Heat treatment effect

0 downloads 0 Views 1MB Size Report
Jun 8, 2016 - 116 (2014) 123504. [19] S. Derom, A. Berthelot, A. Pillonnet, O. Benamara, A.M. Jurdyc, C. Girar, G. Colas · des Francs, Phys. Opt. (2013) 1312.
Journal of Alloys and Compounds 686 (2016) 556e563

Contents lists available at ScienceDirect

Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom

Growth of silver nanoparticles stimulate spectroscopic properties of Er3þ doped phosphate glasses: Heat treatment effect rid a I. Soltani a, S. Hraiech a, *, K. Horchani-Naifer a, H. Elhouichet a, b, B. Gelloz c, M. Fe a Physical Chemistry Laboratory of Mineral Materials and Their Applications, National Center of Research in Materials Science, B.P. 73, 8027, Soliman, Tunisia b Department of Physics, Faculty of Sciences of Tunis, Tunis- El Manar University, El Manar, 2092, Tunisia c Graduate School of Engineering, Nagoya University, 2-24-16 Furo-cho, Chikusa-ku, Nagoya, Aichi, 464-8603, Japan

a r t i c l e i n f o

a b s t r a c t

Article history: Received 17 February 2016 Received in revised form 31 May 2016 Accepted 6 June 2016 Available online 8 June 2016

The melt quenching technique was used to prepare Er3þ ion doped phosphate glasses containing silver nanoparticles (Ag NPs). The amorphous nature of the glass is confirmed by X-ray diffraction patterns. Homogeneous distribution of spherical Ag NPs was shown from the transmission electron microscopy (TEM). The EDX analysis shows the presence of Ag element. The DSC measurements indicate a relatively good thermal stability of the prepared glass samples. The characteristic surface plasmon resonance (SPR) band of Ag NPs was observed in the range of 420e570 nm in the optical absorption spectra. The SPR band stimulated by the presence of Ag NPs enhanced both the photoluminescence (PL) intensity and the fluorescence lifetime relative to the 4I13/2 / 4I15/2 transition of Er3þ ion. Ideal PL enhancement was obtained after 12 h of heat-treatment. The results indicate that there is a large influence of nucleation and growth of silver NPs on the optical and spectroscopic properties of the glass samples. The present results indicate that the glass heat-treated for 12 h has a good prospect as a gain medium applied for 1.53 mm band broad and high-gain erbium-doped fiber amplifiers. © 2016 Elsevier B.V. All rights reserved.

Keywords: Heat treatment Phosphate glasses Nanoparticles Surface plasmon resonance (SPR) PL enhancement

1. Introduction Phosphate-based glasses drew much interest due to the high chemical durability [1,2], good mechanical and good thermal stability. In addition, the optical properties of phosphate glasses include excellent transparency [3]. These favorable features make phosphate glasses useful in optical devices. Phosphate glasses containing rare earth (RE) ions incorporating metallic nanoparticles (NPs) such as gold (Au) or silver (Ag) NPs have been widely used [4e7]. Many compositions of phosphate glasses containing metallic NPs have been reported such us Yb3þ/ Er3þ co-doped PbOeGeO2 glasses containing silver NPs [8e11], phosphate glasses with compositions P2O5eMgOexAgCle0.5Er2O3 [12] also Er3þ doped phosphate glasses with Ag NPs [13e15]. The mechanism of interaction between the metallic NPs and the rare earth (RE) ions is a precondition for the development of photonic devices [16,17]. Trivalent erbium (Er3þ) ion, is the most important active ion in the RE family due to the favorable energy

level structure, has been exploited in a wide variety of glasses for applications in the solid-state lasers and the optical amplifiers [18]. The photoluminescence intensity of RE ions can be enhanced for many reasons. This enhancement is primarily based on the SPR strong local electric field presence in the proximity of RE ions. Also, it is attributed to the energy transfer (ET) from AgNPs to RE ions [19]. This enhancement is mainly an improvement of absorption process: we have a strong exaltation of the excitation field resonant with the resonance frequency, of the localized surface plasmon associated to the metallic nanoparticles [20,21]. The phenomenon of surface plasmon resonance (SPR) is classically described as the oscillation of the free electrons with respect to the ionic background of the nanoparticles, when they are collectively excited by laser irradiation [22,23]. Metal enhanced luminescence is mostly studied using Ag NPs due to a speed-up reduction of the silver than gold (Au) NPs [24e26]. The silver NPs are reduced in one step, while for the gold NPs, it occurs in three steps: Agþ þ 1e / Ag0

* Corresponding author. E-mail address: [email protected] (S. Hraiech). http://dx.doi.org/10.1016/j.jallcom.2016.06.027 0925-8388/© 2016 Elsevier B.V. All rights reserved.

