- Email: [email protected]
Photophysical properties of phenosafranine (PHNS) adsorbed on the TiO2-incorporated zeolite-Y
Microporous and Mesoporous Materials 86 (2005) 185–190 www.elsevier.com/locate/micromeso
Photophysical properties of phenosafranine (PHNS) adsorbed on the TiO2-incorporated zeolite-Y S. Easwaramoorthi a
a,b
, P. Natarajan
a,*
National Centre for Ultrafast Processes, University of Madras, Taramani Campus, Chennai 600 113, India b Department of Inorganic Chemistry, University of Madras, Guindy Campus, Chennai 600 025, India Received 17 September 2004; received in revised form 1 March 2005; accepted 11 July 2005 Available online 29 August 2005
Abstract Investigation of photophysical properties of phenosafranine dye ion-exchanged with zeolite-Y samples has been carried out to understand the host–guest interactions of the dye and the zeolite. Photosensitization of titanium dioxide, encapsulated in the supercages and anchored on the external surface of the zeolite-Y by the visible light excited dye molecules, has been studied using steady state and time resolved fluorescence techniques. Titanium dioxide loaded zeolites at the external surface and at the interior of the supercages are found to show different characteristics in the sensitization process. 2005 Elsevier Inc. All rights reserved. Keywords: Zeolite photochemistry; Photosensitization; Multicomponent systems; Titanium dioxide; Phenosafranine; Semiconductor nanoparticles
1. Introduction Supramolecular organization of molecules in molecular sieves is of interest to develop efficient systems for controlling chirality and the nanoscale-advanced materials [1–3]. The non-deformable zeolite structure offers the ordered assembly of multicomponent systems spatially in a single roof for stabilizing the light induced charge separated state by preventing the energy wasting back electron transfer process [4]. However, a formidable task in these systems is to devise a competent method to organize the guest molecules. Many reports are available that demonstrate the occurrence of energy [5] and electron-transfer [6] photosensitization processes using zeolites as host materials. Titanium dioxide semiconductor nanoparticles have been extensively studied for significant applications in
*
Corresponding author. Tel.: +91 44 24480962; fax: +91 44 24926709. E-mail address: [email protected] (P. Natarajan). 1387-1811/$ - see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.micromeso.2005.07.009
photocatalysis [7] and dye sensitized solar cells [8]. The incorporation of titanium dioxide in zeolite cavity offers advantages due to size quantization resulting in different optical and electronic properties [9–12]. In order to improve the efficiency of the titanium dioxide doped zeolite catalyst to operate in the visible region, attempts are made to assemble the semiconductor and sensitizer in the internal and external zeolite surfaces. 2þ RuðbpyÞ3 ion encapsulated in titanium dioxide codoped zeolites [13–15] shows enhanced photocatalytic activity and also the mechanism of photochemical water splitting using these materials was elucidated by Kim et al. [16]. The efficiency of charge transfer from the sensitizer to the titanium dioxide is modulated by capping the nanocrystals with insulating layers [17,18] for applications like photonic crystals and in paint industry. The role played by the zeolite surface, water molecules, and charge balancing cations [19] on the guest molecules for developing efficient systems of interest is not understood well. In this report, we have used phenosafranine (PHNS) as sensitizer which is an azine class
186
S. Easwaramoorthi, P. Natarajan / Microporous and Mesoporous Materials 86 (2005) 185–190
of dye and the dye behaves as an electron acceptor and donor in the ground and excited state [20]. The dye has been used in photogalvanic cells in homogeneous solution as well as coated on the electrode surface by covalently linking the dye with the polymer [21]. The electron injection from the singlet excited state of phenosafranine to the conduction band of the bulk titanium dioxide semiconductor was investigated by diffuse reflectance laser flash photolysis technique [22]. In this report, we discuss the photophysical properties of phenosafranine dye and the sensitization of titanium dioxide nanoparticles located in the zeolite surface as well as encapsulated in the cavities of the zeolite.
2. Experimental Zeolite-Y (Suid Chemie, India), washed with 1 M NaCl for about 2 h, was calcined at 530 C for 12 h. Titanium dioxide anchored on the external surface of the zeolite (1–5% loading (w/w)) was prepared by sol–gel method using tetraisopropyl orthotitanate as reported [23]. In brief, 15 ml of tetraisopropylorthotitanate (E-Merck) was hydrolyzed in water containing 0.75 ml of conc. HNO3 and was peptized for 15 h. The resulting sol is added to the pre-hydrated zeolite and then the suspension is stirred for 2 h and the solvent is evaporated slowly. Finally, the samples were calcined at 450 C. TiO2 encapsulated in the zeolite cavity was prepared by ion-exchange method [10]. In this method, zeolite was stirred with potassium titanooxalate in aqueous solution for 24 h and the zeolite samples were heated at 150 C for 2–3 h. Different loading of titanium dioxide was achieved by successive ion-exchange followed by heat treatment. Actual amount of titanium present in the zeolite samples was determined by atomic absorption spectroscopy and chemical analysis [16,24]. The calculated quantity of titanium dioxide loaded zeolite is digested in concentrated sulfuric acid for about 5 h and small quantity of this parent solution is diluted with water. To this solution, 10% H2O2 is added in excess and titanium is estimated by using the absorbance of the peak at 408 ± 1 nm. Previously the absorbance value is calibrated using pure anatase titanium dioxide under the same experimental conditions. Phenosafranine (PHNS) chloride (3,7-diamino-5-phenylphenazenium chloride) obtained from Fluka was purified by column chromatography [20]. The aqueous solution of the dye and the zeolite when stirred for 3–4 h at room temperature results in the loading of PHNS in zeolite. The concentration of the dye in zeolite is kept at 3 lM per gram of the zeolite-Y. The studies reported are concerned with the samples, titanium dioxide encapsulated in zeolite TY(e) and titanium dioxide on the external surface of the zeolite TY(s).
