Open Access
Add to BibSonomyAbstract
A set of sodium silicate glass matrices were synthesized to study the influence of Na2O concentration on the optical properties of Eu3+. The samples were characterized by optical absorption (OA), time and energy resolved photoluminescence and Fourier Transform Infrared (FTIR). We observed that decreasing sodium oxide concentration affects the absorption of OH radicals in the host matrix. Adjusting the obtained FTIR spectra by Gaussian functions, we observed the existence of two possible non-radiative transfer channels with the transition from the 5D0 to the 7F2 state of the Eu3+ ions. The first was produced by resonance with the 5th harmonic vibration of OH bonded radicals and the second by resonance with the 14th harmonic vibration of the silicon network. A decrease in OH radicals observed by FTIR was followed by an increase in the lifetime of the 5D0 state of the Eu3+ ions. However, resonance with higher harmonic orders did not affect the optical properties of the Eu3+ ions. Increases in the lifetime of this transition (~3.3 ms) were obtained from the synthesized sample with the lowest sodium oxide concentration. This lifetime is comparable with the well-known YAG system. Molecular dynamic results show that decreasing sodium oxide content in the host matrix produces structural changes such as decreases in non-bridge oxygen species, which may explain the decreases in OH radical absorption seen in the experimental results.
©2013 Optical Society of America
1. Introduction
Glass systems doped with Rare Earth (RE) ions have been the object of several studies. They have attracted the attention of material science researchers in recent decades because of their good mechanical and thermal stability, inexpensive synthesis, favorable thermal conductivity and heat capacity compared to crystals [1–3]. Glass doped with Eu3+ ions has been used in optical devices, field emission technology, phosphors and LEDs with more efficient red luminescence [1–3]. The optical properties of Eu3+ ions embedded in glass matrices have been investigated due to their hypersensitivity [4] of the transition of the 5D0 to 7F2 state to the local environment. The probability of this transition is associated with the covalence around the Eu3+ ions [5]. This transition is located in the red region of the electromagnetic spectrum (around 614 nm) and is due to 4f-4f transitions [6].
The efficiency of technological devices containing Eu3+ ions is related to the optical properties of these ions in specific glass matrices. In particular, sodium silicate glasses need higher concentrations of Eu3+ to enhance their optical quality. However these ions have low solubility in silicate glasses which increases the probability of cluster formation and/or possible luminescence quenching due to ion-ion interactions [7]. Vijaya et al., reported that the Q2 structure of PO2 groups is responsible for the phonon relaxation of Eu3+ ions in a host phosphate matrix [8]. These authors concluded that the lifetime of the 5D0 level of Eu3+ ions in glasses containing this type of group is independent of Eu3+ ion concentration but depends on host material and that the probability of non-radiative energy transfer between Eu3+ ions is negligible [8]. These results have also been found in other host matrices [9]. However, contamination with OH radicals is also known to produce luminescence quenching. To decrease the concentration of this impurity and better understand this luminescence quenching, resarchers have studied OH radicals in sodium silicate glasses [10]. It is well known that non-radiative transitions are prominent in phonon relaxation. There are several theoretical studies about these phonon transitions. Pukhov et al. [11] said that Frenkel was the first to point out that the linear term of the Taylor expansion of the electron-network interaction Hamiltonian may generate n-phonon transitions. This occurs if the vibration modes of the network are dependent on electronic states involved in non-radiative transition. These researchers also assumed that certain terms in the model can cause n-phonon relaxation even though the vibration modes are the same for different electronic states (nonlinear mechanism). Thus, besides the hydroxyl, the host may absorb energy through the non-radiative transition of rare earth ions. The phonon energy of the host can be defined as the highest vibrational energy observed in the infrared spectrum that plays a major role in the optical properties (particularly multiphonon relaxation) of optically active ions [8]. Controlling spontaneously emitted light through photonic structures has always been important [12]. Dejneka et al. found that fluoride glass has low phonon energy and hence more efficient radiative transitions than other types of glass [13]. However, fluoride glass has lower mechanical resistance than silicate glasses used in optical devices [14].
As previously stated, OH radicals and higher concentrations of Eu3+ ions are responsible for decreases in luminescence. Cluster formation of RE ions has been the subject of much research and has been largely explored through Molecular Dynamic teniques by Cormier et al. [15], Afifty and Monteil et al. [16], Du and Cormack [17] and Kokou and Du [18]. Kokou and Du have presented interesting bibliographic research about Eu3+-doped silicate glasses. They concluded that the tendency of rare earth ions to cluster in silica glasses is higher than in modified silicate glasses. In silicate glasses, rare earth ions are better dispersed as a result of the more open network structure and the availability of non-bridge oxygen (NBOs). However correlation between the structure of sodium silicate glass and luminescence properties is lacking.
