Article pubs.acs.org/JPCB

Atypical Energetic and Kinetic Course of Excited-State Intramolecular Proton Transfer (ESIPT) in Room-Temperature Protic Ionic Liquids Arpan Manna,† Mhejabeen Sayed,‡ Anil Kumar,† and Haridas Pal*,‡ †

Physical and Material Chemistry Division, National Chemical Laboratory, H. J. Bhabha Road, Pune 411 008, India Radiation & Photochemistry Division, Bhabha Atomic Research Centre, Trombay, Mumbai 400 085, India

‡

S Supporting Information *

ABSTRACT: The excited-state intramolecular proton-transfer (ESIPT) process in 1,8-dihydroxyanthraquinone (18DHAQ) dye has been investigated in protic ionic liquid (PIL) solvents using photochemical measurements. The results demonstrate noteworthy modulations in both steady-state and time-resolved emission characteristics of excited normal (N*) and tautomeric (T*) forms of the dye. That the emission of T* increases unexpectedly upon increasing solvent viscosity indicates that subsequent to the initial forward ESIPT, there is also a relatively slower back ESIPT process involved for the excited dye. It is inferred that the propensity of this back ESIPT process is determined by the dynamics of the diffusive solvent relaxation, a process that is known to be strongly viscositydependent in ionic liquids. Evidence of both forward and back ESIPT for the dye has been obtained from femtosecond fluorescence up-conversion measurements. While an unusually fast forward ESIPT is clearly observed in all of the PILs studied, the uncommon back ESIPT process is distinctly indicated in PIL solvents having lower viscosities, certainly due to reasonably fast diffusive solvent relaxation in these solvents that causes a temporal modulation in the energies of the normal and tautomeric forms within a reasonably short time and thereby brings down the energy of N* compared to that of T*, triggering the back ESIPT process. Observation of solvent-viscosity-dependent back ESIPT is an intriguing finding for the present study as to the best of our knowledge, such a behavior has so far not been reported in the literature for the ESIPT reaction.

1. INTRODUCTION Proton transfer is one of the most fundamental processes occurring in chemical and biological systems.1−4 Studies on PT reactions, especially those involving excited-state intramolecular proton transfer (ESIPT), have engrossed enormous research interests in photochemical sciences, mainly because of the direct relevance of the process to a variety of applications like fluorescent molecular probes,5−8 sensors,9,10 bioimaging and labeling,11,12 luminescent materials,13−16 photostabilizers,17,18 and so forth. Necessary criteria for the chromophoric dyes to show the ESIPT process is to have proton-donating (PD) and a proton-accepting (PA) substituents in the same molecule. In most ESIPT molecules, there is a preexisting intramolecular hydrogen bond in the ground state between the PD and the PA units.4−16 Upon excitation, the acid−base character of the PD and the PA groups in general undergoes a reversal, that is, PD becomes PA and vice versa, causing the initially photoexcited normal (N*) form at higher energy than the conjugate excited tautomeric (T*) form, and accordingly, an ESIPT process takes place along the preexisting hydrogen bond, converting N* to T*.4−16 In most cases, the ESIPT process involves a very low or negligible activation barrier, and hence, the process occurs with an exceptionally fast rate, typically on the subpicosecond time scale. This ultrafast nature of the ESIPT process has attracted the attention of researchers to explore the various factors that © 2014 American Chemical Society

control the dynamics and mechanism of the ESIPT process in different solvent environments.19−22 For dyes where ESIPT occurs with an exceedingly fast rate, the steady-state (SS) emission spectra show only the largely red shifted emission band for T* as the N* emission cannot compete with the ultrafast ESIPT process.7,8,20−24 In many cases, however, both N* and T* emissions can contribute in the observed emission spectra, with the lower-wavelength emission band (LWEB) attributed to N* and a higher-wavelength emission band (HWEB) attributed to T*. The relative intensities of the two bands, however, can undergo large modulations upon changing the conditions of the medium, like solvent polarity, solvent viscosity, experimental temperature, and so forth.7,8,20−26 In the present work, our interest is to study the ESIPT process in 1,8-dihydroxy-9,10-anthraquinone (18DHAQ) dye using room-temperature protic ionic liquids (PILs) as the solvents. Quinones are the important class of chromophoric dyes having extensive occurrence in nature and play many important roles in biology, especially in energy conversion in the living organisms.27−29 Many quinone derivatives have also extensive application in industries, mainly as coloring dyes30−32 and redox reagents.33−35 Hydroxyanthraquinones have long Received: January 10, 2014 Published: February 13, 2014 2487

dx.doi.org/10.1021/jp500281z | J. Phys. Chem. B 2014, 118, 2487−2498

The Journal of Physical Chemistry B

Article

2. EXPERIMENTAL SECTION The 18DHAQ sample was obtained from Tokyo Kasei Kogyo (TCI), Japan. The dye was purified through repeated crystallization from cyclohexane. The purity of the crystallized sample was checked by comparing the absorption and fluorescence spectra of the dye in standard organic solvents.39,40,44−48 The PILs were synthesized using a onestep atom economic method.49,54,57 The details of the synthesis and characterization are given in the Supporting Information (SI) (section S1). Wherever available, the polarity and viscosity values of the studied PILs were obtained from the published literature. 57,58 In the other cases, these values were independently estimated in the present work. Thus, the viscosity values for the required solvents and solvent mixtures were measured by using a Brookfield ultra rheometer, keeping the temperature constant at 298 ± (0.01) K by using a Julabo water bath. Each measurement was repeated three times to check the reproducibility, which was found to be within 1% of the average value noted. For solvent mixtures, their polarity parameters (ETN) were estimated using the following relation

attracted immense research interests for as they belong to the model compounds for the anthracycline antitumor drugs.36−38 Photophysical studies of hydroxy- and amino-substituted anthraquinones have revealed that the excited states of most of these molecules suffer a rapid internal conversion (IC) process, promoted mainly by the presence of intra- and intermolecular hydrogen bondings.39−43 In the extreme cases, some of these anthraquinone derivatives show excited-state intramolecular or intermolecular proton transfer that causes an extremely fast nonradiative deexcitation for the excited states of these dyes.39−41,44,45 SS fluorescence spectra of 18DHAQ in different organic solvents show the presence of two distinct emission bands, one at the lower-wavelength region (LWEB) and the other at the higher-wavelength region (HWEB), and this behavior is suggested to be due to the occurrence of the ESIPT process.39,40,44,45 Reports on the time-resolved (TR) spectroscopic studies on the ESIPT process in quinonoid dyes, including that in 18DHAQ, are not only very scarce but are also limited to conventional organic solvents only.45−48 In the present work, we have investigated the ESIPT process of 18DHAQ dye in a number of PIL solvents independently synthesized in this work based on the 1-methylimidazolium [HmIm]+ cation with various anions, displaying largely different solvent polarity (ETN) and viscosity (η). In the family of ionic liquids (ILs), the PILs form a distinct class of solvents as they possess exchangeable protons and are thus characteristically different than other aprotic ILs, though in general, the PILs also show very low vapor pressures and high viscosities like other common ILs.49−57 For PILs, the presence of exchangeable protons renders Bronsted acidity to these solvents and thus makes them very efficient catalysts for organic reactions like the Mannich reaction, Fischer’s esterification, the Knoevengal reaction, and so forth.50−53 Being entirely composed of cations and anions, these solvents can provide intrinsic conductivity with wide potential windows and can act as anhydrous proton conductors in polymer-membrane-based fuel cells.54−57 Our preliminary studies on 18DHAQ dye in different ILs of aprotic and protic nature indicated that while the relative intensities of the LWEB and HWEB do not show any appreciable variations upon changing the nature of the aprotic ILs used, there is a dramatic change in the intensity ratios of the two bands upon changing the PILs having different polarities and viscosities. Thus, in the present study, we have systematically carried out a detailed investigation in different PILs and PIL mixtures with an aim to understand the effect of both polarity and viscosity of the present solvents on the ESIPT process of 18DHAQ dye. Chemical structures of the dye and the PILs used in the present study are shown in Chart 1 for a quick visualization.

