Fluorescence characteristics of Nd3+ doped multicomponent fluoro-phosphate glasses for potential solid-state laser applications

Author’s Accepted Manuscript Fluorescence characteristics of Nd3+ doped multicomponent fluoro-phosphate glasses for potential solid-state laser applications Y.C. Ratnakaram, S. Babu, L. Krishna Bharat, C. Nayak www.elsevier.com/locate/jlumin

PII: DOI: Reference:

S0022-2313(15)30630-X http://dx.doi.org/10.1016/j.jlumin.2016.02.009 LUMIN13842

To appear in: Journal of Luminescence Received date: 23 October 2015 Revised date: 11 January 2016 Accepted date: 1 February 2016 Cite this article as: Y.C. Ratnakaram, S. Babu, L. Krishna Bharat and C. Nayak, Fluorescence characteristics of Nd3+ doped multicomponent fluoro-phosphate glasses for potential solid-state laser applications, Journal of Luminescence, http://dx.doi.org/10.1016/j.jlumin.2016.02.009 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Fluorescence characteristics of Nd3+ doped multicomponent fluoro-phosphate glasses for potential solid-state laser applications Y.C. Ratnakaram*1, S. Babu1, L. Krishna Bharat2, C. Nayak3, 1. Department of Physics, Sri Venkateswara University, Tirupati-517 502, A.P. INDIA. 2. Department of Electronics and Radio Engineering, Kyung Hee University, Yongin-si, Gyeonggi-do 446-701, Republic of Korea. 3. Atomic and Molecular Physics Division, Bhabha Atomic Research Centre, Mumbai 400 085, India Abstract The multicomponent fluoro-phosphate glasses of the type, 49.5P2O5 -10AlF3-10BaF210SrF2-10PbO-10M (M= Li2O, Na2O, K2O, ZnO and Bi2O3) doped with 0.5 mol % neodymium were prepared by melt quenching technique. Their structures were characterized by the X-ray diffraction with SEM analysis, Fourier transform infrared (FTIR), Raman spectroscopy and

31

P and

27

Al magic angle spinning (MAS) nuclear magnetic resonance

(NMR) techniques. XPS spectra were studied to know the bridging and non-bridging oxygen groups. X-ray absorption near edge spectroscopy (XANES) was used to study the electronic structure of neodymium in the host glass matrices. The Judd–Ofelt parameters (J-O) (Ω2, Ω4 and Ω6) were evaluated from the intensities of the absorption bands through optical absorption spectra. Further, J-O parameters have been used to calculate various radiative properties like probabilities of radiative transitions, radiative lifetimes and branching ratios for different fluoro-phosphate glasses. The luminescence kinetics from the excited neodymium levels have been studied upon selective excitation through photoluminescence measurements. Neodymium ion emits two near infrared laser emissions: 4F3/2→4I11/2 at 1.06 μm and another one 4F3/2→4I13/2 at 1.32 μm. The major intensity is observed for 1.06 μm for the prepared fluoro-phosphate glasses. The lifetimes of these levels have been experimentally determined through decay profile studies. The above results suggest that the prepared lithium fluoro-phosphate glass system could be a suitable candidate for using it as 1.06 μm laser source in the near infrared region of spectrum.

1

Graphical Abstract

Keywords: Fluoro-phosphate glass; Judd-Ofelt theory; Neodymium (Nd3+); NIR emission; FTIR. * Corresponding author. Tel.: +91 9440751615 Email address: [email protected]

1. Introduction One may observe a technological interest in Ln3+-doped materials due to their potential applications as a laser active media in a variety of hosts such as glasses, transparent ceramics and crystals which exhibit a wealth of sharp fluorescent transitions in the VIS and NIR spectral regions [1]. Glasses are being used as host materials for the solid state lasers and in 2

many cases are more suitable than crystals [2]. From this point of view, researchers investigated a variety of inorganic oxide glasses like borate, silicate, phosphate, tellurite, fluoro-phosphate (FP) and germanate glass hosts. The phonon energy in FP glasses is a smaller than for phosphate glasses, which leads to low non-radiative relaxation in fluorophosphates glasses [3]. In the present work, fluorophosphate glass is chosen as the host material in which P2O5 act as a glass former. Phosphates have many merits such as low dispersion, high rare-earth ions concentration and large emission cross sections [4, 5]. But the phonon energy of phosphate is larger, which is disadvantageous to the luminescent efficiency [6]. This problem somewhat overcome through the addition of fluoride based oxides in the glass host. That’s why, fluorides such as AlF3, BaF2 and SrF2 are often added to modify their properties. The presence of PbO in the FP glass may also result in the formation of P–O–Pb bonds and lead to an improvement of the chemical durability of glass systems [7, 8]. It is essential to understand the structure-composition relationship of glass material in order to develop better material properties. The glass properties are not only related to polymerization state of phosphate glasses and also on the crosslinking cations which depend on the coordination state. Raman, MAS-NMR and XPS are very sensitive tools to study the coordination state of phosphate (P) and aluminum (Al). Of many rare earth active ions, Nd3+ is found to be the most efficient lasing ion due to its multiple absorption levels spreading from UV to NIR through visible spectral range for efficient pumping. Nd3+ doped fibers are used in the field of telecommunications, for tunable and ultrafast laser medium [9]. The laser applications based on various host materials doped with Nd3+ have been used at 1060 and 1350 nm [10]. The energy levels and the spectral broadening of absorption and emission bands depend on the type of host material [11]. In the present study, an attempt has been made to study the optical and lasing properties of Nd3+ doped fluoro-phosphate glasses by changing the glass composition. Recently, Tao Wei et al [12] investigated the structural, absorption, energy transfer, luminescent properties and near infrared applications of neodymium doped germanate glasses. Farouk et al [13] studied optical properties of lead bismuth borate glasses doped with neodymium oxide. Tianfeng Xue et al [14] examined thermal and spectroscopic properties of Nd3+-doped novel fluorogallate glass. Vázquez et al [15] reported spectroscopic analysis of novel Nd3+ activated barium borate glass for broadband laser ampification. In the present work, vibrational modes and structural modifications in the phosphate and aluminate units are investigated using FTIR, Raman and solid state NMR studies respectively. XANES spectra are used to collect information on the electronic structure of 3

doping ion in the FP glasses. We report the spectroscopic properties of Nd3+ doped different FP glasses by calculating the Judd-Ofelt intensity parameters (Ω

2, 4, 6),

radiative transition

probabilities (AR), radiative lifetimes (τR) and absorption cross-sections (Σ) from absorption spectra. The laser characteristic parameters such as branching ratios (β) and stimulated emission cross-sections (σp) are obtained from the emission spectra. Decay times (τexp) of the excited state, 4F3/2 of Nd3+ in FP glasses have been compared with the values of other Nd3+ doped glasses. From these studies, we expect that the lithium fluoro-phosphate glass matrix act as good laser active media.

