Open Access
Add to BibSonomy
Abstract
We successfully demonstrate new heavy metal-gallium-doped oxide glass media, which are capable of accommodating a significant level of Er3+ doping, while avoiding the deleterious effects of concentration quenching. We examine the effects of composition and microstructure of the glass networks on optical properties. Near-infrared and visible emission demonstrate the absence of concentration quenching and Er3+ clustering. Both Raman spectroscopy and X-ray absorption fine structure spectroscopy confirm that gallium enters the glass as a tetrahedral network former. The incorporation of gallium into the glass modifies the energy landscape and creates two distinct crystal field environments, which promote Er3+ radiative transitions.
© 2019 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Erbium-doped glasses can act as functional materials at more than one emission wavelength. They function as gain media in fiber lasers and amplifiers at the ~1.5 μm emission, the standard telecommunications wavelength that falls in the range of minimum transmission loss for silica-based optical fibers [1]. The green Er3+ emission (~550 nm) is used for compact lasers devices and for displays, while its near infrared emissions (~800 nm and ~975 nm) are suitable for diode pumping lasers [2,3]. Achieving optical gain, in any of the aforementioned Er3+ emission regions would yield an improvement in the performance of optical devices.
The constant need for miniaturization of optical devices requires high erbium concentrations on the order of a weight percent, to achieve sufficient pump-light absorption and net gain. It is known that rare-earths (RE) have a very low solubility in silicate glasses [4–7]. This constitutes an important technological obstacle for the improvement of RE-doped optical devices. One of the major hurdles is that, already at very modest concentrations, Er3+ ions tend to agglomerate and form clusters, leading to ‘concentration quenching’, which considerably decreases the excited state lifetime and degrades the emission output. In a pure silica glass, which is still the predominant glass former [4–7] used for RE-doped optical devices, significant Er3+ - Er3+ interactions already start at extremely low concentrations (~100 ppm) and degrade the luminescence quantum yield, through cooperative upconversion and excited-state absorption.
Besides concentration, the performance of erbium-doped glass materials strongly depends on the composition of the host matrix and on the nature of dopants, which together determine the atomic scale microstructure around Er3+ and influence the end optical properties. In terms of composition, heavy-metal oxide glasses, such as tellurite, bismuthate or germanate, are of high interest for optical applications owing to their good solubility for rare-earth ions, low phonon energy, high refractive index, high glass transition temperature, and good infrared (IR) transmission [8–14], all of which are desirable characteristics for applications in high-power fibers lasers and amplifiers. Recent studies have found that the incorporation of gallium as a modifier in a chalcogenide glassy network can dramatically increase the solubility of RE ions [15–17].
We synthesized and investigated 3 series of isostructural, but chemically different glass media and we tuned the material composition to find trade-offs between glass network microstructures and efficient erbium emission. All glass series are based on silica, germania, or their mixture, contain a heavy-metal dopant, gallium, and are co-doped with a variable amount of Er3+. The objective was to lower the phonon energy with respect to pure SiO2 glass in order to enhance erbium radiative transitions, while maintaining a non-esoteric chemistry that is compatible with silica-based optoelectronic devices. For characterization of the fabricated glass materials we use a multi-technique approach. Emission spectroscopy provides information about the optically active species, i.e. erbium. Raman spectroscopy focuses on glass architecture such as type and connectivity of main glass network building blocks. Finally, to clarify the microstructural role and to study the local environment of the heavy-metal and of the rare-earth dopant we use synchrotron x-ray absorption spectroscopy. Such multi-technique diagnostic allows us to link the chemistry, the microstructure and the resulting materials’ optical properties.
2. Experimental details
2.1 Synthesis of glass materials
We selected three isostructural oxide glass systems. The first is comprised of silica-based glasses doped with gallium and co-doped with Er3+, the second system is comprised of germanate glasses doped with heavy-metal, gallium and co-doped with Er3+ while the third system is comprised of mixed silica-germanate glasses doped with gallium and co-doped with Er3+. The three oxide glass systems are referred to in this work as: the ‘(Si)/Ga-series’ with a composition (80SiO2)-(12-x)Ga2O3-8K2O-xEr2O3 mol% where x = 0.0, 0.05, 0.5, 1.0, 2.0, 3.0 mol%; the ‘(Ge)/Ga-series’ with a composition (80GeO2)-(12-x)Ga2O3-8K2O-xEr2O3 mol % where x = 0.0, 0.5, 1.0 mol%; and finally the ‘(Si/Ge)/Ga-series’ with a composition (40SiO2/40GeO2)-(12-x)Ga2O3-8K2O-xEr2O3 mol% where x = 0.0, 0.5, 1.0 mol%. Parentheses indicate the main glass formers. Mixtures of the initial batch components are prepared using analytical grade reagents (99.999% purity) to avoid any spurious effects of unwanted luminescent or other contaminants, found in lower purity reagents. We take several measures to ensure good glass homogeneity. The batch components are thoroughly ground, mixed, sintered, re-ground and re-sintered at low temperature and afterwards melted at high temperature (1500°C) in alumina crucibles, in an air atmosphere, using a research grade programmable furnace. The liquid glass melt is rapidly air cooled to obtain a glass. The fabricated glasses are annealed at low temperature, below the glass transition, to remove internal stresses. For optical and microstructural studies bulk samples are cut into rectangles and polished.