Au3þ þ 3e / Au0

I. Soltani et al. / Journal of Alloys and Compounds 686 (2016) 556e563

On the other hand, the reduction potential of the silver (E0 ¼ 0.7996) in room temperature and at equilibrium with air is smaller than that gold (E0 ¼ 1.498) [27]. To the best of our knowledge, only few studies have been reported in literature, dealing with the effect of silver nanoparticles on optical characteristics at 1.53 mm of phosphate-ZnO glass host [28,29]. In the present study, we report on the thermal, structural, and optical properties of phosphate glass containing Ag NPs to understand the nucleation and growth mechanisms of silver NPs and to investigate their effect on erbium emission. 2. Experimental The glass samples having composition 42P2O5e42Na2Oe15ZnOe0.5Er2O3e0.5Ag2CO3 were prepared using the melt quenching method, from the starting chemical constituents NaH2PO4, ZnO, Er2O3 and Ag2CO3. All the starting chemical constituents are highpurity (99.9%). The reagents were melted in an electric furnace at 900  C at a rate of 10  C/min over 2 h in order to have complete melting. Then, the melt was casted into a steel mould which was annealed at 200  C for 3 h to avoid excess thermal. Immediately after the quench, the samples were annealed at 250  C below their glass transition temperature for 2 h. The samples were then allowed to cool at room temperature before polishing them and heat treated at 503  C during 4 h, 8 h, 12 h, 16 h and 20 h to nucleate and grow the silver NPs. Glass compositions and their corresponding labels are shown in Table 1. The X-ray diffraction patterns of the samples prepared and heattreated were recorded in a X-ray diffractometer X’PERT Pro PAN Analytical diffractometer with CuKa radiation of wavelength (1.5418 Å), the diffractometer setting in the 2q range from 3 to 70 by changing the 2q with a step size of 0.02 . For transmission electron microscopic (TEM) studies, a few mg of powder samples were mixed with acetone and isolated by ultrasonic dispersion for about 20 min. Then, one drop of solution was spattered on the grid copper and dried before analysis. The TEM micrographs of the samples were taken in a transmission electron microscope equipped with high-resolution facility (TEM, Tecnai G20) with an acceleration voltage of 200 kV. The glass transition temperature Tg, crystallization onset temperature Tx and crystallization temperature Tc were determined using an Metler Toldo instrument differential scanning calorimetry (DSC) at a heating rate of 10  C/min from RT to 550  C. Raman scattering spectra have been recorded using HORIBA Scientific (lab RAM HR) spectrometer equipped with laser source (632 nm) and CCD detector. The absorption spectra of the glasses were recorded with a Perkin-Elmer UV-VIS Lambda 20 spectrometer between 200 and 1800 nm. The optical band gap energy and the Urbach energy were calculated from the experimental absorption spectra. Photoluminescence (PL) spectra and time-resolved PL spectra were measured in air at room temperature using a spectrophotometer (Hitachi U-4100). The IR PL spectra were measured with a N2-cooled CCD array from Princeton Instruments coupled to

557

grating. For IR lifetime, it was a PMT from Hamamatsu Photonics (H10330A-75) coupled to oscilloscope. The excitation source is an argon laser (lexc ¼ 488 nm) with a pump power of 2 mW over a spot of 2 mm diameter at the sample. 3. Results and discussion 3.1. Thermal and structural properties The X-ray diffraction (XRD) patterns for glass samples are shown in Fig. 1. The absence of any sharp peak and the observed broad hump between 20  C and 40  C suggests the absence of crystalline phase, which confirms the amorphous nature of the prepared glasses. Fig. 2a exhibits as an example, the TEM images of the NPZEAgref glass sample. Analyzing the image, dispersed spherical Ag NPs were clearly observed, with particle size varying from 20 to 40 nm. Fig. 2b shows the EDX spectra of the NPZEAgref glass sample to confirm the presence of Ag NPs as observed in the TEM images, in addition to the phosphor, zinc, oxygen and erbium. DSC curves of all samples recorded in the range of 25  Ce550  C and at a heating rate of 10  C/min are shown in Fig. 3. The glass transition temperature (Tg), crystallization onset (Tx), crystallization temperature (Tc), and the difference DT ¼ Tx  Tg are listed in Table 2. Thermal stability of glasses is a measure of disorder of glassy state. As shown in Table 2, the values of the thermal stability DT increase after the heat treatment, it can indicate that the heat treatment makes the glass more stable. One can notice that the thermograms exhibit 2 crystallization temperature peaks, which indicates the nucleation and growth of Ag NPs in glasses matrix after the heat treatment [30]. Raman spectroscopy was used to obtain essential information

Fig. 1. X-ray diffraction patterns of glass samples.

