- Email: [email protected]
Photoinduced electron transfer from chlorin p6 to methyl viologen in aqueous micelles
Chemical Physics Letters 407 (2005) 119–123 www.elsevier.com/locate/cplett
Photoinduced electron transfer from chlorin p6 to methyl viologen in aqueous micelles Tushar Kanti Mukherjee, Padmaja P. Mishra, Anindya Datta
*
Department of Chemistry, Indian Institute of Technology, Bombay Powai, Mumbai 400 076, India Received 30 December 2004; in final form 7 March 2005 Available online 7 April 2005
Abstract We report the effect of micellar encapsulation on the photoinduced electron transfer between chlorin p6 and methyl viologen. In aqueous solutions, these molecules form contact ion pairs and the fluorescence quenching observed is mainly static in nature. The quenching efficiency increases markedly in SDS micelles and the mechanism changes partially to dynamic quenching. However, fluorescence quenching of all kinds are suppressed in CTAB and Triton X-100 micelles. This is explained in the light of dissociation of the contact ion pairs and the electrostatic attraction of MV2+ by the headgroups of the surfactants. 2005 Elsevier B.V. All rights reserved.
1. Introduction Aggregation and photoinduced electron and energy transfer in porphyrin based molecules have been studied extensively, in attempts to understand and mimic green plant photosynthesis and to develop efficient solar cells [1–6]. Recently, there has been a spate of activity in the field of PET in organized media [7,8] where the excited state dynamics differ largely from those in homogeneous solutions [9,10]. The motivation for the study of PET in porphyrinic systems in organized media stems from the urge to generate long-lived ion pairs, so that the energy produced thus can be used fruitfully. If the D–A distance is too large for the reactants to diffuse to each other during the excited state lifetimes, then PET is suppressed. On the other hand, if the distance is comparable to the van der Waal radius, then the back electron transfer is also ultrafast, leading to a very shortlived ion pair. The challenge, therefore, lies in the design of systems where this distance can be controlled in order
*
Corresponding author. Fax: + 91 22 2572 3480. E-mail address: [email protected] (A. Datta).
0009-2614/$ - see front matter 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.cplett.2005.03.055
to have an efficient forward electron transfer and a negligible back electron transfer. This goal can be achieved either by synthesis or by entrapment of the donor or acceptor in organized assemblies. The encapsulation effect of organized assemblies can bring the donor and the acceptor together and thus facilitate the forward electron transfer whereas their hydrophobic and hydrophilic microenvironments can provide differential solubilization to the ion pair formed, in case one of the products is more hydrophobic than the other, leading to the formation of solvent separated ion pairs and consequent suppression of the back electron transfer [11,12]. In the present study, we have investigated the effect of micelles on photoinduced electron transfer between chlorin p6 and methyl viologen dication (MV2+) (Fig. 1).
2. Experimental Chlorin p6 is prepared using literature procedure [3–5]. Methyl viologen dichloride and Triton X-100 (TX-100) (AR grade) from Aldrich have been used as received. Sodium dodecyl sulphate (SDS) and cetyl trimethyl ammonium bromide (CTAB) (AR grade) from
120
T.K. Mukherjee et al. / Chemical Physics Letters 407 (2005) 119–123 0.4 (a) Soret
NH
N
N
HN
CO2H
Me +
3xQ
N
CO2H
N
0.2
0.0 0.4
+
(b)
Me
HO2C Chlorin p6
H2O
0.2
Methyl viologen dication
Fig. 1. Structures of chlorin p6 and MV2+.
Aldrich are recrystallized from methanol–water solutions. The concentration of chlorin p6 used is 4 lM. The concentrations of SDS, CTAB and Triton X-100 are 14, 3 and 1 mM, respectively. The absorption and fluorescence spectra have been recorded on JASCO V570 spectrophotometer with 2 nm bandpass and a Perkin–Elmer LS-55 spectrofluorimeter with kex = 401 nm, respectively. Fluorescence decays have been recorded on an IBH Fluorocube time resolved fluorescence spectrometer [13]. The fluorescence decays were fitted to biexponential functions by iterative reconvolution using the IBH DAS 6.0 software. Global analysis of the decays were performed using a homemade program.
O.D.
