List of Publications - INFLIBNETshodhganga.inflibnet.ac.in/bitstream/10603/4804/14/14... · 2015. 12. 4. · List of Publications 11. A. Kathiravan, S. Anandan and R. Renganathan,

List of Publications

1. A. Kathiravan, G. Paramaguru and R. Renganathan, “Studies on the binding

of colloidal ZnO nanoparticles with bovine serum albumin”, J. Mol. Struc. (2009) In press.

2. M. Asha Jhonsi, A. Kathiravan and R. Renganathan, “Spectroscopic studies on interaction of colloidal capped CdS with Bovine serum albumin”, Colloids and Surfaces B 72 (2009) 167-172.

3. A. Kathiravan and R. Renganathan, Photosensitization of colloidal TiO2 nanoparticles with phycocyanin pigment, J. Colloid and Interface Sci. 335 (2009) 196-202.

4. M. Asha Jhonsi, A. Kathiravan and R. Renganathan, “An Investigation on Fluorescence Quenching of Certain Porphyrins by Colloidal CdS”, J. Lumi. 129 (2009) 854-860.

5. A. Kathiravan and R. Renganathan, “Photoinduced interactions between colloidal TiO2 nanoparticles and Calf thymus-DNA”, Polyhedron 28 (2009) 1374-1378.

6. M. Asha Jhonsi, A. Kathiravan and R. Renganathan, “Photoinduced interaction between Xanthene dyes and colloidal CdS”, J. Mol. Struc. 921 (2009) 279-284.

7. A. Kathiravan and R. Renganathan, “Effect of anchoring group on the

Photosensitization of colloidal TiO2 with Porphyrins”, J. Colloid and Interface Sci. 331 (2009) 401-407.

8. A. Kathiravan, M. Chandramohan R. Renganathan and S. Sekar, “Spectroscopic studies on the interaction between phycocyanin and bovine serum albumin”, J. Mol. Struc. 919 (2009) 210-214.

9. A. Kathiravan, V. Anbazhagan, M. Asha Jhonsi and R. Renganathan, “Excited Singlet State Reactions of meso-tetrakis(p-sulfonatophenyl) porphyrin (TSPP) with pyrimidines: A Steady State and Time-Resolved fluorescence quenching study”, J. Mol. Struc. 919 (2009) 79-82.

10. A. Kathiravan, M. Chandramohan R. Renganathan and S. Sekar,

“Photoinduced Electron Transfer from Phycoerythrin to Colloidal Metal Semiconductor Nanoparticles”, Spectrochim. Acta A 72 (2009) 496-501.


11. A. Kathiravan, S. Anandan and R. Renganathan, “Interaction of Colloidal AgTiO2 nanoparticles with Bovine Serum Albumin”, Polyhedron 28 (2009) 157-161.

12. A. Kathiravan, M. Chandramohan R. Renganathan and S. Sekar,

“Cyanobacterial Chlorophyll as a sensitizer for colloidal TiO2”, Spectrochim. Acta A 71 (2009) 1783-1787.

13. A. Kathiravan, S. Anandan and R. Renganathan, “Interaction of Colloidal TiO2 with Human Serum Albumin: A Fluorescence Quenching Study”, Colloids and Surfaces A 333 (2009) 91-95.

14. A. Kathiravan, P. Sathish Kumar, R. Renganathan and S. Anandan,

“Photoinduced electron transfer reactions between meso-tetrakis (4-sulfonatophenyl) porphyrin and metal-semiconductor nanoparticles”, Colloids and Surfaces A 333 (2009) 175-181.

15. A. Kathiravan and R. Renganathan, “Interaction of Colloidal TiO2 with Bovine Serum Albumin: A Fluorescence Quenching Study”, Colloidal and Surfaces A: 324 (2008) 176-180.

16. A. Kathiravan, V. Anbazhagan, M. Asha Jhonsi and R. Renganathan, “Photosensitization of Colloidal TiO2 with ZnTPP and Pyrene”, Z. Phys. Chem. 222 (2008) 647 - 654.

17. A. Kathiravan, V. Anbazhagan, M. Asha Jhonsi and R. Renganathan, “Fluorescence quenching of meso-tetrakis(4-sulfonatophenyl) porphyrin by colloidal TiO2”, Spectrochim. Acta: A 70 (2008) 615-618.

18. A. Kathiravan and R. Renganathan, “Fluorescence quenching of meso-

tetrakis (4-sulfonatophenyl) porphyrin by certain organic dyes”, Z. Phys. Chem. 222 (2008) 987-995.

19. A. Kathiravan, V. Anbazhagan, M. Asha Jhonsi and R. Renganathan, “A Study on the Fluorescence Quenching of Eosin by certain Organic Dyes”, Z. Phys. Chem. 222 (2008) 1013-1021.

20. A. Kathiravan and R. Renganathan, “Photoinduced interaction between Riboflavin and TiO2 colloid”, Spectrochim. Acta: A 71 (2008) 1080-1083.

21. A. Kathiravan and R. Renganathan, “An Investigation on Electron Transfer Quenching of Zinc(II) meso-tetraphenylporphyrin (ZnTPP) by colloidal TiO2”, Spectrochim. Acta: A 71 (2008) 1106-1109.


22. A. Kathiravan, M. Asha Jhonsi, J. Thiruvengadam and R. Renganathan, “Photoinduced Electron Transfer between Triphenylpyrylium Ion (TPP+) and Certain Phenols”, Z. Phys. Chem. 222 (2008) 1591-1599.

23. V. Anbazhagan, V. Kandavelu, A. Kathiravan and R. Renganathan,

“Investigation on the Fluorescence Quenching of 2,3-Diazabicyclo[2.2.2] oct-2-ene {DBO} by certain Estrogens and Catechols”, J. Photochem. Photobiol. A: Chem. 193 (2008) 204-212.

24. M. Asha Jhonsi, A. Kathiravan and R. Renganathan, “Interaction between certain porphyrins and CdS colloid: A steady state fluorescence quenching study”, Spectrochim. Acta: A 71 (2008) 1507-1511.

25. A. Kathiravan, V. Anbazhagan, M. Asha Jhonsi and R. Renganathan,

“Fluorescence quenching of Xanthene dyes by TiO2”, Z. Phys. Chem. 221 (2007) 941-948.

26. V. Anbazhagan, A. Kathiravan, M. Asha Jhonsi and R. Renganathan, “Fluorescence Quenching Study on Electron Transfer from Certain Amines to Excited state Triphenylpyrylium Ion (TPP+)”, Z. Phys. Chem. 221 (2007) 929-939.

Journal of Molecular Structure xxx (2009) xxx–xxx

ARTICLE IN PRESS

Contents lists available at ScienceDirect

Journal of Molecular Structure

journal homepage: www.elsevier .com/ locate /molst ruc

Study on the binding of colloidal zinc oxide nanoparticles with bovineserum albumin

A. Kathiravan, G. Paramaguru, R. Renganathan *

School of Chemistry, Bharathidasan University, Tiruchirappalli 620 024, Tamil Nadu, India

a r t i c l e i n f o a b s t r a c t

Article history:Received 10 January 2009Received in revised form 17 June 2009Accepted 23 June 2009Available online xxxx

Keywords:Colloidal ZnO nanoparticlesFluorescence quenchingBSA

0022-2860/$ - see front matter � 2009 Elsevier B.V. Adoi:10.1016/j.molstruc.2009.06.032

* Corresponding author. Tel.: +91 431 2407053; faxE-mail address: [email protected] (R. Renganath

Please cite this article in press as: A. Kathiravan

The interaction between colloidal zinc oxide (ZnO) nanoparticles and bovine serum albumin (BSA) wasstudied by using absorption, fluorescence, Fourier transform infrared, synchronous and time resolvedfluorescence spectroscopic measurements. The apparent association constant has been deduced(Kapp = 1.1 � 104 M�1) from the absorption spectral changes of BSA–colloidal ZnO nanoparticles usingBenesi–Hildebrand equation. Addition of colloidal ZnO nanoparticles effectively quenched the intrinsicfluorescence of BSA. The number of binding sites (n = 1.06) and apparent binding constant(K = 2.5 � 104 M�1) were calculated by relevant fluorescence data. Based on Forster’s non-radiationenergy transfer theory, distance between the donor (BSA) and acceptor (ZnO) (r0 = 2.88 nm) as well asthe critical energy transfer distance (R0 = 2.49 nm) has also been calculated. The interaction between col-loidal ZnO and BSA occurs through static quenching mechanism. The effect of colloidal ZnO nanoparticleson the conformation of BSA has been analyzed by means of UV–visible absorption spectra and synchro-nous fluorescence spectra.

� 2009 Elsevier B.V. All rights reserved.

1. Introduction

Over the past decades colloidal semiconductor nanoparticleshave attracted considerable attention owing to their unique size-dependent optical and electronic properties [1,2]. Lately, a newdirection has emerged in potential biotechnological applicationssuch as luminescence tagging, immunoassay, drug delivery, andcellular imaging [3–6].

As a fluorescent semiconductor material, ZnO nanocrystals ex-hibit much higher chemical stability and safety relative to othertoxic semiconductor nanocrystals. Thus, fluorescent ZnO nanocrys-tals have become a biofriendly candidate for biological technologyapplications [7,8]. However, it is well known that the ZnO is ratherunstable in solution. Under illumination, the photogenerated holescan thermodynamically oxidize the semiconductor because theZnO decomposition potential is located inside the semiconductorband gap [9].

Serum albumins are the most abundant proteins in plasma [10].As the major soluble protein constituents of the circulatory system,they have many physiological functions [11]. Among them, BSA hasa wide range of functions involving the binding, transport, anddelivery of fatty acids, porphyrins, bilirubin, steroids, etc. It is hometo specific binding sites for metals, pharmaceuticals and dyes [12].BSA has two tryptophan moieties at positions 134 and 212 as well

ll rights reserved.

: +91 431 2407045.an).

et al., J. Mol. Struct. (2009), d

as tyrosine and phenylalanine [13], the protein intrinsic fluores-cence is due to aromatic amino-acid residues.

The binding properties of BSA and drugs were investigated bymany researchers [14–18]. Nanoparticle probes acting as biosen-sors in chemical and biochemical field have been researchedrecently and their applications are becoming more extensive. Threetypes of nanoparticles in biochemical analysis are used: metalnanoparticles [19], silica nanoparticles [20,21] and luminescencequantum dots [22–24].

Critical literature survey reveals that attempts have not beenmade so far to investigate the interaction between colloidal ZnOnanoparticles and BSA by fluorescence technique in our knowl-edge. Recently we have reported the interaction of colloidal tita-nium dioxide with serum albumins [25–27]. In the present work,the fluorescence quenching technique is applied to study the inter-action between colloidal ZnO and BSA in an inoculation to charac-terize the chemical associations taking place. According to the plotof log[(F0 � F)/F] versus log[Q], the binding constant and the num-bers of binding sites were determined. The effect of colloidal ZnOnanoparticles on the conformation of BSA has been analyzed.

2. Materials and methods

2.1. Materials

Zinc acetate dihydrate (98%) was purchased from Aldrich.Lithium hydroxide monohydrate (98%, AG) powder was purchased

oi:10.1016/j.molstruc.2009.06.032

http://dx.doi.org/10.1016/j.molstruc.2009.06.032

mailto:[email protected]

http://www.sciencedirect.com/science/journal/00222860

http://www.elsevier.com/locate/molstruc


2 A. Kathiravan et al. / Journal of Molecular Structure xxx (2009) xxx–xxx

ARTICLE IN PRESS

from Fluka. Bovine Serum Albumin (96% fatty acid free, Product No.A8806, Sigma) was dissolved in double distilled water to preparestock solution (1 � 10�4 M) which was then stored at 0–4 �C. Allmeasurements were performed at room temperature.

2.2. Preparation of colloidal ZnO nanoparticles

Colloidal ZnO nanoparticles were prepared by the method re-ported by Spanhel and Anderson [28]. A solution of 0.1 M Zinc ace-tate dihydrate (2.19 g in 100 ml water) was prepared and refluxedin a round bottom flask for 2 h and this was followed by a gradualcooling to room temperature. This precursor solution (0.1 M) is sta-ble for several weeks. When required, small batches of ZnO colloi-dal suspension were prepared by hydrolysis of the precursor usinglithium hydroxide monohydrate powder. This method was favor-able for maintaining reproducibility of sample preparation andobtaining fresh colloids for experiments. In a typical procedure, a2.5 ml solution of the precursor was diluted to 25 ml with waterand 11 mg of lithium hydroxide is added then hydrolyzing in anultrasonic bath at 273 K (0 �C) for 15 min. The ZnO suspension(1 � 10�2 M) prepared in this manner was transparent and stablewhen kept in a sealed flask. The prepared colloidal ZnO nanoparti-cles have been characterized by using Scanning electron micros-copy (Fig. 1). From the SEM picture we observed that themonodisperse nature of ZnO nanoparticles.

2.3. Determination of particle size of colloidal ZnO nanoparticles

The particle size of the prepared colloidal ZnO nanoparticles hasbeen determined from the relationship between band gap shift(DEg) and the radius (R) of quantum size particles using followingEq. (1),

DEg ¼p2h2

2lR2 �1:8e2

eRþ Polarization terms ð1Þ

where, h is Planck’s constant, R is radius of the particle, e is the rel-ative permittivity of semiconductor and DEg is the band gap shift.The calculated band gap shift for colloidal ZnO is 0.66 eV (bandgap energy of the colloidal ZnO was measured, Eg = 3.96 eV corre-sponding to k = 313 nm, Fig. 10). l is the reduced mass of exciton,i.e., reduced effective mass of the electron and hole (1/l = l/mc + l/mh) in the semiconductor, e is electron charge and c is dielectricconstant of the semiconductor. The value of l = 0.156me (me isthe electron rest mass) is calculated from mc = 0.24me, mh = 0.45me

[29]. Since the optical dielectric constant of bulk semiconductor is

Fig. 1. SEM picture of colloidal ZnO nanoparticles.

Please cite this article in press as: A. Kathiravan et al., J. Mol. Struct. (2009), d

very large (e = 170), coloumbic and polarisation terms in Eq. (1)are neglected. The calculated size of the prepared colloidal ZnO is2.49 nm.

2.4. XRD measurements

X-ray powder diffraction patterns were recorded in a BrukerAXS B8 Discover model using Cu Ka radiation (k = 0.154 nm) anda graphite monochromator in the diffracted beam. ZnO samplewas in the form of powder. A scan rate of 0.05� min�1 was appliedto record the XRD pattern in the 2h range of 2h = 20–80�.

2.5. Steady-state measurements

The fluorescence quenching measurements were carried outwith JASCO FP-6500 spectrofluorimeter. The excitation wave-length of BSA was 280 nm and the emission was monitored at345 nm. The excitation and emission slit widths (each 5 nm)and scan rate (500 nm/min) were maintained constant for allthe experiments. Samples were carefully purged using pure nitro-gen gas for 15 min. Quartz cells (4 � 1 � 1 cm) with high vacuumTeflon stopcocks were used for purging. Absorption spectral mea-surements were recorded using Cary 300 UV–visiblespectrophotometer.

2.6. Time resolved fluorescence measurements

Fluorescence lifetime measurements were carried out in apicosecond time correlated single photon counting (TCSPC) spec-trometer. The excitation source is the tunable Ti-sapphire laser(Tsunami, Spectra Physics, USA). The fluorescence decay was ana-lyzed by using the software provided by IBH (DAS-6).

2.7. UV–visible and fluorescence studies

A 3 ml solution, containing appropriate concentration of BSA(1 � 10�6 M), was titrated by successive additions of a 30 ll stocksolution of colloidal ZnO nanoparticles (1 � 10�2 M). Titrationswere manually done by using micro pipette for the addition of col-loidal ZnO. UV–visible spectra of all the solutions were recorded inthe range of 200–800 nm.

Fluorescence spectra were then measured by using 4 cm cuv-ette (excitation and emission wavelength of BSA is 280 and345 nm respectively) at ambient temperature.

For synchronous fluorescence spectra also the same concentra-tion of BSA and colloidal ZnO were used and the spectra were mea-sured at two different Dk (difference between the excitation andemission wavelengths of BSA) values such as 15 and 60 nm.

2.8. Fourier transform infrared spectroscopy (FT-IR) measurements

Fourier transform infrared spectra were obtained using a Per-kin-Elmer Spectrum RXI FT-IR Spectrometer at room temperaturein the range of 4000–400 cm�1. The samples were placed in aliquid cell between two windows (CaF2). Mirror velocity is0.3 cm/s and number of co-added scans are 4 then total collectiontime is less than 2 min.

3. Results and discussion

3.1. XRD characterization of ZnO nanoparticles

Fig. 2 shows the XRD pattern of powdered ZnO nanoparticlesobtained by rotary evaporation of colloidal ZnO at 30 �C. The XRDpeaks are found to be broad indicating fine size of sample grains.

oi:10.1016/j.molstruc.2009.06.032


Fig. 2. X-ray diffraction (XRD) spectrum of ZnO nanoparticles.

A. Kathiravan et al. / Journal of Molecular Structure xxx (2009) xxx–xxx 3

ARTICLE IN PRESS

From the diffraction pattern, we found that the planes (1 0 0),(1 0 1), (1 0 2) and (1 0 3) are corresponds to wurtzite structure.

3.2. Absorption characteristics of BSA–ZnO

Fig. 3 shows the absorption spectrum of BSA in water and inwater containing colloidal ZnO nanoparticles at different concen-trations. In the presence of colloidal ZnO nanoparticles the absor-bance of BSA is increased markedly, without change in thelocation of the peak (280 nm). This inference is due to, while add-ing colloidal ZnO nanoparticles to the solution of BSA some of theBSA molecules gets adsorbed on the surface of colloidal ZnOnanoparticles and involved in the formation ground state complexof the type BSA. . .ZnO. The newly formed complex also havingabsorption at 280 nm. This is the reason for increase in absor-bance of BSA with the addition of colloidal ZnO as supported by

0

0.17

0.34

0.51

240 290 340

Wav

Abs

orba

nce

0

6

Fig. 3. Absorption spectrum of BSA (1 � 10�6 M) in the presence of colloidal ZnO in the c(Aobs � A0) on the reciprocal concentration of colloidal ZnO.


similar observations made earlier [30]. The main change that wecan observe in the spectra of the complex was not due to theexperimental error, especially in the range of 230–300 nm. Theseresults indicated that there is interaction between colloidal ZnOnanoparticles and BSA is existed through ground state complexformation.

The equilibrium for the formation of complex between BSA andcolloidal ZnO nanoparticles is given by Eq. 2, where Kapp representsthe apparent association constant:

BSAþ nZnO �Kapp

BSA . . . nZnO ð2Þ

Here n = 1,

Kapp ¼½BSA . . . ZnO�½BSA� � ½ZnO�

390 440 490

elength

0

20

40

60

0 0.2 0.4 0.6 0.8 1

1/[ZnO] x 10-5 M

1/A

-A0

oncentration range of 0–6 � 10�5 M. The inset is the straight line dependence of 1/

oi:10.1016/j.molstruc.2009.06.032



ARTICLE IN PRESS

Changes in intensity of the absorption peak at 280 nm as a re-sult of the formation of surface complex, Kapp is obtained accordingto the method reported by Benesi and Hildebrand [31]:

Aobs ¼ ð1� aÞC0eBSAlþ aC0ecl ð3Þ

where Aobs is the observed absorbance of the solution containingdifferent concentrations of the colloidal ZnO nanoparticles at280 nm, a is the degree of association between BSA and ZnO, eBSA

and ec are the molar extinction coefficients at the defined wave-length (k = 280 nm) for BSA and the formed complex, respectively,C0 is the initial concentration of BSA and ‘l’ is the optical path length.Eq. (3) can be expressed as Eq. (4), where A0 and Ac are the absor-bances of BSA and the complex at 280 nm, respectively, with theconcentration of C0:

Aobs ¼ ð1� aÞA0 þ aAc ð4Þ

At relatively high ZnO concentrations, a can be equated to(Kapp[ZnO])/(1 + Kapp[ZnO]). In this case, Eq. (4) can be changed asEq. (5):

1Aobs � A0

¼ 1Ac � A0

þ 1KappðAc � A0Þ½ZnO� ð5Þ

Based on the enhancement of absorbance at 280 nm due toabsorption of surface complex, a linear relationship between 1/(Aobs � A0) and the reciprocal concentration of colloidal ZnO with

0

150

300

450

600

750

290 320 350 380Wave

Inte

nsit

y

0

6

Fig. 4. Fluorescence quenching of BSA (1 � 10�6 M; kexi = 280 nm; kemi = 345 nm) in th6 � 10�5 M.

0.8

1

1.2

1.4

1.6

1.8

2

0 1 2

[ZnO

I0/I

Fig. 5. Stern–Volmer plot for the steady state fluorescence quenching of


a slope equal to 1/Kapp(Ac � A0) and an intercept equal to 1/(Ac � A0) was obtained (Fig. 3, inset). The value of apparent associ-ation constant (Kapp) determined from this plot is 1.1 � 104 M�1

(R2 = 0.9975).

3.3. Fluorescence quenching of BSA by colloidal ZnO nanoparticles

Fig. 4 shows the effect of increasing concentration of colloidalZnO nanoparticles on the fluorescence emission spectrum of BSA.Addition of ZnO colloid to the solution of BSA resulted in thequenching of its fluorescence emission. The fluorescence quench-ing is described by Stern–Volmer relation:

I0=I ¼ 1þ KSV½Q � ¼ 1þ kqs0½Q � ð6Þ

where, I0 and I are the emission intensities of BSA in the absenceand presence of colloidal ZnO, KSV is the Stern–Volmer constant re-lated to the bimolecular quenching rate constant, kq, by KSV = kq�s0,and s0 is the average lifetime of BSA in the excited state which is10�8 s [32], [Q] is concentration of the quencher. According to Eq.(6) we get linear plot [Fig. 5] for I0/I against [ZnO], from the slopewe calculated the quenching rate constant (kq) as 1.38 �1012 M�1 s�1. The possible scattering due to colloidal nature ofZnO has been omitted because of the fact that baseline correctionwas done for all the spectral measurements and also even after add-ing the highest concentration of colloidal ZnO nanoparticles the

410 440 470 500length

e presence of various concentration of colloidal ZnO, [ZnO] = 0, 1, 2, 3, 4, 5, and

3 4 5 6

] x 10-5 M

BSA (1 � 10�6 M) by colloidal ZnO (0, 1, 2, 3, 4, 5, and 6 � 10�5 M).

oi:10.1016/j.molstruc.2009.06.032



ARTICLE IN PRESS

solution remains clear. So we confirmed that the fluorescencequenching is only because of the interaction between BSA and col-loidal ZnO not due to scattering.

3.4. Fluorescence lifetime measurements

Fluorescence lifetime measurement is useful for understandingthe interaction between donor and acceptor systems. The observa-tion of linear Stern–Volmer plot in steady state measurement alonedoes not prove the dynamic nature of quenching. Sometimes staticquenching also results in linear Stern–Volmer plot. In general, timeresolved measurement is the most definitive method for differen-tiating static and dynamic quenching [33]. Fig. 6 shows the fluores-cence decay curve of BSA in the absence and presence of colloidalZnO nanoparticles. BSA exhibits single exponential decay not onlyin dilute solutions but also in the presence of colloidal ZnO nano-particles. While increasing the concentration of colloidal ZnO thereis no change in the fluorescence lifetime of BSA. This observationshows that quenching follows static mechanism. It also supportsthe adsorption of BSA on the surface of colloidal ZnO and the for-mation of ground state surface complex. In Fig. 6 though the decaytraces of BSA in the absence and presence of colloidal ZnO nano-particles were actually plotted however the lifetime of BSA re-mains same in both conditions, hence the merging of the kinetictraces were observed (the plot looks like a single decay curve). Thisshows that the quenching of BSA by colloidal ZnO nanoparticles isstatic in nature. For static quenching, we can deduce the bindingconstant (K) because static quenching arises from the formationof complex between fluorophore and the quencher. Hence thebinding constant (K) was calculated by the method given in the fol-lowing section.

3.5. Binding constant and number of binding sites

In general, maximum collisional quenching constant (kq) of var-ious kinds of quenchers to biopolymers is 2.0 � 1010 M�1 s�1 [32].But for BSA–ZnO system higher quenching rate constant(1.38 � 1012 M�1 s�1) was observed. This shows that the quenchingof BSA by colloidal ZnO is not dynamic in nature. Therefore, itdepends on the formation of complex between colloidal ZnO andBSA. For static quenching, we can deduce the binding constant(K) because static quenching arises from the formation of complexbetween fluorophore and the quencher. Hence the binding con-stant (K) was calculated by following method.

If it is assumed that there are similar and independent bindingsites in the BSA, the relationship between fluorescence intensity

1

10

100

1000

10000

15 25Ti

Cou

nts

Fig. 6. Fluorescence decay profiles for BSA (1 � 10�6 M) in th


and the quencher medium can be deduced from the followingEq. (7):

nQ þ B! Q n . . . B ð7Þ

where B is the fluorophore, Q is the quencher, nQ + B is the postu-lated complex between a fluorophore and n molecules of thequencher. The constant K is given by

K ¼ Qn . . . B½Q �n½B�

ð8Þ

If the overall amount of biomolecules (bound or unbound withthe quencher) is B0, then [B0] = [Qn. . .B] + [B], here [B] is the concen-tration of unbound biomolecules, then the relationship betweenfluorescence intensity and the unbound biomolecule as [B]/[B0] = F/F0 that is:

logF0 � F

F

� �¼ log K þ n log½Q � ð9Þ

where K is the binding constant of colloidal ZnO with BSA, whichcan be determined from the intercept of log[(F0 � F)/F] versus log[Q]as shown in Fig. 7 and thus we obtained the calculated value ofbinding constant K as 2.5 � 104 M�1 and number of binding sites(n) as 1.06. The value of ‘‘n” equals 1, and thus indicates the exis-tence of just a single binding site in BSA for ZnO.

3.6. Characteristics of synchronous fluorescence spectra

Synchronous fluorescence spectroscopy has been applied to avariety of multi-component system. The main advantages of syn-chronous fluorescence spectra are simplified spectra, narrowedbandwidth, high selectivity and sensitivity. The excitation andemission monochromators are synchronously scanned, separatedby a constant wavelength interval (Dk). As it is well known that,synchronous fluorescence spectra can provide the information onthe molecular microenvironment, particularly in the vicinity ofthe fluorophore functional groups [34]. The fluorescence of BSAmay due to the presence of amino-acid residues such as tyrosine,tryptophan and phenylalanine. Hence spectroscopic methods areusually applied to study the conformation of serum protein. In syn-chronous fluorescence spectroscopy, according to Miller [35], thedifference between excitation wavelength and emission wave-length (Dk = kemi � kexc) reflects the spectra of a different natureof chromophores, with large Dk values such as 60 nm, the synchro-nous fluorescence of BSA is characteristic of tryptophan residueand small Dk values such as 15 nm is characteristic of tyrosine[36]. The synchronous fluorescence spectra of BSA with various

35 45 55me (ns)

e absence and presence of colloidal ZnO (0–6 � 10�5 M).

oi:10.1016/j.molstruc.2009.06.032


0

50

100

150

200

250 276 302 328 354 380

Excitation Wavelength

Flu

ores

cenc

e In

tens

ity

Fig. 8. Synchronous fluorescence spectrum of BSA (1 � 10�6 M) at Dk = 15 nm in the absence and presence of colloidal ZnO (0–6 � 10�5 M).

-1

-0.8

-0.6

-0.4

-0.2

0

-5.2-5-4.8-4.6-4.4-4.2-4log [Q]

log

F0-

F/F

Fig. 7. Plot of log[(F0 � F)/F] vs log [Q].


ARTICLE IN PRESS

amounts of colloidal ZnO were recorded at Dk = 15 nm (Fig. 8) andDk = 60 nm (spectra not shown here), respectively. The emissionwavelength of the tyrosine residues is blue-shifted with increasingconcentration of colloidal ZnO nanoparticles. At the same time, thetryptophan fluorescence emission is decreased regularly, but nosignificant change in wavelength was observed. It suggests thatthe interaction of colloidal ZnO nanoparticles with BSA does not af-fect the conformation of tryptophan micro-region. The tyrosinefluorescence spectrum may represent that the conformation ofBSA is somewhat changed, leading to the polarity around tyrosineresidues strengthened and the hydrophobicity weakened [37]. It is

+-H2O

tyrosine

NH2

OHO

HO

H2N

OHO

tryptophan

+

ZnIIHO

ZnIIHO

Scheme 1. Mechanism of interaction


important to note that colloidal ZnO nanoparticles affect only thetyrosine residues present in the BSA. This is because of tyrosinecontains one aromatic hydroxyl group unlike tryptophan.

In our previous study, riboflavin was used as a sensitizer for col-loidal TiO2 [38] and it was observed that riboflavin interactedthrough the hydroxyl group of colloidal TiO2. Hence it is clear thatthe presence of hydroxyl group in tyrosine residues is may respon-sible for the interaction of BSA with colloidal ZnO similar to 2,3-diazabicyclo[2.2.2]oct-2-ene (DBO) interact with aromatic hydro-xyl group of tyrosine in BSA [39] as reported by Anbazhagan andRenganathan [Scheme 1].

conformational changes due to -OH group

no conformational changes due toabsence of -OH group

NH2

OO

HO

ZnII

between colloidal ZnO and BSA.

oi:10.1016/j.molstruc.2009.06.032



ARTICLE IN PRESS

3.7. FT-IR characterization

To verify whether the conformation of BSA has changed or not,FT-IR measurements were performed in the absence and presenceof colloidal ZnO nanoparticles. Usually the amide I peak positionfor BSA occurs in the region of 1600–1700 cm�1 and amide II bandnear 1548 cm�1 [40,41]. Former one is the more sensitive tochange of protein conformation than later. In the present study,Fig. 9 shows the FT-IR spectra of BSA, a sharp peak near1640 cm�1 appears while a broad peak appears near 1550 cm�1.In this figure BSA in the absence of ZnO nanoparticles shows theamide I peak at 1640 cm�1 and in the presence of ZnO nanoparti-cles the amide I peak was shifted to 1645 cm�1. From the shift inpeak position we confirmed that the conformation of BSA has beenaffected by the addition of ZnO nanoparticles as supported by sim-ilar observations made earlier [42].

3.8. Energy transfer efficiency

The decrease in fluorescence intensity is attributed to electrontransfer or energy transfer between BSA and the colloidal ZnOnanoparticles. The excited state energy of BSA (3.5 eV, was calcu-lated from the emission wavelength of BSA by using the equationE = hm, where ‘E’ is excited state energy of BSA, ‘h’ is the Plank’s con-stant and ‘m’ is equal to c/k, in which ‘c’ is speed of light and ‘k’ is the

30

44

58

72

86

100

1000 1250 1

Wave nu

% T

BSA

BSA-ZnO

1640

Fig. 9. The FT-IR spectra of BSA in the absen

0.08

0.15

0.22

0.29

0.36

0.43

0.5

280 330Wavelen

Abs

orba

nce

BA

Fig. 10. Overlap spectrum of (A) absorption spectrum o


fluorescence maximum wavelength of BSA) is greater than theband gap energy of ZnO (3.3 eV, which is reported in [43]) andthe fluorescence emission of BSA is overlapped with the absorptionof colloidal ZnO (Fig. 10). Based on these two facts we confirmedthat energy transfer from the excited state of BSA to ZnO colloidsis possible.

Energy transfer efficiency (E) is given by the following Eq. [13],

E ¼ 1� I=I0 ð10Þ

where, I is the emission intensity of donor in the presence of accep-tor and I0 is the emission intensity of the donor alone. From theabove results it is clear that, in presence of ZnO, the fluorescenceintensity of BSA is reduced (from I0 to I) by energy transfer toZnO. Based on the above Eq. (10), a plot of ‘E’ obtained at differentconcentrations of ZnO is shown in Fig. 11. It is seen from this plot;‘E’ value steadily increases with increasing concentration of ZnO.This may be due to while increasing the concentration of ZnO, num-ber of BSA molecules gets adsorbed on the surface of colloidal ZnOis increased, so that the amount of energy transfer from BSA to ZnOhas also been increased.

3.9. Energy transfer between colloidal ZnO nanoparticles and BSA

According to Forster’s non-radiative energy transfer theory [33],the energy transfer efficiency is related not only to the distance be-

500 1750 2000

mber (cm-1)

1647

ce and presence of ZnO nanoparticles.

380 430gth

0

120

240

360

480

600

720

Inte

nsit

y

f colloidal ZnO and (B) emission spectrum of BSA.

oi:10.1016/j.molstruc.2009.06.032


0

0.1

0.2

0.3

0.4

0.5

0 1 2 3 4 5 6

[ZnO] x 10-5 M

E

Fig. 11. Energy transfer efficiency of BSA with various concentration of colloidal ZnO (1–5 � 10�5 M).


ARTICLE IN PRESS

tween the acceptor and donor (r0), but also to the critical energytransfer distance (R0), that is:

E ¼ R60

R60 þ r6

0

ð11Þ

where, R0 is the critical distance when the transfer efficiency is 50%.

R60 ¼ 8:8� 10�25 K2N�4UJ ð12Þ

where, K2 is the spatial orientation factor of the dipole, N the refrac-tive index of the medium, U the fluorescence quantum yield of thedonor, J the overlap integral of the fluorescence emission spectrumof the donor and the absorption spectrum of the acceptor. The valueof J can be calculated by using the Eq. (13),

J ¼Z

FðkÞeðkÞk4dkFðkÞdk

ð13Þ

where, F(k) is the fluorescence intensity of the donor, e(k) is molarabsorptivity of the acceptor. The parameter J (4.27 �10�11 M�1 cm3) can be evaluated by integrating the spectral param-eters in Eq. (13). Under these experimental conditions, we foundR0 = 2.49 nm from Eq. (12) using K2 = 2/3, U = 0.032, N = 1.3467(the values of K2, U and N were taken from the literature [44]).Obviously, the calculated value of R0 is in the range of maximal crit-ical distance (R0 = 5–10 nm) [34]. This is in accordance with theconditions of Forster’s non-radiative energy transfer theory [34],indicating the static quenching interaction between ZnO and BSA.Using Eq. (11) we found r0 = 2.88 nm, it can be seen that the valueof r0 < 8 nm scale which suggests that non-radiative energy transferoccurs between colloidal ZnO and BSA with high probability [45],while r0 is higher than R0 in the present study also reveals the pres-ence of static-type of quenching mechanism [46]. Furthermore, it issuggested that the binding of colloidal ZnO to BSA is occurs throughenergy transfer.

4. Summary

The interaction between colloidal ZnO nanoparticles and BSAhas been studied by UV–visible, steady state, time resolved andsynchronous fluorescence spectroscopic measurements. The re-sults presented clearly indicated that ZnO quenches the fluores-cence of BSA through complex formation. The quenching rateconstant, binding constant, and number of binding sites were cal-culated according to the relevant fluorescence data. From the syn-chronous fluorescence spectra, it is established that theconformational changes of BSA occurs due to the interaction withcolloidal ZnO nanoparticles. The binding study of colloidal ZnO


nanoparticles with BSA is of great importance in pharmacy, phar-macology, and biochemistry.

Acknowledgements

R.R. and A.K. thanks CSIR, Government of India (Ref: No.01(2217)/08/EMR-II, dt. 06/05/2008) for the project and fellow-ship, respectively.

We are thankful to Prof. P. Ramamoorthy, NCUFP, University ofMadras, Chennai for lifetime measurements.

R.R. thanks DAAD for sponsoring study and research visit andProf. Bahnemann for extending his lab facilities at University ofHannover, Germany.

References

[1] V.L. Colvin, M.C. Schlamp, A.P. Alivisators, Nature 370 (1994) 354.[2] M.V. Artemyev, U. Woggon, R. Wannemacher, H. Jaschinski, W. Langbein, Nano

Lett. 1 (2001) 309.[3] A.P. Alivisatos, Science 271 (1996) 933.[4] M. Bruchez, M. Moronne, P. Gin, S. Weiss, A.P. Alivisatos, Science 281 (1998)

2013.[5] X. Wu, H. Liu, J. Liu, K. Haley, J. Treadway, J. Larson, N. Ge, F. Peals, M. Bruchez,

Nat. Biotechnol. 21 (2003) 41.[6] X. Gao, Y. Cui, R.M. Levenson, L. Chung, S. Nie, Nat. Biotechnol. 22 (2004) 969.[7] R. Bravner, R. Ferrari-Iliou, N. Brivois, S. Djediat, M.F. Benedetti, F. Fievet, Nano

Lett. 6 (2006) 866.[8] X. Wang, X. Kong, Y. Yu, H. Zhang, J. Phys. Chem. C 111 (2007) 3836.[9] J. Domenech, A. Prieto, J. Phys. Chem. 90 (1986) 1123.

[10] D.D. Carter, J.X. Ho, Adv. Protein Chem. 45 (1994) 153.[11] R.E. Olson, D.D. Christ, Ann. Rep. Med. Chem. 31 (1996) 327.[12] S. Ashoka, J. Seetharamappa, P.B. Kandagal, S.M.T. Shaikh, J. Lumin. 121 (2006)

179.[13] L.A. Sklar, B.S. Hudson, R.D. Simoni, Biochemistry 16 (1977) 5100.[14] Y.Q. Wang, H.M. Zhang, G.C. Zhang, W.H. Tao, S.H. Tang, J. Lumin. 126 (2007)

211.[15] N. Zhou, Y.Z. Liang, P. Wang, J. Photochem. Photobiol. A Chem. 185 (2006) 271.[16] B. Zhou, Z. Qi, Q. Xiao, J.X. Dong, Y.Z. Zhang, Y. Liu, J. Biochem. Biophys.

Methods 70 (2007) 743.[17] C.X. Wang, F.F. Yan, Y.X. Zhang, L. Ye, J. Photochem. Photobiol. A Chem. 192

(2007) 23–28.[18] A. Sulkowska, J. Rownicka, B. Bojkoa, W. Sulkowski, J. Mol. Struct. 651–653

(2003) 133.[19] R. Elghanian, J.J. Storhoff, R.C. Mucic, R.L. Letsinger, C.A. Mirkin, Science 277

(1997) 1078.[20] S. Santra, P. Zhang, K. Wang, R. Tapec, W. Tan, Anal. Chem. 73 (2001) 4988.[21] M. Qhobosheane, S. Santra, P. Zhang, W. Tan, Analyst 126 (2001) 1274.[22] X.G. Peng, M.C. Schlamp, A.V. Kadavanich, A.P. Alivistos, J. Am. Chem. Soc. 119

(1997) 7019.[23] W.C.W. Chan, S. Nie, Science 281 (1998) 2016.[24] L. Tan, L.Y. Liu, Q.J. Xie, Y.Y. Zhang, S.Z. Yao, Anal. Sci. 20 (2004) 441.[25] A. Kathiravan, R. Renganathan, Colloids Surf. A Physicochem. Eng. Aspects 324

(2008) 176.[26] A. Kathiravan, S. Anandan, R. Renganathan, Eng. Aspects 333 (2009) 91.[27] A. Kathiravan, R. Renganathan, S. Anandan, Polyhedron 28 (2009) 157.[28] L. Spanhel, M.A. Anderson, J. Am. Chem. Soc. 113 (1991) 2826.

oi:10.1016/j.molstruc.2009.06.032



ARTICLE IN PRESS

[29] S.A. Studenikin, N. Golego, M. Cocivera, J. Appl. Phys. 84 (1998) 2287.[30] J. Wang, J. Wu, Z.H. Zhang, X.D. Zhang, L. Wang, L. Xu, B.D. Guo, H. Li, J. Tong,

Chin. Chem. Lett. 16 (2005) 1105.[31] H.A. Benesi, J.H. Hildebrand, J. Am. Chem. Soc. 71 (1949) 2703.[32] J.R. Lakowicz, G. Weber, Biochemistry 12 (1973) 4161.[33] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, Plenum Press, New

York, 1983.[34] G.Z. Chen, X.Z. Huang, J.G. Xu, Z.B. Wang, Z.Z. Zhang, Method of Fluorescent

Analysis, second ed., Science Press, Beijing, 1990, p. 123, 126 (Chapter 4).[35] J.N. Miller, Proc. Anal. Div. Chem. Soc. 16 (1979) 203.[36] E.A. Brustein, N.S. Vedenkina, M.N. Irkova, Photochem. Photobiol. 18 (1973)

263.[37] B. Klajnert, M. Bryszewska, Bioelectrochemistry 55 (2002) 33.


[38] A. Kathiravan, R. Renganathan, Spectrochim. Acta A 71 (2008) 1080.[39] V. Anbazhagan, R. Renganathan, J. Lumin. 128 (2008) 1454.[40] P. Athina, J.G. Rebecca, A.F. Richard, J. Agric. Food Chem. 53 (2005) 158.[41] S. Molet, P. Gosset, P. Lassalle, W. Czarlewski, A.B. Tonnel, Clin. Exp. Allergy 27

(1997) 1167.[42] A. Kathiravan, M. Chandramohan, R. Renganathan, S. Sekar, J. Mol. Struc. 919

(2009) 210.[43] B. Lin, Z. Fu, Y. Ji, Appl. Phy. Lett. 79 (1991) 943.[44] L. Cyril, J.K. Earl, W.M. Sperry, Biochemists Handbook, E & F.N. Spon, London,

1961.[45] W.Y. He, Y. Li, C.X. Xue, Z.D. Hu, X.G. Chen, F.L. Sheng, Bioorg. Med. Chem. 13

(2005) 1837.[46] Y.J. Hu, Y. Liu, L.X. Zhang, J. Mol. Struct. 750 (2005) 174.

oi:10.1016/j.molstruc.2009.06.032


Z. Phys. Chem. 222 (2008) 647–654 . DOI 10.1524.zpch.2008.5299© by Oldenbourg Wissenschaftsverlag, München

Photosensitization of Colloidal TiO2 with

ZnTPP and Pyrene

By A. Kathiravan, V. Anbazhagan, M. Asha Jhonsi, and R. Renganathan*

School of Chemistry, Bharathidasan University, Tiruchirappalli-620 024 Tamil Nadu, India

(Received July 25, 2006; accepted September 5, 2007)

ZnTPP . Fluorescence Quenching . Photoinduced Electron Transfer . TiO2

The interaction of Zinc(II) tetraphenyl porphyrin (ZnTPP) and pyrene with colloidal TiO2 wasstudied by absorption and fluorescence spectroscopy. Upon excitation of its absorption band, thefluorescence emission of ZnTPP and pyrene was quenched by colloidal TiO2. The bimolecularquenching rate constant (kq) is 4.09 ! 1011 MK1sK1 for ZnTPP and 2.09 ! 109 MK1sK1 forpyrene. ZnTPP and pyrene adsorbed on colloidal TiO2 can participate in the sensitization proc-ess by injecting electrons from their excited states into the conduction band of TiO2. Themechanism is discussed on the basis of free energy change (ΔGet) obtained from Rehm-Wellerequation considering quenching rate constants and conduction band potential of colloidal TiO2.

1. Introduction

Organic dyes are an important class of compounds which usually can undergoboth oxidative and reductive processes. Photoreduction of an excited state dyemolecule by a substrate usually involves an electron or hydrogen atom transferfrom the reducing agent to the dye [1]. During the last few years, a number oforganic dyes, such as phthalocyanines, triphenyl methane, xanthenes, coumarins,and porphyrins have been tested as sensitizers [2]. TiO2 is a semiconductor andgood photocatalyst for the removal of organic and inorganic pollutants [3]. Sensi-tization of semiconductors has been studied extensively in the past [4]. Sensitiza-tion of large band-gap semiconductor electrode with organic dye molecules hasbeen a field of extensive research over past decades due to its potential applica-tions in solar energy conversion [5, 6].

The basic photoinduced reactions can be written as

Dye /hν

Dye* (1)

* Corresponding author. E-mail: [email protected]

648 A. Kathiravan et al.

Dye* KK/TiO2 Dye+ + ecb

K(TiO2) (2)

Dye+ + eK / Dye (3)

Owing to their strong absorption in the 400–450 nm region (B or Soret band) aswell as absorptions in the 500–700 nm region (Q bands), porphyrin derivativesare suitable photosensitizers for the photovoltaic conversion of solar energy.Several photosensitization studies by porphyrins have been reported [7–9]. Wuet al. reported, hypocrellin B (HB) as a sensitizer to study electron injection incolloidal TiO2 particles [10]. Recently we have reported interaction betweenxanthene dyes and colloidal TiO2 via electron transfer mechanism [11].

The present study involves the use of an acetonitrile soluble ZnTPP andpyrene as sensitizers to study the electron injection from excited state of thesedyes to the conduction band of the semiconductor TiO2.

2. Experimental section

2.1 Materials

Zinc(II) tetraphenyl porphyrin (ZnTPP), pyrene and titanium(IV) 2-propoxidewere purchased from Aldrich. The spectroscopic grade solvent acetonitrile wasused for preparing the solutions. All measurements were performed at roomtemperature (28 oC).

2.2 Preparation of colloidal TiO2

The colloidal TiO2 suspension was prepared by the hydrolysis of titanium(IV)2-propoxide in acetonitrile, as described earlier [12, 13]. Freshly prepared colloi-dal TiO2 stock solution (1 ! 10K2 M) was diluted with acetonitrile to obtainthe desired concentration of TiO2. No attempts were made to exclude the tracesof 2-propanol (~ 0.4%) present in the colloidal semiconductor suspension andit was confirmed separately that the presence of 2-propanol did not affect thephotochemical measurements as earlier reported [14].

Photosensitization of Colloidal TiO2 … 649

Fig. 1. Absorption(A) and emission(B) spectrum of ZnTPP.

Fig. 2. Absorption(A) and emission(B) spectrum of Pyrene.

2.3 Instrumentation

The steady state fluorescence quenching measurements were carried out withJASCO FP-6500 spectrofluorometer. The excitation wavelength of ZnTPP andpyrene were 420 nm and 334 nm and the emission was monitored at 602 nmand 375 nm respectively. The excitation and emission slit widths (3 nm) andscan rate (500 nm) were kept constant for all the experiments. Absorption spectrawere recorded using Cary300 UV-Vis spectrophotometer. Samples were pre-pared by dissolving ZnTPP and pyrene in acetonitrile and administering the ap-propriate amounts of colloidal TiO2. The samples were carefully degassed usingpure nitrogen gas for 15 minutes. Quartz cells (4 ! 1 ! 1 cm) with high vacuumTeflon stopcocks were used for degassing.


The absorption and emission spectra of ZnTPP and pyrene (Fig. 1 & 2) weretaken in the absence and in presence of colloidal TiO2 (Fig. 3–5 & 6). The


Fig. 3. Absorption spectrum of ZnTPP in presence of Colloidal TiO2 in acetonitrile.

Fig. 4. Absorption spectrum of pyrene in presence of Colloidal TiO2 in acetonitrile.

following observations were made: (i) The shape and band maxima of absorptionand fluorescence spectra remain unchanged, (ii) no other new emission band ofthe ZnTPP and pyrene is noticed, and (iii) absorbance of the ZnTPP and pyrenedoes not change during the course of experiment.

The above observations suggest that (i) the ZnTPP-TiO2 and pyrene-TiO2

interaction does not change the absorption and fluorescence spectral properties,(ii) the formation of any emissive exciplex may be ruled out.

The Stern-Volmer relationship was used for the analysis of fluorescencequenching.

I0.I = 1 + Ksv [Q] (4)

where I0 and I are the intensities of the fluorescer in the absence and the presenceof quencher, respectively, Ksv is Stern-Volmer quenching rate constant and [Q]is the concentration of the TiO2.


Fig. 5. Steady state fluorescence quenching of ZnTPP (1 ! 10K6 M) with colloidal TiO2 inthe concentration range of 0–5 ! 10K4 M in acetonitrile.

Fig. 6. Steady state fluorescence quenching of pyrene (1 ! 10K6 M) with colloidal TiO2 inthe concentration range of 0–5 ! 10K4 M in acetonitrile.

The bimolecular quenching rate constant (kq) was calculated using Eq. (5).

Ksv = τ · kq (5)

where, τ is the fluorescence lifetime of the ZnTPP and pyrene in the absence ofquencher, i.e., 1.78 ns and 374 ns in acetonitrile [15, 16], kq is the bimolecularquenching rate constant. Plots of I0.I vs. [Q] were linear for both sensitizers,indicating dynamic nature of quenching process (Fig. 7). Such observations ondynamic quenching based on steady state method [17] are quite common. Fromthe slope value we get the KSV and using Eq. (5) we obtained the kq values ofZnTPP and Pyrene (Table 1).


Table 1. Stern-Volmer constants & quenching rate constants of ZnTPP and Pyrene by TiO2.

S.No. Sensitizer KSVa kq

b

1. ZnTPP 7.29 ! 102 MK1 4.09 ! 1011 MK1sK1

2. Pyrene 7.82 ! 102 MK1 2.09 ! 109 MK1sK1

a obtained by Stern-Volmer equation, I0.I = 1 + KSV[Q]b determined by steady state fluorescence quenching in acetonitrile

Fluorescence quenching rate constant decreased in the order:ZnTPP > Pyrene

ZnTPP has a higher quenching rate constant than pyrene, this is due to thelower oxidation potential of ZnTPP than that of pyrene [22, 23].

3.1 Schematic diagram describing the electron transfer quenchingprocess

3.2 Calculation of free energy changes (ΔGet) for the electron transferreactions

The nature of the electron transfer pathway (i.e., oxidative or reductive quench-ing) can be understood by examining the free energy of the corresponding elec-

Fig. 7. Stern-Volmer plot for the fluorescence quenching of ZnTPP and Pyrene with variousconcentrations of TiO2 (0–5 ! 10K4 M) in acetonitrile.


Table 2. Photophysical properties of ZnTPP and pyrene in acetonitrile.

S.No. Sensitizer τ (ns) E(0,0)(eV) E1.2oxi

vs. SCE (V) ΔGet (eV)e

1. ZnTPP 1.78a 2.10 0.94c K0.722. Pyrene 374b 3.34 1.22d K1.68

a,b from Literature [15] & [16].c,d from Literature [22] & [23].e calculated by Rhem-Weller equation ΔGet = E1.2

oxiK E1.2

redK E(0,0) + C

tron transfer reactions. Thermodynamics of electron transfer from the sensitizerto the quencher can be calculated by the well known Rehm-Weller equation [18].

