Effect of silver nanoparticles on fluorescence and nonlinear properties of naturally occurring betacyanin dye

Optical Materials 39 (2015) 211–217

Contents lists available at ScienceDirect

Optical Materials journal homepage: www.elsevier.com/locate/optmat

Effect of silver nanoparticles on fluorescence and nonlinear properties of naturally occurring betacyanin dye Arindam Sarkar a,⇑, Aparna Thankappan a,b, V.P.N. Nampoori a a b

International School of Photonics, Cochin University of Science and Technology, Cochin, Kerala, India Inter University Centre for Nanomaterials and Devices, Cochin University of Science and Technology, Cochin, Kerala, India

a r t i c l e

i n f o

Article history: Received 24 May 2014 Received in revised form 16 October 2014 Accepted 11 November 2014 Available online 5 December 2014 Keywords: Plasmonics Scattering Nonlinear optics Materials for solar cells Silver nanoparticles Betacyanin

a b s t r a c t We present the linear and nonlinear optical studies of a natural dye betacyanin extracted from red beet root in the presence of silver nano particles in colloidal solution. We synthesized silver nano particles and characterized by XRD and HRTEM. We show how appropriate concentration of silver nanoparticles can enable tuning of dye fluorescence efficiency. Nonlinear properties are studied using open aperture Z scan technique with Nd:YAG laser (532 nm, 7 ns, 10 Hz). We show modification of nonlinear properties for the dye to the desired level can be achieved in the presence of silver nanoparticles. High nonlinearity we also demonstrated in PVA/Ag nano/Betacyanin composite films. Theoretical analysis is performed using model based on nonlinear absorption of materials and scattering of metal nanoparticles. Ó 2014 Elsevier B.V. All rights reserved.

1. Introduction The optical material having large reverse saturable absorption (RSA) and nonlinear refraction can produce irradiance dependent transmittance and phase shift, which can limit the throughput fluence (optical energy per unit area). In the present age of advancement of optical technology there is a great importance of such materials in optical limiting related applications [1]. Now we also know that combinations of saturable absorber (SA) and reverse saturable absorber (RSA) type materials can be used for pulse compression, mode locking and different other pulse shaping schemes for high power laser amplifiers [2,3]. SA property and other third order nonlinear properties are also being exploited very highly precise techniques such as frequency combs [4]. Passive all optical diode using asymmetric nonlinear absorption is also demonstrated in recent past [5]. This SA and RSA properties can be used in optical computing and various similar applications such as spatial line modulators (SLM) [3]. So it is evident that the optical materials with large nonlinearity can be used as key component in modern photonic technologies such as optical computing, fluorescence imaging and different applications of optical limiting. There is also a need of optical materials which are easy to synthesize, stable in high power applications, cost effective and ⇑ Corresponding author. E-mail address: [email protected] (A. Sarkar). http://dx.doi.org/10.1016/j.optmat.2014.11.028 0925-3467/Ó 2014 Elsevier B.V. All rights reserved.

hazardless in the safety grounds. Not only pigments like chlorophyll and carotenoids Indian subcontinent is full of natural dye containing plants [6]. Authors of this article believe that these cheap dyes can be good solution as new generation optical materials. One of the common natural pigments is betalain and quite extensively is being investigated for the application in Dye-sensitized solar cells (DESC) [7–9]. Betalain pigments are commonly found in red beet root and can easily be extracted by diffusion in room temperature or hot extraction method. The betalain pigments consists of the red–purple betacyanins (See Fig. 1), betanin (I) and betanidin (II), with maximum absorptivity kmax about 535 nm and 542 nm respectively. The yellow betaxanthins (most commonly Indicaxanthin) have kmax near 480 nm [10]. It would be worth mentioning that the amount of each pigment present in the extracted dye can actually shift the absorption spectra and emission spectra of the dye. Hence, observing the peaks of the sample it can easily be identified the purity of the sample. If with the increase of intensity, excited states show saturation for their long lifetimes, the transmission will show SA characteristics and if the exited excited state has stronger absorption compared to the ground state it will show RSA characteristics. So, it becomes very important to identify the nonlinear absorption effects with determination of saturation intensity for SA material and two photon absorption (TPA) coefficients for two photon absorbing material. The RSA property of silver composite is also being tested recently [11]. It is also reported in recent works that