CTAB 0.0 0.4 (c)
0.2
TX-100 0.0
0.4 (d)
0.2
SDS
3. Results and discussion The absorption peaks of chlorin p6 in water get red shifted and broadened with gradual addition of MV2+ (Fig. 2a), indicating ground state association due to their opposite charges [14]. Little, if any shift or broadening of the absorption peak is observed with addition of MV2+ to chlorin p6 in 3 mM CTAB and in 1 mM TX-100 (Fig. 2b and c). In 14 mM SDS, the changes in absorption spectra on addition of MV2+ are less prominent than in bulk water, but are more prominent than in the other micelles (Fig. 2d). This indicates that the extent of contact ion pair formation is reduced substantially in micelles compared to neat aqueous solutions. The Stern–Volmer plot in water is linear (Fig. 3a). In CTAB and TX-100, the slope of the Stern– Volmer plot is zero, except for an initial small quenching in TX-100 upto a MV2+ concentration of 4 · 10 5 M, (Fig. 3). These indicate a complete incorporation of chlorin p6 in CTAB micelle, resulting in effective sequestering from MV2+, which resides in bulk water. In TX100, chlorin p6 may not be totally bound and MV2+ can initially quench the free fluorophore molecules. The most remarkable result is obtained with SDS, where a linear Stern–Volmer plot, with a steeper slope than
0.0 350
400
650
700
λ (nm)
Fig. 2. Principal absorption peaks of chlorin p6 in the absence (solid line) and presence (dotted line) of 1 mM MV2+ in (a) water, (b) 3 mM CTAB, (c) 1 mM Triton X-100 and (d) 14 mM SDS. The Q bands are magnified for the sake of clarity.
that in neat aqueous solution, is obtained (Fig. 3a, Table 1). This is an indication of the changeover of the mechanism to a more efficient one. In SDS, where the association complexes are disrupted, the quenching mechanism might be predominantly dynamic in nature [15]. Interestingly, the KSV is greater in SDS micellar media than in neat aqueous solutions, which is in contrast to some earlier observations [7,15,16]. The decrease in KSV values in micelles is generally rationalized by a decreased mobility of incorporated quenchers and fluorophores. However, an increase in KSV is also observed in certain cases. For example, an enhanced quenching of ruthenium complexes by MV2+ and Cu2+ is observed in SDS solutions, and is explained by the formation of clusters of the fluorophore with SDS, with which the oppositely charged quenchers associate easier [17]. In our experiment, since chlorin p6 resides near the
Counts
T.K. Mukherjee et al. / Chemical Physics Letters 407 (2005) 119–123 10
4
10
3
10
2
10
1
10
0
121
(a) No MV 1 mM MV
2+
2+
H 2O
10
4
10
3
10
2
10
1
10
0
(b) 2+
No MV 1 mM MV
2+
SDS
0
8
16
Time (ns)
Fig. 3. Stern–Volmer plots with fluorescence quantum yields (/f) of chlorin p6 in (a) water, 3 mM CTAB and in 14 mM SDS micelles and (b) water and 1 mM Triton X-100.
Table 1 Values of KSV and kq from steady state experiments in different media Ksv (lit mol 1) kq (lit mol 1 s 1)
H2O
SDS
Triton X-100
1560 4.58 · 1011
26 370 7.4 · 1012
11 520 3.1 · 1012
negatively charged headgroups of SDS, this charged surface might offer electrostatic attraction to draw the MV2+ ions towards the fluorophore. In order to gain a better insight to the situation, time resolved experiment have been performed. The fluorescence decays of chlorin p6 in water exhibit no change on addition of MV2+ (Fig. 4a). This confirms that in water, only static quenching is operative and can be explained in the light of instantaneous electron transfer in the association complexes between MV2+ and chlorin p6 [14]. Dynamic quenching might be observed in micelles, where these association complexes are disrupted. However, there is no change in the fluorescence decays in CTAB or Triton X-100 micelles, indicating the absence of dynamic quenching by MV2+ in these micelles. In SDS micelle, the fluorescence decays become faster with addition of MV2+ (Fig. 4b), indicating the dynamic nature of quenching [15]. The decays are well fit to biexponential functions and P the Stern–Volmer plots with average lifetime sav = aisi is shown in Fig. 5a. The Stern–Volmer constant is much smaller compared to that obtained from the steady state experiment (Tables 1 and 2). Similar discrepancies have been
Fig. 4. Fluorescence decays in the presence and absence of 1 mM MV2+ in (a) water and (b) 14 mM SDS. The instrument response functions (FWHM = 250 ps) are shown in dashed lines.