E T N(MS) = fA E T N(A) + fB E T N(B)

(1)

ETN

where fA and f B are the volume fractions and (A) and ETN (B) are the polarity parameters of the cosolvents A and B, respectively.42,43,59 The organic solvents used in the present study were of spectroscopic grade, obtained either from Spectrochem (Mumbai, India) or Fluka (Buchs, Switzerland), and used as received. Absorption spectra were recorded using a JASCO UV−vis spectrophotometer (model V-530, Tokyo, Japan). SS fluorescence spectra were recorded using a Hitachi spectrofluorimeter (model F-4010, Tokyo, Japan). Fluorescence decays of 18DHAQ dye in different PILs and PIL mixtures were measured using a TR spectrofluorimeter (Horiba Jobin Yvon IBH, Scotland, U.K.) that works on the time-correlated singlephoton-counting (TCSPC) principle.60 In the present study, a 445 nm diode laser (∼100 ps, 1 MHz repetition rate) was used as the excitation source and a special photomultiplier tube (PMT)-based detection module supplied by Horiba Jobin Yvon IBH was used for the fluorescence detection. All of the measurements were carried out at the magic angle configuration to eliminate the effect of rotational anisotropy on the observed fluorescence decays. Measured decays were analyzed using a reconvolution procedure.60 Instrument response function (IRF) for the TCSPC setup was obtained by replacing the sample cell with a dilute scatterer solution (suspended TiO2 particles in water). In the decay analysis, the reduced χ2 values (close to unity) and the distribution of the weighted residuals (random distribution among the data channels) were used to judge the goodness of the fits.60 The full width at halfmaximum (fwhm) for a typical IRF for the present TCSPC setup is ∼140 ps. The shortest fluorescence lifetime (τf) thus measurable using the present setup following reconvolution analysis is about 30 ps. In the fluorescence measurements, the absorbance of the sample solution at the excitation wavelengths was always kept quite low, only about 0.1, to avoid any selfabsorption or inner filter effect. All of the absorption and fluorescence measurements in the present study were carried out at ambient temperature (298 ± 1 K). Ultrafast fluorescence kinetic traces in the subpicosecond to picosecond time domain were measured using a femtosecond fluorescence up-conversion instrument (model FOG 100, from

Chart 1. Chemical Structures of 18DHAQ Dye and the Protic PILs Used in the Present Study

2488

dx.doi.org/10.1021/jp500281z | J. Phys. Chem. B 2014, 118, 2487−2498

The Journal of Physical Chemistry B

Article

Figure 1. (A) Ground-state absorption spectra and (B) SS fluorescence spectra of 18DHAQ dye in different PILs studied, [HmIm]propionate (black), [HmIm]acetate (magenta), [HmIm]formate (green), and [HmIm]bisulfate (orange).

CDP Inc., Russia). The details of this experimental setup are given elsewhere.61,62 Briefly, the second-harmonic light (400 nm) of a 50 fs Ti:sapphire laser is used for the excitation of the sample taken in a thin (0.4 mm) rotating cell, and the residual fundamental (800 nm) of the Ti:sapphire laser is used as the gate pulse to up-convert the fluorescence light in a 0.5 mm BBO crystal. The up-converted light is detected using a Hamamatshu PMT (model 5000U-09) operated in the photoncounting mode. The IRF of the present setup is found to have a Gaussian intensity profile with a fwhm of about 220 fs.

excitation, the initially produced excited normal form (N*) undergoes a fast ESIPT process whereby one of the hydroxyl hydrogens (at the 1 or 8 position) is transferred from the hydroxyl group to the quinonoid oxygen along the preexisting hydrogen bond, producing the excited tautomeric form (T*) of the dye. Accordingly, in the SS emission spectra, the LWEB is ascribed to the emission from the N* form, and the HWEB is ascribed to the emission from the T* form of the dye. Thus, the SS emission characteristics of 18DHAQ dye in the studied PILs can be represented by Scheme 1.

3. RESULTS AND DISCUSSION 3.1. Ground-State Absorption and SS Fluorescence Studies. The absorption spectra of 18DHAQ in different PILs are shown in Figure 1A. In all of the solvents, the lowest-energy absorption band for the dye appears very similar, with the peak position at around 430 nm. In the literature, the absorption spectra of the dye in conventional organic solvents are also reported to be similar to those shown in Figure 1A and have been assigned to the S0 to S1 transition of π−π* nature, having some charge-transfer character.39−48 As indicated from Figure 1A, the widths of the absorption spectra are quite identical in all of the PILs studied, an observation very similar to that reported in different organic solvents39−48 and suggest that in all of the solvent systems, the dye in the ground state exclusively exists in a single conformational structure. The SS fluorescence spectra of 18DHAQ in different PILs are shown in Figure 1B. As indicated from this figure, in all of the PIL solvents studied, the fluorescence spectra do not show any mirror-image relationship with the absorption spectra, suggesting that the dye undergoes a large structural change in the excited state compared to its conformational structure in the ground state.20−26,44−48 It is evident from Figure 1B that the emission spectra of the dye in the studied PILs are essentially composed of two emission bands, one with a peak at around 505−515 nm and the other with a peak at around 580− 600 nm, henceforth designated as the LWEB and HWEB, respectively, for the convenience of our discussions. Similar dual emission behavior of 18DHAQ dye has also been reported in the conventional organic solvents and convincingly been attributed to the ESIPT process occurring in the excited dye.5−15,44−48 Thus, drawing an analogy with the reported literature,5−15,44−48 it is suggested from the observed results in Figure 1B that in the ground state, the dye exists exclusively in its normal form (N) where the quinonoid oxygen at 9-position is intramolecularly hydrogen bonded to the hydroxyl groups at 1 and 8 positions of the dye (cf. Chart 1). Upon photo-

Scheme 1. ESIPT in 18DHAQ Occurring along the Preexisting Hydrogen Bond in the Ground State between the Phenolic OH Group and the Quinonoid Oxygen of the Dye

An important point to be noted here is the fact that the emission intensities for the LWEB and HWEB of the dye are largely different in different PILs. Present observation indicates that there are largely different stabilizations for the N* and T* forms of the excited dye in the different PIL solvents studied. It is suggested that following photoexcitation and the subsequent ultrafast ESIPT, there is a substantial degree of modulation in the populations of the N* and T* forms caused by the solvent relaxation processes in different PILs. It is likely that the dissimilarities in the solvent relaxation and the consequential differences in the degree of potential energy (PE) stabilizations of N* and T* forms in different PILs leads to drastically different SS populations of N* and T* forms in different 2489

dx.doi.org/10.1021/jp500281z | J. Phys. Chem. B 2014, 118, 2487−2498

The Journal of Physical Chemistry B

Article

is evident from this figure that as the composition of the higher viscosity/polarity cosolvent [HmIm]bisulfate is increased in the solvent mixture (cf. Table 1), the intensity of the HWEB gradually increases at the expense of the intensity of the LWEB of the dye. The inset of Figure 2 shows the plot of the changes in the intensity ratios for the HWEB to LWEB, measured at 590 and 510 nm, respectively, as a function of the solvent composition. It is evident from this plot that there is a systematic increase in the HWEB intensity in lieu of LWEB as the volume percentage of [Hmim]bisulfate is increased in the solvent mixture. As both the solvent polarity and solvent viscosity change concomitantly among the studied PILs and PIL mixtures (cf. Table 1), observations in Figure 2 indicate that either of the two solvent parameters might play a dominating role in determining the effective SS distribution of the N* and T* populations of the dye in the studied PIL solvents. To understand more about which one of the two solvent properties, namely, the polarity (ETN) or the viscosity (η), is mainly responsible for the observed changes in the emission characteristics for the dye, we tried to correlate the changes in the intensity ratios of the two emission bands with the changing ETN and η values of the solvents. Figure 3A shows these changes in the ratios for the 590−510 nm emission intensity as a function of the ETN values of the solvents. As indicated from this figure, the intensity ratio shows only a nominal change at the lower solvent polarity region, but beyond a critical polarity, the ratio increases very sharply. Such an abrupt change in the trends for the intensities of the two bands seems very unusual because for 18DHAQ dye, we do not expect any large conformational change (e.g., conversion of the planar to twisted structure or so on)63,64 in either of the N* or T* forms of the excited dye that would be triggered by the increasing solvent polarity beyond a certain limit. Moreover, solvent-dependent stabilizations in the energies for the N* and T* forms are expected to follow a linear function with solvent polarity,63,64 and accordingly, the relative populations and hence emission contributions of the T* and N* forms would have changed rather smoothly, possibly following a linear correlation, with the changing ETN values, then displaying a kind of switching in the trends, as indicated in Figure 3A. Thus, we infer that the changing solvent polarities is not the main reason for the observed changes in the relative emission contributions for the T* and N* forms of the dye in different PIL solvents and solvent mixtures. This is further anticipated from the literature reports that the N* form of the studied dye has a higher dipole moment than its T* form.46−48 Therefore, one would expect that upon increasing the solvent polarity, the relative stabilization of the N* form will be more than that of the T* form, and accordingly, the population/emission contributions of the N* form (LWEB) are anticipated to increase in lieu of the T* form (HWEB) of the dye, which is just opposite to fact observed experimentally (cf. Figure 3A). Figure 3B shows the changes in the intensity ratios for the HWEB/LWEB of the dye as a function of the viscosity of the solvents and solvent mixtures studied. As indicated from this plot, for the whole range of viscosities studied, the intensity ratio increases more or less smoothly, following a kind of linear correlation. This observation suggests that the solvent viscosity and, accordingly, the diffusive solvent motions play a major role in determining the relative SS emission contributions of N* and T* forms of the dye in the studied solvents. The most striking observation from Figure 3B is the gradual increase in the