2.2 Characterization of glass materials
For the characterization of the glass materials we use a multi-technique approach. Raman spectra are measured using a confocal spectrometer (LabRam HR Evolution) equipped with 1800 lines/mm grating and a 442 nm line of a continuous wavelength laser. All spectra are measured in a backscattering configuration. The incoming and scattered light is focused and collected using an Olympus microscope (BX41) with x100 (NA = 0.95) microscope objective, leading to a theoretical diameter of focal spot of 0.6 μm and an adjustable length of the focal region from 3 μm to several mm. A CCD detector (1024 x 256 pix) is used for recording the spectra in the range between 100 cm−1 and 1200 cm−1 with a spectral resolution of ~1 cm−1.
To generate emission in the visible range, a confocal set-up used a 442 nm laser excitation, a 600 lines/mm grating and a CCD (1024 x 256 pix) detector. The emission signal was measured in a backscattering configuration using an x100 objective of a confocal microscope leading to a focal spot of 0.6 μm.
To investigate the near-infrared emission range (NIR) we used a computer controlled, modular spectrometer (FLS 980). The instrument is capable of measuring a steady state emission spectrum in the ultraviolet to near infrared spectral range with single photon counting sensitivity. The excitation source was a xenon lamp at 645 nm and the signal was measured in (90°) configuration with an InGaAs detector for the NIR range.
The local environment around erbium or gallium atoms was studied using synchrotron radiation-based x-ray absorption spectroscopy: x-ray absorption near-edge structure (XANES). The spectra were recorded at the Er L3 edge. Experiments were carried out at the facilities of the MRCAT (Sector 10) insertion device beamline at the Advanced Photon Source, Argonne National Laboratory (Chicago, IL). The samples were placed at 45° to the incident beam with a fluorescence ion chamber located at 90° from the incident beam. Data are collected in continuous scan mode with at 5~10 scans taken per sample. Using the IFEFFIT-based Athena and Artemis programs [18–20], the data is merged and normalized for XANES analysis.
3. Results and discussion
The three series of glasses investigated in this work are doped by the heavy-metal gallium, and co-doped by Er3+, with the key difference being the type of the main glass network former in each: SiO2 in glasses named ‘(Si)/Ga-series’, GeO2 in glasses named the ‘(Ge)/Ga-series’ and finally a mixture of SiO2 and GeO2 in glasses named the ‘(Si/Ge)/Ga-series’. Our three series of glasses embody the essence of excellent glass network formers and compatible dopants, offering similar microstructures (or isostructural compositions), and are simple enough to search for the basic mechanisms ruling optical properties. In the first section we use emission spectroscopy to illustrate how the heavy-metal-doped glass matrices enable a quench-free enhancement of Er3+ emission. In the next section we use Raman spectroscopy to describe the microstructural role of the glass network formers. In the last section we highlight the microstructural role of the heavy-metal and its influence on the emission of the luminescent dopant, with synchrotron x-ray absorption spectroscopy.
3.1 Quench-free enhancement in emission of erbium
Emission studies were carried out in function of erbium concentration and of glass network composition. For Er3+-doped glasses there are two spectral ranges of interest: near-infrared (NIR, Fig. 1(a)-1(e)) and visible (VIS, Fig. 2(a)-2(d)). The observed luminescence bands, which correspond to the electronic intra-4f transitions of Er3+ from the ground state energy level, are located at the ~1.5 µm telecom emission (4I13/2→4I15/2, Fig. 1), at ~550 nm (2H11/2/4S3/2→4I15/2,), at ~670 nm (4F9/2→4I15/2), at ~850 nm (4I9/2→4I15/2), and finally at ~970 nm (4I11/2→4I15/2, Fig. 2).
Fig. 1 Er3+ telecom emission at ~1.5 μm (4 I 13/2 →4 I 15/2) as a function of Er2O3 concentration for 0.5 and 1.0 mol% for the: (a) ‘(Si)/Ga - series’ (b) the ‘(Si/Ge)/Ga - series’ and (c) the ‘(Ge)/Ga - series’; (d) emission vs. the full span of Er2O3 concentrations for the ‘(Si)/Ga - series’; (e) comparison of the telecom emission intensity vs Er2O3 concentration for all glass series. Excitation was at 645 nm. Dashed lines are guides for the eye. Error bars are smaller than symbols.
Fig. 2 Comparison of Er3+ emission as a function of concentration for the 3 glass matrices with (a) 0.5 mol% of Er2O3 doping, (b) 1.0 mol% doping; (c) intensity of the green Er3+ emission at 547 nm (4S3/2→4I15/2) vs. Er2O3 concentration for all series; (d) Er3+ energy diagram. Excitation was at 442 nm. Error bars are smaller than symbols.
With erbium concentration increase, the erbium emission intensity increases in all glass series (Fig. 1(a)-1(e)). The ‘(Si)/Ga-series’ was synthesized with a wider dopant range comparing to the other two series. The 1.5 μm emission in the ‘(Si)/Ga-series’ shows a close to linear increase of erbium emission intensity as a function Er2O3 doping up to 3 mol% (Fig. 1(d) and 1(e)). A similar behavior is observed for the mixed ‘(Si/Ge)/Ga-series’ and the ‘(Ge)/Ga-series’ with Er2O3 doping up to 1 mol% (Fig. 1(b), 1(c) and 1(e)).