reported in some other studies of bimolecular processes in micelles [7]. We feel that this discrepancy arises out of the oversimplified model of data fitting. The analysis of multiexponential fluorescence decays is by no means a trivial procedure and the results are strongly dependent on the model used to describe the system under investigation [18]. Several such models have been proposed so far [19–24]. For example, Gehlen and De Schryver [19] have investigated the ground state charge transfer complex between MV2+ and pyrene in SDS by a model that allowed them to estimate all parameters related to the static and dynamic quenching processes. PET in micelles is often treated by the model proposed by Tachiya and coworkers [20,21], which involves immobile fluorophores and a migration of quenchers between the micelles and aquous phase. In another model proposed by Almgren and Lofroth [22], the effect of polydispersity on fluorescence quenching in micelles is considered. In this context, a closer look at the fluorescence lifetimes of our system is required, with an appropriate model that is described below. The fluorescence decay of chlorin p6 in water is single exponential. In SDS, the decay becomes biexponential, Table 2 Comparison of KSV for quenching of chlorin p6 by MV2+ in 14 mM SDS from steady state and time resolved experiments using average lifetimes 1
KSV (lit mol ) kq (lit mol 1 s 1)
Steady state
Time resolved
26 370 7.4 · 1012
331 0.94 · 1011
T.K. Mukherjee et al. / Chemical Physics Letters 407 (2005) 119–123
Table 3 Results of global analysis of fluorescence decays of 4 · 10 p6 in 14 mM SDS at different concentrations of MV2+
6
τav(0)/τav
1.3
(a)
1.2
H2O SDS
1.1
1.0 0
30
60 2+
90
5
[MV ]x10 (M) (b)
( τ1)0/τ1
( τav)0/τav
1.4
( τ av)0/τav
indicating that chlorin p6 molecules experience two different kinds of environments in this microheterogeneous medium. Since no dynamic quenching is observed in bulk water, we can safely assign the dynamic quenching observed in SDS to PET between micelle-bound chlorin p6 and methyl viologen. The micellar quenching model proposed by Tachiya yields unacceptable results and poor fits, probably because the migration of MV2+ between aqueous and micellar phase is not significant and MV2+ is mostly localized near the Stern layer of SDS micelle [23]. Moreover, chlorin p6 in the Stern layer may not be totally immobile. Since the absorption spectra indicate a disruption of the ground state association complexes between chlorin p6 and MV2+, a model like that proposed by Gehlen and De schryver would also be inapplicable [19]. A more applicable model would be similar to those proposed by Tominaga and co-workers [24] or by Levstein and van Willigen [25]. Such a model would involve two kinds of fluorophores: one which are available for quenching by MV2+ and the other, which are not. If this model is appropriate, then the decays would be biexponential. Only one component of the lifetime should exhibit a decrease on addition of MV2+ and the other, corresponding to unavailable fluorophores, should remain unchanged. If the free-bound equilibrium of chlorin p6 with SDS is unaltered by the addition of MV2+, then the weightage of the components should remain unaltered as well. Thus, in this model, the only variable in the biexponential decay would be the shorter lifetime and it is this lifetime that should be used to construct the Stern–Volmer plot. A gobal analysis of the decays is required with the amplitudes and one lifetime as the global variables and the second lifetime as the local variable [26–28]. The results of the global analysis are shown in Table 3. The Stern–Volmer plot using lifetimes is shown in Fig. 5b. The plot is linear upto a MV2+ concentration of 20 mM, but exhibits an upward deviation subsequently, indicating a breakdown of the model at such concentrations where the bound-free equilibrium of
( τ1)0/τ1
122
1.2
1.0 0
10
20 2+
30
5
[MV ]x10 (M) Fig. 5. Stern–Volmer plots with (a) average lifetimes obtained from biexponential fitting of the fluorescence decays of 4 · 10 6 M chlorin p6 in 14 mM SDS (hollow circles) and in water (solid circles); (b) shorter component s1 obtained from global analysis of the same data. The dashed line joining the points is a mere aid to the eye. The plot with average lifetime is shown alongside for comparison (hollow circles).
chlorin p6 with micelles might be affected, as MV2+ can cement the negatively charged SDS and chlorin p6, at high concentration. However, the linearity of the Stern–Volmer plot upto 20 mM MV2+ indicates the viability of the model till such concentrations. The KSV and kq values are 1140 M 1 and 3.2 · 1011 M 1 s 1, respectively, which are higher than those obtained using the average lifetimes, but are still much lower than that obtained from the steady state data. The lower value of KSV obtained compared to the steady state analysis even with this model indicates that a large part of the quenching is still instantaneous in nature.
M chlorin
4. Conclusion
[MV2+] · 105 (M)
s1 (ns)
s2 (ns)
a1
a2
v2
0 1 5 10 15 20 30
3.1 3.0 2.89 2.75 2.63 2.5 2.1
4.2 4.2 4.2 4.2 4.2 4.2 4.2
0.57 0.57 0.57 0.57 0.57 0.57 0.57
0.43 0.43 0.43 0.43 0.43 0.43 0.43
1.1 1.17 1.14 1.04 1.04 1.07 1.38
The data were fit to a biexponential function with the amplitudes and the longer lifetime as global parameters and the first lifetime, s1, as the local parameter.
