755 ACIDITY EFFECTS Several physical and chemical processes can be conceived for the irreversible decay in the excited uranyl i0n.~9~ One of them is electron transfer from the ligands to the uranium. Such processes can be described in terms of radiationless transition theories, but involve low-frequency phonon modes and, consequently, have very low activation energies,2o contrary to what happens here.The possibility of a purely physical non-radiative transition cannot be excluded under conditions that will be discussed later, but again such processes are expected to have low activation energies (< 15 kJ mol-l) in the temperature range of this study. The most favoured interpretati~n~-~q 21 is hydrogen abstraction from H 2 0 with significant charge-transfer character. The remaining OH radical recombines with the uranium(v) intermediate, producing no overall photochemical reaction. Theoretical support for such a mecha- nism has also been prod~ced.~ Hydrogen abstraction can occur from the more strongly bonded water in the equatorial position or from loose water molecules that approach towards an axial direction. The former situation can be envisaged and 0 II u-0, II I €4 O H in the equatorial field assumes pyramidal geometry that allows a closer approximation of the hydrogen atoms from the U=O bond.For this situation the activation energy should be high, because the distance of approach O-H.--O=U is not very small; in contrast the activation frequency factor can be of the order of the frequency of vibrations (ca. 1013 s-l). Although we ignore completely the detailed structure of the water molecules and the hydrogen bonds around the aquo-ions, it seems that this situation satisfies the data at pH 2. At pH 3 the rates k , and k , attain a minimum; the activation energies are smaller than at pH 2 (ca. 25 kJ mol-l) revealing that the H,O molecules get very close to the U=O bond and approach it with the appropriate geometry (linear approach).Those are probably water molecules coming from outside the first solvent shell towards an axial position. However, the activation frequency should now be much smaller. For the U* state the situation is intermediate between these two extremes at pH 1 and pH 4. However, for the X* state the acti,vation energies are very low (particularly at pH 1, Ea z 13 kJ mol-l) and the pre-exponential factors are also very low. It may be that this process is a physical non-radiative transition. The structure of the water molecules and OH- ions in a more rigid equatorial position may hinder the hydrogen-abstraction reactions, but favour the non-radiative physical transitions with stronger cation-solvent vibrations and higher matrix elements. The uranyl ion affects the rates of decay of the U* state more strongly and consequently also affects the rate k,.The temperature dependence of the autoquenching rate for k , studied at pH 3 gives Ea = 26 kJ mb1-l and A = 4 x 1O1O s-l. This reveals that excimer formation helps hydrogen abstraction from H,O in the U* state and in fact the observed E, and A values approach the values of the same parameters in the polynuclear cations at pH 4. The intensity of the laser leaves the rates k , and k , unaffected at all the pH values studied. This reveals that radiationless conversion from higher excited states is faster than the hydrogen-abstraction reaction and is mainly to the lowest U* and X* excited states.1756 PHOTOPHYSICS OF URANYL ION IN AQUEOUS SOLUTIONS CONCLUSIONS Reversible crossings between two excited states in condensed media are rare, but the delicate balance between all the rates under different conditions of pH, temperature, coordinated anions, uranyl concentration, ionic strength, nature of the solvent, exciting intensity etc.makes the excited uranyl ion an unique case. The quasi-equili- brium between the two lowest states, revealed by the biexponential decay, can be destroyed by several factors, but possibly the most significant of these is the nature of the coordinated ligands. Ligands can stabilize the 7zL& and n",sl, states by increasing or decreasing the energy gap, but are expected to increase the uranyl lifetimes by decreasing hydrogen abstraction from the coordinated waters and quench the reversible decay by decreasing the solvent-exchange rates.Under such conditions it is no surprise that the uranyl ion behaves as a molecular chameleon by presenting so great a variety of contradictory factors. In spite of these results, uranyl will continue to present new and unexpected features, particularly as it has > 20 different Q states in a very small energy region. The present studies reveal that some of these states can play a significant role on the photophysical processes in condensed phases and it is really a matter of luck that a great number of these features are amenable to analysis in terms of only two of those states, although the electronic configurations may be the determining factor. However, it is still not certain how many different low-energy configurations are important.We thank Prof. J. S. Redinha and Prof. H. D. Burrows for many helpful discussions and one of the referees for useful suggestions. We acknowledge also the financial support of INIC. S. J. Formosinho, M. G. Miguel and H. D. Burrows, J. Chem. SOC., Faraday Trans. I , 1984, 80, 1717. M. G. Miguel, S. J. Formosinho, A. C. Cardoso and H. D. Burrows, J. Chem. Soc., Faraday Trans. I , 1984,80, 1735. H. D. Burrows and T. J. Kemp, Chem. Soc. Rev., 1974,3, 139. C. K. Jrargensen and R. Reisfeld, Struct. Bonding (Berlin), 1982, 50, 121. H. D. Burrows and S. J. Formosinho, J. Chem. Soc., Faraday Trans. 2, 1977, 73, 201. Y. Ikeda, S. Soya, H. Fukutomi and H. Tomiyasu, J. Inorg. Nucl. Chem., 1979, 41, 1333. F. Basolo and R. G. Pearson, Mechanisms of Inorganic Reactions. A Study of Metal Complexes in Solution (John Wiley, New York, 2nd edn, 1967), pp. 137 and 172. J. Burgess, Metal Ions in Solution (Ellis Horwood, Chichester, 1978), chap. 11 and p. 198. ' H. P. Bennett0 and E. F. Caldin, J. Chem. SOC., A , 1971, 2198. lo C. K. Jrargensen, J. Lumin., 1979, 18/19, 63. l 1 R. G. Denning, T. R. Snellgrove and D. R. Woodwark, Mol. Phys., 1979, 37, 1109. l 2 0. Ya. Samoilov, Structure of Aqueous Electrolyte Solutions and the Hydration of Ions (Consultants l3 V. A. Shcherbakov and L. L. Shcherbakova, Radiokhimiya, 1976, 18, 207; C. Gorller-Walrand and l4 L. E. Orgel, J. Chem. SOC., 1952, 4756. l5 G. Herzberg, Spectra of Diatomic Molecules (Van Nostrand, Princeton, 1967), p. 214. l6 C. Gorller-Walrand and L. G. Vanquickenborne, J. Chem. Phys., 1972, 57, 1436. l8 A. Cox, T. J. Kemp and W. J. Reed and 0. Traverso, J. Chem. Soc., Faraday Trans. 1,1980,76,804. l9 C . Becker, J . Chem. Educ., 1964,41, 358. 2o J. Jortner, J. Chem. Phys., 1976, 64, 4860; E. Buhks and J. Jortner, J. Phys. Chem., 1980, 84, 3370. 21 M. D. Marcantonatos, J. Chem. Soc., Faraday Trans. I , 1980,76, 1093. Bureau, New York, 1965). W. Colen, Chem. Phys. Lett., 1982, 93, 82. J. Reuben and D. Fiat, J. Am. Chem. SOC., 1969, 91, 4652. (PAPER 3/ 1284)