ΔGet = E(oxid) K E1 /2(red) K E(0,0) + C (6)

where, E1/2(oxid) is the oxidation potential of the donor, E1 /2

(red) is the reductionpotential of the acceptor, E(0,0) is the singlet state energy of the sensitizer and Cis the coulomb term which describes the electrostatic attraction within the contaction pair and has a value of ca. K0.06 eV in acetonitrile [19]. The oxidationpotential, excited singlet state energy (E(0,0)) and the calculated ΔGet values forthe ZnTPP and pyrene are shown in Table 2. The conduction band potential ofTiO2 is K0.5 V vs. SCE [20]. By using Rehm-Weller equation negative valuesof ΔGet were obtained. The absence of any overlap between the emission spectraof the sensitizers and the absorption spectrum of TiO2 and the well establishedelectron accepting nature of ecb

K of TiO2 (in presence of excited electron donorsnamely the dyes [21]) coupled with the above observation support that thequenching mechanism in this work involves electron transfer (i.e., electron trans-fer from excited ZnTPP and pyrene to TiO2).

4. Conclusions

The fluorescence quenching of singlet excited ZnTPP and Pyrene by colloidalTiO2 was carried out in acetonitrile. Based on the above results, it is suggestedthat (i) the fluorescence quenching involves electron transfer mechanism. ZnTPPand pyrene serve as electron donors and TiO2 as electron acceptor. (ii) noground-state complex is formed between the fluorophore and quenchers.

Acknowledgement

R.R. and V.A thank DST (Ref: SP.S1.H-41.2001, dt: 12-09-2002) (Govern-ment of India) for the Project and fellowship respectively. Authors thank DST-FIST and UGC-SAP for facilities to School of Chemistry.


References

1. P. S. Rao and E. Hayon, J. Phys Chem. 77 (1973) 2753–2756.2. H. N. Ghosh, J. Chem Sci. 119 (2007) 205–215.3. A. Mills and S. L. Hunte, J. Photochem Photobiol. A: Chem. 108 (1997) 1–35.4. A. Nozik and R. Memming, J. Phys Chem. 100 (1996) 13061–13078.5. B. O'Regan, J. M. Anderson, and M. Gratzel, J. Phys Chem. 94 (1990) 8720–8726.6. B. O'Regan and M. Gratzel, Nature 253 (1991) 737–740.7. A. Kay and M. Grätzel, J. Phys Chem. 97 (1993) 6272–6277.8. K. Kalyanasundaram, N. Vlachopoulos, V. Krishnan, A. Monnier, and M. Gratzel,

J. Phys Chem. 91 (1987) 2342–2347.9. S. Cherian and C. C. Wamser, J. Phys Chem. B 104 (2000) 3624–3629.

10. T. Wu, S. J. Xu, J. Q. Shen, S. Chen, M. H. Zhang, and T. Shen, J. PhotochemPhotobiol. A: Chem. 137 (2000) 191–196.

11. A. Kathiravan, V. Anbazhagan, M. Asha Jhonsi, and R. Renganathan, Z. Phys Chem.221 (2007) 941–948.

12. P. V. Kamat, J. Photochem 28 (1985) 513–524.13. P. V. Kamat, Langmuir 1 (1985) 608–611.14. P. V. Kamat, J. P. Chauvet Richard, and W. Fessenden, J. Phys Chem. 90 (1986)

1389–1394.15. Y. Li and R. P. Steer, Chem. Phys. Lett. 373 (2003) 94–99.16. M. Kristiansen, R. D. Scurlock, K. K. Iu, and P. R. Ogilby, J. Phys Chem. 95 (1991)

5190–5197.17. V. K. Sharma, D. Mohan, and P. D. Sahare, Spectrochim. Acta A 66 (2007) 111–

113.18. D. Rehm and A. Weller, Isr. J. Chem 8 (1970) 259–271.19. V. Anbazhagan, A. Kathiravan, M. Asha Jhonsi, and R. Renganathan, Z. Phys Chem.

221 (2007) 929–939.20. P. V. Kamat, J. Phys Chem. 93 (1989) 859–864.21. Q. Dai and J. Rabani, J. Photochem Photobiol. A: Chem. 148 (2002) 17–24.22. K. Kalyanasundaram, Photochemistry of polypyridine and porphyrin complexes.

Chapter 14, Academic press, London (1992).23. A. A. Abdel-Shafi, J. Phys Chem. A 104 (2000) 5747–5757.

P

AS

a

ARA

KRFPC

1

moSsicsdteccscanp

1d

Spectrochimica Acta Part A 71 (2008) 1080–1083


Spectrochimica Acta Part A: Molecular andBiomolecular Spectroscopy

journa l homepage: www.e lsev ier .com/ locate /saa

hotoinduced interaction between riboflavin and TiO2 colloid

. Kathiravan, R. Renganathan ∗

chool of Chemistry, Bharathidasan University, Tiruchirappalli 620024, Tamil Nadu, India

r t i c l e i n f o

rticle history:

a b s t r a c t

The adsorption of riboflavin on the surface of TiO2 colloidal particles and the electron transfer process from
eceived 29 October 2007ccepted 3 March 2008
eywords:iboflavinluorescence quenchinghotoinduced electron transferolloidal TiO2

its singlet excited state to the conduction band of TiO2 were examined by absorption and fluorescencequenching measurements. The apparent association constants (Kapp) were determined. The quenchingmechanism is discussed involving electron transfer from riboflavin to TiO2.

© 2008 Elsevier B.V. All rights reserved.

nifl[

. Introduction

Sensitization processes resulting from photoexcitation of dyeolecules (sensitizers) bound to semiconductor nanoparticles are

f great importance for photochemical solar energy conversion [1].ensitization can be achieved by adsorption of dye molecules at theemiconductor surface by an electrostatic, hydrophobic, or chem-cal interaction that, upon excitation, injects an electron into itsonduction band [2]. Semiconductor particles of colloidal dimen-ions are sufficiently small to yield transparent solutions, allowingirect analysis of electron transfer by a fluorescence quenchingechnique [3]. Sensitization of colloidal TiO2 has been studiedxtensively in the past [4–7]. Recently, we reported the fluores-ence quenching of meso-tetrakis(4-sulfonatophenyl)porphyrin byolloidal TiO2 [8]. Zhang and co workers made some comparativetudies of photophysical properties of fluorescein derivatives inolloidal TiO2 [9]. It was reported that fluorescein derivatives aredsorbed on colloidal TiO2 through their phenolic group [10]. TiO2anoparticles have been used as carriers of photosensitizers likeorphyrins [11] in the treatment of cancer therapy [12].

∗ Corresponding author. Tel.: +91 431 2407053; fax: +91 431 2407045.E-mail address: [email protected] (R. Renganathan).

rmbb

386-1425/$ – see front matter © 2008 Elsevier B.V. All rights reserved.oi:10.1016/j.saa.2008.03.004

Riboflavin (RB) is an yellow-orange compound that undergoesumerous reactions when exposed to solar radiation [13]. The

soalloxazine three-ring substructure in riboflavin behaves like auorescent chromophore and emits greenish-yellow fluorescence14].

We have studied the photoinduced interaction betweeniboflavin and colloidal TiO2 using absorption and fluorescenceethods. It is demonstrated that the association constant could

e explicitly derived. It is found that there is a strong interactionetween riboflavin and TiO2 particle surface.



dx.doi.org/10.1016/j.saa.2008.03.004

A. Kathiravan, R. Renganathan / Spectrochim

Fcr

2

2

At

2

tiobpppdN(in[

2

cwi5tnhts

3

3

atwwrTb

rr

r

K

Toa

ig. 1. Absorption spectrum of riboflavin in the presence of colloidal TiO2 in theoncentration range of 0–5 × 10−4 M. The inset is the dependence of 1/A − A0 on theeciprocal concentration of colloidal TiO2.

. Experimental

.1. Materials

Riboflavin and titanium(IV) 2-propoxide were purchased fromldrich. The doubly distilled water was used for preparing the solu-

ions. All measurements were performed at room temperature.

.2. Preparation of colloidal TiO2

The colloidal TiO2 suspension was prepared by the hydrolysis ofitanium(IV) 2-propoxide [15]. Typically, titanium(IV) 2-propoxiden 2-propanol (l0%, 0.5 ml) was injected by syringe into 40 mlf water kept stirred for 8 h under a N2 atmosphere. No sta-ilizing agents were used. The colloidal suspensions of TiO2repared by this method were stable for 3–5 days. For theresent study, fresh colloidal TiO2 dispersed in water was pre-ared before each set of experiments. The stock suspension wasiluted with water to obtain the desired concentration of TiO2.o attempts were made to exclude the traces of 2-propanol

∼0.4%) present in the colloidal semiconductor suspension andt was confirmed separately that the presence of 2-propanol didot affect the photochemical measurements as earlier reported16].

A

wdd

Scheme 1. Possible modes for adsorption

ica Acta Part A 71 (2008) 1080–1083 1081

.3. Steady-state measurements

The steady-state fluorescence quenching measurements werearried out with JASCO FP-6500 spectrofluorimeter. The excitationavelength of riboflavin was 445 nm and the emission was mon-

tored at 524 nm. The excitation and emission slit widths (eachnm) and scan rate (500 nm/s) were maintained constant for all

he experiments. The samples were carefully degassed using pureitrogen gas for 15 min. Quartz cells (4 cm × 1 cm × 1 cm) withigh vacuum teflon stopcocks were used for degassing. Absorp-ion spectral measurements were recorded using Cary300 UV–vispectrophotometer.

. Results and discussion

.1. Absorption characteristics

Fig. 1 shows the absorption spectra of riboflavin in neat waternd in water containing colloidal TiO2 at different concentra-ions. In the presence of colloidal TiO2 the absorption at bothavelengths (360 nm and 445 nm) increased without change inavelengths. This implies that there is an interaction of TiO2 with

iboflavin through hydroxyl group (Scheme 1) similar to colloidaliO2 through hydroxyl group of ethyl ester of fluorescein reportedy He et al. [17].

Generally the equilibrium for the formation of complex betweeniboflavin and colloidal TiO2 can be given by Eq. (1), where Kapp

epresents the apparent association constant:

iboflavin + TiO2Kapp� riboflavin . . . TiO2 (1)

app = [riboflavin . . . TiO2][riboflavin]•[TiO2]

he changes in intensity of the absorption peak at 445 nm as a resultf formation of the surface complex can be utilized to obtain Kapp

ccording to Benesi and Hildebrand equation [18]:

obs = (1 − ˛)C0εriboflavinl + ˛C0εcl (2)

here Aobs is the observed absorbance of the solution containingifferent concentrations of the colloidal TiO2 at 445 nm; ˛ is theegree of association between riboflavin and TiO2; εriboflavin and

of the riboflavin to the TiO2 surface.

1082 A. Kathiravan, R. Renganathan / Spectrochimica Acta Part A 71 (2008) 1080–1083

Ft1

ε(tAr

A

A(a

Toer1FdTf

3

qist(

Fs

a

wflsir

tK

3t

elinTstt

tR

ig. 2. Fluorescence quenching of riboflavin in the presence of various concentra-ion of colloidal TiO2, [TiO2] = 0,1,2,3,4,5,6 × 10−4 M. The inset is the dependence of/F0 − F on the reciprocal concentration of colloidal TiO2.

c are the molar extinction coefficients at the defined wavelength�max = 445 nm) for riboflavin and the formed complex, respec-ively, in water. Eq. (2) can be expressed by Eq. (3), where A0 andc are the absorbances of riboflavin and the complex at 445 nm,espectively, with the concentration of C0:

obs = (1 − ˛)A0 + ˛Ac (3)

t relatively high TiO2 concentrations, ˛ can be equated toKapp[TiO2])/(1 + Kapp[TiO2]). In this case, Eq. (3) can be expresseds Eq. (4):

1Aobs − A0

= 1Ac − A0

+ 1Kapp(Ac − A0)[TiO2]

(4)

herefore, if the enhancement of absorbance at the wavelengthf 445 nm was due to absorption of surface complex, one wouldxpect a linear relationship between 1/(Aobs − A0) and the recip-ocal concentration of colloidal TiO2 with a slope equal to/Kapp(Ac − A0) and an intercept equal to 1/(Ac − A0). The inset toig. 1 shows the plot for riboflavin, there is a good linear depen-ence of 1/(Aobs − A0) on the reciprocal concentration of colloidaliO2. The value of apparent association constant (Kapp) determinedrom this plot is 8.87 × 104 M−1.

.2. Fluorescence quenching characteristics

Addition of TiO2 colloid to a solution of riboflavin resulted in theuenching of its fluorescence emission. Fig. 2 shows the effect of
ncreasing concentration of TiO2 colloid on the fluorescence emis-ion spectrum of riboflavin. This quenching behaviour is similar tohe studies earlier reported [7]. The apparent association constantKapp) can also be obtained from the fluorescence quenching data
�

wi

Scheme 2. Electron transfe

ig. 3. The overlap of the (A) absorption spectrum of colloidal TiO2 (B) emissionpectrum of riboflavin.

ccording to following equation,

1F0 − F

= 1F0 − F ′ + 1

Kapp(F0 − F ′)[TiO2](5)

here Kapp is the apparent association constant, F0 is the initialuorescence intensity of riboflavin, F′ is the fluorescence inten-ity of TiO2 adsorbed riboflavin and F is the observed fluorescencentensity at its maximum. The plot of 1/F0 − F versus 1/[TiO2] foriboflavin is shown in the inset to Fig. 2.

A good linear relationship was obtained between 1/F0 − F andhe reciprocal concentration of colloidal TiO2. From the slope, theapp has been assessed and the value is 1.42 × 103 M−1.

.3. Calculation of free energy changes (�Get) for the electronransfer reactions

The decrease in fluorescence emission may be attributed to thelectron transfer or energy transfer between the riboflavin and col-oidal TiO2. The bandgap energy of the TiO2 semiconductor particles greater than the singlet excited state energy of the riboflavin ando overlapping of fluorescence emission of riboflavin with colloidaliO2 (Fig. 3), thus ruling out energy transfer from the singlet excitedtate of riboflavin to TiO2 colloids. Therefore, it is concluded thathe fluorescence quenching shown in Fig. 2 is caused by electronransfer (Scheme 2).

The thermodynamic feasibility of the excited singlet state elec-ron transfer reaction was calculated by employing the well knownehm-Weller expression [19].

Get = E(ox)1/2 − E(red)

1/2 − E∗ + C (6)

here, E(ox)1/2 is the oxidation potential of riboflavin (−0.5 V), E(red)

1/2s the reduction potential of TiO2 (i.e. conduction band potential of

r quenching process.

ochim

T(iitce[

4

patae

A

eUB

R

[

[[[[[

[[

[[[

[21] S. Parret, F.M. Savary, J.P. Fouassier, P. Ramamurthy, J. Photochem. Photobiol. A:Chem. 83 (1994) 205–209.

[22] K. Kikuchi, T. Niwa, Y. Takahashi, H. Ikeda, T. Miyashi, J. Phys. Chem. 97 (1993)5070–5073.

[23] S. Nath, H. Pal, D.K. Palit, A.V. Sapre, J.P. Mittal, J. Phys. Chem. A 102 (1998)5822–5830.

A. Kathiravan, R. Renganathan / Spectr

iO2 (−0.1 V) [20], E* is the excited singlet state energy of riboflavin2.7 eV) [it can be obtained from the position of the absorption band.e. � = 445 nm] and C is the coulombic term. Since one of the speciess neutral and the solvent used is polar in nature, the coulombicerm in the above expression is neglected [21]. The �Get value wasalculated as −3.1 eV and this higher negative �Get value indicateslectron transfer processes which is thermodynamically favorable22,23].

. Conclusion

Riboflavin adsorbs on the surface of colloidal TiO2 through itshenolic group, as evidenced by the effects of colloidal TiO2 on thebsorption and fluorescence spectroscopy. The apparent associa-ion constant (Kapp) has been determined from absorption changesnd fluorescence quenching. Electron injection from its singletxcited state into the conduction band of TiO2 is suggested.

cknowledgements

R.R. thanks DST (Ref: SP/S1/H-41/2001,dt: 12-09-2002) (Gov-rnment of India) for the project. Authors also thank DST-FIST andGC-SAP for spectrofluorimeter facility in the School of Chemistry,harathidasan University, Trichy.

eferences

[1] B. O’Regan, M. Gratzel, Nature 353 (1991) 737–740.[2] H.P. Zhang, Y.L. Zhou, M.H. Zhang, T. Shen, Y.L. Li, D.B. Zhu, J. Colloid Interface

Sci. 264 (2003) 290–295.

ica Acta Part A 71 (2008) 1080–1083 1083

[3] C. Chen, X. Qi, B. Zhou, J. Photochem. Photobiol. A: Chem. 109 (1997) 155–158.[4] H. Mao, H. Deng, H. Li, Y. Shen, Z. Lu, H. Xu, J. Photochem. Photobiol. A: Chem.

114 (1998) 209–212.[5] T. Wu, S.J. Xu, J.Q. Shen, S. Chen, M.H. Zhang, T. Shen, J. Photochem. Photobiol.

A: Chem. 137 (2000) 191–196.[6] Z.X. Zhou, S.P. Qian, S.D. Yao, Z.Y. Zhang, Dyes Pigments 51 (2001) 137–144.[7] Z. Zhou, S. Qian, S. Yao, Z. Zhang, Radiat. Phys. Chem. 65 (2002) 241–248.[8] A. Kathiravan, V. Anbazhagan, M.A. Jhonsi, R. Renganathan, Spectrochim. Acta

A (in press).[9] H.P. Zhang, Y.L. Zhou, M.H. Zhang, T. Shen, Y.L. Li, D.B. Zhu, J. Phys. Chem. B 106

(2002) 9597–9603.10] H.P. Zhang, Y.L. Zhou, M.H. Zhang, T. Shen, Y.L. Li, D.B. Zhu, J. Colloid Interface

Sci. 251 (2002) 443–446.11] S. Wang, R. Gao, F. Zhou, M. Selke, J. Mater. Chem. 14 (2004) 487–493.12] A. Wiseman, Handbook of Enzyme Biotechnology, Horwood, Chichester, 1985.13] E. Choe, R. Huang, D.B. Min, J. Food Sci. 70 (2005) 28–36.14] P.S. Song, D.E. Metzler, Photochem. Photobiol. 6 (1967) 691–709.15] D. Bahnemann, A. Henglein, J. Lilie, L. Spanhel, J. Phys. Chem. 88 (1984) 709–

711.16] P.V. Kamat, J.P. Chauvet, R.W. Fessenden, J. Phys. Chem. 90 (1986) 1389–1394.17] J. He, F. Chen, J. Zhao, H. Hidaka, Colloid Surf. A: Physicochem. Eng. Aspects 142

(1998) 49–57.18] H.A. Benesi, J.H. Hildebrand, J. Am. Chem. Soc. 71 (1949) 2703–2707.19] D. Rehm, A. Weller, Isr. J. Chem. 8 (1970) 259–271.20] S.L. Murov, I. Carmichael, G.L. Hug, Handbook of Photochemistry, second ed.,

M. Dekker Inc., New York, 1993, pp. 269–273.

Am

AS

a

ARRA

KZFC

1

atiata

D

D

D

tp4xaeptoTT

1d





n investigation on electron transfer quenching of zinc(II)eso-tetraphenylporphyrin (ZnTPP) by colloidal TiO2

. Kathiravan, R. Renganathan ∗

chool of Chemistry, Bharathidasan University, Tiruchirappalli 620024, Tamil Nadu, India

r t i c l e i n f o

rticle history:

a b s t r a c t

The interaction of zinc(II) meso-tetraphenylporphyrin (ZnTPP) with colloidal TiO2 was studied by absorp-
eceived 3 November 2007eceived in revised form 3 March 2008ccepted 3 March 2008
eywords:

tion, steady state and time-resolved fluorescence spectroscopy. The quenching was found to obey theStern–Volmer equation and the corresponding Stern–Volmer plots were linear in the range of quencherconcentration used 0–5 × 10−4 M. The bimolecular quenching rate constants (kq) were 20.5 × 1010 M−1 s−1

(steady-state) and 2.85 × 1010 M−1 s−1 (time resolved). The quenching process is suggested to involveelectron transfer from the ZnTPP to TiO2 considering the experimental evidences obtained.

qir

2

nTPPluorescence quenchingolloidal TiO2

. Introduction

Semiconductor materials such as TiO2 have been widely useds photocatalysts for solar energy conversion [1] and for the pho-odegradation of organic pollutants [2]. Photochemical studiesnvolving dyes are an active area of research particularly with TiO2s such systems lead to sensitization of TiO2 via electron injectiono its conduction band by excited dyes [3–6], and the steps involvedre usually indicated as:

yeh�−→Dye∗ (1)

ye∗TiO2−→Dye+ + e−cb(TiO2) (2)

ye+ + e− → Dye (3)

Renganathan and co-workers reported TiO2 mediated pho-ooxidation of methylene blue [7], certain pyrimidines [8–11],hotocatalysed reaction of meso-tetraphenylporphyrin on TiMCM-1 [12] and also recently studies on fluorescence quenching ofanthene dyes [13]. Porphyrin and porphyrin derivatives are suit-ble photosensitizers for the photovoltaic conversion of solarnergy due to their strong absorption in the visible region [14]. Por-hyrins find applications in photodynamic therapy [15] of cancer
reatment. Recently we reported [16], the fluorescence quenchingf meso-tetrakis(4-sulfonatophenyl)porphyrin [TSPP] by colloidaliO2, wherein TSPP donating an electron to conduction of band ofiO2 is suggested.

2

pgwp



Due to the above points of significance we have studied theuenching of zinc(II) meso-tetraphenylporphyrin by colloidal TiO2n this work as a preliminary model using steady state and time-esolved techniques.

. Experimental

.1. Materials

Zinc(II) meso-tetraphenylporphyrin and titanium(IV) 2-ropoxide were purchased from Aldrich. The spectroscopicrade solvents N,N′-dimethylformamide (DMF) and 2-propanolere used for preparing the solutions. All measurements wereerformed at room temperature.



dx.doi.org/10.1016/j.saa.2008.03.006

rochimica Acta Part A 71 (2008) 1106–1109 1107

2

2

tlTiNpFpdapfip

2

iwcu

cwarAU

2

pt(lt(Ywbmtntsta

2

tegR(g

3

3

aono

3

S

wpta�oicp0aolthe observed quenching is due to a diffusive process.

The ability of the excited porphyrins to inject electron into theconduction band of the semiconductor is determined by the energydifference between the conduction band potential of the semicon-

A. Kathiravan, R. Renganathan / Spect

.2. Methods

.2.1. Preparation of colloidal TiO2The colloidal TiO2 suspension was prepared by the hydrolysis of

itanium(IV) 2-propoxide [17]. The method of preparation of col-oidal TiO2 in DMF was similar to the one employed earlier [14].ypically, titanium(IV) 2-propoxide in 2-propanol (10%, 0.5 ml) wasnjected by syringe into 40 ml of DMF kept stirred for 8 h under a

2 atmosphere. No stabilizing agents were used. The colloidal sus-ensions of TiO2 prepared by this method were stable for 3–5 days.or the present study, fresh colloidal TiO2 dispersed in DMF wasrepared before each set of experiments. The stock suspension wasiluted with DMF to obtain the desired concentration of TiO2. Nottempts were made to exclude the traces of 2-propanol (∼0.4%)resent in the colloidal semiconductor suspension and it was con-rmed separately that the presence of 2-propanol did not affect thehotochemical measurements as earlier reported [18].

.2.2. Steady-state measurementsSamples were prepared by dissolving ZnTPP in DMF and admin-

stering the appropriate amounts of colloidal TiO2. The samplesere carefully degassed using pure nitrogen gas for 15 min. Quartz

ells (4 cm × 1 cm × 1 cm) with high vacuum Teflon stopcocks weresed for degassing.

The steady-state fluorescence quenching measurements werearried out in a JASCO FP-6500 spectrofluorimeter. The excitationavelength of ZnTPP was 425 nm and the emission was monitored

t 606 nm. The excitation and emission slit width (5 nm) and scanate (500 nm/s) were maintained constant for all the experiments.bsorption spectral measurements were recorded using Cary300V-Vis spectrophotometer.

.2.3. Time-resolved measurementsFluorescence lifetime measurements were carried out in a

icosecond time correlated single photon counting (TCSPC) spec-rometer. The excitation source was the tunable Ti-sapphire laserTSUNAMI, Spectra Physics, USA). The diode laser pumped mil-ennia V (Spectra Physics) CW Nd-YVO4 laser was used to pumphe sapphire rod in the Tsunami mode locked picosecond laserSpectra Physics). The diode laser output was used to pump the Nd-VO4 rod in the Millennia. The time-resolved fluorescence emissionas monitored at 606 nm. The emitted photons were detected

y a MCP-PMT (Hamamtsu R3809U) after passing through theonochromater (f/3). The laser source was operated at 4 MHz and

he signal from the photodiode was used as a stop signal. The sig-al from the MCP-PMT was used as start signal in order to avoidhe dead time of the TAC. The difference between the start andtop signal is due to the time taken by the pulses traveling throughhe cables and electronic relaxation of the excited state. The datanalysis was carried out by the software provided by IBH (DAS-6).

.2.4. Cyclic voltammetric measurementsThe oxidation potential of ZnTPP was measured in DMF with

etrabutylammonium perchlorate (TBAP, 0.1 M) as electrolyte. The
xperimental setup consisted of a platinum working electrode, alassy carbon counter electrode and a silver reference electrode.eversible peak potentials were measured at different scan rates0.05 V/s). All samples were deaerated by bubbling with pure nitro-en gas for ca. 5 min at room temperature.
Fv

Fig. 1. Absorption (A) and emission (B) spectrum of ZnTPP in DMF.


.1. Absorption characteristics of the ZnTPP–TiO2 system

Fig. 1 shows absorption and emission spectrum of ZnTPP. Theddition of TiO2 did not change the shape of absorption spectrumf ZnTPP and specifically, there were no new absorption bands ando peak broadening was detected, thus eliminating the possibilityf the ground state complex formation between TiO2 and ZnTPP.

.2. Fluorescence quenching by colloidal TiO2

The fluorescence quenching behaviour is usually described bytern–Volmer relation:

�0

�= I0

I= 1 + KSV[Q ] = 1 + kq�0[Q ] (4)

here, I0 and I are the fluorescence intensities in the absence andresence of quencher, KSV is the Stern–Volmer constant related tohe bimolecular quenching rate constant, kq, by KSV = kq�0, and �0nd � are fluorescence lifetime of fluorophore (namely ZnTPP with= 1.89 ns) in the absence and presence of the quencher. Additionf TiO2 colloid to a solution of ZnTPP resulted in the quenching ofts fluorescence emission. Typical S–V plot for steady-state fluores-ence quenching of ZnTPP by colloidal TiO2 is shown in Fig. 2. S–Vlot was linear with a correlation coefficient (R2) of greater than.9863 indicating the dynamic nature of quenching process andbsence of static quenching. The kq value (20.5 × 1010 M−1 s−1) wasbtained from the initial slope (3.89 × 102 M−1) of the S–V plot. Thisarge value of kq indicates efficient quenching by colloidal TiO2 and

ig. 2. Stern–Volmer plot for the steady-state fluorescence quenching of ZnTPP witharious concentrations of TiO2 (0–5 × 10−4 M) in DMF.

1108 A. Kathiravan, R. Renganathan / Spectrochimica Acta Part A 71 (2008) 1106–1109

Scheme 1. Schematic diagram describing the conduction and valence bands for TiO2

and the electron-donating energy level for ZnTPP.

Fig. 3. Time-resolved fluorescence quenching of ZnTPP (2 × 10−6 M) with TiO2 int

ds

−−(

Fw

3

tcs

botwiq

3

reTqt

�

wroothe coulombic term in the above expression is neglected [21]. The�Get value thus calculated for the ET processes in the systemsstudied in DMF is negative (−3.19 eV). Hence, the ET processes

he concentration range of 0–5 × 10−4 M in DMF.

uctor and the oxidation potential of the porphyrins in the excitedtate.

The standard oxidation potential of the excited-singlet ZnTPP is0.78 V and the conduction band of colloidal TiO2 which is around
0.5 V yield a favorable energetics for electron transfer to occur.
Scheme 1)sl

Scheme 2. Schematic diagram describing th

ig. 4. Stern–Volmer plot for the time-resolved fluorescence quenching of ZnTPPith various concentrations of TiO2 (0–5 × 10−4 M) in DMF.

.3. Fluorescence lifetime measurements

The fluorescence quenching data clearly highlighted the role ofhe excited-singlet state of porphyrins in injecting electron into theonduction band of the semiconductor TiO2 and dynamic (colli-ional quenching) in nature, supported by the linear S–V plot.

In general, static and dynamic quenching can be distinguishedy their differing lifetime measurements [19]. Fluorescence lifetimef ZnTPP has regularly decreased upon addition of desired concen-ration of colloidal TiO2 [Fig. 3]. S–V plots of �0/� vs. [Q] were linearith a correlation coefficient (R2) of greater than 0.9865 confirm-

ng the dynamic nature of quenching process and absence of staticuenching [Fig. 4]. The kq is found to be 2.85 × 1010 M−1 s−1.

.4. Calculation of free energy changes (�Get)

The nature of the electron transfer pathway (i.e., oxidative oreductive quenching) can be understood by examining the freenergy change of the corresponding electron transfer reactions.hermodynamics of electron transfer from the sensitizer to theuencher can be calculated by the well known Rehm–Weller equa-ion [20].


1/2 − E(0,0) + C (5)

here, E(ox)1/2 is the oxidation potential of the donor, E(red)

1/2 is theeduction potential of the acceptor, E(0,0) is the singlet state energyf the sensitizer (2.91 eV) and C is the coulombic term. Since onef the species is neutral and the solvent used is polar in nature,

tudied are thermodynamically favorable. The absence of any over-ap between the emission spectra of the ZnTPP (606 nm) and the

e electron transfer quenching process.

rochim

aeeoit

4

attd�t

A

eRtS

R

[

[

[

[

[

[[

[

[

[1983, 258-265.

[20] D. Rehm, A. Weller, Isr. J. Chem. 8 (1970) 259–271.

A. Kathiravan, R. Renganathan / Spect

bsorption spectrum of TiO2 (365 nm) and the well establishedlectron accepting nature of e−

cb of TiO2 (in presence of excitedlectron donors namely the dyes [22]) coupled with the abovebservation support that the quenching mechanism in this worknvolves electron transfer (i.e., electron transfer from excited ZnTPPo TiO2) as shown in Scheme 2 [23].

. Conclusions

The interaction of ZnTPP by colloidal TiO2 has been studied bybsorption, fluorescence and time-resolved spectroscopy. Based onhe above results, it is suggested that quenching involves electronransfer mechanism. The quenching rate constant values remainsiffusion controlled limit (∼1010 M−1 s−1) with higher negativeGet value. From the exergonic nature of thermodynamics, electron

ransfer from ZnTPP to TiO2 is suggested.

cknowledgements

R.R. thanks DST (Ref: SP/S1/H-41/2001, date: 12-09-2002) (Gov-rnment of India) for the project. We are thankful to Prof. P.amamoorthy, NCUFP, University of Madras, Chennai. Authors alsohank DST-FIST and UGC-SAP for spectrofluorimeter facility in thechool of Chemistry, Bharathidasan University, Trichy.

eferences

[1] P.V. Kamat, Chem. Rev. 93 (1993) 267–300.[2] M.R. Hoffmann, S.T. Martin, W.Y. Choi, D.W. Bahnemann, Chem. Rev. 95 (1995)

69–96.

[

[[

ica Acta Part A 71 (2008) 1106–1109 1109

[3] C.Y. Wang, C.Y. Liu, W.Q. Wang, T. Shen, J. Photochem. Photobiol. A: Chem. 109(1997) 159–164.

[4] T. Wu, S.J. Xu, J.Q. Shen, S. Chen, M.H. Zhang, T. Shen, J. Photochem. Photobiol.A: Chem. 137 (2000) 191–196.

[5] Z.X. Zhou, S.P. Qian, S.D. Yao, Z.Y. Zhang, Dyes Pigments 51 (2001) 137–144.[6] Q. Dai, J. Rabani, J. Photochem. Photobiol. A: Chem. 148 (2002) 17–24.[7] S. Lakshmi, R. Renganathan, S. Fujita, J. Photochem. Photobiol. A: Chem. 88

(1995) 163–167.[8] M.R. Dhananjeyan, R. Annapoorani, S. Lakshmi, R. Renganathan, J. Photochem.

Photobiol. A: Chem. 96 (1996) 187–191.[9] M.R. Dhananjeyan, R. Annapoorani, R. Renganathan, J. Photochem. Photobiol.

A: Chem. 109 (1997) 147–153.10] M.R. Dhananjeyan, V. Kandavelu, R. Renganathan, J. Mol. Catal. A: Chem. 151

(2000) 217–223.11] M.R. Dhananjeyan, V. Kandavelu, R. Renganathan, J. Mol. Catal. A: Chem. 158

(2000) 577–582.12] V. Kandavelu, M.R. Dhananjeyan, R. Renganathan, S.K. Badamali, P. Selvam, J.

Mol. Catal. A: Chem. 157 (2000) 189–192.13] A. Kathiravan, V. Anbazhagan, M. Asha Jhonsi, R. Renganathan, Z. Phys. Chem.

221 (2007) 941–948.14] H. Mao, H. Deng, H. Li, Y. Shen, Z. Lu, H. Xu, J. Photochem. Photobiol. A: Chem.

114 (1998) 209–212.15] E.A. Lissi, M.V. Encinas, E. Lemp, Chem. Rev. 93 (1993) 699–723.16] A. Kathiravan, V. Anbazhagan, M. Asha Jhonsi, R. Renganathan, Spectrochim.

Acta: A, in press.17] D. Bahnemann, A. Henglein, J. Lilie, L. Spanhel, J. Phys. Chem. 88 (1984) 709–

711.18] P.V. Kamat, J.P. Chauvet, R.W. Fessenden, J. Phys. Chem. 90 (1986) 1389–

1394.19] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, Plenum Press, New York,

21] S. Parret, F.M. Savary, J.P. Fouassier, P. Ramamurthy, J. Photochem. Photobiol. A:Chem. 83 (1994) 205–209.

22] Z. Zhou, S. Qian, S. Yao, Z. Zhang, Radiat. Phys. Chem. 65 (2002) 241–248.23] S. Nad, H. Pal, J. Phys. Chem. A 104 (2000) 673–680.

Colloids and Surfaces A: Physicochem. Eng. Aspects 324 (2008) 176–180


Colloids and Surfaces A: Physicochemical andEngineering Aspects

journa l homepage: www.e lsev ier .com/ locate /co lsur fa

Interaction of colloidal TiO2 with bovine serum albumin:

ia

colloictroscentrat (K) wsfer

A fluorescence quenching study

A. Kathiravan, R. Renganathan ∗

School of Chemistry, Bharathidasan University, Tiruchirappalli 620024, Tamil Nadu, Ind

a r t i c l e i n f o

Article history:Received 12 December 2007Received in revised form 10 March 2008Accepted 7 April 2008Available online 15 April 2008

Keywords:BSAFluorescence quenchingColloidal TiO2

a b s t r a c t

The interaction betweention and fluorescence spepresence of different concapparent binding constandiscussed. The energy tran

1. Introduction

Fluorescence quenching measurement of albumin is an impor-tant tool to investigate the interactions of drugs with serumalbumins. The nature of binding modes of the drugs with albumins
has been reported [1]. Bovine serum albumin (BSA) is a proteincontaining 582 amino-acid residues. It has two tryptophans at posi-tions 134 and 212 as well as tyrosine and phenylalanine [2] andthe protein intrinsic fluorescence is due to aromatic amino-acidresidues. From biopharmaceutical point of view, one of the mostimportant biological functions of albumins is their ability to carrydrugs as well as endogenous and exogenous substances, the bindingcapacity and sites of albumins have been characterized [3,4].
Nanoparticle probes acting as biosensors in chemical andbiochemical field have been researched recently and their applica-tions are becoming more extensive. Three types of nanoparticlesin biochemical analysis are used: metal nanoparticles [5], silicananoparticles [6,7] and luminescence quantum dot [8–10]. Theseprobes have been applied to the ultrasensitive detection of proteins,DNA sequencing, clinical diagnostics, etc. Reports on the interac-tion between BSA with various drugs like glycyrrhetinic acid [11],loratadine [12], flavanoids [13] and dyes including porphyrins [14],eosin [15,16], etc. have been cited.

In the present work, BSA is selected as the protein model. Criticalliterature survey reveals that attempts have not been made so far


0927-7757/$ – see front matter © 2008 Elsevier B.V. All rights reserved.doi:10.1016/j.colsurfa.2008.04.017

dal TiO2 and bovine serum albumin (BSA) was studied by using absorp-opic methods. The quenching of the intrinsic protein fluorescence in thetions of colloidal TiO2 was analyzed and number of binding sites (n) andere measured. The quenching mechanism of albumin by colloidal TiO2 is

efficiency (E) and critical transfer distance (R0) were determined.© 2008 Elsevier B.V. All rights reserved.

to investigate the mechanism of interaction of colloidal TiO2 withBSA. This is the first attempt made to investigate the mode of inter-action of colloidal TiO2 nanoparticles with BSA. TiO2 nanoparticleshave been used as carriers of photosensitizers like porphyrins [17]in the treatment of cancer therapy [18]. The quenching of intrinsictryptophan fluorescence of BSA has been used as a tool to study theinteraction of colloidal TiO2 nanoparticles with this transport pro-tein in an attempt to characterize the chemical associations takingplace. The binding interaction and the efficiency of energy transfer
between colloidal TiO2 and BSA have been investigated by fluores-cence methods. It is demonstrated that the binding constant andthe number of binding sites could be explicitly derived.
2. Materials and methods

2.1. Materials

Titanium(IV) 2-propoxide was purchased from Aldrich. Bovineserum albumin (Sigma) was dissolved in double distilled waterto prepare stock solution (1.0 × 10−4 M) which was then stored at0–4 ◦C. All measurements were performed at room temperature.

2.2. Preparation of colloidal TiO2

The colloidal TiO2 suspension was prepared by the hydrolysis oftitanium(IV) 2-propoxide [19]. Typically, titanium(IV) 2-propoxidein 2-propanol (l0%, 0.5 ml) was injected by syringe into 40 ml ofwater kept stirred for 8 h under a N2 atmosphere. No stabilizingagents were used. The colloidal suspensions of TiO2 prepared by



dx.doi.org/10.1016/j.colsurfa.2008.04.017

aces A

At relatively high TiO2 concentrations, ˛ can be equated to

A. Kathiravan, R. Renganathan / Colloids and Surf

Fig. 1. Absorption spectrum of BSA in the presence of colloidal TiO2 in the concen-tration range of (0–6) × 10−4 M.

this method were stable for 3–5 days. For the present study, freshcolloidal TiO2 dispersed in water was prepared before each setof experiments. The stock suspension was diluted with water toobtain the desired concentration of TiO2. No attempts were madeto exclude the traces of 2-propanol (∼0.4%) present in the colloidalsemiconductor suspension and it was confirmed separately that thepresence of 2-propanol did not affect the photochemical measure-ments as earlier reported [20]. The absorption of the colloidal TiO2in water is at 345 nm. The diameter of the particles determinedfrom the relationship between bandgapshift (�Eg) and radius (R)of quantum-size particles [21] is about 1.4 nm.

2.3. Steady-state measurements

The steady-state fluorescence quenching measurements werecarried out with JASCO FP-6500 spectrofluorimeter. The excitationwavelength of BSA was 280 nm and the emission was monitoredat 345 nm. The excitation and emission slit widths (each 5 nm) andscan rate (500 nm/min) were maintained constant for all the exper-iments. The samples were carefully degassed using pure nitrogengas for 15 min. Quartz cells (4 cm × 1 cm × 1 cm) with high vac-uum teflon stopcocks were used for degassing. Absorption spectralmeasurements were recorded using Cary300 UV–vis spectropho-tometer.


3.1. Absorption characteristics of BSA–TiO2 system

Fig. 1 shows the absorption spectra of BSA in neat water and inwater containing colloidal TiO2 at different concentrations. Fig. 2shows the absorption and emission spectrum of BSA. In the pres-

Fig. 2. Absorption (A) and emission (B) spectrum of BSA.

: Physicochem. Eng. Aspects 324 (2008) 176–180 177

ence of colloidal TiO2 the absorption peak increased markedly, butthe location of the peak did not change. It is due to the adsorptionof BSA partly on the surface of TiO2 colloid as supported by similarobservations made earlier [22].

The equilibrium for the formation of the complex between BSAand colloidal TiO2 can be given by Eq. (1), where Kapp representsthe apparent association constant:

BSA + TiO2Kapp� BSA . . . TiO2 (1)

Kapp = [BSA . . . TiO2][BSA][TiO2]

The changes in intensity of the absorption peak at 280 nm as aresult of formation of the surface complex can be utilized to obtainKapp according to Benesi and Hildebrand [23]:

Aobs = (1 − ˛)C0εBSA1 + ˛C0εc1 (2)

where Aobs is the observed absorbance of the solution contain-ing different concentrations of the colloidal TiO2 at 280 nm, ˛is the degree of association between BSA and TiO2, εBSA and εc

are the molar extinction coefficients at the defined wavelength(� = 280 nm) for BSA and the formed complex, respectively in water.Eq. (2) can be expressed as Eq. (3), where A0 and Ac are theabsorbances of BSA and the complex at 280 nm, respectively, withthe concentration of C0:

Aobs = (1 − ˛)A0 + ˛Ac (3)

(Kapp[TiO2])/(1 + Kapp[TiO2]). In this case, Eq. (3) can be changedto the following equation:

1Aobs − A0

= 1Ac − A0


(4)

Therefore, if the enhancement of absorbance at 280 nm was dueto absorption of surface complex, a linear relationship between1/(Aobs − A0) and the reciprocal concentration of colloidal TiO2 witha slope equal to 1/Kapp(Ac − A0) and an intercept equal to 1/(Ac − A0)(Fig. 1, inset). The value of apparent association constant (Kapp)determined from this plot is 3.16 × 105 M−1.

3.2. Fluorescence quenching by colloidal TiO2

The fluorescence quenching is described by Stern–Volmer rela-tion:

I0I

= 1 + KSV[Q] = 1 + kq�0[Q] (5)

where I0 and I are the fluorescence intensities in the absence andpresence of quencher, KSV is the Stern–Volmer constant related

Fig. 3. Fluorescence quenching of BSA in the presence of various concentration ofcolloidal TiO2, [TiO2] = 0, 1, 2, 3, 4, 5, 6 and 7 × 10−5 M.

Administrator

Highlight

Administrator

Note

it should be "-4"

aces A
178 A. Kathiravan, R. Renganathan / Colloids and Surf
Fig. 4. Stern–Volmer plot for the steady-state fluorescence quenching of BSA bycolloidal TiO2.

to the bimolecular quenching rate constant, kq, by KSV = kq�0, and�0 is the average lifetime of BSA in the absence of colloidal TiO2evaluated as 10−8 s [24], [Q] is the concentration of quencher. Addi-tion of TiO2 colloid to a solution of BSA resulted in the quenchingof its fluorescence emission. Fig. 3 shows the effect of increasingconcentration of TiO2 colloid on the fluorescence emission spec-trum of BSA. According to Eq. (5) from the linear plot (Fig. 4) ofI0/I against [Q], the quenching rate constant was obtained from theslope (kq = 1.01 × 1012 M−1 s−1, r = 0.9963), assuming �0 value of BSAas 1 × 10−8 s [24].

The decrease in fluorescence emission may be attributed to the
electron transfer or energy transfer between BSA and the colloidalTiO2. The singlet excited state energy of BSA (it can be obtained fromthe position of its absorption band) is greater than the band gapenergy of TiO2 (3.2 eV) semiconductor and fluorescence emission ofBSA can be overlapped with the absorption of colloidal TiO2 (Fig. 5);hence energy transfer from the singlet excited state of BSA to TiO2colloids is possible.
3.3. Energy transfer efficiency

Energy transfer efficiency (E) is given by the following equation[25]:

E = 1 − I

I0(6)

where I is the emission intensity of the donor in the presence ofacceptor and I0 is the emission intensity of the donor alone.

From the above results it is clear that, in presence of TiO2, thefluorescence intensity of BSA is reduced (from I0 to I) by energytransfer to TiO2. Based on the above Eq. (6) a plot of the values of ‘E’obtained at different concentrations of TiO2 is shown in Fig. 6. It is

Fig. 5. The overlap of the (A) absorption spectrum of colloidal TiO2 (B) emissionspectrum of BSA.

: Physicochem. Eng. Aspects 324 (2008) 176–180

Fig. 6. Energy transfer efficiency of BSA with various concentrations of colloidalTiO2.

seen from this figure that ‘E’ value steadily increases with increasein TiO2 concentration.

3.4. Energy transfer between colloidal TiO2 and BSA

According to Forster’s non-radiative energy transfer theory [26],the energy transfer efficiency is related not only to the distancebetween the acceptor and donor (r0), but also to the critical energytransfer distance (R0), that is:

E = R60

R60 + r6

0

(7)

where R0 is the critical distance when the transfer efficiency is 50%.

R60 = 8.8 × 10−25K2N−4˚J (8)

where K2 is the spatial orientation factor of the dipole, N the refrac-tive index of the medium, ˚ the fluorescence quantum yield ofthe donor, J the overlap integral of the fluorescence emission spec-trum of the donor and the absorption spectrum of the acceptor.Therefore,

J =∑ F(�)ε(�)�4 d�

F(�) d�(9)

where F(�) is the fluorescence intensity of the donor at wavelength�, ε(�) the molar absorptivity of the acceptor at wavelength �. Theoverlap of the absorption spectrum of TiO2 and the fluorescenceemission spectrum of BSA is shown in Fig. 5. The parameter J canbe evaluated by integrating the spectral parameters in Eq. (9). Underthese experimental conditions, we found R0 = 2.09 nm from Eq. (8)using K2 = 2/3, ˚ = 0.032, n = 1.3467 [27]. Obviously, R0 value is lowerthan the maximal academic critical distance for R0 (5–10 nm) [28].This is in accord with conditions of Forster’s non-radiative energytransfer theory [28], indicating the static quenching interaction

Fig. 7. Plot of log[(F0 − F)/F] vs. log[Q].

A. Kathiravan, R. Renganathan / Colloids and Surfaces A: Physicochem. Eng. Aspects 324 (2008) 176–180 179

ansfer
Scheme 1. Energy tr
between TiO2 and BSA. Using Eq. (7) we found r0 = 2.93 nm, it canbe seen that the data of R0 and r0 are both in the academic range,which suggests that non-radiative energy transfer occurs betweenTiO2 and BSA. Furthermore, it is suggested that the binding of TiO2to BSA is through energy transfer.


If it is assumed that there are similar and independent bindingsites in the biomolecule, the relationship between the fluorescence
intensity and the quencher medium can be deduced from the fol-lowing equation:
nQ + B → Qn + B (10)

where B is the fluorophore, Q is the quencher, Qn + B is the quenchedbiomolecules and the constant Ka is given by

Ka = [Qn + B][Q]n[B]

(11)

If the overall amount of biomolecules (bound or unbound withthe quencher) is B0, then [B0] = [Qn + B] + [B], here [B] is theconcentration of unbound biomolecules, then the relationshipbetween fluorescence intensity and the unbound biomolecule as[B]/[B0] = F/F0 that is:

log[

F0 − F

F

]= log K + n log[Q] (12)

where K is the binding constant of colloidal TiO2 with BSA,which can be determined by the intercept of the log[(F0 − F)/F]vs. log[Q] curves as shown in Fig. 7 and thus we obtained bind-ing constant K as 5.25 × 105 M−1 and number of binding sites (n)as 1.04.

quenching process.

3.6. Energy transfer quenching process

BSA absorbs light energy around 280 nm which is not absorbedby TiO2. The excited state energy of BSA can be transferred to groundstate TiO2, thus TiO2 gets excited and generates conduction bandelectrons and valence band holes that take part in subsequent redoxreactions with the surface-adsorbed molecules to yield the ultimateproducts (Scheme 1).

The photogenerated conduction band electrons are havingsufficient electronegativity to reduce dioxygen to superox-ide/hydroperoxide radicals to effect the deep oxidation of a wide
range of organic pollutants to degraded products. In addition, thephotogenerated holes are highly oxidizing and split water in toH+ and OH•, the produced OH• radicals can oxidize the organicpollutants.
4. Summary

The interaction between colloidal TiO2 and BSA has beenstudied by absorption and fluorescence spectroscopy. The resultspresented clearly indicated the adsorption of BSA molecules onthe surface of colloidal TiO2 and quenching of intrinsic proteinfluorescence by colloidal TiO2 through energy transfer. Quench-ing follows ground state complex formation via static quenchingmechanism. The binding constant and binding sites for BSA withcolloidal TiO2 have been calculated. It is possible to study theinteractions of nanoparticles with biomolecules based on presentmethod.

Acknowledgements

R.R. thanks DST for NSTI project and CSIR for funding.

[

180 A. Kathiravan, R. Renganathan / Colloids and Surfaces A

References

[1] D. Silva, C.M. Cortez, J.C. Bastos, S.R.W. Loura, Toxicol. Lett. 147 (2004) 53–61.[2] U.K. Hansen, Pharmacol. Rev. 33 (1981) 17–53.[3] F. Moreno, M. Cortijo, J.G. Jimenez, Photochem. Photobiol. 70 (1999) 695–700.[4] M. Kessler, O. Wolfbeis, Anal. Biochem. 200 (1992) 254–259.[5] R. Elghanian, J.J. Storhoff, R.C. Mucic, R.L. Letsinger, C.A. Mirkin, Science 277

(1997) 1078–1081.[6] S. Santra, P. Zhang, K. Wang, R. Tapec, W. Tan, Anal. Chem. 73 (2001) 4988–4993.[7] M. Qhobosheane, S. Santra, P. Zhang, W. Tan, Analyst 126 (2001) 1274–1278.[8] X.G. Peng, M.C. Schlamp, A.V. Kadavanich, A.P. Alivistos, J. Am. Chem. Soc. 119

(1997) 7019–7029.[9] W.C.W. Chan, S. Nie, Science 281 (1998) 2016–2018.