Table 1 Glass compositions and their labels. Samples

Na2O (%)

P2O5 (%)

ZnO (%)

Er2O3 (%)

Ag2O (%)

HT > Tg (hours)

NPZEAgref NPZEAg4h NPZEAg8h NPZEAg12h NPZEAg16h NPZEAg20h

42 42 42 42 42 42

42 42 42 42 42 42

15 15 15 15 15 15

0.5 0.5 0.5 0.5 0.5 0.5

0.5 0.5 0.5 0.5 0.5 0.5

0 4 8 12 16 20

558

I. Soltani et al. / Journal of Alloys and Compounds 686 (2016) 556e563

Fig. 2. TEM image (a) and EDX spectrum (b) of NPZEAgref glass sample.

Table 2 Thermal parameters of all glass samples.

Fig. 3. DSC curves of Er3þ/Ag NPs co-doped phosphate glass samples.

concerning the arrangement of the structural units from the studied glasses. Raman spectra of the glass system after the heat treatment are shown in Fig. 4, in the region of 200e1350 cm1. The

Samples

Tg ( C)

Tx ( C)

Tc ( C)

DT ¼ Tx  Tg

NPZEAgref NPZEAg4h NPZEAg8h NPZEAg12h NPZEAg16h NPZEAg20h

287 301 291 296 295 295

394 430 418 439 425 416

434 498 463 499 488 493

107 129 127 143 130 121

spectra were normalized to the band at 1157 cm1. The band assignments for phosphate glasses have been discussed in the literature [31]. The weak overlapping bands near 340 and 521 cm1 ascribed to the internal vibrations of metaphosphate chains. The strongest band near 726 cm1 is assigned to (PeOeP) symmetric stretching vibration, the band at 1033 cm1 is assigned to the asymmetric stretching vibrations (PeOeP). The band at 1157 cm1 is assigned to the symmetric stretching vibration (PO2), and a band at 1265 cm1 for the (PO2) asymmetric vibration. By examining the results in Fig. 4, It is clearly noticed that the Raman intensity increases greatly with the annealing time. The maximum enhancement is observed at 725 cm1 after 12 h of heattreatment. The enhancement may be attributed to the SPR contribution which enlarges the local electric field [32]. In addition, it is clearly detected a red shift from all peaks. The shift is due to the

I. Soltani et al. / Journal of Alloys and Compounds 686 (2016) 556e563

559

respectively. The presence of silver nanoparticles can be proved because of the plasmon band shown in Fig. 5 in the range of 420e570 nm. The intensity of the SPR band increases with the annealing time caused by the nucleation of silver nanoparticles. As we can see, the glass annealed for 20 h shows a prominent plasmon band at 489 nm, the broad band has been observed also in AgPO3 crystals and assigned to AgþeAgþ dimers [37]. The development of the SPR band can be due to both an increase in the number of plasmonic particles and the nanoparticles size [38]. These results are in well conformity to those obtained with other glasses containing silver NPs [30,39,40]. Introduction of silver nanoparticles into the phosphate glass influences both the structural and the optical properties. The formation and aggregation of metallic silver are a quite complex process during the heat treatment, although, the mechanisms of silver NPs formation is, we recall that the Ag2CO3 compound added to the starting glass constituents easily decompose at the temperatures used to prepare the samples by the reaction: Fig. 4. Raman spectra of Er3þ/Ag NPs co-doped phosphate glass samples.

closeness of the excitation wavelength and SPR frequency [32]. It is also due to the growth and nucleation of silver nanoparticles occurring in the heat treated samples, the same behavior is presented in other previous work [33]. Thus, the shift can be attributed to a change in the in-chain PeOeP bond angle depending on the effect of the network modifier on phosphate glass structure [34e36]. 3.2. Spectroscopic properties 3.2.1. Absorption spectra The glass samples containing the same amount of Ag2CO3 and heat-treated at 305  C for different periods (4, 8, 12, 16 and 20 h), are prepared in order to observe the plasmon band (SPR) of silver NPs in this glasses matrix. The room temperature measured absorption spectra of the phosphate glasses doped with Er3þ containing silver NPs, in the visible-NIR range have been presented in Fig. 5. The spectra reveal absorption peaks due to the 4f-4f forced electric dipole transitions from the ground state 4I15/2 to the different excited states 4I13/2, 4I11/2, 4I9/2, 4F9/2, 4S3/2, 2H11/2, 4F7/2, 4F5/ 4 2 3þ ions. The absorption bands are located at 2, F3/2 and G9/2, of Er 1617, 1061, 869, 717, 611, 586, 554, 470, 448, and 424 nm,

Ag2CO3 / Ag2O þ CO2 On the other hand, Ag2O decomposes to metallic Ag0 and to O2 through Ag2O / 2Ag þ 1/2O2 [41]. In the regions where the starting ionic concentration is quite high Agþ can be reduced during its path close to a particle, being there after captured as Ag0 starting to aggregate into nanoparticles. This is a clear indication that different centers are co-existent, probably AgþeAgþ and AgþeAg0 dimers, trimers and other small aggregates, as already observed in silicate glasses [37,42]. Heat treatment process leads to the increase of refractive index due to the more creation of nonbridging oxygen [43,44].