Based on the comparison of our glasses in the NIR range (Fig. 1), we can draw the following conclusions. In the present work the levels of Er2O3 doping without emission quenching are better than those reported for other oxide glasses, such as alumino-silicate, phosphate or chalcogenide glasses [17,21–23]. For the ‘(Si)/Ga-series’ at the ~1.5 μm telecom emission there is no sign of luminescence quenching up to 3 mol% of Er2O3 (Fig. 1(e)). The other two sets: ‘(Ge)/Ga-series’ and ‘(Si/Ge)/Ga-series’ were studied up to 1 mol% of doping and they do not display any NIR emission quenching either.
This is a noteworthy outcome and since the close-to-linear trend of increasing intensity does not show signs of tapering off (Fig. 1(e)), it could imply that higher Er2O3 doping could be accepted in all 3-glass series. In comparing the erbium emission intensity for the same Er2O3 doping level (Fig. 1(a)-1(c)), it is apparent that the ‘(Si/Ge)/Ga-series’ shows the most intense 1.5 μm emission. In addition, the ‘(Si/Ge)/Ga-series’ exhibits the largest spectral width in the NIR, ~55 nm (1 mol% of Er2O3), compared to ~49 and ~53 nm for the ‘(Si)/Ga-series’ and the ‘(Ge)/Ga-series’. These findings make the ‘(Si/Ge)/Ga-series’ a compositional ‘sweet spot’ between a silicate and a germanate glass network and make this composition the best host matrix for augmented Er-doping. Such high FWHM values [24] are desirable for tunable solid-state lasers and broadband optical amplifiers.
In the VIS range (Fig. 2(a)-2(c)) an examination of the strong green emission band at ~550 nm (4S3/2→4I15/2, Fig. 2(d)), reveals that in the ‘(Si)/Ga-series’, emission initially slowly increases up to 2 mol% of Er2O3. and then drops at higher doping, indicating the onset of concentration quenching. At equal Er2O3 doping levels the green emission band is 6 and 3 times more intense for ‘(Ge)/Ga-series’ and ‘(Si/Ge)/Ga-series’, respectively, than that for the ‘(Si)/Ga-series’ (Fig. 2(a) and 2(b)). The 2 ‘heaviest’ glass matrices, the ‘(Ge)/Ga-series’ and the ‘(Si/Ge)/Ga-series’ are the best Er3+ hosts since they do not display any signs of erbium concentration quenching in VIS. The linear increase of emission intensity as a function of erbium concentration (Fig. 2(c)) once again implies that more Er3+ doping would likely be accepted in those glasses without emission quenching in the VIS.
Among the proposed strategies that could reduce the concentration quenching effect, co-doping by aluminum (Al3+) was reported as a particularly promising approach [12]. A model based on theoretical calculations on Er-doped alumino-silicate glass, by Lægsgaard [25] proposes that Er clustering can be counteracted by the formation of Er-Al complexes. Results of numerical simulations by Monteil [26] for Er-doped and Al-co-doped silica glass support the idea of this affinity and suggest that RE ions prefer Al-rich domains in glass. Arai and associates [27] introduced the “solvation-shell” model based on assumption that the Al co-dopant atoms form some kind of “solvation shell” around the RE atoms. Gallium (Ga3+) is known to isostructurally substitute for aluminum (Al3+) in oxide glasses [28] and it is likely that – in our glasses – it would play the same structural role as aluminum does above. According to recent works on chalcogenide glasses, a gallium additive can improve significantly the solubility of rare-earths in those glasses [14–17]. Our emission studies (Fig. 1) at 1.5 µm show the absence of Er emission quenching up to the highest Er concentration of 3 mol% of Er2O3 (or 6 mol% of Er3+). This allows us to hypothesize that the addition of Ga to the silica-based glass benefits Er emission properties by counteracting Er clustering. In fact, in a separate work on x-ray absorption fine structure spectroscopy of similar glasses we demonstrate that Er3+ localizes preferentially in Ga3+-rich domains within the glass network, and that Er clustering is offset by the formation of Ga-O-Er complexes [29].
In the next two sections we examine the glass network microstructure (section 3.2) and the role of gallium in modifying the local energy environment of Er3+ (section 3.3).
3.2 Understanding the glass network architecture: Raman spectroscopy
To understand the enhancement of emission in our glass materials we turn to Raman spectroscopy as a tool to develop a description of the glass’ local microstructure. In fact, Raman spectra are sensitive to variations in the composition of the glass network, to changes in the type of glass forming units (polyhedra: triangles, tetrahedra etc.) as well as to the degree of network polymerization i.e. glass “tetrahedral ring structures”.
3.2.1 Tetrahedral ring structures in glass networks
The glass materials fabricated in this project belong to the oxide family and we can categorize them as classic network glasses. They are made of building blocks of well-ordered tetrahedra (SiO4, GeO4) linked together by bridging oxygens, to form a 3-dimensional, disordered network. The majority of tetrahedra in such glasses are connected to form 6-membered structures, but bigger and smaller rings are also present [30–35].