We have reported the effects of different surfactants on the PET between chlorin p6 and MV2+. The PET is totally static in water due to formation of contact ionpairs between the fluorophore and the quencher. On addition of micelles, these contact ion pairs get disrupted. In the positively charged CTAB and neutral Triton X-100, quenching is suppressed as chlorin p6 is incorporated in the micelle and MV2+is in the aqueous phase, thereby eliminating the chance of electron
T.K. Mukherjee et al. / Chemical Physics Letters 407 (2005) 119–123
transfer. In SDS, however, the negatively charged micellar surface attracts MV2+ towards the chlorin p6 ions and so, diffusion controlled PET is observed. The KSV obtained from time resolved data is significantly lower than those from the steady state results, indicating that a significant portion of the quenching process is instantaneous. Further experiments in the femtosecond time scale are required to investigate such instantaneous quenching. Microsecond to millisecond timescale studies also need to be performed to see whether the products of such PET are long-lived or not. Acknowledgements This work is supported by CSIR research Grant No. 01 (1851)/03/EMR-II to AD and UGC Junior Research Fellowship to TKM. The authors thank Ms. Jaya Bhatnagar for her participation in the initial stage of these experiments and Professor G. Naresh Patwari for some helpful comments. References [1] S.C.M. Gandini, V.E. Yushmanov, I.E. Borissevitch, M. Tabak, Langmuir 15 (1999) 6233. [2] A.S.R. Koti, J. Taneja, N. Periasamy, Chem. Phys. Lett. 375 (2003) 171. [3] K. Das, B. Jain, A. Dube, P.K. Gupta, Chem. Phys. Lett. 401 (2005) 185. [4] A.. Datta, A. Dube, B. Jain, A. Tiwari, P.K. Gupta, Photochem. Photobiol. 75 (2002) 488.
123
[5] P.P. Mishra, J. Bhatnagar, A. Datta, Chem. Phys. Lett. 386 (2004) 158. [6] A.S.R. Koti, N. Periasamy, J. Mater. Chem. 12 (2002) 2312. [7] M. Kumbhakar, S. Nath, H. Pal, A.V. Sapre, T. Mukherjee, J. Chem. Phys. 119 (2003) 388. [8] B.L. Bales, M. Almgren, J. Phys. Chem. 99 (1995) 15153. [9] K. Bhattacharyya, Acc. Chem. Res. 36 (2003) 95. [10] J.C. Dederen, M. Van der Auweraer, F.C. De Schryver, J. Phys. Chem. 85 (1981) 1198. [11] M. Aoudia, M.A.J. Rodgers, J. Phys. Chem. B 107 (2003) 6194. [12] D.M. Togashi, S.M.B. Costa, Phys. Chem. Chem. Phys. 4 (2002) 1141. [13] P.P. Mishra, A.L. Koner, A. Datta, Chem. Phys. Lett. 400 (2004) 128. [14] D.M. Togashi, S.M.B. Costa, New J. Chem. 26 (2002) 1774. [15] R.H. Schmehl, D.G. Whitten, J. Phys. Chem. 85 (1981) 3473. [16] D. Mandal, S.K. Pal, D. Sukul, K. Bhattacharyya, J. Phys. Chem. A 103 (1999) 8156. [17] S.J. Atherton, J.H. Baxendale, B.M. Hoey, J. Chem. Soc. Faraday Trans. 1 78 (1982) 2167. [18] D.R. James, W.R. Ware, Chem. Phys. Lett. 120 (1985) 455. [19] M.H. Gehlen, F.C. De Schryver, J. Phys. Chem. 97 (1993) 11242. [20] A.V. Barzykin, K. Seki, M. Tachiya, J. Phys. Chem. B 103 (1999) 9156. [21] M. Tachiya, Chem. Phys. Lett. 33 (1975) 29. [22] M. Almgren, J.E. Lofroth, J. Chem. Phys. 76 (1982) 2734. [23] S.M. Hubig, J. Phys. Chem. 96 (1992) 2903. [24] S. Pant, H. Ohtaka-Saiki, M. Takezaki, T. Tominaga, J. Mol. Liq. 90 (2001) 121. [25] P.R. Levstein, H. van Willigen, Chem. Phys. Lett. 187 (1991) 415. [26] G. Vermunicht, N. Boens, F.C. de Schryver, Photochem. Photobiol. 53 (1991) 57. [27] L. Crovetto, A. Orte, E.M. Talavera, J.M. Alvarez-Pez, M. Cotlet, J. Thielemans, F.C. De Schryver, N. Boens, J. Phys. Chem. A 108 (2004) 6082. [28] M. Halder, K. Das, P.K. Chowdhury, S. Kundu, M.S. Hargrove, J.W. Petrich, J. Phys. Chem. B 107 (2003) 9933.