2. Experimental In this study, we prepared and investigated five series of Nd3+ doped fluoro-phosphate glasses using chemicals such as phosphorous pentoxide (P2O5), aluminum fluoride (AlF3), barium fluoride

(BaF2), strontium fluoride (SrF2), lead oxide (PbO), lithium carbonate

(Li2CO3), sodium carbonate (Na2CO3), potassium carbonate (K2CO3), zinc oxide (ZnO), bismuth oxide (Bi2O3) and neodymium oxide (Nd2O3). The method of preparation is melt quenching technique. All the chemicals are analar grade with 99.9% purity. Final glass compositions of the prepared glass samples are marked as follows 1 Li:

49.5P2O5 -10AlF3-10BaF2-10SrF2-10PbO-10Li2O-0.5Nd2O3

2 Na:

49.5P2O5 -10AlF3-10BaF2-10SrF2-10PbO-10Na2O-0.5Nd2O3

3 K:

49.5P2O5 -10AlF3-10BaF2-10SrF2-10PbO-10K2O-0.5Nd2O3

4 Zn:

49.5P2O5 -10AlF3-10BaF2-10SrF2-10PbO-10ZnO-0.5Nd2O3

5 Bi:

49.5P2O5 -10AlF3-10BaF2-10SrF2-10PbO-10Bi2O3-0.5Nd2O3

The raw materials were thoroughly mixed in an agate mortar and melted in an electronic furnace at the temperature 1100-11500C for 1h. Then, the mixture was quenched by pouring it on a preheated brass plate and pressed by another brass plate. The obtained glasses are then cooled to room temperature. The refractive index (n) measurements were performed using an Abbe refractometer at sodium wavelength (589 nm). The densities (d) of the glasses were calculated using the Archimedes’s principle. The amorphous nature of the prepared Nd3+ doped fluoro-phosphate glasses are confirmed through the X-ray diffraction (XRD) studies using SIEFERT diffractometer employing Cu Kα radiation at 40 KV applied voltage and 30 mA anode current with a Si detector. The range of diffractometer is from 5O to 70O with step size of 0.02O. SEM images 4

were recorded using Caral Zeiss EVO-MA15 scanning electron microscope. FTIR spectra were recorded using a model of Perkin Elmer spectrum One FT-IR spectrophotometer. Raman spectra were obtained in back scattering geometry with 514 nm line of Ar+ laser as an excitation source at power of 10 mW. The spectral slitwidth was 1 mm and accumulation number was 50 seconds.

Solid state NMR spectra were recorded to further study the

structural evolution of prepared glasses using JEOL DELTA 2 NMR at 9.4 T with a 4 mm probe. The acquisition time was 18 ms and pulse width was 2.9 μs. The spinning speed was 10 kHz. X-ray photoelectron spectroscopy (XPS) spectra were recorded using a Thermo Scientific K-Alpha. XANES measurements of glass samples at L3 edge were carried out in fluorescence mode at the Scanning EXAFS Beamline (BL-9) at the INDUS-2 Synchrotron Source (2.5 GeV, 100 mA) at the Raja Ramanna Centre for Advanced Technology (RRCAT), Indore, India [16]. The beamline uses a double crystal monochromator (DCM) which works in the photon energy range of 4-25 KeV with a resolution of 104 at 10 KeV. A 1.5 m horizontal pre-mirror with meridonial cylindrical curvature is used prior to the DCM for collimation of the beam and higher harmonic rejection. The second crystal of the DCM is a saggital cylinder with radius of curvature in the range 1.28-12.91 meters which provides horizontal focusing to the beam. For measurements in the fluorescence mode, the sample is placed at 45o to the incident X-ray beam and the fluorescence signal ( I f ) is detected using a Si drift detector placed at 90o to the incident X-ray beam. An ionization chamber detector is used prior to the sample to measure the incident X ray flux (I0) and the absorbance of the sample (μ=If/I0) is obtained as a function of energy by scanning the monochromator over the specified energy range. The optical absorption spectral measurements were collected using a Varian Cary 5000 spectrophotometer in UV-VIS region with 0.01 nm steps and in NIR region with 0.04 nm steps, with the resolution of <0.05 and <0.2 in the UV-VIS and NIR regions respectively. The photoluminescence spectra and decay curves of Nd3+ doped glass samples were recorded using Jobin-Yvon Fluorolog-3 spectrofluorimeter (Horiba FL322iHR320). All these measurements were conducted at room temperature.

3. Results and discussion 3.1 XRD and SEM analysis In order to confirm the amorphous state of prepared fluoro-phosphate (FP) glasses, the prepared samples have been subjected to XRD (X-ray Diffraction) analysis. Based on this analysis, it is found that all the glass samples are amorphous in nature because of absence of sharp peaks and reflections from the XRD patterns. The XRD pattern of Nd3+doped lithium FP glass matrix is shown in Fig.1a. SEM image of Nd3+ doped lithium FP glass is shown in 5

Fig.1b. No bubbles or crystals or clusters were observed in the bulk glass, which is very important property when it is to be used in optical applications.

