Optical properties of peralkaline aluminosilicate glasses doped with Sm3+

Optical properties of peralkaline aluminosilicate glasses doped with Sm3+

Accepted Manuscript Optical properties of peralkaline aluminosilicate glasses doped with Sm 3+ R. Turki, M. Zekri, A. Herrmann, C. Rüssel, R. Maalej...

1MB Sizes 0 Downloads 10 Views

Accepted Manuscript Optical properties of peralkaline aluminosilicate glasses doped with Sm

3+

R. Turki, M. Zekri, A. Herrmann, C. Rüssel, R. Maalej, K. Damak PII:

S0925-8388(19)32770-7

DOI:

https://doi.org/10.1016/j.jallcom.2019.07.255

Reference:

JALCOM 51543

To appear in:

Journal of Alloys and Compounds

Received Date: 4 May 2019 Revised Date:

16 July 2019

Accepted Date: 22 July 2019

Please cite this article as: R. Turki, M. Zekri, A. Herrmann, C. Rüssel, R. Maalej, K. Damak, Optical 3+ properties of peralkaline aluminosilicate glasses doped with Sm , Journal of Alloys and Compounds (2019), doi: https://doi.org/10.1016/j.jallcom.2019.07.255. 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 proof before it is published in its final 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.

ACCEPTED MANUSCRIPT

Optical properties of peralkaline aluminosilicate glasses doped with Sm3+ R. Turkia*, M. Zekria, A. Herrmannb,c, C. Rüsselc, R. Maaleja, K. Damaka a

RI PT

Photonics and Advanced Materials Group, Georesources Materials, Environment and Global Changes Laboratory (GEOGLOB), University of Sfax, 3018, Sfax, Tunisia b Kazuo Inamori School of Engineering, New York State College of Ceramics, Alfred University, 2 Pine Street, Alfred, NY 14802, USA c Otto Schott Institute of Materials Research, Jena University, Fraunhoferstrasse 6, 07743 Jena, Germany

EP

TE D

M AN U

SC

ABSTRACT Enhancing the fluorescence and thermo-mechanical properties of rare-earth doped aluminosilicate glasses (AS) by modifying the chemical composition with different alkaline earth oxides opens new opportunities for their application as optical devices. This paper provides a study on 1×1020 ions/cm3 Sm3+ doped aluminosilicate glasses, with the network modifier oxides BaO, SrO, CaO, MgO (35 mol%), Al2O3 (10 mol%) and SiO2 (55 mol%), denoted as BaAS, SrAS, CaAS and MgAS, prepared by the melt quench technique. Based on the measured UV-visible and NIR absorption spectra, the Judd-Ofelt parameters Ωλ, (λ=2, 4, 6) were evaluated to predict the radiative properties of Sm3+ ions such as radiative transition probabilities (Arad), radiative lifetimes (τrad) and branching ratios (βR). By using the Füchtbauer-Ladenburg (FL) formula, the gain cross sections were calculated from the stimulated emission cross sections. The glass BaAS has the highest value of Ω2 which corresponds to a high asymmetry at the rare earth sites and the glass MgAS has the highest values of Ω4 and Ω6 which corresponds to a high rigidity and a low optical basicity of the glass. The same glass has the highest quantum efficiency and the highest gain cross section values of the investigated glasses. On the other hand, the glass BaAS has the smallest efficiency and gain cross section, however, it provides the lowest glass transition temperatures which would be beneficial for the production of high-quality glass. Raman spectra of the AS glasses show a broader variation of Qn species and an increasing depolymerization for glasses with network modifier ions of larger ionic radii.