212

A. Sarkar et al. / Optical Materials 39 (2015) 211–217

Fig. 2. Ag NP absorption spectra. Fig. 1. Structure of betacyanin.

silver nanostructures can be used as plasmonic fluorescence enhancement for the dyes [12]. One of the other recent studies indicates scattering of metal nanoparticles can be used for light trapping and such principle can be used for plasmonic solar cells (PSC) [13,14]. One of the other challenges is improvement of fluorescence efficiency in present generation optical materials. In this article we will try to address all above requirements with our dye under test and silver nanoparticles. 2. Experiments We have used polyol method [15] for synthesis of silver (Ag) nanoparticles (NPs). In this method ethylene glycol (EG) (99% pure) is heated at 120° centigrade. At this temperature silver nitrate (pure) and polyvinyl pyrrolidone (PVP) (Mw = 10,000) (99%) (Sigma–Aldrich) is added. PVP is added to prevent agglomeration in AgNPs. We maintained 12:1 M ratio between AgNO3 and PVP. The solution is allowed to heat for next thirty minutes to have following reactions:

OH—CH2 —CH2 —OH ! CH3 —CHO þ H2 O

ð1Þ

3. Results and discussions

ð2Þ

Fig. 4 is showing XRD results of a thin film made from our Ag NPs. In XRD we got peak at 2h = 39.147°. We can easily calculate the spacing between planes (d) from the same with Bragg’s law [17]:

2ðCH3 —CHOÞ þ 2AgNO3 ! ðCH3 —CO—CO  CH3 Þ þ 2Ag þ 2HNO3

measured in open aperture (OA). We used Q-switched Nd:YAG laser (Spectra Physics LAB-1760, 532 nm, 7 ns, 10 Hz) as light source. 20 cm converging lens is used to focus the laser beam. The beam radius w0 was calculated to be 35.5 lm. The Rayleigh length, Z0 = pw20/k was calculated as 7.4 mm. We have tested Z scan in following samples, B1 = Dye without Ag NPs, A1 = 3 mmole/l solution of Ag NPs in EG, C0 = Dye with 0.27 mmole/l concentration of Ag NPs, C1 = Dye with 5 mmol/l concentration of Ag NPs, C2 = Dye with 9.6 mmol/l of Ag NPs, C3 = Dye with 14 mmol/l of Ag NPs. Dye concentration was kept as 5 ⁄ 104 mM for all these samples. Sample C4 is a composite 70 lm film made of Ag NP, polyvinyl acetate (PVA) and dye. This film is made from 10 ml of dye, 10 ml water and 2 gram PVA. We added 2.7 mmol/l Ag nano in the same. Free standing film is generated by tape casting. For rest of the samples we used 1 mm thickness cuvette to make our sample thickness less than Rayleigh length. We have characterized our samples by calculating two photon absorption (TPA) coefficient (b), third order nonlinear optical susceptibility Im[v(3)] and optical limiting threshold.