N
N+
NH2
NH2 Cl-
Phenosafranine, PHNS
UV–visible diffuse reflectance spectra of the samples were recorded using a Schimadzu spectrophotometer equipped with integrating sphere attachment. The steady state fluorescence measurements for opaque samples were done with Fluromax spectrophotometer as front face configuration at 45. Powder X-ray diffraction measurements were recorded using Philips X-ray diffractometer. EDAX measurements were carried out with JSM-840 scanning electron microscope. Time resolved fluorescence measurements were carried out with time correlated single photon counting fluorescence spectrometer (IBH) by exciting the sample at 470 nm with Ti: Sapphire laser of pulse width 2 ps FWHM and the instrument response time of 50 ps. Fluorescence decay was measured at the front face configuration with suitable cutoff filters to avoid scattered light and the decay curve was analyzed using the software (DAS6, IBH) provided with the instrument.
3. Results and discussion The titanium dioxide nanoparticles present in the zeolites are found to be either amorphous in nature or the particle sizes are too small to be detected by the powder XRD, as no evidence for crystalline titanium dioxide was found in the XRD pattern of the samples used for the experiments. The titanium dioxide loading in zeolite, which is lower than that reported earlier [11] (7% decrease in crystallinity at 0.83% for TY(e) samples), does not change the crystallinity of the zeolite to any appreciable extent as shown by XRD pattern. The external surface area of the zeolite samples is about 1% of the inner surface area, which is close to 6 m2 g1 for the particle size of the zeolite around 0.5–1 lm. TiO2 formed by sol–gel method is known to have particle size larger than ˚ ), and are anthat of the zeolite pore opening (7.4 A chored on the external surface of the zeolite [16], which was confirmed by EDAX measurement as well. When the ion-exchange method was used for the preparation of TiO2 incorporated zeolite samples, titanium dioxide nanoclusters thus formed are entrapped in the supercages of the zeolite-Y. The maximum size of the TiO2 cluster present in the supercage may not be larger than 1.3 nm, the diameter of the supercage [10]. This is further confirmed by the UV–visible spectra shown in
S. Easwaramoorthi, P. Natarajan / Microporous and Mesoporous Materials 86 (2005) 185–190
Absorbance, a.u
Fig. 1 exhibiting considerable blue shift in the spectra of all the encapsulated titanium dioxide samples as compared to the surface anchored titanium dioxide and pure anatase. The band gap energies of the titanium dioxide on the surface and encapsulated zeolite samples were calculated from the absorption band edge of the titanium dioxide. The absorption band edge shows blue shift as compared to bulk anatase. The observed blue shift is more pronounced in the case of encapsulated TiO2 and is due to the size quantization of the TiO2 nanoparticles. The band gap energies of 3.5 eV and 4.0 eV have been obtained for TY(s) and TY(e) samples, respectively, whereas that of bulk anatase is 3.2 eV. The EDAX measurement, which identifies only the surface elements, shows that there is no titanium present on the external surface of zeolites in the case of all TiO2 encapsulated zeolite samples while the surface anchored TiO2 samples showed the presence of titanium. The molecular dimension of PHNS molecule ˚ · 8.21 A ˚ ) is larger than that of the pore open(11.35 A ˚ ing (7.4 A) of the zeolite-Y and hence the dye molecules are suggested to be adsorbed only on the external surface of the zeolite. The dye concentration used in these studies is 1 mg/g of the zeolite, which corresponds to less than 5% monolayer coverage of the external surface. The energetics of the ground and excited states of the dye are altered when the dye interacts with the titanium dioxide [25] nanoparticles. Diffuse reflectance spectra of phenosafranine in TY(s) shows blue shift in the absorption maxima, which suggests the interaction between the dye and the titanium dioxide on the external surfaces of
the zeolite. Encapsulation of TiO2 in the supercages does not alter the absorption maxima of the dye as shown in Table 2, as the dye is not in direct contact with TiO2 present in the host lattice. The arrangements of the dye and the semiconductor nanoparticles in the zeolite host are shown in Scheme 1. The emission and the fluorescence lifetime (Table 1) of PHNS samples are sensitive to the solvent polarity [20]. In zeolite samples used in this investigation, the fluorescence maxima of the dye are blue shifted by 21 nm with respect to that of water and are closer to the value of the dye in 2:3 methanol:water solvent mixture. Fluorescence decay curve of PHNS in all the zeolite samples could be satisfactorily fitted to biexponential function (v2 = 1.2 ± 0.2) while in the homogeneous solution the decay fits to a single exponential function (Tables 1 and 2). Analysis of the observed parameters reveals that the biexponential decay may be originated from the two different emissive species formed upon excitation on the zeolite surface, which are equilibrated. Similar results were observed for Nile-Red—incorporated zeolite [26]. In comparison, the fluorescence lifetime measurement of PHNS adsorbed on the silica surface clearly establishes the difference between the zeolite and silica surface. Similar to that of zeolite, PHNS on the silica particles shows two different lifetimes (Table 2). However, the observed fluorescence lifetimes are considerably longer than those observed in zeolites. The shorter lifetime for PHNS in zeolite is probably due to the influence of various parameters on the excited state properties of the dye like the presence of aluminium in the framework and charge-balancing cations in zeolites, whereas these parameters are absent at the silica surfaces. PHNS is known to sensitize the titanium dioxide semiconductor particles through its singlet excited state [22]. Since the oxidation potential of the PHNS*(S1) is around 1.25 V versus NHE and is more negative than that of titanium dioxide conduction band (0.5 V versus NHE) [22], the charge injection from the excited dye into the conduction band of the semiconductor is thermodynamically favored. The possible mechanism is summarised as follows: hm
PHNS ! 1 PHNS
a
1
b
200
187
250
300
350
400
450
500
550
600
Wavelength, nm
Fig. 1. Diffuse reflectance spectrum of TiO2: (a) on the external surface of the zeolite-Y (dotted line) from top to bottom 1%, 2%, 3%, 4%, 5% TiO2 and (b) encapsulated in zeolite-Y (solid line) 0.22%, 0.68%, 0.83% TiO2.