Photoinduced electron transfer from chlorin p6 to methyl viologen in aqueous micelles Tushar Kanti Mukherjee, Padmaja P. Mishra, Anindya Datta
*
Department of Chemistry, Indian Institute of Technology, Bombay Powai, Mumbai 400 076, India Received 30 December 2004; in final form 7 March 2005 Available online 7 April 2005
Abstract We report the effect of micellar encapsulation on the photoinduced electron transfer between chlorin p6 and methyl viologen. In aqueous solutions, these molecules form contact ion pairs and the fluorescence quenching observed is mainly static in nature. The quenching efficiency increases markedly in SDS micelles and the mechanism changes partially to dynamic quenching. However, fluorescence quenching of all kinds are suppressed in CTAB and Triton X-100 micelles. This is explained in the light of dissociation of the contact ion pairs and the electrostatic attraction of MV2+ by the headgroups of the surfactants. 2005 Elsevier B.V. All rights reserved.
1. Introduction Aggregation and photoinduced electron and energy transfer in porphyrin based molecules have been studied extensively, in attempts to understand and mimic green plant photosynthesis and to develop efficient solar cells [1–6]. Recently, there has been a spate of activity in the field of PET in organized media [7,8] where the excited state dynamics differ largely from those in homogeneous solutions [9,10]. The motivation for the study of PET in porphyrinic systems in organized media stems from the urge to generate long-lived ion pairs, so that the energy produced thus can be used fruitfully. If the D–A distance is too large for the reactants to diffuse to each other during the excited state lifetimes, then PET is suppressed. On the other hand, if the distance is comparable to the van der Waal radius, then the back electron transfer is also ultrafast, leading to a very shortlived ion pair. The challenge, therefore, lies in the design of systems where this distance can be controlled in order
*
Corresponding author. Fax: + 91 22 2572 3480. E-mail address: [email protected] (A. Datta).
0009-2614/$ - see front matter 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.cplett.2005.03.055
to have an efficient forward electron transfer and a negligible back electron transfer. This goal can be achieved either by synthesis or by entrapment of the donor or acceptor in organized assemblies. The encapsulation effect of organized assemblies can bring the donor and the acceptor together and thus facilitate the forward electron transfer whereas their hydrophobic and hydrophilic microenvironments can provide differential solubilization to the ion pair formed, in case one of the products is more hydrophobic than the other, leading to the formation of solvent separated ion pairs and consequent suppression of the back electron transfer [11,12]. In the present study, we have investigated the effect of micelles on photoinduced electron transfer between chlorin p6 and methyl viologen dication (MV2+) (Fig. 1).
2. Experimental Chlorin p6 is prepared using literature procedure [3–5]. Methyl viologen dichloride and Triton X-100 (TX-100) (AR grade) from Aldrich have been used as received. Sodium dodecyl sulphate (SDS) and cetyl trimethyl ammonium bromide (CTAB) (AR grade) from
120
T.K. Mukherjee et al. / Chemical Physics Letters 407 (2005) 119–123 0.4 (a) Soret
NH
N
N
HN
CO2H
Me +
3xQ
N
CO2H
N
0.2
0.0 0.4
+
(b)
Me
HO2C Chlorin p6
H2O
0.2
Methyl viologen dication
Fig. 1. Structures of chlorin p6 and MV2+.
Aldrich are recrystallized from methanol–water solutions. The concentration of chlorin p6 used is 4 lM. The concentrations of SDS, CTAB and Triton X-100 are 14, 3 and 1 mM, respectively. The absorption and fluorescence spectra have been recorded on JASCO V570 spectrophotometer with 2 nm bandpass and a Perkin–Elmer LS-55 spectrofluorimeter with kex = 401 nm, respectively. Fluorescence decays have been recorded on an IBH Fluorocube time resolved fluorescence spectrometer [13]. The fluorescence decays were fitted to biexponential functions by iterative reconvolution using the IBH DAS 6.0 software. Global analysis of the decays were performed using a homemade program.
O.D.