The objective of this study was to examine the effects OH bonded radicals and glass matrix composition have on the optical properties of Eu3+-doped sodium silicate glass. Optical Absorption (OA) and time and energy resolved photoluminescence (TRPL and ERPL, respectively) were used for optical characterization. The Fourier Transform Infrared (FTIR) techinique was used to obtain the mid-infrared absorption spectra. Molecular Dynamic simulations were performed in order to gain understanding of the local structure of the sodium silicate glass system and help understand the interaction between the host and OH radicals with Eu3+ ions.
2. Experimental details
A set of S(98-x)Nx matrices with nominal composition SiO2, Na2CO3 and Eu2O3, were synthesized by fusion at different compositions as shown in Table 1. After mixing, these materials were annealed at 900°C for 4 hours to remove CO2 traces from Na2CO3, yielding Na2O. The mixtures were then melted at 1600°C for 1 hour in a nitrogen rich atmosphere using porcelain crucibles. The resulting melt was rapidly cooled to approximately room temperature. OA spectra (190–3300 nm) were obtained from the synthesized samples using a SHIMADZU UV-3600 spectrophotometer. Absorption spectra (400 – 4500 cm−1) were measured using the KBr pellet method with an FTIR IR Prestige – 21 SHIMADZU spectrophotometer. The luminescence spectra were recorded via the ERPL technique using a spectral CCD JAZ (200 to 1000 nm) with the 405.0 nm line of a diode laser and an interference filter (405.0 nm). The decay characteristics of the excited states of Eu3+ were measured by the TRPL technique using a Tektronix DPO 2012 digital phosphor oscilloscope (100 MHz 1GS/s) and a silicon detector. Densities of the prepared glass samples were determined by Archimedes principle using distilled water as an immersion liquid. All characterizations were performed at room temperature.
Table 1. Nominal Composition of the Synthesized Samplesa
3. Molecular dynamic details
Molecular dynamic simulations were performed using the DL POLY package (version 4.03) developed by Smith and Forrester at Daresbury Laboratory, UK [19]. Coulomb interactions were calculated by the Ewald summation method with a cutoff of 12.0 Å and a precision of 1 · 10−6. The short-range interaction cutoff was 8.0 Å. The motion equation was integrated by the Verlet Leapfrog algorithm with a time step of 2 · 10−15 s. During simulations, the canonical ensemble (NVT) was used with a Berendsen thermostat for equilibration and cooling [20]. A 3D box of each glass system was constructed to mimic the experimental density of the synthesized samples. In addition, two extreme systems were simulated (64.86 and 94.83 mol % SiO2), but not synthesized (Table 2).
Table 2. Concentration (in mol %) of Each Matrix Component, Number of Each Type of Ion, Density (D [g.cm−3]) and Box Length (L [Å]) of Each System Simulated by the Molecular Dynamic
The Buckingham interatomic potential was applied to simulate the sodium silicate systems. The atomic parameters used in the simulations were the same as those published by Du and Cormack [17]. The potential used for short-range interactions had the Buckingham form:
where r is the distance between atom pairs and A, c and ρ are listed in Table 3. Similar to Du and Cormack [17], we included a repulsive term V(r) = B/rn (r < 0.7 Å) in the DL POLY 4 to account for possible variation singularities in the Buckingham potential during glass melting.Table 3. Buckingham Potential Parameters Used in This Studya
Initial configurations of each system were randomly generated using packmol [22] with a minimal distance of 2 Å for each pair of ions. This distance placed the atoms in a cubic simulation cell with a cell volume and density consistent with the experimental glass. The glasses were formed by simulating melting and cooling. First each system was equilibrated at zero K for 40000 steps and then raised to a nominal 6000 K temperature for 40000 steps. This was followed by 5000K, 4000K, 3000K, 2000K and 1000 K temperature runs with 40000 steps each. Only the first 20000 steps were permitted to scale to the velocity of each ion. The glasses were then continuously cooled to 300 K at a rate of 10 K/ps and then equilibrated for 20000 steps (see Fig. 1). After the equilibration step 400 configurations were recorded for statistical analysis and then plotted the distribution functions (RDFs), coordination number (CN), bond angle distribution (BAD) for each set of atoms and silicon Qn distributions for each sample. Structure factors were analyzed by calculated X-ray and Neutron Scattering.
4. Results and discussions
Figure 2 shows the OA spectra of all synthesized samples. Here, bands associated with the presence of Eu3+ ions, the structure of the glass matrix and SiO4 and OH vibrational modes are visible [23]. The bands associated with Eu3+ transition were identified according to Carnall [24]. The most intense of these are indicated in Fig. 2. The spectra obtained by 405 nm excitation are in resonance [25] with the Eu3+ 7F0-5D3 transition. The intensity of the band corresponding to this transition increases weakly with sodium oxide concentration. This was expected since the concentration of Eu2O3 was kept constant for all synthesized samples. In Fig. 2, it is interesting to note that in addition to the bands produced by the presence of Eu3+ ions it is possible to identify a broad band centered at 1000 nm. This band results from the combined vibrational modes of absorbed OH and SiO4 [26]. Another band, between 2700 and 3300 nm is associated with the presence of OH absorbed in the glass matrix at different sites [27]. This is also the only band affected by changing Na2O concentration. It was expected that this variation in OH concentration would affect the spectroscopic properties of the Eu3+ ions.