3.2.2. Optical band gap The direct and indirect optical band gaps of the samples are defined and calculated according to Mott and Davis [45,46], where the curve of absorption coefficient (a(hn)) is plotted as a function of photon energy (hn) at high values of absorption near the UV region, using the following relation:



aðhnÞ ¼ A hn  Eopt

n 

(1)

Where a is the absorption coefficient, hy is the incident photon energy, A is a constant and Eopt is the optical band gap. Values of nare 1/2 and 2 for direct and indirect forbidden transitions, respectively. Fig. 6a shows the variation of (ahn)2 and (ahn)1/2 as a function of (hn) for the NPZEAg16h as an example. The value for Edir lies in range of 3.70e4.05 eV, meanwhile the value of Eind varies between 3.33 and 3.61 eV (Table 3). The optical band gap energy decreases after a certain time of heat treatment assigned to the structural changes in the glass [47e49]. The decrease in the band gap is affected by the increasing of disorder of materials [50]. In glass materials there exists a band tailing in the forbidden energy band gap. The measure of disorder in the material can be estimated from the extent of band tailing using the Urbach equation [51]:

aðnÞ ¼ a0 expðhn=EuÞ

Fig. 5. Absorption spectra of Er3þ/Ag NPs co-doped phosphate glass samples.

hn

(2)

Where a0 a constant, hn is the photon energy and Eu is the width of band tails of electron states. Experimentally, plots are drawn for ln(a) against photon energy, E ¼ hy, case of NPZEAg16h sample is taken as an example (Fig. 6b). Table 3 shows the Urbach energy (DE) of all the studied glasses. The heat treatment process causes further increment in Eu and the value lies between 0.42 and 0.98 eV. The higher value of Eu

560

I. Soltani et al. / Journal of Alloys and Compounds 686 (2016) 556e563

Fig. 7. (a)PL intensity relative to 4I13/2 / 4I15/2 transition (lex ¼ 488 nm), (b) The effect of annealing time on the PL intensity.

Fig. 6. (a) optical energy band gap for direct Edir and indirect Eind transitions,(b) Logarithm of the absorption coefficient, ln (a), against photon energy, hy (eV) of NPZEAg16h for example.

Table 3 The refractive indices (n), Urbach energy (EU, eV), direct (Edir, eV) and indirect (Eind, eV) band gaps Glass

n

Eu

EInd

Edir

NPZEAgref NPZEAg4h NPZEAg8h NPZEAg12h NPZEAg16h NPZEAg20h

1.583 1.593 1.623 1.591 1.599 1.605

0.89 0.85 0.70 0.89 0.52 0.42

3.57 3.33 3.59 3.61 3.59 3.56

3.70 3.80 4.05 3.77 3.85 3.89

indicates higher disorder in the glass matrix as the consequence of more extension of the localized states within the gap [47].

3.3. Luminescence properties Fig. 7a shows the measured 1.53 mm band fluorescence spectra corresponding to Er3þ/Ag NPs co-doped glass samples in the range of 1400e1700 nm. The excitation wavelength (488 nm from Ar laser) is in resonance with the SPR band of Ag NPs. Each of the glass samples exhibits a broad fluorescence emission ranging from 1450 to 1650 nm, originated from the 4I13/2 meta-stable level to the terminal 4I15/2 one, which is ideal as gain media for the 1.53 mm