PHNS þ TiO2 ! TiO2 ðeÞ þ PHNSþ
The fluorescence intensity of PHNS* is found to be quenched with increasing loading of the titanium dioxide and is linear up to the addition of 2–3% TiO2 in the case of TY(s) samples as shown in Fig. 2a. Since at this titanium dioxide concentration complete monolayer coverage of the zeolite [23] may occur, and further increase in the concentration of TiO2 has little effect. Meanwhile with the TY(e) samples, fluorescence quenching process is found to be linear with TiO2 concentration and about 80% of the fluorescence is
188
S. Easwaramoorthi, P. Natarajan / Microporous and Mesoporous Materials 86 (2005) 185–190 D
D
D
TiO 2
TiO 2 TiO 2
D TiO 2
D
D
TiO 2
D
D D
TiO 2
D TiO
D
2
D
D D
a
D
D
TiO 2
D
D
b
D
TiO 2
D TiO 2
TiO 2
TiO 2
D
TiO 2
D D
TiO 2
D TiO 2
D
D
D
TiO 2
TiO 2
D TiO 2
D
D TiO 2
D
D
TiO 2
D TiO 2 D
D
TiO 2
TiO 2
TiO 2
D
D
D
D
c
d
D
Scheme 1. Arrangement of dye and semiconductor in zeolite-Y: (a) dye (D) on the zeolite surface; (b) dye on TY(s) samples at lower concentration of TiO2; (c) dye on TY(s) samples at higher concentration of TiO2; (d) dye on TY(e) samples, TiO2 (d).
Table 1 Photophysical properties of PHNS in various solvents Solvent Water Acetonitrile Methanol Methanol:water (1:4) Methanol:water (2:3) Methanol:water (3:2) a b
kmax, absa 520 517 520 524 526 526
kmax, emsa 595 560 572 584 576 574
Fluorescence lifetimeb, ns 0.93 4.1 2.12 1.08 1.43 1.73
kabs and ems = ±2 nm. Lifetime ±0.04 ns.
quenched at 0.68% titanium dioxide concentration due to its high dispersivity in the zeolite cages (Fig. 2b). Fluorescence lifetime measurement of PHNS in titanium dioxide loaded zeolites elucidates the mode of quenching involved in these systems and the data are given in Table 2. The lifetimes and the relative percentage of the emitting species remain constant within the experimental error for the titanium dioxide anchored on the external surface of the zeolite as shown in Fig. 3. It may be noted that the emission intensity is quenched proportional to the amount of titanium dioxide present on the surface. The observed emission is from the dye molecules that are not in contact with the TiO2, while the excited dye molecules, which are in contact with TiO2, are quenched. The mechanism of quenching is essentially static in nature due to the restricted movement
Table 2 Fluorescence lifetime of PHNS in zeolite/silica with titanium dioxide present at the external surface and encapsulated in supercages Host
kmax, absa
kmax, emsa
Fluorescence lifetimeb, ns s1
A1
s2
A2
savg
NaY TY(s) (1%)c TY(s) (2%) TY(s) (3%) TY(s) (4%) TY(s) (5%) TY(e) (0.22%) TY(e) (0.68%) TY(e) (0.83%) Silica Silica–TiO2 (5%)
522 512 512 512 512 512 522 522 522 520 512
574 574 574 574 574 574 574 574 574 575 575
0.48 0.35 0.41 0.46 0.43 0.36 0.34 0.26 0.23 0.85 0.84
25.22 30.86 23.67 23.71 32.79 33.22 29.71 53.83 55.23 31.91 25.46
1.42 1.35 1.60 1.50 1.40 1.34 1.28 1.10 1.01 2.15 2.20
74.74 69.14 76.33 76.29 67.21 66.78 71.29 46.17 44.77 68.09 74.54
1.17 1.04 1.31 1.24 1.08 1.01 1.04 0.65 0.58 1.74 1.85
a
kabs and ems = ±2 nm. Lifetime ±0.04 ns, amplitude ±5%. c Percentage of titanium dioxide, A1 and A2 are relative amplitudes in percentage. b
of the sensitizer and the semiconductor. The observed result thus confirms the mechanism of charge injection process when the excited dye is in contact with the 2þ semiconductor in zeolite as is observed for RuðbpyÞ3 ion and titanium dioxide systems in zeolite-Y reported earlier [13,14]. Identical trends were observed for PHNS adsorbed in titanium dioxide loaded silica surface (Table 2); the fluorescence intensity decreases with increase in titanium dioxide concentration. However, no
S. Easwaramoorthi, P. Natarajan / Microporous and Mesoporous Materials 86 (2005) 185–190 3.0e+6
90
a
a 75 Relative amplitude
2.5e+6
Intensity
2.0e+6 1.5e+6 1.0e+6
60 45 30 15
0.5e+5 0.0 540
189
0 560
580
600 620 640 Wavelength, nm
660
680
700
0
1
2
3 % of TiO2
4
5
85
3.0e+6
b
b
70 Relative amplitude
2.5e+6 2.0e+6 Intensity
6
1.5e+6 1.0e+6
55
40
25 0.5e+5
10 0.0 540
0 560
580
600
620
640
660
680
0.2
0.4
700
1
Fig. 3. Plot of percentage of titanium dioxide versus relative amplitude: (a) TY(s) and (b) TY(e).
1.3 1.2 1.1 Lifetime, ns
change in fluorescence lifetime is observed with increase in titanium dioxide loading and this observation confirms the static quenching process. In the case of TiO2 encapsulated in the supercages, fluorescence lifetime of PHNS is found to decrease (Fig. 4) with increase in TiO2 concentration with change in the relative amplitude of the emitting species (Table 2, Fig. 3b). The decrease in amplitudes of the longer lifetime component is suggested to be responsible for the quenching of the fluorescence intensity. It is worth noting that the quenching of the fluorescence lifetime of PHNS occurs in the zeolite framework by the encapsulated titanium dioxide particles, although there is no direct contact between semiconductor in the cage and the sensitizer at the external surface. Increase in the titanium dioxide concentration filling up more number of supercages by the TiO2 nanoparticles leads to closer approach towards the dye molecule on the external surface and accelerates the charge transfer process which results in further decrease in lifetime. It is known that
0.8
% of TiO2
Wavelength, nm
Fig. 2. Fluorescence spectra of PHNS in zeolite-Y with increase in concentration of TiO2: (a) TY(s), from top 0%, 1%, 2%, 3%, 4% and 5% TiO2, respectively, (b) TY(e), from top 0%, 0.22%, 0.68% and 0.83% TiO2, respectively.
