Solvation Dynamics in Molten Salts - American Chemical Society


Solvation Dynamics in Molten Salts - American Chemical Societypubs.acs.org/doi/pdf/10.1021/j100093a024Similarby E Bart -...

0 downloads 64 Views 505KB Size

J. Phys. Chem. 1994, 98, 10819-10823

10819

Solvation Dynamics in Molten Salts E. Bart, A. Meltsin, and D. Huppert’ Sackler Faculty of Exact Sciences, School of Chemistry, Tel Aviv University, Ramat Aviv 69978, Israel Received: March 24, 1994; In Final Form: July 8, 1994@

We have observed time-dependent shifts in the fluorescence spectra of solvatochromic probe molecules in several molten tetraalkylammonium salts. The phenomenon is a general one and is observed in a variety of fluorophores. The steady-state emission position shows a pronounced dependence on the molten salt cation size. Frequency shifts in steady-state emission spectra are a function of the solute. The solvation of excited large dipoles occurs on two time scales which are both dependent on the cation size. The larger the cation size, the longer the relaxation time. The average experimental solvation time, (T), for the same molten salt depends on the solute. The observed results suggest that molten salts are an appropriate medium for studying excited large dipole solvation.

Introduction Time-resolved fluorescence spectroscopy of polar fluorescent “probes” has been extensively used to measure microscopic solvation dynamics in a broad range of A probe molecule for transient solvation studies should have a net dipole moment change between the ground state and first excited singlet state. The simplest ultrafast solvation process occurs when a rigid dye molecule is electronically excited in a polar solvent. When its fluorescence spectrum is followed as a function of time, a red shift is observed that reflects the microscopic solvent relaxation.1-8 Our recent works9-12 represent an extension of previous studies on solvation dynamics into the realm of included ionic solutions. We studied the static and dynamic “salt effect” on solvation of fluorescent probe molecules in electrolyte solutions. Results from several groups have led to a better understanding of the solvation statics and dynamics of probe molecules in solutions of organic liquids containing electrolyte^.'^^^^ Molten salts are a very important supplement to nonaqueous and waterlike solvents. They are important as solvents for salts, metals, and g a ~ e s . ’ ~ . ’ ~ Great attention has been devoted to the study of molten organic salts.”J8 Recently, we employed static and timeresolved fluorescence techniques to measure the solvation statics and dynamics of electronically excited coumarin 153 in molten quatemary ammonium salt^.'^.^^ The solvation of coumarin 153 occurs on two time scales which are both dependent on the cation size. The larger the cation size, the longer the relaxation time. The steady-state emission position is also strongly dependent on the molten salt cation size. Molten quatemary ammonium salts possess several advantages over inorganic ones. First of all, they were shown to be better solvents for organic compounds than fused inorganic salts because of their lower melting points and polarities,21and they are able to mix with a wide variety of organic materials. Potential obstacles to their use as solvents might be their instability in acids and possible thermal decomposition. Gordon and co-workersZ2studied the decomposition products of a set of low-melting tetra-n-alkylammonium salts. They were unable to detect decomposition products which occur at 5 “C above the melting point within 3 h. The short time duration of our measurements (up to 3 h) and the choice of the molten salts @

Abstract published in Advance ACS Abstracts, September 1, 1994.

0022-3654/94/2098-10819$04.50/0

used enable us to assume that both the probe and the salts were stable during the experimental measurements. In the present work we have continued our studies of solvation statics and dynamics in molten organic salts, and the work reported here consists of steady-state and time-resolved fluorescence measurements of several common solvatochromic probes in a variety of molten tetraalkylammonium salts. We now report new results on solvation of coumarin 102 and 6-[N-

(4-methylphenyl)amino]-2-naphthalene-N,N-dimethylsulfonamide (TNSDMA) in molten quatemary ammonium salts and present a comparison of the previous results of coumarin 153 and these molecules. A few comments concerning the probe molecules chosen for these studies. The main feature that these probes have in common is that their fluorescence frequencies exhibit a strong dependence on the solvent polarity. The dyes coumarin 1026323and coumarin 1533,4,8,uhave been extensively used as probes in studies of solvation dynamics. These molecules all show very large shifts in their fluoreescence frequencies as a function of solvent polarity, especially in hydrogen-bonding solvents. They all have well-separated SO S1 absorptions as judged by fluorescence polarization studies and quantum yields of near unity. TNSDMA was successfully used in the initial solvation s t ~ d i e s . ~Studies ~ - ~ ~on TNSDMA in various solvents including the alkanols methanol, ethanol, n-butanol, and n-hexanol were proven to be especially valuable in illuminating the mechanism of the fast intramolecular electron transfer process.