[10] L. Tan, L.Y. Liu, Q.J. Xie, Y.Y. Zhang, S.Z. Yao, Anal. Sci. 20 (2004) 441–444.[11] N. Zhou, Y.Z. Liang, P. Wang, J. Photochem. Photobiol A: Chem. 185 (2006)

271–276.12] B. Zhou, Z. Qi, Q. Xiao, J.X. Dong, Y.Z. Zhang, Y. Liu, J. Biochem. Biophys. Methods

70 (2007) 743–747.[13] Z. Hongwei, G. Min, Z. Zhaoxia, W. Wenfeng, W. Guozhong, Spectrochim. Acta

A 65 (2006) 811–817.

[[[[[[[

[

[

: Physicochem. Eng. Aspects 324 (2008) 176–180

[14] Y.B. Yin, Y.N. Wang, J.B. Ma, Spectrochim. Acta A 64 (2006) 1032–1038.[15] D. Gao, Y. Tian, F. Liang, D. Jin, Y. Chen, H. Zhang, A. Yu, J. Lumin. 127 (2007)

515–522.[16] L. Birla, A.M. Cristian, M. Hillebrand, Spectrochim. Acta A 60 (2004) 551–556.[17] S. Wang, R. Gao, F. Zhou, M. Selke, J. Mater. Chem. 14 (2004) 487–493.[18] A. Wiseman, Handbook of Enzyme Biotechnology, Horwood, Chichester, 1985.[19] D. Bahnemann, A. Henglein, J. Lilie, L. Spanhel, J. Phys. Chem. 88 (1984) 709–

711.20] P.V. Kamat, J.P. Chauvet, R.W. Fessenden, J. Phys. Chem. 90 (1986) 1389–1394.21] E. Joselevich, I. Willner, J. Phys. Chem. 98 (1994) 7628–7635.22] C. Chen, X. Qi, B. Zhou, J. Photochem. Photobiol. A: Chem. 109 (1997) 155–158.23] H.A. Benesi, J.H. Hildebrand, J. Am. Chem. Soc. 71 (1949) 2703–2707.24] J.R. Lakowicz, G. Weber, Biochemistry 12 (1973) 4161–4170.25] L.A. Sklar, B.S. Hudson, R.D. Simoni, Biochemistry 16 (1977) 5100–5108.26] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, Plenum Press, New York,

1983.27] L. Cyril, J.K. Earl, W.M. Sperry, Biochemists Handbook, E & F.N. Spon, London,

1961.28] G.Z. Chen, X.Z. Huang, J.G. Xu, Z.B. Wang, Z.Z. Zhang, Method of Fluorescent

Analysis, 2nd ed., Science Press, Beijing, 1990, pp. 123, 126 (Chapter 4).

C

Aa

b

a

ARRA

KTFC

1

atraTucs

D

D

Ssicsdteob

1d





yanobacterial chlorophyll as a sensitizer for colloidal TiO2

. Kathiravana, M. Chandramohanb, R. Renganathana,∗, S. Sekarb

School of Chemistry, Bharathidasan University, Tiruchirappalli 620 024, Tamil Nadu, IndiaDepartment of Biotechnology, Bharathidasan University, Tiruchirappalli 620 024, Tamil Nadu, India

r t i c l e i n f o

rticle history:eceived 6 May 2008eceived in revised form 17 June 2008ccepted 25 June 2008

a b s t r a c t

Chlorophyll has been extracted from cyanobacteria. The adsorption of chlorophyll on the surface of col-loidal TiO2 through electrostatic interaction was observed. The apparent association constant (Kapp) ofchlorophyll–TiO2 obtained from absorption spectra is 3.78 × 104 M−1. The Kapp value of chlorophyll–TiO2

4 −1

eywords:iO2 colloidluorescence quenchinghlorophyll

as determined from fluorescence spectra is 1.81 × 10 M , which matches well with that determinedfrom the absorption spectra changes. These data indicate that there is an interaction between chlorophylland colloidal TiO2 nanoparticle surface. The dynamics of photoinduced electron transfer from chlorophyllto the conduction band of colloidal TiO2 nanoparticle has been observed and the mechanism of electrontransfer has been confirmed by the calculation of free energy change (�Get) by applying Rehm–Wellerequation as well as energy level diagram. Lifetime measurements gave the rate constant (ket) for electron

state

soattta[srbt

injection from the excited

. Introduction

Semiconductor materials such as TiO2 have been widely useds photocatalysts for solar energy conversion [1] and for the pho-odegradation of organic pollutants [2]. However, solar energyeaching the surface of the earth and available to excite TiO2 is rel-tively small and artificial UV light sources are more expensive.herefore, recent efforts have been focused on exploring means totilize effectively the cheaper visible light [3]. Sensitization pro-esses resulting from photoexcitation of dye molecules bound toemiconductor nanoparticles are of great importance [4]:

yehv−→Dye∗ (1)

ye∗TiO2−→Dye•+ + e

•−cb (TiO2) (2)

ensitization can be achieved by adsorption of dye molecules at theemiconductor surface by an electrostatic, hydrophobic, or chem-cal interaction that, upon excitation, injects an electron into itsonduction band [5]. Semiconductor particles of colloidal dimen-ions are sufficiently small to yield transparent solutions, allowingirect analysis of electron transfer by a fluorescence quenching
echnique [6]. Sensitization of colloidal TiO2 has been studiedxtensively in the past [7–10]. Recently we have reported, the flu-rescence quenching of riboflavin, xanthene dyes and porphyrinsy colloidal TiO2 [11–13].

ltw


chlorophyll into the conduction band of TiO2 is 4.2 × 108 s−1.© 2008 Elsevier B.V. All rights reserved.

In this work we have employed cyanobacterial chlorophyll as aensitizer for colloidal TiO2. The cyanophyceae are a diverse groupf prokaryotic organisms that evolved in the pre-Cambrian era,pproximately 3.5 billion years ago [14]. They are responsible forhe oxygenation of the earth’s atmosphere, one billion years afterheir appearance in the fossil record. The ability to survive in warmemperatures, high light, and low carbon dioxide concentration hasllowed the cyanobacteria to radiate into a broad range of habitats15]. They perform oxygenic photosynthesis using the apparatusimilar to that in chloroplast of algae and plants [16]. In the lighteaction of oxygenic photosynthesis in cyanobacteria, the red andlue wavelengths of the visible light are mainly absorbed by cyclicetrapyrroles, the chlorophylls.

Chlorophyll is the dominant pigment on earth and serves as theight trapping and energy transferring chromophore in photosyn-hetic organisms. Although there are several forms of chlorophyll,hich differ only in the side chain, cyanobacteria contain only


http://www.elsevier.com/locate/saa


dx.doi.org/10.1016/j.saa.2008.06.031

1 ica Ac

cbbstTeccatia

fb

2

2

dm

2

t(4(apFsotToe

2

ctBtapoaa1as

2

cw6

sigumt

2

pt(l

3

3

mr

�

wts(he(a�dtT

3

tvdye (sensitizer) molecules when they adsorb on the nanoparticlesurface. Fig. 1 shows the absorption spectra of chlorophyll in theabsence and presence of colloidal TiO2 at different concentrations.In the presence of colloidal TiO2 the optical density at the wave-length of 480 nm was increased without change in wavelength. This

784 A. Kathiravan et al. / Spectrochim

hlorophyll a [14]. Chlorophylls are effective photoreceptorsecause they contain a network of alternating single and dou-le bonds, and the orbitals can delocalize electrons stabilizing thetructure. Such delocalized polyenes have very strong absorption inhe visible region, allowing the absorption of energy from sunlight.his starts a chain of electron-transfer steps, which ends with anlectron transferred to carbon dioxide. Thus, chlorophyll is at theenter of the photosynthetic oxidation–reduction reaction betweenarbon dioxide and water. Chlorophyll a is the predominant light-bsorbing pigment of photosystem I (PS I), while the phycobilins arehe predominant energy collectors of photosystem II (PS II), pass-ng absorbed energy to the photosynthetic reaction centre throughrelatively small number of chlorophyll molecules [17].

In the present work we have investigated the electron transferrom excited chlorophyll a to the conduction band of TiO2 colloidy using absorption and fluorescence spectroscopy.

. Experimental

.1. Materials

Titanium(IV) 2-propoxide was purchased from Aldrich. Theoubly distilled water was used for preparing the solutions. Alleasurements were performed at room temperature.


The colloidal TiO2 suspension was prepared by the hydrolysis ofitanium(1V) 2-propoxide [18]. Typically, titanium(IV) 2-propoxide153 �l) in 2-propanol (10 ml) was injected by using syringe into0 ml of water with constant stirring under nitrogen atmosphere8 h) it will give 1 × 10−2 M titania stock solution. No stabilizinggents were used during the hydrolysis process. The colloidal sus-ensions of TiO2 prepared by this method were stable for 3–5 days.resh colloidal TiO2 dispersed in water was prepared before eachet of experiments. The stock suspension was diluted with water tobtain the desired concentration of TiO2. No attempts were madeo exclude the traces of 2-propanol (∼0.4%) present in the colloidaliO2 suspension and it was confirmed separately that the presencef 2-propanol did not affect the photochemical measurements asarlier reported [19].

.3. Preparation of chlorophyll

Chlorophyll employed in this study was obtained from theyanobacteria namely Spirulina sp. (marine form) from the cul-ure collections maintained in the Department of Biotechnology,harathidasan University, Tiruchirappalli. Spirulina sp. was cul-ured in ASN III synthetic marine medium [20] at 27 ± 2 ◦C withrtificial illumination from cool white fluorescent lamps. Chloro-hyll was extracted from the freshly harvested biomass. Five gramsf freshly harvested biomass was crushed in a tissue homogenizernd the cell suspension was centrifuged at 10,000 rpm for 10 mint 4 ◦C, the pellet was extracted twice with 90% (v/v) methanol forh at 4 ◦C in dim light. The solution was clarified by centrifugationt 10,000 rpm for 10 min and was stored in dark at 4 ◦C [21]. Theolution was used for further investigations.


The steady-state fluorescence quenching measurements werearried out with JASCO FP-6500 spectrofluorimeter. The excitationavelength of chlorophyll was 480 nm and the emission was at74 nm. The excitation and emission slit widths (each 5 nm) and

Fcd

ta Part A 71 (2009) 1783–1787

can rate (500 nm/min) were maintained constant for all the exper-ments. The samples were carefully degassed using pure nitrogenas for 15 min. Quartz cells (4 cm × 1 cm × 1 cm) with high vac-um Teflon stopcocks were used for degassing. Absorption spectraleasurements were recorded using Cary300 UV-Vis spectropho-

ometer.

.5. Time resolved measurements

Fluorescence lifetime measurements were carried out in aicosecond time correlated single photon counting (TCSPC) spec-rometer. The excitation source is the tunable Ti-sapphire laserTsunami, Spectra Physics, USA). The fluorescence decay was ana-yzed by using the software provided by IBH (DAS-6).


.1. Determination of particle size of colloidal TiO2

The particle size of the prepared colloidal TiO2 has been deter-ined from the relationship between band gap shift (�Eg) and

adius (R) of quantum size particles using the following equation:

Eg = �2h2

2�R2− 1.8e2

ε+ polarization terms (3)

here h is the Planck’s constant, R is the radius of the particle, ε ishe relative permittivity of the semiconductor, �Eg is the band gaphift, the calculated bandgap shift for the colloidal TiO2 is 0.3 eVusing absorption spectrum of colloidal TiO2, spectra not shownere), as compared to bulk anatase, � is the reduced mass of thexciton, i.e., the reduced effective mass of the electron and the hole1/� = l/m∗

c + l/m∗h) in the semiconductor, e is the electron charge,

nd c is the dielectric constant of the semiconductor. The value of= 1.63 me (me is the electron rest mass) [22]. Since the optical

ielectric constant of bulk titanium dioxide is very large (ε = 170),he coloumbic and polarisation terms in the equation are neglected.he calculated size of the prepared colloidal TiO2 is 1.12 nm.


To study the dye-sensitized electron transfer reactions of sensi-izer with semiconductor nanoparticles in the excited state, it isery important to know the type of ground state interaction of

ig. 1. Absorption spectrum of chlorophyll (2 × 10−6 M) in the presence of variousoncentration of colloidal TiO2, [TiO2] = 0, 1, 2, 3, 4 and 5 × 10−4 M. The inset is theependence of 1/(A − A0) on the reciprocal concentration of colloidal TiO2.

A. Kathiravan et al. / Spectrochimica Acta Part A 71 (2009) 1783–1787 1785

Ss

ipcs

pt

Toa

A

wddε(taw

A

A(a

Toer1Fat(

gihbetbotS

Fot

3

TfssT[S

wicIoaa

wooichlorophyll is shown in the insert of Fig. 2. There is a good linearrelationship between 1/(F0 − F) and the reciprocal concentration ofcolloidal TiO2. From the slope, the Kapp has been assessed and thevalue is 1.81 × 104 M−1. The values of Kapp obtained from the data offluorescence quenching matches well with that determined from

cheme 1. Electrostatic interaction of chlorophyll with the positively charged TiO2

urface.

mplies that there is an interaction of colloidal TiO2 with chloro-hyll through carboxyl group (Scheme 1) similar to interaction ofolloidal TiO2 with fluorescein reported by Hilgendorff and Sund-trolm [23].

The equilibrium for the formation of complex between chloro-hyll and colloidal TiO2 is given by Eq. (4), where Kapp representshe apparent association constant:

chlorophyll + TiO2Kapp� chlorophyll· · ·TiO2,

Kapp = [chlorophyll· · ·TiO2][chlorophyll] · [TiO2]

(4)

he changes in intensity of the absorption peak at 480 nm as a resultf formation of the surface complex were utilized to obtain Kapp

ccording to Benesi and Hildebrand equation [24]:

obs = (1 − ˛)C0εchlorophylll + ˛C0εcl (5)

here Aobs is the observed absorbance of the solution containingifferent concentrations of the colloidal TiO2 at 480 nm; ˛ is theegree of association between chlorophyll and TiO2; εchlorophyll andc are the molar extinction coefficients at the defined wavelength�max = 480 nm) for chlorophyll and the formed complex, respec-ively. Eq. (5) can be expressed by Eq. (6), where A0 and Ac are thebsorbances of chlorophyll and chlorophyll complex respectively,ith the concentration of C0:

obs = (1 − ˛)A0 + ˛Ac (6)

t relatively high TiO2 concentrations, ˛ can be equated toKapp[TiO2])/(1 + Kapp[TiO2]). In this case, Eq. (6) can be expresseds Eq. (7):

1Aobs − A0

= 1Ac − A0


(7)

herefore, if the enhancement of absorbance at the wavelengthf 480 nm was due to absorption of surface complex, one wouldxpect a linear relationship between 1/(Aobs − A0) and the recip-ocal concentration of colloidal TiO2 with a slope equal to/Kapp(Ac − A0) and an intercept equal to 1/(Ac − A0). The inset ofig. 1 shows the Benesi and Hildebrand plot for chlorophyll, there isgood linear dependence of 1/(Aobs − A0) on the reciprocal concen-

ration of colloidal TiO2. The value of apparent association constantKapp) determined from this plot is 3.78 × 104 M−1.

The ground state absorption study reveals that the anchoringroup plays an important role in adsorption. The absence of peakn the region of 345 nm (at which bare TiO2 absorbs) even in theigher concentration of TiO2 shows that the sensitization of TiO2y chlorophyll occurs through complex formation and subsequentlectron transfer from the excited state of chlorophyll molecules tohe conduction band of TiO2. The process of electron transfer can
e explained based on the Rehm–Weller equation as well as thexidation potential of excited state chlorophyll with the conduc-ion band potential of TiO2 using energy level diagram as shown incheme 3.
Fb

ig. 2. Fluorescence Quenching of chlorophyll (2 × 10−6 M) in the presence of vari-us concentration of colloidal TiO2, [TiO2] = 0, 1, 2, 3, 4 and 5 × 10−5 M. The inset ishe dependence of 1/(F0 − F) on the reciprocal concentration of colloidal TiO2.

.3. Fluorescence quenching characteristics

Fig. 2 shows the effect of increasing concentration of colloidaliO2 on the fluorescence emission spectrum of chlorophyll. As seenrom the figure, we observed that, addition of colloidal TiO2 to theolution of chlorophyll resulted in the gradual decrease in emis-ion intensity of chlorophyll which indicates the quenching occurs.his quenching behaviour is similar to the studies reported earlier10]. The fluorescence quenching behaviour is usually described bytern–Volmer relation (8):

I0I

= 1 + KSV[Q ] (8)

here I0 and I are the fluorescence intensities of the fluorophoren the absence and presence of quencher, KSV is the Stern–Volmeronstant and [Q] is the quencher concentration. The plot between0/I vs. [TiO2] gave upward curvature suggesting the static naturef quenching as shown in Fig. 3. From fluorescence quenching datalso, the apparent association constant (Kapp) has been calculatedccording to the following Eq. (9):

1F0 − F

= 1F0 − F ′ + 1

Kapp(F0 − F ′)[TiO2](9)

here Kapp is the apparent association constant, F0 is the initial flu-rescence intensity of chlorophyll, F′ is the fluorescence intensityf TiO2 adsorbed chlorophyll and F is the observed fluorescencentensity at its maximum. The plot of 1/(F0 − F) vs. 1/[TiO2] for

ig. 3. Stern–Volmer plot for the fluorescence quenching of chlorophyll (2 × 10−6 M)y colloidal TiO2 in the concentration range of (0–5) × 10−5 M.

1786 A. Kathiravan et al. / Spectrochimica Acta Part A 71 (2009) 1783–1787

tvf

vgl

gtcstss

pcsTr

ieotSS(o

Fs

S

cst

if

3t

tR

�

wiT

Scheme 2. Energy level diagram for chlorophyll and TiO2.

he absorption spectra changes. The good agreement between thesealues of Kapp highlighted the validity of the assumption proposedor the association between chlorophyll and colloidal TiO2.

The decrease in fluorescence emission may be attributed to thearious possibilities such as energy transfer, electron transfer orround state complex formation between the chlorophyll and col-oidal TiO2.

As shown in Scheme 2 the bandgap energy of TiO2 (3.2 eV) isreater than the excited state energy (1.84 eV) of chlorophyll andhere is no overlap between the fluorescence emission spectrum ofhlorophyll with the absorption spectrum of colloidal TiO2 (Fig. 4),o the above two inferences excluded the possibility of energyransfer from chlorophyll to colloidal TiO2. From the above discus-ion we confirmed that the fluorescence quenching shown in Fig. 2hould not be caused by energy transfer.

The possibility of surface complex formation between chloro-hyll and colloidal TiO2 is the reason for quenching of chlorophyll byolloidal TiO2. While increasing the concentration of colloidal TiO2ome of the chlorophyll molecules were adsorbed on the surface ofiO2, so the number of molecules available for the fluorescence iseduced, which is the reason for decrease in fluorescence intensity.

The ability of the excited state chlorophyll to inject its electronsnto the conduction band of TiO2 is determined by energy differ-nce between the conduction band of TiO2 and oxidation potentialf excited state chlorophyll. According to equation Es*/s+ = Es/s+ − Es,
he oxidation potential of excited state chlorophyll is −1.27 vs.CE, where Es/s+ is the oxidation potential of chlorophyll, 0.57 V vs.CE [25] and Es is the excited state energy of chlorophyll, 1.84 eVexcited state energy of the chlorophyll calculated from the flu-rescence maximum based on the reported method [26]). The
ig. 4. The overlap of the (A) absorption spectrum of colloidal TiO2 and (B) emissionpectrum of chlorophyll.

cuihw

3

as

cheme 3. Schematic energy level diagram showing electron transfer process.

onduction band potential of TiO2 is −0.1 eV vs. SCE [27]. Scheme 3uggested that electron transfer from excited state chlorophyll tohe conduction band of TiO2 is energetically favorable.

Therefore we conclude that the fluorescence quenching shownn Fig. 2 is caused by electron transfer. The process of electron trans-er mechanism is shown in Scheme 4.

.4. Calculation of free energy changes (�Get) for the electronransfer reactions

The thermodynamic feasibility of the excited state electronransfer reactions was calculated by employing the well-knownehm–Weller expression [28]:


1/2 − Es + C (10)

here E(ox)1/2 is the oxidation potential of chlorophyll (0.57 V), E(red)

1/2s the reduction potential of TiO2 (i.e.) conduction band potential ofiO2 −0.1 V, Es is the excited state energy of chlorophyll and C is theoulombic term. Since one of the species is neutral and the solventsed is polar in nature, the coulombic term in the above expression

s neglected [29]. The �Get value was calculated as −1.17 eV and thisigher negative �Get value indicates electron transfer processeshich is thermodynamically favorable [30,31].


As shown by previous reports [19,32,33], the dye moleculesdsorbed on the semiconductor particle surface had significantlyhorter fluorescence lifetime than the unadsorbed molecules. The

Scheme 4. Electron transfer quenching process.

A. Kathiravan et al. / Spectrochimica Ac

Fc

flccw9tdlal

esf

k

wakcua

4

cscdliebl

A

0rl

DI

R

[[[

[

[

[[[

[

[[

[

[[[[[[

[[

[30] K. Kikuchi, T. Niwa, Y. Takahashi, H. Ikeda, T. Miyashi, J. Phys. Chem. 97 (1993)

ig. 5. Fluorescence decay curve of (A) chlorophyll (2 × 10−6 M) in presence of (B)olloidal TiO2 (5 × 10−5 M).

uorescence decay of chlorophyll in the absence and presence ofolloidal TiO2 nanoparticles were shown in Fig. 5. In the absence ofolloidal TiO2 nanoparticles, the fluorescence decay of chlorophyllas fitted with single exponential and it shows the lifetime (�) of.1 ns. Upon addition of colloidal TiO2 nanoparticles (5 × 10−4 M),he decay of chlorophyll was deviated from single exponentialecay and fitted with bi-exponential. It shows both the shorter-

ifetime and longer-lifetime components. The shorter lifetime isttributed to that of adsorbed chlorophyll (�ads = 1.8 ns) and theonger one is for free/unadsorbed chlorophyll (� = 7.2 ns).

The observed decrease in lifetime could be correlated with thelectron transfer processes in the dye molecules adsorbed on theemiconductor nanoparticles and it may be correlated by using theollowing equation:

et = 1�ads

− 1�

(11)

here � and �ads are the lifetime of free chlorophyll molecules inqueous solution and adsorbed on the colloidal TiO2 surface andet is the specific rate of electron transfer process from excited statehlorophyll to the conduction band of TiO2. By substituting the val-es of � and �ads in the above Eq. (11) the value of ket were calculatednd is found to be 4.2 × 108 s−1.

. Conclusion

Chlorophyll adsorbs on the surface of colloidal TiO2 through itsarboxyl group, as evidenced by the absorption and fluorescencepectroscopy. The apparent association constant (Kapp) has beenalculated according to absorption and fluorescence quenchingata, both values are in same order of magnitude. Quenching fol-

ows static mechanism through ground complex formation whichs confirmed by the curvature of Stern–Volmer plot. The process oflectron transfer from excited state chlorophyll to the conductionand of TiO2 has been confirmed by the decrease in fluorescence

ifetime and the rate of electron injection has also been calculated.

[

[[

ta Part A 71 (2009) 1783–1787 1787

cknowledgements

R.R. and A.K. thanks CSIR, Government of India (Ref. No.1(2217)/08/EMR-II, dt.06/05/2008) for the project and fellowship,espectively. Authors thankful to Laser Spectra of Services, Banga-ore for lifetime measurements.

M.C. and S.S. acknowledge the financial support extended by theepartment of Science and Technology, New Delhi, Government of

ndia, in the form of a project research grant with fellowship.

eferences

[1] P.V. Kamat, Chem. Rev. 93 (1993) 267–300.[2] M.R. Hoffmann, S.T. Martin, W.Y. Choi, D.W. Bahnemann, Chem. Rev. 95 (1995)

69–96.[3] T. Wu, T. Lin, J. Zhao, H. Hidaka, N. Serpone, Environ. Sci. Technol. 33 (1999)

1379–1387.[4] B. O’Regan, M. Gratzel, Nature 353 (1991) 737–740.[5] H.P. Zhang, Y.L. Zhou, M.H. Zhang, T. Shen, Y.L. Li, D.B. Zhu, J. Colloid Interface

Sci. 264 (2003) 290–295.[6] C. Chen, X. Qi, B. Zhou, J. Photochem. Photobiol. A: Chem. 109 (1997) 155–158.[7] H. Mao, H. Deng, H. Li, Y. Shen, Z. Lu, H. Xu, J. Photochem. Photobiol. A: Chem.

114 (1998) 209–212.[8] T. Wu, S.J. Xu, J.Q. Shen, S. Chen, M.H. Zhang, T. Shen, J. Photochem. Photobiol.

A: Chem. 137 (2000) 191–196.[9] Z.X. Zhou, S.P. Qian, S.D. Yao, Z.Y. Zhang, Dyes Pigments 51 (2001) 137–144.10] Z. Zhou, S. Qian, S. Yao, Z. Zhang, Radiat. Phys. Chem. 65 (2002) 241–248.11] A. Kathiravan, R. Renganathan, Spectrochim. Acta 71 (2008) 1080–1083.12] A. Kathiravan, V. Anbazhagan, M.A. Jhonsi, R. Renganathan, Z. Phys. Chem. 221

(2007) 941–948.13] A. Kathiravan, V. Anbazhagan, M.A. Jhonsi, R. Renganathan, Spectrochim. Acta

A 70 (2008) 615–618.14] T.D. Brock, Evolutionary and ecological aspects of the cyanophytes, in: N.G.

Whitton (Ed.), The Biology of Cyanobacteria, Blackwell Scientific, Oxford, 1973,pp. 487–500.

15] L.J. Rothschild, R.L. Mancinelli, Nature 345 (1990) 710–712.16] W. Vermaas, Rev. Plant Physiol. Plant Mol. Biol. 44 (1993) 457–482.17] J.P. Thornber, J.P. Markwell, S. Reinman, Photochem. Photobiol. 29 (1979)

1205–1216.18] D. Bahnemann, A. Henglein, J. Lilie, L. Spanhel, J. Phys. Chem. 88 (1984)

709–711.19] P.V. Kamat, J.P. Chauvet, R.W. Fessenden, J. Phys. Chem. 90 (1986) 1389–1394.20] P. Rippka, J. Deruelles, J.B. Waterbury, M. Herdman, R.Y. Stanier, J. Gen. Microbiol.

111 (1979) 1–61.21] J.F. Talling, D. Driver, in: M. Doty (Ed.), Primary Productivity Measurement,

Marine and Freshwater, Atomic Energy Commission, Washington, DC, 1963,p. 142.

22] E. Joselevich, I. Willner, J. Phys. Chem. 98 (1994) 7628–7635.23] M. Hilgendorff, V. Sundstrolm, J. Phys. Chem. B 102 (1998) 10505–10514.24] H.A. Benesi, J.H. Hildebrand, J. Am. Chem. Soc. 71 (1949) 2703–2707.25] K. Hasegawa, T. Noguchi, Biochemistry 44 (2005) 8865–8872.26] E.J. Shin, D. Kim, J. Photochem. Photobiol. A: Chem. 152 (2002) 25–31.27] S.L. Murov, I. Carmichael, G.L. Hug, Handbook of Photochemistry, second ed.,

Marcel-Dekker, Inc., New York, 1993, pp. 269–273.28] G.J. Kavarnos, N.J. Turro, Chem. Rev. 86 (1986) 401–449.29] S. Parret, F.M. Savary, J.P. Fouassier, P. Ramamurthy, J. Photochem. Photobiol. A:

Chem. 83 (1994) 205–209.

5070–5073.31] S. Nath, H. Pal, D.K. Palit, A.V. Sapre, J.P. Mittal, J. Phys. Chem. 102A (1998)

5822–5830.32] P.V. Kamat, J. Phys. Chem. 93 (1989) 859–864.33] P.V. Kamat, M.A. Fox, Chem. Phys. Lett. 102 (1983) 379–384.

This article was published in an Elsevier journal. The attached copyis furnished to the author for non-commercial research and

education use, including for instruction at the author’s institution,sharing with colleagues and providing to institution administration.

Other uses, including reproduction and distribution, or selling orlicensing copies, or posting to personal, institutional or third party

websites are prohibited.

In most cases authors are permitted to post their version of thearticle (e.g. in Word or Tex form) to their personal website orinstitutional repository. Authors requiring further information

regarding Elsevier’s archiving and manuscript policies areencouraged to visit:

http://www.elsevier.com/copyright


Author's personal copy

Available online at www.sciencedirect.com

Journal of Photochemistry and Photobiology A: Chemistry 193 (2008) 204–212

Investigation on the fluorescence quenching of2,3-diazabicyclo[2.2.2]oct-2-ene (DBO) by

certain estrogens and catechols

V. Anbazhagan, V. Kandavelu, A. Kathiravan, R. Renganathan ∗School of Chemistry, Bharathidasan University, Tiruchirappalli 620 024, Tamil Nadu, India

Received 16 February 2007; received in revised form 15 June 2007; accepted 21 June 2007Available online 26 June 2007

Abstract

The fluorescence quenching of singlet excited 2,3-diazabicyclo[2.2.2]oct-2-ene (DBO), a fluorescent probe for antioxidants, by various estrogensand certain catechols has been investigated. The time resolved and steady state fluorescence quenching experiments were conducted in acetonitrileand dichloromethane. The bimolecular quenching rate constant (kq) values are in the range of 107–109 M−1 s−1 for estrogens and 109–1010 M−1 s−1

for catechols. In the case of estrogens, a direct hydrogen atom abstraction is proposed, while exciplex induced quenching becomes competitivefor catechols with electron donating (ED) substituents. The quenching mechanism was analysed on the basis of exciplex formation, deuteriumisotopic effects and cyclicvoltammetric studies. Further, the invitro-antioxidant activity of estrogens and catechols were evaluated with rat livercatalase by gel electrophoresis. The data suggest the involvement of hydrogen atom transfer in the fluorescence quenching of DBO by estrogensand catechols.© 2007 Elsevier B.V. All rights reserved.

Keywords: 2,3-Diazabicyclo[2.2.2]oct-2-ene; Fluorescence quenching; Estrogens; Catechols

1. Introduction

The photochemical reactions of ketones with n,�* states[1] resembling simple alkoxyl radicals [2–4] in their reactiv-ity are well studied. Such reactive radicals are responsible forcellular damage in biological systems [5]. Antioxidants play avital role in medicine, biology, polymer chemistry, cosmeticsand in food industry. By intercepting oxidizing species, pre-dominantly reactive radicals, they prevent cellular damage andpolymer or food degradation. The important aim in such researchareas is the quantification of their reactivity [6,7]. The reactiv-ity of t-butoxyl (t-BuO•) radical towards antioxidants [8,9] hasbeen studied by transient absorption spectroscopy. Alternatively,the reactivity of n,�* triplet-excited ketones, mostly benzophe-none (Ph2CO), has been analysed by laser flash photolysis toobtain direct information on antioxidant reactivity [8,10]. Thehydrogen abstraction by n,�*-excited chromophores, in par-


ticular ketones, is the most extensively studied photoreaction[11]. The mechanism of charge transfer processes [12,13] isvital to the understanding of many photochemical and photobi-ological reactions. The reactions of singlet-excited states, whichoccur often on faster time-scales, and the reactions of other n,�*chromophores except the carbonyl group (C O), which may beof minor practical relevance, have received comparatively lessattention. Some aspects of hydrogen abstraction reactions ofn,�*-excited states have appeared recently [14–16]. Of late thereis great interest on fluorescent probe for biomolecules [17–19].However, the fluorescence lifetime of common fluorophores aretypically in the range of several nanoseconds or even less. Thisis too short to monitor nanosecond-to-microsecond processes,which is important to understand the functions and dynamics ofmany biomolecules.

The azoalkane, 2,3-diazabicyclo[2.2.2]oct-2-ene (DBO) hasan extremely long fluorescence lifetime (up to 1 �s) [20]. Itsn,�*-excited states behave in a radical-like way, thus the reac-tivity resembles simple alkoxyl radicals [2–4,21] and has beenextensively employed as a fluorescent probe for antioxidants[22–24]. DBO shows a pronounced tendency to undergo direct

1010-6030/$ – see front matter © 2007 Elsevier B.V. All rights reserved.doi:10.1016/j.jphotochem.2007.06.026


V. Anbazhagan et al. / Journal of Photochemistry and Photobiology A: Chemistry 193 (2008) 204–212 205

Table 1Fluorescence quenching rate constants and thermodynamic data of DBO by estrogens

S. No. Quencher (Estrogens) kq/(109 M−1 s−1)a D EOX1/2 vs. SCE (V)b �Get (eV)c

A B C

1 DES 1.38 [1.46] 1.03 [1.18]d 1.05 1.46 1.83 1.272 EST 0.70 [0.75] 0.47 [0.82]e 0.55 1.91 1.42 0.863 EED 0.65 [0.69] 0.44 [0.18]f 0.39 1.08 1.57 1.014 BEST 0.77 [0.83] 0.43 0.58 1.72 1.44 0.885 EEDME 0.09 [0.09] 0.09 0.06 1.12 1.60 1.04

a Bimolecular fluorescence quenching rate constants (A) measured by steady state method in CH3CN (τ = 620 ns); values in brackets are measured by time resolvedmethod in CH3CN, (B) measured by steady state method in CH2Cl2 (τ = 185 ns), (C) measured by steady state method in wet acetonitrile (10% H2O) (D) deuteriumisotopic effect measured by steady state fluorescence quenching in CH3CN containing 10% H2O or D2O, i.e. [kq (O H)/kq (O D)].

b Irreversible oxidation potential of estrogens in V vs. SCE in CH3CN.c Calculated by Rhem–Weller equation �Get = EOX(D) − Ered (A) − E* + C, the reduction potential of DBO is −2.8 V vs. SCE, E* = 3.3 eV and C = −0.06 (for

acetonitrile). Error ±10%.d Data for model compounds from [20].e Data for model compounds from [20].f Data for model compound from [48].

hydrogen abstraction reactions [25–28] with alcohols, ethers,alkanes, phenols and alkylbenzenes. Taking advantage of thisproperty, DBO has been used in this work to probe the reactivityof the mentioned estrogens and catechols.

Steroidal estrogens have been reported to function as antioxi-dants and free radical scavengers under a variety of experimentalconditions. For example, phenolic and catecholic estrogensprevent lipid peroxidation induced by diverse pro-oxidantsin microsomes [29–31] and liposomes [32,33]. They exerttheir protective effects to different extents with different effi-ciencies depending on the phenolic and catecholic structureof the molecule [5]. In the present work, we have stud-ied the fluorescence quenching of DBO by various estrogenssuch as estrone (EST), �-estradiol (BEST), diethylstilbesterol(DES), 17-�-ethynylestradiol (EED), ethynylestradiol-3-methyl

ether (EEDME) as well as certain catechols such as catechol(CAT), tetrachlorocatechol (TCC), 3,5-di-tert-butylcatechol(DTBC), 4-nitrocatechol (4-NC), 4-tert-butylcatechol (4-TBC),3-methylcatechol (3-MC). The quenching rate constants (kq) ofDBO were determined by both steady state and time resolvedfluorescence technique in acetonitrile and dichloromethane(Tables 1 and 2). Rehm–Weller equation was applied to measurefree energetics for electron transfer process.

2. Experimental

2.1. Materials

DBO was obtained as a gift sample from Prof. W.M.Nau, International University of Bremen, Germany. The estro-gens, catechols and D2O were obtained from Sigma–Aldrich,USA. Potassium ferricyanide, ferric chloride, H2O2, KH2PO4,K2HPO4, glycerol and Tris-buffer were obtained from Merck,India. The spectroscopic grade solvents, CH3CN and CH2Cl2were used for preparing the solutions. Fresh rat liver tissuehomogenate was used as enzyme source for gel electrophoresis.All measurements were performed at ambient temperature.

Table 2Fluorescence quenching rate constants and thermodynamic data of DBO by catechols

S. No. Quencher (Catechols) kq/(109 M−1 s−1)a D EOX1/2 vs. SCE (V)b �Get (eV)c

A B C

1 CAT 1.10 [1.19] 0.92 0.95 1.85 0.95 0.392 TCC 2.64 [2.77] 1.77 2.13 2.10 1.36 0.803 DTBC 2.42 [2.36] 4.10 2.08 1.45 1.40 0.844 4-NC 43.6 [40.1] 42.9 42.2 1.62 2.09 1.535 4-TBC 1.73 [1.57] 2.83 1.39 1.32 1.28 0.726 3-MC 1.24 [1.40] 1.95 1.02 1.19 1.17 0.61

a Bimolecular fluorescence quenching rate constants (A) measured by steady state method in CH3CN (τ = 620 ns); values in brackets are measured by time resolvedmethod in CH3CN, (B) measured by steady state method in CH2Cl2 (τ = 185 ns), (C) measured by steady state method in wet acetonitrile (10% H2O) (D) deuteriumisotopic effect measured by steady state fluorescence quenching in CH3CN containing 10% H2O or D2O, i.e. [kq (O H)/kq (O D)].

b Irreversible oxidation potential of catechols in V vs. SCE in CH3CN.c Calculated by Rhem–Weller equation �Get = EOX(D) − Ered (A) − E* + C, the reduction potential of DBO is −2.8 V vs. SCE, E* = 3.3 eV and C = −0.06 (for

acetonitrile). Error ±10%.


206 V. Anbazhagan et al. / Journal of Photochemistry and Photobiology A: Chemistry 193 (2008) 204–212

2.2. Methods

2.2.1. Fluorescence quenching experimentsSamples were prepared by dissolving DBO (ca. 0.1 mM) in

proper (acetonitrile and dichloromethane) solvents and admin-istering the appropriate amounts of estrogens and catechols witha GC syringe. The samples were carefully degassed using purenitrogen gas for 30 min. Quartz cells (4 cm × 1 cm × 1 cm) withhigh vacuum Teflon stopcocks were used for degassing. Forthe (Stern–Volmer) quenching plots, various concentrations ofquenchers were chosen. Lifetimes and fluorescence intensitieswere obtained for different quencher concentration and plottedaccording to the equation: τ0/τ = I0/I = 1 + kq τ0 [Q]. The slopesafforded the bimolecular quenching rate constant (kq). The life-time (τ0) of DBO in degassed acetonitrile and dichloromethanewithout any added quencher was 620 ns and 185 ns, respectively.Deuterium isotope effects for estrogens and catechols were mea-sured in wet acetonitrile (either with 10% H2O or D2O) assuminga complete and fast OH/OD exchange of hydrogen.

2.2.1.1. Steady-state measurements. The steady-state fluo-rescence quenching measurements were carried out in aPerkin-Elmer LS55 Luminescence spectrometer. The excita-tion wavelength was 365 nm and the emission was monitored at420 nm. The excitation and emission slit widths (15 nm) and scanrate (500 nm) were maintained constant for all the experiments.

2.2.1.2. Time-resolved experiments. Fluorescence lifetimemeasurements were carried out in a picosecond time correlatedsingle photon counting (TCSPC) spectrometer. The excitationsource is the tunable Ti-sapphire laser (TSUNAMI, SpectraPhysics, USA). The diode laser pumped millennia V (SpectraPhysics) CW Nd-YVO4 laser was used to pump the sapphirerod in the Tsunami mode locked picosecond laser (SpectraPhysics). The diode laser output was used to pump the Nd-YVO4 rod in the Millennia. The DBO was excited by the laserpulse at 365 nm. The time resolved fluorescence emission wasmonitored at 420 nm. The emitted photons were detected bya MCP-PMT (Hamamtsu R3809U) after passing through themonochromator (f/3). The laser source is operated at 4 MHzand the signal from the photodiode is used as a stop signal.The signal from the MCP-PMT is used as start signal in orderto avoid the dead time of the TAC. The difference betweenthe start and stop signal is due to the time taken by the pulsestraveling through the cables and electronic relaxation of theexcited state. The data analysis was carried out by the softwareprovided by IBH (DAS-6). The kinetic trace was analyzed bynon-linear least square fitting of monoexponential function.

2.2.2. Cyclic voltammetric measurementsThe reduction potential for DBO was reported as −2.8 V

versus SCE [34] in acetonitrile. In the present study, the oxi-dation potential of estrogens and catechols were measured inacetonitrile with tetrabutylammonium perchlorate (0.1 M) aselectrolyte. The experimental setup consisted of a platinumworking electrode, a graphite counter electrode and a silver ref-erence electrode. Irreversible peak potentials were measured at

different scan rates (50–100 mV/s). All values are reported rela-tive to ferrocene as internal standard. All samples were deaeratedby bubbling with pure nitrogen gas for ca. 5 min at room tem-perature.

2.2.3. Invitro-antioxidant activityLiver sample (100 mg/ml buffer) was homogenized in 50 mM

phosphate buffer (pH 7.0); and then centrifuged at 10,000 rpmfor 15 min; the supernatant thus obtained was used for this exper-iment as a source of catalase enzyme. This experiment wasdesigned as follows:

Sample 1 contains liver homogenate (30 �l), sample 2 con-tains liver homogenate (30 �l) and 10 �l of H2O2 (30%).Samples 3–8 contain 30 �l of liver homogenate, 10 �l of H2O2(30%) and 10 �l of antioxidant compounds as DES, EST, EED,BEST and EEDME, respectively. All samples were shaken welland incubated at 37 ◦C for 30 min. After the incubation period,equal volume (50 �l) of sample buffer (7.25 ml distilled water,1.25 ml of 0.5 M Tris buffer (pH 6.8) and 1 ml of glycerol) wereadded to all the samples and were subjected to non-denaturingpolyacrylamide gel electrophoresis (Native – PAGE).

Non-denaturing polyacrylamide gel electrophoresis was per-formed essentially as described by Laemmli [35], except thatSDS was omitted from all the buffers and the samples were notboiled before electrophoresis. Electrophoretic separation wasperformed on 8% gel at 4 ◦C with a constant power supply of50 V for stacking gel and 100 V for separating gel.

Catalase activity was detected by the method of Woodbury etal. [36]. The gel was soaked in 5 mM H2O2 solution for 10 minand was washed with water and stained with a reaction mixturecontaining 1% potassium ferricyanide (w/v) and 1% ferric chlo-ride. The enzyme appeared as a yellow band superimposed on adark green background. The reaction was terminated by addingwater and the gel was photographed at once. Quantification of theenzyme bands was performed by a densitometer (GS-300 trans-mittance/reflectance scanning densitometer, Hoefer ScientificInstruments, USA).


The absorption and emission spectrum of DBO are shown inFig. 1. The structures of the quenchers used in the present studyare given below.

Fig. 1. Absorption (a) and emission (b) spectrum of DBO.



3.1. Fluorescence quenching of DBO by estrogens

The time resolved and steady state quenching of DBO byestrogens provide the same quenching rate constants (withinexperimental error) as shown in Figs. 2 and 3. The Stern–Volmerquenching plots obtained by the plot of τ0/τ or I0/I versus [Q]were linear, indicating that the quenching process is dynamic innature. As can be seen from literature [20], estrogens possessless kq values compared with that of glutathione, ascorbic acid,uric acid and �-tocopherol. From the kq values given in theTable 1, the following trend is observed in acetonitrile among

Fig. 2. Stern–Volmer plots for the time resolved fluorescence quenching of DBO(1 × 10−4 M) by estrogens in acetonitrile.

the estrogens:

DES > BEST ≈ EST ≈ EED > EEDME

In the estrogen structure, the hydroxyl group attached to thearomatic ring is an important feature for the contribution of theantioxidant activity. DES possesses highest kq among the estro-gens because DES contains two monohydroxylated benzenerings (phenolic units) separated by an ethane bridge. Almostsimilar reactivities among BEST, EST and EED are in accordwith the fact that phenolic estrogens possess similar antioxidantactivities [37]. These three estrogens possess only one phenolic

Fig. 3. Stern–Volmer plots for the steady state fluorescence quenching of DBO(1 × 10−4 M) by estrogens in acetonitrile.



group unlike DES. Even though BEST and EED possess oneadditional hydroxyl group in the “D” ring, the kq values remainalmost the same. Hence, the hydroxyl group in the “D” ringdoes not have any significant effect on the antioxidant activity.While analyzing the kq values of BEST, EST and EED, only aslight variation among these three compounds could be identi-fied. EEDME possesses one order lower magnitude kq, becausethe reactive 3-hydroxyl group of the aromatic ring is replaced by

OCH3 group, which proves the importance of aromatic O Hhydrogen in the fluorescence quenching of DBO by estrogens. Itmay be noted that the antioxidant activity of the estrogens wasalready established based on the reduction of the blue–green2,2′-azinobis-(3-ethylbenzothiazoline-6-sulphonicacid) radicalcation (ABTS•+) as indicated by the suppression of its character-istic absorption at 734 nm [38]. Such dependence of antioxidantactivity on the availability of OH groups has been reportedin the case of flavonoids [39] which also possess more thanone aromatic ring just like estrogens. The similar trend withlow kq values has been observed in the case of low polardichloromethane solvent. The kq decreases with decreasingsolvent polarity [40,49,50], which clearly indicates that thequenching mechanism of singlet excited DBO involves hydro-gen atom abstraction. In the present study, the observed trendssuggest that the phenolic ring of estrogens could be a crucialcomponent for their antioxidant activity.

3.2. Fluorescence quenching of DBO by catechols

The fluorescence quenching experiments were conducted indichloromethane and acetonitrile which have different polari-ties. The Stern–Volmer plots (Figs. 4 and 5) clearly suggest theabsence of any static component in quenching. Table 2 providesthe fluorescence quenching rate constant for catechols and showsthe following trend for quenching rate constants in acetonitrile:

4 − NC > TCC > DTBC > 4 − TBC > 3 − MC > CAT

Among catechols, 4-NC shows an unusually high quench-ing rate constant, which is not expected from the electronwithdrawing nature of nitro group. To get more insightinto the high quenching rate constant, 4-nitrobenzene wasused as a quencher to infer the importance of hydroxylgroups. It was found that there is no significant reactivity

Fig. 4. Stern–Volmer plots for the time resolved fluorescence quenching of DBO(1 × 10−4 M) by catechols in acetonitrile.

Fig. 5. Stern–Volmer plots for the steady state fluorescence quenching of DBO(1 × 10−4 M) by catechols in acetonitrile.

difference between 4-nitrocatechol (43.6 × 109 M−1 s−1) and4-nitrobenzene (37.8 × 109 M−1 s−1) which speaks against theinvolvement of aromatic hydroxyl group in hydrogen abstrac-tion. Energy transfer from the excited DBO to 4-NC presents anefficient competitive quenching mechanism in this case [28]. Allthe substituted catechols show higher kq than the parent catechol.The kq values of DTBC, 4-TBC and 3-MC were reduced in highpolar solvent (acetonitrile) and increased in low polar solvent(dichloromethane), that is, inverted solvent effect is observed[40]. It is presumed that the initial reaction step involves theformation of an exciplex with partial charge transfer charac-ter [11,41–43]. The exciplex is stabilized to a lesser degree bypolar solvents than the reactants. These catechols may undergothrough the exciplex formation mechanism as electron donat-ing groups tend to stabilize the partial development of positivecharge in the aromatic ring. The electron donating groups gener-ally tend to increase the electron density of the aromatic ring by+I effect which might facilitate exciplex formation [Scheme 2]mediated hydrogen atom transfer. The parent catechols and TCChave shown increase in the kq values with increasing solventpolarity from dichloromethane to acetonitrile. Hence, these cat-echols may undergo direct hydrogen abstraction mechanism.

The DTBC (2.42 × 109 M−1 s−1) shows higher kq than 4-TBC, 3-MC and catechol, which reveals that the fluorescencequenching of DBO increases quite systematically with increasein the number of alkyl groups due to electron donating ability. Allcatechols show slightly higher kq value than those of estrogens.The above results infer that catechols are potential antioxidantsthan estrogens. From these observations, it is proposed that cate-chols may undergo two different mechanisms (Schemes 1 and 2)based on electronic effects of the substituents on the aromaticring.

3.3. Deuterium isotope effects

Fluorescence quenching of DBO by estrogens and catecholsshows substantial deuterium isotope effects in (10% H2O orD2O) CH3CN/H2O and CH3CN/D2O mixtures (Tables 1 and 2).A small but significant deuterium isotope effect was observedfor estrogens between 1.08 and 1.91, which are characteristic forhydrogen abstraction close to diffusion controlled limit [14]. Theparent catechol has isotope effect of 1.89 and quenching rate con-



Scheme 1. Fluorescence quenching mechanism of DBO through a direct hydrogen atom transfer.

stant of 1.10 × 109 M−1 s−1. A high isotope effect is observedfor TCC (2.10) with relatively high quenching rate constant(2.64 × 109 M−1 s−1). This possibly suggests a direct hydrogenabstraction leading to dominant quenching. The low values ofisotope effect for DTBC, 4-TBC and 3-MC (catechols with elec-tron donating substitutions) with relatively high quenching rateconstants compared to that of the parent catechol justifies com-petitive quenching through exciplex formation. The deuteriumisotope effects provide evidence that hydrogen abstraction isinvolved in the quenching process of singlet excited DBO byestrogens and catechols.

3.4. Invitro-antioxidant activity

Antioxidant activity of various estrogens and catechols weredetermined by the electrophoretic method. An analysis ofthe electrophoretic pattern of catalase enzyme in the liverhomogenate revealed one band with variations in the stainingintensity of the band among the seven samples of estrogens(Fig. 6) and eight samples of catechols (Fig. 7), the lane 1(only liver homogenate) exhibited a single band of high inten-sity (band area 230), while that of lane 3– 7 exhibited a band ofalmost similar intensity in estrogen compounds (Fig. 6a) (bandarea 226.00, 217.68, 216.82, 218.16 & 212.98 respectively) andin catecholic compounds (Fig. 7a) (band area 213.90, 220.12,218.97, 228.14, 217.85 and 216.72, respectively). However, lane2 (liver homogenate damaged by H2O2) revealed a single bandwith less intensity (band area 211.29).