A comparison in Fig. 3 shows one of our glasses from the ‘(Si)/Ga-series’ (doped with 1 mol % of Er2O3) to a reference SiO2 glass. The deconvolution of Raman spectra illustrates how the spectral shape of the main Raman band (~450 cm−1) can be used to extract information about the average ring structure. It is apparent that 6-membered and larger tetrahedral rings dominate for both glasses, but our heavy-metal doped, erbium co-doped glass has a network built of fewer of the larger and more of the smaller than 6-fold rings (Fig. 3(b) and 3(d)).
Fig. 3 Example of quantitative analysis (Lorentzian and Gaussian convolutions) of ring statistics illustrating the diversity of tetrahedral structures, from 3-fold to >6-fold rings: (a) a reference SiO2 glass with (b) its ring structure distribution in the network, and (c) a glass from the ‘(Si)/Ga - series’ synthesized in this work with (d) its ring structure distribution in the network. (e) Schematic representation of different ring structures, where colors correspond to the x-fold tetrahedral rings.
3.2.2 Changes in glass microstructure
Raman bands are associated with the vibrational density of states of glass forming units - network building blocks - and their widths reflect the degree of structural disorder specific to a given glass composition.
Figure 4(a)-4(c) shows Raman spectra of the ‘(Si)/Ga-series’, the ‘(Ge)/Ga-series’, and the mixed ‘(Si/Ge)/Ga-series’ of glasses synthesized in this work. A GeO2 glass network is structurally compatible to a SiO2 network, allowing the mixing of these two glass formers, and this is what occurs in the ‘(Si/Ge)/Ga-series’. The Raman band positions in the spectra of our glass series can be compared to those of reference SiO2 and GeO2 glasses (Fig. 4(d)-4(f)). Extensive discussion of Raman mode assignments of reference glasses can be found in the literature [30–36]. A summary of Raman band assignments for our glasses is given in Table 1 along with those of reference glasses. Here we discuss the key features that can be extracted from mode assignments. In the ‘(Si)/Ga-series’, the overall spectral shape of Raman bands is similar to that of pure SiO2 glass, in the ‘(Ge)/Ga-series’ the bands are comparable to that of pure GeO2 glass and in the mixed ‘(Si/Ge)/Ga-series’ the bands reflect the mixing of the SiO4 and GeO4 tetrahedral units. Our previous works on mixed SiO2-GeO2 glass matrices [37,38] have shown that Raman spectra will more resemble a pure GeO2 or a pure SiO2 glass, depending on which glass network former is predominant.
Fig. 4 Raman spectra of the heavy-metal doped (gallium) and erbium co-doped (a) ‘(Si)/Ga-series’, (b) ‘(Si/Ge)/Ga-series’ and (c) ‘(Ge)/Ga-series’ glasses synthesized in this work. The spectra illustrate the change of width of the main glass band (M) with increasing erbium concentration, indicating an increase in network disorder. For all glass series the D1 and D2 defect bands increase in intensity with erbium concentration, indicating an increase in the average number of smaller rings. Raman spectra of reference glasses are shown for comparison: (d) pure SiO2 glass, (e) 1:1 mix SiO2-GeO2, and (f) pure GeO2 glass.
Table 1. Optical Raman band assignments for the ‘(Si)/Ga-series’, the ‘(Ge)/Ga-series’, and the mixed ‘(Si/Ge)/Ga-series’ of glasses synthesized in this work as well as reference SiO2 and GeO2 glasses.
For glasses with 3-dimensional tetrahedral networks, the interpretation of Raman bands usually refers to the vibrational modes of the tetrahedral units forming the network. It follows that the spectra of all glasses in Fig. 4 are dominated by a broad, intense, asymmetric band, called the main glass band (M), which reflects symmetric stretching motions of bridging oxygens (T-O-T, where T = Si, Ge) in predominantly 6-membered SiO4 tetrahedral ring structures. The main glass band moves from ~445 cm−1 (‘(Si)/Ga-series’ and (‘(Si/Ge)/Ga-series’) to ~435 cm−1 (‘(Ge)/Ga-series’) and progressively narrows as the GeO2 content in the network increases, due to a decrease in the mean value and the width of the T–O–T bond- and intertetrahedral-angle distribution (where T = Si, Ge). Overlapped with the main glass band is a sharp D1 ‘defect’ band, assigned to the stretching motions of bridging oxygens in 4-membered TO4 (T = Si, Ge) rings. D1 shifts from ~480 cm−1 (‘(Si)/Ga-series’) to ~465 cm−1 (‘(Si)/Ga-series’) upon substitution of silicon by heavier germanium.