Fig. 2 Optical absorption spectra of the synthesized sodium silicate glasses doped with Eu3+ ions. The variation in the sodium oxide concentration affects only the OA band attributed to the presence of OH radicals in the samples.
Figure 3 shows the luminescence spectra of Eu3+ ions in the VIS region from 550 to 750 nm for all the synthesized samples. This range of the electromagnetic spectrum shows electric dipole (ED) and magnetic dipole (MagD) transitions. These bands are centered at approximately 578.8 (5D0 to 7F0; ED); 592.5 (5D0 to 7F1; MagD); 611.0 (5D0 to 7F2; ED); 654.4 (5D0 to 7F3; ED); 704.6 (5D0 to 7F4; ED) and 689.0 nm (5D1 to 7F4; ED). Only transitions where ΔJ = 2, 4 and 6 are allowed by the induced electric-dipole mechanism, if luminescence starts from a level where J = 0. The MagD 5D0 to 7F1 transition is independent of the host matrix [8]. Thus, the obtained spectra were normalized with respect to the maximum intensity of this transition (5D0 to 7F1). The full-width at half-maximum (FWHM) of the band centered at 611 nm is approximately (9.32 ± 0.91 nm) and was not affected by decreasing sodium concentration. Another author [28] observed that FWHM changes from 13.2 nm for glass to 0.6 nm for silica glass-ceramic. This behavior is attributed to the ordering of the local environment around Eu3+. Therefore, the change in Na2O content did not affect the local environment in our samples. The intensity of the band representing the transition from the 5D0 state to the 7F2 state increases with decreasing Na2O concentration. This behavior can be explained by the fact that decreasing Na2O concentration favors Eu3+ radiative transitions.
Luminescence decay can be used to experimentally determine the lifetime of any observed radiative transition. Luminescence intensity is proportional to the rate of spontaneous decay, which is proportional to the population of this state. Therefore, in this simplest case, the luminescence decay is single exponential I(t) = I(0) exp[-t/τexp], where I(t) is the emission intensity after the pulsed excitation, τexp the lifetime of the excited state, I(0) a constant and t is time [2]. Figure 4(a) shows the luminescence decay of the transition from 5D0 to 7F2 as a function of time for the S69N29 samples. The values of the luminescence decay reported for each composition have been obtained as a mean of 32 measurement acquisitions. The ration of the signal/noise is not significant (signal/noise >>> 1). Figure 4(b) shows the lifetime of this state calculated as previously mentioned and as function of sodium oxide concentration. Here, lifetime increases linearly as sodium oxide concentration decreases. This behavior can be explained by the interaction between the host and the Eu3+ ions. In general, Eu3+ lifetime increases as Eu2O3 concentration decreases in the glass matrix [29]. However, in Fig. 4(b), the lifetime increases without a decrease in Eu2O3 concentration. This fact plus the conclusions from Fig. 1 indicate that the reduction in the OH group is responsible for the increase in Eu3+ lifetime and luminescence intensity. Note that the lifetime of the sample with the lowest concentration of sodium oxide is approximately 3.3 ms. This value is comparable to those of crystal [30] systems such as YAG (3.46 ms).
Fig. 4 (a) Luminescence decay of the transition from 5D0 to 7F2 for the S69N29 samples as function of the time. (b) Luminescence lifetime measured for the transition from 5D0 to 7F2 as function of sodium oxide concentration in the glass matrix.
Further investigations of all the synthesized samples were performed by analyzing the FTIR spectra (Fig. 5). It is known that a hydroxyl radical attached to the sodium silicate glass network produces a band in the infrared region of the absorption spectra. In Fig. 5, this band is centered at 3250 cm−1. It is also well known that this band is a result of hydroxyl radicals attached at different sites. According to Nawrocki et al. [27], hydroxyls are absorbed at 4 different sites. Consequently, we have adjusted the band using 4 gaussian functions centered at 3630, 3470, 3260 and 3030 cm−1. These are represented by blue curves in Fig. 5 and in Table 4, respectively. According to Nawrocki the band at 3030 cm−1 is attributed to physically adsorbed water, and the 3260, 3470 and 3630 cm−1 bands are attributed to bonded H- and –OH at different OH...H distances.
Fig. 5 FTIR spectra in the region showing a band associated with the presence of OH in the glass matrix for all synthesized samples.