AC C

Keywords: alkaline earth aluminosilicate glasses; Sm3+; luminescence; Judd-Ofelt analysis; cross section; Raman spectra. *corresponding author. E-mail address: Rim Turki: [email protected] / [email protected]

1

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

1. Introduction Among different host materials used for optical applications, aluminosilicate glasses are very attractive materials due to their advantageous properties, which include high thermal and chemical stability, high mechanical strength, good mechanical properties and a good glass forming ability in a very broad compositional range [1-6]. While rare earth doped silica glass prepared by CVD processes is commonly used to produce laser fibers, aluminosilicate glasses (AS) produced by melt quenching techniques allow a higher solubility of the rare earth dopants and therefore avoid negative effects such as clustering of these ions. This should result in higher fluorescence lifetimes and higher quantum efficiencies at comparably high doping concentrations. Furthermore, these glasses have lower melting and processing temperatures than silica glass. The addition of the so-called network modifier ions decreases the melting and processing temperature by disruption of the continuity of the glass network and also affects all other physical properties in dependence of the used modifier ions. By appropriate selection of the network modifier oxide and its concentration, most physical properties of the AS glasses can be altered on demand. This has been shown in numerous earlier publications of our group [e.g. 2-8]. Good mechanical properties such as high elastic moduli and high hardness values can be obtained in the AS glass system with network modifier ions of high field strength (e.g. Mg2+, Zn2+, La3+) [6]. Advantageous optical properties such as long fluorescence lifetimes can be found for AS glasses of low refractive indices [2, 3, 5, 8]. However, recently, two interesting effects that strongly depend on the choice of the network modifier oxide were studied: for peralkaline Ba, Sr, K, and Na AS glasses, i. e. glasses which molar network modifier concentration is larger than the molar Al2O3 concentration, unusually high luminescence lifetimes of the rare earth ions were found [2, 7, 8]. Secondly, a systematic change in the emission spectra of Sm3+ was reported for AS glasses with low field strength network modifier ions and low network modifier concentrations [7]. In this work, four specific peralkaline glass compositions were studied taking into account Judd-Ofelt (J-O) theory to investigate the radiative properties (J-O parameters, dipole line strength, radiative transition possibility, stimulated emission cross section and branching ratio) of the glass samples for analyzing the potential applications of the studied glasses. In order to correlate results from Judd-Ofelt theory with the structure of the –SiO4– (SiO4, AlO4-)– network Raman spectroscopy was performed for the different studied glasses.

2. Experimental and theoretical procedures 2.1. Glass preparation Using the conventional melt-quenching technique, a series of peralkaline aluminosilicate (AS) glasses was prepared with the molar composition of 55 mol% SiO2, 10 mol% Al2O3, 35 mol% M (M=MgO, CaO, SrO, BaO) and doped with 1×1020 Sm3+ ions/cm3 (about 0.2 mol% Sm2O3) using high purity (Fe < 10 ppm, other contaminating metals < 0.5 ppm) oxides and carbonates: MgO (Merck KG, Germany, >97.0 %), CaCO3 (Merck KG, Germany, >99.0 %), SrCO3 (Reachim,

2

ACCEPTED MANUSCRIPT

2.2. Glass characterization

M AN U

SC

RI PT

Ukraine, >99.0 %), BaCO3 (Reachim, Ukraine, >99.0 %), Al2O3 (Pengda Munich GmbH, Germany, >99.99 %), SiO2 (Sipur A1, Bremthaler Quarzitwerk, Germany, >99.98 %), Sm2O3 (Auer-Remy GmbH, Germany, >99.99 %). The doping concentration of 1×1020 Sm3+/cm3 generally offers a good compromise between relatively high emission and absorption intensities and low concentration quenching as described in numerous previous publications [e.g. 2-8]. The raw material batches of 150 g were thoroughly mixed with an agate mortar and pestle before being melted in platinum crucibles at air for 3 hours in an electric furnace at temperatures in the range from 1500 to 1600 °C depending on the glass composition. The melt was stirred from time to time to achieve a better homogeneity. Afterwards, the glasses were cast onto a brass mold and transferred to a cooling furnace preheated to temperatures depending on the glass transition temperature Tg of the respective glass composition to reduce the mechanical stress, in the range from 720 to 920 °C. Subsequently, the cooling furnace was switched off to cool the samples slowly to room temperature with a rate of approximately 3 K/min. Later, the samples were cut to pieces of about 1 cm × 2 cm and a thickness of 1 cm, and ground and polished for the optical measurements.