CTAB 0.0 0.4 (c)
0.2
TX-100 0.0
0.4 (d)
0.2
SDS
3. Results and discussion The absorption peaks of chlorin p6 in water get red shifted and broadened with gradual addition of MV2+ (Fig. 2a), indicating ground state association due to their opposite charges [14]. Little, if any shift or broadening of the absorption peak is observed with addition of MV2+ to chlorin p6 in 3 mM CTAB and in 1 mM TX-100 (Fig. 2b and c). In 14 mM SDS, the changes in absorption spectra on addition of MV2+ are less prominent than in bulk water, but are more prominent than in the other micelles (Fig. 2d). This indicates that the extent of contact ion pair formation is reduced substantially in micelles compared to neat aqueous solutions. The Stern–Volmer plot in water is linear (Fig. 3a). In CTAB and TX-100, the slope of the Stern– Volmer plot is zero, except for an initial small quenching in TX-100 upto a MV2+ concentration of 4 · 10 5 M, (Fig. 3). These indicate a complete incorporation of chlorin p6 in CTAB micelle, resulting in effective sequestering from MV2+, which resides in bulk water. In TX100, chlorin p6 may not be totally bound and MV2+ can initially quench the free fluorophore molecules. The most remarkable result is obtained with SDS, where a linear Stern–Volmer plot, with a steeper slope than
0.0 350
400
650
700
λ (nm)
Fig. 2. Principal absorption peaks of chlorin p6 in the absence (solid line) and presence (dotted line) of 1 mM MV2+ in (a) water, (b) 3 mM CTAB, (c) 1 mM Triton X-100 and (d) 14 mM SDS. The Q bands are magnified for the sake of clarity.
that in neat aqueous solution, is obtained (Fig. 3a, Table 1). This is an indication of the changeover of the mechanism to a more efficient one. In SDS, where the association complexes are disrupted, the quenching mechanism might be predominantly dynamic in nature [15]. Interestingly, the KSV is greater in SDS micellar media than in neat aqueous solutions, which is in contrast to some earlier observations [7,15,16]. The decrease in KSV values in micelles is generally rationalized by a decreased mobility of incorporated quenchers and fluorophores. However, an increase in KSV is also observed in certain cases. For example, an enhanced quenching of ruthenium complexes by MV2+ and Cu2+ is observed in SDS solutions, and is explained by the formation of clusters of the fluorophore with SDS, with which the oppositely charged quenchers associate easier [17]. In our experiment, since chlorin p6 resides near the
Counts
T.K. Mukherjee et al. / Chemical Physics Letters 407 (2005) 119–123 10
4
10
3
10
2
10
1
10
0
121
(a) No MV 1 mM MV
2+
2+
H 2O
10
4
10
3
10
2
10
1
10
0
(b) 2+
No MV 1 mM MV
2+
SDS
0
8
16
Time (ns)
Fig. 3. Stern–Volmer plots with fluorescence quantum yields (/f) of chlorin p6 in (a) water, 3 mM CTAB and in 14 mM SDS micelles and (b) water and 1 mM Triton X-100.
Table 1 Values of KSV and kq from steady state experiments in different media Ksv (lit mol 1) kq (lit mol 1 s 1)
H2O
SDS
Triton X-100
1560 4.58 · 1011
26 370 7.4 · 1012
11 520 3.1 · 1012
negatively charged headgroups of SDS, this charged surface might offer electrostatic attraction to draw the MV2+ ions towards the fluorophore. In order to gain a better insight to the situation, time resolved experiment have been performed. The fluorescence decays of chlorin p6 in water exhibit no change on addition of MV2+ (Fig. 4a). This confirms that in water, only static quenching is operative and can be explained in the light of instantaneous electron transfer in the association complexes between MV2+ and chlorin p6 [14]. Dynamic quenching might be observed in micelles, where these association complexes are disrupted. However, there is no change in the fluorescence decays in CTAB or Triton X-100 micelles, indicating the absence of dynamic quenching by MV2+ in these micelles. In SDS micelle, the fluorescence decays become faster with addition of MV2+ (Fig. 4b), indicating the dynamic nature of quenching [15]. The decays are well fit to biexponential functions and P the Stern–Volmer plots with average lifetime sav = aisi is shown in Fig. 5a. The Stern–Volmer constant is much smaller compared to that obtained from the steady state experiment (Tables 1 and 2). Similar discrepancies have been
Fig. 4. Fluorescence decays in the presence and absence of 1 mM MV2+ in (a) water and (b) 14 mM SDS. The instrument response functions (FWHM = 250 ps) are shown in dashed lines.