3.2 Fourier Transform Infra Red Spectra (FTIR) Fourier transform infrared analysis is used to obtain information regarding the arrangement of structural units for the studied glass system. FTIR transmission spectra of neodymium doped different FP glasses in the frequency range 400–4000 cm-1 are recorded and are shown in Fig. 2. The band around 525 cm-1 is due to bending vibrations of O-P-O linkages in the Q3 units. The band around 750 cm-1 is ascribed to the symmetric stretching of P-O-P in Q1 units [17]. The band around 900 cm-1 is due to asymmetric stretching of P-O-P linked with linear meta-phosphate chain and P-F groups in Q2 units [18]. This band also indicates that phosphate network is built from short chains. The band around 1085 cm-1 is due to the symmetric stretching of P-O-P and O-P-F in Q1 units. The broad features at 1275 cm-1 band is asymmetric stretching owing to P=O bonds [19]. The three small bands nearly at 1613 cm-1, 2900 cm-1 and 3428 cm-1 are ascribed to bending vibrations of O-H, stretching vibrations of O-H and hydroxyl groups respectively[20].

3.3 Raman spectra Raman spectra are recorded in the region 200-1500 cm-1 for neodymium doped different FP glasses and are shown in Fig. 3. The spectra show characteristic peaks related to phosphate groups. Raman spectra have been employed to look into the structure of FP glass. The band at ~257 cm-1 (a) ascribed to symmetric stretching vibrations of Bi-O bonds in BiO3 and BiO6 units [21]. This band is observed only in the bismuth FP glass matrix. The band at 330-380 cm-1 (b) is caused to F-PO3 scissoring mode and originates from the presence of corresponding modifier oxide [22]. The band at 525-550 cm-1(c) is due to the deformation mode of P-O and P-F vibrations. The band at 700-725 cm-1(d) is due to symmetric stretching mode of P-O-P bridge oxygen. The band at 930-950 cm-1 (e) is an asymmetric stretching mode of P-O-P and P-F bonds [23]. The metaphosphate group is found to be in the 1090-1110 cm-1 (f) region. The band at 1130-1160 cm-1 (g) is with symmetric stretching of O-P-O nonbridging oxygen atoms of a phosphate group. The existence of this band indicates interaction of different modifiers with a phosphate group through non bridging oxygen. The band at 1240-1250 cm-1 (h) is ascribed to symmetric stretching of P=O terminal oxygen group [24]. The peak at about 1130 cm-1 is due to low phonon energy of bismuth FP glass matrix compared to the other FP glasses. The structural depolymorization is observed from lithium FP glass matrix to bismuth FP glass matrix through the peak located at 1160 cm-1 in the 6

lithium FP glass matrix and shifted to lower wavenumber in the bismuth FP glass matrix (1130 cm-1). 3.4 Magic angle spin nuclear magnetic resonance (MAS NMR) analysis The

31

P and

27

Al MAS NMR data is recorded for all the prepared Nd3+ doped FP

glasses and are shown in Figs. 4a and 4b, respectively. The MAS NMR spectrum of lithium fluoro-phosphate glass is dominated by the major signal of Q2 units (phosphate tetrahedral with two bridge oxygens) with chemical shift of about –20 ppm and others were sidebands arise due to anisotropic shifts (Table 1). These units are related to metaphosphate units. The signal has a sort of single symmetric peak which indicates the existence of no other QO (phosphate tetrahedral with zero bridge oxygens), Q1 (phosphate tetrahedral with one bridge oxygens) and Q3 (phosphate tetrahedral with three bridge oxygens) structural units. The absence of these structural units in lithium fluoro-phosphate glass matrix is due to availability of sufficient quantity of oxygen bonds. The observed large linewidths indicate the formation of non-crystalline state of the prepared glasses. It is observed that Li, Na, K, Zn and Bi fluoro-phosphate glasses exhibit strong signal with chemical shifts of about -20, -31, -33, -40 and -44 ppm respectively. With the modification in the modifier oxide, chemical shifts become more negative. As more cross-connections are formed in the glass, the charge on the oxygen is moved away from the P–O bonds, running to the nuclei becoming more shielded, producing a change in chemical shift and leads to increased network connectivity. It can thus be distinctly stated that modifier ions are connected (P-O-M, M= Li, Na, K, Zn, Bi) exclusively to the phosphate group. The structural depolymorization is observed from lithium fluoro-phosphate glass matrix to bismuth fluoro-phosphate glass matrix through the appearance of other structural units (Q1, Q3). This observation is coherent with the Raman spectral analysis also [25-30]. From Fig. 4b, it is observed that Li, Na, K, Zn and Bi fluoro-phosphate glasses show intense resonance signals nearly at 2, -27, -21, -30 and -26 ppm respectively. Most of the peaks are assigned due to the octahedrally coordinated Al. Hence, all the prepared glasses tend to co-ordinate in the Al(6) structure except lithium FP glass matrix [31]. In the case of lithium FP glass matrix single resonance peak (2 ppm) is observed which is assigned to Al(5) coordination of Al3+ ions based on the

27

Al NMR spectra. However, other glass matrices,

consist of other additional peaks i.e. Na (0, 6 ppm); K (-3, 33 ppm); Zn (-1, 26 ppm) and Bi (0, 28 ppm). These additional peaks in the regions -1 to -3 ppm and 6 to 33 ppm are assumed to be Al(5) and Al(4) structural units [32, 33]. All these glass matrices have symmetric in

7

resonance peak, which suggests that there is no other dominated structural coordination units. Hence, they coordinate to other atoms in order to charge compensation.