Table 4. Areas of the 4 Adjusted Curves in the FTIR Band Associated with the Presence of Hydroxyl in the Glass Matrixa
Energy transfer processes involving non-radiative transitions are favored by the resonance mechanism. Therefore, before energy can transfer from Eu3+ ions to OH bonded to the glass matrix, resonance is needed between the excited state of the Eu3+ ions and any OH vibrational frequency. The ratios between the transition energy from 5D0 to 7F2 of the Eu3+ ions and the OH radical modes (3030, 3260, 3470 and 3630 cm−1) are 5.40, 5.02, 4.72 and 4.51 respectively. The transition from the 5D0 to 7F2 states in Fig. 3 reaches a maximum of 611 nm (16366.6 cm−1) which corresponds to the fifth harmonic of the hydroxyl vibrational frequency. Note that in Table 4 the area of this band decreases as Na2O content decreases. This indicates that increased lifetime is due to a reduction in non-radiative energy transfer to OH in this specific configuration.
Figure 6 shows the FTIR spectra from 700 to 1500 cm−1. The bands in this region are associated with the vibrational modes of the silicon tetrahedral (SiO4) structure [31]. These bands were also adjusted using Gaussian function. The centers and areas of each of obtained curve are presented in Table 5. The Si-centered tetrahedral structures are designated as Qn, where Q refers to a silicon atom and n denotes the number of bridging oxygen (BOs) in the structure. The band from approximately 1179.0 to 1169.9 cm−1 is assigned to Si–O0 (O0, bridging oxygen) stretching in fully polymerized structural species (Q4). The band from approximately 1054.0 to 1048.7 cm−1 was assigned to Si–O0 vibrations in structural species that contain bridging oxygen, but are not necessarily fully polymerized, or to vibrations associated with alkali metals. These assignments are somewhat controversial [32]. Other authors have attributed this band to the manifestation of Q3 species [33]. The band from about 930 to 939.4 cm−1 is assigned to Si–O− stretching in species with NBO/Si = 2 and is referred to as Q2 in the text [32]. On the right side of this wide band (~1460 cm−1) P6 is a small shoulder that is more evident in the Na-rich glass. This can be assigned to stretching modes of structures containing carbonates. However, this assignment is also controversial. Some authors, such as Potapova et al., attribute this structure only to bands with higher wavenumbers (1700 - 1730 cm−1). In Fig. 6, the area of this band decreases with decreasing Na2O concentration and is smaller than the areas attributed to SiO4 structures [34].
Fig. 6 FTIR spectra in the region showing a band associated with the SiO4 structures in the glass matrix for all synthesized samples.
Table 5. Obtained Center and Area for the Fitted Curves of the FTIR Bands Associated with the SiO4 Structurea
The band centered between 760.4 and 782.2 cm−1 is associated with Si–O–Si bending vibrations [33]. The absorption band from 457.6 to 466.2 cm−1 is attributed to the bending mode of (O–Si–O) NBO vibrations. The maximum intensity of this band shifts to lower wavenumbers as Na2O concentration decrease. Similar behavior has been observed in glasses containing Mg+ and attributed to the increasing strength of cation-oxygen bonds [31].
Figure 6 and Table 5 show the change in concentrations of various types of Si-O based on Gaussian curve areas. The integrated peak intensities (P1, P3 and P4) decrease with decreasing Na2O concentration and consequently, non-bridge oxygen (NBO) formation in the silicon network also decreases. Nevertheless, bridging oxygen (BO) in the system favors increasing SiO4 in the Q4 state, as can be confirmed by the increased P5 area.
The intense band in the FTIR is caused by absorption of tetrahedral SiO4, which is predominantly Q2, Q3 and Q4 (Fig. 6 and Table 5). There is also a relationship between the transition energy from 5D0 to 7F2 of the Eu3+ ions (S89N9 system) and the frequency modes of Q2,3,4 (17.42; 15.61; 13.99; respectively), as observed in other ions [35]. This result indicates that the fourteenth harmonic vibration mode Q4 is resonant with the transition at 611 nm of the Eu3+ ions. Thus, the Q4 configuration of the SiO4 tetrahedral contributes to non-radiative energy transfer from Eu3+ ions to the host [11]. This is valid for synthesized samples with low Na2O concentrations. This can be seen Fig. 6 by the dislocation of P5 (Q4) from 1179 to 1170 cm−1 resulting from decreasing Na2O concentration (Table 5). This shift should increase non-radiative energy transfer from Eu3+ ions to the silicon network (Q4). However, the linear increase in the lifetime of the Eu3+ ions (Fig. 4) indicates that this resonance is not sufficient to change lifetime gain. This is because non-radiative energy transfer with high-order harmonics requires strong-field interactions [36]. This is not the case for Eu3+ ions given that their inner 4f electrons are shielded from their outer filled 5s and 5p shells [37].
4.1 Molecular dynamic results
This section presents the results of the molecular dynamic (MD) calculations for the Eu3+-sodium silicate glass structure.