0.6
1.0 0.9 0.8 0.7 0.6 0.5 0.0
0.2
0.4
0.6
0.8
1.0
% of TiO2
Fig. 4. Plot of average lifetime of PHNS versus percentage of TiO2 for TY(e) samples.
capping the titanium dioxide with polymer [17] suppresses the charge injection process to certain extent from PHNS* to the conduction band of titanium dioxide. However, when titanium dioxide nanocrystals are
190
S. Easwaramoorthi, P. Natarajan / Microporous and Mesoporous Materials 86 (2005) 185–190
coated with insulating silica layer [18], the fluorescence quantum yield and lifetime of rhodamine6G increases gradually with increase in thickness of the silica layer by suppressing the electron transfer and the complete coverage of the titania nanocrystal causes total prevention of the electron transfer. The present investigation suggests that while covering the semiconductor nanoparticles in the supercages of zeolites, sensitization is facilitated by the interaction between the semiconductor in the supercage and the dye present in the external surface presumably driven by the dynamics of hydration, porous structure of the zeolite and through the silicate network. The increase in amplitude of the shorter component of PHNS in the TY(e) samples as the concentration of TiO2 is increased indicates that the TiO2 encapsulated into the supercage acts as an electron donor synergistically through the zeolite frameworks. However, the amplitude of the shorter component is not changed when the TiO2 concentration increased on the surface which may be due to the electron transfer caused from the direct interaction between the dye and TiO2 without involvement of the zeolite framework. Further investigations are in progress with other mesoporous silicate materials.
4. Conclusion The photophysical properties of the dye elucidate the role of the internal and external surface of the zeolite in the sensitization of titanium dioxide by the PHNS. Decrease in fluorescence intensity of the dye in presence of titanium dioxide shows that the sensitization of the semiconductor occurs through the singlet excited state of the dye. The operative mechanism for the quenching process in the external surface of the zeolite is static in nature and the interfacial charge transfer is very fast due to static quenching. Decrease in fluorescence lifetime of the dye in titanium dioxide encapsulated systems is suggested to be due to the interaction of the excited state with the semiconductor facilitated by the zeolite network. The increase in the relative amplitude of the shorter component may be due to the TiO2 encapsulated zeolite which acts as an electron donor for PHNS* synergistically. Acknowledgments The investigations reported here are supported by Department of Science and Technology, India. Fellow-
ship received by S.E. from Council of Scientific and Industrial Research, India is gratefully acknowledged.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/ j.micromeso.2005.07.009.
References [1] V. Ramamurthy, in: V. Ramamurthy (Ed.), Photochemistry in Organized and Constrained Media, VCH, New York, 1991. [2] S. Hashimoto, J. Photochem. Photobiol. C: Photochem. Rev. 4 (2003) 19. [3] G. Schulz-Ekloff, D. Wohrle, B.V. Duffel, R.A. Schoonheydt, Micropor. Mesopor. Mater. 51 (2002) 91. [4] P.K. Dutta, M. Ledney, Prog. Inorg. Chem. 44 (1997) 209. [5] H. Maas, G. Calzaferri, Spectrum 16 (2003) 18. [6] K.B. Yoon, Chem. Rev. 93 (1993) 321. [7] A.L. Linsebigler, G. Lu, J.T. Yates Jr., Chem. Rev. 95 (1995) 735. [8] A. Hagfeldt, M. Gratzel, Chem. Rev. 95 (1995) 49. [9] M. Matsuoka, M. Anpo, J. Photochem. Photobiol. C: Photochem. Rev. 3 (2003) 225. [10] S. Corrent, G. Cosa, J.C. Scaiano, M.S. Galletero, M. Alvaro, H. Garcia, Chem. Mater. 13 (2001) 715. [11] G. Cosa, M.S. Galletero, L. Fernandez, F. Marquez, H. Garcia, J.C. Scaiano, New J. Chem. 26 (2002) 1448. [12] A. Corma, H. Garcia, Chem. Commun. (2004) 1443. [13] S.H. Bossmann, C. Turro, C. Schnabel, M.R. Pokhrel, L.M. Payawan Jr., B. Baumeister, M. Worner, J. Phys. Chem. B 105 (2001) 5374. [14] G. Cosa, M.N. Chretien, M.S. Galletero, V. Fornes, H. Garcia, J.C. Scaiano, J. Phys. Chem. B 106 (2002) 2460. [15] S.H. Bossmann, S. Jockusch, P. Schwarz, B. Baumeister, S. Gob, C. Schnabel, L. Payawan Jr., M.R. Pokhrel, M. Worner, A.M. Braun, N.J. Turro, Photochem. Photobiol. Sci. 2 (2003) 477. [16] Y. Kim II, S.W. Keller, J.S. Krueger, E.H. Yonemoto, G.B. Saupe, T.E. Mallouk, J. Phys. Chem. B 101 (1997) 2491. [17] L. Ziolkowski, K. Vinodgopal, P.V. Kamat, Langmuir 13 (1997) 3124. [18] L. Bechger, A.F. Koenderink, W.L. Vos, Langmuir 18 (2002) 2444. [19] S. Uppili, K.J. Thomas, E.M. Crompton, V. Ramamurthy, Langmuir 16 (2000) 265. [20] K.R. Gopidas, P.V. Kamat, J. Photochem. Photobiol. A: Chem. 48 (1989) 291. [21] P. Natarajan, J. Macromol. Sci. Chem. 81 (1988) 1285. [22] P.V. Kamat, K.R. Gopidas, D. Weir, Chem. Phys. Lett. 149 (1988) 491. [23] Y. Xu, C.H. Langford, J. Phys. Chem. 99 (1995) 11501. [24] Z. Marcezenko, Spectrophotometric Determination of Elements, Wiley & Sons, New York, 1976. [25] P.V. Kamat, Chem. Rev. 93 (1993) 267. [26] M. Yoon, S.Y. Ryu, Res. Chem. Intermed. 30 (2004) 207.