solvents, resulting in the largely different intensities for the LWEB and HWEB of the dye, as indicated in Figure 1B. It is important at this point to understand if the changes in the relative intensities observed for the LWEB and HWEB of the dye in different PIL solvents are related to some solvent property. For a quick comparison, the solvent characteristics, namely, the polarity parameters (ETN) and the solvent viscosity (η) values of the PILs studied, are listed in Table 1. Following Table 1. Solvent Polarity and Viscosity Parameters of the PILs and PIL Mixtures Used in the Present Study PILs [HmIm]propionate [HmIm]acetate [HmIm]formate [HmIm]bisulfate 10% [HmIm]bisulfatea 15% [HmIm]bisulfatea 30% [HmIm]bisulfatea 60% [HmIm]bisulfatea 80% [HmIm]bisulfatea 90% [HmIm]bisulfatea

polarity (ETN)b b

0.50 0.61b 0.78b 1.02b 0.804 0.816 0.852 0.924 0.972 0.996

viscosity (cP)c 5.5 5.6 6.7 406 36 68 175 276 350 385

a

Representatives of the [HmIm]formate and [HmIm]bisulfate solvent mixtures with the volume percent of the [HmIm]bisulfate indicated in the table. bThe polarity (ETN) values for the pure PILs were obtained from ref 57, and those of the mixed solvents were estimated using eq 1. c The viscosity (η) values of first three PILs were obtained from ref 58, and those of the other solvent and solvent mixtures were independently measured in the present work.

Figure 1B and Table 1, it is indicated that the increase in the emission intensity for the HWEB and the consequential decrease in the emission intensity for the LWEB apparently have a concurrence with both the increasing trend of solvent polarity and solvent viscosity. To understand this effect better, we also carried out the SS fluorescence measurements of the dye in a series of [HmIm]formate and [HmIm]bisulfate solvent mixtures, where the above two solvent parameters are systematically varied (cf. Table 1). Fluorescence spectra obtained in these solvent mixtures are shown in Figure 2. It

Figure 2. Changes in the SS fluorescence spectra of 18DHAQ dye in the mixtures of [Hmim]formate and [Hmim]bisulfate solvents. Spectra 1−8 correspond to 0, 10, 15, 30, 60, 80, 90, and 100% [Hmim]bisulfate in the solvent mixtures, respectively. The inset shows the changes in the intensity ratios for the HWEB to LWEB as a function of volume percentage of [HmIm]bisulfate in the solvent mixtures. 2490

dx.doi.org/10.1021/jp500281z | J. Phys. Chem. B 2014, 118, 2487−2498

The Journal of Physical Chemistry B

Article

Figure 3. Changes in the relative intensity of the HWEB to LWEB for 18DHAQ dyes as a function of (A) the ETN values and (B) the viscosities of the PILs and PIL mixtures studied. Open circles correspond to the pure PILs, and closed diamonds correspond to the [Hmim]formate and [Hmim]bisulfate solvent mixtures (cf. Table 1).

Figure 4. Comparison of the fluorescence decays at (A) the LWEB (measured at 510 nm) and (B) the HWEB (measured at 510 nm) of 18DHAQ dye in different PILs and PIL solvent mixtures, as measured using the TCSPC technique (cf. section S2 of the SI). The decays 1−7 correspond to [HmIm]propionate, [HmIm]acetate, [HmIm]formate, 15% [HmIm]bisulfate, 30% [HmIm]bisulfate, 60% [HmIm]bisulfate, and [HmIm]bisulfate (cf. Table 1). The IRFs are also shown in the respective panels.

from picoseconds to nanoseconds, and the longer components in these cases are found to be strongly dependent on the viscosity of the solvent used.65−68 Therefore, one would expect that while in lower viscosity PILs the back ESIPT will be quite efficient (due to faster diffusive solvent relaxation), in very high viscosity PILs, the back ESIPT process will be practically improbable (due to very slow diffusive solvent relaxation), leading to a large increase in the relative population of T*. Support for this proposition is in fact obtained from fluorescence up-conversion results for the dye in different PILs and will be discussed later. In the present context, however, it is evident that the solvent viscosity plays an important role in determining the relative contributions of N* and T* forms of the dye, though a minor role of the solvent polarity on this aspect cannot be ignored completely. 3.2. TR Fluorescence Studies Using TCSPC Measurements. TR fluorescence measurements for 18DHAQ dye in different PILs were carried out using a picosecond TCSPC setup to understand the role of the solvent properties on the fluorescence decay kinetics of the dye. From the measured fluorescence decays, it is observed that in all of the PILs, there is always a long decay tail in the kinetic traces in addition to the major contribution arising from an unusually fast decay component. It is important to mention here that all of the PILs used in the present work show very weak inherent fluorescence in the spectral region of 18DHAQ emission, which is reported to be characteristic of the imidazolium-based ILs.69,70 Though this inherent fluorescence from the PILs is quite weak to cause any significant change in the spectral features of the SS emission of the dye, its presence significantly

relative contributions of the T* form of the dye as the solvent viscosity is increased. This is certainly an unusual observation because as per our common perception, the rate of conversion of N* to T* would have been retarded upon increasing the solvent viscosity if solvent diffusive motions were coupled to the forward ESIPT process. To rationalize the observed results, thus, we need to consider that subsequent to the forward ESIPT process, there is some back transfer of T* to N* by participation of a back ESIPT process, and the latter is actually controlled by the solvent diffusive motions rather than the former. This, in other words, suggests that the forward ESIPT occurs with an unusually fast rate7,8,20−26 and hence without any influence of the solvent diffusion. The back ESIPT process, however, occurs with a relatively slower rate,46−48 influenced largely by the solvent diffusive motions, and accordingly, the effectiveness of the back ESIPT reduces largely upon increasing the solvent viscosity, resulting in an increase in the contribution of the T* form, as indicated in Figures 2 and 3B. As one can understand, the guiding force for the back ESIPT process will actually be the differential solvent stabilizations (PE stabilizations) for the N* and T* forms, caused by the diffusive solvent relaxations, the latter being known to be reasonably slow for ILs.65−68 Because the dipole moment of N* is higher than that of T*,46−48 as long as the solvent relaxation rate is competitive with the deexcitation rates of N* and T*, the former will cause a more time-dependent stabilization for N* than T*, sustaining the back ESIPT process. The rationale behind this assertion is that for ILs, the solvent relaxation process in general occurs for a very long period of time span, with time constants for the relaxation components ranging 2491