Figure 7 shows the radial distribution function and the coordination number for the Eu-O, Si-O, Na-O and Eu-Eu interactions in the simulated S64N34 and S94N4 systems. The Si-O peak is narrower and more intense than the others. The Si-O bond lengths obtained from the first peak position change from 1.54 to 1.60 Å due to decreasing sodium oxide concentration. The coordination number of the Si-O bond, 4 coordinated, was not affected by decreasing Na2O concentration. The first peak of the Na-O bond decreases considerably, as expected, and the average distance changes from 2.41 to 2.48 Å. The bond distance of Eu-O however, is less affected by decreasing sodium oxide concentration. Nevertheless, its coordination number decreases from 5.72 to 4.33. This can be explained by the decrease in non-bridging oxygen induced by the decrease in network modifiers, which according to Kokou and Du [18] represent more than 60% of the oxygen in the first coordination of europium. This change also reveals that there is no preferential symmetry around the Er3+ ions.
Fig. 7 Radial distribution function of (a) S64N34 and (b) S94N4 and the coordination number (c) and (d) at the respective concentrations.
The Eu-Eu interaction also decreased (6.10 to 8.95 Å) with diminishing Na2O concentration. This result was unexpected given RE ions have low solubility in silicate glass and decreasing network modifier concentration should limit diffusion and enhance the probability of RE-cluster formation. Only one peak was observed for the Eu-Eu interaction (S94N4) suggesting that Eu-cluster formation occurs only as pairs. This conclusion is in agreement with Du and Cormack’s work [17] on Er-cluster formation in low doping sodium silicate glasses.
Figure 8 shows the bond angle distribution functions of the O-Eu-O, O-Si-O and Si-O-Si bonds in the S79N19 system. All other glass simulations behaved the same. The Si-O-Si bond has a broad peak on the BAD plot between 120° to 180° and a maximum close to 150°, which correctly represents the silica network [38]. The O-Si-O bond has another narrower peak at about 110°, which indicates a higher degree of symmetry around the Si atoms compared to the oxygen. The O-Eu-O bond has a less defined BAD distribution with three major regions at approximately 70°, 90° and 170°. This implies a lack of the defined symmetry expected in a glass system.
Fig. 8 Fraction of bond angle distribution for the O-Eu-O, O-Si-O and Si-O-Si bonds of the S79N19 simulated glass system.
Figure 9 compares the structural factors from the simulation of neutron and x-ray scattering for all simulated systems. Both sets of results reproduce the shape of the x-ray and neutron experimental scattering presented by Qin et al. and Jin et al. respectively [38,39]. It is interesting to note that in both spectra types there are changes in peak intensity and position that are related to Na2O concentration. Except for the second peak of the X-ray, every peak increases in intensity and shifts towards lower Q values as sodium oxide decreases (see arrows in Fig. 4). This is due to the impact of Na2O concentration on the bond distance of Si-O, O—O, Eu-O as explained by Johnson and colleagues for praseodymium and europium doped sodium silicate glass [40].
Fig. 9 (a) X-ray diffraction and (b) neutron diffraction for all simulated glass samples. Note the change in the first peak of the two graphs. Decreasing Na2O concentration in the glass matrix reduces the intensity of the first peak and causes a shift to lower Q.
Figure 10 shows the fractions of non-bridging and bridging oxygen in each simulated system as a function of Na2O concentration. These were calculated assuming a cutoff radius of 3 Å. It is well known that sodium oxides act as network modifiers [41] that break Si-O-Si bonds and create NBO (Si-O-—Na+). Thus, the linear decrease in NBO and sodium oxide concentration is correctly reproduced by molecular dynamic simulations. Furthermore, our molecular dynamic simulations reproduce the experimental scenario in which the luminescence gain in Eu3+ ions embedded in sodium silicate glasses occurs as Na2O concentration decreases.
The concentration of the various silicon Qn species is generally used to characterize the intermediate range structure of glasses. The fraction of the silicon Qn distribution was calculated and is shown in Fig. 11 as a function of the silicon oxide concentration in the glass matrix. Q4 represents the fraction of silicon atoms connected with just bridging oxygen, Q3 with one type of non-bridging oxygen and 3 bridging, Q2 with two types of non-bridging oxygen and two bridging, Q1 with one type of bridging oxygen and 3 types of non-bridging and Q0 with 4 or more non-bridging oxygen. The fraction of non-bridging oxygen decreases with sodium oxide concentration as shown in Fig. 10. Thus all Qn distributions that have NBO connected to silicon atoms should also decrease. The fraction of each Qn in our simulation accurately reproduced the one presented by Prasada and colleagues, which was obtained by reported values from NMR and Raman spectroscopy [42]. The broad distribution for the low concentration of silicon oxides is due to the presence of sodium that de-polymerizes the silicon network, creating non-bridging oxygen.
Fig. 11 Fraction of Qn, n = 0,1,2,3,4 as function of silicon oxide concentration in the matrix. Q4 represents the number of silicon atoms connected with only oxygen type bridges, Q3 represents the number of silicon atoms connected with one type of non-bridge oxygen, Q2 represents the number of silicon atoms connected with two types of non-bridge oxygen, and so on. As expected the number of non-bridge oxygen decreases with increasing SiO2 concentration in the glass composition.