AC C

EP

TE D

A helium pycnometer, AccPyc 1330, (Micromeritics, Germany) was used to measure the density of each glass sample. The refractive indices were measured at 5 different wavelengths by using an Abbe refractometer. The Tg of the synthesized glasses were determined by using a differential scanning calorimeter (DSC 822, Mettler Toledo, Switzerland). Using a commercial double beam spectrometer UV3102PC (Shimadzu, Japan), the absorption spectra of the glasses were determined at wavelengths in the range from 190 to 1800 nm. The fluorescence emission as well as the excitation spectra in the UV to visible range were recorded using a spectrofluorometer RF 5301 PC (Shimadzu, Japan). The fluorescence lifetimes were measured for the strongest fluorescence transition of Sm3+ by using a home-made experimental setup with a high intensity pulsed LED at 395 nm (LED 395-66-60110, Roithner Lasertechnik GmbH, Austria) for excitation, a monochromator (H.25, Jobin Yvon, France) for wavelength selection, a photomultiplier tube (R5929, Hamamatsu, Japan) as detector and an oscilloscope (TDS2012, Tektronix, USA) for data acquisition. All measurements were conducted at room temperature. The theoretical optical basicity was calculated from the molar compositions of the glasses, according to Duffy [9], from the averaged partial basicity values of Duffy and Lebouteiller/Courtine given in Ref. [10]. The Raman spectra were recorded using a Senterra dispersive Micro-Raman spectrometer (Bruker, Germany). All samples were investigated under the same optical conditions and with the same incident power (20 mW) using a 532 nm excitation laser. The chromaticity coordinate values (x, y,) were evaluated from the emission spectra following the same method used in previous publications, e.g. [11].

3. Results and discussion

3

ACCEPTED MANUSCRIPT

3.1. Physical properties

1 Es E2 = − n 2 ( E ) − 1 Ed E s E d

RI PT

The sample names, and some physical properties such as glass densities, glass transition temperatures, optical basicity and refractive indices are summarized in Table 1. The dependency of the refractive index on the wavelength can be described by Wemple relation [12], which enables the calculation of the Sellmeier energy gap Es, and dispersion energy Ed. (1)

EP

TE D

M AN U

SC

For each studied glass composition, Es and Ed were calculated from the linear regression of the plot 1/(n2(E)-1) versus E2, which allows for modeling of the refractive index dependence on the wavelength. The densities and the refractive indices of the glasses increase (as expected) and the Es values decrease with increasing atomic weight and increasing ionic radius of the network modifier ion. In general, atoms of higher atomic number (and higher number of electrons) have a higher polarizability due to their larger electron clouds, which can be distorted more easily by electro-magnetic fields, which in turn is the cause of the higher refractive indices. However, a high refractive index is likely to be disadvantageous for laser applications because of increased second order effects, which might lead to selffocusing and self-phase modulation deteriorating the temporal and spatial characteristics of the laser pulses. The transition temperatures Tg of the studied glasses are in the range from 730 to 831 °C. The Tg values are consistent with published data for similar compositions [e.g. 2-8]. The data shown in Table 1 indicates that the addition of alkaline earth ions of higher atomic numbers tends to decrease Tg of the investigated samples. The lowest Tg (730 °C) was observed for the BaAS glass with the highest molecular weight, while the highest Tg was measured for the MgAS glass with the lowest molecular weight. A low Tg might be advantageous for glass production because in most cases it enables lower melting viscosities and/or temperatures. Hence, it enables a better and easier homogenization of the glasses [13-18]. 3.2. Raman spectroscopy

AC C

Fig. 1 shows the Raman spectra of all four glass samples. The strong bands in the range from 800 to 1200 cm-1 are due to Si–O stretching vibrations in the –SiO4– (SiO4, AlO4-)– tetrahedral network with [AlO4]- acting only as a perturbation on these. In comparison to binary silicate glasses, the peak in this region is notably broadened due to the Al2O3 addition since Al–O bonds have a lower vibrational energy than Si–O bonds [19-21]. The addition of network modifiers results in a depolymerization of the glass network and non-bridging oxygen sites SiO3–O- are created which are, besides the [AlO4]- groups, needed for additional charge compensation of the network modifying ions. Since the Si–O- bond has a lower vibrational energy than the Si-bond with bridging oxygen, the peak is broadened even more. Obviously, the addition of different network modifier ions results in different shapes of this peak. With increasing size (decreasing field strength) of the