3.5 X-ray photoelectron spectroscopy (XPS) The binding energy (B.E.) of the core level spectra is very informative in terms of the structural role of cations occupy in glass host. The deconvolution of the P 2p spectra into two peaks could arise from the spin–orbit splitting of the P 2p core level resulting in the distinguishable P 2P3/2 and P 2P1/2 core levels with the lower and higher binding energy peaks respectively. Such fitting for lithium fluoro-phosphate glass matrix is depicted in Fig. 5a. Oxygen atoms are more covalently bonded to phosphate units on both sides. This formation is called as bridging oxygens (BO) (133.5 eV) like P–O–P, M–O–M and P–O–M (M= Al, Ba, Sr, Pb, Li, Na, K Zn and Bi). Oxygen atoms that are more ionically bonded (double bonded) to phosphate units are referred to as non-bridging oxygen (NBO) atoms like P=O bonding (132.7 eV). Bridging and non-bridging oxygens have higher and lower binding energies, respectively. The B.E. values in the prepared FP glasses centred at 132.7 eV suggest that due to the O–P–O bonding of the PO42− mode corresponds to the metal-phosphate network [34]. The presence of non-briding oxygens indicate more than one valance states such as Sr2+ and Al3+ linked to phosphate units. The first component with B.E. 133.5 eV (P 2P1/2) is attributed to pentavalent tetra coordinated phosphorus units (pyrophosphate and orthophosphate) surrounded by different chemical environment. The second component observed in P 2p spectrum with B.E.132.7 eV (P 2P3/2) should be attributed to remaining metaphosphate units [35]. The deconvoluted spectrum of O 1s for lithium FP glass matrix is presented in Fig. 5b. The peak located at higher BE related to BO in P-O-P binding, while peak located at lower B.E related to NBO coming from P-O- and M-O-. From XPS spectrum, it is observed that in between bridging and non bridging oxygens, asymmetric bridging oxygens (ABO) are present due to formation of P-O-M and M-O-M bondings [36]. The BE energy of NBO O 1s spectra is lower than that of ABO bonding. The fraction of BO and NBO are very less. The dominated peak is asymmetric bridging oxygen group in all the FP glasses. The peak profile of core level spectra is independent of glass host composition. But the variation in intensity is observed.

3.6 X-ray absorption near edge spectroscopy (XANES) analysis Fig. 6 shows the XANES spectra of Nd doped glass samples at Nd L3 edge along with Nd2O3 standard. The white line positions indicate that chemical valence of the 8

absorption atoms. In the present work, all samples have an intense white line resonance, which can be attributed to the 2P3/2–5d electronic transition of neodymium atoms. From this XANES spectra, it is clear that Nd can be efficiently doped into all the prepared glass matrices as expected Nd3+ state i.e. standard Nd and Nd doped FP glass samples have stable trivalent oxidation state. The intensities of white line in the Nd doped glass samples are much larger than that in Nd2O3 standard. The variations in intensity in different samples suggest that variation of electron density around the Nd3+ ion. Bi/K FP glasses show the lowest/highest white line intensity. A significant variation of intensity between standard Nd and doped glasses is observed due to change in the local coordination environment. From Fig.6, it is observed that there is no shift in energy of resonance peaks and also there is no additional resonance peaks indicating the non-dependence of compositional variation on energies in these FP glasses [37].

3.7 Absorption spectra and Judd-Ofelt theory The absorption spectra of Nd3+ doped different doped fluoro-phosphate glasses are recorded in the 350-900 wavelength region and are depicted in Fig. 7. The spectra consist of inhomogeneously broadened bands assigned the transitions from the 4I9/2 ground state to the various excited states of Nd3+ ions. The observed absorption bands at 360, 429, 456, 476, 511, 525, 582, 620, 682, 745, 800 and 866 nm are corresponding to transitions from 4I9/2 to 2

I11/2+ 4D3/2 + 4D5/2, 2P1/2, 4G11/2, 2D3/2 + 2G9/2 + 2K13/2, 4G9/2, 4G7/2, 4G5/2 +2G7/2, 2H11/2, 4F9/2,

4

S3/2 + 4F7/2, 4F5/2+ 2H9/2 and 4F3/2 transitions, respectively [38]. The visible region absorption

bands can imply that the FP glass material is transparent and exhibits characteristic absorption bands of Nd3+ ions. There is no obvious change in the position of absorption bands for different FP glasses owing to shielding nature of 4f electrons by outermost orbital’s and only the intensities are affected by surrounding ligand environment due to composition variation. The optical absorption spectra are analyzed to find the experimental spectral intensities for the observed bands and then Judd- Ofelt intensity parameters ( Ω2, Ω4, Ω6 ) are obtained using the formulae given in Ref. [39, 40]. These parameters are important for the investigation of local structure and bonding in the vicinity of Nd3+ ions. The reliability between experimental (fexp) and calculated spectral intensities (fcal) can assist from root mean square (RMS) deviations which are low in the present work (Table 2). The theory behind the work has been taken from Ref. [41]. 9

The strong absorption bands at 582 and 745 nm, in Fig. 7 correspond to transitions from the ground state, 4I9/2 to the excited states, 4G5/2 +2G7/2 and 4S3/2 + 4F7/2 of Nd3+ ions, respectively. The intensities of certain f–f transitions are larger compared to other transitions and are characterized by the higher values of reduced matrix element ||U2||2. These transitions are known as ‘hypersensitive transitions’, i.e. sensitive to inhomogeneity of ligand environment and follows the selection rules, ΔL≤2, ΔJ≤2 and ΔS= 0 [42]. The most intense absorption band at 586 nm corresponding to 4I9/2→4G7/2+4G5/2 transition is a hypersensitive for Nd3+ ions which is in the visible region. If any changes made on surrounding ligand environment of neodymium ion, first this hypersensitive transition responses large variations in intensities and shapes of the transition, while the rest of the transitions shows very small variation. This transition reveals changes in the symmetry of the crystal field around the Nd3+ ions. Among the three intensity parameters, Ω2 is the most sensitive to the local structure and glass composition, which reflects the amount of covalent bonding, short range structural disorder effects and asymmetry of local environment near Nd3+ site while Ω4 and Ω6 are related to the bulk property and rigidity of the samples [13]. In the present work, the covalency parameter (Ω2) follow the trend Bi>Li>Na>Zn>K for different FP glasses. As shown in Table 3, Ω2 of Nd3+-doped bismuth FP glass is relatively large. Large value of Ω2 suggests less centro symmetric coordination environment around Nd3+ in bismuth FP glass matrix (8.76x10-20) while in the case of potassium FP glass matrix (4.48x10-20) centro symmetric coordination is more. The wide variations are observed in Ω2 parameter from one host to other hosts in oxide glassy material, it is due to the existence of non-uniform ligand fields caused by different Nd–O distances. A partly replacement of oxide ligands for fluoride ones should change the rare-earth coordination environment and crystal field strength. Even though, co-existence of oxygen and fluorine coordination shells in the FP glasses, neodymium prefers to connect to the oxygen ligand than fluorine ligand. This is evident from high Ω2 parameter than the other two parameters for all FP glasses except for potassium and zinc FP glasses. The higher value of Ω2 parameter in bismuth FP glass sample indicates higher covalency and the glass system is more asymmetric. The Ω2 parameter or covalent parameter is higher in bismuth fluoro-phosphate glass matrix compared with other reported glasses such as fluorogallate (6.42x10-20) [14], tellurite (4.71x10-20) [43] and phosphate (5.27x10-20) [44]. The covalency is low when compared with barium borate glass matrix (9.95x10-20) [15]. Ω4 parameter or rigidity of glass sample is higher and lower in zinc (4.60x10-20) and lithium (1.47x10-20) fluoro-phosphate glass matrices respectively. 10