Following the mode assignment for pure GeO2 glass, D1 in our ‘(Ge)/Ga-series’, located at ~340 cm−1 is assigned to Ge-deformation motions. The second glass ‘defect’ band, D2, is assigned to the stretching motions of bridging oxygens in 3-membered TO4 (T = Si, Ge) rings. With increased GeO2 content D2 moves from ~565 cm−1 (‘(Si)/Ga-series’) to ~520 cm−1 (‘(Si/Ge)/Ga-series’) and its intensity increases relative to the M band. In the region of ~600~800 cm−1 the broad bands are due to the transverse optical (TO) and longitudinal optical (LO) split of bending (deformation) modes of the T–O–T bridging oxygens, where T = Si, Ge. They shift to lower frequencies from ~800 cm−1 for the ‘(Si)/Ga-series’ to 600 cm−1 for the mixed ‘(Si/Ge)/Ga-series’. In the region 850~1200 cm−1 the broad bands encompass a combination of two high-frequency bands (TO/LO split) of asymmetric stretching of the T–O–T bridging oxygens. This spectral region contains information about the mixing of the SiO4 and GeO4 vibrational modes. The TO/LO split in ‘(Si)/Ga-series’ shows up as the weak intensity doublet ~950 cm−1 and ~1050 cm−1 and it transforms into one broad band in the mixed glass compositions and finally appears as a low frequency doublet, at ~840 cm−1 and 930 cm−1, for ‘(Ge)/Ga-series’ of glasses.
The results of Raman spectroscopy can be summarized into a few key findings. The substitution of light silicon by heavy germanium changes the vibrational energy of the glass network and moves the position of all bands to lower frequencies. Remarkably, the main glass band has narrower width in the ‘(Ge)/Ga-series’, indicating a narrower distribution of intra- and inter-tetrahedral angles compared to such distribution in the ‘(Si)/Ga-series’. This is consistent with angle distributions from neutron and X-ray data, where intertetrahedral angles for pure GeO2 glass were ~132° versus ~148–151° in pure SiO2 glass and T–O bond lengths were 1.74 Å versus 1.61 Å [32–36]. The second key finding is that the intensity of D2 relative to M in the ‘(Ge)/Ga-series’ is much stronger than the equivalent D2 band in the ‘(Si)/Ga-series’. Thus our ‘(Ge)/Ga-series’ has a network built of a larger proportion of 3-membered tetrahedral ring structures, relative to the two other series, and this finding is consistent with reports on pure SiO2 and GeO2 glasses. Finally, we note that the mixed ‘(Si/Ge)/Ga-series’ has the largest width of the main glass band, indicating the highest level of structural disorder in the network of this glass series. This is caused by interconnected mixed linkages (Si–O–Si, Ge–O–Ge and Si–O–Ge) resulting in an overlap of SiO4 and GeO4 vibrations.
3.2.3 The role of Ga and Er dopants in the glass network architecture
Using Raman spectroscopy, we find no direct evidence of a six-fold coordinated (octahedral) gallium. Gallium can be a so-called ‘conditional’ glass former, because it does not form a glass by itself, but does so when combined with other suitable network-forming oxides. In an alkali-silica glass system, Si4+ ions can be substituted by Al3+ ions, which are then charge compensated by alkali ions (Na+, K+), thus reducing the number of non-bridging oxygens and allowing for Al tetrahedral coordination with oxygens [28,39–42]. In our glasses the structural role of Ga3+ is comparable to that of Al3+ in silicate glasses and could potentially counteract the rare-earth clustering issue. Gallium may assume more than one type of coordination with oxygens [28], thus acting both as glass network-former and modifier, conditioned by glass composition, and by the availability of charge balancing entities. Further evidence that gallium acts as network former comes from our ongoing studies of the Ga local environment in the present glasses [29] and in our previous investigations of similar glasses [43] using synchrotron x-ray absorption fine structure (XAFS) spectroscopy. We conclude that in our three series of glasses gallium acts as a secondary network former and forms GaO4 tetrahedra, charge-balanced by a network modifier (alkali ion K+).
A comparative analysis of Raman spectra of all synthesized glass series shows that the introduction of the Er3+ dopant in the network does not bring new vibrational bands nor change the overall spectral shape. However, the width of the main glass band increases with increasing Er3+ concentration for each glass series (Fig. 4(a)-4(c)). We interpret this as a larger distribution of inter-tetrahedral angles as a function of Er3+ concentration and indicative of an increase of network disorder. The intensity of D1 and D2 defect bands also increases, as more of the small, 3- and 4-membered rings are created, which points to a gradual densification of the glass network with raising Er2O3 concentration.
3.3 Modifications of Er3+ energy landscape
The optical properties of RE ions depend on the ligand field of the specific host matrix in which these ions are incorporated. In other words, the local environment around Er3+ in a host glass can have a large influence on its optical absorption and emission, which are the key factors controlling the optical performance of the whole material. We turn to results of synchrotron x-ray absorption spectroscopy to study the local environment of the rare-earth dopant. This section presents key results of our x-ray absorption near-edge spectroscopy (XANES) studies, while a comprehensive analysis by extended X-ray absorption fine structure (EXAFS) is reported elsewhere [29].