4

ACCEPTED MANUSCRIPT

EP

TE D

M AN U

SC

RI PT

network modifier ion, a shoulder at around 950 cm-1 becomes apparent in the presented peralkaline compositions. This is the opposite effect that was observed for met-aluminous AS glasses (glasses with equimolar concentrations of MO and Al2O3) of the composition 20 MO - 20 Al2O3 - 60 SiO2 (mol%). Here the peak splitting is increased with increasing field strength of the network modifier ion [3, 4]. Recent molecular dynamic simulations show that in met-aluminous glasses charge compensation of smaller network modifier ions such as Mg2+ cannot exclusively be achieved by coordination with [AlO4]- tetrahedra, most likely due to the limited space [22]. Therefore, small network ions additionally create non-bridging oxygen (NBO) sites in these glasses, resulting in the described shoulder due to the lower vibrational energy of the SiO3–O- (Q3) groups. In the met-aluminous AS glasses with larger network modifier ions (e.g. Ba2+) much fewer Q3 and mostly Q4 groups exist. In peralkaline AS glasses, the opposite effect occurs. Here network modifier ions of larger radii create more NBO sites in the glass network than smaller ions [21]. This is also reflected by the opposing Tg trends of mixed BaMg-AS glasses of metaluminous or peralkaline compositions [7]. The bands between 700 and 850 cm−1 of AS glasses are clearly dominated by aluminum and are ascribed to the Al-O stretching vibrations [23]. Due to the low Al2O3 concentrations in the prepared glasses they are relatively low in intensity. Another prominent feature in the Raman spectra is the peak at around 500-600 cm-1. Its relative intensity clearly increases with decreasing size of the network modifier ion. Additionally, a slight shift to lower energies can be observed. The bands in the range from 450 to 700 cm-1 are mostly due to bending vibrations of bridging oxygen in Si–O– (Si, Al) linkages. In binary alkali and alkaline earth silicate glasses, the peak of this band shifts to higher frequencies with decreasing polymerization of the glass network (increasing network modifier concentration) [23, 24]. This corresponds to the observations described for the high energy stretching vibrations. Network modifier ions of larger ionic radii increase the depolymerization of the glass network in peralkaline AS glasses. However, the maximum vibrational (phonon) energy, which may have an influence on the optical properties (multi-phonon quenching) is barely touched by the network modifier substitution, since it is only determined by the Si–O–Si bonds in Q4 groups which occur in all four investigated glasses.

AC C

3.3. Absorption spectra

The absorption spectra of the 1020 ions/cm3 Sm3+ doped AS glasses are shown in Fig. 2. The glasses show many absorption bands which can be assigned to the transitions from the ground state 6H5/2 to various excited states 6H13/2, 6F1/2, 6F3/2, 6 F5/2, 6H15/2, 6F7/2, 6F9/2, 6F11/2, 4G5/2, 4M15/2 (4I9/2, 4I11/2, 4I13/2, 4G7/2), 6P3/2, 6P7/2, 4D3/2, and 4D7/2 in the visible and near infrared range. The absorption spectra for the present glasses are quite similar and the characteristic peaks are observed at the same wavenumbers with only slight differences in their intensities. Table 2 shows the measured properties of the 6H13/2, 6F1/2, 6F3/2, 6F5/2, 6H15/2, 6F7/2, 6F9/2, 6F11/2, 4G5/2, 4 M15/2, 4I9/2, 4I11/2, 4I13/2, 4G7/2, 6P3/2, 6P7/2 absorption transitions from the ground state 6 H5/2. These transitions were selected and used for the Judd-Ofelt analysis. For this

5

ACCEPTED MANUSCRIPT

the measurement offset was removed from the spectra. However, as seen in Fig. 1 there are some measurement anomalies in the spectral ranges around the 6H13/2 (~5,200 cm-1) and 4G5/2 (~17,800 cm-1) peaks for the BaAS and at wavenumbers above 25,000 cm-1 (6P7/2) for the CaAS glasses. These peaks were omitted for the J-O calculation of these samples.