dx.doi.org/10.1021/jp500281z | J. Phys. Chem. B 2014, 118, 2487−2498

The Journal of Physical Chemistry B

Article

modulates the observed fluorescence decay traces recorded using TCSPC measurements, incorporating additional components in the decays. In fact, in the TCSPC measurements, it is observed that the PILs alone show quite nonexponential fluorescence decays accompanying very long decay tails (cf. Figure S2, SI). Interestingly, these decay tails for the PILs alone appear to be very similar to the ones observed in the decays measured for the dye in the PILs (cf. Figure S2, SI). Thus, it is evident that the decay tails that appear in the fluorescence kinetic traces for 18DHAQ dye are actually due to the inherent emissions of the PIL solvents involved. Therefore, to recover the true fluorescence decays of the dye in different PILs, we subtracted the tail-matched decays of the PILs alone from the observed decays of the dye in the respective solvents and wavelengths (cf. Figure S2, SI). The true fluorescence decays thus obtained at the LWEB and HWEB of the dye in different PIL solvents and solvent mixtures are shown in Figure 4A and B, respectively. As indicated from Figure 4, the decays are, in general, very fast in all if the PILs studied at both the LWEB and HWEB of the dye. Present observation suggests that the deexcitation rate for both N* and T* forms of the excited dye are exceedingly fast, suggestive of the involvement of an unusually fast nonradiative deexcitation process. In many hydroxy- and amino-substituted quinines, the fluorescence decays are found to be exceptionally fast, and the reason is ascribed to the involvement of an unusually fast IC process, the latter being strongly coupled with the motions associated with the intramolecular and intermolecular hydrogen bonds present in these molecules.39−43 The recovered fluorescence decays of the 18DHAQ dye in all of the PILs and PIL mixtures (cf. Figure 4) were seen to fit reasonably well with a single-exponential function. The fluorescence lifetime values thus obtained in different cases are listed in Table 2.

the time scales probed by TCSPC measurements. The other important observation from Figure 4 and Table 2 is that the fluorescence decays for the dye gradually become slower as the solvent viscosity is increased (cf. Table 1). This observation suggests that the inherent decays of both N* and T* forms of the dye are largely controlled by the diffusive solvent motions around the excited dye. To validate it further, we have also plotted the τf values of the dye at both the LWEB and HWEB as a function of the ETN values and also as a function of the viscosity of the solvents studied, as shown in Figure 5A and B, respectively. As indicated from Figure 5A, the changes in the τf values show an unexpected break in the plot at some intermediate polarity, quite similar to that observed in Figure 3A. Such a behavior is quite impracticable because no major solvent-polarity-induced conformational change (planar to twisted etc.) is expected in either the N* or T* form of the excited dye.63,64 As indicated from Figure 5B, the changes in the τf values show quite a linear correlation with the viscosity of the solvents used. This observation in combination with the results in Figure 3B clearly suggest the strong involvement of the solvent diffusive motions in determining not only the population distributions of the N* and T* forms but also their deexcitation kinetics in the studied PIL solvents. Support for the involvement of solvent diffusion in determining the effective populations of N* and T* is also indicated from the studies of Smulevich et al.,71 where it was observed in conventional organic solvents that while the SS emission spectra of the dye at room temperature have a much higher intensity for LWEB, the spectra almost exclusively exhibit the HWEB in a Spolski matrix at very low temperature (10 K) with very negligible emission for the LWEB. Thus, from the present results, we infer that the relative populations of the N* and T* forms of 18DHAQ dye in PILs undergo large time-dependent modulations at the earlier time scales before attaining a SS situation as accessed by the TCSPC measurements, and these modulations are mainly governed by the solvent diffusive relaxation guided back ESIPT process, the propensity of which decreases sharply upon increasing the solvent viscosity, hence showing almost a linear increase in the relative intensity of the HWEB compared to that of the LWEB.46−48,65−68,71−73 3.3. Ultrafast Fluorescence Studies Using Up-Conversion Measurements. Ultrafast fluorescence kinetic traces for 18DHAQ dye in different PILs were recorded using femtosecond fluorescence up-conversion measurements to understand the time evolutions of N* and T* forms of the dye at the early time scales. These measurements were carried out at two selected emission wavelengths, namely, 510 (LWEB) and 590 nm (HWEB), to monitor the N* and T* emissions, respectively. The kinetic traces thus obtained for the dye at the two monitoring wavelengths in different PILs are shown in Figure 6A and B, respectively, mainly zooming in on the initial part of the kinetic traces. The traces as actually recorded for relatively longer time spans (up to about 35 ps) are shown in Figure S3A and B of the SI. In all cases, a triexponential function was uniformly used to analyze the observed traces, to obtain good fits for the traces, and also to find logical correlations among the estimated time constants at the two measuring wavelengths and in different solvent systems (cf. Figure S3 of SI). The time constants of different decay/growth components (τi) and their absolute contributions (Ai), as obtained from the analysis of the up-conversion traces at the LWEB and HWEB of the dye in different PIL solvents and solvent mixtures, are listed in Table 3. It should be mentioned

Table 2. Fluorescence Lifetimes (τf) of 18DHAQ Estimated at the HWEB (515 nm) and LWEB (590 nm) of the Dye in Different PILs Using TCSPC Measurementsa solvent

η (cP)

[HmIm]propionate

5.5

[HmIm]acetate

5.6

[HmIm]formate

6.7

15% [HmIm]bisulfate

55

30% [HmIm]bisulfate

120

60% [HmIm]bisulfate

255

[HmIm]bisulfate

406

λem (nm)

τf (ps)b

515 590 515 590 515 590 515 590 515 590 515 590 515 590

59 64 66 62 70 65 114 113 146 150 286 290 422 430

a

Samples were excited with a 445 nm diode laser. bError limits in the estimated fluorescence lifetimes (τf) are within ±5%.

The important observation from Table 2 is that in all of the solvents studied, the τf values are very close to each other at both the LWEB (510 nm) and HWEB (610 nm) of the dye. This observation suggests that the deexcitation rates for both N* or T* forms of the dye are very similar, indicating a kind of kinetic equilibrium between the two forms of the excited dye at 2492

dx.doi.org/10.1021/jp500281z | J. Phys. Chem. B 2014, 118, 2487−2498

The Journal of Physical Chemistry B

Article

Figure 5. Changes in the τf values for 18DHAQ dye with (A) the ETN values and (B) the viscosities of the PILs and PIL mixtures studied. The symbols ○ and ◇ represent the τf values estimated at the LWEB and HWEB of 18DHAQ dye.

Figure 6. Comparison of the fluorescence kinetic traces (for a shorter time span) as measured at (A) the LWEB (recorded at 510 nm) and (B) the HWEB (recorded at 590 nm) of 18DHAQ dye in different PILs solvents and solvent mixtures using the up-conversion technique. The traces 1−7 correspond to [HmIm]propionate, [HmIm]acetate, [HmIm]formate, 15% [HmIm]bisulfate, 30% [HmIm]bisulfate, 60% [HmIm]bisulfate, and [HmIm]bisulfate (cf. Table 1). The IRFs are also shown in the respective panels.

here that in these analyses, we justifiably kept the longest time constant (τ3) fixed as equal to the average τf values estimated from TCSPC measurements (cf. section 3.2, Table 2). As indicated from Table 3, for all of the kinetic traces recorded at 510 nm, irrespective of the solvents used, there is an ultrafast decay component τ1, having a time constant in the range of 0.2−0.27 ps. On the basis of the fact that the ESIPT process in most of the ESIPT dyes occurs with an unusually fast rate,7,8,19−26 the exceptionally short τ1 decay component observed at 510 nm is justifiably attributed to the ultrafast forward ESIPT process (conversion of initially photoproduced N* to T*). This assignment is further supported by the observation that in all of the solvents, the kinetic traces at 590 nm (HWEB) display a exceptionally short τ1 growth component that corresponds very nicely with the τ1 decay component observed for the 510 nm traces. A point to be noted from Table 3 is that the τ1 value apparently increases to a small extent as the solvent viscosity is increased (cf. Table 1). Though we cannot give much emphasis to this increase because the changes are almost within the time resolution of the present experimental setup (IRF ≈ 220 fs), a small effect of viscosity of the PIL solvents on the ultrafast forward ESIPT process cannot be ruled out completely. The involvement of an unusually fast ESIPT process in the studied system suggests that the initial N* to T* conversion for 18DHAQ dye essentially occurs along a barrierless PE surface.7,8,19−26 Such a situation can schematically be presented as in Figure 7, where the uppermost curve in the middle panel qualitatively represents the PE changes for the excited dye as a function of the ESIPT coordinate, without involving any energy