3.8 Radiative properties By means of the obtained J-O (Ω

2, 4, 6)

intensity parameters, various radiative

parameters have been calculated to explore their luminescence behavior. Radiative properties such as total radiative transition probabilities (AT), radiative lifetimes (τR), branching ratios (R) and absorption cross-sections (Σ) for the excited states, 4G9/2, 4G7/2, 4G5/2, 2H11/2, 4F9/2, 4

F5/2 and 4F3/2 of Nd3+ doped different FP glass matrices are calculated and are presented in

Table 4. The radiative properties of Nd3+ ions depend on the property of network formers and network modifiers of the glass environment. The radiative relaxation of an exited state to all its lower levels depends upon radiative transition rates. Radiative transition rates depend upon Judd-Ofelt parameters and energy gap between initial stage and end point. The levels having the relatively large values of transition rates, branching ratios and energy gap to the next lower level may exhibit laser action. As can be seen from Table 4, it is observed that 4

G5/2 and 2H11/2 states have higher and lower values of radiative transition probabilities,

respectively. The excited states which have higher radiative lifetimes are useful for laser activity. In the present work, 2H11/2 and 4G5/2 excited states have relatively higher and lower radiative lifetimes than the other excited states in all the FP glasses. The branching ratios (βR) and integrated absorption cross-sections (Σ) are calculated for Nd3+ doped different FP glasses. The magnitudes of the branching ratios, which have greater than 50% are useful for laser action. Among different transitions, 4G5/2→4I9/2 (80-87%), 4F5/2→4I9/2 (66-68%), 2H11/2→4I15/2 (50-51%) and 4F3/2→4I11/2 (52-60%) transitions have greater than 50% of population for different FP glasses. From the magnitude of integrated absorption cross-sections, it is observed that among various transitions, 4G5/2→4I9/2, 4F5/2→4I9/2 and 4F3/2→4I11/2 transitions are having higher absorption cross-sections (Σ) and these values are in the range 23.4640.81x10-16 cm, 8.50-11.39x10-16 cm and 8.73-11.82x10-16 cm respectively. 3.9 Near infrared emission (NIR) spectra The emission spectra of different Nd3+ doped fluoro-phosphate glasses are recorded between 1000-1400 nm wavelength region using 586 nm as an excitation wavelength. Fig. 8 shows the fluorescence spectra of Nd3+ doped for different fluoro-phosphate glasses indicating two emission bands. From the spectra, it is observed that the fluorescence intensity of 4F3/2→4I11/2 transition centered at 1.06 μm is higher than the 4F3/2→4I13/2 transition. So, 4

F3/2→4I11/2 transition at 1.06 μm is the dominated laser emission in the prepared different FP

glasses. The pump at 586 nm excites the Nd3+ ions from the ground state, 4I9/2 to 4G5/2 + 2G7/2 excited states and then decay non-radiatively (NR) to the 4F5/2 metastable state. Due to very 11

small energy gap (~1000 cm-1) between the adjacent energy levels (4F5/2 and 4F3/2), the lower level (4F3/2) is populated through multiphonon relaxation at room temperature [12]. The energy difference between the 4F3/2 level and the next lower level 4I15/2 is sufficiently large enough and favors radiative emission over non-radiative emission. It is observed that in the present work, the luminescence intensity is decreasing in the order: Li→Bi→Na→K→Zn. Due to variation of the crystal field interaction around neodymium ions, the luminescence intensity varies. So, in the present work lithium FP glass matrix which has high intensity contributes efficiently as active media for Nd3+ ion. The magnitude of various emission parameters for the Nd3+ doped different glass samples are given in Table 5. It is noted that the predicted radiative transition probability (AR) for the transition 4F3/2→4I11/2 of Nd3+ doped lithium fluoro-phosphate glass (which has high emission intensity) is 2129 s-1, which is higher than another kind of fluorogallate (1576 s-1) [14], barium borate (1592 s-1) [15], and fluoroindate glasses (385 s-1) [45] and lower than the tellurite (3515 s-1) [43], phosphate (2658 s-1) [44], sulphide (3800 s-1) [46] and lead borate glasses (2197 s-1) [47]. Higher spontaneous emission probability provides better opportunity to obtain laser action. The experimental branching ratios (βexp) are obtained from the integrated areas under the emission peaks. It is worth to note that the NIR emission transition, 4F3/2→ 4I11/2 at 1060 nm possess higher magnitudes of βexp than the 4F3/2→ 4I13/2 emission transition at 1324 nm for all the prepared glass matrices. Among all the FP matrices, lithium FP and zinc FP glasses show higher (97%) and lower (93%) magnitudes of branching ratios for the 4F3/2→4I11/2 emission transition, respectively. Lithium FP glass matrix shows higher βexp than the other reported fluorogallate (46%) [14], lead borate (51 %) [47] and fluoro-phosphate (64 %) [48]. According to Fuchtbauer–Ladenburg theory, the emission cross sections can be calculated using the formula

 p aJ , bJ  

4p 8cn eff 2

A rad (aJ , bJ ' )

where P is the emission peak wavelength and λeff, is the effective linewidth determined by given formula eff =

I( )

 I max d

where Imax is the maximum intensity at fluorescence emission peaks. To obtain good solid state laser material, the effective linewidth should be narrow. In the present work, the 4