RI PT

3.4. Luminescence spectra

AC C

EP

TE D

M AN U

SC

The excitation spectra of all glasses were recorded in the spectral range 250 to 600 nm by monitoring the emission at 600 nm corresponding to the transition 4 G5/2→6H7/2 (Fig. 3). Among all excitation bands, the band 6H5/2 →6P3/2 at 402 nm is the most intense and it is used as excitation wavelength for the measurement of the emission spectra of the studied glasses. Fig. 4 shows the typical four fluorescence emission lines of the Sm3+ ions in the visible range 4G5/2→6H5/2 at 562 nm, 4 G5/2→6H7/2 at 600 nm, 4G5/2→6H9/2 at 645 nm and 4G5/2→6H11/2 at 706 nm. The two most intense emission bands are observed at 600 and 645 nm corresponding to the transitions 4G5/2→6H7/2 (orange) and 4G5/2→6H9/2 (red), respectively. The energy level scheme of the emission mechanism of Sm3+ ions in the glass samples is shown in Fig. 5. In the emission spectra, a small shift towards higher wavelengths can be observed with higher glass basicity. The optical basicity of the AS glasses calculated increases in the order MgAS
6

ACCEPTED MANUSCRIPT

Sedmeas ( J , J ') =

 9n  3hc(2 J + 1) 2.303 α ( λ ) d λ − nS ( J , J ')   md ∫ (n² + 2)²  8π 3e² N λ J →J ' 

(2)

2

J L + 2S J

M AN U

 h  S md ( J , J ') =   4π mc 

SC

RI PT

where J and J’ are the total angular momentum quantum numbers of the initial and final states, respectively; λ is the mean wavelength of the absorption band, n is the refractive index of the host with respect to λ , c is the vacuum velocity of light, h is the Planck constant, N is the Sm3+ concentration, e is the electron OD(λ ) charge, α (λ ) = is the experimental absorption coefficient of the glass with L OD(λ) and L representing the optical density and the thickness of the sample under study (L=1 cm), respectively. Smd (J, J’) is the magnetic dipole line strength which can be calculated from Eq. 3. It is independent of the ligand field and was determined by C. K. Jayasankar et al. [28].

(3)

Here m is the electron mass, L and S are respectively the orbital and spin angular momentum. The electric dipole line strengths ( S edcal ) are calculated using Eq. (4) from [26]. S edcal = ∑ k =2,4,6 Ω k J U K J '

2

(4)

TE D

Here the Ω k ( k = 2, 4, 6 ) are the Judd-Ofelt intensity parameters. The equality between equation (2) and equation (4) was used to calculate the set of Ωk parameters by a least-square fitting approach. To justify the results obtained, a measure of accuracy of the fit is given by the root mean square (rms) deviation δ rms [26].

AC C

EP

 ( S meas − Sedcal )²  (5) δ rms =  ∑ ed  ( p − 3)   Here p is the number of observed transitions in the absorption spectrum. The measured and calculated dielectric dipole line strengths of different transitions, their wavenumbers and the magnitude of δ rms of all studied glasses are presented in Table 2. The evaluated J-O intensity parameters were compared with available literature data as presented in Table 3. Generally, Ω2 is relatively small for the present glasses, which suggests a more ionic nature and low covalency of the Sm–O bonds and/or a less asymmetric environment around the Sm3+ ions [27, 29]. When comparing the Ω2 parameters of the prepared AS glasses with values of different glass systems, it can be noted that very ionic host materials such as fluorides and fluoride containing glasses as well as fluoride crystals generally have very low Ω2 parameters (Table 3) [29-32]. Medium range Ω2 values are mostly observed in silicate and aluminosilicate glasses [29, 31, 33-38, 42], while especially high Ω2 values are found for instance in sulphate glasses [29]. It has been reported that Ω2 is rather correlated with short-