band broad amplifiers. Fig. 7b presents the variation of the fluorescence intensity of the 4 I13/2 / 4I15/2 transition as function of the annealing time. It is clearly noticed that the fluorescence intensity increases greatly with the heat-treating temperature. The maximum enhancement is observed after 12 h of heat-treatment. The main reasons for improvement for the 1.53 mm band fluorescence is attributed to the enhanced local electric field induced by the SPR of metallic NPs as a result of the aggregation and growth of silver NPs during the heattreating process [14,52,53]. Meanwhile, the Ostwald ripening and NPs migration followed by coalescence lead to the increase in size of silver NPs and enhance the local electric field around the Er3þ sites. The strong local electric field can increase the transition probability of Er3þ from the excited state to the ground state, resulting in an intensified fluorescence emission [54,55]. The Er3þ ions trapped at the inter-particle junctions of anisotropic hexagonal metal particles are expected to experience greater electric field than that produced by spherical nanoparticles [56]. Another proposed reason for the enhancement in fluorescence intensity could be the energy transfer from silver NPs to Er3þ ions [52,57e59]. It is clearly noticed that the quenching of the PL intensity for annealing time above 12 h is due to a process of energy transfer from the Er3þ ions to the surface of clustered silvers (reabsorption by NPs) [53]. The quenching phenomenon occurs because of short-distance between Er3þ and NPs, the dipole-dipole interaction between the ion and the NPs becomes large [20]. Thus, the PL quenching takes place with the precipitation of the NPs in the matrix, in accord with an excitation energy transfer channel becoming more effective with the increase in Ag NPs produced by heat treatment. A significant contribution of the present work is related to the optimization of the nucleation process that enables a larger enhancement of the emission spectra of Er3þ/Ag NPs co-doped phosphate glasses [8]. Fig. 8 shows the decay curves recorded for the 4I13/2 / 4I15/2 transition in the Er3þ doped phosphate glasses, the decay curves do not follow single-exponential decay, which verified the presence of energy transfer from Ag NPs to Er3þ ions. The measured lifetime corresponds to the mean lifetime, given by:

I. Soltani et al. / Journal of Alloys and Compounds 686 (2016) 556e563

561

Fig. 9. Absorption and emission cross-sections of 4I13/2 / 4I15/2 transition for the NPZEAgref glass.

Fig. 8. Room temperature PL decay curves of Er3þ: 4I13/2 level.

Table 4 Measured lifetimes, radiatives lifetimes and quantum efficiencies of Er3þ in the 4I13/2 / 4I15/2 transition. Samples

NPZEAgref

NPZEAg4h

NPZEAg8h

NPZEAg12h

NPZEAg16h

NPZEAg20h

tmes (ms) trad (ms) h (%)

2.23 3.38 66

4.06 4.01 101

4.19 4.09 104

4.87 4.23 115

3.88 4.72 82

2.94 3.96 74

tmes

experimental lifetimetmes, is given by:

 Z ¼ 1 I0 IðtÞdt

(3)

Where I(t) and I0 are the PL intensities at time t and at t ¼ 0, respectively. The resulting fluorescence decay times are collected in Table 4. As we can see, the fluorescence lifetime first increases greatly from 2.23 ms to 4.87 ms and then drops slightly to 2.94 ms when the annealing time increases from 0 to 20 h. The increase of fluorescence lifetime is due to the enhanced local electric field effect and energy transfer from the silver NPs to Er3þ levels. On the other hand, reverse energy transfer from Er3þ to silver NPs as the local electric field may reach saturation as a consequence of the increase in particle size of Ag NPs with the annealing time resulting in PL quenching should cause a lifetime decrease [52,61,62]. The quantum efficiency (h) calculated from the radiative lifetime trad, obtained from Judd- Ofelt theory, and the measured

hð%Þ ¼ tmeas =trad

(4)

Table 4 compares the values of quantum efficiency of samples glass heat treated. As we can see, the results show a significant increase of the quantum efficiency of the 4I13/2 level after the increase of heat treatment time and the more quantum efficiency is 115% in NPZEAg12h sample. It indicates that this glassy system could be considered as a good candidate for the realization of a laser emission at 1.53 mm. 3.4. Emission cross-section at 1.53 mm The determination of sem is a fundamental parameter for the characterization of an optically active material. According to the measured absorption cross section, the emission cross section was calculated using the theory developed by Mc-Cumber [20,58] as applied to rare-earth ions by Miniscalco and Quimby [60]. The

Table 5 Comparison of FWHM, effective bandwidths (Dleff), stimulated emission cross-sections (se) and se x FWHM for the 4I13/2 / 4I15/2 transition. Glass samples

FWHM (nm)

Dleff (nm)

sa (1021 cm2)

se (1021 cm2)

se  FWHM (1021 nm cm2)

NPZEAgref NPZEAg4h NPZEAg8h NPZEAg12h NPZEAg16h NPZEAg20h Tellurite [14] Bismuth [62] Phosphate [63] Aluminosilicate [64]