Figure 5 shows XANES spectra of the Er L3 (2p3/2) edge for the ‘(Si)/Ga-series’. By fitting each absorption edge to an arctangent and a Gaussian white line (Table 2) we conclude that the absorption edge position is unchanged as a function of Er3+ content. The white line is indicative of the number of empty d-states at the Er edge. As shown in Fig. 5 and quantitatively in Table 2, the white line amplitude decreases as more Er3+ is doped into the glass. XANES also shows the presence of isosbestic points, or points that cross the same energy points for the ‘(Si)/Ga-series’ (Fig. 5). Specifically, the presence of isosbestic points demonstrates the formation of two distinct crystal field environments, corresponding to two discrete microstructural erbium environments [29]. Therefore, the role of gallium in the matrix of the ‘(Si)/Ga-series’ is to modify the energy environment in the vicinity of Er3+ ions. It is worthy to note that there is an ongoing discussion about differentiated local crystal fields (in the sense of localized energy sites) for rare earth ions in glasses. The combination of our results of emission spectroscopy and XAFS leads to the conclusion that Er3+ radiative transitions are promoted by the presence of gallium through a structuring of the local environment of Er3+in the glass matrix.
Fig. 5 Example of XANES of the Er L3 (2P3/2) edge for the ‘(Si)/Ga-series’ of glasses, where µ is the absorption coefficient. The ‘white line’ spectral shape illustrates the presence of isosbestic points. The incorporation of gallium into the network results in an alteration of the energy environment and creates 2 distinct erbium microstructural surroundings. The arrow marks the direction of increase (0.05 - 3.0 mol%) of Er2O3 concentration.
Table 2. The results of fitting of the Er L3 white line edge for ‘(Si)/Ga-series’ of glasses, to an arctangent and a Gaussian white line peak.
Regarding the role of gallium, it may assume more than one type of coordination with oxygens [28], thus acting both as glass network-former and modifier, depending on the host glass composition and conditioned by the availability of charge balancing entities and potentially counteract the rare-earth clustering issue. Our previous extended X-ray Absorption Fine Structure (EXAFS) spectroscopy investigations of similar glass systems, performed at the gallium L3-edge, demonstrated that gallium is tetrahedrally coordinated with oxygens [43]. Our ongoing EXAFS studies of the current 3 glass series [29] show that the heavy-metal dopant, gallium, is in four-fold coordination and acts as a secondary glass network former.
4. Conclusions
We successfully demonstrate new, non-esoteric glass media - of practical interest because of their compatibility with silica – that accept high erbium concentrations and display enhanced emission, while limiting Er3+-Er3+ clustering. Put together, both emission and XAFS results demonstrate that the presence of heavy-metal gallium promotes Er3+ radiative transitions through micro-structuring of its local environment. This can reduce or even prevent luminescence concentration quenching. The presence of more than one, distinct crystal field environments of erbium ions could also lead to a wider range of luminescence decay rates and consequently could reduce the detrimental resonant energy transfer mechanisms that lead to loss of pumping energy. The results of this work point the way to designing host glass matrices that limit interactions between neighboring rare-earth ions and thus minimize loss of excitation by energy migration and maximize light output. Detailed knowledge of the relationship between the glass network structure and chemistry, as well as understanding the microstructural and optical role of all components provides a path forward to develop higher performance glass-based optical devices, such as lasers and amplifiers.
Funding
Air Force Office of Scientific Research (AFOSR) (FA9550-16-1-0242); National Science Foundation (NSF CAREER) (1753012); Department of Energy (DoE) (DE-AC02-06CH11357).
Acknowledgements