F3/2→4I11/2 emission spectral line has relatively sharp and narrow bandwidth (~ 19 nm) than 12

other emission line (~ 31 nm) which is suggested for solid state lasers. It has been reported so far that the values of Δλeff at 1060 nm are 49.00 nm (fluorogallate) [14], 39.44 nm (barium borate) [15] and 27.80 (tellurite) [43] glasses. The stimulated emission cross-section (σp) is one of the key parameters to determine the lasing action and its value signifies the rate of energy extraction from the lasing material. It is noted that the emission cross section of 1.06 μm emission in the prepared glasses pumped by 586 nm laser diode is significantly higher than that of other emission transition. It should be noted that the σp depends on the radiative lifetimes. The 4F3/2→ 4I11/2 transition at 1060 nm exhibits high stimulated cross section i.e. 6.33x10-20 cm2 in lithium FP glass matrix. The σp values are found to be larger than that in barium borate (2.49x10-20 cm2) [15], tellurite glasses (3.89x10-20 cm2) [43], fluoro-phosphate (3.69x10-20 cm2) [48], sodium zinc borate (2.26x10-20 cm2) [49] and phosphate glass (4.12x10-20 cm2) [50]. Thus, this kind of fluorophosphate glasses can be selected as an appropriate laser host material. Zinc FP glass exhibits little emission cross-section value.

3.10 Luminescence decay kinetics Fluorescence decay profiles of the 4F3/2 multiplet of Nd3+ doped different FP glasses are shown in Fig. 9. The decay profiles show characteristics of a non-exponential behavior. The presence of small trace impurities (OH-) or defects in the structure (from FTIR analysis) is likely to have impact on the decay levels. The average lifetime decay constants are estimated using the formulae given in ref. [51] and are found to be 124, 145, 150, 148, and 180 μs for Li, Na, K, Zn and Bi fluoro-phosphate glasses, respectively. The luminescence lifetime of 4F3/2 excited level of Nd3+ is shorter and hence, decays quickly from the 4F3/2 excited level to ground state level.

4. Conclusions All the prepared glass materials have amorphous nature, as confirmed by X-ray diffraction and SEM analysis. FTIR and Raman spectra enabled detection of bands related to structure and it was pointed that the prepared fluoro-phosphate glasses built with phosphate and various network modifiers. On the basis of 31P MAS NMR analysis, it was stated that a decrease of the proportion of Q2 units in the sequence Li→Na→K→Zn→Bi fluorophosphate glasses and show that these oxides are incorporated into glass as network modifier oxides. From 27Al analysis, it is observed that all the prepared glasses tend to co-ordinate in the Al(6) structure except lithium FP glass matrix. From XANES spectra, it is established that 13

the prepared Nd doped glasses have +3 as oxidation state.

Among the three intensity

parameters in all fluoro-phosphate glasses, particularly in zinc fluoro-phosphate glass, there is higher magnitude in Ω2 parameter which is attributed to higher covalency and asymmetry. Moreover, higher value of Ω2 parameter indicates, a preferential coordination of oxygen ligand to neodymium ion compared to fluorine ligand. Among various transitions, the transitions, 4G5/2→4I9/2, 4F5/2→4I9/2 and 4F3/2→4I11/2 have higher magnitude of integrated absorption cross-sections and branching ratios. It is clear from photoluminescence measurements that the narrowing of the emission peak and modulation of emission peak intensity (in the NIR region) suggests dependence of emission properties on the host glass matrix. It is clear that among various glass matrices, lithium fluro-phosphate glass matrix has better performance in terms of high luminescence intensity, sharp and narrow bandwidth character, larger emission cross sections and better lifetimes. So, the present optical characterizations shown that lithium fluoro-phosphate glass matrix is the most favourable optimized host glass matrix for near infrared laser emission transition, 4F3/2→4I11/2 operated at 1060 nm wavelength.

Acknowledgment One of the authors S. Babu would like to thank University Grants Commission (UGC), New Delhi for the sanction of Senior Research Fellowship (SRF) under Research Fellowship in Sciences for Meritorious students (RFSMS) scheme. The authors are thankful to Dr. S.N. Jha, Scientific officer (SOH), A&MPD, RRCAT, Indore, for providing EXAFS measurements. The support of Dr. Parasmani Rajput, Scientific officer (SOD), A&MPD, RRCAT, Indore to record the data is gratefully acknowledged. The authors thank SAIF, IISc, Bangalore, for providing Solid State NMR facilities.

References

[1] M. Sobczyk, Opt. Mater. 35 (2013) 852–859. [2] F. Zhang, Z. Bi, A. Huang, Z. Xiao, J.Lumin. 160 (2015) 85–89. [3] Meng Wang, Lixia Yi, Guonian Wang, Lili Hu, Junjie Zhang, Solid State Commun. 149 (2009) 1216-1220. [4] M. Sołtys, J. Janek, Lidia Zur, J. Pisarska, W. A. Pisarski, Opt. Mater. 40 (2015) 91–96. 14

[5] S. Hraiech, M. Ferid, Y. Guyot, G. Boulon, J. Rare Earth., 31 ( 2013) 685. [6] C. Ming, F. Song, Y. Qin, X. Ren, Liqun An, Opt. Commun. 321 (2014) 195–197. [7] B.C. Jamalaiah, L. Rama Moorthy, Hyo Jin Seo, J. Non-Cryst. Solids 358 (2012) 204-209. [8] H. Doweidar, Y.M. Moustafa, K. El-Egili, I. Abbas, Vib. Spectrosc. 37 (2005) 91–96. [9] K.H. Mahmoud, Opt. Commun. 320 (2014) 68–72 [10] J.H. Choia , Alfred Margaryanb , Ashot Margaryanb , F.G. Shi, J.Lumin. 114 (2005) 167–177. [11] K. Lu and N. K. Dutta, 89, 3079 (2001) [12] T. Wei, Ying Tian, Cong Tian, Xufeng Jing, Muzhi Cai, Junjie Zhang, Long Zhang, Shiqing Xu, J. Alloys Compds. 618 (2015) 95–101 [13] M. Farouk, A. Abd El-Maboud, M. Ibrahim, A. Ratep, I. Kashif, Spectrochim. Acta Part A 149 (2015) 338–342. [14] Tianfeng Xue, Liyan Zhang, Junjiang Hu, Meisong Lia, Lili Hu, Opt. Mater. 47 (2015) 24–29. [15] G.V. Vázque, G. Muñoz H, I. Camarillo, C. Falcony, U. Caldiño, A. Lira, Opt. Mater. 46 (2015) 97–103. [16] A. K. Poswal, A. Agrawal, A. K. Yadav, C. Nayak, S. Basu, S. R. Kane, C. K. Garg, D. Bhattachryya, S. N. Jha and N. K. Sahoo. AIP Conf. Proc. 1591 . 2014. 649