42.12 49.78 56.63 58.17 50.42 43.05 59.02 78 43.84 43

52.87 59.35 58.59 69.58 56.17 52.95 70.68 84 65.41 e

6.59 6.97 7.49 7.95 6.82 5.64 7.69 4.60 e e

7.14 8.20 8.93 9.05 8.84 7.36 9.80 e 7.70 5.7

300.73 408.19 505.70 526.43 445.71 316.84 578.39 280.60 337.50 245.10

562

I. Soltani et al. / Journal of Alloys and Compounds 686 (2016) 556e563

stimulated emission cross section, sem, and the measured sabs are related by:

se ðyÞ ¼ sa ðyÞexpðε  hnÞ=kT

(5)

Where h is the Planck constant, k the Boltzmann constant, ʋ is the photon frequency, and ε the net free energy required to excite one Er3þ ion from the 4I15/2 level to 4I13/2 level at temperature T. The effective bandwidth is an important parameter for the Er3þ doped fiber amplifier (EDFA) used in the wavelength division multiplexing (WDM) network system of optical communication. The definition of the effective bandwidth according to Weber et al. [59] is:

Dleff ¼

Z

IðlÞdl=Imax

(6)

Where I(l) is the emission intensity at the wavelength l, and Imax is the intensity at the peak wavelength l. The emission cross section se and absorption cross-section sa of phosphate glasses are shown in Fig. 9. The results obtained for the glasses have been reported in Table 5. As we can see, the emission cross-section intensity increases with the heat treatment. The high value for the emission cross-section relative to NPZEAg12h after 12 h of heat-treatment, while further annealing resulted in decrease of the emission cross-section intensity. We find that the emission cross section se relative to NPZEAg12h is larger than that of phosphate [63] and aluminosilicate glasses [64]. The emission cross-section se enhance the luminescence intensity of 4I13/ 4 2 / I15/2 transition in phosphate glasses and confirms the energy transfer from Ag NPs to Er3þ ion [14]. Also it is observed that the Dleff increases from 52.87 to 69.58 nm and the FWHM increases from 42.12 to 58.17 nm after 12 h of heat-treatment. The broadening of the luminescence band in these glasses is mainly due to the interaction between trivalent erbium ions and silver NPs, These high values indicate that the NPZEAg12h sample glass is a favorable material for optical amplifiers.

4. Conclusion Phosphate glasses doped trivalent erbium containing silver NPs are prepared using melt quenching method. The influence of the heating time on the structural and luminescence properties of the glasses is investigated. The amorphous nature of the prepared glasses is confirmed using the X-ray diffraction method and the TEM image reveals the formation of Ag NPs with homogenous distribution on glass host. The Ag NPs have a spherical shape with average diameter in the range of 20e40 nm. Raman spectroscopy confirmed a slightly red shift after the heat treatment. A broad absorption band was observed due to the surface plasmon resonance (SPR) of Ag NPs. The optical band gap energy decreases after certain time of heat treatment assigned to the structural changes in the glass. Furthermore, the presence of silver NPs has a remarkable effect on the enhancement of both the fluorescence intensity and the fluorescence lifetime, relative to the 4I13/2 / 4I15/2 transition. The enhancement is attributed to local field enhancement induced by SPR of Ag NPs, while the energy transfer from Er3þ ions to Ag NPs and/or the conglomerates of silver NPs are responsible factors for quenching of the emission intensity and fluorescence lifetime of Er3þ: 4I13/2 level. High quantum efficiency, up to 115%, is reported after heat treatment. The results show that the phosphate glasses would be promising host materials for 1.53 mm broad band amplification.

Acknowledgements This work is supported by the Tunisian Ministry of Higher Education and Scientific Research. We thank Professor Ben Salem Mohamed, Laboratory of Physics of Materials-Structures and Properties, Department of Physics, Faculty of Sciences of Bizerte, University of Carthage, for his kind assistance with measuring the TEM of the samples.