Photophysical properties of phenosafranine (PHNS) adsorbed on the TiO2-incorporated zeolite-Y S. Easwaramoorthi a
a,b
, P. Natarajan
a,*
National Centre for Ultrafast Processes, University of Madras, Taramani Campus, Chennai 600 113, India b Department of Inorganic Chemistry, University of Madras, Guindy Campus, Chennai 600 025, India Received 17 September 2004; received in revised form 1 March 2005; accepted 11 July 2005 Available online 29 August 2005
Abstract Investigation of photophysical properties of phenosafranine dye ion-exchanged with zeolite-Y samples has been carried out to understand the host–guest interactions of the dye and the zeolite. Photosensitization of titanium dioxide, encapsulated in the supercages and anchored on the external surface of the zeolite-Y by the visible light excited dye molecules, has been studied using steady state and time resolved fluorescence techniques. Titanium dioxide loaded zeolites at the external surface and at the interior of the supercages are found to show different characteristics in the sensitization process. 2005 Elsevier Inc. All rights reserved. Keywords: Zeolite photochemistry; Photosensitization; Multicomponent systems; Titanium dioxide; Phenosafranine; Semiconductor nanoparticles
1. Introduction Supramolecular organization of molecules in molecular sieves is of interest to develop efficient systems for controlling chirality and the nanoscale-advanced materials [1–3]. The non-deformable zeolite structure offers the ordered assembly of multicomponent systems spatially in a single roof for stabilizing the light induced charge separated state by preventing the energy wasting back electron transfer process [4]. However, a formidable task in these systems is to devise a competent method to organize the guest molecules. Many reports are available that demonstrate the occurrence of energy [5] and electron-transfer [6] photosensitization processes using zeolites as host materials. Titanium dioxide semiconductor nanoparticles have been extensively studied for significant applications in
*
Corresponding author. Tel.: +91 44 24480962; fax: +91 44 24926709. E-mail address: [email protected] (P. Natarajan). 1387-1811/$ - see front matter 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.micromeso.2005.07.009
photocatalysis [7] and dye sensitized solar cells [8]. The incorporation of titanium dioxide in zeolite cavity offers advantages due to size quantization resulting in different optical and electronic properties [9–12]. In order to improve the efficiency of the titanium dioxide doped zeolite catalyst to operate in the visible region, attempts are made to assemble the semiconductor and sensitizer in the internal and external zeolite surfaces. 2þ RuðbpyÞ3 ion encapsulated in titanium dioxide codoped zeolites [13–15] shows enhanced photocatalytic activity and also the mechanism of photochemical water splitting using these materials was elucidated by Kim et al. [16]. The efficiency of charge transfer from the sensitizer to the titanium dioxide is modulated by capping the nanocrystals with insulating layers [17,18] for applications like photonic crystals and in paint industry. The role played by the zeolite surface, water molecules, and charge balancing cations [19] on the guest molecules for developing efficient systems of interest is not understood well. In this report, we have used phenosafranine (PHNS) as sensitizer which is an azine class
186
S. Easwaramoorthi, P. Natarajan / Microporous and Mesoporous Materials 86 (2005) 185–190
of dye and the dye behaves as an electron acceptor and donor in the ground and excited state [20]. The dye has been used in photogalvanic cells in homogeneous solution as well as coated on the electrode surface by covalently linking the dye with the polymer [21]. The electron injection from the singlet excited state of phenosafranine to the conduction band of the bulk titanium dioxide semiconductor was investigated by diffuse reflectance laser flash photolysis technique [22]. In this report, we discuss the photophysical properties of phenosafranine dye and the sensitization of titanium dioxide nanoparticles located in the zeolite surface as well as encapsulated in the cavities of the zeolite.
2. Experimental Zeolite-Y (Suid Chemie, India), washed with 1 M NaCl for about 2 h, was calcined at 530 C for 12 h. Titanium dioxide anchored on the external surface of the zeolite (1–5% loading (w/w)) was prepared by sol–gel method using tetraisopropyl orthotitanate as reported [23]. In brief, 15 ml of tetraisopropylorthotitanate (E-Merck) was hydrolyzed in water containing 0.75 ml of conc. HNO3 and was peptized for 15 h. The resulting sol is added to the pre-hydrated zeolite and then the suspension is stirred for 2 h and the solvent is evaporated slowly. Finally, the samples were calcined at 450 C. TiO2 encapsulated in the zeolite cavity was prepared by ion-exchange method [10]. In this method, zeolite was stirred with potassium titanooxalate in aqueous solution for 24 h and the zeolite samples were heated at 150 C for 2–3 h. Different loading of titanium dioxide was achieved by successive ion-exchange followed by heat treatment. Actual amount of titanium present in the zeolite samples was determined by atomic absorption spectroscopy and chemical analysis [16,24]. The calculated quantity of titanium dioxide loaded zeolite is digested in concentrated sulfuric acid for about 5 h and small quantity of this parent solution is diluted with water. To this solution, 10% H2O2 is added in excess and titanium is estimated by using the absorbance of the peak at 408 ± 1 nm. Previously the absorbance value is calibrated using pure anatase titanium dioxide under the same experimental conditions. Phenosafranine (PHNS) chloride (3,7-diamino-5-phenylphenazenium chloride) obtained from Fluka was purified by column chromatography [20]. The aqueous solution of the dye and the zeolite when stirred for 3–4 h at room temperature results in the loading of PHNS in zeolite. The concentration of the dye in zeolite is kept at 3 lM per gram of the zeolite-Y. The studies reported are concerned with the samples, titanium dioxide encapsulated in zeolite TY(e) and titanium dioxide on the external surface of the zeolite TY(s).