-

Experimental Section Time-resolved fluorescence was measured using time-correlated single-photon counting (TCSPC) technique. As an excitation source, we used a CW mode-locked Nd:YAG-pumped dye laser (Coherent Nd:YAG Antares and a 702 dye laser) providing a high repetition rate (< 1 MHz) of short pulses (1 ps at full width at half-maximum, fwhm). The (TCSPC) detection system is based on Hamamatsu 1564U and 3809U photomultipliers, a Tennelec 864 TAC, a Tennelec 454 discriminator, and a personal computer-based multichannel analyzer (nucleus PCA-II). The overall instrumental response at full width at halfmaximum was about 50 ps (fwhm). Measurements were taken at a 10 nm spectral width. Steady-state fluorescence spectra were taken using a Perkin-Elmer spectrofluorimeter MPF-4. Fluorescence spectra were corrected for instrumental response. The fluorescence probe molecules used in this study included coumarin 102 and 153 dyes ((2.102, C.153) and 6-[N-(40 1994 American Chemical Society

Bart et al.

10820 J. Phys. Chem., Vol. 98, No. 42, 1994

SCHEME 1

f'i

-

5 2 4 -

H

u

'b23-. :22n

C.102 R=CH, C.153 R=CF,

TNSDM4

. .

.

W

21-

TABLE 1: Results of Steady-State Fluorescence Spectra of Several Probmolten Salt Systems

TBAHS THAPCl TOAPCl TDDAPCl

170

4.8

455

5.5 6.1 7.0

450

535 520

470

105

445

515

440

497

450 403

136 120

Values for the effective radii of tetraalkylammonium salts were taken from ref 33.

methylphenyl)amino]-2-naphthalene-N,N-dimethylsulfonamide (TNSDMA), illustrated in Scheme 1. The C.153 and C.102 were obtained from Exciton, and TNSDMA was obtained from Professor E. M. Kosower. These probe molecules were used without further purification. The concentration of the probe dyes was M. Tetrabutylammonium hydrogen sulfate (TBAHS, mp 172.2172.6 "C), tetrahexylammonium perchlorate (THAPC1, mp 105.3- 105.9 "C), tetraoctylammonium perchlorate (TOAPCl, mp 134.0- 134.9 "C), and tetradodecylammonium perchlorate (TDDAPCl, mp 116.3-117.3 "C) were puriss grade (Fluka) and used without further purification. The probe molecule samples in tetraalkylammonium salts Torr at were kept in quartz vessels under a vacuum of room temperature. The samples were heated by a home-built oven with a temperature control and stability of f 2 "C. The temperature of the irradiated samples was 3-5 "C above the melting point.

Results The results of the steady-state (time-integrated) fluorescence spectra of C.153, C.102, and TNSDMA in several molten tetralkylammonium salts (see Table 1) demonstrate that the emission spectra of the solvatochromic probes shift to the red as the radius of the cation decreased. On the basis of the relative position of the fluorescence band maxima, the solvent polarities of the various molten tetraalkylammonium salts used in this study were found.*O A steady-state spectral shift appears to reflect a decrease of the solvation energies with an increase of the cation size. In order to explain this observation,we assumed the following qualitative model. The total solvation energy, E,, of the excited probe consists of two contributions, Es+ f Es-, where Es+ is the solvation energy of the probe positive charge by the negative ions and E,- is the solvation energy contribution of its negative charge due to the close proximity of salt cations. Both E,+ and Es- in this crude picture are inversely dependent on the distance between salt ions and the probe charges and hence on the molten salt ions' size. Since in most of the salts used in this study the anionic species were identical, we assumed that E,+ is constant. From the above qualitative model it is expected to find a linear relation between fluorescence band maximum and 1/R+ (R+ is the cation size). Figure 1 shows the fluorescence band maximum as a function of 1/R+ for three solvatochromic probes, C.153, C.102, and