KL gratefully acknowledges the generous support from the Air Force Office of Scientific Research (AFOSR, Dr. Ali Sayir) under award no. FA9550-16-1-0242. AJA thanks IIT and the National Science Foundation under a CAREER grant no. 1753012 for the generous support to this research. MRCAT operations are supported by the Department of Energy and the MRCAT member institutions. This research used resources of the Advanced Photon Source, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC02-06CH11357.
References
1. A. Polman and F. C. J. M. van Veggel, “Broadband sensitizers for erbium-doped planar optical amplifiers: review,” J. Opt. Soc. Am. B 21(5), 871–892 (2004). [CrossRef]
2. R. F. de Morais, E. Oliveira Serqueira, and N. Oliveira Dantas, “Effect of thermal annealing on the spectroscopic parameters of Er3+-doped sodium silicate glass,” Opt. Mater. 35(12), 2122–2127 (2013). [CrossRef]
3. A. C. Lira, E. Martín-Rodríguez, R. Martínez-Martínez, I. Camarillo, G. H. Muñoz, J. Garcia-Solé, and U. Caldiño, “Spectroscopy of the Bi4Si3O12:Er3+ glass for optical amplification and laser application,” Opt. Mater. 32(9), 1266–1273 (2010). [CrossRef]
4. W. J. Miniscalco, “Erbium-doped glasses for fiber amplifiers at 1500 nm,” J. Lightwave Technol. 9(2), 234–250 (1991). [CrossRef]
5. M. P. Hehlen, T. Gosnell, N. J. Cockroft, and A. Bruce, “Spectroscopic properties of Er3+ and Yb3+ doped soda-lime silicate and aluminosilicate glasses,” Phys. Rev. B Condens. Matter 56(15), 9302–9318 (1997). [CrossRef]
6. A. B. Seddon, Z. Tang, D. Furniss, S. Sujecki, and T. M. Benson, “Progress in rare-earth-doped mid-infrared fiber lasers,” Opt. Express 18(25), 26704–26719 (2010). [CrossRef] [PubMed]
7. S. Sen, R. Rakhmatullin, R. Gubaidullin, and A. Pöppl, “Direct spectroscopic observation of the atomic-scale mechanisms of clustering and homogenization of rare-earth dopant ions in vitreous silica,” Phys. Rev. B Condens. Matter Mater. Phys. 74(10), 100201 (2006). [CrossRef]
8. S. Simon and V. Simon, “Thermal characterization of gallium-bismuthate oxide glasses,” Mater. Lett. 58(29), 3778–3781 (2004). [CrossRef]
9. R. Balda, A. Oleaga, J. Fernández, and J. M. Fdez-Navarro, “Spectroscopy and frequency upconversion of Er3+ ions in lead niobium germanate glasses,” Opt. Mater. 24(1–2), 83–90 (2003). [CrossRef]
10. S. Zhao, X. Wang, D. Fang, S. Xu, and L. Hu, “Spectroscopic properties and thermal stability of Er3+-doped tungsten-tellurite glass for waveguide amplifier application,” J. Alloys Compd. 424(1–2), 243–246 (2006). [CrossRef]
11. G. Gao, L. Hu, H. Fan, G. Wang, K. Li, S. Feng, S. Fan, and H. Chen, “Effect of Bi2O3 on physical, optical and structural properties of boron silicon bismuthate glasses,” Opt. Mater. 32(1), 159–163 (2009). [CrossRef]
12. T. Skopak, P. Hee, Y. Ledemi, M. Dussauze, S. Kroeker, T. Cardinal, E. Fargin, and Y. Messaddeq, “Mixture experimental design applied to gallium-rich GaO3/2-GeO2-NaO1/2 glasses,” J. Non-Cryst. Solids 455, 83–89 (2017). [CrossRef]
13. X. Wen, G. Tang, J. Wang, X. Chen, Q. Qian, and Z. Yang, “Tm3+ doped barium gallo-germanate glass single-mode fibers for 2.0 μm laser,” Opt. Express 23(6), 7722–7731 (2015). [CrossRef] [PubMed]
14. K. Wei, D. P. Machewirth, J. Wenzel, J. E. Snitzer, and G. H. Sigel, “Spectroscopy of Dy3+ in Ge–Ga–S glass and its suitability for 1.3 μ m fiber-optical amplifier applications,” Opt. Lett. 19, 904–906 (1994). [CrossRef] [PubMed]
15. T. H. Lee, S. I. Simdyankin, J. Hegedus, J. Heo, and S. R. Elliott, “Spatial distribution of rare-earth ions and GaS4 tetrahedra in chalcogenide glasses studied via laser spectroscopy and ab initio molecular dynamics simulation,” Phys. Rev. B Condens. Matter Mater. Phys. 81(10), 760–762 (2010). [CrossRef]
16. Z. G. Ivanova, V. S. Vassilev, E. Cernoskova, and Z. Cernosek, “Physicochemical, structural and fluorescence properties of Er-doped Ge–S–Ga glasses,” J. Phys. Chem. Solids 64(1), 107–110 (2003). [CrossRef]
17. Q. Jiao, G. Li, L. Li, C. Lin, G. Wang, Z. Liu, S. Dai, T. Xu, and Q. Zhang, “Effect of gallium environment on infrared emission in Er3+-doped gallium- antimony- sulfur glasses,” Sci. Rep. 7(1), 41168 (2017). [CrossRef] [PubMed]
18. A. J. Kropf, J. Katsoudas, S. Chattopadhyay, T. Shibata, E. A. Lang, V. N. Zyryanov, B. Ravel, K. McIvor, K. M. Kemner, K. G. Scheckel, S. R. Bare, J. Terry, S. D. Kelly, B. A. Bunker, and C. U. Segre, “The new MRCAT (Sector 10) Bending Magnet Beamline at the Advanced Photon Source,” AIP Conf. Proc. 1234, 299–302 (2010).