N
N+
NH2
NH2 Cl-
Phenosafranine, PHNS
UV–visible diffuse reflectance spectra of the samples were recorded using a Schimadzu spectrophotometer equipped with integrating sphere attachment. The steady state fluorescence measurements for opaque samples were done with Fluromax spectrophotometer as front face configuration at 45. Powder X-ray diffraction measurements were recorded using Philips X-ray diffractometer. EDAX measurements were carried out with JSM-840 scanning electron microscope. Time resolved fluorescence measurements were carried out with time correlated single photon counting fluorescence spectrometer (IBH) by exciting the sample at 470 nm with Ti: Sapphire laser of pulse width 2 ps FWHM and the instrument response time of 50 ps. Fluorescence decay was measured at the front face configuration with suitable cutoff filters to avoid scattered light and the decay curve was analyzed using the software (DAS6, IBH) provided with the instrument.
3. Results and discussion The titanium dioxide nanoparticles present in the zeolites are found to be either amorphous in nature or the particle sizes are too small to be detected by the powder XRD, as no evidence for crystalline titanium dioxide was found in the XRD pattern of the samples used for the experiments. The titanium dioxide loading in zeolite, which is lower than that reported earlier [11] (7% decrease in crystallinity at 0.83% for TY(e) samples), does not change the crystallinity of the zeolite to any appreciable extent as shown by XRD pattern. The external surface area of the zeolite samples is about 1% of the inner surface area, which is close to 6 m2 g1 for the particle size of the zeolite around 0.5–1 lm. TiO2 formed by sol–gel method is known to have particle size larger than ˚ ), and are anthat of the zeolite pore opening (7.4 A chored on the external surface of the zeolite [16], which was confirmed by EDAX measurement as well. When the ion-exchange method was used for the preparation of TiO2 incorporated zeolite samples, titanium dioxide nanoclusters thus formed are entrapped in the supercages of the zeolite-Y. The maximum size of the TiO2 cluster present in the supercage may not be larger than 1.3 nm, the diameter of the supercage [10]. This is further confirmed by the UV–visible spectra shown in
S. Easwaramoorthi, P. Natarajan / Microporous and Mesoporous Materials 86 (2005) 185–190
Absorbance, a.u
Fig. 1 exhibiting considerable blue shift in the spectra of all the encapsulated titanium dioxide samples as compared to the surface anchored titanium dioxide and pure anatase. The band gap energies of the titanium dioxide on the surface and encapsulated zeolite samples were calculated from the absorption band edge of the titanium dioxide. The absorption band edge shows blue shift as compared to bulk anatase. The observed blue shift is more pronounced in the case of encapsulated TiO2 and is due to the size quantization of the TiO2 nanoparticles. The band gap energies of 3.5 eV and 4.0 eV have been obtained for TY(s) and TY(e) samples, respectively, whereas that of bulk anatase is 3.2 eV. The EDAX measurement, which identifies only the surface elements, shows that there is no titanium present on the external surface of zeolites in the case of all TiO2 encapsulated zeolite samples while the surface anchored TiO2 samples showed the presence of titanium. The molecular dimension of PHNS molecule ˚ · 8.21 A ˚ ) is larger than that of the pore open(11.35 A ˚ ing (7.4 A) of the zeolite-Y and hence the dye molecules are suggested to be adsorbed only on the external surface of the zeolite. The dye concentration used in these studies is 1 mg/g of the zeolite, which corresponds to less than 5% monolayer coverage of the external surface. The energetics of the ground and excited states of the dye are altered when the dye interacts with the titanium dioxide [25] nanoparticles. Diffuse reflectance spectra of phenosafranine in TY(s) shows blue shift in the absorption maxima, which suggests the interaction between the dye and the titanium dioxide on the external surfaces of
the zeolite. Encapsulation of TiO2 in the supercages does not alter the absorption maxima of the dye as shown in Table 2, as the dye is not in direct contact with TiO2 present in the host lattice. The arrangements of the dye and the semiconductor nanoparticles in the zeolite host are shown in Scheme 1. The emission and the fluorescence lifetime (Table 1) of PHNS samples are sensitive to the solvent polarity [20]. In zeolite samples used in this investigation, the fluorescence maxima of the dye are blue shifted by 21 nm with respect to that of water and are closer to the value of the dye in 2:3 methanol:water solvent mixture. Fluorescence decay curve of PHNS in all the zeolite samples could be satisfactorily fitted to biexponential function (v2 = 1.2 ± 0.2) while in the homogeneous solution the decay fits to a single exponential function (Tables 1 and 2). Analysis of the observed parameters reveals that the biexponential decay may be originated from the two different emissive species formed upon excitation on the zeolite surface, which are equilibrated. Similar results were observed for Nile-Red—incorporated zeolite [26]. In comparison, the fluorescence lifetime measurement of PHNS adsorbed on the silica surface clearly establishes the difference between the zeolite and silica surface. Similar to that of zeolite, PHNS on the silica particles shows two different lifetimes (Table 2). However, the observed fluorescence lifetimes are considerably longer than those observed in zeolites. The shorter lifetime for PHNS in zeolite is probably due to the influence of various parameters on the excited state properties of the dye like the presence of aluminium in the framework and charge-balancing cations in zeolites, whereas these parameters are absent at the silica surfaces. PHNS is known to sensitize the titanium dioxide semiconductor particles through its singlet excited state [22]. Since the oxidation potential of the PHNS*(S1) is around 1.25 V versus NHE and is more negative than that of titanium dioxide conduction band (0.5 V versus NHE) [22], the charge injection from the excited dye into the conduction band of the semiconductor is thermodynamically favored. The possible mechanism is summarised as follows: hm
PHNS ! 1 PHNS
a
1
b
200
187
250
300
350
400
450
500
550
600
Wavelength, nm
Fig. 1. Diffuse reflectance spectrum of TiO2: (a) on the external surface of the zeolite-Y (dotted line) from top to bottom 1%, 2%, 3%, 4%, 5% TiO2 and (b) encapsulated in zeolite-Y (solid line) 0.22%, 0.68%, 0.83% TiO2.