>

0.6-

z

w

+-

-

5 0.4-

w

u z

. 02-

w

a" LL =

00bd0'8

IO

1'2

1'4

1'6

I:8

20

T I M E ( ns 1

Figure 2. Time-resolved emission of C.102 in molten TOAPCl measured at the following wavelengths (top to bottom): 540,470,440, 420, and 400 nm. TNSDMA. A monotonic decrease of the fluorescence band maximum with an increase of 1/R+ is found. Compared to coumarin 102, TNSDMA has the strongest sensitivity to the molten salt's polarity. The magnitude of the observed spectral shift depends on the identity of the probe molecule. The solvatochromic behavior of a probe depends on its dipole moment. Table 2 lists results of steady-state fluorescence measurements of three probes in pure regular organic solvents, electrolyte solutions (LiC104 in acetonitrile), and tetraalkylammonium molten salts. From the pure solvent shifts we can conclude that all of the probes must have relatively large dipole moments in both the ground and excited states and substantial increase in dipole moment between ground and excited statesz9 We also note that although C.102 and C.153 are structurally very similar (Scheme l), they exhibit significantly different solvent sensitivities. The largest emission shift was observed between DMSO and cyclohexane with TNSDMA and the smallest, with (2.102. Similar observations were also obtained for the molten tetraalkylammoniumsalts (see Table 2). The time-resolved fluorescence decay was measured using the TCSPC technique. Figure 2 illustrates data obtained with the C. 102 probe in molten tetraoctylammonium perchlorate measured at 10 nm intervals. The time-resolved spectra have been reconstructed from these series of fluorescence decays in the manner described in ref 30. To deconvolute the instrument response from the decay data, each decay was fit to a sum of exponentials. The multiexponential forms of the fluorescence dynamics are combined to obtain emission spectra at various times after The time-dependent fluorescence

Solvation Dynamics in Molten Salts

J. Phys. Chem., Vol. 98, No. 42, 1994 10821

TABLE 2: Steady-State Fluorescence of Several Solvatochromic Probes in Different Media (Cm-') C.153 1.84 103 2.60 x 103 1.54 103

Av,,,, (cm-l)

&bS

medium pure solvenv

c.102 1.47 x 103

electrolyte solutionsb

0.64 x 103 0.68 x 103

molten saltsC

TNSDMA 1.26 x 103

(2.102 2.94 x 103

2.03 x 103

0.98 x 103

1.38 x 103

0.75 x 103

(2.153 3.85 x 103 0.71 x 103 1.43 x 103

TNSDMA 4.90 x 103 0.64 x 103

3.54 x 103

Frequency shifts in absorption (abs) and emission (em) spectra between cyclohexane and dimethyl sulfoxide solvents for C.102,13C.153,I3 and TNSDMA. Frequency shifts in absorption (abs) and emission (em) spectra between 1 M LiC104/acetonitrilesolutions and pure acetonitrile solvent. Frequency shifts in absorption (abs) and emission (em) spectra between molten TBAHS and TDDAPCl.

WAVELENGTH ( n m )

Figure 3. Time-resolved emission spectra of TNSDMA in molten TBAHS. The times shown are 10, 50, 200, 800, and 2000 ps after

The parameter /3 can vary between 0.65 and 0.75; but the solvation dynamic data is best reproduced by p = 0.7. The two lifetimes (tl, t2) and the relative amplitude of the two phases (HA) (given in Table 3) show marked dependence on the cation size. An increase of the lifetimes is observed with the increase of the cation size. The magnitude of the longer lifetime, t2. is more affected by the cation size. The same results were obtained for C. 153 in several molten tetraalkylammonium salts.20 These observations are viewed as the result of two separate solvation energy contributions of both the cation and anion, which occur on two separate time scales and are dependent on the cation size. The average relaxation time, defined by

excitation (top to bottom). frequency change is obtained by fitting the spectrum at each time to a log-normal line shape.31 Transient spectra of TNSDMA in molten TBAHS are shown in Figure 3. The normalized spectral shift correlation function is defined by eq 1.

Here, ~ ( t )Y(=), , and ~ ( 0represent ) the fluorescence frequencies at time, t, after excitation, at infinite time after the probe/ solvent system has reached equilibrium, and at zero time immediately after photon excitation, respectively. It has been found previously that a shift of v+(t)(Y+ is defined as the halfheight point on the high-frequency side) is larger than that of ~ , , ( t ) . ~ ~That is why the high-frequency normalized correlation function C+(t) was used in order to quantify the solvation dynamics. This function is constructed in the same way using the values of v+(t),Y+(=), and v+(O). @(t) for C.102 in a series of four molten salts (TBAHS, THAPC1, TOAPCl, and TDDAPC1) and for TNSDMA in a series of three molten salts (TBAHS, TOAPC1, and TDDAPCl) are shown in parts a and b of Figure 4,respectively. The experimental data (dots) were fitted by two stretched exponentials (solid lines). The fitting parameters of C+(t)for these systems are given in Table 3. Since the fwhm of the instrumental time response is ~ 5 ps, 5 it is hard to determine components of the dynamics that are faster than the time resolution. General features of solvation dynamics have been observed with a variety of probe/molten salt combinations and can be summarized as follows: 1. In all of these probes the decay of C+(t) is generally nonexponential. 2. With decreasing cation size, the process of solvation becomes faster. 3. The observed @(t) correlation functions indicate that solvation dynamics in molten quaternary salts is biphasic. Each phase is described by a stretched exponential decay.