[17] H.A. ElBatal, A.M. Abdelghany, F.H. ElBatal, Kh.M. ElBadry, F.A. Moustaff, Physica B 406 (2011) 3694–3703. [18] Liyan Zhanga, Lei Wena, Junjie Zhanga, Lili Hua, Mater. Chem. Phys. 91 (2005) 166–171. [19] Yuanming Lai, Xiaofeng Liang, Guangfu Yin, Shiyuan Yang, Junxia Wang, Haixiang Zhu, J. Mol. Struct. 1004 (2011) 188–192 [20] R.M.M. Morsi, S. Ibrahiem, Physica B 406 (2011) 2982–2989. 15

[21] Hong-Wei Li, Shi-Qing Man, Opt. Commun. 282 (2009) 1579-1583. [22] G.S. Henderson, R.T. Amos, J. Non-Cryst. Solids 328 (2003) 1-19. [23] J. York-Winegar, T. Harper , C. Brennan, J. Oelgoetz, A. Kovalskiy, Physics Procedia 44 ( 2013 ) 159 – 165. [24] S. Babu, M. Seshadri, V. Reddy Prasad, Y.C. Ratnakarama, Mater. Res. Bull. 70 (2015) 935–944. [25] M.R. Sahar, A. Wahab M.A. Hussein, R. Hussin, J. Non-Crystalline Solids 353 (2007) 1134–1140. [26] R. James Kirkpatrick, Richard K. Brow, Solid State Nucl. Mag. 5 (1995) 9-21. [27] Che-Zoue Weng, Jia-Hong Chen, Ping-Yu Shih, Mater. Chem. Physics 115 (2009) 628– 631. [28] P.Y. Shih, J.Y. Ding, S.Y. Lee, Mater. Chem. Physics 80 (2003) 391–396. [29] D. Qiua, P. Guerry, I. Ahmed, D. M. Pickupa, D. Cartaa, J. C. Knowles, Mark E. Smithb, Robert J. Newport, Mater. Chem. Physics 111 (2008) 455–462. [30] J.W. Wiench, B. Tischendorf, J.U. Otaigbe, M. Pruski, J.Mol. Struct. 602-603 (2002) 145157. [31] S.M. Hsu, S.W. Yung, R.K. Brow, W.L. Hsu, C.C. Lu, F.B. Wu, S.H. Ching, Mater. Chem. Phys. 123 (2010) 172–176 [32] B. Tiwari, M. Pandey, V. Sudarsan, S.K. Deb, G.P. Kothiyal, Physica B 404 (2009) 47–51 [33] G. Gongyi, Solid State Nucl. Mag.13 (1998) 119–122.

[34] I.M. Nagpure, Shreyas. Pitale, E. Coetsee, O.M. Ntwaeaborwa, J.J. Terblans, H.C. Swart, Physica B 407 (2012) 1505–1508. [35] A. Majjane, A. Chahine, M. Et-tabirou, B. Echchahed, Trong-On Do, P. Mc Breen, Mater. Chem. Phys. 143 (2014) 779-787.

[36] M.A. Tshabalala, F.B. Dejene, Shreyas S. Pitale, H.C. Swart, O.M. Ntwaeaborwa, Physica B 439 (2014) 126–129. 16

[37] Huilong Guo, Jifu Bi, Jiayi Wang, Xuequan Zhang, Shichun Jiang, Zhonghua Wu, Dalton Trans., 44 (2015) 9130.

[38] W.T. Carnall, H. Crosswhite, H.M. Crosswhite, Argonne National Laboratory Report II, 1977. [39] B.R. Judd, Phys. Rev. 127 (1962) 750. [40] G.S. Ofelt, J. Chem. Phys. 37 (1962) 511 [41] S. Babu, A. Balakrishna, D. Rajesh, Y.C. Ratnakaram, Spectrochim. Acta Part A 122 (2014) 639. [42] S.Sunil Kumar, K. Jayakrishna, S. Kasthurirengan, K.S.R. Koteswara Rao, K.P. Ramesh, J. Non-Cryst. Solids, 357 (2011) 842–846. [43] H. Kalaycioglu, H. Cankaya , G. Ozen , L. Ovecoglu , A. Sennaroglu, Opt. Commun. 281 (2008) 6056-6060. [44] G.Takebe, Y. Nageno, K. Morinaga, J. Am. Chem. Soc. 78 (1995) 2287 [45] A Fl re , J.F Mart ne , M Fl re , P Porcher, J. Non-Cryst. Solids, 284, (2001) 261-267. [46] B. Viana, M. Palazzi, O. LeFol, J. Non-Cryst. Solids, 215 (1997) 96-102. [47] Joanna Pisarska, Wojciech A. Pisarski, Witold Ryba-Romanowski, Optics & Laser Technology 42 (2010) 805–809. [48] S. Surendra Babu, P. Babu, C.K. Jayasankar, A.S. Joshi, A. Speghini, M. Bettinelli, J. Non-Cryst. Solids, 353 (2007) 1402-1406. [49] B. Karthikeyan, S. Mohanb, M.L. Baesso, Physica B 337 (2003) 249–254. [50] N. O. Dantas, E. O. Serqueira, A.C.A. Silva, A.A. Andrade, S. A. Lourenço, Braz J Phys 43 (2013) 230–238. [51] B. J. Chen, L. F. Shen, E. Y. B. Pun, and H. Lin, 5 (2014) 113-123.