7

1/2

ACCEPTED MANUSCRIPT

M AN U

SC

RI PT

range coordination effects, whereas Ω6 and Ω4 are (within a compositional series) more related to the macroscopic properties such as the basicity, rigidity and hence viscosity of the medium in which the ions are located [27, 29]. For the samples reported here the Ω6 and Ω4 parameters are highest in the magnesium containing glass, which also has the lowest basicity and highest rigidity and hence highest viscosity of the glass. Ω6 (and Ω4) decrease steadily for glasses with network modifiers of larger ionic radii. This correlates with increasing optical basicities of these glasses. This correlation has been found also for large sets of Er3+ doped phosphate and alumino silicate glasses [39, 40]. However, sample CaAS does not follow this trend. The parameter Ω2 shows a completely different tendency. It is highest for BaAS which should correspond to the highest covalency and/or asymmetry of the local rare earth site. However, it is second highest for MgAS, and CaAS and SrAS show the lowest values. This trend is very similar to the measured luminescence lifetimes and even corresponds to the abnormal lifetime behavior found for peralkaline BaAS and SrAS glasses that was already described in the introduction section. Interestingly a similar trend was observed for the same glass compositions doped with Dy3+ [43]. From the obtained Ωk parameters, the radiative transition probabilities Aed, Amd, AR, the branching ratios βR and the radiative lifetime τR for the 4G5/2 level were estimated using the following relations [13, 26, 44]: AR ( J → J ’) = Aed ( J → J ’) + Amd ( J → J ’) (6)

Amd =

64π 4 n3S md 3h(2 J + 1)λ 3

64π 4 n(n ² + 2)² ( ) Sed 3 3h(2 J + 1)λ 9 A ( J → J ') β R ( J → J ') = R ∑ AR ( J → J ')

EP

TE D

Aed =

τ R (J ) =

(7) (8) (9)

J'

1 ∑ AR ( J → J ')

(10)

AC C

The results are presented in Table 4. The calculated radiative lifetimes for the 4G5/2 level are 3671, 3123, 3154 and 2796 µS for the glasses BaAS, SrAS, CaAS and MgAS, respectively. These values are higher than the measured lifetimes, which is probably caused by luminescence quenching processes such as cross-relaxation [45, 46, 47]. This is possible because of the matching energy differences between 6H5/2 and 6H15/2 (and adjacent levels) and the energy gap between 4G5/2 and 5F11/2, which are both around 7000 cm-1. Therefore, the 4G5/2 level of excited Sm3+ ions can be deactivated by exciting a nearby Sm3+ ion in the ground state and the red/orange luminescence of the excited ion is quenched.

3.6. Fluorescence lifetime measurements The emission decay curves of the 4G5/2 emission of Sm3+ doped AS glasses were recorded and analyzed. These curves provide information on the lifetime of the 4G5/2

8

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

excited state of the rare-earth ion in the respective host glass composition. Fig. 6 shows the emission decay profiles for all glasses. In a semi logarithmic scale, a mono-exponential decay appears as a straight line. The slope of this line represents the reciprocal fluorescence lifetime. For all samples, a nearly mono-exponential decay is observed, which indicates homogeneously distributed fluorophores and an almost undisturbed fluorescence emission process, i.e. almost no concentration quenching. The observed slight deviation from the mono-exponential decays can be understood with the help of the Sm3+ electronic energy level structure and is supposedly due to cross-relaxation processes as explained earlier. The measured luminescence lifetimes are almost constant for the samples MgAS, CaAS, SrAS at about 2.6 ms. Only BaAS shows a notably higher lifetime value. Exceptionally high luminescence lifetimes of doped rare earth ions in peralkaline BaAS glasses were also reported in previous publications (Sm3+ [8], Tb3+ [7], Eu3+ [2], Dy3+ [43]). An even stronger increase was observed for peralkaline NaAS and KAS glasses [2, 8]. So far, no explanation for this unexpected effect could be found. However, the J-O investigations presented here and earlier investigations on Dy3+ doped AS glasses [43] suggest a correlation with the Ω2 parameter. But generally, a higher asymmetry (higher Ω2) should decrease the fluorescence lifetime, since the parity selection rule for the forbidden f-f-transition is relaxed by a non-centrosymmetric crystal field. Therefore, the correlation of Ω2 and the measured lifetimes may be accidental. However, the actual site symmetry in glasses is generally very difficult to determine. In addition, ionic materials such as fluoride glasses and crystals (low Ω2) generally provide high fluorescence lifetimes, but the investigated samples with the lowest Ω2 parameters also show the lowest fluorescence lifetime. Therefore, the cause of the high fluorescence lifetime of the BaAS glass remains unclear. Here, further investigations are needed. The quantum efficiency is a crucial factor for the performance of the material and its potential to be used for laser light generation. The luminescence quantum efficiency η is the ratio of the measured to the calculated radiative lifetime using [43]: τ (11) η = meas τR τmeas and τR are the measured and the calculated radiative lifetime values of the state 4 G5/2, respectively. In the present investigations, the calculated values of the quantum efficiency are between 78.47 and 93.7 %. The glass MgAS has the highest quantum efficiency of the investigated glasses, most likely because of the lower intrinsic (calculated) lifetime. So, the excited ions have less time for cross relaxation (see Table 5). However, this high value should be proven by experiments, which is no easy task because of many disadvantageous effects such as OH-quenching and quenching by different impurities, which cannot be ruled out completely during the glass melting process. On the other hand, high quantum efficiencies have been proven in similar ZnAS and LiAS glasses, doped with Yb3+ ions [48]. The transition cross section σ(λP) is an important index used to quantify the likelihood of optically induced transition events and to evaluate the energy extraction efficiency for laser materials. The σ(λP) can be estimated for the transitions arising from the 4G5/2 level of Sm3+ by using the fact that the fluorescence intensity I(λ) is