Particles are deposited using centrifuge and washed with ethanol. We have taken absorption spectra of both AgNP and beetroot extraction using spectrometer (Jasco V-570 UV/VIS/IR). We got 420 nm as absorption peak for silver nano sample (Fig. 2). We have extracted dye from fresh cut beet root by diffusion process of extraction where EG is used as solvent. Absorbance of the extracted dye is shown in Fig. 3(a). The sample has peak absorption of 2.38833 near 542 nm, so we can easily conclude with this peak and shape of the spectra that the pigment present is betacyanin dye with betanidin as main component [10]. Fig. 3(b) shows variation of absorbance with introducing different concentrations of Ag NP in the dye. The fluorescence studies are done for the dye and dye with the presence of Ag NP in different concentration using a Cary Eclipse fluorescence spectrometer (Varian). Ag NPs were characterized using high resolution transmission electron microscopy (HRTEM) and X-ray diffraction (XRD). TEM was taken on JEOL 3010. XRD was taken on AXS Bruker D% diffractometer using Cu Ka-radiation (k = 0.1541 nm). We have checked the third order optical characteristics of samples using single beam Z-scan technique as proposed by SheikBahae et al. [16]. The transmission characteristics that changes near the focal point during the process of sample translation were

2d sin h ¼ nk

ð3Þ

where h is the scattering angle, k be the wavelength of the X ray, and n is any integer. With this d becomes 2.2984 Å. We take our lattice constant a = 4.0867 Å when face centred cube (FCC) is assumed [18]. Using (1/d2) = (h2 + k2 + l2)/a2 with Miller indices (h, k, l), we understand that our peak belongs to the plane (1, 1, 1). This also matches to the JCPDS data (file No. 04-0783). With XRD data crystallite size can easily be determined by Sherrer’s equation [19,20]:

Bð2hÞ ¼ Kk=L cos h

ð4Þ

where L is the particle size, B(2h) is the half value breadth. Full width half maximum (dhFWHM) is 0.29327°, so we get B(2h) = 0.005119. Assuming K = 0.93 with Eq. (4) XRD average this size becomes nearly 30 nm. We determined the particle size with HRTEM (Fig. 5). TEM images also confirm particles prepared by us are approximately spherical. The average particle size calculated from TEM image is 39.58 nm.

213

A. Sarkar et al. / Optical Materials 39 (2015) 211–217

Fig. 3. (a) Absorption spectra of dye. (b) Absorption spectra of dye with different concentrations of silver nanoparticles.

Fig. 4. XRD of silver nanoparticle thin film.

The dye has peak absorbance of 2.38833. If we take one third of the peak absorption the absorption range is from 408 nm to 592 nm. We introduced different concentration of Ag NPs to observe variation in the level the fluorescence output. In our experiments we observed that initially if small amount of the particles are present in betacyanin solution the level of fluorescence starts decreasing. But after a certain concentration it starts increasing and with optimum level of Ag NP concentration it even exceeds the fluorescence level of the dye. But if we continue to increase the Ag NP concentration it starts decreasing again. The effect can be understood observing the peak values of fluorescence intensity for different Ag NP concentration as shown in Table 1 for different excitation wavelengths. The variation of florescence intensity with different concentration of Ag NP can be also seen in our Fig. 6(a)– (d) for different excitation wavelengths. Fig. 6(e) is showing how peak fluorescence intensity of the dye varies in the presence of different concentration of Ag NPs. We know that relative strength of absorption and scattering of the individual metal nanoparticles can be determined from absorption and scattering cross sections Cabs and Csca respectively. They are given by for wavelength (k) [21,22]:

C abs ¼

    4 2p 1 2p Im½a; C sca ¼ jaj2 6p k k

aðIÞ ¼ a0

 1þ

I Is

 ð7Þ

And nonlinear absorption coefficient in terms of TPA coefficient is given by [16],

aðIÞ ¼ a0 þ bI

ð8Þ

where Is be the saturable intensity and b is TPA coefficient, a0 is the linear absorption coefficient with intensity I. Now, if we assume Eq. (8) as generalized equation for nonlinear absorption coefficient, then we can equate Eqs. (7) and (8) for SA material. Equating these two equations we get:

bI ¼ a0

    I I 1þ Is Is

ð9Þ

Now putting I = Is in Eq. (9) we get:

ð5Þ

where a is polarizability of the particle and if we assume spherical shape sub-wavelength size particle then it is given by,

a ¼ 3Vð  1Þ=ð þ 2Þ

where V is the volume and e be the dielectric constant of the metal. So, the absorption is related to volume and scattering is related to square of the volume for metal NPs. In our nano colloidal solution other than NPs betacyanin dye is present as fluorophore. So initially when small amount metal particles are present then scattering from the NPs is again absorbed by the dye other than the absorption of light by the NPs for the corresponding emission wavelength. Now after certain increase of the concentration of metal NPs total scattering increase substantially and the fluorescent output starts increasing since it overcomes the barrier from dye absorption. But this enhancement cannot be exceeded after certain limit for the emission wavelength since emission intensity is limited. So after a certain increment of the concentration of metal NPs the output starts to decrease again since new NPs will have some amount of absorption also but no additional photons at emission wavelength to increase the efficiency. We enhanced the efficiency of the fluorescence up to level of 23.99% compared the dye fluorescent level without the presence of Ag NPs in our experiment. Though this efficiency varies with the excitation wavelength since stokes shifted emission can be again absorbed by the metal NPs and the fluorophore, which is also dependent on the wavelength as shown earlier in Figs. 2 and 3. This effect results the fluorescence tuning of the dye as per requirement with varying concentration of Ag NPs. Now as per common mathematical model nonlinear absorption coefficient for SA material is given by [23],

ð6Þ

  a0 Is ¼  2b

ð10Þ

Since Is and a0 are positive then b becomes evidently negative which is actually the consequence of negative nonlinear absorption of SA.

214

A. Sarkar et al. / Optical Materials 39 (2015) 211–217

Fig. 5. (a) and (b) showing TEM images of Ag nanoparticles, (c) is showing their size distribution.

Table 1 Fluorescence peak for different concentration of silver in the dye with different excitation wavelengths (k). Concentration of silver NP in dye (mmol/l)

k = 420 nm (a.u)

k = 440 nm (a.u)

k = 520 nm (a.u)

k = 540 nm (a.u)

0 0.025 0.105 0.41 0.42 0.435 0.95

137.42 105.32 97.87 119.74 152.47 121.71 92.52

203.62 162.88 137.52 197.24 239.71 199.78 155.89

85.36 83.24 77.20 88.26 105.84 99.66 79.85

87.32 77.50 75.01 90.76 107.57 95.54 79.08

10.95

17.73

23.99

23.18

Enhancement (%)

We have carried open aperture Z scan and analyzed the data following techniques described by Sheik-Bahae et al. [16]. Now in Z scan the normalized transmittance is given by [16],

1 Tðz; S ¼ 1Þ ¼ pffiffiffiffi pq0 ðz; 0Þ where

q0 ðz; 0Þ ¼ bI0 Leff

,

Z1

ln½1 þ q0 ðz; 0Þe

s2

 ds

ð11Þ

ð12Þ

Here I0 is the laser intensity in the focal plane, b is the TPA coefficient, Leff is the effective thickness of the sample of length l,

Leff ¼ ð1  expða0 lÞÞ=a0 For |q0| < 1 normalized transmittance can be written as,

1 X ½q0 ðz; 0Þm =ðm þ 1Þ3=2

ð13Þ

ð14Þ

m¼0

From OA Z scan third order nonlinearity Im[v(3)] can also be calculated. Same is given by [16],

Im½vð3Þ  ¼ bðe0 n20 c2 =wÞ

1

 2 ! Z 1þ Z0

Tðz; S ¼ 1Þ ¼

ð15Þ

where e0 is the permittivity of free space, c is the speed of light in vacuum and n0 is the linear refractive index of the sample. Normalized transmittance curves from OA Z scan for different samples are shown in Fig. 7(a)–(c). For the pure dye solution (B1) we got SA characteristics and for pure Ag NP solution (A1) we got RSA characteristics. All these coefficients are calculated considering spherical AgNP of a particular size synthesized by authors and it is also seen that pure AgNP curve nicely fits into the mathematical models of Z scan [14] (Fig. 7(a)). We calculated b for all