varies significantly with solute. Values of (T) (obtained by numerical integration) for three probe molecules in several molten quaternary salts are listed in Table 3. This variability is greater than is observed among different probes in the same pure polar solvent and electrolyte solutions. (Examples of the latter case are listed in Table 4.) (T) values obtained with (2.102 were longer than those observed with (2.153 in the same molten salts. In these two cases the compared probe structures are very similar, differing mainly in their overall size due to different alkyl substituents peripheral to the aromatic system. Among the solvatochromicprobes we have studied, the smallest average relaxation time is in the case of TNSDMA (see Table 3). From the three studied probe molecules, the solvation dynamics of TNSDMA is less sensitive to the cation size (see Figure 5). Figure 6 compares the time-dependent Stokes shift of the fluorescence spectrum in molten TOAPCl for the probe molecules used in this study. Differencesbetween the results obtained with probe molecules C.153, C.102, and TNSDMA clearly indicate a noticeable probe dependence on the solvation dynamics. However, the general tendencies in the solvation behavior of the studied probe molecules support the proposed model*O of the solvation process in molten organic salts. Summary Steady-state fluorescence emission, as well as dynamic measurements of coumarin 153, coumarin 102, and TNSDMA in molten quaternary ammonium salts, reveals the existence of solvation processes in a new medium, molten salts. The quaternary ammonium salts were chosen to demonstrate the solvation of fluorescent probe molecules in this medium. N-butyl and n-dodecyl were the shortest and longest alkyl chain ammonium cations, respectively. !I these salts the cation size varies monotonically from 4.9 to 7 A.33 Identical counteranion was used for the hexyl, octyl, and dodecyl salts (perchlorate) and hydrogen sulfate anion for the butyl salt.

Bart et al.

10822 J. Phys. Chem., Vol, 98, No. 42, 1994

a .

OL

20

IO

30

TIME (ns)

TABLE 4: Time-Resolved Fluorescence Results of Several Solvatochromic Probes in Different Media media probe pure solvent electrolyte solution molten salt molecule (t)solv" (ps) W b(PSI (PS) c.102 200 1250 1580 C.153 280 810 1150 TNSDMA 245 750 317 a Average solvation times measured in 1-propanolat 253 K. Results for C.102 and C.153 are taken from ref 13. Average relaxation times of the time-resolved fluorescence shifts, measured in 1 M NaC10.J acetonitrile solution. Results for C.102 and C.153 are taken from ref 13. Average relaxation time (as defined by eq 3) measured in molten TOAPCl.

"i i

. . v1

I

'0°40 TlME(ns)

50 C A T I O N SIZE

Figure 4. (a) Time-dependence of the correlation function C(t)of C.102 in molten tetraalkylammonium salts. From top to bottom the salts are TDDAPCl, TOAPC1, THAPC1, and TBAHS. The points are the experimental data, and the solid lines are the fits to these data (see eq 2). (b) Time-dependence of the correlation function p(t)of TNSDMA in molten tetraalkylammonium salts. From top to bottom the salts are TDDAPCl, TOAPC1, and TBAHS. The points are the experimental data, and the solid lines are the fits to these data (see eq 2). TABLE 3: Results of Time-Resolved Fluorescence Experiments stretch average probe relaxation time (PS) amplitude factor relaxation time molecule tl t2 A B B (PS) (0 TBAHS 625 0.54 0.46 0.7 c.102 53 380 416 0.69 0.31 0.7 C.153 42 190 125 0.73 0.27 0.7 TNSDMA 40 71 THAPCl 1694 0.27 0.73 1210 0.40 0.60

0.7 0.7

1290 819

c.102 C.153

88 60

c.102 C.153

2040 1416 500

TOAPCl 0.18 0.82 0.25 0.75 0.64 0.36

0.7

TNSDMA

189 136 127

0.7

1580 1150 317

c.102 C.153 TNSDMA

250 176 159

TDDAPCl 4350 0.11 0.89 3268 0.13 0.87 625 0.52 0.48

0.7 0.7 0.7

2455 2123 463

0.7

Measurements of solvation energies and dynamics of three electmonically excited solvation probes in molten tetraalkylammonium salts demonstrate that although solvation statics and dynamics quantitatively differ and depend on the solute, the solvation of highly polar aromatic molecules in molten organic salts occurs in a similar manner in all of the examined systems. The solvation dynamics in these probe/molten salt systems was found to be biphasic and occurs on picosecond-nanosecond time scales. The biphasic solvation dynamics is in contrast to

60

I

i

1

70

(1)

Figure 5. Semilogarithmicplot of the average relaxationrate constants as a function of cation size: (A)C.102; (0)C.153; (m) TNSDMA.