Among the estrogen compounds tested DES possess highantioxidant activity whereas compounds EST, EED and BESTshow similar activity. However, EEDME has low activity thanthe other compounds. In the case of catechol compounds, 4-NC possesses high antioxidant activity whereas the compoundsTCC, DTBC, 4-TBC and 3-MC possess antioxidant activity inthe following order: TCC > DTBC > 4-TBC > 3-MC. However,CAT has low activity than the other catechol compounds. Theability to scavenge free radicals by the estrogen and catecholcompounds is correlated with the number of hydroxyl groupbound with the aromatic ring and their structure, position ofaromatic OH ring.

Catalase is an antioxidant enzyme, which is present in mostcells and catalyses the decomposition of H2O2 to water andO2. In the present study the enzyme catalase was inactivated byH2O2 (lane 2) and then activated by the estrogen and catecholcompounds (Figs. 6 and 7). This might be due to the potentialantioxidant activity of the estrogens and catechols. The H2O2could be scavenged by estrogen and catechol compounds. Thismight be the reason for increasing intensity of the bands in lanes3– 8. This result strongly indicates the potential antioxidantactivity of studied estrogens and catechol compounds.

3.5. Thermodynamic Properties of Estrogens and Catechols

The oxidation potential of estrogens and catechols are givenin the Tables 1 and 2, respectively. The relevant mechanis-tic pathways are illustrated in Schemes 1 and 2. The question



Scheme 2. Exciplex-induced quenching of DBO by electron donating (ED) substituents of catechols.

whether the photoreduction operates through sequential elec-tron and proton transfers or by direct H-atom abstraction fromthe estrogens and catechols by the excited azoalkane can onlybe answered speculatively on the basis of the free energetics forelectron transfer estimated by the Rehm–Weller equation [44]

i.e., ΔGet = EOX(D) − Ered(A) − E∗ + C

where EOX (D) is the oxidation potential of donor, Ered (A) isthe reduction potential of acceptor, E* is the excitation energyof the acceptor and C is the coulomb term which describes theelectrostatic attraction within the contact ion pair and has a valueof ca.— 0.06 eV in acetonitrile. The calculated �Get values arelisted in Tables 1 and 2.

The electrochemical data imply positive �Get values forexcited 1DBO with estrogens and catechols (�Get values inTables 1 and 2). The variation in the oxidation potential ofestrogens and catechols arises due to their structure. Amongthe estrogens studied, DES possesses higher oxidation potentialbecause of its readily oxidisable nature and resonance stabi-lization. EED and EEDME possess almost similar oxidation

potential due to its similarity in structure (presence of C CHgroup). EST and BEST shows lower oxidation potential.

Among catechols, 4-nitrocatechol shows highest Eox (2.09 Vversus SCE) due to its electron withdrawing nature of thenitro group (inductive effect). The alkyl substituted catecholsshow higher Eox values than the parent catechol. Among thealkyl substituted catechols the following trend is noticeable:DTBC > 4-TBC > 3-MC. The Eox increases with increasingnumber of alkyl group. The driving force for electron transfer issignificantly endergonic for estrogens and catechols. An ender-gonic thermodynamics reveals that quenching through hydrogenabstraction may be a possible mechanism.

The overall mechanistic pathways of the photoreduc-tion of DBO by estrogens and catechols are illustrated inSchemes 1 and 2. The photoinduced electron transfer can beexcluded on the basis of endergonic energetics. The quench-ing may occur along two pathways: (i) the formation of anexcited state charge transfer complex (exciplex) or (ii) directhydrogen abstraction. In the transition state for direct hydro-gen abstraction, polar resonance structure becomes important,



Fig. 6. Invitro-antioxidant activity of estrogens by gel electrophoresis (a) densit-ometric pattern of catalase enzyme from liver homogenate treated with estrogens.

which accounts for the possible observation of polar substituenteffects in the hydrogen abstraction. In the case of estrogens,parent catechols, and TCC, quenching involves direct hydrogenabstraction. Excited state hydrogen abstraction not only may beachieved by direct abstraction, but also may occur from exci-plex (mediated abstraction) involving lone pair coordination.Catechols with electron donating substituents involve hydro-gen abstraction through formation of exciplex. Alternatively,the exciplex may induce deactivation to the ground state. Hydro-gen atom abstraction from the phenolic hydroxyl group is theproposed quenching mechanism. The estrogen phenoxyl radicalformed after the donation of hydrogen atom to DBO would berelatively long-lived and stabilized by internal delocalization ofthe electron deficiency around its aromatic structure.

The same quenching mechanism of singlet excited DBOhas been characterized in theoretical and experimental detailwith respect to the involvement of conical intersections alongthe quenching pathways [11,25,34,45]. Quenching of DBOproceeds via hydrogen atom transfer [26,44,46] as does the scav-enging of reactive radicals by vitamin E in vitro [47]. Fromthe quenching rate constant and endergonic thermodynamics,

Fig. 7. Invitro-antioxidant activity of catechols by gel electrophoresis (a) densit-ometric pattern of catalase enzyme from liver homogenate treated with catechols.

the fluorescence quenching of DBO by estrogens and catecholsoccurs through a reaction mechanism that “mainly” involveshydrogen atom from the hydroxyl group. The experimentaldata obtained from various experiments i.e., quenching stud-ies (steady state and time resolved), cyclic voltammetric studiesand gel electrophoresis studies are fully consistent with thisinterpretation.

3.6. Tentative mechanism

Scheme 1.

4. Conclusions

The compound DBO is quenched by estrogens and catecholsin acetonitrile and dichloromethane and the corresponding kq(fluorescence quenching rate constant) depends on the availabil-ity of OH group in the aromatic ‘A’ ring of the estrogens. Theinvolvement of exciplexes of DBO and catechols with electron



donating substituents, i.e. partial charge transfer intermediates,is suggested by the observation of an “inverted solvent effect”.The trend in quenching efficiency and antioxidant capacitywas similar in all the three methods studied. It clearly sug-gests that the quenching of singlet excited DBO by estrogensand catechols involves hydrogen atom abstraction based onthe experimental data obtained from time resolved and steadystate fluorescence quenching technique, cyclic voltammetric andinvitro-antioxidant activity studies.

Acknowledgments

R.R thanks DST (Government of India) for the ResearchProject (ref: SP/S1/H-41/2001, dated 12-09-2002) and V.Anbazhagan thanks DST for the fellowship. We thank Prof.W.M. Nau, School of Engineering and Science, InternationalUniversity Bremen, Germany for the DBO sample. We arethankful to Profs. P. Natarajan and P. Ramamoorthy, NCUFP,University of Madras, Chennai and Laser Spectra ServicesIndia Pvt. Ltd., Bangalore for the lifetime measurement facility.Authors also thank DST-FIST and UGC-SAP for spectroflu-orimeter facility in the School. We thank Prof. P. Geraldine,Department of Animal Science, Bharathidasan University forgel electrophoresis study.

References

[1] N.J. Turro, Modern Molecular Photochemistry, University Science Books,Mill Valley, 1991.

[2] C. Walling, M.J. Gibian, J. Am. Chem. Soc. 87 (1965) 3361–3364.[3] H. Paul, R.D. Small, J.C. Scaiano, J. Am. Chem. Soc. 100 (1978)

4520–4527.[4] D. Griller, J.A. Howard, P.R. Marriott, J.C. Scaiano, J. Am. Chem. Soc.

103 (1981) 619–623.[5] M.B. Ruiz-Larrea, C. Martin, R. Martinez, R. Navarro, M. Lacort, N.J.

Miller, Chem. Phys. Lipids 105 (2000) 179–188.[6] H. Sies (Ed.), Oxidative Stress, Oxidants and Antioxidants, Academic

Press, New York, 1991.[7] P.W. Washko, R.W. Welch, K.R. Dhariwal, Y. Wang, M. Levine, Anal.

Biochem. 204 (1992) 1–14.[8] C. Evans, J.C. Scaiano, K.U. Ingold, J. Am. Chem. Soc. 114 (1992)

4589–4593.[9] L. Valgimigli, J.T. Banks, K.U. Ingold, J. Lusztyk, J. Am. Chem. Soc. 117

(1995) 9966–9971.[10] M.V. Encinas, E.A. Lissi, A.F. Olea, Photochem. Photobiol. 42 (1985)

347–352.[11] U. Pischel, W.M. Nau, J. Phys. Org. Chem. 13 (2000) 640–647.[12] R.A. Marcus, Angew. Chem. 105 (1993) 1161–1172.[13] H. Kurreck, M. Huber, Angew. Chem. 107 (1995) 929–947.[14] W.M. Nau, F.L. Cozens, J.C. Scaiano, J. Am. Chem. Soc. 118 (1996)

2275–2282.[15] W. Adam, J.N. Moorthy, W.M. Nau, J.C. Scaiano, J. Org. Chem. 61 (1996)

8722–8723.

[16] W. Adam, J.N. Moorthy, W.M. Nau, J.C. Scaiano, J. Am. Chem. Soc. 119(1997) 5550–5555.

[17] A.P. Demchenko, Biochem. Biophys. Acta 1209 (1994) 149–164.[18] V. Muhoz, P.A. Thompson, J. Hofrichter, W.A. Eaton, Nature 390 (1997)

196–199.[19] J.R. Lakowicz, R. Nair, G. Piszczek, I. Gryczynski, Photochem. Photobiol.

71 (2000) 157–161.[20] W.M. Nau, J. Am. Chem. Soc. 120 (1998) 12614–12618.[21] A. Padwa, Tetrahedron Lett. 5 (1964) 3465–3468.[22] X. Zhang, C. Erb, J. Flammer, W.M. Nau, Photochem. Photobiol. 71 (2000)

524–533.[23] G. Gramlich, J. Zhang, W.M. Nau, J. Am. Chem. Soc. 124 (2002)

11252–11253.[24] G. Gramlich, J. Zhang, W.M. Nau, J. Am. Chem. Soc. 126 (2004)

5482–5492.[25] D. Klapstein, U. Pischel, W.M. Nau, J. Am. Chem. Soc. 124 (2002)

11349–11357.[26] W.M. Nau, G. Greiner, J. Wall, H. Rau, M. Olivucci, M.A. Robb, Angew.

Chem. Int. Ed. 37 (1998) 98–101.[27] U. Pischel, W.M. Nau, Photochem. Photobiol. Sci. 1 (2002) 141–147.[28] U. Pischel, D. Patra, A.L. Koner, W.M. Nau, Photochem. Photobiol. 82

(2006) 310–317.[29] H. Wiseman, B. Halliwell, FEBS Lett. 332 (1993) 159–163.[30] M. Lacort, A.M. Leal, M. Liza, C. Matin, R. Martinez, M.B. Ruiz-Larrea,

Lipids 30 (1995) 141–146.[31] T. Miura, S. Muraoka, T. Ogiso, Steroids 61 (1996) 379–383.[32] K. Sugioka, Y. Shimosegawa, M. Nakano, FEBS Lett. 210 (1987) 37–

39.[33] H. Wiseman, M. Cannon, H.R.V. Arnstein, B. Halliwel, FEBS Lett. 274

(1990) 107–110.[34] U. Pischel, X. Zhang, B. Hellrung, E. Haselbach, P.A. Muller, W.M. Nau,

J. Am. Chem. Soc. 122 (2000) 2027–2034.[35] U.K. Laemmli, Nature 227 (1970) 680–685.[36] W. Woodbury, A.K. Spencer, M.A. Stahmann, Anal. Biochem. 44 (1971)

301–305.[37] N.J. Miller, C.A. Rice-Evans, M.J. Davis, V. Gopinathan, N.J. Milner, Clin.

Sci. 84 (1993) 407–412.[38] N.J. Millar, C.A. Rice-Evans, Redox Rep. 2 (1996) 161–171.[39] K. Mukai, W. Oka, K. Watanabe, Y. Egawa, S. Nagaoka, J. Terao, J. Phys.

Chem. A 101 (1997) 3746–3753.[40] W.M. Nau, U. Pischel, Angew. Chem. Int. Ed. 38 (1999) 2885–2888.[41] S. Inbar, H. Linschitz, S.G. Cohen, J. Am. Chem. Soc. 103 (1981)

1048–1054.[42] J.B. Guttenplan, S.G. Cohen, J. Am. Chem. Soc. 94 (1972) 4040–4042.[43] P.J. Wagner, A.E. Kemppainen, J. Am. Chem. Soc. 91 (1969) 3085–

3087.[44] D. Rehm, A. Weller, Ber Bunsen–Ges, Phys. Chem. 73 (1969) 834–

839.[45] A. Sinicropi, U. Pischel, R. Basosi, W.M. Nau, M. Olivucci, Angew. Chem.

Int. Ed. 39 (2000) 4582–4586.[46] W.M. Nau, G. Greiner, H. Rau, M. Olivucci, M.A. Robb, Ber. Bunsen-Ges.

Phys. Chem. 102 (1998) 486–492.[47] G.W. Burton, K.U. Ingold, J. Am. Chem. Soc. 103 (1981) 6472–6477.[48] C.V. Lundgren, A.L. Koner, M. Tinkl, U. Pischel, W.M. Nau, J. Org. Chem.

71 (2006) 1977–1983.[49] C. Coenjarts, J.C. Scaiano, J. Am. Chem. Soc. 122 (2000) 3635–3641.[50] W.M. Nau, G. Greiner, H. Rau, J. Wall, M. Olivucci, J.C. Scaiano, J. Phys.

Chem. A 103 (1999) 1579–1584.

Z. Phys. Chem. 221 (2007) 929–939 / DOI 10.1524/zpch.2007.221.7.929© by Oldenbourg Wissenschaftsverlag, München

Fluorescence Quenching Studyon Electron Transfer from Certain Aminesto Excited State Triphenylpyrylium Ion (TPP+)

By V. Anbazhagan, A. Kathiravan, M. Asha Jhonsi, and R. Renganathan∗School of Chemistry, Bharathidasan University, Tiruchirappalli-620 024, Tamil Nadu,India

(Received February 8, 2007; accepted March 5, 2007)

2,4,6-Triphenylpyrylium Tetrafluoroborate / Fluorescence Quenching /Photoinduced Electron Transfer / Amines

The fluorescence quenching of excited singlet state of 2,4,6-triphenylpyrylium tetrafluo-roborate (TPPBF4 or TPP+), a very good electron acceptor by amines were investigatedin a acetonitrile solution using steady state technique. The bimolecular quenching rateconstants lie in the range 2.11–10.26×1010 M−1 s−1. The driving force (∆G et) for electrontransfer process was calculated from the oxidation potential of amines and the reductionpotential of TPP+. The observed kq values correlated well with the driving force for theelectron transfer reactions. Aromatic amines show higher kq than aliphatic amines. Fromthe oxidation potential of amines and the quenching rate constant values, a mechanisminvolving photoinduced electron transfer from amines to excited state TPP is suggested.

1. IntroductionPhotoinduced electron transfer (PET) processes have attracted growing interestin the last decade [1–5]. This leads to a better understanding of the involvedmechanism. In particular, PET reactions such as cycloaddition, cyclorever-sions, oxygenations and fragmentations have been extensively studied [6]. Insuch a process, amines are excellent electron donors because of their low ion-ization potentials. The lone pair electrons on the nitrogen atom lead to facileelectron transfer reactions of amines in a number of chemical, electrochem-ical, photochemical and biochemical redox processes [7–9]. Electron transferreactions of amines are also important in certain technological applicationssuch as imaging [10] and photopolymerization [11] etc. In addition, the elec-tron donating capability of amino functionality has been extensively used for

* Corresponding author. E-mail: [email protected]


930 V. Anbazhagan et al.

designing new materials such as organic conductors [12], electro luminescentmaterials [13], photovoltaics [14] and materials with nonlinear optical activ-ity [15]. The donor properties of amines are of considerable practical interestfor photoreduction of ketones [16–19], azoalkanes [20–22], dyes as well as forpolymer technology, where amines serve both as photoinitiators and light sta-bilizers. In addition, fluorescence quenching by amines has been the subject ofnumerous mechanistic investigations [23–30]. Obviously, aliphatic amines do-nate electrons through their lone pair orbital and are expected to interact withexcited states in a different manner than aromatic donors. The formation ofamine cation radical, the intermediate was observed by several groups [34–37, 45].

2,4,6-Triphenylpyrylium salts are easily photoexcited by visible light,and the photosensitizing behavior of these species has been widely appliedin photoinduced electron transfer (PET) processes [31]. Among these salts,2,4,6-triphenylpyrylium tetrafluoroborate (TPPBF4) has received most atten-tion. TPPBF4 is a very strong oxidant in both singlet and triplet excited state,capable to oxidize several organic donors (D) to the corresponding radicalcations (D•+) with concomitant formation of triphenylpyranyl radical (TPP•).The neutral character of TPP• reduces the significance of the competing back-electron-transfer (BET) process [32], thus favoring the follow-up reactionsof D•+.

Considerable efforts have been made by several groups to explain themechanism of fluorescence quenching by amines. Fluorescence quenching ofTPP by benzene derivatives were thoroughly studied by Jayanthi et al. [33].Thanasekaran et al. [34] studied the PET quenching of rhenium (I) rectangleswith amines very recently. An electron transfer reaction of amines with elec-tron acceptor such as triplet C60 was studied by Arbogast et al. in 1991 [35].Nau reported the fluorescence quenching of azoalkanes by amines, photore-duction of azoalkanes through exciplex formation [36]. Taking the advan-tage of such properties, we aim to investigate the mechanism of fluorescencequenching of 2,4,6-triphenylpyrylium tetrafluoroborate by electron donors.The kq values are correlated with oxidation potential and ionization potentialof amines.

Fluorescence Quenching Study on Electron Transfer from Certain Amines . . . 931

2. Experimental2.1 Materials

TPP was obtained from Sigma and used as such. Amines such as TMPBand TMBD were purchased from Aldrich. Other amines were obtained fromMerck, which were further purified by distillation and recrystallization beforeuse. The spectroscopic grade solvent, CH3CN was used for preparing the solu-tions. All measurements were performed at room temperature (24 ◦C).

2.2 Methods

2.2.1 Fluorescence quenching experiments

The steady state fluorescence quenching measurements were carried out inPerkin Elmer LS55 Luminescence spectrometer. The excitation wavelengthwas 355 nm and the emission was monitored at 460 nm. The excitation andemission slit width (15 nm), scan rate (500 nm) were constantly maintained forall the experiments. Absorption spectra were recorded using cary300 UV-Visspectrophotometer. Samples were prepared by dissolving TPP (ca. 0.01 mM) inacetonitrile and administering the appropriate amounts of amines with a GC sy-ringe. The samples were carefully degassed using pure nitrogen gas for 30 min.Quartz cells (4×1×1 cm) with high vacuum Teflon stopcocks were used fordegassing.

3. Results and discussionThe fluorescence quenching of TPP by a variety of amines was carried outin acetonitrile medium. The absorption and emission spectrum of TPP areshown in Fig. 1. To check the ground state complex formation between TPP

Fig. 1. Absorption (a) and emission (b) spectrum of TPP.


and amines, we recorded the absorption spectra of a mixture of TPP and amineusing concentration similar to those used in the quenching studies. If there isany ground state charge transfer complex formation, a new absorption peakshould appear from the tail end of the absorption spectrum of TPP. No evi-dence was obtained for ground state complex formation between amines andTPP (Fig. 3). The quenching rate constant and free energy change for electrontransfer processes data are given in Table 1.

The emission intensity of TPP was effectively quenched by amines (Fig. 2)and the quenching rate constants were analyzed in terms of Stern–Volmer equa-tion.

I0/I = 1+ KSV[Q] = 1+ kqτ0[Q] ,where I0 and I are the emission intensities in the absence and presence ofamines, [Q] represents concentration of amines, KSV is the Stern–Volmer con-stant, kq is the bimolecular quenching rate constant and τ0 represents lifetimeof TPP in the absence of amines, i.e. 4.2 ns in acetonitrile [39]. Plot of I0/Ivs. [Q] were linear (Figs. 6 and 7) for all quenchers, indicating the dynamicnature of the quenching process. The quenching rate constants decrease in thefollowing trend: TMBD (2,3,5,6-tetramethyl-p-phenylenediamine) > TMPD(N,N,N ′ ,N ′-tetramethyl benzidine) > TPA (triphenylamine) > DPA (diphenyl-amine) > Aniline > TEOA (triethanol amine) > TEA (triethyl amine) >


Fig. 2. Steady state fluorescence quenching of TPP (1 × 10−6 M) with TMPD in theconcentration range of (a) 0, (b) 1 × 10−3 , (c) 2× 10−3 , (d) 3× 10−3 , (e) 4× 10−3 and(f) 5×10−3 M in acetonitrile.

Fig. 3. Absorption spectrum of TPP in the presence of TPA in acetonitrile solution.

DEA (diethyl amine). Among these, aromatic amines possess highest kq thanaliphatic amines, because the former may donate electrons from their aro-matic rings (π-donors), while later act as pure “n-donors”. The quenching rateconstant increases with the decrease of ionization potential values, which sup-port the above trend. The quenching rate constant (kq) can be correlated withthe free energy change for the electron transfer (∆G et) and is given by theRehm–Weller equation [37],

i.e.,∆G et = EOX(D)− E red(A)− E* +C ,

where EOX(D) is the oxidation potential of donor, E red (A) is the reduction po-tential of acceptor, E* is the excitation energy of the acceptor and C is thecoulomb term which describes the electrostatic attraction within the contact ion


Table 1. Bimolecular rate constants and thermodynamics for the quenching of the singletexcited TPP with amines.

S. No. Quencher IP/eVa kq/ E1/2 vs. SCE ∆G et

(Amines) (109 M−1 s−1)b (V)c (eV)d

1 DEA 8.01 2.11 1.31 −1.172 TEA 7.50 2.61 0.96 −1.523 TEOA 7.35 2.85 1.06 −1.424 Aniline 7.72 4.07 0.87 −1.615 DPA 7.16 5.38 0.84 −1.646 TPA 6.86 5.92 0.85 −1.637 TMPD 6.20 9.35 0.38 −2.108 TMBD 5.85 10.26 0.32 −2.16

a Adiabatic ionization potentials taken from [41, 42]. b Determined by steady state fluo-rescence quenching in CH3CN (τ = 4.2 ns), c irreversible oxidation potential of amines,[43–47] and d calculated by Rehm–Weller equation ∆G et = EOX(D)− E red(A)− E* +C,the reduction potential of TPP is −0.38 eV, E* = 2.8 eV and C = −0.06 (for acetonitrile).

Fig. 4. Plot of log kq vs. the oxidation potential of amines for the oxidation of amines byTPP.

pair and has a value of ca. – 0.06 eV in acetonitrile. The calculated ∆G et valuesare listed in Table 1.

Among the aromatic amines TMBD and TMPD shows approximately twotimes higher kq than other aromatic amines. This is due to the fact that,TMBD and TMPD possess two amino (donor) groups in the para position witheach other, which leads to easy delocalization of electrons. TMBD (10.26×1010 M−1 s−1) shows higher kq than TMPD due to the presence of one ad-ditional phenyl ring possessing π-electrons, which may facilitate the donorstrength. The same trend was also found in the case of TPA and DPA. The


Fig. 5. Plot of log kq vs. IP for the fluorescence quenching of TPP by the amines.

Fig. 6. Stern-Volmer plots for the reductive quenching of TPP with amines (TEA, aniline,DPA and TMBD) in acetonitrile.

kq value of TMPD (9.35×1010 M−1 s−1) shows more than twice than that ofaniline (4.07 × 1010 M−1 s−1), because TMPD possess one additional aminogroup and four methyl groups, which may enhance the electron donating abil-ity. Aliphatic amines having higher oxidation potential and ionization potentialthan aromatic amines, show lower kq values than aromatic amines. Among thealiphatic amines the order of kq value is TEOA > TEA > DEA, the trend showsTEOA possesses higher kq than TEA and DEA, because electron donating ca-pability is increased due to the presence of alcoholic group (+M effect) in theformer case. Among the TEA and DEA, TEA (2.61×1010 M−1 s−1) possessessomewhat higher kq value than DEA (2.11×1010 M−1 s−1). This is due to thefact that electron density is more on nitrogen atom in the TEA because of + Ieffect.


Fig. 7. Stern-Volmer plots for the reductive quenching of TPP with amines (DEA, TEOA,TPA and TMPD) in acetonitrile.

The quenching of TPP by amines can be explained by a number of pos-sible mechanisms such as electron transfer, energy transfer and hydrogen atomtransfer. Since the triplet energy of amines (3.1 eV) [38] is above the excita-tion energy of TPP (2.8 eV) [39], electronic energy transfer is not operativehere. A mechanism involving hydrogen atom transfer from amines to TPP canalso be excluded because the TPA (which has no abstractable α C–H hydrogen)also efficiently quenches the luminescence of TPP. Even though DPA possessone abstractable hydrogen atom, the kq is lower than TPA. The results fromthe fluorescence quenching rate constant and oxidation potential values indi-cates that the effect of amines on the luminescence quenching of TPP must beassociated with only a dynamic quenching process through electron transfer(Scheme 1).

Scheme 1. Photoinduced electron transfer mechanism of luminescence quenching of TPPby amines.


The photoinduced electron transfer mechanism of luminescence quench-ing of TPP by amines are shown in scheme 1. The parameter k12 and k21 arethe diffusion controlled rate constants for the formation and dissociation of en-counter complex respectively. k23 and k32 are the forward and reverse electrontransfer rate constants, k34 is the sum of all the rate constants causing the disap-pearance of ion-pair state. According to the scheme the excited state acceptor(TPP) collide with ground-state donor (Amine) to form an encounter complex1[TPP+*. . . . . . Amine]. This encounter complex then undergoes a reorganiza-tion to reach the transition state where electron transfer takes place from donorto acceptor to give an ion-pair species [TPP•. . . . . . Amine•+]. This ion-pairspecies may undergo dissociation into either (i) formation of TPP radical andamine radical cation or (ii) TPP cation and amine (Scheme 1).

To better understand the nature of the quenching process, the bimolecu-lar quenching rate constants were correlated with the free energy change ofthe electron transfer processes. From Table 1, amines with lower oxidation po-tential exhibit higher quenching rate constant, this trend perfectly supports theelectron transfer quenching process. In addition, the plot of log kq vs. oxida-tion potential of amines is linear (Fig. 4), which proves an additional supportfor electron transfer mechanism.

4. ConclusionThe fluorescence quenching of singlet excited TPP by various amines werecarried out in acetonitrile. The efficiency of the quenching rate constants isfound to be higher than those of TPP quenched by benzene derivatives [33] andthe kq values remain diffusion-controlled limit (∼ 1010 M−1 s−1) with highernegative ∆G et values. The donor strength of an amine is an important factorwhich governs the kinetics of quenching of TPP. A good correlation among thelogarithmic quenching rate constants, adiabatic ionization potential and oxida-tion potential of amines is observed (Figs. 4 and 5). From the exergonic (high−∆G et) thermodynamics and quenching rate constant, the electron transferfrom amines to singlet excited TPP is suggested.

Acknowledgement

V.A. thanks DST for the fellowship and R.R. thanks DST (Government of In-dia) for the Research Project (Ref: SP/S1/H-41/2001, dt: 12-09-2002).

References

1. S. L. Mattes and S. Farid, Org. Photochem. 6 (1983) 233.2. G. J. Kavarnos and N. Turro, J. Chem. Rev. 86 (1986) 401.3. M. A. Fox, Adv. Photochem. 13 (1986) 237.4. J. Mattay, Angew. Chem., Int. Ed. Engl. 26 (1987) 825.


5. M. A. Fox and M. Chanon, Photoinduced Electron Transfer. Parts A-D Elsevier,Amsterdam (1988).

6. M. A. Miranda and H. Garcia, Chem. Rev. 94 (1994) 1063.7. (a) D. Chiapperino, S. McIlroy, and D. E. Falvey, J. Am. Chem. Soc. 124 (2002)

3567; (b) M. Goto, K. Otsuka, X. Chen, Y. Tao, and M. Oyama, J. Phys. Chem. A108 (2004) 3980.

8. (a) J. Stubbe and W. A. van der Donk, Chem. Rev. 98 (1998) 705; (b) N. Leven-tis, J. Yang, E. F. Fabrizio, A. M. M. Rawashdeh, W. S. Oh, and C. Sotiriou-Leventis,J. Am. Chem. Soc. 126 (2004) 4094.

9. (a) N. Fujieda, A. Satoh, N. Tsuse, K. Kano, and T. Ikeda, Biochemistry 43 (2004)10 800; (b) L. B. Anderson, A. J. A. Ouellette, J. E. Rye, M. Maderia, M. J. Mac-Coss, J. R. Yates III, and B. A. Barry, J. Am. Chem. Soc. 126 (2004) 8399.

10. (a) R. D. Theys and G. Sosnovsky, Chem. Rev. 97 (1997) 83; (b) S. K. Bergstrom,N. Edenwall, M. Laven, I. Velikyan, B. Langstrom, and K. E. Markides, Anal.Chem. 77 (2005) 938.

11. L. J. Mathias, B. S. Shemper, M. Alirol, and J. F. Morizur, Macromolecules 37(2004) 3231.

12. (a) P. Novak, K. Muller, K. V. S. Santhanam, and O. Haas, Chem. Rev. 97 (1997)207; (b) P. Bandyopadhyay, P. Bandyopadhyay, and S. L. Regen, BioconjugateChem. 13 (2002) 1314.

13. (a) J. Kido, M. Kimura, and K. Nagai, Science 275 (1996) 1267; (b) H. Mochizuki,T. Hasui, M. Kawamoto, T. Ikeda, C. Adachi, Y. Taniguchi, and Y. Shirota, Macro-molecules 36 (2003) 3457.

14. (a) H. Imahori, M. Kimura, K. Hosomizu, T. Sato, T. K. Ahn, S. K. Kim, D. Kim,Y. Nishimura, I. Yamazaki, Y. Araki, O. Ito, and S. Fukuzumi, Chem. Eur. J. 10(2004) 5111; (b) J. Kruger, R. Plass, M. Gratzel, P. J. Cameron, and L. M. Peter,J. Phys. Chem. B 107 (2003) 7536.

15. (a) F. Wurthner, S. Yao, T. Debaerdemaeker, and R. Wortmann, J. Am. Chem. Soc.124 (2002) 9431; (b) B. J. Coe, J. A. Harris, B. S. Brunschwig, J. Garin, J. Orduna,S. J. Coles, and M. B. Hursthouse, J. Am. Chem. Soc. 126 (2004) 10 418.

16. S. G. Cohen, A. Parola, and G. H Parsons, Jr., Chem. Rev. 73 (1973) 141.17. J. C. Scaiano, J. Photochem. 2 (1973) 81.18. D. Griller, J. A. Howard, P. R. Marriott, and J. C. Scaiano, J. Am. Chem. Soc. 103

(1981) 619.19. H. Miyasaka and N. Mataga, Bull. Chem. Soc. Jpn. 63 (1990) 131.20. W. M. Nau, Ber. Bunsen-Ges. Phys. Chem. 102 (1998) 476.21. W. Adam, J. N. Moorthy, W. M. Nau, and J. C. Scaiano, J. Org. Chem. 61 (1996)

8722.22. W. Adam, J. N. Moorthy, W. M. Nau, and J. C. Scaiano, J. Am. Chem. Soc. 119

(1997) 6749.23. H.-J. Timpe, S. Jockusch, and K. Körner (Eds.), Dye Sensitized Photopolymeriza-

tion. Vol. II, Elsevier, London (1993) 575.24. O. Brede, D. Beckert, C. Windolph, and H. A. Göttinger, J. Phys. Chem. A 102

(1998) 1457.25. E. N. Step, N. J. Turro, M. E. Gande, and P. P. Klemchuk, J. Photochem. Photobiol.

A: Chem. 74 (1993) 203.26. H. Knibbe, D. Rehm, and A. Weller, Ber. Bunsen-Ges. Phys. Chem. 73 (1969) 839.27. N. J. Turro and R. Engel, J. Am. Chem. Soc. 91 (1969) 7113.28. W. Hub, S. Schneider, F. Dörr, J. D. Oxman, and F. D. Lewis, J. Am. Chem. Soc. 106

(1984) 701.29. K. Tominaga, S. Yamauchi, and N. Hirota, J. Phys. Chem. 92 (1988) 5160.30. P. Jacques, X. Allonas, D. Burget, E. Haselbach, P.-A. Muller, A.-C. Sergenton, and

H. Galliker, Phys. Chem. Chem. Phys. 1 (1999) 1867.


31. B. Branchi, M. Bietti, G. Ercolani, M. A. Izquierdo, M. A. Miranda, and L. Stella,J. Org. Chem. 69 (2004) 8874.

32. (a) M. A. Miranda and H. Garcia, Chem. Rev. 94 (1994) 1063; (b) A. Sanjuan,M. N. Pillai, M. Alvaro, and H. Garcia, Chem. Phys. Lett. 341 (2001) 153; (c)R. Akaba, M. Iwasaki, T. Matsumura, M. Kamata, and H. Itoh, J. Phys. Org. Chem.9 (1996) 187; (d) R. Akaba, M. Kamata, H. Sakuragi, and K. Tokumaru, Tetrahe-dron Lett. 33 (1992) 8105; (e) R. Akaba, H. Sakuragi, and K. Tokumaru, J. Chem.Soc., Perkin Trans. 2 (1991) 291.

33. S. S. Jayanthi and P. Ramamurhty, J. Phys. Chem. A 101 (1997) 2016.34. P. Thanasekaran, R.-T. Liao, B. Manimaran, Y.-H. Liu, P.-T. Chou, S. Rajagopal,

and K.-L. Lu, J. Phys. Chem. 110 (2006) 10 683.35. J. W. Arbogast, C. S. Foote, and M. Kao, J. Am. Chem. Soc. 114 (1992) 2277.36. (a) U. Pischel, X. Zhang, B. Hellrung, E. Haselbach, P.-A. Muller, and W. M. Nau,

J. Am. Chem. Soc. 122 (2000) 2027; (b) U. Pischel and W. M. Nau, J. Phys. Org.Chem. 13 (2000) 640.

37. D. Rehm and A. Weller, Ber. Bunsen-Ges. Phys. Chem. 73 (1969) 834.38. V. Wintgens, J. Pouliquen, J. Kossanyi, and M. Heintz, New. J. Chem. 10 (1986)

345.39. A. P. Darmanyan, W. Lee, and W. S. Jenks, J. Phys. Chem. A 103 (1999) 2705.40. Y. Kuriyama, T. Arai, H. Sakuragi, and K. Tokumaru, Chem. Lett. 17 (1988) 1193.41. S. G. Lias, J. E. Bartmess, J. F. Liebman, J. L. Holmes, R. D. Levin, and W. G. Mal-

lard, J. Phys. Chem. Ref. Data 17 (1988) Suppl. 1.42. R. C. Weast, D. R. Lide, M. J. Astle, and W. H. Beyer (Eds.), CRC Handbook of

Chemistry and Physics. CRC Press, Boca Raton, FL (1990).43. W. Adam, J. N. Moorthy, W. M. Nau, and J. C. Scaiano, J. Am. Chem. Soc. 119

(1997) 6749.44. W. Adam and A. Nikolaus, J. Am. Chem. Soc. 122 (2000) 884.45. J.-M. Chen, T.-I. Ho, and C.-Y. Mou, J. Phys. Chem. 94 (1990) 2889.46. K. Sasaki, A. Kitani, and M. Tsuboi, Nippon Kagaku Kaishi 12 (1973) 2269.47. A. P. Darmanyan and W. S. Jenks, J. Phys. Chem. A 102 (1998) 7420.

Is

MS

a

ARRA

KPFP

1

uissp[

yiibeTvaawb[ifg[

1d





nteraction between certain porphyrins and CdS colloids: A steadytate and time resolved fluorescence quenching study

. Asha Jhonsi, A. Kathiravan, R. Renganathan ∗

chool of Chemistry, Bharathidasan University, Tiruchirappalli 620 024, Tamil Nadu, India

r t i c l e i n f o

rticle history:eceived 14 March 2008

a b s t r a c t

The interaction between porphyrins namely, meso-tetrakis (4-methoxyphenyl)porphyrin (TMeOPP), pro-

eceived in revised form 29 April 2008ccepted 12 May 2008

eywords:orphyrins

toporphyrin IX (PPIX) and Zinc(II) meso-tetraphenylporphyrin (ZnTPP) with colloidal CdS has beenstudied by using steady state and time resolved fluorescence quenching measurements. The porphyrinsadsorbed on the surface of colloidal CdS due to electrostatic interaction. This adsorption leads to changesin the absorption spectra related to the complex formation. The apparent association constant (Kapp)was in the order of 4.34–5.58 × 105 M−1 from the effect of colloidal CdS on the absorption spectra and0.64–1.6 × 105 M−1 from fluorescence quenching data. Quenching is attributable mainly to static mecha-

e com

de[chotdtps

Cscca

2

2

luorescence quenchingorphyrins with CdS colloids nism through ground stat

. Introduction

Photosensitization of stable, large band-gap semiconductorssing dyes in the visible light has been the subject of active

nvestigation [1,2]. These types of reactions were successfullytudied by Hagfeldt and Gratzel [3]. Colloidal suspensions of theemiconductor nanoparticles are luminescent, and the spectralroperties of these colloids depended on the size of colloids4].

There are molecules, whose interaction with nanoparticlesields surface-modified nanoparticles, which results in chang-ng their optical properties [5]. Datta et al. reported the possiblenteraction of cadmium-enriched CdS nanoparticles with certainiomolecules like tyrosine [6] and tryptophan [7]. The band-gapnergy of CdS (Eg = 2.6 eV) [8] is lower than the band-gap ofiO2 (3.2 eV) [9] corresponding to the absorption of CdS in theisible region. Cadmium sulphide has applications as photocat-lyst [10] and non-linear optical material [11] in solar cells [12]nd in display devices [13]. The interaction of CdS nanoparticlesith bovine serum albumin in reversed micelles has been studied

y Liang Tan et al. [14].Porphyrins are common photosensitizers15]. Chemical reactivity between a porphyrin and other molecule
nvolves either direct docking/binding of the two compounds or theormation of a reactive intermediate (e.g., proton, activated oxy-en, etc.) that moves between the two non-complexed molecules16]. The spectrophotometric characteristics of the porphyrin are

p(pudwh


plex formation as confirmed by lifetime measurements.© 2008 Elsevier B.V. All rights reserved.

ictated by binding or interaction with another molecule and arexpected to alter the porphyrin spectrophotometric characteristics17]. Porphyrins are of great importance in the fields of catalytichemistry, biochemistry and photochemistry [18]. Several groupsave studied physical and chemical processes of the porphyrinsn colloid surfaces [19]. Electronic coupling of porphyrins withhe semiconductor is certainly one of the key parameters for theesign of efficient sensitizers [20]. In our previous work we studiedhe fluorescence quenching of meso-tetrakis (4-sulfonatophenyl)orphyrin by colloidal TiO2 [21]. Structure of porphyrins used arehown below:

Bhamro et al. studied the photochemistry of ZnTPP induced bydS nanoparticles in 2-propanol [22]. In this present work, we havetudied the interaction between substituted porphyrins with CdSolloids as a quencher. It is expected that this study might lead tolues on how to use the colloidal CdS nanoparticles for widespreadpplications in biochemical research.

. Experimental

.1. Materials

Protoporphyrin IX (PPIX), meso-tetrakis (4-methoxyphenyl)orphyrin (TMeOPP) and Zinc(II) meso-tetraphenylporphyrinZnTPP) were obtained form Sigma and used without further
urification. Cadmium chloride (CdCl2) purchased from Fluka wassed as such. N,N′-dimethylformamide (DMF) was distilled underiminished pressure prior to use, and hydrogen sulphide gasas generated in a Kipp’s apparatus from ferrous sulphide andydrochloric acid.


dx.doi.org/10.1016/j.saa.2008.05.011

1 ica Acta Part A 71 (2008) 1507–1511

2

2

ctisAtot

2

accetwr

TAea

S

123

2

KogttrU

bt

�

w

508 M.A. Jhonsi et al. / Spectrochim

.2. Instrumentation

.2.1. Steady state and time resolved measurementsThe steady state fluorescence quenching measurements were

arried out with JASCO FP-6500 spectrofluorometer. The excita-ion and the emission wavelengths for all porphyrins are shownn Table 1. The excitation and emission slit width (each 5 nm) andcan rate (500 nm/min) were kept constant for all the experiments.bsorption spectra were recorded using Cary 300 UV–vis spec-

rophotometer. Fluorescence lifetime measurements were carriedut in a picosecond time correlated single photon counting (TCSPC)echnique.

.2.2. Cyclic voltammetric measurementsIn the present study, the oxidation potential of TMeOPP, ZnTPP

nd PPIX were measured in DMF with tetrabutylammonium per-hlorate (TBAP, 0.1 M) as electrolyte. The experimental setup
onsisted of a platinum working electrode, a glassy carbon counterlectrode and a silver reference electrode. Reversible peak poten-ials were measured at different scan rates (0.05 V/s). All samplesere deaerated by bubbling with pure nitrogen gas for ca. 5 min at
oom temperature.

able 1bsorption (�ex) and emission (�em) wavelengths of the porphyrins and appar-nt association constant (Kapp) of porphyrins with colloidal CdS calculated frombsorption and fluorescence study

. No. Sensitizer �ex (nm) �em (nm) Kapp × 105 M−1

Absorbance Fluorescence

ZnTPP 425 606 5.58 1.60PPIX 407 634 4.34 0.67TMeOPP 421 657 4.51 0.64

mso

ntas

3

3

o

.3. Preparation of CdS colloids

CdS colloids were prepared by the procedure reported byhanna et al. [23]. For the typical preparation, 0.02 M DMF solutionf CdCl2 was taken in a round bottom flask and hydrogen sulphideas was passed for a few seconds with continuous stirring at roomemperature in air. The reaction mixture was stirred at the sameemperature for a few minutes to obtain colorless (<5 min) or fluo-escent yellow (>5 min) solutions. The solutions were analyzed byV–vis spectroscopy to identify the presence of nanoparticles.

The diameter of the particles determined from the relationshipetween band-gap shift (�Eg) and radius (R) of quantum size par-icles using the following equation.

Eg = �2h2

2R2

[1

me∗ + 1mh∗

]− 1.8e2

εR+ Polarisation

terms(1)

here h is Planck’s constant, R is the radius of the particle, me* andh* are the effective masses of the e− and h+, respectively, in the

emiconductor, e is the electron charge, ε is the relative permittivityf the semiconductor.

We used this equation to estimate the size of colloidal CdSanoparticles. A value of 0.153 me was used for the reduced effec-ive mass of the exciton (1/� = 1/me + 1/mh) of CdS, the coloumbicnd polarisation terms in the equation are neglected. The particleize of the prepared colloidal CdS is 3.35 nm.



Fig. 1 shows the normalized absorption and emission spectraf the porphyrins (TMeOPP, PPIX and ZnTPP). The absorption and

M.A. Jhonsi et al. / Spectrochimica Acta Part A 71 (2008) 1507–1511 1509

Fig. 1. Normalized absorption and emission spectra of porphyrins in DMF (1, 2,3 are absorption and 1a, 2a, 3a are emission spectra of PPIX, ZnTPP and TMeOPP,respectively).

FCl

esitssiiipssep

Fid

Fid

s

P

K

Bce

A

wd(apficta

A

[l

ig. 2. Absorption spectrum of TMeOPP (�ex 421 nm) in the presence of colloidaldS in the concentration range of 0–5 × 10−3 M in DMF. The insert is the straight

ine dependence of 1/Aobs–A0 on the reciprocal concentration of CdS colloid.

mission wavelengths are shown in Table 1. To study the dye-ensitized electron transfer reaction in the excited state, it is verymportant to know the type of interaction of the dye molecule whenhey adsorb on the nanoparticle surface. Figs. 2–4 show the steadytate absorption spectra of porphyrins in DMF with colloidal CdSolution at different concentrations. It has been observed that withncreasing CdS concentration, the absorbance increases, and theres neither a spectral shift nor a new peak was noticed. These changesn the absorption spectral features clearly indicate that the por-hyrins adsorbed on the surface of colloidal CdS to form a ground
tate complex. Similar adsorption of sensitizers on colloidal CdSurface was reported earlier [24]. In the present system porphyrinsxists in two forms, unadsorbed porphyrin monomer and adsorbedorphyrin · · · CdS.
ig. 3. Absorption spectrum of PPIX (�ex 407 nm) in the presence of colloidal CdSn the concentration range of 0–5 × 10−3 M in DMF. The insert is the straight lineependence of 1/Aobs–A0 on the reciprocal concentration of CdS colloid.

4lcat1cs

3

pcZtsi

ig. 4. Absorption spectrum of ZnTPP (�ex 421 nm) in the presence of colloidal CdSn the concentration range of 0–5 × 10−3 M in DMF. The insert is the straight lineependence of 1/Aobs–A0 on the reciprocal concentration of CdS colloid.

We can express equilibrium between the adsorbed and unad-orbed porphyrins by the following equation,

orphyrin + CdSKapp� Porphrin · · · CdS (2)

app = [Porphyrin · · · CdS][Porphyrin] · [CdS]

(3)

y the well known Benesi and Hildebrand method [25] we can cal-ulate the apparent association constant Kapp using the followingquation:

obs = (1 − ˛)C0εporphyrinl + ˛C0εcl (4)

here Aobs is the observed absorbance of the solution containingifferent concentrations of colloidal CdS at 421, 407 and 425 nmthe absorbance band maximum of the porphyrins TMeOPP, PPIXnd ZnTPP, respectively); ˛ is the degree of association betweenorphyrins and CdS; εporphyrin and εc are the molar extinction coef-cients at the defined wavelengths for porphyrins and the formedomplex, respectively. Eq. (4) can be expressed as Eq. (5), wherehe A0 and Ac are the absorbances of porphyrins and the complext 421, 407 and 425 nm, respectively, with the concentration of C0:

obs = (1 − ˛)A0 + ˛Ac (5)

At relatively high CdS concentrations, ˛ can be equated to (Kapp

CdS])/(1 + Kapp[CdS]). In this case, Eq. (5) can be expressed as fol-ows:

1Aobs − A0

= 1Ac − A0

+ 1Kapp(Ac − A0)[CdS]

(6)

Therefore, if the enhancement of absorbance at 421, 407 and25 nm was due to absorption of complex, one would expect a

inear relationship between 1/(Aobs − A0) and the reciprocal con-entration of colloidal CdS with a slope equal to 1/Kapp(Ac − A0)nd an intercept equal to 1/(Ac − A0). The inset of Figs. 2–4 showshe plot for porphyrins, there is a good linear dependence of/(Aobs − A0) on the reciprocal concentration of colloidal CdS. Thealculated apparent association constant (Kapp) values from thetraight line are shown in Table 1.

.2. Fluorescence quenching of porphyrins by CdS colloids

The effects of CdS colloids on the fluorescence spectra of por-hyrins were examined. Fig. 5 shows the effect of increasing
oncentration of CdS colloids on the fluorescence spectrum ofnTPP. The other two porphyrins such as TMeOPP and PPIX also gavehe similar type of fluorescence behaviour (not shown here). Aseen, addition of CdS colloids to the solution of porphyrins resultsn the quenching of fluorescence intensity. The Stern–Volmer plot

1510 M.A. Jhonsi et al. / Spectrochimica Acta Part A 71 (2008) 1507–1511

Fig. 5. Fluorescence quenching of ZnTPP (1 × 10−6 M) with CdS colloids in the con-centration range of 0–5 × 10−3 M in DMF. The insert is the straight line dependenceof 1/I0–I on the reciprocal concentration of CdS colloid.

Fc

bt

tcsoseiqfcpe

P

TPs

S

123

m

m

ps

S

tcdfliptcpS

oe

P

Tnoflaa[

3

tI

ig. 6. Stern–Volmer plot for the quenching of porphyrins by colloidal CdS in theoncentration range of 0–5 × 10−3 M in DMF.

etween I0/I vs [CdS] gave upward curvature (Fig. 6) indicated thathe static nature of quenching occurred in these systems.

The quenching of fluorescence emission may be attributed tohe electron transfer, energy transfer or the formation of certainomplex which has no fluorescence. The band-gap energy of CdSemiconductor (Eg = 2.6 eV) is greater than the excited state energyf porphyrins (Es) in Table 2 [26,27] and no fluorescence emis-ion of the porphyrins can be absorbed by the colloidal CdS. Thus,nergy transfer from the excited state of porphyrins to colloidal CdSs impossible. It can therefore be concluded that the fluorescenceuenching shown in Fig. 5 should not be caused by energy trans-er.The other possible mechanism of quenching of porphyrins by
olloidal CdS is electron transfer between excited singlet of por-hyrins and the conduction band of CdS shown in the followingquation.
orphyrin∗ + CdS → Porphyrin+ + CdS(e−) (7)

able 2hotophysical properties of porphyrins such as fluorescence lifetime (�0), excitedtate energy (Es), ground (Es/s+ ) and excited state oxidation potential (Es∗/s+ )

. No. Porphyrins �0 (ns)a Es (eV)b Es/s+ (V)c Es∗/s+ (V)d

ZnTPP 1.89 2.04 1.06 −0.98PPIX 13.4 1.95 1.01 −0.94TMeOPP 9.2 1.87 1.04 −0.83

a Obtained from time-resolved measurements.b Singlet state energy of the porphyrins calculated from the fluorescence maxi-um wavelength based on the reported method [27].c The oxidation potentials are in DMF vs NHE from cyclic voltammetric measure-ents.d Calculated from the equation, Es∗/s+ = Es/s+ − Es, where Es/s+ is the oxidationotential of the ground state porphyrins and Es∗/s+ is the oxidation potential of theinglet excited state porphyrins and Es is the excitation energy.

dItccwsei

TF(

[

01234

cheme 1. Diagram illustrating the energetics of porphyrins and colloidal CdS.