5. Discussion
Our results indicate that two facts are connected to the increase in the lifetime of the 5D0 states of the Eu3+ ions. First, nonradiative energy transfer from Eu3+ ions to OH radicals (3260 cm−1) decreases and second, the Q4 configuration of the SiO4 tetrahedral (1170 cm−1), (Fig. 12 schematic). Initially, energy transfer to OH bonded species decreases, which is indicated by the decrease in integrated intensity of the band at 3260 cm−1 as Na2O concentration decreases (Fig. 5). Consequently, radiative transition from 5D0 to 7F2 of the Eu3+ ions increases, which induces lifetime growth of the 5D0 state of the Eu3+ ions. On the other hand, at low Na2O concentrations, we observe a possible energy transfer from Eu3+ ions to the Q4 configuration of the SiO4 tetrahedral in the host [11]. This happens because the fourteenth harmonic vibrational mode of Q4 is resonant with the luminescence transition of the Eu3+ ions centered at 611 nm (16366.6 cm−1), (Fig. 12(b) schematic). Thus, decreasing Na2O concentration reduces absorbed OH density and Q4 growth, which requires different gain rates for non-radiative energy transfer. However, resonance between the Q4 structures and the Eur3+ ions occurs at a higher order of vibration and thus does not cause luminescence quenching and explains the linear lifetime increases (Fig. 4). These results are in agreement with our molecular dynamic simulations where the experimental lifetime gain of the Eu3+ ions embedded in sodium silicate glasses were related to the decrease in NBOs.
Fig. 12 Graphical representation of non-radiative energy transfer from Eu3+ ions to bonded OH. Transfer to the SiO4 (Q4) structure is negligible because it requires a strong field interaction while the Rare-Earth ions have been weakened by 4f-4f transitions located below the 5s and 5p shells.
6. Conclusions
Our results clearly show that lowering Na2O concentration increases the lifetime of the 5D0 state of the Eu3+ ions when embedded in sodium silicate glasses. This lifetime increased linearly with decreasing sodium oxide concentration and was explained by the possible decrease in non-radiative energy transfer from Eu3+ ions to OH radicals. Resonance exists between this transition and the SiO4 structure (Q4) as sodium oxide concentration decreases. However, this resonance does not affect lifetime gain since the phonon decay produced by resonance with high order harmonic vibration requires a strong-field interaction. This is not the case for Eu3+ ions since the 4f electrons are shielded by the 5p and 6p orbitals. The sample synthesized with a high concentration of silicon oxide had a transition lifetime at 611 nm of about 3.3 ms. This is similar to that of crystal systems, such as YAG (3.46 ms).
By using a large simulation cell we were able to study the effect of sodium oxide concentration on Eu3+-doped sodium silicate glass at a low doping level (~0.2 mol%) using molecular dynamic simulations. The results show that a decrease in sodium oxide concentration affects the glass network structure by decreasing the number of non-bridging oxygen species and the Q0,1,2,3 configurations of the SiO4 tetrahedral and consequently reducing the preferential sites for OH radical absorption. This decreases the possibility of non-radiative transition from Eu3+ ions to the OH radical, which in turn explains the lifetime gain of the luminescence transition from the 5D0 to the 7F2 state obtained experimentally in this study. We believe that our results may inspire further research on these systems for optical devices applications.
Acknowledgments
The authors acknowledge financial support from the Brazilian agencies CNPq, FAPEMIG and CAPES.
References and links
1. P. Krishnapuram, S. K. Jakka, C. Thummala, and R. M. Lalapeta, “Photoluminescence characteristics of Eu2O3 doped calcium fluoroborate glasses,” J. Mol. Struct. 1028, 170–175 (2012). [CrossRef]
2. B. Han, H. Liang, H. Ni, Q. Su, G. Yang, J. Shi, and G. Zhang, “Intense red light emission of Eu3+-doped LiGd(PO3)4 for mercury-free lamps and plasma display panels application,” Opt. Express 17(9), 7138–7144 (2009). [CrossRef] [PubMed]
3. H. Guo, H. Zhang, R. Wei, M. Zheng, and L. Zhang, “Preparation, structural and luminescent properties of Ba2Gd2Si4O13:Eu3+ for white LEDs,” Opt. Express 19(S2Suppl 2), A201–A206 (2011). [CrossRef] [PubMed]
4. R. Peacock, “The intensities of lanthanide f ↔ f transitions,” in Rare Earths (Springer Berlin Heidelberg, 1975), pp. 83–122.