L0 02

0

20

IO

30

TIME (ns)

Figure 6. Time-dependence of the correlation function p(t)of three solvatochromic probe molecules in molten TOAPC1. From top to bottom the molecules are C.102, (2.153, and TNSDMA. The points are the experimental data, and the solid lines are the fits to these data (see eq 2). the monophasic solvation dynamics of coumarin 153 and other probe molecules in molecular liquids like simple n-alcohols.8,19,34,35 Similarity in solvation behavior of several solvatochromic probes confirms that the solvation process in molten salts is not specific for individual probe molecules. Therefore, a new medium, molten organic salts, is appropriate for further study of the excited large dipole solvation. A better understanding of the main features of the solvation process occumng in molten salts requires more experimental evidence. Research of the anion size effects on solvation energetics and dynamics might help to clarify this complex solvation process. Acknowledgment. We thank Professor N. Agmon for providing the software for data analysis and for many helpful

Solvation Dynamics in Molten Salts discussions. This work was supported by grants from the Ministry of Science and Technology, thee U.S.-Israel Binational Science Foundation, and the James Franck Binational German Israel Program in Laser Matter Interaction. References and Notes (1) 127. (2) (3) 1674. (4)

Kosower, E. M.; Huppert, D. Annu. Rev. Phys. Chem. 1986, 37, Barbara, P. F.; Jarzeba, W. Acc. Chem. Res. 1988, 21, 195. Maroncelli, M.; MacInnis, J.; Fleming, G. R. Science 1989, 243, Barbara, P. F.; Jarzeba, W. Adv. Photochem. 1990, 15, 1. S. K.: Brant. G. S.: Choe. P. P. J . Chem. Phvs.

( 5 ) Ware. W. R.: Lee. 1971,54?4729.

(6) Kahlow, M. A.; Jarzeba, W.; Kang, T. S.; Barbara, P. F. J . Chem. Phys. 1989, 90, 151. (7) Castner, E. W., Jr.; Maroncelli, M.; Fleming, - G. R. J . Chem. Phys. 1987, 86, 1090. (8) Maroncelli, M.; Fleming, G. R. J . Chem. Phys. 1987, 86, 6221. (9) Huppert, D.; Ittah, V.; Kosower, E. M. Chem. Phys. Lett. 1989, 159, 967.

J. Phys. Chem., Vol. 98, No. 42, 1994 10823 (10) Huppert, D.; Ittah, V. In Perspectives in Photosynthesis; Jortner, J., Pullman, B., Eds.; Kluwer: Dordrecht, 1990; pp 310-316. (11) Ittah, V.; Huppert, D. Chem. Phys. Lett. 1990, 173, 496. (12) Bart, E.: Huppert, D. Chem. Phys. Lett. 1992, 195, 37. (13) Chapman, C. F.; Maroncelli, M. J . Phys. Chem. 1991, 95, 9095. (14) Neria, E.; Nitzan, A. J . Chem. Phys. 1994, 100, 3855. (15) Sandemeyer, W. Angew. Chem., In?. Ed. Engl. 1965, 4, 222. (16) Sandhein, B. R. Fused Salts; Department of Chemistry: New York

University, 1964. (17) Gordon, J. E. J . Am. Chem. SOC. 1964, 86, 4492. (18) Gordon, J. E. J. Am. Chem. SOC. 1965, 87, 4347. (19) Bart, E.; Meltsin, A.; Huppert, D. Chem. Phys. Lerr. 1992, 220, 592. (20) Bart, E.; Meltsin, A.; Huppert, D. J . Phys. Chem. 1994. 98, 3295. (21) Ford, W. J.; Mauri, R. S.;Hart,D. J. J . Org. Chem. 1973, 38 (22), 3916. (22) Gordon, J. E. J . Org. Chem. 1965, 30, 2760. (23) Maroncelli, M.; Fee, R. S.;Chapman, F.; Fleming, G. R. J . Phys. Chem. 1991, 95, 1012. (24) Jarzeba, W.; Walker, G. G.; Johnson, A. E.; Barbara, P. F. Chem. Phys. 1991, 57, 152.