PHNS þ TiO2 ! TiO2 ðeÞ þ PHNSþ
The fluorescence intensity of PHNS* is found to be quenched with increasing loading of the titanium dioxide and is linear up to the addition of 2–3% TiO2 in the case of TY(s) samples as shown in Fig. 2a. Since at this titanium dioxide concentration complete monolayer coverage of the zeolite [23] may occur, and further increase in the concentration of TiO2 has little effect. Meanwhile with the TY(e) samples, fluorescence quenching process is found to be linear with TiO2 concentration and about 80% of the fluorescence is
188
S. Easwaramoorthi, P. Natarajan / Microporous and Mesoporous Materials 86 (2005) 185–190 D
D
D
TiO 2
TiO 2 TiO 2
D TiO 2
D
D
TiO 2
D
D D
TiO 2
D TiO
D
2
D
D D
a
D
D
TiO 2
D
D
b
D
TiO 2
D TiO 2
TiO 2
TiO 2
D
TiO 2
D D
TiO 2
D TiO 2
D
D
D
TiO 2
TiO 2
D TiO 2
D
D TiO 2
D
D
TiO 2
D TiO 2 D
D
TiO 2
TiO 2
TiO 2
D
D
D
D
c
d
D
Scheme 1. Arrangement of dye and semiconductor in zeolite-Y: (a) dye (D) on the zeolite surface; (b) dye on TY(s) samples at lower concentration of TiO2; (c) dye on TY(s) samples at higher concentration of TiO2; (d) dye on TY(e) samples, TiO2 (d).
Table 1 Photophysical properties of PHNS in various solvents Solvent Water Acetonitrile Methanol Methanol:water (1:4) Methanol:water (2:3) Methanol:water (3:2) a b
kmax, absa 520 517 520 524 526 526
kmax, emsa 595 560 572 584 576 574
Fluorescence lifetimeb, ns 0.93 4.1 2.12 1.08 1.43 1.73
kabs and ems = ±2 nm. Lifetime ±0.04 ns.
quenched at 0.68% titanium dioxide concentration due to its high dispersivity in the zeolite cages (Fig. 2b). Fluorescence lifetime measurement of PHNS in titanium dioxide loaded zeolites elucidates the mode of quenching involved in these systems and the data are given in Table 2. The lifetimes and the relative percentage of the emitting species remain constant within the experimental error for the titanium dioxide anchored on the external surface of the zeolite as shown in Fig. 3. It may be noted that the emission intensity is quenched proportional to the amount of titanium dioxide present on the surface. The observed emission is from the dye molecules that are not in contact with the TiO2, while the excited dye molecules, which are in contact with TiO2, are quenched. The mechanism of quenching is essentially static in nature due to the restricted movement
Table 2 Fluorescence lifetime of PHNS in zeolite/silica with titanium dioxide present at the external surface and encapsulated in supercages Host
kmax, absa
kmax, emsa
Fluorescence lifetimeb, ns s1
A1
s2
A2
savg
NaY TY(s) (1%)c TY(s) (2%) TY(s) (3%) TY(s) (4%) TY(s) (5%) TY(e) (0.22%) TY(e) (0.68%) TY(e) (0.83%) Silica Silica–TiO2 (5%)
522 512 512 512 512 512 522 522 522 520 512
574 574 574 574 574 574 574 574 574 575 575
0.48 0.35 0.41 0.46 0.43 0.36 0.34 0.26 0.23 0.85 0.84
25.22 30.86 23.67 23.71 32.79 33.22 29.71 53.83 55.23 31.91 25.46
1.42 1.35 1.60 1.50 1.40 1.34 1.28 1.10 1.01 2.15 2.20
74.74 69.14 76.33 76.29 67.21 66.78 71.29 46.17 44.77 68.09 74.54
1.17 1.04 1.31 1.24 1.08 1.01 1.04 0.65 0.58 1.74 1.85
a
kabs and ems = ±2 nm. Lifetime ±0.04 ns, amplitude ±5%. c Percentage of titanium dioxide, A1 and A2 are relative amplitudes in percentage. b
of the sensitizer and the semiconductor. The observed result thus confirms the mechanism of charge injection process when the excited dye is in contact with the 2þ semiconductor in zeolite as is observed for RuðbpyÞ3 ion and titanium dioxide systems in zeolite-Y reported earlier [13,14]. Identical trends were observed for PHNS adsorbed in titanium dioxide loaded silica surface (Table 2); the fluorescence intensity decreases with increase in titanium dioxide concentration. However, no
S. Easwaramoorthi, P. Natarajan / Microporous and Mesoporous Materials 86 (2005) 185–190 3.0e+6
90
a
a 75 Relative amplitude
2.5e+6
Intensity
2.0e+6 1.5e+6 1.0e+6
60 45 30 15
0.5e+5 0.0 540
189
0 560
580
600 620 640 Wavelength, nm
660
680
700
0
1
2
3 % of TiO2
4
5
85
3.0e+6
b
b
70 Relative amplitude
2.5e+6 2.0e+6 Intensity
6
1.5e+6 1.0e+6
55
40
25 0.5e+5
10 0.0 540
0 560
580
600
620
640
660
680
0.2
0.4
700
1
Fig. 3. Plot of percentage of titanium dioxide versus relative amplitude: (a) TY(s) and (b) TY(e).
1.3 1.2 1.1 Lifetime, ns
change in fluorescence lifetime is observed with increase in titanium dioxide loading and this observation confirms the static quenching process. In the case of TiO2 encapsulated in the supercages, fluorescence lifetime of PHNS is found to decrease (Fig. 4) with increase in TiO2 concentration with change in the relative amplitude of the emitting species (Table 2, Fig. 3b). The decrease in amplitudes of the longer lifetime component is suggested to be responsible for the quenching of the fluorescence intensity. It is worth noting that the quenching of the fluorescence lifetime of PHNS occurs in the zeolite framework by the encapsulated titanium dioxide particles, although there is no direct contact between semiconductor in the cage and the sensitizer at the external surface. Increase in the titanium dioxide concentration filling up more number of supercages by the TiO2 nanoparticles leads to closer approach towards the dye molecule on the external surface and accelerates the charge transfer process which results in further decrease in lifetime. It is known that
0.8
% of TiO2
Wavelength, nm
Fig. 2. Fluorescence spectra of PHNS in zeolite-Y with increase in concentration of TiO2: (a) TY(s), from top 0%, 1%, 2%, 3%, 4% and 5% TiO2, respectively, (b) TY(e), from top 0%, 0.22%, 0.68% and 0.83% TiO2, respectively.