5. E. W. J. L. Oomen and A. M. A. van Dongen, “Europium (III) in oxide glasses: dependence of the emission spectrum upon glass composition,” J. Non-Cryst. Solids 111(2-3), 205–213 (1989). [CrossRef]
6. M. Kumar, T. K. Seshagiri, and S. V. Godbole, “Fluorescence lifetime and Judd–Ofelt parameters of Eu3+ doped SrBPO5,” Physica B 410, 141–146 (2013). [CrossRef]
7. E. Snoeks, P. G. Kik, and A. Polman, “Concentration quenching in erbium implanted alkali silicate glasses,” Opt. Mater. 5(3), 159–167 (1996). [CrossRef]
8. N. Vijaya and C. K. Jayasankar, “Structural and spectroscopic properties of Eu3+-doped zinc fluorophosphate glasses,” J. Mol. Struct. 1036, 42–50 (2013). [CrossRef]
9. Z. Hao, J. Zhang, X. Zhang, and X. Wang, “CaSc2O4:Eu3+: a tunable full-color emitting phosphor for white light emitting diodes,” Opt. Mater. 33(3), 355–358 (2011). [CrossRef]
10. H. Desirena, E. De la Rosa, L. A. Díaz-Torres, and G. A. Kumar, “Concentration effect of Er3+ ion on the spectroscopic properties of Er3+ and Yb3+/Er3+ co-doped phosphate glasses,” Opt. Mater. 28(5), 560–568 (2006). [CrossRef]
11. K. K. Pukhov, T. T. Basiev, Y. V. Orlovskii, and M. Glasbeek, “Multiphonon relaxation of the electronic excitation energy of rare-earth ions in laser crystals,” J. Lumin. 76–77, 586–590 (1998). [CrossRef]
12. Y. Zhu, W. Xu, H. Zhang, W. Wang, L. Tong, S. Xu, Z. Sun, and H. Song, “Highly modified spontaneous emissions in YVO4:Er3+ inverse opal and refractive index sensing application,” Appl. Phys. Lett. 100(8), 081104 (2012). [CrossRef]
13. M. Dejneka, E. Snitzer, and R. E. Riman, “Blue, green and red fluorescence and energy transfer of Eu3+ in fluoride glasses,” J. Lumin. 65(5), 227–245 (1995). [CrossRef]
14. M. C. Paul, S. Bysakh, S. Das, S. K. Bhadra, M. Pal, S. Yoo, M. P. Kalita, A. J. Boyland, and J. K. Sahu, “Yb2O3-doped YAG nano-crystallites in silica-based core glass matrix of optical fiber preform,” Mater. Sci. Eng. B 175(2), 108–119 (2010). [CrossRef]
15. G. Cormier, J. A. Capobianco, and A. Monteil, “Molecular dynamics simulation of the trivalent europium ion doped in silica and sodium disilicate glasses,” J. Non-Cryst. Solids 152(2-3), 225–236 (1993). [CrossRef]
16. N. D. Afify and G. Mountjoy, “Molecular-dynamics modeling of Eu3+-ion clustering in SiO2 glass,” Phys. Rev. B 79(2), 024202 (2009). [CrossRef]
17. J. Du and A. N. Cormack, “The structure of erbium doped sodium silicate glasses,” J. Non-Cryst. Solids 351(27-29), 2263–2276 (2005). [CrossRef]
18. L. Kokou and J. Du, “Rare earth ion clustering behavior in europium doped silicate glasses: Simulation size and glass structure effect,” J. Non-Cryst. Solids 358(24), 3408–3417 (2012). [CrossRef]
19. I. T. Todorov, W. Smith, K. Trachenko, and M. T. Dove, “DL_POLY_3: new dimensions in molecular dynamics simulations via massive parallelism,” J. Mater. Chem. 16(20), 1911–1918 (2006). [CrossRef]
20. H. J. C. Berendsen, J. P. M. Postma, W. F. van Gunsteren, A. DiNola, and J. R. Haak, “Molecular dynamics with coupling to an external bath,” J. Chem. Phys. 81(8), 3684–3690 (1984). [CrossRef]
21. J. Du and L. Kokou, “Europium environment and clustering in europium doped silica and sodium silicate glasses,” J. Non-Cryst. Solids 357(11-13), 2235–2240 (2011). [CrossRef]
22. L. Martínez, R. Andrade, E. G. Birgin, and J. M. Martínez, “PACKMOL: a package for building initial configurations for molecular dynamics simulations,” J. Comput. Chem. 30(13), 2157–2164 (2009). [CrossRef] [PubMed]
23. Y. Bai, Y. Wang, K. Yang, X. Zhang, G. Peng, Y. Song, Z. Pan, and C. H. Wang, “The effect of Li on the spectrum of Er3+ in Li- and Er-codoped ZnO nanocrystals,” J. Phys. Chem. C 112(32), 12259–12263 (2008). [CrossRef]
24. W. T. Carnall, H. Crosswhite, and H. M. Crosswhite, “Energy level structure and transition probabilities in the spectra of the trivalent lanthanides in LaF3,” in Argonne National Laboratory Report(1977).