References [1] I.O. Mazali, L.C. Barbosa, O.L. Alves, J. Mater. Sci. 39 (2004) 1987. [2] Z. Teixeira, O.L. Alves, I.O. Mazali, J. Amer. Ceram. Soc. 90 (2007) 256. [3] L. Sirleto, M.G. Donato, G. Messina, S. Santangelo, A.A. Lipovskii, D.K. Tagantsev, S. Pelli, G.C. Righini, Appl. Phys. Lett. 94 (2009) 031105. [4] L.P. Naranjo, C.B. de Araujo, O.L. Malta, P.A.S. Cruz, L.R.P. Kassab, Appl. Phys. Lett. 87 (2005) 241914. [5] R. de Almeida, D.M. da Silva, L.R.P. Kassab, C.B. de Araújo, Opt. Commun. 281 (2008) 108. [6] L.P.R. Kassab, L. Ferreira Freitas, K. Ozga, M.G. Brik, A. Wojciechowski, Opt. Laser. Technol. 42 (2010) 1340. [7] A.P. Carmo, J.V. Bell, V. Anjos, R. de Almeida, D.M. da Silva, L.R.P. Kassab, J. Phys. D Appl. Phys. 42 (2009) 155404. [8] F.A. Bomfima, J.R. Martinelli, L.R.P. Kassab, T.A.A. Assumpç~ ao, C.B. de Araújo, J. Non Cryst. Solids 356 (2010) 2598. [9] L.R.P. Kassab, F.A. Bonfim Junior, J.R. Martinelli, N.U. Wetter, J. Jakutis Neto, C.B. de Araújo, Appl. Phys. B 94 (2009) 239. ~o, C.B. de Araújo, [10] L.R.P. Kassab, R. de Almeid, D.M. da Silva, T.A.A. de Assumpça J. Appl. Phys. 105 (2009) 103505. [11] T.A.A. de Assumpç~ ao, D.M. da Silva, L.R.P. Kassab, C.B. de Araújo, J. Appl. Phys. 106 (2009) 063522. [12] Raja J. Amjad, M.R. Sahara, S.K. Ghoshal, M.R. Dousti, S. Riaz, B.A. Tahir, J. Lumin. 132 (2012) 2714. [13] P.N. Prasad, Nanophotonics, WileyInterscience, New York (, 2004. [14] M.R. Dousti, M. Amjad, A. Hussain, R.J. Amjad, S.F. Shaukat, Chalcogenide Lett. 12 (2015) 123. [15] T. Som, B. Karmakar, J. Appl. Phys. 105 (2009) 013102. [16] A.L. Falk, F.H.L. Koppens, C.L. Yu, K. Kang, N.D.L. Snapp, A.V. Akimov, M.H. Jo, M.D. Lukin, H. Park, Nat. Phys. 5 (2009) 475. [17] S. Kühn, U. H akanson, L. Rogobete, V. Sandoghdar, Phys. Rev. Lett. 97 (2006) 017402. [18] H. Fares, H. Elhouichet, B. Gelloz, M. Ferid, J.Appl. Phys. 116 (2014) 123504. [19] S. Derom, A. Berthelot, A. Pillonnet, O. Benamara, A.M. Jurdyc, C. Girar, G. Colas des Francs, Phys. Opt. (2013) 1312. [20] M.R. Sahar, E.S. Sazali, Raja J. Amjad, Siminar Nasional Fisika, Jakarta, 9 Juni 2012, pp. 14e19. [21] A. Pillonnet, A. Berthelot, A. Pereira, O. Benamara, S. Derom, G. Colas des Francs, A.-M. Jurdyc, Appl. Phys. Lett. 100 (2012) 153115. [22] S.A. Maier, H.A. Atwater, J. Appl. Phys. 98 (2005) 011101. [23] W.L. Barnes, A. Dereux, T.W. Ebbesen, Nature 424 (2003) 824. [24] K.L. Kelly, E. Coronado, L.L. Zhao, G.C. Schatz, J. Phys. Chem. B 107 (2003) 668. [25] M. Reza Doustia, Raja J. Amjad, JNS 3 (2013) 435. [26] T. Som, B. Karmakar, J. Quant. Spectrosc. Radiat. Transf. 112 (2011) 2469. [27] D.R. Lide, CRC Handbook of Chemistry and Physics, 75th ed., CRC Press, Boca Raton, 1975. [28] V.A.G. Rivera, S.P.A. Osorio, Y. Ledemi, D. Manzani, Y. Messaddeq, L.A.O. Nunes, E. Marega Jr., Opt. Express 18 (2010) 25321. [29] Y. Qi, Y. Zhou, L. Wu, F. Yang, S. Peng, S. Zheng, D. Yin, J. Lumin. 153 (2014) 401. [30] M. Reza Dousti, M.R. Sahar, S.K. Ghoshal, Raja J. Amjad, R. Arifin, J. Non Cryst. Solids 358 (2012) 2939. [31] J. Wong, C.A. Angell, Glass Structure by Spectroscopy, Marcel Dekker, New York, 1976, p. 459. [32] M. Reza Dousti, M.R. Sahar, Raja J. Amjad, S.K. Ghoshal, Asmahani Awang, J. Lumin. 143 (2013) 368. [33] K. Maed, A. Yasumori, J. Non Cryst. Solids 427 (2015) 152. [34] J. Koo, B.-S. Bae, H.-K. Na, J. Non Cryst. Solids 212 (1997) 173. [35] Siti Amlah M. Azmi, M.R. Sahar, S.K. Ghoshal, R. Arifin, J. Non Cryst. Solids 411 (2015) 53. [36] M.R. Dousti, R.J. Amjad, M.R. Sahar, Z.M. Zabidi, A.N. Alias, A.S.S. de Camargo, J. Non Cryst. Solids 429 (2015) 70. [37] H. Portales, M. Mattarelli, M. Montagna, A. Chiasera, M. Ferrari, A. Martucci, P. Mazzoldi, S. Pelli, G.C. Righini, J. Non Cryst. Solids (2005) 1738e1742. nez, M. Sendova, H. Liu, F.E. Fern [38] J.A. Jime andez, Plasmonics 6 (2011) 399. [39] T. Som, B. Karmakar, Solid State Sci. 13 (2011) 887. [40] Sh Murai, R. Hattori, T. Matoba, K. Fujita, K. Tanaka, J. Non Cryst. Solids 357 (2011) 2259. , C.B. Garcia deAraújo, J. Alloys [41] M.E. Camilo, E.O. Silva, L.R.P. Kassab, A.M. Jose Compd. 644 (2015) 155. [42] M. Mattarelli, M. Montagna, E. Moser, K.C. Vishnubhatla, C. Armellini, A. Chiasera, M. Ferrari, G. Speranza, M. Brenci, G. Nunzi Conti, G.C. Righini,