9

ACCEPTED MANUSCRIPT

proportional to the emission cross section. This leads to the Füchtbauer-Ladenburg (FL) formula [49, 50]: λp4 I (λ ) (12) σ (λ p )(J → J ') = 2 8π n τ R ∫ I (λ )d λ

M AN U

SC

RI PT

Here λP is the emission peak wavelength and n is the refractive index of the material. If the intensity is normalized to 1 at the peak of the optical spectrum, the integral in the denominator can be interpreted as the effective emission bandwidth ΔλP. Among the four emission transitions in the visible range, the transition 4G5/2→6H7/2 is the most intense in the investigated glasses. The value of σ(λP) for this transition is found to be highest for the MgAS glass and smallest for BaAS when compared with that of the other studied samples. The gain bandwidths (σ(λP)· ΔλP) and optical gains (σ(λP)· τR) together with the other above-mentioned parameters are summarized in Table 6. The optical parameters of the studied glasses are in the same range as for other potential laser host glasses doped with Sm3+ [51, 52] but obviously higher than those in cadmium aluminosilicate glasses [38]. 3.7. CIE chromaticity coordinates

EP

TE D

In order to understand the color emitted by the Sm3+ doped AS glasses, the chromaticity coordinate values (x, y, z) were evaluated from the emission spectra following the method described in [11, 53]. These coordinates were calculated from the emission spectra obtained after exciting the samples at 402 nm. The chromaticity coordinates x and y are given in Table 7 and displayed in the CIE diagram in Fig. 7. It is observed that the CIE coordinates for all studied AS glasses are in the orangered region. Only very little change in the (x, y) values is observed for the studied glasses, which could be expected due to the high similarity of the emission spectra. The calculated chromaticity co-ordinates (0.59, 0.40) in the orange to red spectral region may be useful in lighting application. 3. Conclusion

AC C

Sm3+ doped peralkaline aluminosilicate (AS) glasses with different network modifier ions were synthesized and their physical and optical properties were characterized. The results show that the densities and the refractive indices of the studied glasses increase with increasing atomic weight of the network modifying elements. The barium aluminosilicate glass (BaAS) provides the lowest glass transition temperature which would be beneficial for the production of high-quality glass. The differences in the Raman bands of the glasses could be correlated to an increasing depolymerization for glasses with network modifier ions of large radii (low field strength). The four glasses show broad fluorescence excitation and emission spectra that are barely affected by the glass composition. However, the Judd-Ofelt calculations revealed interesting details: The BaAS glass has the highest value of Ω2 which is often correlated to a low local symmetry at the rare earth sites and a high covalency of the Sm-O bonds. The covalency and asymmetry decrease in

10

ACCEPTED MANUSCRIPT

SC

RI PT

the order CaAS>SrAS>MgAS>BaAS. This correlate well with the measured luminescence lifetime values, however, the correlation with the site symmetry and covalency is not plausible. Despite its high refractive index and atomic weight, the BaAS glass has the highest luminescence lifetime. The Ω6 and Ω4 parameters suggest a higher rigidity and viscosity in the order of BaAS
AC C

EP

TE D

M AN U

This work was supported by the German Federal Ministry of Education and Research (BMBF) and the Tunisian Ministry for Higher Education and Scientific Research (MESRS) through the joint German-Tunisian announcement "Regulations Governing the Funding of Scientific and Technological Cooperation (STC) between Germany and Tunisia" (project number IB-TUNGER15-067).