0.6
1.0 0.9 0.8 0.7 0.6 0.5 0.0
0.2
0.4
0.6
0.8
1.0
% of TiO2
Fig. 4. Plot of average lifetime of PHNS versus percentage of TiO2 for TY(e) samples.
capping the titanium dioxide with polymer [17] suppresses the charge injection process to certain extent from PHNS* to the conduction band of titanium dioxide. However, when titanium dioxide nanocrystals are
190
S. Easwaramoorthi, P. Natarajan / Microporous and Mesoporous Materials 86 (2005) 185–190
coated with insulating silica layer [18], the fluorescence quantum yield and lifetime of rhodamine6G increases gradually with increase in thickness of the silica layer by suppressing the electron transfer and the complete coverage of the titania nanocrystal causes total prevention of the electron transfer. The present investigation suggests that while covering the semiconductor nanoparticles in the supercages of zeolites, sensitization is facilitated by the interaction between the semiconductor in the supercage and the dye present in the external surface presumably driven by the dynamics of hydration, porous structure of the zeolite and through the silicate network. The increase in amplitude of the shorter component of PHNS in the TY(e) samples as the concentration of TiO2 is increased indicates that the TiO2 encapsulated into the supercage acts as an electron donor synergistically through the zeolite frameworks. However, the amplitude of the shorter component is not changed when the TiO2 concentration increased on the surface which may be due to the electron transfer caused from the direct interaction between the dye and TiO2 without involvement of the zeolite framework. Further investigations are in progress with other mesoporous silicate materials.
4. Conclusion The photophysical properties of the dye elucidate the role of the internal and external surface of the zeolite in the sensitization of titanium dioxide by the PHNS. Decrease in fluorescence intensity of the dye in presence of titanium dioxide shows that the sensitization of the semiconductor occurs through the singlet excited state of the dye. The operative mechanism for the quenching process in the external surface of the zeolite is static in nature and the interfacial charge transfer is very fast due to static quenching. Decrease in fluorescence lifetime of the dye in titanium dioxide encapsulated systems is suggested to be due to the interaction of the excited state with the semiconductor facilitated by the zeolite network. The increase in the relative amplitude of the shorter component may be due to the TiO2 encapsulated zeolite which acts as an electron donor for PHNS* synergistically. Acknowledgments The investigations reported here are supported by Department of Science and Technology, India. Fellow-
ship received by S.E. from Council of Scientific and Industrial Research, India is gratefully acknowledged.
Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/ j.micromeso.2005.07.009.
References [1] V. Ramamurthy, in: V. Ramamurthy (Ed.), Photochemistry in Organized and Constrained Media, VCH, New York, 1991. [2] S. Hashimoto, J. Photochem. Photobiol. C: Photochem. Rev. 4 (2003) 19. [3] G. Schulz-Ekloff, D. Wohrle, B.V. Duffel, R.A. Schoonheydt, Micropor. Mesopor. Mater. 51 (2002) 91. [4] P.K. Dutta, M. Ledney, Prog. Inorg. Chem. 44 (1997) 209. [5] H. Maas, G. Calzaferri, Spectrum 16 (2003) 18. [6] K.B. Yoon, Chem. Rev. 93 (1993) 321. [7] A.L. Linsebigler, G. Lu, J.T. Yates Jr., Chem. Rev. 95 (1995) 735. [8] A. Hagfeldt, M. Gratzel, Chem. Rev. 95 (1995) 49. [9] M. Matsuoka, M. Anpo, J. Photochem. Photobiol. C: Photochem. Rev. 3 (2003) 225. [10] S. Corrent, G. Cosa, J.C. Scaiano, M.S. Galletero, M. Alvaro, H. Garcia, Chem. Mater. 13 (2001) 715. [11] G. Cosa, M.S. Galletero, L. Fernandez, F. Marquez, H. Garcia, J.C. Scaiano, New J. Chem. 26 (2002) 1448. [12] A. Corma, H. Garcia, Chem. Commun. (2004) 1443. [13] S.H. Bossmann, C. Turro, C. Schnabel, M.R. Pokhrel, L.M. Payawan Jr., B. Baumeister, M. Worner, J. Phys. Chem. B 105 (2001) 5374. [14] G. Cosa, M.N. Chretien, M.S. Galletero, V. Fornes, H. Garcia, J.C. Scaiano, J. Phys. Chem. B 106 (2002) 2460. [15] S.H. Bossmann, S. Jockusch, P. Schwarz, B. Baumeister, S. Gob, C. Schnabel, L. Payawan Jr., M.R. Pokhrel, M. Worner, A.M. Braun, N.J. Turro, Photochem. Photobiol. Sci. 2 (2003) 477. [16] Y. Kim II, S.W. Keller, J.S. Krueger, E.H. Yonemoto, G.B. Saupe, T.E. Mallouk, J. Phys. Chem. B 101 (1997) 2491. [17] L. Ziolkowski, K. Vinodgopal, P.V. Kamat, Langmuir 13 (1997) 3124. [18] L. Bechger, A.F. Koenderink, W.L. Vos, Langmuir 18 (2002) 2444. [19] S. Uppili, K.J. Thomas, E.M. Crompton, V. Ramamurthy, Langmuir 16 (2000) 265. [20] K.R. Gopidas, P.V. Kamat, J. Photochem. Photobiol. A: Chem. 48 (1989) 291. [21] P. Natarajan, J. Macromol. Sci. Chem. 81 (1988) 1285. [22] P.V. Kamat, K.R. Gopidas, D. Weir, Chem. Phys. Lett. 149 (1988) 491. [23] Y. Xu, C.H. Langford, J. Phys. Chem. 99 (1995) 11501. [24] Z. Marcezenko, Spectrophotometric Determination of Elements, Wiley & Sons, New York, 1976. [25] P.V. Kamat, Chem. Rev. 93 (1993) 267. [26] M. Yoon, S.Y. Ryu, Res. Chem. Intermed. 30 (2004) 207.