This inference can be further defined by the difference betweenhe oxidation potential of the porphyrins and the energy level of theonduction band potential of CdS. With the use of porphyrins oxi-ation potential and the excited state energy (calculated from theuorescence maximum wavelength of the porphyrins [27]), accord-

ng to the equation, Es∗/s+ = Es/s+ − Es, one can obtain the oxidationotential of excited singlet of porphyrins shown in Table 2. Sincehese levels are energetically higher than the energy level of theonduction band of CdS (−1.0 V vs NHE) [28], the electron-transferath in Eq. (7) is impossible thermodynamically (shown below incheme 1).

So the fluorescence quenching may be due to the formationf complex which has no fluorescence as shown in the followingquation.

orphyrin + CdS � [Porphyrin· · ·CdS] (8)

he observed fluorescence quenching is due to the non-fluorescentature of complexed porphyrins with colloidal CdS and the flu-rescence only from the uncomplexed porphyrins. Based on theuorescence quenching data we can calculate the apparent associ-tion constant (Kapp) for the equilibrium between unadsorbed anddsorbed porphyrins using the method reported by Kamat and Fox28] and the calculated values are shown in Table 1.


Fluorescence lifetime measurement is useful in understandinghe interaction between the colloidal semiconductor–dye systems.n general, the measurement of fluorescence lifetime is the mostefinitive method to distinguish static and dynamic quenching [29].

n the present work we have studied the effect of CdS colloids onhe fluorescence lifetime of porphyrins, Table 3 shows the fluores-ence lifetime of porphyrins with different concentrations of CdS
olloids. Fig. 7 shows the fluorescence decay of TMeOPP with andithout CdS colloids (the decay for other two porphyrins are as
ame as TMeOPP, not shown here). The fluorescence of porphyrinsxhibit biexponential decay not only in the dilute solutions but alson the presence of CdS colloids. While increasing the concentration

able 3luorescence lifetimea of porphyrins at different concentrations of CdS colloids0–5 × 10−3 M) in dimethyl formamide

CdS] (×10−3 M) ZnTPP TMeOPP PPIX

1.89 9.2 13.41.89 9.2 13.711.88 9.1 13.701.89 9.2 13.41.89 9.11 13.33

a Lifetime given in nanoseconds.

M.A. Jhonsi et al. / Spectrochimica Act

Fc

optotqtC

4

sTspil

A

t

tUrtl

R

[

[

[[[[[[[[[

[

[[[

[

ig. 7. Fluorescence decay profiles for TMeOPP (1 × 10−6 M) with and without CdSolloid (5 × 10−3 M).

f CdS colloids there is no change in the fluorescence lifetime oforphyrins, they were almost constant (Table 3). This result showshat no electron-injection process occurs between the excited statef porphyrins and the conduction band of CdS. It also indicates thathe fluorescence quenching processes mainly belongs to the staticuenching. Moreover, these observations also support the sugges-ion of the adsorption of porphyrins onto the surface of colloidaldS.

. Conclusion

The effect of CdS colloids on the absorption and fluorescencepectra of porphyrins TMeOPP, PPIX and ZnTPP in DMF was studied.he result in perturbation of the absorption spectrum of porphyrinshows the ground state complex formation through adsorption oforphyrins on the surface of colloidal CdS. Static nature of quench-

ng has been confirmed by the unaltered lifetime of porphyrins byifetime measurements.

cknowledgments

R.R. thank CSIR (Ref: 01(2217)/08/EMR-II, dt. 06-05-2008) forhe project. Authors also thank DST-FIST and UGC-SAP for spec-

[

[[[

a Part A 71 (2008) 1507–1511 1511

rofluorimeter facility in the School of Chemistry, Bharathidasanniversity, Trichy, and Dr. S. Anandan (NIT, Trichy) and Dr. R. Rama-

aj (MKU, Madurai) for providing their CV facilities. We are thankfulo Prof. P. Ramamoorthy, NCUFP, University of Madras, Chennai, forifetime measurements.

eferences

[1] B. O’Regan, M. Gratzel, Nature 353 (1991) 737–740.[2] J.Z. Zhang, R.H. O’Neil, T.W. Roberti, J. Phys. Chem. 98 (1994) 3859–3864.[3] A. Hagfeldt, M. Gratzel, Chem. Rev. 95 (1995) 49–68.[4] J.J. Ramsden, M. Gratzel, Faraday Trans. 80 (1984) 919–933.[5] M. Bruchez, M. Moronne, P. Gin, S. Weiss, A.P. Alivisatos, Science 281 (1998)

2013.[6] A. Datta, A. Saha, A.K. Sinha, S.N. Bhattacharyya, S. Chatterjee, J. Photochem.

Photobiol. B: Biol. 78 (2005) 69–75.[7] A. Datta, S. Chatterjee, A.K. Sinha, S.N. Bhattacharyya, A. Saha, J. Lumin. 121

(2006) 553–560.[8] A.L. Linsebigler, G. Lu, J.T. Yates, Chem. Rev. 95 (1995) 735–758.[9] J.C. Tristao, F. Magalhaes, P. Corio, M.T.C. Sansiviero, J. Photochem, Photobiol. A:

Chem. 181 (2006) 152–157.10] S. Shiojiri, T. Hirai, I. Komasawa, J. Chem. Soc., Chem. Commun. (1998)

1439–1440.11] K. Murakoshi, H. Hosokawa, M. Saitoh, Y. Wada, H. Mori, S. Yanagida, J. Chem.

Soc., Faraday Trans. 94 (1998) 579–586.12] J. Britt, C. Ferekides, Appl. Phys. Lett. 62 (1993) 2851–2852.13] S.S. Rink, D.S. Chemla, D.A.B. Miller, Adv. Phys. 38 (1989) 89–188.14] L. Tan, L.Y. Liu, Q.J. Xie, Y. Zhang, S. Yao, Anal. Sci. 20 (2004) 441–445.15] S. Cherian, C. Wamser, J. Phys. Chem. B 104 (2000) 3624–3629.16] M. Rahman, H.J. Harmon, Spectrochim. Acta A 65 (2006) 901–906.17] D. Mauzerall, Biochemistry 4 (1965) 1801–1810.18] F.J. Vergeldt, R.B.M. Koehorst, A. Van Hoek, J. Phys. Chem. 99 (1995) 4397–4405.19] C.B. Mou, D.M. Chen, X.Y. Wang, Chem. Phys. Lett. 179 (1991) 237–242.20] F. Odobel, E. Blart, M. Lagre’e, M. Villieras, H. Boujtita, N.E. Murr, S. Caramoric,

C.A. Bignozzi, J. Mater. Chem. 13 (2003) 502–510.21] A. Kathiravan, V. Anbazhagan, M.A. Jhonsi, R. Renganathan, Spectrochim. Acta

A 70 (2008) 615–618.22] J. Bhamro, J. Chrysochoos, Photochem, Photobiol. A: Chem. 111 (1997) 187–198.23] P.K. Khanna, R. Gokhale, V.V.V.S. Subbarao, Mater. Lett. 57 (2003) 2489–2493.24] X. He, Y. Zhou, M. Yalin Zhou, T. Zhang, Shen, J. Colloid. Interface Sci. 225 (2000)

128–133.25] H.A. Benesi, J.H. Hildebrand, J. Am. Chem. Soc. 71 (1949) 2703–2707.
26] H. Zhang, Y. Zhou, M. Zhang, T. Shen, Y. Li, D. Zhu, J. Colloid. Interface Sci. 264
(2003) 290–295.27] E.J. Shin, D. Kim, J. Photochem. Photobiol. A: Chem. 152 (2002) 25–31.28] P.V. Kamat, J. Phys. Chem. 93 (1989) 859–864.29] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, Plenum Press, New York,

1983, pp. 258–265.

Journal of Molecular Structure 919 (2009) 79–82




Interaction of meso-tetrakis (p-sulfonatophenyl) porphyrin (TSPP) with pyrimidines:A steady state and time-resolved fluorescence quenching study

A. Kathiravan, V. Anbazhagan, M. Asha Jhonsi, R. Renganathan *


a r t i c l e i n f o a b s t r a c t

Article history:Received 13 July 2008Received in revised form 13 August 2008Accepted 13 August 2008Available online 26 August 2008

Keywords:TSPPFluorescence quenchingPhotoinduced electron transferPyrimidines

N

NH N

HN

SO3

O3S

SO3

Structure of meso-tetrakis (p-sulfonatophe



Fluorescence quenching of meso-tetrakis (p-sulfonatophenyl) porphyrin (TSPP) by certain pyrimidineshas been investigated by using steady state and time-resolved fluorescence techniques. The pyrimidinesused are uracil, 5-fluorouracil, 5-chlorouracil, 5-bromouracil, 5-iodouracil and 5-aminouracil. Thequenching was found to obey the Stern–Volmer equation and the corresponding Stern–Volmer plotswere linear from both steady state and time-resolved measurements, indicating dynamic nature ofquenching and the bimolecular quenching rate constants (kq) agreed well. Electron transfer from TSPPto pyrimidines were confirmed by the calculation of free energy change (DGet) using Rehm–Wellerequation.


1. Introduction

Porphyrin and porphyrin derivatives are suitable photosensitiz-ers for photovoltaic conversion of solar energy due to their strongabsorption in the region of 400–450 nm (B or Soret band) as well as500–700 nm (Q bands) [1,2]. The application of porphyrins in thefield of medicine has increased over the last decade due to theirefficacy in photodynamic therapy (PDT) for the treatment of cancer[3,4]. Investigations show that some porphyrins are accepted notonly as photosensitizers [5] but also as potential anti-cancer andanti-virus drugs [6,1]. Porphyrin demonstrates the photodynamicactivity against psoriasis atheromatous plaque, viral and bacterialinfections including HIV as well [7].

SO3

nyl) porphyrin [TSPP]

ll rights reserved.

: +91 431 2407045.an).

The 5-substituted pyrimidines comprise a biologically importantclass of base analogues. In particular, uracil and thymine (with ahydrogen atom or a methyl group in the 5-position, respectively)are major constituents of RNA and DNA, respectively. The thymine5-methyl group in DNA is a frequent site of oxidation [8–11], gen-erating the 5-hydroxymethyl, formyl and carboxyl derivatives. Theoxidation of DNA is an established source of genomic instability[12–14], possibly because of the oxidation of thymine. Indeed,the 5-halo uracil derivatives have demonstrated anti-tumor andanti-viral properties [15–18]. Addition of an electron to DNA is aprocess of general interest. One of the mechanisms by which ion-izing radiation damages DNA involves attachment of solvated elec-trons to the DNA bases [19–21]. It is generally understood that theattached electrons localize on the pyrimidines.

Recently, we have reported [22] the fluorescence quenching ofmeso-tetrakis(4-sulfonatophenyl)porphyrin by colloidal TiO2,which suggested the transfer of electron from TSPP to the conduc-tion band of TiO2. Scannell et al. reported the photoinduced elec-tron transfer to pyrimidines using fluorescence quenchingkinetics [23]. Based on the above reports, in the present work weaim to investigate the interaction between TSPP and pyrimidinesby using steady state and time-resolved fluorescence quenchingtechniques. Rehm–Weller equation has been successfully appliedto this system for the confirmation of electron transfer mechanism.

2. Experimental

2.1. Materials

TSPP was obtained from Sigma. Pyrimidines such as 5-fluoro-uracil, 5-chlorouracil, 5-bromouracil, 5-iodouracil and 5-amino-




NH

O

HN

O

Uracil

NH

O

HN

OF

5-Fluorouracil

NH

O

HN

ONH2

5-Aminouracil

OCl

OBr

OI

Structures of the pyrimidines

0

0.2

0.4

0.6

0.8

1

300

Wavelength

Abs

orba

nce

0

150

300

450

600

750

900

Inte

nsit

y

A B

700600500400

Fig. 1. Absorption (1 � 10�5 M) (A) and emission (1 � 10�6 M) (B) spectrum of TSPPin water.

80 A. Kathiravan et al. / Journal of Molecular Structure 919 (2009) 79–82

uracil were purchased from Aldrich. Double distilled water wasused for preparing the solutions. All measurements were per-formed at room temperature (28 �C).

2.2. Methods

2.2.1. Fluorescence quenching experiments2.2.1.1. Steady-state measurements. A stock solution of TSPP(1 � 10�3 M, 0.01 g) is prepared by using double distilled water.Three milliliters of 1 � 10�6 M TSPP from the stock solution isplaced in a quartz cuvette, sealed with a septum, lined with Teflontape to prevent contamination and then purged with nitrogen for15 min. A stock solution of the pyrimidines (1 � 10�2 M) was pre-pared by dissolving the appropriate amount in double distilledwater. Aliquots (30 ll) of the quencher solution are injected intothe sealed cuvette containing the TSPP. The excitation wavelengthof TSPP (413 nm) is chosen in order to ensure that none of the lightis absorbed by the quencher. The fluorescence quenching measure-ments were carried out in a JASCO FP-6500 spectrofluorimeter. Thewavelength of TSPP is monitored at 640 nm. The excitation andemission slit width (each 5 nm) and scan rate (500 nm/min) weremaintained constant for all the experiments. The fluorescencequenching rate constants kq were determined from Stern–Volmeranalysis.

Ground-state absorption measurements were recorded usingCary 300 UV–vis spectrophotometer.

2.2.1.2. Time-resolved measurements. Fluorescence lifetime mea-surements were carried out in a picosecond time correlated singlephoton counting (TCSPC) spectrometer. The excitation source is thetunable Ti-sapphire laser (TSUNAMI, Spectra Physics, USA). Thediode laser pumped millennia V (Spectra Physics) CW Nd-YVO4 la-ser was used to pump the sapphire rod in the Tsunami modelocked picosecond laser (Spectra Physics). The diode laser outputwas used to pump the Nd-YVO4 rod in the Millennia. The time-re-solved fluorescence emission was monitored at 640 nm. The emit-ted photons were detected by a MCP-PMT (Hamamtsu R3809U)after passing through the monochromator (f/3). The laser sourceis operated at 4 MHz and the signal from the photodiode is usedas a stop signal. The signal from the MCP-PMT is used as start sig-nal in order to avoid the dead time of the TAC. The difference be-tween the start and stop signal is due to the time taken by thepulses traveling through the cables and electronic relaxation ofthe excited state. The data analysis was carried out by the softwareprovided by IBH (DAS-6).

2.2.2. Cyclic voltammetric measurementsIn the present study, the reduction potential of pyrimidines was

measured in water with potassium chloride (KCl, 0.1 M) as sup-porting electrolyte. The experimental setup consisted of a platinumworking electrode, a glassy carbon counter electrode and a silverreference electrode. Reversible peak potentials were measured atdifferent scan rates (0.05 V/s). All samples were deaerated by bub-bling with pure nitrogen gas for ca. 5 min at room temperature.

NH

O

HN

5-Chlorouracil

NH

O

HN

5-Bromouracil

NH

O

HN

5-Iodouracil
3.1. Absorption characteristics of TSPP–pyrimidines system

Fig. 1 shows the absorption and emission spectra of TSPP inwater. The addition of pyrimidines does not give any changes inthe absorption spectrum of TSPP such as shift in wavelength aswell as there is no new peak formation. This precludes the possibil-ity of ground-state complex formation between TSPP andpyrimidines.

3.2. Fluorescence quenching of TSPP by pyrimidines

The emission of TSPP measured in water was quenched effec-tively by halogenated pyrimidine derivatives. For all the pyrimi-dine systems studied, there was no change in the shape of thefluorescence spectra, even in the higher concentration of pyrimi-dines used. Thus, the formation of exciplex formation has been ru-led out.

The fluorescence quenching behavior can be analyzed by theStern–Volmer relation

s0=s ¼ I0=I ¼ 1þ KSV½Q � ¼ 1þ kqs0½Q � ð1Þ

where, I0 and I are the emission intensity of fluorophore in the ab-sence and presence of quencher, KSV is Stern–Volmer constant re-lated to the bimolecular quenching rate constant (kq), byKSV = kq � s0. Fluorescence lifetime of TSPP (10.4 ns), in the absenceand presence of quencher is indicated as s0 and s, respectively.Addition of pyrimidines to a solution of TSPP resulted in thequenching of its fluorescence emission. The plot of I0/I vs [Q] werelinear with the correlation coefficient (R2) of greater than 0.9863.Comparison of typical S–V plots for the steady-state fluorescencequenching of TSPP by pyrimidines were shown in Fig. 2. The kq val-ues obtained from the initial slopes of Stern–Volmer plot are shownin Table 1. The larger value of kq (1010 M�1 s�1, diffusion controlledlimit) indicates the efficient quenching of TSPP by pyrimidines andthe quenching is due to diffusive process.

The quenching rate constants (kq) decreased in the following order:5-fluorouracil > 5-aminouracil > 5-chlorouracil > 5-bromouracil >5-iodouracil.

In the presence of uracil there is no quenching of TSPP was ob-served. This may be due to during the quenching process, anionicradical of the pyrimidine was formed, which lying between twoadjacent electron rich functional groups such as @NH and olefin.

0.9

1

1.1

1.2

1.3

0

[Q] x 10-4 M

5-Fluorouracil 5-Aminouracil

5-Chlorouracil 5-Bromouracil

5-Iodouracil

54321

τ 0/τ

Fig. 3. Stern–Volmer plots for the time-resolved fluorescence quenching of TSPPwith various concentrations of pyrimidines (0–5 � 10�4 M) in water.

TSPP* + Q (TSPP.....Q)* (TSPP +.....Q )* TSPP + + Q

hυk -d

kd

k -et

ket k esc

kRkb

_ _

0.9

1

1.1

1.2

1.3

0

[Q] x 10-4 M

I 0/I

5-Fluoro uracil 5-Amino uracil

5-Chloro uracil 5-Bromo uracil

5-Iodo uracil

54321

Fig. 2. Stern–Volmer plots for the steady-state fluorescence quenching of TSPP(1 � 106 M) with various concentrations of pyrimidines (0–5 � 10�4 M) in water.

Table 1Steady state and time-resolved quenching rate constants, oxidation potential ofpyrimidines and free energy change for TSPP–pyrimidines system

S.No. Quenchers kq � 1010 (M�1 s�1) E1/2 (V) SCE DGet (eV)

Steady state Time resolved

1 Uracil – – – –2 5-Fluorouracil 4.61 5.11 �0.56 �0.263 5-Aminouracil 4.03 4.46 �0.58 �0.244 5-Chlorouracil 2.59 2.27 �0.67 �0.155 5-Bromouracil 2.11 2.42 �0.72 �0.106 5-Iodouracil 2.01 1.75 �0.75 �0.07

A. Kathiravan et al. / Journal of Molecular Structure 919 (2009) 79–82 81

NH

O

HN

O

TSPP + no quenching take place

These two functional groups being an electron rich which willdestabilize the anionic radical by increasing the electron densityof the system. Hence pyrimidine alone does not exhibit anyquenching.

On the other hand, if the 5th position is substituted with elec-tronegative atoms (F, Cl, Br, I and NH2), it shows quenching.

NH

O

HN

O

TSPP + quenching take plac eX

X = F, NH2, Cl, Br & I

When electronegative atoms are present in the 5th position of ura-cil, stability of the anionic radical intermediate will be more. The �I

TSPP TSPP + Q Products

Scheme 1.

10.2

10.3

10.4

10.5

10.6

10.7

10.8

-0.8-0.75-0.7-0.65-0.6-0.55-0.5

E1/2 (V)

log

kq

Fig. 4. Plot between logkq and E1/2 (V) of pyrimidines.

effect of halogen atoms will disperse the negative charge, thus itwill stabilize the anionic radical and therefore facilitate the quench-ing process. Hence, 5-fluorouracil has high magnitude of quenchingrate constant than others, this is due to fluorine has more electro-negative than other halogen atoms (F > N > Cl > Br > I).


The fluorescence quenching data clearly highlighted the role ofexcited-singlet state TSPP in injecting its electron into pyrimidinesand dynamic (collisional) nature of quenching, supported by thelinear S–V plot.

In general, static and dynamic quenching can be distinguishedby lifetime measurements [24]. Fluorescence lifetime of TSPP hasregularly decreased upon addition of pyrimidines. S–V plots ofs0/s vs [Q] were linear with the correlation coefficient (R2) of great-er than 0.9865 indicating the dynamic nature of quenching processFig. 3. The calculated kq (Table 1) values are matched well with

steady state measurements. The good agreement between thesevalues of kq highlighted the validity of the assumption proposedfor the interaction between TSPP and pyrimidines.

3.4. Calculation of free energy changes (DGet) for the electron transferreactions

The thermodynamic feasibility of excited state electron transferreaction was confirmed by employing the well known Rehm–Wel-ler expression [25].

DGet ¼ EðoxÞ1=2 � EðredÞ

1=2 � Eð0;0Þ þ C ð2Þ

where, E½(ox) is the oxidation potential of TSPP (1.1 V vs SCE) [18],

E½(red) is the reduction potential of the pyrimidines (Table 1), E(0,0)

is the singlet state energy of the TSPP (1.92 eV) [26] and C is colum-bic term. Since one of the species is neutral and the solvent used ispolar in nature, the coulombic term in the above expression can beneglected [27]. The DGet value thus calculated are listed in Table 1.The negative value of DGet indicates the electron transfer processesstudied are thermodynamically favorable. The propensity of kq toincrease with decreasing reduction potential of the acceptors isthe incontrovertible proof for the electron transfer mechanism,and based on this, following scheme (Scheme 1) is conceivable.The plot of log kq vs E1/2 (V) is represented in Fig. 4, and the valuesare collected in Table 1.

where, kd and k�d are the rate constants of diffusion and disso-ciation of the encounter complex, respectively. ket and k�et are theactivation controlled rate constants for electron transfer, and kesc is

82 A. Kathiravan et al. / Journal of Molecular Structure 919 (2009) 79–82

the rate constant for the separation of the radicals. kb is the rateconstant for the recombination of the radical pair. kR is the rateconstant for the decay of the TSPP radical.

4. Conclusion

Based on the above results, the electron transfer mechanismwas proposed for the fluorescence quenching of TSPP by haloge-nated pyrimidines in water, and it was established using thetime-resolved fluorescence technique. Porphyrin serves as electrondonor and pyrimidines as electron acceptors. There is no ground-state complex formation between TSPP and pyrimidines. The logkq is correlated satisfactorily with the reduction potential of thepyrimidines.

Acknowledgments

R. R. thanks DST for NSTI Project (SR/S5/NM-16/2007, dt:3.7.’08). R. R. and A. K. thank CSIR (Ref: No. 01(2217)/08/EMR-II,dt. 06/05/2008) for the Project and Fellowship respectively.

We are thankful to Prof. P. Ramamoorthy, NCUFP, University ofMadras, Chennai for lifetime measurements. Authors also thank Dr.S. Anandan, NIT, Trichy, for providing his CV facilities.

References

[1] R. Bonnett, Chem. Soc. Rev. (1995) 19.[2] M. Ochsner, J. Photochem. Photobiol. B Biol. 39 (1997) 1.

[3] A. Kay, M. Gratzel, J. Phys. Chem. 97 (1993) 6272.[4] S. Cherian, C.C. Wamser, J. Phys. Chem. B 104 (2000) 3624.[5] M.A. Fox, A.J. Baud, Science 261 (1993) 897.[6] A.M. Del, C. Batlle, J. Photochem. Photobiol. 20 (1993) 5.[7] I.B. Haur, B. Horowitz, Photochem. Photobiol. 62 (1995) 383.[8] J.P.E. Spencer, M. Whiteman, A. Jenner, B. Halliwell, Free Radical Biol. Med. 28

(2000) 1039.[9] G.W. Teebor, K. Frenkel, M.S. Goldstein, Proc. Natl. Acad. Sci. USA 81 (1984)

318.[10] E. Privat, L. Sowers, Mutat. Res. 354 (1996) 151.[11] E. Privat, L. Sowers, Chem. Res. Toxicol. 9 (1996) 745.[12] T. Douki, J. Onuki, M.H.G. Medeiros, E.J.H. Bechara, J. Cadet, P.D. Mascio, FEBS

Lett. 428 (1998) 93.[13] K. Sakano, S. Oikawa, Y. Hiraku, S. Kawanishi, Free Radical Biol. Med. 33 (2002)

703.[14] P. Liu, A. Burdzy, L.C. Sowers, DNA Repair 2 (2003) 199–210.[15] S. Cao, D.P. Baccanari, Y.M. Rustum, S.T. Davis, R.L. Tansik, D.J.T. Porter, T.

Spector, Biochem. Pharmacol. 59 (2000) 953.[16] V. Kristafor, S.R. Malic, M. Cetina, M. Kralj, L. Suman, K. Pavelic, J. Balzarini, E.D.

Clerca, M. Mintas, Bioorg. Med. Chem. 14 (2006) 8126.[17] T. Gazivoda, S.R. Malic, M. Marjanovic, M. Kralj, K. Pavelic, J. Balzarini, E.D.

Clerca, M. Minton, Bioorg. Med. Chem. 15 (2007) 749.[18] S.M. Morris, Mutat. Res. 297 (1993) 39.[19] C.V. Sonntag, The Chemical Basis of Radiation Biology, Taylor & Francis,

Philadelphia, 1987.[20] D. Becker, M.D. Sevilla, Adv. Radiat. Biol. 17 (1993) 121.[21] P. O’Neill, E.M. Fielden, Adv. Radiat. Biol. 17 (1993) 53.[22] A. Kathiravan, V. Anbazhagan, M. Asha Jhonsi, R. Renganathan, Spectrochim.

Acta A 70 (2008) 615.[23] M.P. Scannell, G. Prakash, D.E. Falvey, J. Phys. Chem. A 101 (1997) 4332.[24] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, Plenum press, New

York, 1983. p. 258.[25] C. Tablet, M. Hilebrand, J. Photochem. Photobiol. A: Chem. 189 (2007) 73.[26] K. Kalyanasundaram, M.N. Spallart, J. Phys. Chem. 86 (1982) 5163.[27] S. Parret, F.M. Savary, J.P. Fouassier, P. Ramamurthy, J. Photochem. Photobiol.

A: Chem. 83 (1994) 205.

IA

Aa

b

a

ARRAA

KCFH

1

tstSttTm

6gdi

cTnlacct

0d

Colloids and Surfaces A: Physicochem. Eng. Aspects 333 (2009) 91–95


Colloids and Surfaces A: Physicochemical andEngineering Aspects

journa l homepage: www.e lsev ier .com/ locate /co lsur fa

nteraction of colloidal TiO2 with human serum albumin:fluorescence quenching study

. Kathiravana, S. Anandanb, R. Renganathana,∗

School of Chemistry, Bharathidasan University, Tiruchirappalli 620 024, Tamil Nadu, IndiaNanomaterials & Solar Energy Conversion Lab, Department of Chemistry, National Institute of Technology, Tiruchirappalli 620 015, Tamil Nadu, India

r t i c l e i n f o

rticle history:eceived 28 March 2008

a b s t r a c t

The interaction between colloidal TiO2 and human serum albumin (HSA) was studied by using absorp-tion, fluorescence and synchronous fluorescence spectroscopic measurements. Colloidal TiO2 effectively

eceived in revised form 5 September 2008ccepted 11 September 2008vailable online 19 September 2008

eywords:olloidal TiO2

quenched the intrinsic fluorescence of HSA. The number of binding sites (n) and apparent binding con-stant (K) were calculated by fluorescence quenching data. The interaction between colloidal TiO2 and HSAoccurs through static quenching and conformational changes of HSA were observed.


Tsaad

2

2

(pp

2

ti

luorescence quenchingSA

. Introduction

One of the most important biological functions of albumins isheir ability to carry drugs as well as endogenous and exogenousubstances. Numerous experiments with the aim of characterizinghe binding capacity and sites of albumins have been reported [1,2].erum albumins are most abundant proteins in plasma [3]. Amonghem, HSA has a wide range of physiological functions involvingransport and delivery of fatty acids, bilirubin, steroids, etc. [4,5].he binding properties of HSA with drugs were investigated byany researchers [6–10].HSA is a water-soluble protein with the molecular weight of

6,500 containing 585 amino-acid residues. There is only a sin-le tryptophan (Trp) residue within HSA at the position of 214 inomain II, which makes it very convenient to study the protein

ntrinsic fluorescence [11–13].Nanoparticle probes acting as biosensors in chemical and bio-

hemical fields and their applications are becoming more extensive.hree types of nanoparticles in biochemical analysis are usedamely metal nanoparticles [14], silica nanoparticles [15,16] and

uminescence quantum dots [17–19]. These probes have been
pplied to the ultrasensitive detection of proteins, DNA sequencing,linical diagnostics, etc. TiO2 nanoparticles have also been used asarriers for photosensitizer like porphyrins [20] in photodynamicherapy for cancer treatment [21].

wspfstmc

927-7757/$ – see front matter © 2008 Elsevier B.V. All rights reserved.oi:10.1016/j.colsurfa.2008.09.027

In this paper, we have studied the interaction between colloidaliO2 nanoparticles with HSA using absorption and fluorescencepectral measurements. Critical literature survey reveals thatttempts have not been made so far to investigate this type of inter-ction studies. The binding constant and binding sites have beenetermined.

. Experimental

.1. Materials

Titanium(IV) 2-propoxide was purchased from Aldrich. HSAfatty acid free, Sigma) was dissolved in double distilled water torepare the stock solution (1.0 × 10−4 M). All measurements wereerformed at ambient temperature.


The colloidal TiO2 suspension was prepared by the hydrolysis ofitanium(IV) 2-propoxide [22]. Typically, titanium(IV) 2-propoxiden 2-propanol (l0%, 0.5 ml) was injected by syringe into 40 ml ofater and kept stirred for 8 h under N2 atmosphere. There is no

tabilizing agents were used. The colloidal TiO2 nanoparticles pre-ared by this method were stable for 3–5 days. In the present study,
resh colloidal TiO2 dispersed in water was prepared before eachet of measurements. The stock solution was diluted with watero obtain the desired concentration of TiO2. No attempts were
ade to exclude the traces of 2-propanol (∼0.4%) present in theolloidal TiO2 and it was confirmed separately that the presence


http://www.elsevier.com/locate/colsurfa


dx.doi.org/10.1016/j.colsurfa.2008.09.027

92 A. Kathiravan et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 333 (2009) 91–95

oe

2

woT(SQcw

3

3

sootaoooetg

ca

H

Ft

Fc

K

Toa

A

wdoeHac

A

A(t

Tt1s(d

3

Fig. 1. Absorption (A) and emission (B) spectrum of HSA.

f 2-propanol did not affect the photochemical measurements asarlier reported [23].


The fluorescence quenching measurements were carried outith JASCO FP-6500 spectrofluorimeter. The excitation wavelength

f HSA was 280 nm and the emission was monitored at 337 nm.he excitation and emission slit widths (each 5 nm) and scan rate500 nm/min) were constantly maintained for all measurements.amples were carefully degassed by pure nitrogen gas for 15 min.uartz cells (4 cm × 1 cm × 1 cm) with high vacuum teflon stop-ocks were used for degassing. Absorption spectral measurementsere recorded using Cary300 UV–vis spectrophotometer.


.1. Absorption characteristics of HSA–TiO2

Fig. 1 shows the absorption and emission spectra of HSA. Fig. 2hows the absorption spectrum of HSA in the absence and presencef colloidal TiO2 at different concentrations. From this study webserved that upon increasing the concentration of colloidal TiO2he absorbence of HSA increases regularly without change in itsbsorption wavelength (280 nm). It may be due to the adsorptionf HSA partly on the surface of colloidal TiO2 supported by similarbservations reported earlier [24]. The main changes that we canbserve in the spectra of complex were not due to the experimentalrror, especially in the range of 230–300 nm. These results indicatedhat there is an interaction between colloidal TiO2 and HSA throughround state complex formation.

The equilibrium for the formation of complex between HSA and
olloidal TiO2 is given by equation 1, where Kapp represents thepparent association constant:
SA+TiO2Kapp� HSA . . . TiO2 (1)

ig. 2. Absorption spectrum of HSA in the absence and presence of colloidal TiO2 inhe concentration range of 0–6 × 10−4 M.

t

wpbototeTfiot

ig. 3. Fluorescence quenching of HSA in the presence of various concentration ofolloidal TiO2, [TiO2] = 0–6 × 10−5 M.

app = [HSA . . . TiO2][HSA][TiO2]

he change in intensity of the absorption peak (280 nm) as a resultf formation of the surface complex were utilized to obtain Kapp

ccording to the method reported by Benesi and Hildebrand [25]:

obs = (1 − ˛)C0εHSAl + ˛C0εcl (2)

here Aobs is the observed absorbance of the solution containingifferent concentrations of colloidal TiO2 at 280 nm, ˛ is the degreef association between HSA and TiO2, εHSA and εc are the molarxtinction coefficients at the defined wavelength (� = 280 nm) ofSA and the formed complex, respectively. Eq. (2) can be expresseds Eq. (3), where A0 and Ac are the absorbances of HSA and theomplex at 280 nm, respectively, with the concentration of C0:

obs = (1 − ˛)A0 + ˛Ac (3)

t relatively high TiO2 concentrations, ˛ can be equated toKapp[TiO2])/(1 + Kapp[TiO2]). In this case, Eq. (3) can be changedo the following equation:

1Aobs − A0

= 1Ac − A0


(4)

he enhancement of absorbance at 280 nm was due to absorp-ion of surface complex, based on the linear relationship between/(Aobs − A0) vs. reciprocal concentration of colloidal TiO2 with alope equal to 1/Kapp(Ac − A0) and an intercept equal to 1/(Ac − A0)Fig. 2, insert). The value of apparent association constant (Kapp)etermined from this plot is 9.12 × 105 M−1.

.2. Fluorescence quenching of HSA by colloidal TiO2

The fluorescence quenching is described by Stern–Volmer rela-ion:

I0I

= 1 + KSV[Q ] = 1 + kq�0[Q ] (5)

here I0 and I are the fluorescence intensities in the absence andresence of quencher, KSV is the Stern–Volmer constant, kq is theimolecular quenching rate constant, and �0 is the average lifetimef HSA, 10−8 s [26], [Q] is the concentration of quencher. Fig. 3 showshe effect of increasing concentration of colloidal TiO2 on the flu-rescence emission spectrum of HSA. Addition of TiO2 colloid tohe solution of HSA resulted in the quenching of its fluorescencemission and there is no peak shift and no new peak was noticed.
he result from the fluorescence study indicated that the complexormed between colloidal TiO2 and HSA is responsible for quench-ng. The possible scattering due to colloidal nature of TiO2 has beenmitted because of the fact that baseline correction was done for allhe spectral measurements and also even after adding the highest

A. Kathiravan et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 333 (2009) 91–95 93

Fc

ccf

asqmqsncbmtf

3

st

n

wlT

K

Itcb[

l

wccaatmbic

3

vcbebcmotHctwtvaaowsswtsgtsosIt is important to note that colloidal TiO2 affects only the tyrosineresidues present in HSA molecule. This is because tyrosine containsone aromatic hydroxyl group unlike tryptophan.

In our previous study, riboflavin was used as a sensitizer forcolloidal TiO2 [31] and it was observed that riboflavin interacted

ig. 4. Stern–Volmer plot for the steady-state fluorescence quenching of HSA byolloidal TiO2.

oncentration of colloidal TiO2 nanoparticles the solution remainslear. So we confirmed that the quenching of HSA is only resultingrom interaction with colloidal TiO2 and not due to scattering.

According to Eq. (5) we got a linear plot (Fig. 4) between I0/Igainst [TiO2], from the slope we calculated the quenching rate con-tant (kq) as 6.58 × 1013 M−1 s−1. In general, maximum collisionaluenching constant (kq) of various kinds of quenchers to biopoly-ers is 2.0 × 1010 M−1 s−1 [26]. But for HSA–TiO2 system higher

uenching rate constant (6.58 × 1013 M−1 s−1) was observed. Thishows that the quenching of HSA by colloidal TiO2 is not dynamic inature. Therefore, it depends on the formation of complex betweenolloidal TiO2 and HSA. For static quenching, we can deduce theinding constant (K) because static quenching arises from the for-ation of complex between fluorophore and the quencher. Hence

he binding constant (K) was calculated by the method given in theollowing section.

.3. Binding constant and number of binding sites

If it is assumed that there are similar and independent bindingites in HSA, the relationship between fluorescence intensity andhe quencher medium can be deduced from the following equation:

Q + B → Qn . . . B (6)

here B is the fluorophore, Q is the quencher, nQ + B is the postu-ated complex between a fluorophore and n molecules of quencher.he constant K is given by

= [Qn . . . .B][Q ]n[B]

(7)

f the overall amount of biomolecules (bound or unbound withhe quencher) is B0, then [B0] = [Qn. . .B] + [B], here [B] is theoncentration of unbound biomolecules, then the relationshipetween fluorescence intensity and the unbound biomolecule asB]/[B0] = F/F0 that is

og[

F0 − F

F

]= log K + n log[Q ] (8)

here K is the binding constant of colloidal TiO2 with HSA, whichan be determined from the intercept of log[(F0 − F)/F] vs. log[Q]urve as shown in Fig. 5. Thus we can obtain binding constant (K)
s 3.49 × 105 M−1 and binding sites “n” (0.92) from the interceptnd slope of Fig. 5. The value of binding constant obtained fromhe data of fluorescence quenching matches well with that deter-
ined from the absorption spectral changes. The good agreementetween these values of binding constant highlighted the valid-

ty of assumption proposed for the association between HSA andolloidal TiO2.

Fa

Fig. 5. Plot of log[(F0 − F)/F] vs. log[Q].

.4. Characteristics of synchronous fluorescence spectra

Synchronous fluorescence spectroscopy has been applied to aariety of multi-component system. The main advantages of syn-hronous fluorescence spectra are simplified spectra, narrowedandwidth, high selectivity and sensitivity. The excitation andmission monochromators are synchronously scanned, separatedy a constant wavelength interval (��). As it is well known syn-hronous fluorescence spectra can provide information on theolecular microenvironment, particularly in the vicinity of the flu-

rophore’s functional groups [27]. Fluorescence of HSA may be dueo the presence of tyrosine, tryptophan and phenylalanine residues.ence spectroscopic methods are usually applied to the study ofonformation of serum protein. In synchronous fluorescence spec-roscopy, according to Miller [28], the difference between excitationavelength and emission wavelength (�� = �emi − �exc) reflects

he spectra of a different nature of chromophores, with large ��alues such as 60 nm, the synchronous fluorescence of HSA is char-cteristic of tryptophan residue and with small �� values suchs 15 nm is characteristic of tyrosine [29]. The synchronous flu-rescence spectra of HSA with various amounts of colloidal TiO2ere recorded at �� = 15 nm (Fig. 6) and �� = 60 nm (spectra not

hown here), respectively. The emission wavelength of the tyro-ine residues is blue-shifted (�emi from 306 to 302 nm in Fig. 6)ith increasing concentration of colloidal TiO2. At the same time,

he tryptophan fluorescence emission is decreased regularly, but noignificant change in its emission wavelength was observed. It sug-ests that the interaction of colloidal TiO2 with HSA does not affecthe conformation of tryptophan micro-region. The change in tyro-ine fluorescence spectrum may represent that the conformationf HSA is somewhat disturbed, leading to the polarity around tyro-ine residues strengthened and the hydrophobicity weakened [30].

ig. 6. Synchronous fluorescence spectrum of HSA with �� = 15 nm in the absencend presence of colloidal TiO2 (0–3 × 10−5 M).

94 A. Kathiravan et al. / Colloids and Surfaces A: Physicochem. Eng. Aspects 333 (2009) 91–95

ractio

ttftma

4

istseiibp

A

(

0

r

b

R

[

[

[

[

[

[

[

[

[

[

[

[[

[

[

[

Scheme 1. Mechanism for the inte

hrough the hydroxyl group of colloidal TiO2. Hence it is clear thathe presence of hydroxyl group in tyrosine residues is responsibleor the interaction of HSA with colloidal TiO2 (Scheme 1) similar tohe interaction of 2,3-diazabicyclo[2.2.2]oct-2-ene (DBO) with aro-

atic hydroxyl group of tyrosine in BSA as reported by Anbazhagannd Renganathan [32].

. Summary

The interaction between colloidal TiO2 and HSA has been stud-ed by absorption, fluorescence and synchronous fluorescencepectral measurements. The results presented clearly indicatedhat colloidal TiO2 quenches the fluorescence of HSA throughtatic quenching. From the synchronous fluorescence spectra, it isstablished that the conformational change of HSA occurs due tonteraction with colloidal TiO2. The quenching rate constant, bind-ng constant, and number of binding sites have been calculated. Theinding study of colloidal TiO2 with HSA is of great importance inharmacy, pharmacology and biochemistry.

cknowledgements

R. R. thanks DST (Ref. No. SR/NM/NS-16/2007, dt: 26.9.2008)Government of India) for the Project.

R. R. and A.K. thank CSIR (Ref. No. 01(2217)/08/EMR-II, dt.6/05/2008) for the Project and Fellowship, respectively.

The author SA thanks CSIR, New Delhi for the sanction of majoresearch fund (Ref. No. 01(2197)/07/EMR-II, dt. 26/11/2007).

Authors thank Dr. M. Ashok kumar (Professor, University of Mel-ourne, Australia) for gift of HSA sample.

eferences

[1] F. Moreno, M. Cortijo, J.G. Jimenez, Interaction of acrylodan with human serumalbumin. A fluorescence spectroscopic study, Photochem. Photobiol. 70 (1999)695–700.

[2] M. Kessler, O. Wolfbeis, Laser-induced fluorometric determination of albu-min using longwave absorbing molecular probes, Anal. Biochem. 200 (1992)254–259.

[3] D.C. Carter, J.X. Ho, Structure of serum albumin, Adv. Protein Chem. 45 (1994)153–203.

[4] R.E. Olson, D.D. Christ, Plasma protein binding of drugs, Ann. Rep. Med. Chem.31 (1996) 327–337.

[5] S. Ashoka, J. Seetharamappa, P.B. Kandagal, S.M.T. Shaikh, Investigation of theinteraction between trazodone hydrochloride and bovine serum albumin, J.Lumin. 121 (2006) 179–186.

[6] R.K. Nanda, N. Sarkar, R. Banerjee, Probing the interaction of ellagic acid withhuman serum albumin: a fluorescence spectroscopic study, J. Photochem. Pho-tobiol. A: Chem. 192 (2007) 152–158.

[

[

[

n between colloidal TiO2 and HSA.

[7] F.Q. Cheng, Y.P. Wang, Z.P. Li, C. Dong, Fluorescence study on the interaction ofhuman serum albumin with bromsulphalein, Spectrochim. Acta A 65 (2006)1144–1147.

[8] F. Cuia, J. Wang, Y. Cui, J. Li, X. Yao, Y. Lu, J. Fan, The binding of3-(p-bromophenyl)-5-methyl-thiohydantoin with human serum albumin:investigation by fluorescence spectroscopy and molecular model, J. Lumin. 127(2007) 409–415.

[9] J. Jin, J. Zhu, X. Yao, L. Wu, Study on the binding of farrerol to human serumalbumin, J. Photochem. Photobiol. A: Chem. 191 (2007) 59–65.

10] Z.D. Qi, B. Zhou, X. Qi, S. Chaun, Y. Liu, J. Dai, Interaction of rofecoxib withhuman serum albumin: determination of binding constants and the bindingsite by spectroscopic methods, J. Photochem. Photobiol. A: Chem. 193 (2008)81–88.

11] J. Valanciunaite, S. Bagdonas, G. Streckyte, R. Rotomskis, Spectroscopic studyof TPPS4 nanostructures in the presence of bovine serum albumin, Photochem.Photobiol. Sci. 5 (2006) 381–388.

12] U.K. Hansen, Molecular aspects of ligand binding to serum albumin, Pharmacol.Rev. 33 (1981) 17–53.

13] S. Patel, A. Datta, Steady-state and time-resolved fluorescence investigation ofthe specific binding of two chlorine derivatives with human serum albumin, J.Phys. Chem. B 111 (2007) 10557–10562.

14] R. Elghanian, J.J. Storhoff, R.C. Mucic, R.L. Letsinger, C.A. Mirkin, Selec-tive colorimetric detection of polynucleotides based on the distance-dependent optical properties of gold nanoparticles, Science 277 (1997) 1078–1081.

15] S. Santra, P. Zhang, K. Wang, R. Tapec, W. Tan, Conjugation of biomolecules withluminophore-doped silica nanoparticles for photo stable biomarkers, Anal.Chem. 73 (2001) 4988–4993.

16] M. Qhobosheane, S. Santra, P. Zhang, W. Tan, Biochemically functionalized silicananoparticles, Analyst 126 (2001) 1274–1278.

17] X.G. Peng, M.C. Schlamp, A.V. Kadavanich, A.P. Alivistos, Epitaxial growth ofhighly luminescent CdSe/CdS core/shell nanocrystals with photostability andelectronic accessibility, J. Am. Chem. Soc. 119 (1997) 7019–7029.

18] W.C.W. Chan, S. Nie, Quantum dot bioconjugates for ultrasensitive nonisotopicdetection, Science 281 (1998) 2016–2018.

19] L. Tan, L.Y. Liu, Q.J. Xie, Y.Y. Zhang, S.Z. Yao, Fluorescence quenching of bovineserum albumin in reversed micelles by CdS nanoparticles, Anal. Sci. 20 (2004)441–444.

20] S. Wang, R. Gao, F. Zhou, M. Selke, Nanomaterials and singlet oxygen photo-sensitizers: potential applications in photodynamic therapy, J. Mater. Chem. 14(2004) 487–493.

21] A. Wiseman, Handbook of Enzyme Biotechnology, Horwood, Chichester, 1985.22] D. Bahnemann, A. Henglein, J. Lilie, L. Spanhel, Flash photolysis observation

of the absorption spectra of trapped positive holes and electrons in colloidaltitanium dioxide, J. Phys. Chem. 88 (1984) 709–711.

23] P.V. Kamat, J.P. Chauvet, R.W. Fessenden, Photoelectrochemistry in particu-late systems. 4. Photosensitization of a titanium dioxide semiconductor with achlorophyll analog, J. Phys. Chem. 90 (1986) 1389–1394.

24] C. Chen, X. Qi, B. Zhou, Photosensitization of colloidal TiO2 with a cyanine dye,J. Photochem. Photobiol. A: Chem. 109 (1997) 155–158.

25] H.A. Benesi, J.H. Hildebrand, A spectrophotometric investigation of the inter-action of iodine with aromatic hydrocarbons, J. Am. Chem. Soc. 71 (1949)2703–2707.

26] J.R. Lakowicz, G. Weber, Quenching of fluorescence by oxygen. Probe for struc-tural fluctuations in macromolecules, Biochemistry 12 (1973) 4161–4170.

27] G.Z. Chen, X.Z. Huang, J.G. Xu, Z.Z. Zheng, Z.B. Wang, Methods of FluorescenceAnalysis, second ed., Science Press, Beijing, 1990.

28] J.N. Miller, Recent advances in molecular luminescence analysis, Proc. Anal. Div.Chem. Soc. 16 (1979) 203–208.

A: Phy

[

[

A. Kathiravan et al. / Colloids and Surfaces

29] J.H. Tang, F. Luan, X.G. Chen, Binding analysis of glycyrrhetinic acid to humanserum albumin: fluorescence spectroscopy, FTIR, and molecular modeling,Bioorg. Med. Chem. 14 (2006) 3210–3217.

30] B. Klajnert, M. Bryszewska, Fluorescence studies on PAMAM dendrimers inter-actions with bovine serum albumin, Bioelectrochemistry 55 (2002) 33–35.

[

[

sicochem. Eng. Aspects 333 (2009) 91–95 95

31] A. Kathiravan, R. Renganathan, Photoinduced interaction between riboflavinand TiO2 colloid, Spectrochim. Acta A 71 (2008) 1080–1083.

32] V. Anbazhagan, R. Renganathan, Study on the binding of 2,3-diazabicyclo[2.2.2]oct-2-ene with bovine serum albumin by fluorescence spectroscopy, J.Lumin. 128 (2008) 1454–1458.

Polyhedron 28 (2009) 157–161


Polyhedron

journal homepage: www.elsevier .com/locate /poly

Interaction of colloidal AgTiO2 nanoparticles with bovine serum albumin

A. Kathiravan a, R. Renganathan a,*, S. Anandan b

a School of Chemistry, Bharathidasan University, Tiruchirappalli 620 024, Tamil Nadu, Indiab Nanomaterials and Solar Energy Conversion Lab, Department of Chemistry, National Institute of Technology, Tiruchirappalli 620 015, Tamil Nadu, India


Article history:Received 5 July 2008Accepted 26 September 2008Available online 27 November 2008

Keywords:Colloidal AgTiO2

Fluorescence quenchingBSA

0277-5387/$ - see front matter � 2008 Elsevier Ltd. Adoi:10.1016/j.poly.2008.09.023


a b s t r a c t

The interaction between colloidal AgTiO2 nanoparticles and bovine serum albumin (BSA) was studied byusing absorption, steady state, time resolved and synchronous fluorescence spectroscopy measurements.Absorption spectroscopy proved the formation of a ground state BSA� � �AgTiO2 complex. Upon excitationof BSA, colloidal AgTiO2 nanoparticles effectively quenched the intrinsic fluorescence of BSA. The numberof binding sites (n = 1.06) and apparent binding constant (K = 3.71 � 105 M�1) were calculated by thefluorescence quenching method. A static mechanism and conformational changes of BSA were observed.

� 2008 Elsevier Ltd. All rights reserved.

1. Introduction

One of the most important biological functions of albumins istheir ability to carry drugs as well as endogenous and exogenoussubstances [1]. The binding capacity and sites of albumins havebeen characterized [2]. Serum albumins are the most abundantproteins in plasma [3]. As the major soluble protein constituentsof the circulatory system, they have many physiological functions[4]. Among the serum albumins, BSA has a wide range of physio-logical functions involving binding, transport and delivery of fattyacids, porphyrins, bilirubin and steroids, etc. It is home to specificbinding sites for metals and pharmaceutical dyes [5]. The bindingproperties of BSA with various drugs have been fully investigatedby many researchers [6–10], which are useful for understandingthe reaction mechanism as well as for providing guidance for theapplication and design of new drugs [11].