Table 3. Time Constants (τi), Their Relative Contributions (Ai) for 18DHAQ Dye Estimated at the LWEB and HWEB in Different PILs Using Fluorescence Up-Conversion Measurementsa ILs [HmIm] propionate [HmIm] acetate [HmIm] formate 15% [HmIm] bisulfate 30% [HmIm] bisulfate 60% [HmIm] bisulfate [HmIm] bisulfate

λem (nm)

A1

τ1 (ps)b

A2

τ2 (ps)b

A3 (ps)

τ3 (ps)b

510

112.1

0.22

39.4

3.4

46.6

60

590 510

−27.9 105.4

0.20 0.21

51.6 41.9

3.4 3.3

83.9 48.1

60 60

590 510

−18.1 85.6

0.19 0.24

51.3 38.8

3.5 3.3

83.6 54.1

60 60

590 510

−25.8 95.2

0.29 0.24

49.8 39.4

3.9 3.3

84.7 52.5

60 90

590 510

−24.8 83.2

0.23 0.26

27.1 43.0

5.2 3.5

101.6 54.6

90 135

590 510

−21.8 62.7

0.25 0.26

24.1 25.2

8.5 3.4

103.2 78.7

135 290

590 510

−19.9 48.5

0.27 0.27

2.5 26.5

18 3.2

123.0 83.4

290 425

590

−19.0

0.27

2.4

22

122.3

425

a

Samples were excited with 400 nm laser pulses. bIn the analysis of the up-conversion traces, the τ3 component was kept fixed similar to the τf values estimated in different solvents using TCSPC measurements. Error limits in the estimated time constants are within ±5%. 2493

dx.doi.org/10.1021/jp500281z | J. Phys. Chem. B 2014, 118, 2487−2498

The Journal of Physical Chemistry B

Article

Figure 7. Qualitative PE diagrams to represent the solvent relaxation guided differential stabilization of the excited N* and T* forms of 18DHAQ dye and the consequential involvement or noninvolvement of the back ESIPT process. The left panel shows the situation prevailing for the lower viscosity PILs, and the right panel shows the situation prevailing for the high viscosity PILs.

Figure 8. Correlation of the τ2 decay component at 590 nm with (A) the viscosity (η) and (B) the polarity (ETN) of the solvents studied. While the τ2 correlates more or less linearly with the solvent viscosity, it does not show a single correlation for the whole range of the solvent polarities studied.

As reported in the literature,46−48 the dipole moment possessed by the N* form of the dye is higher than that of the T* form. One can therefore expect that subsequent to the photoexcitation and ultrafast forward ESIPT, there will be a significant time-dependent stabilization in the energies for both N* and T* forms caused by the diffusive solvent relaxation around the excited dye, and this stabilization will be relatively more pronounced for the N* form than the T* form as the dipole moment is higher for the former species. Qualitatively, this can be represented as in the left and/or right panels in Figure 7, where the reduction in the PE for the N* form is shown to be more than that of the T* form as time progresses. In the left panel, it is assumed that the diffusive solvent relaxation is reasonably fast such that the differential stabilization eventually brings down the N* energy lower than that of T* in a reasonably short time. Such a modulation would thus assist the back ESIPT process to occur quite efficiently, converting a large part of the previously formed T* back to N*. On the other hand, in the right panel, it is assumed that the diffusive solvent relaxation is significantly slow such that the N* energy never becomes lower than that of T* within the excited-state lifetime of the dye. Under such a situation, the back ESIPT process can either occur with an extremely low efficiency or may not occur at all, and hence, the majority of the excited dye population will effectively remain in the T* form,

barrier. According to this presentation, following photoexcitation, the dye is initially placed at a very high PE configuration of N* where the excited dye undergoes quick conformational changes along ESIPT coordinate and thus slides down the barrierless PE curve, introducing an ultrafast forward ESIPT process. It is expect that subsequent to the photoexcitation, there could be a significant time-dependent reorganization of the solvents around the excited dye, and accordingly, the energies of both N* and T* forms would also undergo a timedependent modulation. For the present systems, as the τ1 growth component at 590 nm always matches with the τ1 decay component at 510 nm in all of the PIL solvents studied, it is quite intuitive to assume that during the ultrafast ESIPT, there is hardly any solvent relaxation guided modulation in the PE surfaces for the excited dye. This is also in accordance with the reported fact that in ILs, the ultrafast inertial component of the solvation contributes only to a small extent toward the overall solvent relaxation dynamics.65−68 Therefore, we can justifiably assume that during the ultrafast forward ESIPT process, the energy of N* essentially remains higher than that of T* (cf. the middle panel in Figure 7), and the modulations in the energies of the N* and T* forms, if any, effectively occur during the diffusive solvent relaxation (cf. left and/or right panel in Figure 7), which is known to be quite slow in the ILs.65−68 2494

dx.doi.org/10.1021/jp500281z | J. Phys. Chem. B 2014, 118, 2487−2498

The Journal of Physical Chemistry B

Article

diffusive solvent relaxation guided modulation in the ESIPT process of the studied dye. To the best of our knowledge, such a viscosity-dependent modulation in the energetics and kinetics of the ESIPT process has so far not been reported in the literature.

giving higher intensity for the HWEB than that of the LWEB, as observed in the high viscosity solvents (cf. Figure 2). For the 510 nm kinetic traces, in addition to the ultrafast τ1 decay component, there is a second relatively slower decay component τ2 with time constants in the range of 3.2−3.5 ps in all of the solvents studied. The presence of this τ2 decay component at 510 nm clearly suggests that in the studied PIL solvents, the forward ESIPT process in 18DHAQ dye is not a single-step ultrafast process (corresponding to τ1) but actually occurs following a non-single-exponential kinetics that can be effectively fitted in terms of the two decay components τ1 and τ2. An interesting point to note from the up-conversion results is that related to the τ2 decay component at 510 nm, there is no corresponding growth component present in the kinetic traces at 590 nm. On the contrary, we actually find a distinctly different decay component τ2 in the traces at 590 nm that clearly does not match with the τ2 component observed at 510 nm. Moreover, unlike the τ2 component at 510 nm, the τ2 component at 590 nm shows a clear and significant increase upon increasing the viscosity of the solvents. In fact, a plot of these τ2 values at 590 nm with the viscosity of the solvents shows a reasonably linear correlation, as shown in Figure 8A, suggesting that the diffusive solvent motion has a direct influence on the observed τ2 component at 590 nm. On the basis of the schematics shown in Figure 7, we feel that this τ2 decay component at 590 nm is due to the diffusive solvent relaxation guided back ESIPT process. Because diffusive solvent relaxation in ILs is strongly dependent on the viscosity of the solvents,65−68 upon increasing solvent viscosity, the PE surfaces of the excited dye undergo increasingly fewer modulations, and accordingly, the back ESIPT process gradually becomes less prominent. It should be mentioned that the changes in the τ2 decay component at 590 nm do not show a single correlation with the ETN values of the solvents but demonstrate a break at some intermediate polarity (cf. Figure 8B). Because no major structural change (planar to nonplanar etc.) was expected for 18DHAQ dye upon changing the solvent polarity,63,64 such an observation suggests that the solvent polarity is not responsible for the changes in the τ2 decay constants observed at 590 nm in different PIL solvents. It is thus evident from the observed upconversion results that solvent viscosity plays a significant role in modulating the back ESIPT process. It is apparent from the fluorescence up-conversion results that in the PIL solvents with relatively lower viscosities, the back ESIPT process is highly supported. In the PIL solvents of very high viscosities, however, the propensity of the back ESIPT process is so exceedingly reduced that it hardly contributes in converting T* back to N* within the excitedstate lifetime of the dye. It should be recalled here that the participation of a back ESIPT process that largely modulates with the changing viscosity of the PIL solvents has also been indicated from the SS fluorescence results, as already discussed in section 3.1. Thus, fluorescence up-conversion results nicely corroborate with the observations made from the SS fluorescence measurements for the dye in different PIL solvents and intuitively demonstrate how the adiabatic PE surface coupling the N* and T* forms of the dye undergoes changes with the progress of the diffusive solvent relaxation and thereby causes a temporal modulation in the populations of the N* and T* forms of the dye as the time elapses following the photoexcitation of the dye. Observation of the back ESIPT process for 18DHAQ dye in PIL solvents is certainly the most fascinating result from the present work, demonstrating the