25. K. K. Haldar and A. Patra, “Fluorescence enhancement and quenching of Eu3+ ions by Au–ZnO core-shell and Au nanoparticles,” Appl. Phys. Lett. 95(6), 063103 (2009). [CrossRef]
26. O. Humbach, H. Fabian, U. Grzesik, U. Haken, and W. Heitmann, “Analysis of OH absorption bands in synthetic silica,” J. Non-Cryst. Solids 203, 19–26 (1996). [CrossRef]
27. J. Nawrocki, “The silanol group and its role in liquid chromatography,” J. Chromatogr. A 779(1-2), 29–71 (1997). [CrossRef]
28. S. N. B. Bhaktha, F. Beclin, M. Bouazaoui, B. Capoen, A. Chiasera, M. Ferrari, C. Kinowski, G. C. Righini, O. Robbe, and S. Turrell, “Enhanced fluorescence from Eu3+ in low-loss silica glass-ceramic waveguides with high SnO2 content,” Appl. Phys. Lett. 93(21), 211904 (2008). [CrossRef]
29. S. W. Park, H. Kyoung Yang, J. Won Chung, Y. Chen, B. Kee Moon, B. Chun Choi, J. H. Jeong, and J. Hwan Kim, “Photoluminescent properties of LaVO4:Eu3+ by structural transformation,” Physica B 405(18), 4040–4044 (2010). [CrossRef]
30. H. C. Jung, J. Y. Park, G. Seeta Rama Raju, J. H. Jeong, B. K. Moon, J. H. Kim, and H. Y. Choi, “Crystalline structure dependence of luminescent properties of Eu3+-activated Y2O3–Al2O3 system phosphors,” Curr. Appl. Phys. 9(3), S217–S221 (2009). [CrossRef]
31. A. M. B. Silva, C. M. Queiroz, S. Agathopoulos, R. N. Correia, M. H. V. Fernandes, and J. M. Oliveira, “Structure of SiO2–MgO–Na2O glasses by FTIR, Raman and 29Si MAS NMR,” J. Mol. Struct. 986(1-3), 16–21 (2011). [CrossRef]
32. B. O. Mysen and J. D. Frantz, “Silicate melts at magmatic temperatures: in-situ structure determination to 1651°C and effect of temperature and bulk composition on the mixing behavior of structural units,” Contrib. Mineral. Petrol. 117(1), 1–14 (1994). [CrossRef]
33. A. G. Kalampounias, “IR and Raman spectroscopic studies of sol–gel derived alkaline-earth,” Bull. Mater. Sci. 34(2), 299–303 (2011). [CrossRef]
34. E. Potapova, M. Grahn, A. Holmgren, and J. Hedlund, “The effect of calcium ions and sodium silicate on the adsorption of a model anionic flotation collector on magnetite studied by ATR-FTIR spectroscopy,” J. Colloid Interface Sci. 345(1), 96–102 (2010). [CrossRef] [PubMed]
35. J. Yang, S. Dai, N. Dai, L. Wen, L. Hu, and Z. Jiang, “Investigation on nonradiative decay of 4I13/2→4I15/2 transition of Er3+-doped oxide glasses,” J. Lumin. 106(1), 9–14 (2004). [CrossRef]
36. J. Bajer and A. Miranowicz, “Quantum versus classical descriptions of sub-Poissonian light generation in three-wave mixing,” J. Opt. B Quantum Semiclassical Opt. 3(4), 251–259 (2001). [CrossRef]
37. G. Liu and B. Jacquier, “Spectroscopic properties of rare earths in optical materials,” in Spectroscopic Properties of Rare Earths in Optical Materials (Springer, 2005), p. 550.
38. T. Qin, G. Mountjoy, N. Afify, M. Reid, Y. Yeung, A. Speghini, and M. Bettinelli, “Link between optical spectra, crystal-field parameters, and local environments of Eu3+ ions in Eu2O3-doped sodium disilicate glass,” Phys. Rev. B 84(10), 104206 (2011). [CrossRef]
39. J. Jin, T. Yoko, F. Miyaji, S. Sakka, T. Fukunaga, and M. Misawa, “Neutron diffraction study on the structure of Na-Si-O-N oxynitride glasses,” J. Am. Ceram. Soc. 76(3), 630–634 (1993). [CrossRef]
40. J. A. Johnson, C. J. Benmore, D. Holland, J. Du, B. Beuneu, and A. Mekki, “Influence of rare-earth ions on SiO2-Na2O-RE2O3 glass structure,” J. Phys. Condens. Matter 23(6), 065404 (2011). [CrossRef] [PubMed]
41. J. Zarzycki, W. D. Scott, and C. Massart, Glasses and the Vitreous State (Cambridge University Press, 1991).
42. R. Prasada Rao, T. D. Tho, and S. Adams, “Ion transport pathways in molecular dynamics simulated alkali silicate glassy electrolytes,” Solid State Ion. 192(1), 25–29 (2011). [CrossRef]