Figure captions

17

Fig.1a. XRD pattern of Nd3+ doped lithium fluoro-phosphate glass matrix. Fig.1b. SEM image of Nd3+ doped lithium fluoro-phosphate glass matrix. Fig.2 FTIR transmittance spectra of Nd3+ doped different fluoro-phosphate glasses. Fig.3 Raman spectra of Nd3+ doped different fluoro-phosphate glasses. Fig.4a 31P MAS NMR spectra of Nd3+ doped different fluoro-phosphate glasses Fig.4b 27Al MAS NMR spectra of Nd3+ doped different fluoro-phosphate glasses Fig.5a P 2p XPS spectra of Nd3+ doped different fluoro-phosphate glasses Fig.5b O 1s XPS spectra of Nd3+ doped different fluoro-phosphate glasses Fig.6. XANES spectra of standard Nd3+ and Nd3+ doped different fluoro-phosphate glasses. Fig.7. Absorption spectra of Nd3+ doped different fluoro-phosphate glasses. Fig.8 Emission spectra of Nd3+ doped different fluoro-phosphate glasses. Fig.9 Decay profiles of Nd3+ doped different fluoro-phosphate glasses.

18

Fig.1a

Fig.1b

19

Fig. 2

20

Fig.3

21

Fig. 4a

22

Fig. 4b

23

24

25

Fig. 6

26

27

28

29

Table 1 31

P MAS NMR spectra of Nd3+ doped different fluoro-phosphate glasses.

Glass

Chemical shift δ (ppm)

Units

Li Na

-20 -31 -19 6 -33 -18 5 -40 -20 -44 -28 -2

Q2 Q2 Q1 QO Q2 Q1 QO Q3 Q2 Q3 Q2 QO

K

Zn Bi

30

Table 2

S

Transition

Li

Na

K

Zn

Bi

.NO fex p

l

1

2

P1/2

81

K13/2+2G9/2+ 2

D3/2

1.

3

0.

1. 4

G9/2

91

4

1. 42

2. 4

G7/2

5

71

2

G5/2+ G7/2

.70

6 F9/2

7

07 12

4

S3/2+4F7/2

.10

8

11 .24

7. 4

F5/2+2H9/2

79

.00

8. 91

70

10

59

00

55

.80 7.

12

3

2.40

2.13

2

0

10

.80 1

1

1.20

0.36

8

9

9. 58

.59

.00

.39

3

3

0.

8. 90

3

23 .11

1 .87

.97

80

10

2

3.

1. 93

7.

6. 86

.20

0.

8.

8. 35

.11

1 .27

.10

14

23

1

1.

3.

0 .60

.93

88

82

19

1. 66

9. 31 23

7. 50

.20

0.

2.

1

1.

3.

fc al

.19

30

23

56

19

1.

1.

3.

0. 64

86

44

50

.56

1. 09

1.

2.

24

1.

fe xp

1. 14

00

85

86

.60

0. 81

1.

3.

24

1.

fca l

0. 51

91

61

67

.55

1. 4

1.

2.

25

1.

fex p

0. 52

10

55

09

25 4

1.

fca l

0. 50

78

fex p

0. 82

97

fca l

0. 21

54

fex p

0. 2

2

fca

.56

.50

9

1. 4

F3/2

RMS deviation

99

1. 43 ±0.79

2. 70

2. 20 ±0.70

1. 84

2. 10 ±0.75

2. 81

2. 70 ±0.89

3 .29

2 .56 ±0.88

Experimental (fexp) and calculated (fcal) spectral intensities (x10-6) of Nd3+ doped different fluoro-phosphate glasses.

32

Table 3 Judd-Ofelt (J-O) intensity parameters (Ωλ, x10-20) of Nd3+ doped different fluoro-phosphate glasses



Glass 2

Li

4

8. 04

Na

1

6.

K Zn Bi

Fluorog allate

42

Barium Borate

95

.29

Phospha 27

3 1

4

3

2>6

Present work

2>6

Present work

6>2

Present work

6>2

Present work

2>6

Present work

2>4

[14]

6>2

[15]

2>4

[43]

2>6

[44]

>6 4

.31

Referenc

>4

.89 3

.69

>6

2.00

.06 5.

>4

.71 8

4.

>4 6

4

.91

71

6

.86

.16 9.

>4

.86 4

6.

Tellurite

4

4

8.

>4

.97

.60

76

6

3

5.

>4

.13

.68

27

7

3

4.

Trend es

.67

.62

48

 6

.47

44

te



>4 33

Table 4 Total radiative transition probabilities (AT) ( s-1) and radiative lifetimes (τR) (μs) of certain excited states of Nd3+ doped different fluoro-phosphate glasses

4

G

4

G9/2

4

G7/2

2

G5/2

4

H11/2

4

F9/2

4

F5/2

F3/2

lass τ

A T

L i

R

1 6012

N a K

n

5937 B

i

0633

8

371 6

3

9

9 4

7

1 3782

2

9140

77

5

59

7

7

4

697

65

248

150 5

79

728

34

400

371

2

3 11

4 280

1 86

74

219

83

2

3

1

5

86

650

45

459

371

3

3

2

5

2 29

13

079

27

τR

497 2

4

2

4

11

692

08

2

4

3

4

1

729

60

A T

4

2

3

2

T

36

842

τR

A

2

3

2

4

3

232

178

71

R

4

2

3

4

T

095

τ

A

2

4

5

2

R

4

4

2 1479

7

3

1 7602

9

4

2 2374

1

1

2

τ

A T

3128

19

0266 4

0

8

τR

A T

9

1

7

2 0406

1075

3

1

R

1

6

1 2856

Z

6

1 5950

T

3

τ

A

2 34

4 192

2 39

Table 5 Emission band positions (p, nm), effective bandwidths (∆λeff, nm), radiative transition probabilities (A, s-1), peak stimulated emission cross-sections (σp, x10-20cm2) and branching ratios (exp, %) of emission transitions of Nd3+ doped different fluorophosphate glasses.

S

Transition

.No

Param

Li

Na

K

Zn

Bi

106

535

53

535

535

19

20

220

2187

eters

p ∆λeff 4

1 4

F3/2→

I11/2

0

19

AR

19

σp

212 5

βexp

5 19

192

9

16 5.48 26

6.3 3

∆λeff 4

2 4

I13/2

F3/2→

96

93

3 97

p

6.15 5.35

5.4

95

5

95

132 4

650 31

65 0

650

650

32

31

AR

32

387

30

433

435

σp

483

3.13

31

2.37

3.35

βexp

3.6

5

7

4

3 2.8

3 1

5

35