4. CONCLUSION Investigations using steady-state (SS) and time-resolved (TR) fluorescence measurements on 1,8-dihydroxyanthraquinone (18DHAQ) dye in protic ionic liquids (PILs) unravel interesting new aspects of excited-state intramolecular proton transfer (ESIPT) kinetics of the dye. The observation that the SS fluorescence intensity ratio for the tautomeric (T*) to normal (N*) forms of the excited dye increases with solvent viscosity is an unusual finding, and it impels us to invoke the participation of the back ESIPT process for the excited dye subsequent to its ultrafast forward ESIPT process, the former being considered as strongly viscosity-dependent. Interestingly, the evidence for both the forward and the back ESIPT process has been convincingly obtained from the kinetic traces recorded using femtosecond fluorescence up-conversion measurements. From the observed results, it is clearly indicated that the back ESIPT can contribute quite efficiently only in the lower viscosity PILs, while the process is very negligible in a very high viscosity PIL solvent. As the dipole moment of N* is higher than that of T*, it is inferred based on the experimental findings and the literature reports on the viscosity dependence of solvent diffusive relaxation dynamics in ILs that unlike the ultrafast forward ESIPT, the relatively slower back ESIPT process is mainly governed by the differential stabilizations of N* and T* forms caused by the diffusive solvent relaxations around the excited dye. As the diffusive solvent relaxation is reasonably faster in lower viscosity PILs, the potential energy (PE) of the initially produced N* can quickly come down below that of T*, triggering the back ESIPT process, though immediately after photoexcitation, the higher-energy N* supports the ultrafast forward ESIPT process. In very high viscosity PILs, as the diffusive solvent relaxation is extremely slow, it cannot bring down the N* energy below that of T* within the lifetime of the excited dye, and hence, back ESIPT becomes quite inefficient or improbable. Accordingly, qualitative PE diagrams have been presented to envisage the solvent relaxation guided stabilizations of N* and T* forms and the consequential participation of the back ESIPT depending on the viscosity of the PILs studied. The present study clearly demonstrates the involvement of the back ESIPT process for 18DHAQ dye from the ultrafast fluorescence measurements of the N* and T* emissions in PIL solvents, which could not be unambiguously established earlier in conventional organic solvents using photochemical measurements. It is evident from the present study that not only the ultrafast forward ESIPT but also a reasonably fast back ESIPT plays a significant role in determining the relative contributions of N* and T* forms of the excited dye, especially in the lower viscosity solvents, and this back ESIPT is the main reason for the unexpectedly higher contributions of N* emissions observed in most of the solvents including conventional organic solvents, even though the forward ESIPT process for the dye is known to be unusually fast. 2495

dx.doi.org/10.1021/jp500281z | J. Phys. Chem. B 2014, 118, 2487−2498

The Journal of Physical Chemistry B

■

ASSOCIATED CONTENT

■

AUTHOR INFORMATION

Article

Sensitive Dual-Fluorescence Labels For Monitoring Peptide−Nucleic Acid Interactions. Bioconjugate Chem. 2011, 22, 101−107. (12) Kim, T.; Kang, H. J.; Han, G.; Chung, S. J.; Kim, Y. A Highly Selective Fluorescent ESIPT Probe for the Dual Specificity Phosphatase MKP-6. Chem. Commun. 2009, 5895−5897. (13) Mutai, T.; Tomoda, H.; Ohkawa, T.; Yabe, Y.; Araki, K. Switching of Polymorph-Dependent ESIPT Luminescence of an Imidazo[1,2-a]pyridine Derivative. Angew. Chem., Int. Ed. 2008, 47, 9522−9524. (14) Rode, M. F.; Sobolewski, A. L. Effect of Chemical Substituents on the Energetical Landscape of a Molecular Photoswitch: An Ab Initio Study. J. Phys. Chem. A 2010, 114, 11879−11889. (15) Tang, K.; Chang, M.; Lin, T.; Pan, H.; Fang, T.; Chen, K.; Hung, W.; Hsu, Y.; Chou, P. Fine Tuning the Energetics of ExcitedState Intramolecular Proton Transfer (ESIPT): White Light Generation in a Single ESIPT System. J. Am. Chem. Soc. 2011, 133, 17738−17745. (16) Kwon, J. E.; Park, S. Y. Advanced Organic Optoelectronic Materials: Harnessing Excited-State Intramolecular Proton Transfer (ESIPT) Process. Adv. Mater. 2011, 23, 3615−3642. (17) Yang, P.; Zhao, J.; Wu, W.; Yu, X.; Liu, Y. Accessing the LongLived Triplet Excited States in Bodipy-Conjugated 2-(2-Hydroxyphenyl) Benzothiazole/Benzoxazoles and Applications as Organic Triplet Photosensitizers for Photooxidations. J. Org. Chem. 2012, 77, 6166−6178. (18) Keck, J.; Kramer, H. E. A.; Port, H.; Hirsch, T.; Fischer, P.; Rytz, G. J. Investigations on Polymeric and Monomeric Intramolecularly Hydrogen-Bridged UV Absorbers of the Benzotriazole and Triazine Class. J. Phys. Chem. 1996, 100, 14468−14475. (19) Hsieh, C. C.; Chou, P. T.; Shih, C. W.; Chuang, W. T.; Chung, M. W.; Lee, J. H.; Joo, T. H. Comprehensive Studies on an Overall Proton Transfer Cycle of the ortho-Green Fluorescent Protein Chromophore. J. Am. Chem. Soc. 2011, 133, 2932−2943. (20) Hojniak, A. S.; Wisniewski, L.; Deperasinska, I.; Makarewicz, A.; Jerzykiewicz, L.; Puszko, A.; Erez, Y.; Huppert, D. The Impact of Solvent Polarity on Intramolecular Proton and Electron Transfer in 2Alkylamino-4-nitro-5-methyl Pyridine N-Oxides. Phys. Chem. Chem. Phys. 2012, 14, 8147−8159. (21) Barroso, M.; Arnaut, L. G.; Formosinho, S. J. Absolute Rate Calculations. Proton Transfers in Solution. J. Phys. Chem. A 2007, 111, 591−602. (22) Ishikawa, H.; Iwata, K.; Hamaguchi, H. Picosecond Dynamics of Stepwise Double Proton-Transfer Reaction in the Excited State of the 2-Aminopyridine/Acetic Acid System. J. Phys. Chem. A 2002, 106, 2305−2312. (23) Wu, Y.; Lawson, P. D. V.; Henary, M. M.; Schmidt, K.; Bredas, J. L.; Fahrni, C. J. Excited State Intramolecular Proton Transfer in 2-(2′Arylsulfonamidophenyl)benzimidazole Derivatives: Insights into the Origin of Donor Substituent-Induced Emission Energy Shifts. J. Phys. Chem. A 2007, 111, 4584−4595. (24) Huang, J.; Peng, A. D.; Fu, H. B.; Ma, Y.; Zhai, T. Y.; Yao, J. N. Temperature-Dependent Ratiometric Fluorescence from an Organic Aggregates System. J. Phys. Chem. A 2006, 110, 9079−9083. (25) Yushchenko, D. A.; Shvadchak, V. V.; Klymchenko, A. S.; Duportail, G.; Pivovarenko, V. G.; Mely, Y. Modulation of ExcitedState Intramolecular Proton Transfer by Viscosity in Protic Media. J. Phys. Chem. A 2007, 111, 10435−10438. (26) Das, K.; English, D. S.; Petrich, J. W. Solvent Dependence on the Intramolecular Excited-State Proton or Hydrogen Atom Transfer in Hypocrellin. J. Am. Chem. Soc. 1997, 119, 2763−2764. (27) Garrett, R. H.; Grisham, C. M. Biochemistry, 3rd ed.; Thomson Brooks/Cole: Belmont, CA, 2005; pp 645−657 and pp 681−685. (28) Rich, P. R. A Generalized Model for the Equilibration of Quinone Pools with Their Biological Donors and Acceptors in Membrane-Bound Electron Transfer Chains. FEBS Lett. 1981, 130, 173−178. (29) Quan, M.; Sanchez, D.; Wasylkiw, M. F.; Smith, D. K. Voltammetry of Quinones in Unbuffered Aqueous Solution: Reassessing the Roles of Proton Transfer and Hydrogen Bonding in

S Supporting Information *

Synthesis and characterization of the PILs, additional figures on time-correlated single-photon-counting measurements, and correlation of time constants for back ESIPT with solvent viscosity are given. This material is available free of charge via the Internet at http://pubs.acs.org. Corresponding Author

*E-mail: [email protected]. Tel: 91-22-2559 5396. Fax: 91-222550 5151 and 2551 9613. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors gratefully acknowledge the National Chemical Laboratory, Pune, and the Bhabha Atomic Research Centre, Mumbai, for the generous support provided during the course of the present work. A.M. is grateful to CSIR New Delhi for awarding him a Research Fellowship for carrying out this work. A.K. thanks the Department of Science and Technology (DST), New Delhi, for a J. C. Bose National Fellowship (SR/ S2/JCB-26/2009).