reported in some other studies of bimolecular processes in micelles [7]. We feel that this discrepancy arises out of the oversimplified model of data fitting. The analysis of multiexponential fluorescence decays is by no means a trivial procedure and the results are strongly dependent on the model used to describe the system under investigation [18]. Several such models have been proposed so far [19–24]. For example, Gehlen and De Schryver [19] have investigated the ground state charge transfer complex between MV2+ and pyrene in SDS by a model that allowed them to estimate all parameters related to the static and dynamic quenching processes. PET in micelles is often treated by the model proposed by Tachiya and coworkers [20,21], which involves immobile fluorophores and a migration of quenchers between the micelles and aquous phase. In another model proposed by Almgren and Lofroth [22], the effect of polydispersity on fluorescence quenching in micelles is considered. In this context, a closer look at the fluorescence lifetimes of our system is required, with an appropriate model that is described below. The fluorescence decay of chlorin p6 in water is single exponential. In SDS, the decay becomes biexponential, Table 2 Comparison of KSV for quenching of chlorin p6 by MV2+ in 14 mM SDS from steady state and time resolved experiments using average lifetimes 1
KSV (lit mol ) kq (lit mol 1 s 1)
Steady state
Time resolved
26 370 7.4 · 1012
331 0.94 · 1011
T.K. Mukherjee et al. / Chemical Physics Letters 407 (2005) 119–123
Table 3 Results of global analysis of fluorescence decays of 4 · 10 p6 in 14 mM SDS at different concentrations of MV2+
6
τav(0)/τav
1.3
(a)
1.2
H2O SDS
1.1
1.0 0
30
60 2+
90
5
[MV ]x10 (M) (b)
( τ1)0/τ1
( τav)0/τav
1.4
( τ av)0/τav
indicating that chlorin p6 molecules experience two different kinds of environments in this microheterogeneous medium. Since no dynamic quenching is observed in bulk water, we can safely assign the dynamic quenching observed in SDS to PET between micelle-bound chlorin p6 and methyl viologen. The micellar quenching model proposed by Tachiya yields unacceptable results and poor fits, probably because the migration of MV2+ between aqueous and micellar phase is not significant and MV2+ is mostly localized near the Stern layer of SDS micelle [23]. Moreover, chlorin p6 in the Stern layer may not be totally immobile. Since the absorption spectra indicate a disruption of the ground state association complexes between chlorin p6 and MV2+, a model like that proposed by Gehlen and De schryver would also be inapplicable [19]. A more applicable model would be similar to those proposed by Tominaga and co-workers [24] or by Levstein and van Willigen [25]. Such a model would involve two kinds of fluorophores: one which are available for quenching by MV2+ and the other, which are not. If this model is appropriate, then the decays would be biexponential. Only one component of the lifetime should exhibit a decrease on addition of MV2+ and the other, corresponding to unavailable fluorophores, should remain unchanged. If the free-bound equilibrium of chlorin p6 with SDS is unaltered by the addition of MV2+, then the weightage of the components should remain unaltered as well. Thus, in this model, the only variable in the biexponential decay would be the shorter lifetime and it is this lifetime that should be used to construct the Stern–Volmer plot. A gobal analysis of the decays is required with the amplitudes and one lifetime as the global variables and the second lifetime as the local variable [26–28]. The results of the global analysis are shown in Table 3. The Stern–Volmer plot using lifetimes is shown in Fig. 5b. The plot is linear upto a MV2+ concentration of 20 mM, but exhibits an upward deviation subsequently, indicating a breakdown of the model at such concentrations where the bound-free equilibrium of
( τ1)0/τ1
122
1.2
1.0 0
10
20 2+
30
5
[MV ]x10 (M) Fig. 5. Stern–Volmer plots with (a) average lifetimes obtained from biexponential fitting of the fluorescence decays of 4 · 10 6 M chlorin p6 in 14 mM SDS (hollow circles) and in water (solid circles); (b) shorter component s1 obtained from global analysis of the same data. The dashed line joining the points is a mere aid to the eye. The plot with average lifetime is shown alongside for comparison (hollow circles).
chlorin p6 with micelles might be affected, as MV2+ can cement the negatively charged SDS and chlorin p6, at high concentration. However, the linearity of the Stern–Volmer plot upto 20 mM MV2+ indicates the viability of the model till such concentrations. The KSV and kq values are 1140 M 1 and 3.2 · 1011 M 1 s 1, respectively, which are higher than those obtained using the average lifetimes, but are still much lower than that obtained from the steady state data. The lower value of KSV obtained compared to the steady state analysis even with this model indicates that a large part of the quenching is still instantaneous in nature.
M chlorin
4. Conclusion
[MV2+] · 105 (M)
s1 (ns)
s2 (ns)
a1
a2
v2
0 1 5 10 15 20 30
3.1 3.0 2.89 2.75 2.63 2.5 2.1
4.2 4.2 4.2 4.2 4.2 4.2 4.2
0.57 0.57 0.57 0.57 0.57 0.57 0.57
0.43 0.43 0.43 0.43 0.43 0.43 0.43
1.1 1.17 1.14 1.04 1.04 1.07 1.38
The data were fit to a biexponential function with the amplitudes and the longer lifetime as global parameters and the first lifetime, s1, as the local parameter.