Bovine serum albumin (BSA) has been selected as our proteinmodel due to its water-soluble nature which is important for inter-action studies [12,13]. It contains 582 amino-acid residues with amolecular weight of 69,000, and two tryptophan moieties at posi-tions 134 and 212 as well as tyrosine and phenylalanine [14]. Theintrinsic fluorescence of BSA is due to aromatic amino-acid resi-dues. Nanoparticle probes act as biosensors in the chemical andbiochemical fields, and their applications are becoming moreextensive. These probes have been applied to ultrasensitive detec-tion of proteins, DNA sequencing, clinical diagnostics etc. Recentlywe reported the interaction of colloidal TiO2 with bovine serumalbumin using fluorescence spectroscopy [15].

ll rights reserved.

: +91 431 2407045.an).

Metal semiconductor nanoparticles have become an attractivetopic of research because of their potential applications in differentfields, such as charge-transfer processes [16], optoelectronics [17]and medicine [18]. Among such nanocomposite structures, AgTiO2

has more attention paid to it because silver is an extremely attrac-tive noble metal on the nanoscale due to its remarkable catalyticactivity [19], size and shape-dependent optical properties, promis-ing applications in chemical and biological sensing [20], and alsodue to its antimicrobial activity [21]. Elechiguerra et al. studiedmetal nanoparticle interactions with the human immunodefi-ciency virus-1 (HIV-1), and silver nanoparticles of the size 1–10 nm were attached to HIV-1, and this prevented the virus frombinding to host cells [22].

Based on the above strategy we have investigated the interac-tion of BSA with colloidal AgTiO2 nanoparticles (Scheme 1). Thisis the first challenge made to investigate the mode of interactionof colloidal AgTiO2 nanoparticles with BSA.

BSA absorbs light energy around 280 nm, which is not absorbedby TiO2. The excited state energy of BSA is transferred to theground state TiO2, thus TiO2 gets excited and generates conductionband electrons and valence band holes (Scheme 1). The conductionband electrons are transferred to the metal core, and then the elec-trons in the metal and holes in the valence band of TiO2 take part insubsequent redox reactions with the surface adsorbed moleculesto yield the ultimate products. The photogenerated electrons havesufficient electronegativity to reduce dioxygen to superoxide/hydroperoxide radicals, effecting the deep oxidation of a widerange of organic pollutants to degraded products. In addition, thephotogenerated holes are highly oxidizing which can split waterinto H+ and OH�, and then the produced OH� radicals can oxidizethe organic pollutants.



http://www.elsevier.com/locate/poly

Scheme 1. Interaction of colloidal AgTiO2 nanoparticles with BSA.

Fig. 1. TEM image of AgTiO2 nanoparticles.

158 A. Kathiravan et al. / Polyhedron 28 (2009) 157–161

2. Experimental

2.1. Materials

Bovine serum albumin and titanium-(triethanolaminato) iso-propoxide were purchased from Aldrich. All measurements wereperformed at ambient temperature.

2.2. Methods

2.2.1. Preparation of colloidal AgTiO2 nanoparticlesThe method of preparation of colloidal AgTiO2 nanoparticles in

water is similar to the one reported earlier [23]. Colloidal AgTiO2

nanoparticles were prepared by a one pot synthesis that involvedreduction of metal ions and hydrolysis of titanium-(triethanolam-inato)isopropoxide [TTEAIP] in dimethylformamide. The solvent,DMF, plays an important role in reducing the Ag+ ions first, fol-lowed by the slow hydrolysis of TTEAIP to form colloidal AgTiO2

nanoparticles.

2.2.2. InstrumentationSamples for spectroscopic measurements were prepared by dis-

solving bovine serum albumin in water and administering theappropriate concentration of colloidal AgTiO2 nanoparticles. Thesamples were carefully degassed using pure nitrogen gas for15 min. Quartz cells (4 x 1 � 1 cm) with high vacuum Teflon stop-cocks were used for degassing.

The fluorescence quenching measurements were carried out ina JASCO FP-6500 spectrofluorimeter. The slit width (each 5 nm)and scan rate (500 nm/min) were constantly maintained for allmeasurements. Absorption spectral measurements were recordedusing a Cary 300 UV–Vis spectrophotometer. Fluorescence lifetimemeasurements were carried out in a picosecond time correlatedsingle photon counting (TCSPC) spectrometer. The excitationsource was the tunable Ti-sapphire laser (TSUNAMI, Spectra Phys-ics, USA). The fluorescence decay was analyzed by using the soft-ware provided by IBH (DAS-6). The particle size of the preparedAgTiO2 nanoparticles were analyzed by transmission electronmicroscopy (recorded using TECNAI G2 model).


3.1. Characterization of the AgTiO2 nanoparticles

Transmission electron microscope pictures have been taken forthe prepared Ag–TiO2 nanoparticles (Fig. 1) to give an idea aboutthe particle size and surface modifications effected during dopingof the metal on TiO2 nanoparticles. The TEM pictures permit easy

differentiation of metal nanoparticles (small dark areas) and TiO2

(large bright areas), that is, an Ag nanoparticle is seen on the sur-face of a TiO2 particle as a dark dot. Also, there is the possibilityof metal particles to be incorporated into the interstitial positionsof the semiconductor particles. Further, it is observed that theaverage size of silver in the Ag–TiO2 particles is in the range 10–20 nm.

3.2. XRD characterization of AgTiO2 nanoparticles

Fig. 2 shows the XRD spectrum of AgTiO2 nanoparticles. Diffrac-tions that are attributable to the anatase phase of TiO2 crystals(101) are clearly detectable at 2h = 25� (JCPDS 21-1272) in Fig. 2.Peaks at 2h = 38.4�, 44.5� and 64.6� in AgTiO2 are assigned to(111), (200) and (220) planes of silver (JCPDS), which proves thatthe TiO2 surfaces are covered with metal particles.

3.3. Absorption characteristics of BSA–AgTiO2 nanoparticles

Fig. 3 shows the absorption spectra of BSA in the presence andabsence of colloidal AgTiO2 nanoparticles at different concentra-tions. From this study we observed that upon increasing the con-centration of colloidal AgTiO2 the absorption of BSA increasesregularly with the peak shift of around 5 nm. It is due to the

0 10 20 30 40 50 60 70 80 90

0

100

200

300

400

500

600

700

800

Cou

nts

(arb

. uni

ts)

Degree 2θ

Fig. 2. XRD pattern of AgTiO2 and nanoparticles.

0

0.15

0.3

0.45

0.6

240 290 340 390 440

Wavelength

Abs

orba

nce

0204060

0 0 .2 0.4 0 .6 0.8 1

1/[AgTiO2] x 10-4 M

1/A

-A0

Fig. 3. Absorption spectrum of BSA in the presence of colloidal TiO2 in theconcentration range 0–5 � 10�4 M. Inset is the linear dependence of 1/A � A0 on thereciprocal concentration of colloidal AgTiO2.

BSA + AgTiO2 BSA.....AgTiO2

BSA.....AgTiO2 BSA*.....AgTiO2hν

Scheme 2. Mechanism of complex formation.

0

130

260

390

520

650

285 360 435 510 585 660

Wavelength

Inte

nsit

y 0.8

1.21.6

22.4

[AgTiO2] x 10-6 M

I0/I

3 4 50 1 2

Fig. 4. Fluorescence Quenching of BSA in the presence of various concentrations ofcolloidal AgTiO2 nanoparticles, [TiO2] = 0, 1, 2, 3, 4 and 5 � 10�6 M. Insert shows theStern–Volmer plot.

A. Kathiravan et al. / Polyhedron 28 (2009) 157–161 159

adsorption of BSA on the surface of colloidal AgTiO2 nanoparticles,supported by similar observations reported earlier [24].

The equilibrium for the formation of complex between BSA andcolloidal AgTiO2 nanoparticles can be given by Eq. (1), where Kapp

represents the apparent association constant:

BSAþ AgTiO2 �Kapp

BSA � � �AgTiO2 ð1Þ

Kapp ¼½BSA � � �AgTiO2�½BSA� � ½AgTiO2�

The change in intensity of the absorption peak (280 nm) as a resultof the formation of the surface complex were utilized to obtain Kapp

according to Benesi and Hildebrand [25]:

Aobs ¼ ð1� aÞC0eBSA1þ aC0ec1 ð2Þ

where Aobs is the observed absorbance of the solution containingdifferent concentrations of colloidal AgTiO2 at 280 nm, a is the de-gree of association between BSA and AgTiO2, eBSA and ec are the mo-lar extinction coefficients at the defined wavelength (k = 280 nm) ofBSA and the formed complex, respectively, in water. Eq. (2) can beexpressed as Eq. (3), where A0 and Ac are the absorbances of BSA andthe complex at 280 nm, respectively, with a concentration of C0:


At relatively high AgTiO2 concentrations, a can be equated to(Kapp[AgTiO2])/(1 + Kapp[AgTiO2]). In this case, Eq. (3) can be chan-ged to Eq. (4):

1Aobs � A0

¼ 1Ac � A0

þ 1KappðAc � A0Þ½AgTiO2�

ð4Þ

The enhancement of absorbance at 280 nm was due to absorption ofthe surface complex, based on the linear relationship between

1/(Aobs � A0) vs reciprocal concentration of colloidal AgTiO2 with aslope equal to 1/Kapp(Ac � A0) and an intercept equal to 1/(Ac � A0)(Fig. 3, insert). The value of the apparent association constant (Kapp)determined from this plot is 5.58 � 105 M�1. Further, the reason forthe higher association constant of AgTiO2 compared to that reportedfor TiO2 [15] may be due to the larger surface area of the AgTiO2

nanoparticles.

3.4. Fluorescence quenching of BSA by colloidal AgTiO2 nanoparticles

The fluorescence quenching is described by the Stern–Volmerrelation:

I0=I ¼ 1þ Ksv½Q � ¼ 1þ Kqs0½Q � ð5Þ

where I0 and I are the fluorescence intensities of BSA in the absenceand presence of quencher, KSV is Stern–Volmer constant, kq is thebimolecular quenching rate constant, and s0 is the average lifetimeof BSA, 10�8 s [26], [Q] is the concentration of quencher. Fig. 4shows the effect of increasing the concentration of colloidal AgTiO2

on the fluorescence emission spectrum of BSA. Addition of colloidalAgTiO2 resulted in the quenching of BSA fluorescence emission andthere is no peak shift and no new peak was observed. It is noted thata complex formed between colloidal AgTiO2 and BSA is responsiblefor the quenching of BSA. According to Eq. (5), a linear plot [Inset,Fig. 4] between I0/I against [AgTiO2] was obtained and from theslope we calculated the quenching rate constant (kq) as1.64 � 1015 M�1s�1. In general, the maximum collisional quenchingconstant (kq) of various kinds of quenchers to biopolymers is2.0 � 1010 M�1s�1 [26], but for the BSA–AgTiO2 system a higherquenching rate constant (1.64 � 1015 M�1s�1) was obtained. Thisshows that the quenching is not dynamic in nature, it depends onthe formation of a complex between AgTiO2 and BSA (Scheme 2).Further the type of interaction between BSA and colloidal AgTiO2

was also confirmed by time resolved spectroscopy.


Fig. 5 shows the fluorescence decay of BSA in the absence andpresence of colloidal AgTiO2 nanoparticles. BSA exhibits singleexponential decay not only in dilute solutions but also in the pres-

1

10

100

1000

10000

15 20 25 30 35 40 45 50 55

Time (ns)

Cou

nts

Fig. 5. Fluorescence decay curve of BSA in the absence and presence of colloidalAgTiO2 in the concentration range 0–5 � 10�6 M.

-1

-0.8

-0.6

-0.4

-0.2

0

-6-5.9-5.8-5.7-5.6-5.5-5.4-5.3-5.2

log [Q]

log

[F0-

F]/

F

Fig. 6. Plot of log[(F0 � F)/F] vs. log[Q].

0

100

200

300

300 320 340 360 380 400

Wavelength

Inte

nsit

y

Fig. 7. Synchronous fluorescence spectrum of BSA with Dk = 15 nm in the absenceand presence of colloidal AgTiO2 nanoparticles.

160 A. Kathiravan et al. / Polyhedron 28 (2009) 157–161

ence of colloidal AgTiO2 nanoparticles. On increasing the concen-tration of colloidal AgTiO2 there is no change in the lifetime ofBSA. This observation shows that the quenching follows a staticmechanism. It also supports the formation of a ground state sur-face complex. (In Fig. 5 although the decay traces of BSA in the ab-sence and presence of colloidal AgTiO2 were actually plotted, thelifetime of BSA remained the same under both conditions, hencemerging of the kinetic traces was observed, and the plot looks likea single decay curve.) For static quenching, we can deduce thebinding constant (K) resulting from the formation of a ground statecomplex between fluorophore and the quencher.


The relationship between the fluorescence intensity andquenching medium can be deduced from the following formula:

nQ þ B �K

Qn � � �B ð6Þ

where B is the biomolecule with the fluorophore, Q is the quenchermolecule, Qn� � �B is the quenched biomolecules and the resultantconstant K is given by

K ¼ ½Q n � � �B�½Q �n � ½B�

ð7Þ

If the overall amount of biomolecules (bound or unbound with thequencher) is B0, then [B0] = [Qn� � �B] + [B], here [B] is the concentra-tion of unbound biomolecules. Thus the relationship between fluo-rescence intensity and the unbound biomolecule is [B]/[B0] = F/F0,that is

logF0 � F

F

� �¼ n log½Q � þ log K ð8Þ

where K is the binding constant of AgTiO2 with BSA, which can bedetermined from the plot of log[(F0 � F)/F] versus log[Q] as shownin Fig. 6. Thus we obtained the binding constant ‘‘K” as3.71 � 105 M�1 and binding sites ‘‘n” (1.06) for AgTiO2 with BSAfrom the intercept and slope of Fig. 5. The value of K obtained fromthe data of fluorescence quenching matches well with that deter-mined from the absorption spectral changes. The good agreementbetween these values of K highlights the validity of assumption pro-posed for the association between BSA and colloidal AgTiO2

nanoparticles.


To explore the structural changes of BSA by the addition of col-loidal AgTiO2, we measured the synchronous fluorescence spectraof BSA with the concentrations of colloidal AgTiO2 used for thefluorescence quenching study. Synchronous fluorescence spectra

provide information on the molecular microenvironment, particu-larly in the vicinity of fluorophore functional groups [27]. The fluo-rescence of BSA is due to the presence of tyrosine, tryptophan andphenylalanine residues. Hence spectroscopic methods are usuallyapplied to study the conformation of the serum protein. In syn-chronous fluorescence spectroscopy, according to Miller [28], thedifference between the excitation and emission wavelength(Dk = kemi � kexc) reflects the spectra of chromophores with the dif-ferent natures. With larger Dk values, such as 60 nm, the synchro-nous fluorescence of BSA is characteristic of the tryptophanresidue, and smaller Dk values, such as 15 nm, are characteristicof tyrosine [29]. The synchronous fluorescence spectra of BSA withvarious concentrations of colloidal AgTiO2 were recorded atDk = 15 nm (Fig. 7) and Dk = 60 nm (spectra not shown here). Withan increasing concentration of colloidal AgTiO2 the intensity oftyrosine decreased, and a red-shift of the emission wavelength(Fig. 7) was observed. At the same time, the tryptophan fluores-cence emission decreased regularly, but no significant change inwavelength was observed. It suggests that the interaction of colloi-dal AgTiO2 with BSA affects the conformation of the tyrosine re-gion, but not the tryptophan micro-region. The tyrosinefluorescence spectrum may represent that the conformation ofBSA has changed, leading to a strengthening of the polarity aroundthe tyrosine residues and a weakening of the hydrophobicity [30].It is important to note that colloidal AgTiO2 affects only the tyro-sine residues in the BSA moiety. This is because tyrosine containsone aromatic hydroxyl group, unlike tryptophan.

In our previous study, riboflavin was used as a sensitizer for col-loidal TiO2 [31] and it was observed that riboflavin interactedthrough the hydroxyl group of colloidal TiO2. In addition, tyrosinecan undergo an excited state ionization, resulting in the loss of aproton from the aromatic hydroxyl group. Hence it is clear thatthe presence of a hydroxyl group in the tyrosine residues may be

+-H2O

NH2

OO

HO

tyrosine

NH2

OHO

HO

conformational changes due to presence of -OH group

H2N

OHO

tryptophan

+no conformational changes due toabsence of -OH group

TiIV

HO

Ag

TiIV

HO Ag TiIV

Ag

Scheme 3. Mechanism for the interaction of AgTiO2 nanoparticles with BSA.

A. Kathiravan et al. / Polyhedron 28 (2009) 157–161 161

responsible for the interaction of BSA with AgTiO2, similar to 2,3-diazabicyclo [2.2.2] oct-2-ene (DBO) interacting with the aromatichydroxyl group of tyrosine in BSA [32], as reported by Anbazhaganand Renganathan (Scheme 3).

4. Summary

The interaction between colloidal AgTiO2 nanoparticles and BSAhas been studied by various spectroscopic measurements. The re-sults presented clearly indicate that colloidal AgTiO2 nanoparticlesquench the fluorescence emission of BSA through a static mecha-nism, which is further confirmed by the unaltered lifetime of BSAby time resolved measurements. The quenching rate constant,binding constant and number of binding sites were calculatedaccording to the relevant fluorescence quenching data. From thesynchronous fluorescence spectra, it is shown that the conforma-tional change of BSA is induced by the interaction of colloidal Ag-TiO2 nanoparticles with the tyrosine micro-region of the BSAmolecules. The binding study of drugs with nanoparticles is ofgreat importance in pharmacy, pharmacology and biochemistry.This study is expected to provide important insight into the inter-actions of the physiologically important protein BSA with metalnanoparticles. Information is also obtained about the effect of theenvironment on the BSA structure, which may be correlated toits physiological activity.

Acknowledgements

R.R. thanks DST (SR/NM/NS-16/2007, dt. 26-09-2008) (Govern-ment of India) for the NSTI Project. R.R. and A.K. thank CSIR (Ref:No. 01(2217)/08/EMR-II, dt. 06/05/2008) for the Project and Fel-lowship respectively. We are thankful to Prof. P. Ramamoorthy,NCUFP, University of Madras, Chennai for lifetime measurements.The author, S.A. thanks CSIR, New Delhi for the sanction of a majorresearch fund (CSIR Reference No. 01(2197)/07/EMR-II dated 26thNovember, 2007).

References

[1] F. Moreno, M. Cortijo, J.G. Jimenez, Photochem. Photobiol. 70 (1999) 695.[2] M. Kessler, O. Wolfbeis, Anal. Biochem. 200 (1992) 254.[3] D.D. Carter, J.X. Ho, Adv. Protein Chem. 45 (1994) 153.[4] R.E. Olson, D.D. Christ, Ann. Rep. Med. Chem. 31 (1996) 327.[5] S. Ashoka, J. Seetharamappa, P.B. Kandagal, S.M.T. Shaikh, J. Lumin. 121 (2006)

179.[6] D. Gao, Y. Tian, F. Liang, D. Jin, Y. Chen, H. Zhang, Aimin Yu, J. Lumin. 127 (2007)

515.[7] N. Zhou, Y.Z. Liang, P. Wang, J. Photochem. Photobiol. A: Chem. 185 (2006) 271.[8] B. Zhou, Z. Qi, Q. Xiao, J.X. Dong, Y.Z. Zhang, Y. Liu, J. Biochem. Biophys. Meth.

70 (2007) 743.[9] Z. Hongwei, G. Min, Z. Zhaoxia, W. Wenfeng, W. Guozhong, Spectrochim. Acta,

Part A 65 (2006) 811.[10] Y.B. Yin, Y.N. Wang, J.B. Ma, Spectrochim. Acta, Part A 64 (2006) 1032.[11] Y.J. Hu, Yi. Liu, R.M. Zhao, J.X. Dong, S.S. Qu, J. Photochem. Photobiol. A: Chem.

179 (2006) 324.[12] J. Valanciunaite, S. Bagdonas, G. Streckyte, R. Rotomskis, Photochem. Photobiol.

Sci. 5 (2006) 381.[13] U.K. Hansen, Pharmacol. Rev. 33 (1981) 17.[14] L.A. Sklar, B.S. Hudson, R.D. Simoni, Biochemistry 16 (1977) 5100.[15] A. Kathiravan, R. Renganathan, Colloids Surf. A 324 (2008) 176.[16] K. Vinodgopal, S. Hotchandani, P.V. Kamat, J. Phys. Chem. 97 (1993) 9040.[17] T. Hirakawa, P.V. Kamat, Langmuir 20 (2004) 5645.[18] V. Alt, T. Bechert, P. Steinrucke, M. Wagener, P. Seidel, E. Dingeldein, U.

Domann, R. Schnettler, Biomaterials 25 (2004) 4383.[19] A. Roucoux, J. Schulz, H. Patin, Chem. Rev. 102 (2002) 3757.[20] A. Tao, F. Kim, C. Hess, J. Goldberger, R. He, Y. Sun, Y. Xia, P. Yang, Nano Lett. 3

(2003) 1229.[21] C. Hu, Y. Lan, J. Qu, X. Hu, A. Wang, J. Phys. Chem. B 110 (2006) 4066.[22] J.L. Elechiguerra, J.L. Burt, J.R. Morones, A. Camacho-Bragado, X. Gao, H.H. Lara,

M.J. Yacaman, J. Nanobiotechnol. 3 (2005).[23] P.K. Sudeep, K. Takechi, P.V. Kamat, J. Phys. Chem. C 111 (2007) 488.[24] C. Chen, X. Qi, B. Zhou, J. Photochem. Photobiol. A: Chem. 109 (1997) 155.[25] H.A. Benesi, J.H. Hildebrand, J. Am. Chem. Soc. 71 (1949) 2703.[26] J.R. Lakowicz, G. Weber, Biochemistry 12 (1973) 4161.[27] G.Z. Chen, X.Z. Huang, J.G. Xu, Z.Z. Zheng, Z.B. Wang, Methods of Fluorescence

Analysis, 2nd ed., Science Press, Beijing, 1990.[28] J.N. Miller, Proc. Anal. Div. Chem. Soc. 16 (1979) 203.[29] J.H. Tang, F. Luan, X.G. Chen, Bioorg. Med. Chem. 14 (2006) 3210.[30] B. Klajnert, M. Bryszewska, Bioelectrochemistry 55 (2002) 33.[31] A. Kathiravan, R. Renganathan, Spectrochim. Acta A 71 (2008) 1080.[32] V. Anbazhagan, R. Renganathan, J. Lumin. 128 (2008) 1454.

Journal of Molecular Structure 921 (2009) 279–284




Photoinduced interaction between xanthene dyes and colloidal CdS nanoparticles

M. Asha Jhonsi, A. Kathiravan, R. Renganathan *



Article history:Received 24 September 2008Received in revised form 19 December 2008Accepted 5 January 2009Available online 12 January 2009

Keywords:Xanthene dyesFluorescence quenchingColloidal CdS



a b s t r a c t

Xanthene derivatives namely fluorescein, eosin, erythrosine and rose bengal were examined as sensitiz-ers for colloidal CdS nanoparticles. The interaction of these dyes with colloidal CdS nanoparticles wasstudied by absorption, infra-red, steady state and time resolved fluorescence spectroscopic measure-ments. The adsorption of dyes on the surface of colloidal CdS nanoparticles through electrostatic interac-tion was observed. This adsorption leads to increase in optical density as well as quenching of theemission intensity of dye molecules. The apparent association constant was calculated from fluorescencedata. The fluorescence quenching is attributable to electron transfer from excited state dyes to the con-duction band of colloidal CdS is established.


1. Introduction

Bulk semiconductor materials have been widely used as photo-catalysts for solar energy conversion [1,2]. An important defect ofmetal oxide semiconductors is that their photoactivity is limitedto the UV region. Therefore the photosensitization of stable,large-band gap semiconductors by visible light using dyes is a longterm goal [3,4].

The interaction between semiconductors such as CdS with or-ganic dyes is an interesting and useful phenomenon that was usedto extend their absorptive range and thus carry out photoelectro-chemical reactions under irradiation [5–8]. In this system, a dyeadsorbs directly on the surface of semiconductor particle actingas an electron donor and can transfer an electron from its excitedstate into the conduction band of the semiconductor [9]. Semicon-ductor particles of colloidal dimensions are sufficiently small toyield transparent solutions, allowing direct analysis of interfacialcharge-transfer by a fluorescence quenching technique.

Fluorescent dyes are widely employed in both qualitative andquantitative chemical and biological analysis and in other areasas well [10]. A great diversity of such dyes were used becausethe physico chemical properties of the dyes vary widely, and differ-ent combinations of properties (e.g. absorption and emission max-imum of chromophoric system, polarity, micro environmentaldependence of the fluorescence) suit different applications [11–13]. One property that is nearly always beneficial is high stability,both chemically and physically including photostability [14].

Xanthene dyes are suitable model compounds for the studies ofcharge injection, because they exhibit intense ground-state

ll rights reserved.

: +91 431 2407045.an).

absorption (maximum around 500 nm) which could make themsuitable sunlight harvesting agents [15]. These dyes are interestingmolecules with spectral luminescence properties that make themuseful as a marker, as a probe in studies of biological systems insensors etc. [16–18]. In literature photophysical properties of xan-thene dyes is well established [19,20,13].

Sensitization is achieved by adsorption of dye molecules at thesemiconductor surface which upon excitation inject an electroninto its conduction band. The first successful experiment of thistype was described by Putzeiko and Terenin [21,22]. The Dombereffect of ZnO powder in visible light was sensitized by xantheneand cyanine dyes [23]. In literature various reports are availablefor the electron injection process from excited state dyes to theconduction band of semiconductors [24–31]. Recently we have re-ported the interaction between xanthene dyes and TiO2 which fol-lows electron transfer [32]. The structure of xanthene dyes used isshown below (Scheme 1).

The present study employs the interaction between dyesnamely eosin, rose bengal, erythrosine and fluorescein with colloi-dal CdS nanoparticles and the effect of substituents has also beenstudied. It is expected that this study might lead to clues on howto use CdS for widespread applications in biochemical and photo-chemical research.

2. Experimental

2.1. Materials

The xanthene dyes were obtained from Aldrich and used with-out further purification. Cadmium chloride (CdCl2), thioacetamideand sodium hexametaphosphate were purchased from Fluka andthey were used as such. Double distilled water was used for pre-




2000

2500(111)

O

COO_

O_

O

Y

YY

YX

X X

X

X = H & Y = H ; FluoresceinX = Br & Y = H ; EosinX = I & Y = H ; ErythrosineX = I & Y = Cl; Rose bengal

Scheme 1. Structure of xanthene dyes.

280 M. Asha Jhonsi et al. / Journal of Molecular Structure 921 (2009) 279–284

paring the solutions throughout. All measurements were per-formed at ambient temperature.

2.2. Instrumental methods

Absorption spectra were recorded using Cary 300 UV/visiblespectrophotometer. Samples were prepared by dissolving dyes indouble distilled water and administering the appropriate amountsof colloidal CdS nanoparticles. The samples were carefully purgedusing pure nitrogen gas for 10 min. Quartz cells (4 � 1 � 1 cm)with high vacuum Teflon stopcocks were used for purging.

The fluorescence quenching measurements were carried outwith JASCO FP-6500 spectrofluorometer. The excitation and theemission wavelengths for all dyes are shown in Table 1. The exci-tation and emission slit width (each 5 nm) and scan rate (500 nm/min) were kept constant for all the experiments. Fluorescence life-time measurements were carried out in a picosecond time corre-lated single photon counting (TCSPC) spectrometer.

FT-IR spectra were recorded by using a JASCO-460 plus modelspectrometer using KBr pellet with the resolution of 4 cm�1.

X-ray powder diffraction patterns were recorded on a BrukerAXS B8 Discover model using Cu-Ka radiation (k = 0.154 nm) anda graphite monochromator in the diffracted beam. TiO2 samplewas in the form of powder. A scan rate of 0.05� min�1 was appliedto record a pattern in the 2h range of 2h = 20–80�.

The oxidation potential of dyes was measured in water withpotassium chloride [KCl, 0.1 M] as electrolyte. Cyclic Voltammetrywas performed on Princeton EG and G-Parc Model Potentiostat.The experimental setup consisted of a platinum working electrode,a glassy carbon counter electrode and a silver reference electrode.Reversible peak potentials were measured at different scan rates(0.05 V/s). All samples were deaerated by bubbling with purenitrogen gas for ca. 5 min at room temperature.

2.3. Preparation of colloidal CdS nanoparticles

Water-soluble colloidal CdS were synthesized according to theliterature method [33]. Ten milliliters of 0.1 M CdCl2, 10 ml of0.1 M thioacetamide (TAA) and 10 ml of 0.1 M sodium hexameta-phosphate were mixed with constant stirring. The pH of the mix-ture was adjusted to 10.4 by 0.1 M NaOH, under N2 atmosphere.The mixture was kept at room temperature for 35 min for the

Table 1Absorption (kex) and emission (kem) wavelengths of the xanthene dyes and apparentassociation constant (Kapp) from fluorescence data for dyes with colloidal CdS.

S. No. Dyes kex (nm) kem (nm) Kapp � 103 M�1

1 Rose bengal 546 574 5.922 Eosin 511 537 4.793 Erythrosine B 526 554 3.424 Fluorescein 490 514 1.02

growth of colloidal CdS. The resulting yellow colloid was dilutedto a required concentration and it was stable for over 2 weeks inthe dark stored at 4 �C. The absorption and emission spectra of pre-pared colloidal CdS is shown in Fig. 7.


3.1. Determination of particle size of colloidal CdS nanoparticles

The diameter of the prepared colloidal CdS has been determinedfrom the relationship between band gap shift (DEg) and radius (R)of quantum size particles using equation (1)

DEg ¼p2h2

2R2 1=m�e þ 1=m�h� �

� 1:8e2

eRþ Polarisation terms ð1Þ

where h is Planck’s constant, R is the radius of the particle,m�e and m�h are the effective masses of the e and h+, respectively,in the semiconductor, e is the electron charge, e is the relative per-mittivity of the semiconductor.

A value of 0.153 me was used for the reduced effective mass ofthe exciton (1/l = 1/me + 1/mh) of CdS, the columbic and polariza-tion terms in the equation are neglected. The particle size of theprepared colloidal CdS is �3.35 nm. Properties such as particle size,absorption and emission wavelengths of the prepared colloidal CdSare similar to the reported values [34,35].

3.2. XRD characterization of CdS nanoparticles

Fig. 1 shows the XRD pattern of CdS nanoparticles obtainedfrom colloidal CdS by rotary evaporation. The XRD patterns ofthe nanoparticles are considerably broadened due to very smallsize of the CdS. The XRD pattern exhibits prominent broad peaksat 2h values of 27�, 44� and 52� which are identified for cubicCdS phase, the 2h values are similar to the reported [34]. The aver-age size of the sample is in the range of 6.02 nm from full-widthand half-maximum (FWHM) of the most intense peak makinguse of the Scherrer’s equation (2),

d ¼ 0:9k=bcosh ð2Þ

where k is the wavelength of X-ray radiation, b is the FWHM in radi-ans of the XRD peak (2.585) and h is the angle of diffraction (27�).The particle size calculated from XRD method (6.02 nm) is largerthan the value obtained from UV method (�3.35 nm). This maybe due to aggregation of colloidal CdS upon aging.

3.3. Absorption characteristics

The absorption and emission wavelength of dyes is shown inTable 1. Fig. 2 shows the absorption spectrum of eosin in the pres-ence and absence of colloidal CdS. From the absorption study, we

0

500

1000

1500

10 20 30 40 50 60 70 802θ

Inte

nsit

y

(311)(220)

Fig. 1. X-ray diffraction (XRD) spectrum of CdS nanoparticles.

0

0.2

0.4

0.6

0.8

400 450 500 550 600Wavelength

Abs

orba

nce

Fig. 2. Absorption spectrum of Eosin (1 � 10�5 M) in the presence of colloidal CdSin the concentration range of 0–5 � 10�4 M in water. The arrow indicates theabsorbance increases with increasing concentration of colloidal CdS.

M. Asha Jhonsi et al. / Journal of Molecular Structure 921 (2009) 279–284 281

observed that while increasing the concentration of colloidal CdSto the eosin solution, the dye molecules gets adsorbed on the sur-face of CdS through electrostatic interaction. This leads to increasein optical density of eosin without neither peak shift and nor newpeak formation, so this type of absorption changes indicates theground state complex formation.

The other dyes (erythrosine, rose bengal and fluorescein) alsogave the same type of interaction behaviour in their absorptionband maxima (the spectra were not shown here). The possiblescattering due to colloidal nature of CdS has been omitted becauseof the fact that baseline correction was done for all the UV–visspectral measurements and also even after adding the highest con-centration of colloidal CdS the solution remains transparent. So weconfirmed that the optical density increase is only because of theinteraction between dyes and colloidal CdS not due to scattering.Similar type of interaction has been previously reported [36]. Theinteraction between dyes and CdS surface may be through carbox-ylic group and also the keto/enol group of the dyes. We have omit-ted the keto/enol interaction due to halogen atoms surrounded byketo group, so its interaction is very weak [37]. We consider onlythe adsorption of the carboxylic group on the CdS surface. This isfurther confirmed by the FT-IR characterization.

3.4. FT-IR characterization of dyes-colloidal CdS system

UV–visible measurement is not enough to investigate whichgroup interacts with the surface of semiconductor nanoparticles(dyes on colloidal CdS). So Fourier transform infrared (FT-IR) tech-nique was used to gain further information about the nature ofinteraction between them.

Fig. 3 shows the FT-IR spectra of rose bengal (solid line) and rosebengal bound CdS (broken line). The spectrum of rose bengal alone

60

70

80

90

100

4001150190026503400

Wavenumber [cm-1]

%T

Rose bengal Rose bengal-CdS

Fig. 3. FT-IR spectra of rose bengal (solid line) and rose bengal bound with colloidalCdS (broken line).

shows the C@O stretching vibration at 1620 cm�1. The stretchingfrequency is shifted to lower wave number from the normal fre-quency for carbonyl group (1700 cm�1) and is due to more electro-negative halogen substituents. This band is completely absent inthe spectrum of rose bengal bound with colloidal CdS and newband appears at 1130 cm�1. For other dyes also we observed thatthe absence of bands responsible for the carbonyl groups in the re-gion of 1650 cm�1 (the spectra were not shown here). From theseobservations we confirmed that the dye molecules adsorbed on thesurface of colloidal CdS through their carboxyl group via electro-static interaction.

In order to confirm which group of dyes is responsible for inter-action with the surface of colloidal CdS, we have done comparativeexperiments with the model compound of xanthene. In xantheneCdS system while increasing the concentration of colloidal CdSthere is no change in the absorption spectrum as shown in Fig. 4,indicating that the interaction between xanthene and the surfaceof colloidal CdS is very weak, simply xanthene can not be adsorbedon the surface of the CdS [Scheme 2]. Therefore we conclude that,dyes adsorbed on the surface of colloidal CdS through their anchor-ing group (–COO�).

3.5. Fluorescence quenching of dyes by colloidal CdS

Interaction between dyes and colloidal CdS has also been stud-ied by fluorescence quenching measurements. Fig. 5 shows the ef-fect of colloidal CdS on the fluorescence spectra of rose bengal. Thefluorescence spectra shows that increasing concentration of colloi-dal CdS leads to gradual decrease in the emission intensity of rosebengal, this shows the quenching occurs (other three dyes alsogave the same quenching behaviour, not shown here).

We can express the equilibrium between the adsorbed andunadsorbed dye molecules by using the equation (3), in this equa-tion Kapp is the apparent association constant which can be calcu-lated from the fluorescence data by the following equation (5) [38](shown in the inset of the Fig. 5) and the calculated values areshown in Table 1.

Dyeþ CdS �Kapp½Dye . . . CdS� ð3Þ

Kapp ¼½Dye . . . CdS�½Dye� � ½CdS� ð4Þ

1F0 � F

¼ 1F0 � F 0

þ 1KappðF0 � F 0Þ½CdS�

ð5Þ

where Kapp is the apparent association constant, F0 is the initial fluo-rescence intensity of dye molecules, F0 is the fluorescence intensityof CdS adsorbed dyes and F is the observed fluorescence intensity atits maximum.The Kapp decreases in the following order:

0

0.3

0.6

0.9

300 320 340 360 380 400

Wavelength

Abs

orba

nce

Fig. 4. Absorption spectrum of xanthene (1 � 10�5 M) in the presence of colloidalCdS (5 � 10�4 M) in water (note that no changes were observed on addition ofcolloidal CdS at different concentrations).

........CdS

OO_

O

Br

Br Br

Br

COO

OO_

O

Br

Br Br

Br

COO

+ CdS

a

b

CdSO + No interaction

Scheme 2. The mode of interaction between (a) eosin and (b) xanthene and the surface of colloidal CdS.

0

75

150

225

300

375

540 560 580 600 620 640 660

Wavelength

Inte

nsit

y

0

50

0.010.020.030.040.050.06

0.1 0.3 0.5 0.7

1/[CdS] x 10-4M

1/(F

0 -F)

Fig. 5. Fluorescence quenching of rose bengal (1 � 10�6 M, kex: 546 nm) withcolloidal CdS in the concentration range of 0–5 � 10�4 M in water. The insert is thelinear straight line dependence of 1/(F0-F) on the reciprocal concentration ofcolloidal CdS.

1

10

100

1000

10000

15 18 21 24 27 30

Time (ns)

Cou

nts

Fig. 6. Fluorescence decay curve of Fluorescein (1 � 10�6 M, kex: 490 nm) in theabsence and presence of colloidal CdS in the concentration range of 0–5 � 10�4 M.

Table 2Fluorescence lifetimea (in ns) of xanthene dyes at different concentrations of CdScolloids (0–5 � 10�4 M) in water.

[CdS] (�104 M) Fluorescein Eosin Erythrosine Rose bengal

0 3.65 1.25 1.02 1.801 3.46 1.18 1.18 1.692 3.70 1.19 1.05 1.733 3.60 1.20 1.14 1.814 3.55 1.24 1.08 1.645 3.51 1.18 1.03 1.62

a Lifetime given in nanoseconds obtained from time resolved measurements.


Rosebengal > Eosin > Erythrosine > Fluorescein

Among the dyes the highest apparent association constant va-lue is for rose bengal due to the presence of more electronegativecarboxyl group with four chlorine substituents in its structure, itincreases the interaction of this dye with colloidal CdS. Eosin andErythrosine, both the dyes show more or less equal associationconstant values because of the more similarity in their structures,the only difference is the halogenated substituents. In these twodyes eosin is more efficient than erythrosine due to the more elec-tronegative bromine substituents. Fluorescein shows less interac-tion with colloidal CdS is may be due to the absence ofelectronegative halogenated substituents.


The steady state measurement alone is not enough to confirmthe mechanism of fluorescence quenching whether it follows dy-namic or in static nature. In general, lifetime measurement is themost definitive method to distinguish the static and dynamicquenching process [39].

In the present work, we have studied the effect of colloidal CdSon the fluorescence lifetime of dyes. Fig. 6 shows the fluorescencedecay curve of fluorescein in the absence and presence of colloidalCdS (for other dyes the decay curves are not shown here). The dyes(fluorescein, eosin, erythrosine and rose bengal) exhibit singleexponential decay not only in the dilute solutions but also in thepresence of colloidal CdS. While increasing the concentration ofcolloidal CdS there is no change in the fluorescence lifetime ofdyes, they were almost constant (Table 2). This observation showsthe quenching follows static mechanism. It also supports theadsorption of dyes on the surface of CdS colloids and the formation

of ground state surface complex. In Fig. 6 though the decay tracesof fluorescein in the absence and presence of colloidal CdS wereactually plotted however the lifetime of fluorescein remained thesame in both conditions, hence the merging of the kinetic traceswere observed (the plot looks like a single decay curve).

3.7. Mechanism of quenching

The quenching of dyes may occur through two possible mecha-nisms such as energy or electron transfer. There is no overlap be-tween the emission spectra of dyes with the absorption spectrumof colloidal CdS (Fig. 7) and also the band-gap energy of CdS(Eg = 2.6 eV) is greater than the excited state energy of dyes (Es)in Table 3 [40]. Thus, energy transfer from excited dyes to colloidalCdS is not possible. It can therefore be concluded that the fluores-cence quenching shown in Fig. 5 is should not be caused by energytransfer.

The possible way of quenching is through electron transfer fromexcited state dye molecules to the conduction band of colloidal CdSas shown Scheme 3.

0

0.4

0.8

1.2

360 460 560 660 760Wavelength (nm)

Abs

orba

nce

-0.2

0.1

0.4

0.7

1

1.3

Inte

nsit

y

1a

2a

3a4a

2b

1b

Fig. 7. Absorption and emission spectra of colloidal CdS (1b and 2b) and theEmission spectra of dyes alone (1a-Eosin; 2a-Rose bengal; 3a-Erythrosine and 4a-Fluorescein; kex: 511, 546, 526 and 490 nm for Eosin, Rose bengal, Erythrosine andFluorescein, respectively).

Table 3Photophysical properties of xanthene dyes.

S. No. Dyes E(S) (eV)a Es/s+ (V)b Es�=sþ (V)c DGet (eV)d

1 Rose bengal 2.16 0.83 �1.33 �0.332 Eosin 2.31 1.10 �1.21 �0.213 Erythrosine B 2.24 0.95 �1.29 �0.174 Fluorescein 2.41 1.30 �1.11 �0.14

a Excited state energy of the dyes calculated from the fluorescence maximumbased on the reported method [40].

b The oxidation potentials are in water vs NHE.c Calculated from the equation, Es�=sþ = Es/s+ � Es, where Es/s+ is the oxidation

potential of the ground state dyes and Es�=sþ is the oxidation potential of the excitedstate dyes and Es is the excitation energy.

d Calculated from the Rehm–Weller equation.

[Dye..... CdS] [Dye*.....CdS]hν

[Dye*.....CdS] [Dye .....ecb CdS]

Dye + ecb CdS

_+

Dye + CdS [Dye.....CdS]

[Dye .....ecb CdS]_+ _

+

Scheme 3. Proposed electron transfer mechanism.

M. Asha Jhonsi et al. / Journal of Molecular Structure 921 (2009) 279–284 283

The feasibility of electron transfer from dyes to CdS can be ex-plained on the basis of energy level diagram based on the excitedstate oxidation potential of dyes (obtained from the oxidation po-tential of dyes and their singlet state energy according to the equa-tion Es

*/s+ = Es/s+ Es, in Table 3) and the conduction band potential of

CdS lies around 1.0 eV [41] as shown in Scheme 4. From the

1

2

0

-1

-2

CdS

CB

VB

(0.83 V)

(-1.0 V)

V vs NHE

(1.5 V)

Es/s+

Es*/s+

(1.10V)(0.95V)

(-1.33V)(-1.21 V)(-1.32 V)

νh(2.16 eV) (2.31 eV)(2.17 eV)

EosinErythrosineRose bengal

(-1.11 V)

(2.41 eV)

(1.30V)Fluorescein

Scheme 4. Schematic diagram describing the conduction and valence bandpotentials of CdS and the electron-donating energy levels of dyes.

scheme we observed that the electron transfer process is feasible.Similar type of electron transfer between dyes and semiconductorshas been previously reported [42–46].

4. Conclusion

The effect of colloidal CdS on the absorption and fluorescencespectra of dyes such as eosin, rose bengal, erythrosine and fluores-cein has been studied. The result in perturbation of the absorptionspectrum shows that the surface complex formation throughadsorption of dyes on the surface of colloidal CdS. Static natureof quenching has been confirmed by unaltered fluorescence life-time measurements. Based on the energetic calculations the mech-anism of electron transfer from excited state dyes to theconduction band of colloidal CdS was suggested.

Acknowledgments

R.R. and M.A. thank DST (Ref.: SR/NM/NS-16/2007, dt.: 26-09-08) (Government of India) for the Project and Fellowshiprespectively.

R.R. and A.K. thank CSIR, Government of India (Ref.: 01(2217)/08/EMR-II, dt.: 06/05/2008) for the project and fellowship, respec-tively. Authors also thank Dr. S. Anandan, (NIT, Trichy) and Dr. R.Ramaraj (MKU, Madurai) for providing their CV facilities.

We are thankful to Laser Spectra of Services, Bangalore for life-time measurements.

References

[1] P.V. Kamat, Chem. Rev. 93 (1993) 267.[2] C.A. Koval, J.N. Howaxd, Chem. Rev. 92 (1992) 411.[3] H. Gerischer, F. Willig, Top. Curr. Chem. 61 (1976) 31.[4] C. Wang, C. Liu, W. Wang, T. Shen, J. Photochem. Photobiol. A: Chem. 109

(1997) 159.[5] M.R. Hoffmann, S.T. Martin, W.Y. Choi, D.W. Bahnemann, Chem. Rev. 95 (1995)

69.[6] B. O’Regan, M. Graetzel, Nature 353 (1991) 37.[7] H. Xu, H. Li, K. Liu, Dyes Pigments 49 (2001) 9.[8] A. Hagfeldt, M. Graetzel, Chem. Rev. 95 (1995) 49.[9] Z.X. Zhou, S.P. Qian, S.D. Yao, Z.Y. Zhang, Dyes Pigments 51 (2001) 137.

[10] J. Gao, P. Wang, R.W. Giese, Anal. Chem. 74 (2002) 6397.[11] H.H. Wasserman, R.W. Murray, Singlet Oxygen, Academic Press, New York,

1979.[12] D.D. Hawn, N.R. Armstrong, J. Phys. Chem. 82 (1978) 1288.[13] P.V. Kamat, M.A. Fox, J. Phys. Chem. 88 (1984) 2297.[14] W. Reed, M.J. Politi, J.H. Fendler, J. Am. Chem. Soc. 103 (1981) 4591.[15] M. Hilgendorff, V. Sundstrom, J. Phys. Chem., B 102 (1998) 10505.[16] F.L. Arbeloa, P.R. Ojeda, I.L. Arbeloa, J. Lumin. 44 (1989) 105.[17] N.O.M. Petrossyan, A.N. Vodolazkaya, A. Doroshenko, J. Fluorescence 13 (2003)

235.[18] I.M. Villoslada, M. Jofre, V. Miranda, R. Gonzalez, B.L. Rivas, J. Phys. Chem. B 110

(2006) 11809.[19] G.R. Fleming, A.W.E. Knight, J.M. Morris, R.J.S. Morrison, G.W. Robinson, J. Am.

Chem. Soc. 99 (1977) 4306.[20] O.V. Aguilera, D.C. Neckers, Acc. Chem. Res. 22 (1989) 171.[21] E.K. Putzeiko, A. Terenin, Zh. Fiz. Khim. 23 (1949) 676.[22] A. Terenin, I. Akimov, J. Phys. Chem. 69 (1965) 730.[23] J. Moser, M. Graetzel, J. Am. Chem. Soc. 106 (1984) 6557.[24] C. Chen, X. Qi, B. Zhou, J. Photochem. Photobiol. A: Chem. 109 (1997) 155.[25] Z. Zhou, S. Qian, S. Yao, Z. Zhang, Radiat. Phys. Chem. 65 (2002) 241.[26] A. Kathiravan, V. Anbazhagan, M.A. Jhonsi, R. Renganathan, Spectrochim. Acta

A 70 (2007) 615.[27] H. Zhang, Y. Zhou, M. Zhang, T. Shen, Y. Li, D. Zhu, J. Colloid Interface Sci. 251

(2002) 443.[28] T. Hasobe, S. Hattori, P.V. Kamat, S. Fukuzumi, Tetrahedron 62 (2006) 1937.[29] J.M. Rehm, G.L. McLendon, Y. Nagasawa, K. Yoshihara, J. Moser, M. Gra1tzel, J.

Phys. Chem. 100 (1996) 9577.[30] Q. Wang, W.M. Campbell, E.E. Bonfantani, K.W. Jolley, D.L. Officer, P.J. Walsh, K.

Gordon, R. Humphry-Baker, M.K. Nazeeruddin, M. Gra1tzel, J. Phys. Chem. B109 (2005) 15397.

[31] H. Zhang, Y. Zhou, M. Zhang, T. Shen, Y. Li, D. Zhu, J. Phys. Chem. B 106 (2002)9597.

[32] A. Kathiravan, V. Anbazhagan, M.A. Jhonsi, R. Renganathan, Z. Phys. Chem. 221(2007) 941.

[33] X. Shen, X. Liou, Li-Ping Ye, H. Liang, Z. Wang, J. Colloid Interface Sci. 311(2007) 400.


[34] P.S. Chowdhury, P. Ghosh, A. Patra, J. Lumi. 124 (2007) 327.[35] A. Priyam, A. Chatterjee, S.K. Das, A. Saha, Chem. Commun. (2005) 4122.[36] H. Zhang, Y. Zhou, M. Zhang, T. Shen, Y. Li, D. Zhu, J. Colloid Interface Sci. 264

(2003) 290.[37] M. Hilgendorff, V. Sundstrolm, J. Phys. Chem. B 102 (1998) 10505–10514.[38] P.V. Kamat, J. Phys. Chem. 93 (1989) 859.[39] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, Plenum Press, New

York, 1983. p. 258.

[40] E.J. Shin, D. Kim, J. Photochem. Photobiol. A: Chem. 152 (2002) 25.[41] K.R. Gopidas, M. Bohorquez, P.V. Kamat, J. Phys. Chem. 94 (1990) 6435.[42] P.V. Kamat, M.A. Fox, Chem. Phys. Lett. 102 (1983) 379.[43] M. Hilgendorff, V. Sundstrolm, J. Phys. Chem. B 102 (1998) 10505.[44] J. He, F. Chen, J. Zhao, H. Hidaka, Colloids Surf. A: Physicochem. Eng. Aspects

142 (1998) 49.[45] L. Zang, M.J. Rodgers, J. Phys. Chem. B 104 (2000) 468.[46] A. Kathiravan, R. Renganathan, J. Colloid Interf. Sci. 331 (2009) 401.