■

REFERENCES

(1) Jaramillo, P.; Coutinho, K.; Canuto, S. Solvent Effects in Chemical Processes. Water-Assisted Proton Transfer Reaction of Pterin in Aqueous Environment. J. Phys. Chem. A 2009, 113, 12485− 12495. (2) Malmstrom, B. G. Cytochrome c Oxidase as a Redox-Linked Proton Pump. Chem. Rev. 1990, 90, 1247−1260. (3) Cui, G.; Lan, Z.; Thiel, W. Intramolecular Hydrogen Bonding Plays a Crucial Role in the Photophysics and Photochemistry of the GFP Chromophore. J. Am. Chem. Soc. 2012, 134, 1662−1672. (4) Paul, B. K.; Guchhait, N. Modulated Photophysics of an ESIPT Probe 1-Hydroxy-2-naphthaldehyde within Motionally Restricted Environments of Liposome Membranes Having Varying Surface Charges. J. Phys. Chem. B 2010, 114, 12528−12540. (5) Wu, J. S.; Liu, W. M.; Ge, J. C.; Zhang, H. Y.; Wang, P. F. New Sensing Mechanisms for Design of Fluorescent Chemosensors Emerging in Recent Years. Chem. Soc. Rev. 2011, 40, 3483−3495. (6) Deperasinska, I.; Gryko, D. T.; Karpiuk, E.; Kozankiewicz, B.; Makarewicz, A.; Piechowska, J. 12-Hydroxy-1-azaperylene-Limiting Case of the ESIPT System: Enol−Keto Tautomerization in S0 and S1 States. J. Phys. Chem. A 2012, 116, 2109−2116. (7) Zhao, J.; Ji, S.; Chen, Y.; Guo, H.; Yang, P. Excited State Intramolecular Proton Transfer (ESIPT): From Principal Photophysics to the Development of New Chromophores and Applications in Fluorescent Molecular Probes and Luminescent Materials. Phys. Chem. Chem. Phys. 2012, 14, 8803−8817. (8) Park, S.; Kwon, J. E.; Park, S. Y. Strategic Emission Color Tuning of Highly Fluorescent Imidazole-Based Excited-State Intramolecular Proton Transfer Molecules. Phys. Chem. Chem. Phys. 2012, 14, 8878− 8884. (9) Landge, S. M.; Tkatchouk, K.; Benitez, D.; Lanfranchi, D. A.; Elhabiri, M.; Goddard, W. A., III; Aprahamian, I. Isomerization Mechanism in Hydrazone-Based Rotary Switches: Lateral Shift, Rotation, or Tautomerization? J. Am. Chem. Soc. 2011, 133, 9812− 9823. (10) Klymchenko, A. S.; Demchenko, A. P. Electrochromic Modulation of Excited-State Intramolecular Proton Transfer: The New Principle in Design of Fluorescence Sensors. J. Am. Chem. Soc. 2002, 124, 12372−12379. (11) Zamotaiev, O. M.; Postupalenko, V. Y.; Shvadchak, V. V.; Pivovarenko, V. G.; Klymchenko, A. S.; Mely, Y. Improved Hydration2496

dx.doi.org/10.1021/jp500281z | J. Phys. Chem. B 2014, 118, 2487−2498

The Journal of Physical Chemistry B

Article

the Aqueous Electrochemistry of Quinones. J. Am. Chem. Soc. 2007, 129, 12847−12856. (30) Hattori, M. In Kirk−Othmer Encyclopedia of Chemical Technology, 5th ed.; Seidel, A., Ed.; John Wiley & Sons: Hoboken, NJ, 2004; Vol. 9, pp 300−349. (31) Miyazaki, K.; Tabata, I.; Hori, T. Effects of Molecular Structure on Dyeing Performance and Colour Fastness of Yellow Dyestuffs Applied to Polypropylene Fibres in Supercritical Carbon Dioxide. Coloration Technol. 2012, 128, 51−59. (32) Rembold, M. W.; Kramer, H. E. A. The Role of Anthraquinonoid Dyes in the ‘Catalytic Fading’ of Dye Mixtures. Substituent−Dependent Triplet Yield of Diaminoanthraquinones. J. Soc. Dyers Colour. 1980, 96, 122−126. (33) Abhayawardhana, A. D.; Sutherland, T. C. Heterogeneous Proton-Coupled Electron Transfer of an Aminoanthraquinone SelfAssembled Monolayer. J. Phys. Chem. C 2009, 113, 4915−4924. (34) Campos-Martin, J. M.; Blanco-Brieva, G.; Fierro, J. L. G. Hydrogen Peroxide Synthesis: An Outlook beyond the Anthraquinone Process. Angew. Chem., Int. Ed. 2006, 45, 6962−6984. (35) Eul, M.; Moeller, A.; Steiner, N. In Kirk−Othmer Encyclopedia of Chemical Technology, 5th ed.; Seidel, A., Ed.; John Wiley & Sons: Hoboken, NJ, 2004; Vol. 14, pp 42−51. (36) Huang, Q.; Lu, G.; Shen, H.-M.; Chung, M. C. M.; Choon, N. O. Anti-Cancer Properties of Anthraquinones from Rhubarb. Med. Res. Rev. 2007, 27, 609−630. (37) Doi, A. M.; Irwin, R. D.; Bucher, J. R. Influence of Functional Group Substitutions on the Carcinogenicity of Anthraquinone in Rats and Mice: Analysis of Long-Term Bioassays by the National Cancer Institute and the National Toxicology Program. J. Toxicol. Environ. Health, Part B 2005, 8, 109−126. (38) Lavie, G.; Barliya, T.; Mandel, M.; Blank, M.; Ron, Y.; Orenstein, A.; Livnat, T.; Friedman, N.; Weiner, L.; Sheves, M.; Weinberger, D. Competitive Quenching: A Mechanism by Which Perihydroxylated Perylenequinone Photosensitizers Can Prevent Adverse Phototoxic Damage Caused by Verteporfin during Photodynamic Therapy. Photochem. Photobiol. 2007, 83, 1270−1277. (39) Flom, S. R.; Barbara, P. F. Proton Transfer and Hydrogen Bonding in the Internal Conversion of S1 Anthraquinones. J. Phys. Chem. 1985, 89, 4489−4494. (40) Palit, D. K.; Pal, H.; Mukherjee, T.; Mittal, J. P. Photodynamics of the S1 State of Some Hydroxy- and Amino-Substituted Naphthoquinones and Anthraquinones. J. Chem. Soc., Faraday Trans. 1990, 86, 3861−3869. (41) Yatsuhashi, T.; Inoue, H. Molecular Mechanism of Radiationless Deactivation of Aminoanthraquinones through Intermolecular Hydrogen-Bonding Interaction with Alcohols and Hydroperoxides. J. Phys. Chem. A 1997, 101, 8166−8173. (42) Dahiya, P.; Kumbhakar, M.; Maity, D. K.; Mukherjee, T.; Tripathi, A. B. R.; Chattopadhyay, N.; Pal, H. Solvent Polarity and Intramolecular Hydrogen Bonding Effects on the Photophysical Properties of 1-Amino-9,10-anthraquinone Dye. J. Photochem. Photobiol., A 2006, 181, 338−347. (43) Dahiya, P.; Kumbhakar, M.; Mukherjee, T.; Pal, H. Effect of the Amino and Hydroxy Substituents on the Photophysical Properties of 1,4-Disubstituted-9,10-anthraquinone Dyes. J. Mol. Struct. 2006, 798, 40−48. (44) Marzocchi, M. P.; Mantini, A. R.; Casu, M.; Smulevich, G. Intramolecular Hydrogen Bonding and Excited State Proton Transfer in Hydroxyanthraquinones as Studied by Electronic Spectra, Resonance Raman Scattering, and Transform Analysis. J. Chem. Phys. 1998, 108, 534−549. (45) Arzhantsev, S. Y.; Takeuchi, S.; Tahara, T. Ultrafast ExcitedState Proton Transfer Dynamics of 1,8-Dihydroxyanthraquinone (Chrysazin) Studied by Femtosecond Time-Resolved Fluorescence Spectroscopy. Chem. Phys. Lett. 2000, 330, 83−90. (46) Schmidtke, S. J.; Underwood, D. F.; Blank, D. A. Following the Solvent Directly during Ultrafast Excited State Proton Transfer. J. Am. Chem. Soc. 2004, 126, 8620−86021.