We have reported the effects of different surfactants on the PET between chlorin p6 and MV2+. The PET is totally static in water due to formation of contact ionpairs between the fluorophore and the quencher. On addition of micelles, these contact ion pairs get disrupted. In the positively charged CTAB and neutral Triton X-100, quenching is suppressed as chlorin p6 is incorporated in the micelle and MV2+is in the aqueous phase, thereby eliminating the chance of electron
T.K. Mukherjee et al. / Chemical Physics Letters 407 (2005) 119–123
transfer. In SDS, however, the negatively charged micellar surface attracts MV2+ towards the chlorin p6 ions and so, diffusion controlled PET is observed. The KSV obtained from time resolved data is significantly lower than those from the steady state results, indicating that a significant portion of the quenching process is instantaneous. Further experiments in the femtosecond time scale are required to investigate such instantaneous quenching. Microsecond to millisecond timescale studies also need to be performed to see whether the products of such PET are long-lived or not. Acknowledgements This work is supported by CSIR research Grant No. 01 (1851)/03/EMR-II to AD and UGC Junior Research Fellowship to TKM. The authors thank Ms. Jaya Bhatnagar for her participation in the initial stage of these experiments and Professor G. Naresh Patwari for some helpful comments. References [1] S.C.M. Gandini, V.E. Yushmanov, I.E. Borissevitch, M. Tabak, Langmuir 15 (1999) 6233. [2] A.S.R. Koti, J. Taneja, N. Periasamy, Chem. Phys. Lett. 375 (2003) 171. [3] K. Das, B. Jain, A. Dube, P.K. Gupta, Chem. Phys. Lett. 401 (2005) 185. [4] A.. Datta, A. Dube, B. Jain, A. Tiwari, P.K. Gupta, Photochem. Photobiol. 75 (2002) 488.
123
[5] P.P. Mishra, J. Bhatnagar, A. Datta, Chem. Phys. Lett. 386 (2004) 158. [6] A.S.R. Koti, N. Periasamy, J. Mater. Chem. 12 (2002) 2312. [7] M. Kumbhakar, S. Nath, H. Pal, A.V. Sapre, T. Mukherjee, J. Chem. Phys. 119 (2003) 388. [8] B.L. Bales, M. Almgren, J. Phys. Chem. 99 (1995) 15153. [9] K. Bhattacharyya, Acc. Chem. Res. 36 (2003) 95. [10] J.C. Dederen, M. Van der Auweraer, F.C. De Schryver, J. Phys. Chem. 85 (1981) 1198. [11] M. Aoudia, M.A.J. Rodgers, J. Phys. Chem. B 107 (2003) 6194. [12] D.M. Togashi, S.M.B. Costa, Phys. Chem. Chem. Phys. 4 (2002) 1141. [13] P.P. Mishra, A.L. Koner, A. Datta, Chem. Phys. Lett. 400 (2004) 128. [14] D.M. Togashi, S.M.B. Costa, New J. Chem. 26 (2002) 1774. [15] R.H. Schmehl, D.G. Whitten, J. Phys. Chem. 85 (1981) 3473. [16] D. Mandal, S.K. Pal, D. Sukul, K. Bhattacharyya, J. Phys. Chem. A 103 (1999) 8156. [17] S.J. Atherton, J.H. Baxendale, B.M. Hoey, J. Chem. Soc. Faraday Trans. 1 78 (1982) 2167. [18] D.R. James, W.R. Ware, Chem. Phys. Lett. 120 (1985) 455. [19] M.H. Gehlen, F.C. De Schryver, J. Phys. Chem. 97 (1993) 11242. [20] A.V. Barzykin, K. Seki, M. Tachiya, J. Phys. Chem. B 103 (1999) 9156. [21] M. Tachiya, Chem. Phys. Lett. 33 (1975) 29. [22] M. Almgren, J.E. Lofroth, J. Chem. Phys. 76 (1982) 2734. [23] S.M. Hubig, J. Phys. Chem. 96 (1992) 2903. [24] S. Pant, H. Ohtaka-Saiki, M. Takezaki, T. Tominaga, J. Mol. Liq. 90 (2001) 121. [25] P.R. Levstein, H. van Willigen, Chem. Phys. Lett. 187 (1991) 415. [26] G. Vermunicht, N. Boens, F.C. de Schryver, Photochem. Photobiol. 53 (1991) 57. [27] L. Crovetto, A. Orte, E.M. Talavera, J.M. Alvarez-Pez, M. Cotlet, J. Thielemans, F.C. De Schryver, N. Boens, J. Phys. Chem. A 108 (2004) 6082. [28] M. Halder, K. Das, P.K. Chowdhury, S. Kundu, M.S. Hargrove, J.W. Petrich, J. Phys. Chem. B 107 (2003) 9933.