19. C. U. Segre, N. E. Leyarovska, L. D. Chapman, W. M. Lavender, P. W. Plag, A. S. King, A. J. Kropf, B. A. Bunker, K. M. Kemner, P. Dutta, R. S. Duran, and J. Kaduk, “The MRCAT Insertion Device Beamline at the Advanced Photon Source”, CP521, Synch. Rad. Instr.: 11th U.S. National Conference, ed. P. Pianetta, 419–422, AIP, New York (2000). [CrossRef]
20. G. S. Henderson, F. M. F. de Groot, and B. J. A. Moulton, “X-ray absorption near-edge structure (XANES) spectroscopy,” Rev. Mineral. Geochem. 78(1), 75–138 (2014). [CrossRef]
21. I. Savelii, H. El Hamzaoui, L. Bigot, G. Bouwmans, I. Fsaifes, B. Capoen, and M. Bouazaoui, “Nanostructuring an erbium local environment inside sol–gel silica glasses: toward efficient erbium optical fiber lasers,” Laser Phys. Lett. 13(2), 025108 (2016). [CrossRef]
22. N. G. Boetti, G. C. Scarpignato, J. Lousteau, D. Pugliese, L. Bastard, J.-E. Broquin, and D. Milanese, “High concentration Yb-Er co-doped phosphate glass for optical fiber amplification,” J. Opt. 17(6), 065705 (2015). [CrossRef]
23. Z. Yang, H. Pan, Y. Chen, R. Wan, and X. Shen, “Emission properties of Er3+-doped Ge20Ga5Sb10Se65 glasses in near- and mid-infrared,” Infrared Phys. Technol. 89, 277–281 (2018). [CrossRef]
24. W. A. Pisarski, J. Pisarska, R. Lisiecki, and W. Ryba-Romanowski, “Erbium-doped lead silicate glass for near-infrared emission and temperature-dependent up-conversion applications,” Opto-Electron. Rev. 25(3), 238–241 (2017). [CrossRef]
25. J. Lægsgaard, “Dissolution of rare-earth clusters in SiO2 by Al codoping: A microscopic model,” Phys. Rev. B Condens. Matter Mater. Phys. 65(17), 1741141 (2002). [CrossRef]
26. A. Monteil, S. Chaussedent, G. Alombert-Goget, N. Gaumer, J. Obriot, S. J. L. Ribeiro, Y. Messaddeq, A. Chiaser, and M. Ferrari, “Clustering of rare earth in glasses, aluminum effect: experiments and modeling,” J. Non-Cryst. Solids 348, 44–50 (2004). [CrossRef]
27. K. Arai, H. Namikawa, K. Kumata, T. Honda, Y. Ishii, and T. Handa, “Aluminum or phosphorus co‐doping effects on the fluorescence and structural properties of neodymium‐doped silica glass,” J. Appl. Phys. 59(10), 3430–3436 (1986). [CrossRef]
28. H. Doweidar, “Optical properties and structure of R2O-Ga2O3-SiO2 and RO- Ga2O3-SiO2 glasses,” J. Mater. Sci. 44(11), 2899–2906 (2009). [CrossRef]
29. K. Lipinska and C. Segre, “Local order around Er3+ in Ga-doped oxide glasses probed by XAFS spectroscopy” – to be submitted.
30. G. S. Henderson, G. M. Bancroft, M. E. Fleet, and D. J. Rogers, “Raman spectra of gallium and germanium substituted silicate glasses: variation in intermediate order,” Am. Mineral. 70, 946–960 (1985).
31. F. L. Galeener and A. E. Geissberger, “Vibrational dynamics in 30Si-substituted vitreous SiO2,” Phys. Rev. B Condens. Matter 27(10), 6199–6204 (1983). [CrossRef]
32. G. S. Henderson, D. R. Neuville, B. Cochain, and L. Cormier, “The structure of GeO2–SiO2 glasses and melts: A Raman spectroscopy study,” J. Non-Cryst. Solids 355(8), 468–474 (2009). [CrossRef]
33. M. Micoulaut, M. L. Cormier, and G. S. Henderson, “The structure of amorphous, crystalline and liquid GeO2,” J. Phys. Condens. Matter 18(45), R753–R784 (2006). [CrossRef]
34. G. S. Henderson, “The structure of silicate melts: a glass perspective,” Can. Mineral. 43(1), 1921–1958 (2005). [CrossRef]
35. P. S. Salmon, A. C. Barnes, R. A. Martin, and G. J. Cuello, “Structure of glassy GeO2,” J. Phys. Condens. Matter 19(41), 415110 (2007). [CrossRef] [PubMed]
36. O. Majerus, L. Cormier, D. R. Neuville, L. Galoisy, and G. Calas, “The structure of SiO2–GeO2 glasses: A spectroscopic study,” J. Non-Cryst. Solids 354(18), 2004–2009 (2008). [CrossRef]
37. K. E. Lipinska and D. J. Mowbray, “The Structure of Al, Fe, K silica-germanate glasses investigated by Raman and infrared-spectroscopy,” J. Non-Cryst. Solids 122(1), 1–9 (1990). [CrossRef]
38. K. E. Lipinska and D. J. Mowbray, “Vibrational Study of Mixed SiO2-GeO2 Glasses,” J. Mol. Struct. 219, 107–110 (1990). [CrossRef]
39. C. I. Merzbacher and D. A. Mckeown, “X-ray absorption studies of Ge and Ga environments in BaO-Ga2O3-GeO2 glasses,” J. Non-Cryst. Solids 162(1–2), 81–100 (1993). [CrossRef]
40. L.-G. Hwa, J.-G. Shiau, and S.-P. Szu, “Polarized Raman scattering in lanthanum gallogermanate glasses,” J. Non-Cryst. Solids 249(1), 55–61 (1999). [CrossRef]
41. F. Miyaji, K. Tadanaga, T. Yoko, and S. Sakka, “Coordination of Ga3+ ions in PbO-Ga2O3 glasses as determined by 71Ga NMR,” J. Non-Cryst. Solids 139, 268–270 (1992). [CrossRef]
42. S. Sakka, H. Kozuka, K. Fukumi, and F. Miyaji, “Structures of gallate, aluminate and titanate glasses,” J. Non-Cryst. Solids 123(1–3), 176–181 (1990). [CrossRef]
43. M. Otten, C. Segre, J. Cecil, M. Chavara, K. Lipinska, Y. Ohki, and P. E. Kalita, “Luminescence and local structure correlation of Er-doped glasses and composites,” Bull. Am. Phys. Soc. 2013, M2300006 (2013).