This article appeared in a journal published by Elsevier. The attachedcopy is furnished to the author for internal non-commercial researchand education use, including for instruction at the authors institution

and sharing with colleagues.








An investigation on fluorescence quenching of certain porphyrinsby colloidal CdS

M. Asha Jhonsi, A. Kathiravan, R. Renganathan �



Article history:

Received 15 April 2008

Received in revised form

5 March 2009

Accepted 12 March 2009Available online 2 April 2009

Keywords:

Porphyrins

Fluorescence quenching

Colloidal CdS

a b s t r a c t

Certain porphyrin derivatives namely meso-tetraphenylporphyrin (TPP), meso-tetrakis(4-carboxyphe-

nyl)porphyrin (TCPP), meso-tetrakis(4-sulfonatophenyl)porphyrin (TSPP) were examined as sensitizers

for colloidal CdS. The interaction of these porphyrins and colloidal CdS were studied by absorption,

infrared, steady state and time resolved fluorescence spectroscopy and transient absorption techniques.

The apparent association constants (Kapp) resulting from adsorption of porphyrins on CdS surface were

calculated from both absorption and fluorescence studies and they agree well. Using all the

spectroscopic measurements we confirmed that the interaction between porphyrins and colloidal

CdS occurs through ground state complex formation and the quenching follows static mechanism.

& 2009 Elsevier B.V. All rights reserved.

1. Introduction

The photosensitization of large band gap semiconductors usingvisible light absorbing dyes has been the subject of activeinvestigation over the past 15 years [1–7]. The interactionbetween colloidal CdS with porphyrins is an interesting anduseful phenomenon to extend its absorptive range and hasapplications in the field of photocatalysis [8] non-linear opticalmaterials [9] in solar cells [10] and display devices [11]. There arevarious molecules [12–16] whose interaction with colloidal CdSyields surface modified nanoparticles that results in changingtheir absorbance properties as well as interactions with theenvironment [17]. Among the various molecules, porphyrins are ofgreat importance in the field of physical and chemical processeson colloid surfaces [18]. Extensive information accumulated overthe years on various porphyrin derivatives allow a choice for thedesign of efficient sensitizer with appropriate charge and groundand excited state redox properties [19]. There are some reportsavailable for the study of interaction between various dyes andquantum dots (QDs) which is related to the energy transfer fromQDs to the dye molecules [20–24].

Porphyrins also find use in biochemistry, catalysis andphotochemistry [25]. Xuezhong He and co workers made somecomparative studies of different interactions between isomericporphyrins and CdS nanoparticles [26]. Recently, we have reportedfluorescence quenching of meso-tetrakis(4-sulfonatophenyl)por-phyrin (TSPP) and meso-tetrakis(N-methylpyridyl)porphyrin by

colloidal TiO2 and colloidal CdS, respectively [27,28]. It was foundthat porphyrins adsorbed on colloidal semiconductor nanoparti-cles surface through their anchoring groups, and there is anelectron injection from excited state porphyrins into the conduc-tion band of colloidal nanoparticles. In this work we have reportedthe photoinduced interaction between certain porphyrins(structures shown in Scheme 1) and colloidal CdS. Importance ofmolecular structure on the interaction between porphyrins andcolloidal CdS has also been studied.

2. Experimental

2.1. Materials

All the porphyrins were purchased from Aldrich and usedwithout further purification. Cadmium chloride (CdCl2)was obtained from Fluka and N,N0-dimethylformamide (DMF)was distilled under diminished pressure prior to use and hydrogensulphide gas was generated in a Kipp’s apparatus from ferroussulphide and hydrochloric acid.

2.2. Instrumentation

2.2.1. Spectroscopic measurements

Absorption spectra were recorded using Cary 300 UV�visiblespectrophotometer. The samples were carefully purged by usingpure nitrogen gas for 10 min. Quartz cells (4�1�1 cm3) withhigh-vacuum Teflon stopcocks were used for purging.

Steady state fluorescence quenching measurements werecarried out with JASCO FP-6500 spectrofluorimeter. The excitation

ARTICLE IN PRESS


journal homepage: www.elsevier.com/locate/jlumin

Journal of Luminescence

0022-2313/$ - see front matter & 2009 Elsevier B.V. All rights reserved.

doi:10.1016/j.jlumin.2009.03.013

� Corresponding author. Tel.: +914312407053; fax: +91431 2407045.

E-mail address: [email protected] (R. Renganathan).

Journal of Luminescence 129 (2009) 854–860


and emission slit width (each 5 nm) and scan rate (500 nm/min)were kept constant for all the experiments. The excitation andemission wavelengths of porphyrins are shown in Table 1.

Fourier transform infrared (FT-IR) spectra were recorded byusing a JASCO-460 plus model spectrometer using KBr pellet withthe resolution of 4 cm�1.

Fluorescence lifetime measurements were carried out in apicosecond time correlated single photon counting (TCSPC)spectrometer. The excitation source is the tunable Ti–sapphirelaser (TSUNAMI, Spectra Physics, USA). The fluorescence decaywas analyzed by using the software provided by IBH (DAS-6).

The Nd-YAG laser source produces nanosecond pulses (8 ns) of355 nm light and energy of the laser pulse was around 150 mJ.Dichroic mirrors were used to separate the third harmonic fromthe second harmonic and the fundamental output of the Nd-YAGlaser. The monitoring source was a 150 W pulsed xenon lamp,which was focused on the sample at 901 to the incident laserbeam. The beam emerging through the sample was focused on toa Czerny-Turner monochromator using a pair of lenses. Thedetection was carried out using a Hamamatsu R-928 photomul-tiplier tube. The transient signals were captured with an Agilentinfiniium digital storage oscilloscope interfaced to a computer.

2.2.2. Cyclic voltammetric measurements

The oxidation potential of meso-tetraphenylporphyrin (TPP),TSPP and meso-tetrakis(4-carboxyphenyl)porphyrin (TCPP) weremeasured in DMF with tetrabutylammonium perchlorate (TBAP)(0.1 M) as electrolyte. The experimental setup consisted of aplatinum working electrode, a glassy carbon counter electrodeand a silver reference electrode. Reversible peak potentials weremeasured at different scan rates (0.05 V/s). All samples weredeaerated by bubbling with pure nitrogen gas for ca. 5 min atroom temperature.

2.3. Preparation of colloidal CdS

Although there are several methods known to generatenanosized cadmium sulphide, it is still desirous to have a verysimple one-step preparation method reported by Khanna et al.[29]. Typically, 0.02 M solution of CdCl2 in DMF was used for thepreparation, which was taken in a round bottom flask and

hydrogen sulphide gas was passed for a few seconds withcontinuous stirring at room temperature in air. No protectingagents were added for the preparation of colloidal CdS. Thereaction mixture was stirred at the same temperature for fewminutes to obtain fluorescent yellow solution. The resultingyellow colloid was excess in Cd2+ which was diluted to a requiredconcentration and was stored at 4 1C. The solutions were analyzedby UV-visible spectroscopy to identify the presence of nanopar-ticles.

The absorption studies showed the formation of nanosized CdShaving absorption in the range of 425–450 nm in DMF in a muchdiluted solution (typically o1�10�4 M). It is well understoodwith available literature that milder experimental conditionsfavour the formation of smaller particle size and the stability ofthese nanoparticles. The stability of the prepared colloidal CdSdepended on the reaction time. In the present study where thereaction time was kept o5 min resulted in absorption bands inthe visible region indicating better quality of CdS quantum dots.The absorption bands of these samples are stable for 24–48 h.Furthermore, these solutions have shown bright light emissionwhen illuminated with ultraviolet lamp in the region of 520 nm(Fig. 9). Upon prolonged exposure (more than a week in somecases) to normal environment it became cloudy which isconsistent with the observations made by Chen et al. [30].

The particle size of the prepared colloidal CdS has beendetermined from the relationship between band gap shift (DEg)and radius (R) of quantum size particles using Eq. (1)

DEg ¼p2h2

2R2½1=me � þ1=mh��

1:8e2

�Rþ Polarisation terms (1)

where h is Planck’s constant, R is the radius of the particle, me*and mh* are the effective masses of the e� (electron) and h+ (hole),respectively, in the semiconductor, e is the electronic charge, e isthe relative permittivity of the semiconductor.

A value of 0.153 me was used for the reduced effective mass ofthe exciton (1/m ¼ 1/me+1/mh) of CdS, the columbic and polariza-tion terms in the equation are neglected. The calculated size of theprepared colloidal CdS is 3.35 nm.


3.1. Absorption characteristics of porphyrins with colloidal CdS

Fig. 1 shows the normalized absorption and emission spectra ofporphyrins (TCPP, TSPP and TPP) in DMF. Fig. 2 shows the steadystate absorption spectrum of TCPP in the absence and presence ofcolloidal CdS in different concentrations. The absorption spectrareveals that while increasing CdS concentration, the opticaldensity of the mixture regularly increases (Fig. 2, insert b) butthere is no new peak formation and no peak shift was observed. Itis due to the partial adsorption of TCPP on the surface of colloidalCdS to form a surface complex as shown in Eq. (2) (other twoporphyrins such as TSPP and TPP also gave the similar type of

ARTICLE IN PRESS

Scheme 1. Structure of Porphyrins.

Table 1

Absorption (lexi) and Emission (lemi) wavelengths of the porphyrins (TCPP, TSPP

and TPP) and apparent association constants of porphyrins with colloidal CdS

obtained from both absorption and fluorescence study.

S. No Sensitizer lexi (nm) lemi (nm) Kapp�102 M�1

Absorbance Fluorescence

1 TCPP 419 649 16.8 51.54

2 TSPP 418 652 9.75 31.05

3 TPP 416 650 1.82 12.48

M. Asha Jhonsi et al. / Journal of Luminescence 129 (2009) 854–860 855


observations, spectra are not shown here). Similar type ofadsorption behavior of charged molecules on the surface ofcolloidal TiO2 was reported by Hilgendorff et al. [31].

The equilibrium for the formation of complex betweenporphyrins and CdS is defined by Eq. (2), where Kapp is theapparent association constant,

Porphyrinþ CdS ÐKapp

Porphyrin . . .CdS (2)

Kapp ¼½Porphyrin . . .CdS�

½Porphyrin� � ½CdS�(3)

The Kapp values are calculated by the method reported by Benesiand Hildebrand method [32] using the following equation:

Aobs ¼ ð1� aÞC0�porphyrin1þ aC0�c1 (4)

where Aobs is the observed absorbance of the porphyrin solutioncontaining different concentrations of colloidal CdS at 419, 418and 416 nm; a is the degree of association between porphyrinsand CdS; eporphyrin and ec are the molar extinction coefficients atthe defined wavelengths for porphyrins and the formed complex,respectively. Eq. (4) can be expressed by Eq. (5), where the A0 andAc are the absorbances of porphyrins and the complex at 419, 418

and 416 nm, respectively, with the concentration of C0

Aobs ¼ ð1� aÞA0 þ aAc (5)

At relatively high CdS concentrations, a can be equated to(Kapp [CdS])/(1+Kapp[CdS]). In this case, Eq. (5) can be expressedas Eq. (6)

1

Aobs � A0¼

1

Ac � A0þ

1

KappðAc � A0Þ½CdS�(6)

Therefore, if the enhancement of absorbance at 419, 418 and416 nm was due to absorption of complex, one would expect alinear relationship between 1/(Aobs�A0) and the reciprocalconcentration of colloidal CdS (Fig. 2, insert a) with a slope equalto 1/Kapp(Ac�A0) and an intercept equal to 1/(Ac�A0).The calculated Kapp values from the straight line are shown inTable 1 and the values decreased in the following order:

TCPP4TSPP4TPP

TCPP and TSPP show more Kapp values due to the presence ofmore electronegative carboxylic and sulphonyl groups on theirperiphery, so they have higher electrostatic interaction withcolloidal CdS surface which is having excess Cd2+. TPP shows verylow Kapp value because of the absence of any charged anchoring

ARTICLE IN PRESS

Fig. 1. Normalized absorption and emission spectra of porphyrins (1�10�5 M) in DMF (TSPP—black, TCPP—red and TPP—blue in colors). Inset shows the ‘‘Q’’ bands of

porphyrins in the wavelength region of 450–750 nm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this

article.)

Fig. 2. Absorption spectrum of TCPP (1�10�5 M) in the presence of colloidal CdS in the concentration range of 0–5�10�4 M in DMF. Insets are the (a) linear dependence of

1/A�A0 vs. reciprocal concentration of colloidal CdS and (b) increase in absorbance at 419 nm.

M. Asha Jhonsi et al. / Journal of Luminescence 129 (2009) 854–860856


group in its structure, hence anchoring groups has significanteffect on the adsorption process.

3.2. FT-IR characteristics

Absorption spectroscopic measurement is not enough toinvestigate the molecular structure of adsorbed porphyrins onthe CdS surface. So Fourier transform infrared technique was usedto gain further information about the nature of interactionbetween them.

Fig. 3 (a) shows the FT-IR spectra of TCPP (solid line) and TCPPbound colloidal CdS (broken line). The spectrum of pure TCPPshows the CQO stretching vibration at 1693 cm�1. This band iscompletely absent in the spectrum of TCPP bound colloidal CdSand new band appears at 1643 cm�1. For TSPP also we observedthe absence of bands responsible for the sulphonato group in theFT-IR spectrum of TSPP bound colloidal CdS in the region of

1480 cm�1, Fig. 3(b). There is no significant change observedbetween the FT-IR spectra of TPP alone and TPP with colloidal CdS,Fig. 3(c). From these observations it is confirmed that theinteraction of porphyrins with colloidal CdS surface occursthrough anchoring groups such as COO� and SO3

�.

3.3. Steady state fluorescence characteristics

The effect of colloidal CdS on the fluorescence spectra ofporphyrins has also been examined. Fig. 4 shows the steady statefluorescence spectrum of TSPP in the absence and presence ofvarious concentrations of colloidal CdS. As seen from figure, weobserved a gradual decrease in emission intensity of TSPP uponincreasing the concentration of colloidal CdS (other porphyrinsalso gave the same fluorescence behavior, not shown here).

Fluorescence quenching can be described by Stern–Volmerrelation

I0=I ¼ 1þ KSV½Q � (7)

where I0 and I are the fluorescence intensities of porphyrins in theabsence and presence of quencher, respectively, KSV is theStern–Volmer constant and [Q] is the concentration of quencher.

The corresponding Stern–Volmer plots for the effect ofcolloidal CdS on the steady state fluorescence of porphyrins arecollectively indicated in Fig. 5. The curvature (non-linear) of S-Vplots indicated that the quenching follows static mechanism

ARTICLE IN PRESS

Fig. 3. FT-IR spectra of (a) TCPP (solid line) and TCPP with colloidal CdS (broken

line), (b) TSPP (solid line) and TSPP with colloidal CdS (broken line) and (c) TPP

(solid line) and TPP with colloidal CdS (broken line).

Fig. 4. Steady state fluorescence quenching of TSPP (1�10–5 M) by colloidal CdS in

the concentration range of 0–5�10�4 M in DMF. Inset is the comparison linear

dependence of 1/F0�F vs. reciprocal concentration of colloidal CdS for all

porphyrins.

Fig. 5. Comparison of Stern–Volmer plots for the steady state fluorescence

quenching of porphyrins (TCPP, TSPP and TPP) by colloidal CdS in the concentration

range of 0–5�10�4 M in DMF.



through ground state complex formation and it may rule out thedynamic mechanism.

From the fluorescence quenching data also, we can calculatethe apparent association constant (Kapp) by using the followingEq. (8),

1

F0� F¼

1

F0� F 0þ

1

KappðF0� F 0Þ½CdS�

(8)

where Kapp is the apparent association constant, F0 is the initialfluorescence intensity of porphyrins, F0 is the fluorescenceintensity of CdS adsorbed porphyrins and F is the observedfluorescence intensity in their maximum. The plot of 1/F0

�F vs. 1/[CdS] for all the three porphyrins are collectively shown in theinsert of Fig. 4. There is a good linear relationship between 1/F0

�F

and the reciprocal concentration of colloidal CdS. From the slope,the Kapp has been assessed and the calculated values are shown inTable 1. The value of Kapp obtained from the data of fluorescencemeasurement matches well with that determined from theabsorption study. The good agreement between these values ofKapp highlighted the validity of assumption proposed for theassociation between porphyrins and colloidal CdS.

3.4. Time resolved fluorescence characteristics

Fluorescence lifetime measurement is useful for understandingthe type of interaction in the colloidal semiconductor�dyesystems. For example, it has been shown earlier [33] that thesensitizer molecules adsorbed on TiO2 surface had a significantlyshorter excited singlet lifetime than in homogeneous solution andthis decrease in lifetime could be correlated with charge injectionprocess. In general, the measurement of fluorescence lifetime isthe most definitive method to distinguish static and dynamicquenching processes [34].

In the present work we have studied the effect of colloidal CdSon the fluorescence lifetime of porphyrins. Fig. 6 shows the timeresolved fluorescence decay of TCPP in the absence and presenceof colloidal CdS. The fluorescence of porphyrins exhibitmonoexponential decay not only in the dilute solutions but alsoin the presence of colloidal CdS. While increasing theconcentration of colloidal CdS there is no change in thefluorescence lifetime of the TCPP, it was almost constant (Fig. 7,the fluorescence decay for other two porphyrins TSPP and TPP areas same as TCPP, not shown here). This result resembles withthe steady state measurement which indicates that thefluorescence quenching of porphyrins belongs to the staticquenching mechanism.

3.5. Mechanism of quenching

The fluorescence quenching can be enlightened by differentmechanisms such as electron transfer or energy transfer from theporphyrins to the colloidal CdS and also the formation of acomplex (Eq. (2)) which has no fluorescence.

The band gap energy of CdS semiconductor (Eg ¼ 2.6 eV) [35] isgreater than the excited state energy of porphyrins (�1.90 eV) [36]and no fluorescence emission of the porphyrins is absorbed bycolloidal CdS (Fig. 8). Thus, the energy transfer from excited stateporphyrins to the colloidal CdS is ruled out. It can therefore beconcluded that the fluorescence quenching shown in Fig. 4 is notcaused by energy transfer (Fig. 9).

The other possible mechanism is electron transfer from theexcited state of porphyrins to the conduction band of colloidal CdSas shown in Eq. (9)

Porphyrin� þ CdS! Porphyrinþ þ CdSðe�Þ (9)

This inference can be further defined by the difference betweenexcited state oxidation potential of porphyrins and energy level ofthe conduction band potential of CdS. With the use of oxidationpotential of porphyrins and the excited state energy (excited stateenergy has been calculated from the emission wavelength of theporphyrins by reported method [36]), according to the equationEs*/s+ ¼ Es/s+�Es, one can obtain the oxidation potential of theexcited state porphyrins as shown in Table 2. Since these levels

ARTICLE IN PRESS

Fig. 6. Fluorescence decay of TSPP (1�10�5 M, lexi ¼ 400 nm) by colloidal CdS in

the concentration range of 0–5�10�4 M in DMF.

Fig. 7. Fluorescence lifetime of porphyrins with various concentrations of CdS

colloid (0–4�10–4 M) in DMF.

Fig. 8. Time-resolved transient absorption spectra recorded after 400 nm laser

pulse excitation of TCPP in absence and presence of colloidal CdS in DMF. Inset is

the time decay plot for TCPP (solid line) and TCPP–CdS (broken line).



are energetically higher (Scheme 2) than the energy level of theconduction band of CdS (�1.0 V vs. NHE) [37] and hence theelectron transfer path in Eq. (9) is not allowedthermodynamically. This inference is the major differencebetween our porphyrins�CdS system and the earlier reportedPorphyrins�TiO2 [27] system. Similar type of reports is availablein literature for example fluorescein ester derivatives with CdScolloids system by Zhang et al. [35].

From the absorption study of porphyrins with colloidalCdS we observed that there is an increase in absorption of

porphyrinyCdS complex in the region below 400 nm (Fig. 2) butno emission peak appeared in the fluorescence spectra. Theseresults indicate that the complex (porphyrinsyCdS) has nofluorescence and the observed fluorescence is only from theuncomplexed porphyrins. So in the present system the fluores-cence quenching of porphyrins (TCPP, TSPP and TPP) is mainly dueto the ground state complex formation (Eq. (2)) which has nofluorescence.

3.6. Transient absorption characteristics

It has been shown earlier [38] that laser flash photolysis couldbe a convenient technique to investigate the electron transferprocesses in colloidal semiconductor systems. If the observedfluorescence quenching of porphyrins results in an electroninjection into the conduction band of colloidal CdS (as shown inEq. (9)) one would expect to see the production of cation radical ofporphyrin. The results of the flash photolysis experiments carriedout with the excitation of porphyrin are described below.

In order to study the interaction between excited porphyrinsand colloidal CdS, we measured the transient absorption spectra.Nanosecond laser flash photolysis experiments were carried outusing 400 nm laser pulse as the excitation source. The transientabsorption spectra recorded 20ms after laser pulse excitation forTCPP in the absence and presence of colloidal CdS are shown inFig. 8. TCPP upon excitation with the laser pulse shows transientwhich exhibited a complete decay with a lifetime of 46.5ms(insert of Fig. 8) at 450 nm was attributed to the triplet TCPP. Thebleaching in the 420 nm region indicates the depletion ofporphyrin absorption.

The presence of colloidal CdS (5�10�4 M) did not produce anychanges such as peak shift and new species formation in thetransient absorption recorded in the microsecond time scale, sothere is no excited state electron transfer in the presence ofcolloidal CdS. There is only change in the lifetime of TCPP (9.85ms)in the presence of colloidal CdS. This further confirmed theargument made that the quenching does not follow electrontransfer from excited porphyrins to the conduction band ofcolloidal CdS.

3.7. Importance of molecular structure in energy conversion

implications

In recent years, many researchers have tried to design anddevelop suitable low cost organic dyes, which can be used in solarcells with higher efficiency. Molecular structure of the organicdyes is an important factor to get higher efficiency [39–41]. In ourprevious studies [42–44] we have observed that even a smallstructural change can increase the electron injection efficiency ofxanthene dyes to semiconductor nanoparticles. In the presentinvestigation we observed that TCPP shows more efficiencythan the others due to its structure containing more electro-negative carboxyl group as substituents for interaction with thecolloidal CdS.

4. Conclusion

The interaction of porphyrins with colloidal CdS has beenstudied by using various spectroscopic techniques. Adsorption ofporphyrins on the surface of colloidal CdS has been observed andconfirmed by FT-IR measurements. The apparent associationconstant (Kapp) was calculated from absorption changes as wellas fluorescence quenching data. The fluorescence quenching isdue to ground state complex formation which has no fluorescence

ARTICLE IN PRESS

Fig. 9. Normalized absorption, emission spectra of colloidal CdS (solid and dotted

line respectively) and the emission spectra of porphyrins (solid lines) in DMF.

Table 2Photophysical properties of porphyrins like fluorescence lifetime (t), excited state

energy (Es), oxidation potential in ground (Es/s+) and excited state (Es*/s+).

S. No Porphyrins t (ns)a E(S) (eV)b Es/s+ (V)c Es*/s+ (V)d

1 TCPP 10.5 1.91 1.08 �0.83

2 TSPP 10.9 1.90 0.98 �0.92

3 TPP 10.6 1.90 1.02 �0.88

a Obtained from time-resolved measurementsb Singlet state energy of the porphyrins calculated from the fluorescence

maximum based on the reported method [36].c The oxidation potentials in DMF vs. NHE.d Calculated from the equation, Es*/s+ ¼ Es/s+�Es, where Es/s+ is the oxidation

potential of the ground state porphyrins and Es*/s+ is the oxidation potential of the

excited state porphyrins and Es is the excitation energy.

Scheme 2. Diagram illustrating the energetics of porphyrins and colloidal CdS.



and the static nature of quenching mechanism was confirmedby ground and excited state absorption and fluorescencemeasurements.

Acknowledgments

R.R. and M.A. thank DST (Ref: SR/NM/NS-16/2007, dt.: 26-09-08) (Government of India) for the project and fellowshiprespectively. Authors also thank DST-FIST and UGC-SAP forspectrofluorimeter facility in the School of Chemistry, Bharathi-dasan University, Trichy and Dr. S. Anandan, (NIT, Trichy) forproviding CV facilities. We are thankful to Prof. P. Ramamoorthy,NCUFP, University of Madras, Chennai for lifetime measurements.

References

[1] C.A. Koval, J.N. Howaxd, Chem. Rev. 92 (1992) 411.[2] C. Wang, C. Liu, W. Wang, T. Shen, J. Photochem. Photobiol. A: Chem. 109

(1997) 159.[3] B. O’Regan, M. Gratzel, Nature 353 (1991) 737.[4] J.Z. Zhang, R.H. O’Neil, T.W. Roberti, J. Phys. Chem. 98 (1994) 3859.[5] A. Hagfeldt, M. Gratzel, Chem. Rev. 95 (1995) 49.[6] E.K. Putzeiko, A. Terenin, Z. Fiz. Khim. 23 (1949) 676.[7] J. Moser, M. Gratzel, J. Am. Chem. Soc. 106 (1984) 6557.[8] S. Shiojiri, T. Hirai, I. Komasawa, J. Chem. Soc. Chem. Commun. (1998) 1439.[9] K. Murakoshi, H. Hosokawa, M. Saitoh, Y. Wada, H. Mori, S. Yanagida, J. Chem.

Soc., Faraday Trans. 94 (1998) 579.[10] J. Britt, C. Ferekides, Appl. Phys. Lett. 62 (1993) 2851.[11] S. Schmitt-Rink, D.S. Chemla, D.A.B. Miller, Adv. Phys. 38 (1989) 89.[12] A. Datta, A. Saha, A.K. Sinha, S.N. Bhattacharyya, S. Chatterjee, J. Photochem.

Photobiol. B: Biol. 78 (2005) 69.[13] A. Datta, S. Chatterjee, A.K. Sinha, S.N. Bhattacharyya, A. Saha, J. Lumin. 121

(2006) 553.[14] L. Tan, L.Y. Liu, Q.J. Xie, Y. Zhang, S. Yao, Anal. Sci. 20 (2004) 441.[15] C. Chen, Xiaoyan Qi, B. Zhou, J. Photochem. Photobiol. A. Chem. 109 (1997)

155.[16] Z. Zhou, S. Qian, S. Yao, Z. Zhang, Radiat. Phys. Chem. 65 (2002) 241.

[17] C.B. Mou, D.M. Chen, X.Y. Wang, Chem. Phys. Lett. 179 (1991) 237.[18] K. Kalyanasundaram, N. Vlachopoulos, V. Krishnan, J. Phys. Chem. 91 (1987)

2342.[19] X.Z. He, Y.L. Zhou, M.H. Zhang, Dyes Pigments 39 (1998) 173.[20] E. Zenkevich, T. Blaudeck, M. Abdel-Mottaleb, F. Cichos, A. Shulga, C. von

Borczyskowski, Int. J. Photoenergy (2006) 1 Article ID 90242.[21] E. Zenkevich, F. Cichos, A. Shulga, E.P. Petrov, T. Blaudeck, C. von

Borczyskowski, J. Phys. Chem. B 109 (2005) 8679.[22] S. Kang, M. Yasuda, H. Miyasaka, H. Hayashi, M. Kawasaki, T. Umeyama,

Y. Matano, K. Yoshida, S. Isoda, H. Imahori, ChemSusChem 1 (2008) 254.[23] I. Potapova, R. Mruk, C. Hubner, R. Zentel, T. Basche, A. Mews, Angew. Chem.

117 (2005) 2490.[24] A.R. Clapp, I.L. Medintz, J.M. Mauro, B.R. Fisher, M.G. Bawendi, H. Mattoussi,

J. Am. Chem. Soc. 126 (2004) 301.[25] F.J. Vergeldt, R.B.M. Koehorst, A.V. Hoek, J. Phys. Chem. 99 (1995) 4397.[26] X. He, Y. Zhou, Y. Zhou, M. Zhang, T. Shen, J. Colloid Interface Sci. 225 (2000)

128.[27] A. Kathiravan, V. Anbazhagan, M.A. Jhonsi, R. Renganathan, Spectrochim. Acta

A 70 (2007) 615.[28] M. Asha Jhonsi, R. Renganathan, Z. Phys. Chem. 222 (2008) 1601.[29] P.K. Khanna, R. Gokhale, V.V.V.S. Subbarao, Mater. Lett. 57 (2003) 2489.[30] W. Chen, Z. Lin, Z. Wang, Y. Xu, L. Lin, J. Mater. Sci. Technol. 14 (1998) 389.[31] M. Hilgendorff, V. Sundstrom, J. Phys. Chem. B 102 (1998) 10505.[32] H.A. Benesi, J.H. Hildebrand, J. Am. Chem. Soc. 71 (1949) 2703.[33] P.V. Kamat, J.P. Chauvet, R.W. Fessenden, J. Phys. Chem. 90 (1986) 1389.[34] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, Plenum press, New

York, 1983, 258.[35] H. Zhang, Y. Zhou, M. Zhang, T. Shen, Y. Li, D. Zhu, J. Colloid Interface Sci. 264

(2003) 290.[36] E.J. Shin, D. Kim, J. Photochem. Photobiol. A: Chem. 152 (2002) 25.[37] P.V. Kamat, J. Phys. Chem. 93 (1989) 859.[38] P.V. Kamat, J.P. Chauvet, R.W. Fessenden, J. Phys. Chem. 90 (1986) 1389.[39] K. Hara, H. Sugihara, L.P. Singh, A. Islam, R. Katoh, M. Yanagida, K. Sayama,

S. Murata, H. Arakawa, J. Photochem. Photobiol. A: Chem. 145 (2001)117.

[40] K. Hara, H. Horiuchi, R. Katoh, L.P. Singh, H. Sugihara, K. Sayama, S. Murata,M. Tachiya, H. Arakawa, J. Phys. Chem. B 106 (2002) 374.

[41] K. Hara, T. Sato, R. Katoh, A. Furube, Y. Ohga, A. Shinpo, S. Suga, K. Sayama,H. Sugihara, H. Arakawa, J. Phys. Chem. B 107 (2003) 597.

[42] A. Kathiravan, V. Anbazhagan, M.A. Jhonsi, R. Renganathan, Z. Phys. Chem.221 (2007) 941.

[43] M. Asha Jhonsi, A. Kathiravan, R. Renganathan, J. Mol. Struct. 921 (2009) 279.[44] A. Kathiravan, R. Renganathan, J. Colloid Interface Sci. 331 (2009) 401.

ARTICLE IN PRESS











Photosensitization of colloidal TiO2 nanoparticles with phycocyanin pigment

A. Kathiravan, R. Renganathan *



Article history:Received 20 December 2008Accepted 7 March 2009Available online 8 April 2009

Keywords:TiO2 colloidFluorescence quenchingPhycocyanin

a b s t r a c t

Bluish phycocyanin is a water soluble pigment having strong visible light absorption at 615 nm. Thedynamics of photoinduced electron injection from phycocyanin to colloidal TiO2 nanoparticles has beenstudied by absorption, FT-IR, steady state and time-resolved fluorescence spectroscopic methods. Thephycocyanin adsorbed on the surface of colloidal TiO2 nanoparticles, the apparent association constant(Kapp) for the association between colloidal TiO2 nanoparticles and phycocyanin was measured from bothabsorption changes (Kapp = 4.01 � 102 M�1) and fluorescence quenching data (Kapp = 5.20 � 102 M�1). Thevalue of Kapp obtained from the data of fluorescence quenching matches with that determined from theabsorption spectral changes. The good agreement between these values of Kapp highlighted the validity ofassumption proposed for the association between phycocyanin and colloidal TiO2 nanoparticles. The freeenergy change (DGet) for electron injection has been calculated by applying Rehm–Weller equation. Elec-tron injection from excited state phycocyanin into the conduction band of TiO2 is suggested.

� 2009 Elsevier Inc. All rights reserved.

1. Introduction

Solar energy conversion based on dye-sensitization of wideband gap nanocrystalline semiconductor film is an area of intenseinvestigation [1–5]. The most efficient dye-sensitized solar cells(DSSCs) to date are based on ruthenium-containing metallorganicdyes adsorbed on nanocrystalline TiO2, the best of which have beenreported to convert solar energy to electrical energy with an effi-ciency of 10–11% [6,7]. Photoexcitation of the N3 dye results inan intramolecular metal-to-ligand charge-transfer transition. Thephotoexcited electrons located in the bipyridyl ligands can be effi-ciently injected into the conduction band of the TiO2 electrode onan ultrafast time scale via carboxyl groups anchored to the TiO2

surface. Conversely, recombination between the electrons injectedinto TiO2 and the cations of the N3 dye is a slow process [8], appar-ently due to the large separation between the TiO2 and the Ru3+

imposed by the bipyridyl ligands. Organic dyes have also been uti-lized as photosensitizers in DSSCs. Organic dyes have severaladvantages as photosensitizers: (a) they are cheaper than Ru com-plexes, (b) they have large absorption coefficients due to intramo-lecular d � d* transitions, and (c) there are no concerns aboutlimited resources, because they do not contain noble metals suchas ruthenium. DSSCs based on metal-free organic dyes [9–26], por-phyrin dyes [27–32] and natural dyes [33–35] have been studiedand developed. Since the efficiencies of DSSCs have not yet ap-proached the theoretical limit and are not competitive with themore expensive silicon-based solar cells, their main advantage ofcost-effectiveness depends on the utilization of cheap and readily

available sensitizer dyes. The use of nontoxic natural pigments assensitizers would definitely enhance the environmental and eco-nomic benefits of this alternative form of solar energy conversion.

The phycobiliproteins are antennae-protein pigments involvedin light harvesting in cyanobacteria, rhodophytes, cryptomonadsand cyanelles [36]. In cyanobacteria and red algae, the phycobili-proteins are organized in supramolecular complexes called phy-cobilisomes which are assembled in regular arrays on the outersurface of the thylakoid membranes. Phycobiliproteins are oligo-meric and built up from chromophore bearing polypeptidesbelonging to the a and b families of polypeptides [37]. The colorsof phycobiliproteins originate mainly from covalently boundprosthetic groups that are open-chain tetrapyrrole chromophoresnamely phycobilins (possessing A, B, C and D rings). They areeither blue colored phycocyanobilin (PCB), red colored phycoery-throbilin (PEB), yellow colored phycourobilin (PUB) and purplecolored phycobiliviolin (PXB), also named cryptoviolin. Thesechromophores are generally bound to the polypeptide chain atconserved positions either by one cysteinyl thioether linkagethrough the vinyl substituent on the pyrrole ring A of the tetra-pyrrole or occasionally by two cysteinyl thioether linkagesthrough the vinyl substituent on both A and D pyrrole rings[38]. Four main classes of phycobiliproteins exist: Allophycocya-nin (APC, bluish green), phycocyanin (PC, blue), phycoerythrin(PE, purple) and phycoerythrocyanin (PEC, orange) havingabsorption in the range of 650–655 nm, 615–640 nm, 565–575 nm and 575 nm, respectively, and emit light at 660 nm,637 nm, 577 nm and 607 nm, respectively [39]. Phycobiliproteinsare used as colorants in food, cosmetic and pharmaceuticalindustry [40], possess curative properties and used as fluores-cence tags in biomedical research [41,42].

0021-9797/$ - see front matter � 2009 Elsevier Inc. All rights reserved.doi:10.1016/j.jcis.2009.03.076

* Corresponding author. Fax: +91 431 2407045.E-mail address: [email protected] (R. Renganathan).

Journal of Colloid and Interface Science 335 (2009) 196–202


Journal of Colloid and Interface Science

www.elsevier .com/locate / jc is


The choice of phycocyanin pigment (Scheme 1) is due to its sig-nificant visible light absorption and the presence of –COOH groupwhich could serve as an anchoring group between the dye and TiO2

surface. The spectral properties, such as (i) it contains multiple bi-lin chromophores and hence high absorbance coefficients over awide region of the visible spectra (e = 2.5 � 106 M�1 cm�1 at615 nm); (ii) high fluorescence quantum yield (U = 0.8) indepen-dent of pH; (iii) strong absorption at 615 nm and strong emissionat 642 nm. It extends well into the red region of the visible spec-trum, where interference from biological molecules is minimal;(iv) large stokes shift that minimizes interferences from Rayleighand Raman scattering and other fluorescing species; (v) highlysoluble in aqueous solutions; and (vi) stable in solution as wellas solid phase, thus can be stored for long periods. The phycocya-nin pigment have longer lifetime (nano seconds) than widelyemployed N3 or N719 Ru based dyes (femto seconds) [43–45].

Photosensitization of wide band gap semiconductors such asTiO2 by visible light absorbing dyes has become more practicalfor solar cell applications in the conversion of light into electricity[46]. Sensitization of colloidal TiO2 has been studied extensively inthe past [47–51]. Recently we have reported the sensitization ofcolloidal TiO2 nanoparticles using porphyrins [52]. Semiconductorparticles of colloidal dimensions are sufficiently small to yieldtransparent solutions, allowing direct analysis of electron transferby fluorescence quenching technique [53].

In the present work we have investigated the electron transferfrom excited phycocyanin to the conduction band of TiO2 byabsorption, steady state and time resolved fluorescence spectros-copy. This is the first challenge made to investigate the photosen-sitization of colloidal TiO2 nanoparticles with phycocyanin.

2. Experimental section

2.1. Materials

Titanium (IV) 2-propoxide was purchased from Aldrich. Phyco-cyanin (PC) is obtained as gift sample from Dr. S. Sekar (Bharath-idasan University, Trichy). The doubly distilled water was usedfor preparing the solutions. All measurements were performed atambient temperature.

2.2. Methods

2.2.1. Preparation of colloidal TiO2 nanoparticlesThe colloidal TiO2 nanoparticles was prepared by the hydrolysis

of titanium (IV) 2-propoxide based on the literature procedure[54]. Typically, titanium (IV) 2-propoxide (153 ll) in 2-propanol(10 ml) was injected by using syringe into 40 ml of water [pH 1.5(adjusted with HNO3)] with constant stirring under nitrogen atmo-sphere (8 h) gave 1 � 10�2 M titania stock solution. There is no sta-bilizing agent used for the preparation of colloidal TiO2

nanoparticles. The colloidal suspensions of TiO2 prepared by this

method were stable for 3–5 days. Fresh colloidal TiO2 dispersedin water was prepared before each set of experiments. The stocksuspension was diluted with water to obtain the desired concen-tration of TiO2. No attempts were made to exclude the traces of2-propanol present in the colloidal TiO2 nanoparticles and it wasconfirmed separately that the presence of 2-propanol did not affectthe photochemical measurements as earlier reported [55].

2.2.2. XRD measurementsX-ray powder diffraction patterns were recorded on a Bruker

AXS B8 Discover model using CuKa radiation (k = 0.154 nm) and agraphite monochromator in the diffracted beam. TiO2 samplewas in the form of powder. A scan rate of 0.05� min�1 was appliedto record a pattern in the 2h range of 2h = 20–80�.

2.2.3. Fluorescence quenching experimentsSamples were prepared by dissolving phycocyanin in water and

administering the appropriate amounts of colloidal TiO2. The sam-ples were deoxygenated by bubbling with pure nitrogen. Quartzcells (4 � 1 � 1 cm) with high vacuum Teflon stopcocks were usedfor bubbling.

2.2.3.1. Steady-state measurements. The steady-state fluorescencequenching measurements were carried out in a JASCO FP-6500spectrofluorimeter. Excitation and emission wavelengths of phyco-cyanin are 615 nm and 642 nm, respectively. The slit widths (each5 nm) and scan rate (500 nm/min) were maintained constant forall the experiments. Absorption spectral measurements wererecorded using Cary 300 UV–Visible spectrophotometer.

2.2.3.2. Time-resolved measurements. Fluorescence lifetime mea-surements were carried out in a picosecond time correlated singlephoton counting (TCSPC) spectrometer. The excitation source is thetunable Ti-sapphire laser (TSUNAMI, Spectra Physics, USA). Thefluorescence decay was analyzed by using the software providedby IBH (DAS-6).

2.2.4. FT-IR measurementsFT-IR spectra were obtained by using Perkin-Elmer Spectrum

RXI FT-IR spectrometer at room temperature in the range of4000–400 cm�1. The samples were placed in a liquid cell betweentwo windows (CaF2). Mirror velocity is 0.3 cm/s and number of co-added scans is 4 then total collection time is less than 2 min.

2.2.5. Cyclic voltammetric measurementsIn the present study, the oxidation potential of phycocyanin

was measured in water with potassium chloride (KCl, 0.1 M) assupporting electrolyte (spectra not shown). The experimental set-up consisted of a platinum working electrode, a glassy carboncounter electrode and a standard calomel electrode. Reversiblepeak potentials were measured at different scan rates (0.05 V/s).All samples were deaerated by bubbling with pure nitrogen gasfor ca. 5 min at room temperature.


3.1. Characterization of TiO2 nanoparticles

Fig. 1 shows the XRD pattern of TiO2 nanoparticles obtainedfrom colloidal TiO2 by rotary evaporation at 30 �C. The XRD peaksare found to be broad indicating fine size of the sample grains. TheXRD pattern exhibits prominent peaks at 2h values of 25.6�, 38.4�,48.4�, 54.8� and 63.2� which are similar to the reported [56] valuesfor anatase TiO2. The average size of the sample determined is inthe range of 4.67 nm from full-width and half-maximum (FWHM)of the most intense peak using Scherrer’s equation d = 0.9k/b cosh,

N N N N

H H H H

O O

HOOC COOHSHH3C

H3CH

H

H

Phycocyanin

Scheme 1. Structure of phycocyanin.

A. Kathiravan, R. Renganathan / Journal of Colloid and Interface Science 335 (2009) 196–202 197


where k is the wavelength of X-ray radiation (0.154 nm), b is theFWHM in radians of the XRD peak (3.29) and h is the angle ofdiffraction (25.6�).

The phase contents of titania are obtained from the followingformula [57],

WR ¼ 1=ð1þ 0:8IA=IRÞWA ¼ 1�WR

where WR and WA are the content of rutile and anatase titania,respectively, IA and IR represent the diffraction intensities of anatase(101) and rutile (110). The phase contents of anatase and rutile arefound to be 81% and 19%. Hence, the existence of rutile (in our case19%) in titania has not decreased the amount of adsorbed watermolecules and hydroxyl groups [58,59]. Anatase is usually consid-ered to be more active than rutile phase, and pure rutile phasehas limited photocatalytic activity [60]. TEM image of the colloidalTiO2 nanoparticles is shown in Fig. 2. The particle size of TiO2 nano-particles from TEM measurement is about 3–8 nm.

Fig. 3 shows the FT-IR spectrum of colloidal TiO2 nanoparticles.This spectrum shows broad band around 3400 cm�1 and anotherpeak at 1650 cm�1. It has been reported that adsorbed water hasbands around 3400 and 1630 cm�1 [61–63,58], while Ti–OH bond-ing has bands around 3563, 3172, and 1600 cm�1 [64]. Moreover,small crystallites could result in the broadness of the peaks [64].Therefore, it is believed that the two observed peaks at 3400 and1650 cm�1 correspond to the surface adsorbed water and hydroxylgroups.

3.2. Determination of particle size of colloidal TiO2 nanoparticles

The absorption wavelength (k) and the corresponding band gapenergy (Eg) of TiO2 material are well known to be k = 385 nm andEg = 3.2 eV, respectively, for anatase phase. The particle size ofthe prepared colloidal TiO2 nanoparticles has been determined

from the relationship between band gap shift (DEg) and radius(R) of quantum size particles using following Eq. (1) [65],

DEg ¼p2h2

2lR2 �1:8e2

eRþ Polarization terms ð1Þ

where h is Planck’s constant, R is the radius of the particle, e is thedielectric constant of the semiconductor, DEg is the band gap shift,the calculated bandgap shift for the colloidal TiO2 is 0.3 eV (extrap-olating the spectral curve, the band gap energy of the colloidal TiO2

was measured to be Eg = 3.5 eV corresponding to k = 350 nm usingE = hm, Fig. 7b), as compared to bulk anatase. ‘l’ is the reduced massof the exciton, i.e., the reduced effective mass of the electron andthe hole 1=l ¼ l=m�c þ l=m�h

� �in the semiconductor and ‘e’ is the

electron charge. The value of l = 1.63 me (me is the electron restmass) [66]. Since the optical dielectric constant of bulk titaniumdioxide is very large (e = 170), the coloumbic and polarization termsin the equation are neglected. The calculated DEg = 0.3 eV, this blueshift compared to bulk anatase TiO2 particles indicated a size of TiO2

crystallites smaller than 10 nm due to so called quantum size effect[67]. The calculated size of the prepared colloidal TiO2 is 1.1 nm.

The particle size of TiO2 is calculated from absorption spectrumas well as by the analysis of XRD data. The particle size of TiO2

Fig. 1. X-ray diffraction (XRD) spectrum of prepared TiO2 nanoparticles.

Fig. 2. TEM image of colloidal TiO2 nanoparticles.

0

20

40

60

80

100

1500 2000 2500 3000 3500 4000

Wavenumber (cm-1)

% T

Fig. 3. FT-IR spectrum of colloidal TiO2 nanoparticles.

198 A. Kathiravan, R. Renganathan / Journal of Colloid and Interface Science 335 (2009) 196–202


nanoparticles from XRD method (4.67 nm) is larger than the valueobtained from UV method. For XRD measurement we have donethe isolation of TiO2 particles from colloidal solution by rotaryevaporation at 30 �C.


Fig. 4 shows the absorption spectrum of phycocyanin in the ab-sence and presence of colloidal TiO2 nanoparticles at different con-centrations. In the presence of colloidal TiO2 the optical density ofphycocyanin at 615 nm was increased with the peak shift around9 nm (red shift). It may be due to adsorption of phycocyanin on thesurface of colloidal TiO2 nanoparticles through its anchoring group(–COOH) (Scheme 2). These changes are related to the formation of

surface complex between phycocyanin and colloidal TiO2 nanopar-ticles. Similar type of adsorption of sensitizers on the surface ofcolloidal TiO2 nanoparticles was reported earlier [68–70].

The equilibrium for the formation of complex between phyco-cyanin and colloidal TiO2 nanoparticles is given by Eq. (2), whereKapp represents the apparent association constant:

phycocyaninþ TiO2 �Kapp

phycocyanin . . . TiO2

Kapp ¼ ½phycocyanin...:TiO2 �½phycocyanin��½TiO2 �

ð2Þ

The changes in intensity of the absorption peak at 615 nm as a re-sult of formation of the surface complex were utilized to obtain Kapp

according to Benesi and Hildebrand equation [71]:

Aobs ¼ ð1� aÞC0ephycocyaninlþ aC0ecl ð3Þ

where Aobs is the observed absorbance of the solution containingdifferent concentrations of the colloidal TiO2 at 615 nm; a is the de-gree of association between phycocyanin and TiO2; ephycocyanin andec are the molar extinction coefficients at the defined wavelength(kmax = 615 nm) for phycocyanin and the formed complex, respec-tively, in water. Eq. (3) can be expressed as Eq. (4), where A0 andAc are the absorbances of phycocyanin and phycocyanin complex,respectively, with the concentration of C0:


At relatively high TiO2 concentrations, a can be equated to(Kapp[TiO2])/(1 + Kapp [TiO2]). In this case, Eq. (4) can be expressedas Eq. (5):

1Aobs � A0

¼ 1Ac � A0

þ 1KappðAc � A0Þ½TiO2�

ð5Þ

Therefore, if the enhancement of absorbance at the wavelength of615 nm was due to absorption of surface complex, we would expecta linear relationship between 1/(Aobs � A0) and the reciprocal con-centration of colloidal TiO2 with a slope equal to 1/Kapp(Ac � A0)

0

0.2

0.4

0.6

0.8

500 550 600 650 700 750 800 850 900

Wavelength (nm)

Abs

orba

nce

0

10

20

30

0 0.2 0.4 0.6 0.8 1

1/[TiO2] x 10-4 M

1/A

-A0

Fig. 4. Absorption spectrum of phycocyanin (1 � 10�6 M) in the absence (redcolored) and presence (green colored) of colloidal TiO2 nanoparticles in theconcentration range of 0, 1, 2, 3, 4 and 5 � 10�4 M. The inset is the straight linedependence of 1/(Aobs � A0) on the reciprocal concentration of colloidal TiO2. (Forinterpretation of the references to color in this figure legend, the reader is referredto the web version of this paper.)

N N N N

H H H H

O O

CSHH3CH3C

HH

H

Ti

OHHO

IV

HOOC

OH

O

+

N N N N

H H H H

O O

CSHH3CH3C

HH

H

Ti IV

HOOC

O

O-H2O

N N N N

H H H H

O O

CSHH3C

H3CH

H

H

Ti

OHHO

IV

OH

O

+

-2H2OC

HO

O

N N N N

H H H H

O O

CSHH3CH3C

HH

H

Ti IV

O

OC

O

O

Scheme 2. Possible condensation type interaction of phycocyanin to the hydroxyl group of TiO2 surface.



and an intercept equal to 1/(Ac � A0). The inset of Fig. 4 shows theBenesi and Hildebrand plot for phycocyanin, there is a good lineardependence of 1/(Aobs � A0) on the reciprocal concentration of col-loidal TiO2. The value of apparent association constant (Kapp) deter-mined from this plot is 4.01 � 102 M�1.

3.4. FT-IR characterization for phycocyanin–TiO2 system

Absorption measurement is not enough to investigate the ad-sorbed phycocyanin on the TiO2 surface. So Fourier transforminfrared (FT-IR) technique was used to gain further informationabout the nature of interaction between them. Fig. 5 shows theFT-IR spectra of phycocyanin (solid line) and phycocyanin boundcolloidal TiO2 (broken line). The spectrum of pure phycocyaninshows the C@O stretching vibration at 1629 cm�1. This band isshifted to 1634 cm�1 (around 5 cm�1) in phycocyanin bound col-loidal TiO2 spectrum. From this observation it is inferred that theinteraction of phycocyanin with TiO2 surface occurs through itscarboxyl anchoring group (COOH).