(47) Schmidtke, S. J.; Underwood, D. F.; Blank, D. A. Probing Excited-State Dynamics and Intramolecular Proton Transfer in 1Acylaminoanthraquinones via the Intermolecular Solvent Response. J. Phys. Chem. A 2005, 109, 7033−7045. (48) Neuwahl, F. V. R.; Bussotti, L.; Righini, R.; Buntinx, G. Ultrafast Proton Transfer in the S1 State of 1-Chloroacetylaminoanthraquinone. Phys. Chem. Chem. Phys. 2001, 3, 1277−1283. (49) Welton, T., Wasserscheid, P., Eds. Ionic Liquid Synthesis, 2nd ed.; Wiley VCH: Weinheim, Germany, 2008; Chapter 2. (50) Yoshizawa, M.; Xu, W.; Angell, C. A. Ionic Liquids by Proton Transfer: Vapor Pressure, Conductivity, and the Relevance of ΔpKa from Aqueous Solutions. J. Am. Chem. Soc. 2003, 125, 15411−15419. (51) Gao, H. X.; Han, B. X.; Li, J. C.; Jiang, T.; Liu, Z. M.; Wu, W. Z.; Chang, Y. H.; Zhang, J. M. Preparation of Room-Temperature Ionic Liquids by Neutralization of 1,1,3,3-Tetramethylguanidine with Acids and Their Use As Media for Mannich Reaction. Synth. Commun. 2004, 34, 3083−3089. (52) Hangarge, R. V.; Jarikote, D. V.; Shingare, M. S. Knoevenagel Condensation Reactions in an Ionic liquid. Green Chem. 2002, 4, 266− 268. (53) Greaves, T. L.; Weerawardena, A.; Fong, C.; Drummond, C. J. Formation of Amphiphile Self-Assembly Phases in Protic Ionic Liquids. J. Phys. Chem. B 2007, 111, 4082−4088. (54) Ohno, H.; Yoshizawa, M. Ion Conductive Characteristics of Ionic Liquids Prepared by Neutralization of Alkylimidazoles. Solid State Ionics 2002, 154, 303−309. (55) Xu, W.; Angell, C. A. Solvent-Free Electrolytes with Aqueous Solution-Like Conductivities. Science 2003, 302, 422−425. (56) Zhao, C.; Burrell, G.; Torriero, A. A. J.; Separovic, F.; Dunlop, N. F.; MacFarlane, D. R.; Bond, A. M. Electrochemistry of Room Temperature Protic Ionic Liquids. J. Phys. Chem. B 2008, 112, 6923− 6936. (57) Shukla, S. K.; Khupse, N. D.; Kumar, A. Do Anions Influence the Polarity of Protic Ionic Liquids? Phys. Chem. Chem. Phys. 2012, 14, 2754−2761. (58) MacFarlane, D. R.; Pringle, J. M.; Johansson, K. M.; Forsyth, S. A.; Forsyth, M. Lewis Base Ionic Liquids. Chem. Commun. 2006, 1905−1907. (59) Nad, S.; Pal, H. Photophysical Properties of Coumarin-500 (C500): Unusual Behavior in Nonpolar Solvents. J. Phys. Chem. A 2003, 107, 501−507. (60) O’Connor, D. V.; Philips, D. Time Correlated Single Photon Counting; Academic Press: New York, 1984. (61) Singh, P. K.; Nath, S.; Bhasikuttan, A. C.; Kumbhakar, M.; Mohanty, J.; Sarkar, S. K.; Mukherjee, T.; Pal, H. Effect of Donor Orientation on Ultrafast Intermolecular Electron Transfer in Coumarin−Amine Systems. J. Chem. Phys. 2008, 129, 114504− 114511. (62) Kumbhakar, M.; Singh, P. K.; Nath, S.; Bhasikuttan, A. C.; Pal, H. Ultrafast Bimolecular Electron Transfer Dynamics in Micellar Media. J. Phys. Chem. B 2008, 112, 6646−6652. (63) Dahiya, P.; Kumbhakar, M.; Mukherjee, T.; Pal, H. Effect of Protic Solvents on Twisted Intramolecular Charge Transfer State Formation in Coumarin-152 and Coumarin-481 Dyes. Chem. Phys. Lett. 2005, 414, 148−154. (64) Barik, A.; Kumbhakar, M.; Nath, S.; Pal, H. Evidence for the TICT Mediated Nonradiative Deexcitation Process for the Excited Coumarin-1 Dye in High Polarity Protic Solvents. Chem. Phys. 2005, 315, 277−285. (65) Zhang, X.-X; Liang, M.; Ernsting, N. P.; Maroncelli, M. Complete Solvation Response of Coumarin 153 in Ionic Liquids. J. Phys. Chem. B 2013, 117, 4291−4304. (66) Jin, H.; Baker, G. A.; Arzhantsev, S.; Dong, J.; Maroncelli, M. Solvation and Rotational Dynamics of Coumarin 153 in Ionic Liquids: Comparisons to Conventional Solvents. J. Phys. Chem. B 2007, 111, 7291−7302. (67) Samanta, A. Solvation Dynamics in Ionic Liquids: What We Have Learned from the Dynamic Fluorescence Stokes Shift Studies. J. Phys. Chem. Lett. 2010, 1, 1557−1562. 2497

dx.doi.org/10.1021/jp500281z | J. Phys. Chem. B 2014, 118, 2487−2498

The Journal of Physical Chemistry B

Article

(68) Muramatsu, M.; Nagasawa, Y.; Miyasaka, H. Ultrafast Solvation Dynamics in Room Temperature Ionic Liquids Observed by ThreePulse Photon Echo Peak Shift Measurements. J. Phys. Chem. A 2011, 115, 3886−3894. (69) Paul, A.; Mandal, P. K.; Samanta, A. On the Optical Properties of the Imidazolium Ionic Liquids. J. Phys. Chem. B 2005, 109, 9148− 9153. (70) Mandal, P. K.; Paul, A.; Samanta, A. Excitation Wavelength Dependent Fluorescence Behavior of the Room Temperature Ionic Liquids and Dissolved Dipolar Solutes. J. Photochem. Photobiol. A: Chem. 2006, 182, 113−120. (71) Smulevich, G.; Foggi, P.; Feis, A.; Marzocchi, M. P. Fluorescence Excitation and Emission Spectra of 1, 8-Dihydroxyanthraquinone-d0 and -d2 in n-Octane at 10 K. J. Chem. Phys. 1987, 87, 5664−5669. (72) Smith, T. P.; Zaklika, K. A.; Thakur, K.; Walker, G. C.; Tominaga, K.; Barbara, P. F. Spectroscopic Studies of Excited-State Intramolecular Proton Transfer in 1-(Acylamino) Anthraquinones. J. Phys. Chem. 1991, 95, 10465−10475. (73) Smith, T. P.; Zaklika, K. A.; Thakur, K.; Barbara, P. F. Excited State Intramolecular Proton Transfer in 1-(Acylamino)anthraquinones. J. Am. Chem. Soc. 1991, 113, 4035−4036.

2498

dx.doi.org/10.1021/jp500281z | J. Phys. Chem. B 2014, 118, 2487−2498