A. Sarkar et al. / Optical Materials 39 (2015) 211–217

215

Fig. 6. (a)–(d) showing variation of fluorescence intensity of dye for different concentration of Ag NPs, (e) is showing variation of peak fluorescence intensity of dye for different Ag NP concentration, with different excitation wavelengths.

the samples at 399.5 MW/cm2 b for B1 is 43.95 cm/GW. From Eq. (10) we calculate Is = 26.81 MW/cm2. The parameters obtained from OA Z scan i.e. b, Im[v(3)] and optical limiting threshold for A1 and other samples containing Ag NPs are shown in Table 2. Now, as shown in Table 2 (with sample C0, C1, C2, C3) with increase of the concentration of silver NPs in the colloidal solution the value of b increases and optical threshold of the samples decreases. The transition between SA to RSA of the dye with Ag

NPs can be well observed in the sample C0 where both properties are visibly present in the Z scan curve (Fig. 7(b)). We have also checked with Ag/Betacyanin/PVA composite film (C4) shows very good RSA characteristics (Fig. 7(c)) and have very high b value (Table 2). Optical limiting threshold can also be understood by Fig. 7(d). It is evident from this figure that Ag NP concentration variation can result threshold tuning for optical limiting application for the material.

216

A. Sarkar et al. / Optical Materials 39 (2015) 211–217

Fig. 7. Open aperture Z scan for different samples (a) A1 for two different power 91 MW/cm2 and 399.5 MW/cm2 (b) B1,C0,C1,C2 at 399.5 MW/cm2 (c) C3,C4 at 399.5 MW/ cm2. (d) Optical limiting response of different samples obtained from Z scan measured at 399.5 MW/cm2.

Table 2 Variation of TPA coefficient (b), Im[v(3)] and optical limiting threshold for different samples. Sample

b (cm/ GW)

Optical limiting threshold (MW/ cm2)

Im[v(3)] (1019) (m2/V2)

A1 C0 C1 C2 C3 C4

45.352 10.8869 46.4146 55.408 59.8823 687.98

205.99 – 203.19 94.07 41.21 172.57

2.0593 0.4834 2.0609 2.4602 2.6589 39.6159

4. Conclusion In this article we show that betacyanin dye extracted in EG which has good SA characteristics. We also synthesized Ag NPs of average size 39.58 nm. We showed that with varying concentration of Ag NPs, the dye fluorescence efficiency can be increased as well as decreased and thus resulting fluorescence tuning of the dye. Nonlinear absorption in nano-clusters can be caused by various processes like transient absorption, two-photon or multiphoton absorption, inter-band and intra-band transitions, and

nonlinear scattering. Different studies already have shown that at higher input fluence Ag NPs can exhibit RSA characteristics at 532 nm which generally shows SA in lower intensity. Detailed physical process for the SA and RSA behavior for Ag NP is already discussed elsewhere [24,25]. These properties allowed us to achieve desired value of nonlinear absorption coefficients and optical limiting threshold modification for the dye by varying concentration of Ag NPs. Other than this we should note that the optical limiting curve can further be modified using AgNPs of different size and shape since it is also influenced by scattering effects [26]. But effect of such variation is much smaller than the effect caused by concentration based tuning. We also observed RSA characteristic is exhibited in PVA/Ag nano/Betacyanin composite film. Thus in our work we showed that the linear and nonlinear optical property modification of the dye is possible with appropriate concentration of Ag NPs. It increases the applicability of the corresponding natural dye and extends its application domain in the related field.