I. Soltani et al. / Journal of Alloys and Compounds 686 (2016) 556e563 J. Non Cryst. Solids 353 (2007) 498. [43] R.J. Amjad, M.R. Sahar, S.K. Ghoshal, M.R. Dousti, S. Riaz, B.A. Tahir, Chin. Phys. Lett. 29 (2012) 087304. [44] R. El-Mallawany, M. DirarAbdalla, I. Abbas Ahmed, Mater. Chem. Phys. 109 (2008) 291. [45] E.A. Davis, N.F. Mott, Philos. Mag. 22 (1970) 903. [46] Y. Wang, S. Dai, F. Chen, T. Xu, Q. Nie, Mater.Chem. Phys. 113 (2009) 410. [47] M.K. Halimah, W.M. Daud, H.A.A. Sidek, A.W. Zaidan, A.S. Zainal, Mater. Sci. Pol. 28 (2010) 173. [48] M. Reza Dousti, M.R. Sahar, S.K. Ghoshal, Raja J. Amjad, A.R. Samavati, Adv. J. Mol. Struct. 1035 (2013) 6. [49] M.R. Dousti, M.R. Sahar, R.J. Amjad, S.K. Ghoshal, A. Khorramnazari, A. DordizadehBasirabad, A. Samavati, Eur. Phys. J. D. 66 (2012) 237. [50] S.K. AsmahaniAwang, M.R. Ghoshal, M. Sahar, R. Dousti, F. Nawaz, Adv. Mater. Res. 895 (2014) 254. [51] F. Urbach, Phys. Rev. 92 (1953) 1324. [52] H. Fares, H. Elhouichet, B. Gelloz, M. Ferid, J. Appl. Phys. 117 (2015) 193102.

563

[53] Wu Libo, Bo Huang, F. Yang, Qi Yawei, P. Shengxi, Z. Yaxun, Li Jun, Mater. Lett. 152 (2015) 220. [54] G.H. Silva, D.P.A. Holgado, V. Anjos, et al., Opt. Mater. 37 (2014) 281. [55] E.S. Sazali, M.R. Sahar, S.K. Ghoshal, et al., J. Non Cryst. Solids 410 (2015) 174. [56] K.H. Su, Q.H. Wei, X. Zhang, J.J. Mock, D.R. Smith, S. Schultz, Nano Lett. 3 (2003) 1087. [57] G. Speranza, L. Minati, A. Chiasera, M. Ferrari, G.C. Righini, G. Ischia, J. Phys. Chem. C 113 (2009) 4445. [58] V.A.G. Rivera, Y. Ledemi, S.P.A. Osorio, D. Manzani, F.A. Ferri, S.J.L. Ribeiro, L.A.O. Nunes, E. Marega Jr., J. Non Cryst. Solids 378 (2013) 126. [59] Y. Qi, Y. Zhou, L. Wu, F. Yang, S. Peng, S. Zheng, D. Yin, J. Lumin. 153 (2014) 401. [60] O. Malta, M. Couto dos Santos, Chem. Phys. Lett. 174 (1990) 13. [61] T. Hayakawa, S. Selvan, M. Nogami, Appl. Phys. Lett. 74 (1999) 1513. [62] J. Qi, T. Xu, Y. Wu, X. Shen, S. Dai, Y. Xu, J. Opt. Mater. 35 (2013) 2502. [63] A. Langar, C. Bouzidi, H. Elhouichet, M. Ferid, J. Lumin. 148 (2014) 249. [64] A. Polman, J. Appl. Phys. 82 (1997) 1.