3.5. Fluorescence quenching characteristics

Fig. 6 shows the effect of increasing concentration of colloidalTiO2 nanoparticles on the emission spectrum of phycocyanin.Addition of TiO2 colloid to the solution of phycocyanin resultedin the quenching of its emission. This quenching behavior is similar

to the studies reported earlier [72]. The apparent association con-stant (Kapp) has also been calculated from the fluorescence quench-ing data according to the following equation,

1ðF0 � FÞ

¼ 1ðF0 � F 0Þ

þ 1KappðF0 � F 0Þ½TiO2�

ð6Þ

where Kapp is the apparent association constant, F0 is the initial fluo-rescence intensity of phycocyanin, F0 is the fluorescence intensity ofTiO2 adsorbed phycocyanin and F is the observed fluorescenceintensity at its maximum. The plot of 1/(F0 � F) versus 1/[TiO2] isshown in the inset of Fig. 6. A good linear relationship was obtainedbetween 1/(F0 � F) and the reciprocal concentration of colloidalTiO2. From the slope, the Kapp has been assessed and the value is5.20 � 102 M�1. The value of Kapp obtained from the data of fluores-cence quenching matches with that determined from the absorp-tion spectral changes. The good agreement between these valuesof Kapp highlighted the validity of assumption proposed for the asso-ciation between phycocyanin and colloidal TiO2 nanoparticles.

The decrease in fluorescence emission may be attributed toelectron transfer or energy transfer process. The band gap energyof TiO2 (3.2 eV) is greater than the excited state energy of phycoc-anin (1.94 eV) and there is no overlap between the fluorescenceemission spectrum of phycocyanin with the absorption spectrumof colloidal TiO2 (Fig. 7a), the above two inferences excluded thepossibility of energy transfer from phycocyanin to colloidal TiO2

nanoparticles. So, from the above discussion we confirmed thatthe fluorescence quenching shown in Fig. 6 could be caused byelectron transfer process.

The ability of the excited state phycocyanin to inject its elec-trons into the conduction band of TiO2 is determined by the en-ergy difference between the conduction band of TiO2 andoxidation potential of the excited state phycocyanin. Accordingto the equation Es�=sþ ¼ Es=sþ � Es, the oxidation potential of ex-cited state phycocyanin is �1.41 vs SCE, where, Es/s+ is the oxi-dation potential of phycocyanin 0.53 V vs SCE, Es is the excitedstate energy 1.94 eV (excited state energy of the phycocyanincalculated [73] from the fluorescence maximum based on the re-ported method). The energy level of the conduction band of TiO2

is �0.52 V vs SCE [74]. It suggests that the electron transfer fromthe excited state phycocyanin to the conduction band of TiO2 isenergetically favorable (Scheme 3).


As shown in previous reports [52,72,75,76], the dye moleculesadsorbed on the semiconductor particle surface had significantlyshorter fluorescence lifetime than the unadsorbed molecules, thisdecrease in lifetime can be correlated with the electron transferprocess. Similar to the above, the fluorescence decay of phycocya-nin in colloidal TiO2 system is shown in Fig. 8. In the absence of col-loidal TiO2, the fluorescence of phycocyanin has a singleexponential decay with lifetime of 1.8 ns. However addition of col-loidal TiO2 makes the fluorescence emission of phycocyanin fittedto a bi-exponential decay, with short-lived (0.9 ns) and longer-lived(1.7 ns) components. The longer lifetime is for phycocyanin alone,and the shorter one is for phycocyanin adsorbed on TiO2. The twodifferent lifetimes could be attributed to the molecules of phycocy-anin that are unadsorbed (s) and adsorbed (sads) on the surface ofTiO2 (Table 1). If it is assumed that the observed decrease in fluores-cence lifetime is entirely due to the electron transfer process, therate constant for electron transfer (ket) from excited state sensitizerinto semiconductor can be calculated by using Eq. (7) [75]

ket ¼ 1=sads � 1=s ð7Þ

The calculated value of ket is found to be 5.2 � 108 s�1.

35

55

75

95

800 1100 1400 1700 2000 2300 2600 2900

Wavenumber (cm-1)

% T

Phycocyanin-TiO2

Phycocyanin

Fig. 5. The FTIR spectra of phycocyanin in the absence (blue colored) and presence(red colored) of TiO2. (For interpretation of the references to color in this figurelegend, the reader is referred to the web version of this paper.)

0

114

228

342

456

570

620 640 660 680 700 720 740

Wavelength (nm)

Inte

nsit

y

00.0050.01

0.0150.02

0.025

0 0.2 0.4 0.6 0.8 1

1/[TiO2] x 10-4 M

1/F0 -F

Fig. 6. Fluorescence quenching of phycocyanin (1 � 10�6 M) in the absence (redcolored) and presence (blue colored) of colloidal TiO2 nanoparticles in theconcentration range of 0, 1, 2, 3, 4 and 5 � 10�4 M. The inset is the dependence of1/(F0 � F) on the reciprocal concentration of colloidal TiO2, kexi: 615 nm and kemi:642 nm. (For interpretation of the references to color in this figure legend, thereader is referred to the web version of this paper.)



3.7. Calculation of free energy change (DGet) for the electron transferreactions

The thermodynamic feasibility of the excited state electrontransfer reaction was calculated by employing the well knownRehm–Weller expression [77]

DGet ¼ EðoxÞ1=2 � EðredÞ

1=2 � Es þ C ð8Þ

where, EðoxÞ1=2 is the oxidation potential of phycocyanin (0.53 V), EðredÞ

1=2

is the reduction potential of TiO2 (i.e.) conduction band potential ofTiO2 is �0.52 V, Es is the excited state energy of phycocyanin(1.94 eV) and C is the coulombic term. Since one of the species is

neutral and the solvent used is polar in nature, the coulombic termin the above expression is neglected [78]. The DGet value was calcu-lated as �0.89 eV and this higher negative value indicates the elec-tron transfer process is thermodynamically favorable [79,80].

4. Conclusion

Phycocyanin adsorbed on the surface of colloidal TiO2 nanopar-ticles through its carboxyl group, as evidenced by the effect of col-loidal TiO2 on the absorption and FT-IR spectroscopy. The apparentassociation constant (Kapp) has been determined from absorptionchanges and fluorescence quenching data. Based on the above re-sults, electron injection from excited state phycocyanin into theconduction band of TiO2 is suggested. Future work will be directedto optimizing the performance of phycocyanin sensitized solarcells by exploring variations in the electrolyte composition andfilm properties, as well as the use of purified phycocyanin extractsand other class of phycobiliproteins from various cyanobacteriaand algae systems.

Acknowledgments

R.R. and A.K. thank CSIR, Government of India (Ref. No.01(2217)/08/EMR-II, dt. 06/05/2008) for the project and fellow-ship, respectively.

We are thankful to Prof. P. Ramamoorthy, NCUFP, University ofMadras, Chennai, for lifetime measurements. Authors also thankDr. S. Anandan, (NIT, Trichy) for providing CV facilities.

References

[1] B. O’Regan, M. Gratzel, Nature 353 (1991) 737.[2] M. Gratzel, Nature 414 (2001) 338.[3] A. Hagfeldt, M. Gratzel, Acc. Chem. Res. 33 (2000) 269.[4] A. Hagfeldt, M. Gratzel, Chem. Rev. 95 (1995) 49.[5] M.K. Nazeeruddin, A. Kay, I. Rodicio, R. Humphry-Baker, E. Muller, P. Liska, N.

Vlachopoulos, M. Gratzel, J. Am. Chem. Soc. 115 (1993) 6382.

0.05

0.24

0.43

0.62

0.81

1

300 400 500 600 700 800

Wavelength (nm)

Abs

orba

nce

0

100

200

300

400

500

600

Inte

nsit

y

TiO2 absorption

Phycocyanin emission

(b)

(a)

0

0.2

0.4

0.6

0.8

1

300 350 400 450 500 550 600 650 700 750 800

Wavelength (nm)

Ab

sorb

an

ce

Fig. 7. (a) Spectra of TiO2 absorption and phycocyanin emission. (b) UV–visabsorption spectra of colloidal TiO2.

1

2

3

0

-1

-2

TiO2

CB

VB

(0.53 V)

(-0.52 V)

V vs SCE

(2.9 V)

Es/s+

Es*/s+(-1.41 V)

νh

Phycocyanin

Scheme 3. Schematic energy level diagram showing electron transfer process.

1

10

100

1000

10000

0 5 10 15 20

Time (ns)

Cou

nts

Fig. 8. Fluorescence decay profiles for phycocyanin (1 � 10�6 M) in the absence(pink colored) and presence (black colored) of 5 � 10�4 M colloidal TiO2, kexi:615 nm and kemi: 642 nm. (For interpretation of the references to color in this figurelegend, the reader is referred to the web version of this paper.)

Table 1The data of the fluorescence lifetime measurements.

Samples s (ns) A1 (%) sads (ns) A2 (%) v2 ket � 108 s�1

PC 1.8 100 – – 1.2 –PC + TiO2 1.7 26.8 0.9 73.2 1.1 5.2

[PC] = 1 � 10�6 M; [TiO2] = 5 � 10�4 M.A1, A2 are the percentages of components with the defined fluorescence lifetime.The excitation wavelength is 615 nm and the emission wavelength is 642 nm.ket is the rate of electron transfer.s and sads are the unadsorbed and adsorbed fluorescence lifetimes.



[6] R. Argazzi, N.Y.M. Iha, H. Zabri, F. Odobel, C.A. Bignozzi, Coord. Chem. Rev. 248(2004) 1299.

[7] A.S. Polo, M.K. Itokazu, N.Y.M. Iha, Coord. Chem. Rev. 248 (2004) 1343.[8] Y. Tachibana, J.E. Moser, M. Gratzel, D.R. Klug, J.R. Durrant, J. Phys. Chem. 100

(1996) 20056.[9] B. O’Regan, D.T. Schwartz, J. Appl. Phys. 80 (1996) 4749.

[10] S. Ferrere, A. Zaban, B.A. Gregg, J. Phys. Chem. 101 (1997) 4490.[11] T.N. Rao, L. Bahadur, J. Electrochem. Soc. 144 (1997) 179.[12] A.C. Khazraji, S. Hotchandani, S. Das, P.V. Kamat, J. Phys. Chem. B 103 (1999)

4693.[13] A. Ehret, L. Stuhl, M.T. Spitler, J. Phys. Chem. B 105 (2001) 9960.[14] F.G. Gao, A.J. Bard, L.D. Kispert, J. Photochem. Photobiol. A: Chem. 130 (2000)

49.[15] H. Tian, P.H. Liu, W. Zhu, E. Gao, D.J. Wu, S. Cai, J. Mater. Chem. 10 (2000) 2708.[16] T. Yoshida, K. Terada, D. Schlettwein, T. Oekermann, T. Sugiura, H. Minoura,

Adv. Mater. 12 (2000) 1214.[17] K. Hara, T. Horiguchi, T. Kinoshita, K. Sayama, H. Sugihara, H. Arakawa, Sol.

Energy Mater. Sol. Cells 64 (2000) 115.[18] Z.S. Wang, F.Y. Li, C.H. Huang, Chem. Commun. (2000) 2063.[19] Q.H. Yao, L. Shan, F.Y. Li, D.D. Yin, C.H. Huang, New J. Chem. 27 (2003) 1277.[20] K. Sayama, S. Tsukagoshi, K. Hara, Y. Ohga, A. Shinpou, Y. Abe, S. Suga, H.

Arakawa, J. Phys. Chem. B 106 (2002) 1363.[21] K. Sayama, S. Tsukagoshi, T. Mori, K. Hara, Y. Ohga, A. Shinpou, Y. Abe, S. Suga,

H. Arakawa, Sol. Energy Mater. Sol. Cells 80 (2003) 47.[22] T. Horiuchi, H. Miura, S. Uchida, Chem. Commun. (2003) 3036.[23] L.S. Mende, U. Bach, R.H. Baker, T. Horiuchi, H. Miura, S. Ito, S. Uchida, M.

Gratzel, Adv. Mater. 17 (2005) 813.[24] K. Hara, M. Kurashige, S. Ito, A. Shinpo, S. Suga, K. Sayama, H. Arakawa, Chem.

Commun. (2003) 252.[25] T. Kitamura, M. Ikeda, K. Shigaki, T. Inoue, N.A. Anderson, X. Ai, T. Lian, S.

Yanagida, Chem. Mater. 16 (2004) 1806.[26] R. Mosurkal, J.A. He, K. Yang, L.A. Samuelson, J. Kumar, J. Photochem. Photobiol.

A: Chem. 168 (2004) 191.[27] A. Kay, M. Gratzel, J. Phys. Chem. 97 (1993) 6272.[28] S. Cherian, C.C. Wamser, J. Phys. Chem. B 104 (2000) 3624.[29] T. Ma, K. Inoue, K. Yao, H. Noma, T. Shuji, E. Abe, J. Yu, X. Wang, B. Zhang, J.

Electroanal. Chem. 537 (2002) 31.[30] F. Odobel, E. Blart, M. Lagree, M. Villieras, H. Boujtita, N.E. Murr, S. Caramori,

C.A. Bignozzi, J. Mater. Chem. 13 (2003) 502.[31] K. Nazeeruddin, R.H. Baker, D.L. Officer, W.M. Campbell, A.K. Burrell, M.

Gratzel, Langmuir 20 (2004) 6514.[32] W.M. Campbell, A.K. Burrell, D.L. Officer, K.W. Jolley, Coord. Chem. Rev. 248

(2004) 1363.[33] N. Cherepy, G.P. Smestad, M. Gratzel, J.Z. Zhang, J. Phys. Chem. B 101 (1997)

9342.[34] K. Tennakone, A.R. Kumarasinghe, G.R.R.A. Kumara, K.G.U. Wijayantha, P.M.

Sirimanne, J. Photochem. Photobiol. A: Chem. 108 (1997) 193.[35] Q. Dai, J. Rabani, New J. Chem. 26 (2002) 421.[36] A.N. Glazer, J. Appl. Phycol. 6 (1994) 105.[37] D.A. Bryant, G. Guglielmi, N.T. Marsac, A.M. Castets, Arch. Microbiol. 123

(1979) 113.[38] A.N. Glazer, Annu. Rev. Biochem. Phys. Chem. 14 (1985) 47.[39] A.N. Glazer, J. Cell Biol. 264 (1989) 1.[40] P. Spolaore, C.J. Cassan, E. Duran, J. Biosci. Bioeng. 101 (2006) 87.[41] V.B. Bhat, K.M. Madyastha, Biochem. Biophys. Res. Commun. 275 (2000) 20.

[42] C.H. Romay, R. Gonzalez, N. Ledon, Curr. Protein Peptide Sci. 4 (2003) 207.[43] K. Hara, Z.S. Wang, T. Sato, A. Furube, R. Katoh, H. Sugihara, Y. Dan-oh, C.

Kasada, A. Shinpo, S. Suga, J. Phys. Chem. B 109 (2005) 15476.[44] G. Benko, J. Kallioinen, J.E.I.K. Tommola, A.P. Yartsev, V. Sundstrom, J. Am.

Chem. Soc. 124 (2002) 489.[45] R. Katoh, A. Furube, M. Kasuya, N. Fuke, N. Koide, L. Han, J. Mater. Chem. 17

(2007) 3190.[46] A. Kay, M. Gratzel, J. Phys. Chem. 97 (1993) 6272.[47] H. Mao, H. Deng, H. Li, Y. Shen, Z. Lu, H. Xu, J. Photochem. Photobiol. A: Chem.

114 (1998) 209.[48] Z.X. Zhou, S.P. Qian, S.D. Yao, Z.Y. Zhang, Dyes Pigments 51 (2001) 137.[49] Z. Zhou, S. Qian, S. Yao, Z. Zhang, Radiat. Phys. Chem. 65 (2002) 241.[50] G. Ramakrishna, S. Verma, D.A. Jose, D.K. Kumar, A. Das, D.K. Palit, H.N. Ghosh,

J. Phys. Chem. B 110 (2006) 9012.[51] M.K.I. Senevirathne, P.K.D.D.P. Pitigala, V. Sivakumar, P.V.V. Jayaweera, A.G.U.

Perera, K. Tennakone, J. Photochem. Photobiol. A: Chem. 195 (2008) 364.[52] A. Kathiravan, R. Renganathan, J. Colloid Interface Sci. 331 (2009) 401.[53] C. Chen, X. Qi, B. Zhou, J. Photochem. Photobiol. A: Chem. 109 (1997) 155.[54] D. Bahnemann, A. Henglein, J. Lilie, L. Spanhel, J. Phys. Chem. 88 (1984) 709.[55] P.V. Kamat, J.P. Chauvet, R.W. Fessenden, J. Phys. Chem. 90 (1986) 1389.[56] J.C. Tristao, F. Magalhaes, P. Corio, M.T.C. Sansiviero, J. Photochem. Photobiol.

A: Chem. 181 (2006) 152.[57] R.A. Spurr, W. Myers, Anal. Chem. 29 (1957) 760.[58] Z. Ding, G.Q. Lu, P.F. Greenfield, J. Phys. Chem. B 104 (2000) 4815.[59] G. Liu, Z. Chen, C. Dong, Y. Zhao, F. Li, G.Q. Lu, H.M. Cheng, J. Phys. Chem. B 110

(2006) 20823.[60] S. Tsai, S. Cheng, Catal. Today 33 (1997) 227.[61] T. Nakayama, J. Electrochem. Soc. 141 (1994) 237.[62] Y. Suda, T. Morimoto, Langmuir 3 (1987) 786.[63] K. Tanaka, J.M. White, J. Phys. Chem. 86 (1982) 4708.[64] E. Sanchez, T. Lopez, R. Gomez, Bokhimi, A. Morales, O. Novaro, J. Solid State

Chem. 122 (1996) 309.[65] C. Kormann, D.W. Bahnemann, M.R. Hoffmann, J. Phys. Chem. 92 (1988) 5196.[66] E. Joselevich, I. Willner, J. Phys. Chem. 98 (1994) 7628.[67] A. Henglein, Ber. Bunsen-Ges. Phys. Chem. Chem. Phys. 86 (1982) 241.[68] A. Kathiravan, R. Renganathan, Spectrochim. Acta A 71 (2008) 1080.[69] J. He, F. Chen, J. Zhao, H. Hidaka, Colloid Surf. A: Physicochem. Eng. Aspects 142

(1998) 49.[70] G. Ramakrishna, S. Verma, D.A. Jose, D.K. Kumar, A. Das, D.K. Palit, H.N. Ghosh,

J. Phys. Chem. B 110 (2006) 9012.[71] H.A. Benesi, J.H. Hildebrand, J. Am. Chem. Soc. 71 (1949) 2703.[72] A. Kathiravan, P. Sathish Kumar, R. Renganathan, S. Anandan, Colloids Surf. A:

Physicochem. Eng. Aspects 333 (2009) 175.[73] E.J. Shin, D. Kim, J. Photochem. Photobiol. A: Chem. 152 (2002) 25.[74] G. Ramakrishna, H.N. Ghosh, A.K. Singh, D.K. Palit, J.P. Mittal, J. Phys. Chem. B

105 (2001) 12786.[75] P.V. Kamat, J. Phys. Chem. 93 (1989) 859.[76] P.V. Kamat, M.A. Fox, Chem. Phys. Lett. 102 (1983) 379.[77] G.J. Kavarnos, N.J. Turro, Chem. Rev. 86 (1986) 401.[78] S. Parret, F.M. Savary, J.P. Fouassier, P. Ramamurthy, J. Photochem. Photobiol.

A: Chem. 83 (1994) 205.[79] K. Kikuchi, T. Niwa, Y. Takahashi, H. Ikeda, T. Miyashi, J. Phys. Chem. 97 (1993)

5070.[80] S. Nath, H. Pal, D.K. Palit, A.V. Sapre, J.P. Mittal, J. Phys. Chem. A 102 (1998)

5822.











Colloids and Surfaces B: Biointerfaces 72 (2009) 167–172


Colloids and Surfaces B: Biointerfaces

journa l homepage: www.e lsev ier .com/ locate /co lsur fb

Spectroscopic studies on the interaction of colloidal capped CdS nanoparticleswith bovine serum albumin

M. Asha Jhonsi, A. Kathiravan, R. Renganathan ∗



Article history:Received 13 December 2008Received in revised form 27 March 2009Accepted 29 March 2009Available online 5 April 2009

Keywords:BSACapped CdSFluorescence quenching

a b s t r a c t

Colloidal uncapped and starch capped CdS (SCdS) nanoparticles were prepared and interaction withbovine serum albumin (BSA) have been studied by UV–visible, FT-IR, steady state, time resolved and syn-chronous fluorescence spectroscopic measurements. BSA molecules adsorbed on the surface of colloidalCdS through the capping agent. The apparent association constant (Kapp = 2.54 × 102 M−1) and degreeof association has been calculated (˛ = 1.12) from absorption studies. The binding constant from fluores-cence quenching method (6.6 × 102 M−1) matches well with that determined from the absorption spectralchanges. Static quenching mechanism and conformational changes on BSA molecules were confirmed bytime resolved and synchronous fluorescence measurements respectively. The effect of starch capped CdSon the conformation of BSA has been analyzed by means of UV–visible absorption and synchronousfluorescence spectra.


1. Introduction

Nanoparticle probes compared with organic dyes acting asbiosensors in chemical and biochemical fields have been researchedrecently and their applications are becoming more extensive. Theseprobes have been applied to the ultrasensitive detection of proteins,DNA sequencing, clinical diagnostics, etc. [1]. Compared to conven-tional dyes, confinement of electronic states of quantum dots makesthem quite attractive, showing some unique optical properties suchas high quantum yield, symmetrical emission spectra, broad-bandexcitation, photostability, and readily tunable spectra [2–6].

The interaction of proteins with inorganic surfaces involvingmainly silica particles and to a lesser extent hematite [7], hydroxi-apatite [8] and titanium dioxide [9,10] were reported. Renganathanand co-workers have successfully studied the binding of TiO2 col-loid with serum albumin [11,12]. Major interactions involved inprotein adsorption can be classified as electrostatic, hydrophobicand hydrogen-bonding, etc. [13]. One of the goals in the study ofprotein adsorption is the follow-up of the process dynamics throughin situ techniques that allow the analysis of the possible conforma-tional changes that may take place during adsorption [14,15]. Thisstudy has been studied actively for decades because of its impor-tance in the wide range of biomedical applications, such as artificialtissue and organ [16], drug delivery system [17], biosensor [18],solid-phase immunoassay [19], immunomagnetic cell separation[20] and immobilized enzyme or catalyst [21] and so on.


Quenching measurement of albumin fluorescence is an impor-tant method to investigate the interactions of drugs with serumalbumins. It can reveal the accessibility of quenchers to albumin’sfluorophore groups, help to understand the binding mechanisms ofalbumins with drugs and provide clues to the essential of bindingphenomenon [22].

In this work, bovine serum albumin (BSA) is selected as our pro-tein model because of its medicinal importance, low cost, readyavailability, and unusual ligand-binding properties [23]. BSA hastwo tryptophan residues that possess intrinsic fluorescence, Trp-212 is located within a hydrophobic binding pocket of the proteinand Trp-134 is located on the surface of the molecule [24–26].

The two ions Cd2+ and S2− are component parts of quantumdots which are harmful to human body. By researching the inter-action of CdS nanoparticles with BSA, the effect of CdS to proteinin real cells can be simulated. The interaction of the CdS nanocrys-tals with its environment or capping agents plays a crucial role indetermining its luminescent properties. Organic and inorganic cap-ping agents such as polymers, amines, tri-n-octyl phosphine oxide(TOPO), thiols and silica are used during the wet-chemical syn-thesis for capping the surface of particles to prevent nonradiativerecombination at surface sites and also control of growth kineticsto prevent the aggregation via steric hindrance [27,28]. It is alsobelieved that the capping agent on the surface of particles plays animportant role on transfer of photogenerated electrons and holesto capping agents [29].

Among the direct techniques, fluorimetry is extensively usedand is considered to be superior to the indirect techniques (equilib-rium and dynamic dialysis, ultrafiltration, gel filtration) because, toa first approximation, they do not disturb the binding equilibrium

0927-7765/$ – see front matter © 2009 Elsevier B.V. All rights reserved.doi:10.1016/j.colsurfb.2009.03.030


168 M. Asha Jhonsi et al. / Colloids and Surfaces B: Biointerfaces 72 (2009) 167–172

upon separation [30]. In the present work, synthesis of uncappedand starch capped CdS nanoparticles and the effect of these pre-pared capped CdS on intrinsic tryptophan fluorescence quenchingof BSA has been studied and to characterize the type of chemicalassociation taking place.

2. Experimental

2.1. Materials

Cadmium chloride (CdCl2), cadmium acetate dihydrate(Cd(OAc)2·2H2O), thioacetamide and sodiumhexametaphos-phate were purchased from Qualigens and they were used assuch. Starch, glacial acetic acid, 2-methoxy ethanol and acetonewere purchased from Loba Chemicals. Bovine serum albumin (96%fatty acid free, Product Number A8806, Sigma) was dissolved indouble distilled water to prepare stock solution (1 × 10−4 M) whichwas then stored at 0–4 ◦C. All measurements were performed atambient temperature.

2.2. Instrumentation

X-ray powder diffraction patterns were recorded on a BrukerAXS B8 Discover model using Cu K� radiation (� = 0.154 nm) and agraphite monochromator in the diffracted beam. CdS sample wasin the form of powder. A scan rate of 0.05◦ min−1 was applied torecord a pattern in the 2� range of 2� = 20–80◦.

Absorption spectra were recorded using Cary 300 UV–visiblespectrophotometer. The samples were carefully purged using purenitrogen gas for 10 min. Quartz cells (4 cm × 1 cm × 1 cm) with highvacuum Teflon stopcocks were used for purging.

The fluorescence quenching measurements were carried outwith JASCO FP-6500 spectrofluorometer. The excitation and emis-sion slit width (each 5 nm) and scan rate (500 nm min−1) weremaintained constant for all the measurements.

Fluorescence lifetime measurements were carried out in apicosecond time correlated single photon counting (TCSPC) spec-trometer. The excitation source is the tunable Ti-sapphire laser(Tsunami, Spectra Physics, USA). The fluorescence decay was ana-lyzed by using the software provided by IBH (DAS-6).

FT-IR spectra were obtained by using PerkinElmer SpectrumRXI FT–IR spectrometer at room temperature in the range of4000 − 400 cm−1. The samples were placed in a liquid cell betweentwo windows (CaF2). Mirror velocity is 0.3 cm s−1 and number ofco-added scans are 4 then total collection time is less than 2 min.

2.3. Preparation procedures

The preparation of uncapped and starch capped CdS followed bythe modified methods were adopted from literature [29,31].

2.3.1. Preparation of colloidal CdS nanoparticles10 ml of 0.1 M cadmium chloride, 10 ml of 0.1 M thioacetamide

and 10 ml of 0.1 M sodium hexametaphosphate were mixed withconstant stirring. The pH of the mixture was adjusted to 10.4 by0.1 M NaOH, under N2 atmosphere. The mixture was kept at roomtemperature for 35 min for the growth of colloidal CdS. The result-ing yellow colloid was diluted to a required concentration and wasstored at 4 ◦C.

2.3.2. Preparation of starch capped CdS nanoparticles0.1 gm of starch was dissolved in 5 ml of hot water to get a

clear solution. Then 0.1 mM (0.0266 g) Cd(OAc)2·2H2O dissolving in20 ml of glacial acetic acid was added into the starch solution andstirred for 5 min at room temperature. After this 0.1 mM (0.0075 g)thioacetamide in 20 ml 2-methoxy ethanol was added to the above

mixture and stirred in an oil bath at 85–90 ◦C for 1 h. The resultinglemon yellow colored solid products were centrifuged, washed andfinally dispersed in water for optical study. In starch solution, thehydroxyl groups acted as stabilizer agent of the synthesized CdSnanoparticles.

2.4. Methods

A 3-ml solution, containing appropriate concentration of BSA(1 × 10−5 M), was titrated by successive additions of 3 �L stock solu-tion of colloidal uncapped and starch capped CdS nanoparticlesseparately (1 × 10−3 M). Titrations were manually done by usingmicro-pipette for the addition of colloidal CdS and starch cappedCdS. UV–visible spectra of all the solutions were recorded in therange of 200–800 nm.

Fluorescence spectra were then measured by using Quartz cells(4 cm × 1 cm × 1 cm, excitation and emission wavelength of BSA is278 and 345 nm, respectively) at ambient temperature.

In synchronous fluorescence spectra also the same concentra-tion of BSA and colloidal CdS and starch capped CdS were used andthe spectra were measured at two different �� values such as 15and 60 nm.


3.1. Determination of particle size of colloidal CdS nanoparticles

The prepared colloidal CdS were analyzed by UV–visible andfluorescence spectroscopy to identify the presence of nanoparticles(Fig. 8). The absorption spectra showed the formation of nanosizedCdS having absorption in the range of 425–450 nm in water in amuch diluted solution (typically <1 × 10−4 M). (The fine emissionspectrum of CdS also shown in Fig. 8 indicates the uniform size ofthe CdS.)

It is well understood with available literature that milder exper-imental conditions favour the formation of smaller particle size andthe stability of these nanoparticles is around 7–8 days stored at 4 ◦C.

The diameter of the prepared colloidal CdS has been determinedfrom the relationship between band gap shift (�Eg) and radius (R)of quantum size particles using Eq. (1)

�Eg = �2h2

2R2[1/m∗

e + 1/m∗h] − 1.8 e2

εR+ polarisation terms (1)

where h is Planck’s constant, R is the radius of the particle, m∗e and

m∗h are the effective masses of the e− and h+ respectively in the

semiconductor, e is the electron charge, ε is the relative permittivityof the semiconductor.

A value of 0.153 me was used for reduced effective mass of theexciton (1/� = 1/me + 1/mh) of CdS, the columbic and polarizationterms in this equation are neglected. The particle size of the pre-pared colloidal CdS is 3.35 nm.

3.2. XRD characterization of CdS nanoparticles

For XRD measurement the resulting lemon yellow colored solidproducts were centrifuged, washed and dried in a vacuum ovenat 40 ◦C. Fig. 1 shows the XRD pattern of CdS and starch cappedCdS nanoparticles. The patterns are considerably broadened dueto very small size of the CdS. The XRD pattern exhibits prominentbroad peaks at 2� values of 27◦, 44◦ and 52◦ which are identified forcubic CdS phase, the 2� values are similar to the reported [32]. Theaverage size of the sample determined is in the range of 6.02 nmfrom full-width and half-maximum (FWHM) of the most intensepeak making by using Scherrer’s Eq. (2),

d = 0.9�/ˇ cos � (2)


M. Asha Jhonsi et al. / Colloids and Surfaces B: Biointerfaces 72 (2009) 167–172 169

Fig. 1. X-Ray diffraction (XRD) spectrum of uncapped and starch capped CdSnanoparticles.

where � is the wavelength of X-ray radiation ˇ is the FWHM inradians of the XRD peak (2.585) and � is the angle of diffraction(27◦). The particle size calculated from XRD method (6.02 nm) islarger than the value obtained from UV method (∼3.35 nm). Thismay be due to aggregation of colloidal CdS upon aging.


To study the excited state reactions between BSA and CdS it isimportant to know the type of interaction between them in theground state. The absorption spectra of BSA in the absence and pres-ence of starch capped CdS is shown in Fig. 2. Absorption of BSA ischaracterized by a strong band in the UV region at 278 nm. Addi-tion of starch capped CdS led to gradual increase in BSA absorptionwith a blue shift (shorter wavelength) of 2 nm. These observationsindicate that there is a structural change (microenvironment) inBSA which has occurred upon interaction with the surface of starchcapped CdS. The above results can be rationalized in terms of stronginteraction between starch capped CdS and BSA in the ground statethrough complex formation. Similar type of interaction betweenBSA with capped CdTe quantum dots and TiO2 has been previouslyreported [6,11].

In the present system the possible scattering due to colloidalnature of CdS (CdS also absorbs in the UV region but its range ofabsorbance was far away (more than 400 nm) from the studied BSAtryptophan absorption as shown in Fig. 8) has been omitted becauseof the fact that baseline correction was done for all UV spectralmeasurements. So we confirmed that the increase in optical densityis only resulting from the interaction between BSA and capped CdSand not due to scattering.

The equilibrium for the formation of complex between BSA andstarch capped CdS is defined by Eq. (3) where Kapp is the apparent

Fig. 2. Absorption spectrum of BSA (1 × 10−6 M) in the absence and presence ofstarch capped CdS (0–6 × 10−4 M). The insert is the straight line dependence of1/Aobs − A0 on the reciprocal concentration of capped CdS.

Scheme 1. Interaction between BSA with starch capped and uncapped colloidal CdSnanoparticles.

association constant,

BSA + capped CdSKapp� BSA . . . capped CdS (3)

Kapp = [BSA . . . capped CdS][BSA] · [capped CdS]

(4)

The Kapp value was calculated by the method reported by Benesiand Hildebrand [33] using the following Eq. (5):

Aobs = (1 − ˛)C0εBSA1 + ˛C0εc1 (5)

where Aobs is the absorbance of the BSA solution containing differ-ent concentrations of starch capped CdS at 278 nm, ˛ is the degreeof association between BSA and capped CdS, εBSA and εc are themolar extinction coefficients at the defined wavelengths for BSAand the formed complex, respectively, C0 is the initial concentra-tion of BSA and ‘l’ is the optical path length, which has been takenas unity. Eq. (5) can be expressed by Eq. (6), where the A0 and Ac

are the absorbance of BSA and the complex at 278 nm, with theconcentration of C0:

Aobs = (1 − ˛)A0 + ˛Ac (6)

At relatively high concentrations of capped CdS, ˛ can be equated to(Kapp [capped CdS])/(1 + Kapp[capped CdS]). In this case, Eq. (6) canbe expressed as Eq. (7):

1Aobs − A0

= 1Ac − A0

+ 1Kapp(Ac − A0)[capped CdS]

(7)

Therefore, if the enhancement of absorbance at 278 nm was dueto absorption of complex, one would expect a linear relationshipbetween 1/(Aobs − A0) and the reciprocal concentration of cappedCdS with a slope equal to 1/Kapp(Ac − A0) and an intercept equalto 1/(Ac − A0) (shown in the inset of Fig. 2). The calculated valueof Kapp from the straight line of such plot is about 2.54 × 102 M−1

(R2 = 0.9998). Based on Kapp, the degree of association (˛) has alsobeen calculated and it was found to be 1.12.

In order to determine which group of the CdS is responsiblefor interaction with BSA molecules, we have done the comparativeexperiments with uncapped colloidal CdS. In uncapped CdS–BSAsystem while increasing the amount of CdS there is only a slightchange in optical density without any shift in wavelength in theabsorption spectrum of BSA (not shown here), indicating that theinteraction between uncapped CdS and BSA is very weak, simplyBSA cannot be adsorbed on the surface of uncapped CdS. There-fore we conclude that the BSA molecule strongly interacts withthe capping agent on the surface of the colloidal CdS nanoparticlesand is involved in the ground state complex formation as shown inScheme 1.

3.4. Fluorescence quenching studies

3.4.1. Steady-state measurementThe interaction of BSA with starch capped CdS was studied

by spectrofluorometer at room temperature. An aqueous solutionof BSA (1 × 10−6 M) was titrated with increasing concentration of



Fig. 3. Steady-state fluorescence quenching of BSA (1 × 10−6 M, �exi: 278 nm and�emi: 345 nm) by starch capped CdS in the concentration range of 0–5 × 10−4 M inwater. The insert is the Stern–Volmer plot between I0/I versus [Q].

(0–5 × 10−4 M) starch capped CdS solution as shown in Fig. 3. TheSCdS also has luminescent property but its excitation (415 nm) andemission (522 nm) wavelengths are much far away from the BSAabsorption and emission as mentioned above (Fig. 8), so we did notquantify the emission nature of CdS nanoparticles. While increas-ing the concentration of starch capped CdS the emission intensityof BSA was found to decrease progressively with the blue shift ofaround 6 nm. The quenching of BSA fluorescence by capped CdS canbe described by Stern–Volmer Eq. (8).

I0/I = 1 + KSV[Q] (8)

where I0 and I are the fluorescence intensities of BSA in the absenceand presence of capped CdS, respectively. KSV is Stern–Volmer con-stant and [Q] is the concentration of respective quencher, cappedCdS. The ratios I0/I were calculated and plotted against quencherconcentration according to Eq. (8) (shown in the inset of Fig. 3).The quenching constant is calculated from the slope of the plotis in the order of 0.1955 × 104 M−1. The Stern–Volmer constant(KSV) is related to quenching rate constant by kq = KSV/�. From thequenching constant we have calculated the value of quenching rateconstant by using the lifetime � of BSA (6 ns, from time resolvedmeasurement) is 3.25 × 1011 M−1 s−1. In general, maximum colli-sional quenching constant (kq) of various kinds of quenchers tobiopolymers is 2.0 × 1010 M−1 s−1 [34]. But for BSA–SCdS systemhigher quenching rate constant (3.25 × 1011 M−1 s−1) was obtained.This proves that the quenching is static in nature, it depends onthe formation of complex between starch capped CdS and BSA[Scheme 2]. Further the type of interaction between BSA and cappedCdS was also confirmed by time resolved spectroscopy.

3.4.2. Time resolved measurementFluorescence lifetime measurement is useful for under-

standing the type of interaction between the colloidalsemiconductor–sensitizer systems. In general, the measure-ment of fluorescence lifetime is the most definitive method todistinguish static and dynamic quenching [35].

In the present work we have studied the effect of uncapped andstarch capped CdS on the fluorescence lifetime of BSA. Fig. 4(a)shows the fluorescence decay of BSA in the absence and presenceof starch capped CdS. Initially BSA in the absence of capped CdSshowed that monoexponential decay with the life time of 6 ns,and then first addition of capped CdS changed the decay curveform mono to biexponential with two life times such as 5.27 and

Scheme 2. Mechanism of complex formation.

Fig. 4. (a) Fluorescence decay of BSA (1 × 10−6 M) in the absence and presence ofstarch capped CdS in the concentration range of 0–5 × 10−4 M. (b) Fluorescencedecay of BSA (1 × 10−6 M) in the absence and presence of uncapped CdS in theconcentration range of 0–5 × 10−4 M.

2.02 ns, respectively. But further increasing the concentration ofstarch capped CdS the fluorescence lifetime of BSA remains unal-tered. So there is no change in the fluorescence lifetime of BSA in thepresence of highest concentration of capped CdS which indicatedthat the quenching follows static mechanism.

Fig. 4(b) shows the fluorescence decay of BSA in the absenceand presence of uncapped CdS nanoparticles which indicates thereis no change in fluorescence lifetime of BSA in the presence ofuncapped CdS representing the absence of interaction between thetwo. In Fig. 4(b) though the decay traces of BSA in both the absenceand presence of uncapped CdS were actually plotted however thelifetime of BSA remained the same in both conditions, hence themerging of kinetic traces were observed (the plot looks like a sin-gle decay curve). For static quenching, we can deduce the bindingconstant (K) resulted from the formation of ground state complexbetween fluorophore and the quencher. Hence the binding constant(K) was calculated by the method given in the following section.

3.5. Binding constant and binding sites

The intrinsic tryptophan fluorescence quenching of BSA bycapped CdS also follows complex formation, so we can deduce thebinding constant (K) which is calculated by using the followingmethod.

If it is assumed that there are similar and independent bindingsites in the BSA, the relationship between the fluorescence intensityand the quencher medium can be deduced from the following Eq.(11) [36]:

nQ + BK�Qn . . . B (11)

where B is the fluorophore, Q is the quencher, nQ + B is the pos-tulated complex between a fluorophore and n molecules of the


M. Asha Jhonsi et al. / Colloids and Surfaces B: Biointerfaces 72 (2009) 167–172 171

Fig. 5. The plot of log [(F0 − F)/F] versus log[Q] for BSA with starch capped CdS.

quencher. The constant K is given by

K = [Qn . . . B][Q ]n · [B]

(12)

If the overall amount of biomolecules (bound or unbound withthe quencher) is B0, then [B0] = [Qn. . .B] + [B], here [B] is theconcentration of unbound biomolecules, then the relationshipbetween fluorescence intensity and the unbound biomolecule as[B]/[B0] = F/F0 that is

log

[F0 − F

F

]= log K + n log[Q ] (13)

where K is the binding constant of BSA with starch capped CdS,which can be determined from the plot of log [(F0 − F)/F] versuslog [Q] curve as shown in Fig. 5 and thus we obtained binding con-stant (K) 6.6 × 102 M−1 (R2 = 0.9944) and number of binding sites (n)0.8631 from intercept and slope of the plot respectively. The valuesof binding constant obtained from the fluorescence data matcheswell with that determined from the absorption spectral changes.The good agreement between these values of binding constanthighlighted the validity of assumption proposed for the associationbetween BSA and starch capped CdS. The number of binding siteswhich is close to 1 indicates, there is only one type of interactionbetween BSA and starch capped CdS.


Influences of capped CdS on the conformational changes ofBSA were assessed by synchronous fluorescence method. Syn-chronous fluorescence measurements provide information aboutthe molecular microenvironment in the vicinity of fluorophorefunctional groups. Synchronous fluorescence spectra were obtainedby simultaneous scanning of excitation and emission monochroma-tors. According to Miller [37], as the difference between excitationand emission wavelength (��) is 15 nm, synchronous fluorescence

Fig. 6. Synchronous spectra of BSA (1 × 10−6 M) in the absence and presence of starchcapped CdS (0–5 × 10−4 M) in the wavelength difference of �� = 60 nm.

Fig. 7. FT-IR spectrum of (a) uncapped CdS (pink), SCdS (red), BSA (blue) and BSAbound with SCdS (green) in the range of 400–4000 cm−1. (b) Free BSA (red) and SCdSbound with BSA (green) in the range of 1000–2200 cm−1. (For interpretation of thereferences to color in this figure legend, the reader is referred to the web version ofthe article.)

offers the characteristics of tyrosine residues (spectra not shownhere), while when �� is 60 nm, it provides the characteristic infor-mation of tryptophan residues. Synchronous fluorescence spectraof BSA upon addition of starch capped CdS gained at 60 nm is shownin Fig. 6.

The fluorescence intensity of both tryptophan and tyrosine weredecreased but the emission wavelength of tryptophan (340 nm) isblue shifted (2 nm) with increasing concentration of capped CdS.Comparing the emission wavelength of tyrosine, no significantchange was observed. It indicated that the interaction of capped CdSwith BSA does not affect the conformation of tyrosine micro-region.Firstly, combined with tyrosine capped CdS gradually interacts withtryptophan and brings changes to BSA and results in blue-shiftof fluorescence wavelength. It is likely due to that the hydropho-bic amino acid structure surrounding tryptophan residues inBSA tends to collapse slightly and thus tryptophan residues areexposed more to the aqueous phase. Similar observation has beenreported [38].



Scheme 3. Mode of interaction between starch capped CdS and BSA molecules.

Fig. 8. Absorption and emission spectra of BSA (red and blue) and SCdS nanoparticles(green and pink) respectively. (For interpretation of the references to color in thisfigure legend, the reader is referred to the web version of the article.)

3.7. FT-IR characterization

In order to confirm the binding of BSA with SCdS, FT-IR mea-surements were performed. Fig. 7(a) shows the FT-IR spectra ofuncapped CdS, starch capped CdS, free BSA and BSA bound withSCdS. In the free BSA and uncapped CdS there is no peak in theregion of 1390 cm−1 but it was observed in SCdS which is for thecapping agent starch. This peak is completely disappeared in thespectrum of BSA bound with SCdS, so the interaction between BSAand SCdS through the capping agent (starch) has been proved asshown in Scheme 3.

Fig. 7(b) indicates that whether the conformation of BSA haschanged or not, Usually the amide I peak position for BSA occurs inthe region of 1600–1700 cm−1 and amide II band near 1548 cm−1

[39,40]. Former one is the more sensitive to change of protein con-formation than later. In this figure BSA in the absence of capped CdSshows the amide I peak at 1638 cm−1 and in the presence of cappedCdS the amide I peak was shifted to 1647 cm−1. From the shift inpeak position we confirmed that the conformation of BSA has beenaffected by the addition of starch capped CdS.

4. Conclusions

Uncapped and starch capped CdS nanoparticles were preparedin aqueous medium and their interaction with BSA was studied.The results presented clearly indicated that capped CdS quenchesthe fluorescence of BSA through complex formation. The quench-ing rate constant, binding constant, and number of binding siteswere calculated according to the relevant fluorescence data. Fromthe synchronous fluorescence spectra, it is established that theconformational changes of BSA occurred. The static quenchingmechanism was confirmed by fluorescence lifetime measure-ments.

Acknowledgements

R.R. and M.A thank DST (Ref: SR/S5/NM-16/2007 dt: 3.7.3008)(Government of India) for the project and fellowship, respectively.

We are thankful to Laser Spectra of Services, Bangalore for life-time measurements.

References

[1] L. Tan, L. Liu, Q.J. Xie, Y.Y. Zhang, S.Z. Yao, Anal. Sci. 20 (2004) 441.[2] M.B.M. Moronne Jr., P. Gin, S. Weiss, A.P. Alivisatos, Science 281 (1998) 2013.[3] W.C.W. Chan, S. Nie, Science 281 (1998) 2016.[4] F.X. Wu, J.W. Lewis, D.S. Kliger, J.Z. Zhang, J. Chem. Phys. 118 (2003) 12.[5] X.H. Zhong, Y.Y. Feng, W. Knoll, M.Y. Han, J. Am. Chem. Soc. 125 (2003) 13559.[6] M. Idowu, E. Lamprech, T. Nyokong, J. Photochem. Photobiol. A: Chem. 198

(2008) 7.[7] C. Haynes, W. Norde, J. Colloid Interf. Sci. 169 (1995) 313.[8] K. Kandori, T. Shimizu, A. Yasukawa, T. Ishikawa, Colloids Surf. B 5 (1995) 81.[9] V. Ball, A. Bentaleb, J. Hemmerle, J.C. Voegel, P. Schaaf, Langmuir 12 (1996) 1614.

[10] C.E. Giacomelli, M.J. Esplandiu, P.I. Ortiz, M.J. Avena, C.P. De Pauli, J. Colloid Interf.Sci. 218 (1999) 404.

[11] A. Kathiravan, R. Renganathan, Colloids Surf. A: Physicochem. Eng. Aspects 324(2008) 176.

[12] A. Kathiravan, S. Anandan, R. Renganathan, Colloids Surf. A: Physicochem. Eng.Aspects 333 (2009) 91.

[13] W. Li, S. Li, Colloids Surf. A: Physicochem. Eng. Aspects 295 (2007) 159.[14] T.J. McMaster, M.J. Miles, P.R. Shewry, A.S. Tatham, Langmuir 16 (2000) 1463.[15] F.Y. Oliva, L.B. Avalle, O.R. Camara, C.P. De Pauli, J. Colloid Interf. Sci. 261 (2003)

299.[16] V.N. Vasilets, A.V. Kuznetsov, V.I. Sevastianov, J. Biomed. Mater. Res. A 69 (2004)

428.[17] N.O. Sahin, D.J. Burgess, II Farmaco 58 (2003) 1017.[18] H.R. Byon, H.C. Choi, J. Am. Chem. Soc. 128 (2006) 2188.[19] H. Zhang, M.E. Meyerhoff, Anal. Chem. 78 (2006) 609.[20] J.E. Butler, Methods 22 (2000) 4.[21] C.H. Jang, B.D. Stevens, P.R. Carlier, M.A. Calter, W.A. Ducker, J. Am. Chem. Soc.

124 (2002) 12114.[22] S. Dilson, M.C. Celia, C.B. Jayme, L. Sonia, Toxicol. Lett. 147 (2004) 53.[23] R.E. Olson, D.D. Christ, Ann. Rep. Med. Chem. 31 (1996) 327.[24] M. Dockal, D.C. Carter, F. Ruker, J. Biol. Chem. 275 (2000) 3042.[25] Y.J. Hu, Y. Liu, R.M. Zhao, J.X. Dong, S.S. Qu, J. Photochem. Photobiol. A: Chem.

179 (2006) 324.[26] B. Zhou, Z.D. Qi, Q. Xiao, J.X. Dong, Y.Z. Zhang, Y. Liu, J. Biochem. Biophys. Meth-

ods 70 (2007) 743.[27] J. Aldana, Y.A. Wang, X. Peng, J. Am. Chem. Soc. 123 (2001) 8844.[28] S.F. Wuister, A. Meijerink, J. Lumin. 105 (2003) 35.[29] S. Sadhu, P.S. Chowdhury, A. Patra, J. Lumin. 128 (2008) 1235.[30] Y.P. Wang, Y. Wei, C. Dong, J. Photochem. Photobiol. A: Chem. 177 (2006) 6.[31] X. Shen, X. Liou, H. Li-Ping Ye, Z. Liang, J. Wang, Colloid Interf. Sci. 311 (2007)

400.[32] P. Rodriguez, N.M. Aguirre, E.S.M. Martinez, G.G. Cruz, S.A. Tomas, O.Z. Angel, J.

Crystal Growth 310 (2008) 160.[33] H.A. Benesi, J.H. Hildebrand, J. Am. Chem. Soc. 71 (1949) 2703.[34] J.R. Lakowicz, G. Weber, Biochemistry 12 (1973) 4161.[35] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, Plenum Press, New York,

1983, p. 258.[36] V. Anbazhagan, R. Renganathan, J. Lumin. 128 (2008) 1454.[37] J.N. Miller, Proc. Anal. Div. Chem. Soc. 16 (1979) 203.[38] N. Zhou, Y.Z. Liang, P. Wang, J. Photochem. Photobiol. A: Chem. 185 (2007) 271.[39] P. Athina, J.G. Rebecca, A.F. Richard, J. Agric. Food Chem. 53 (2005) 158.[40] S. Molet, P. Gosset, P. Lassalle, W. Czarlewski, A.B. Tonnel, Clin. Exp. Allergy 27

(1997) 1167.

Documents

List of Publications - INFLIBNETshodhganga.inflibnet.ac.in/bitstream/10603/4804/14/14... · 2015. 12. 4. · List of Publications 11. A. Kathiravan, S. Anandan and R. Renganathan,