Acknowledgements Authors wish to acknowledge SAIF-KOCHI for XRD facility and SAIF, IIT Madras for TEM facility. Author AS acknowledges UGC

A. Sarkar et al. / Optical Materials 39 (2015) 211–217

and author AT acknowledges to IUCND for financial support. Authors also acknowledge S. Divya, S. Mathew and Roopa Venkataraj for their support and encouragement. References [1] T.-H. Wei, T.-H. Huang, T.-T. Wu, P.-C. Tsai, M.-S. Lin, Chem. Phys. Lett. 318 (2000) 53. [2] Yehuda B. Band, D.J. Harter, R. Bavli, Chem. Phys. Lett. 126 (1986) 280. [3] K.P.J. Reddy, Curr. Sci. 61 (8) (1991) 520. [4] A. Schliesser, N. Picque, T.W. Hansch, Nat. Photonics 6 (2012) 440–449. [5] R. Philip, M. Anija, C.S. Yelleswarapu, D.V.G.L.N. Rao, Appl. Phys. Lett. 91 (2007) 1411181. [6] S.B. Gokhale, A.U. Tatiya, S.R. Bakliwal, R.A. Fursule, Nat. Prod. Radiance 3 (4) (2004). 228-224. [7] A.R. Hernandez-Martinez, M. Estevez, Susana Vargas, Fracisco Quintanilla, Int. J. Mol. Sci. 12 (2011) 5565–5576. [8] Aparna Thankappan, Sheenu Thomas, V.P.N. Nampoori, J. Appl. Phys. 112 (2012) 123104. [9] Aparna Thankappan, Sheenu Thomas, V.P.N. Nampoori, Opt. Mater. 35 (2013) 2332–2337. [10] D. Zhang, S.M. Laniera, J.A. Downing, J.L. Avent, J. Lumc, J.L. McHalea, J. Photochem. Photobiol. A 195 (2008) 72.

217

[11] Q. Ouyang, X. Di, Z. Lei, L. Qi, C. Li, Y. Chen, Phys. Chem. Chem. Phys. 15 (2013) 11048. [12] F. Xie, J.S. Pang, A. Centeno, M.P. Ryan, D.J. Riley, N.M. Alford, Nano Res. 6 (7) (2013) 496–510. [13] K.R. Catchpole, A. Polman, Appl. Phys. Lett. 93 (2008) 191113. [14] K.R. Catchpole, A. Polman, Opt. Express 16 (26) (2008) 21794. [15] D. Kim, S. Jeong, J. Moon, Nanotechnology 17 (2006) 4019–4024. [16] M. Sheik-Bahae, A.A. Said, T.H. Wei, D.J. Hagan, E.W. Van Stryland, IEEE J. Quant. Electron. 26 (1990) 760. [17] G.E.M. Jauncey, Proc. National Acad. Sci. 10 (1924). [18] P.J. Schields, N. Dunwoody, M. Mamak, C. Gendron, S. Bates, JCPDS – Int. Centre Diffr. Data 162–168 (2008). [19] P. Scherrer, Göttinger Nachr. Math. Phys. 2 (1918) 98–100. [20] A.L. Patterson, Phys. Rev. 39 (1939) 979–982. [21] C.F. Bohren, D.R. Huffman, Absorption and Scattering of Light by Small Particles, Wiley, New York, 1983. [22] Howard R. Stuart, Dennis G. Hall, Appl. Phys. Lett. 73 (1998) 3815. [23] W. Robert Boyd, Nonlinear Optics, third ed., Academic Press, 2008. [24] U. Gurudas, E. Brooks, D.M. Bubb, S. Heiroth, T. Lippert, A. Wokaun, J. Appl. Phys. 104 (2008) 073107. [25] M. Hari, S. Mathew, B. Nithyaja, S.A. Joseph, V.P.N. Nampoori, P. Radhakrishnan, Opt. Quant. Electron. 43 (2012) 49–58. [26] O. Muller, S. Dengler, G. Ritt, B. Eberle, Appl. Opt. 52 (2) (2013) 140–149.