11

ACCEPTED MANUSCRIPT

References

[7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18]

AC C

[19] [20]

RI PT

[6]

SC

[5]

M AN U

[4]

TE D

[2] [3]

H. Scholze, Glass - Nature, Structure, and Properties, Springer New York Inc., New York, 1991. A. Herrmann, S. Kuhn, M. Tiegel, C. Rüssel, Optical Materials 37 (2014) 293-297. A. Herrmann, S. Kuhn, M. Tiegel, C. Rüssel, J. Körner, D. Klöpfel, J. Hein, M. C. Kaluza, Journal of Materials Chemistry C 2 (2014) 4328-4437. M. Tiegel, A. Herrmann, C. Rüssel, J. Körner, D. Klöpfel, J. Hein, M.C. Kaluza, Journal of Materials Chemistry C1 (2013) 5031-5039. M. Tiegel, S. Kuhn, A. Herrmann, C. Rüssel, J. Körner, D. Klöpfel, R. Seifert, J. Hein, M. C. Kaluza, Laser Physics Letters 11 (2014) 115811-1-115811-6. M. Tiegel, R. Hosseinabhadi, S. Kuhn, A. Herrmann, C. Rüssel, Ceramics International 41 (2015) 7267-7275. A.A. Assadi, A. Herrmann, M. Tewelde, R. Maalej, C. Rüssel, Journal of Luminescence 199 (2018) 384-390. A. Herrmann, M. Tewelde, S. Kuhn, M. Tiegel, C. Rüssel, Journal of NonCrystalline Solids 502 (2018) 190-197. J.A. Duffy, Geochimica et Cosmochimica Acta 57 (1993) 3961-3970. V. Dimitrov and T. Komatsu, J. Univ. Chem. Tech. Met. 45(2010) 219–250. K. Damak, E. S. Yousef, C. Rüssel, R. Maâlej, Journal of Quantitative Spectroscopy and Radiative Transfer 134 (2014) 55-63. S. H. Wemple, Journal of Chemical Physics 67 (1977) 2151. B.R. Judd, Phys. Rev. 127 (1962) 750-761. M. Sheik-Bahae, E. W. van Stryland, Semiconductors and Semimetals 58 (1999) 257-318. T. Töpfer, J. Hein, J. Philipps, D. Ehrt, R. Sauerbrey, Applied Physics B 71 (2000) 203-206. R. Reisfeld, Inorganica Chimica Acta 95 (1984) 69-74. W. J. M. van der Kemp, J. G. Blok, P. R. van der Linde, H. A. J. Oonk, A. Schuijff, Calphad 18 (1994) 255-267. D. A. Pinnow, L. G. van Uitert, T. C. Rich, F. W. Ostermayer, W. K. Grodkiewicz, OSA Technical Digest Optical Society of America, (1975) paper TuA3. S. A. Brawer, W. B. White, Journal of Chemical Physics 63 (1975) 2421. E. I. Kamitsos, J. A. Kapoutsis, H. Jain, C. H. Hsieh, Journal of Non-Crystalline Solids 171 (1994) 31-45. P. Mc Millan, B. Piriou, J. Non-Cryst. Solids 53 (1982) 279-298. M. Zekri, A. Erlebach, A. Herrmann, K. Damak, C. Rüssel, M. Sierka, R. Maâlej, Materials 11 (2018) 1790. P. McMillan, B. Piriou, Bull. Mineral. 106 (1983) 57-75. S. A. Brawer, W. B. White, Journal of Non-Crystalline Solids 23 (1977) 261-278. W. T. Carnall, P. R. Fields, K. Rajnak, Journal of Chemical Physics 49 (1968) 4424-4442. K. Damak, E. S. Yousef, A. S. Al-Shihri, H. J. Seo, C. Rüssel, R. Maâlej, Solid State Sciences 38 (2014) 74-80.

EP

[1]

[21] [22]

[23] [24] [25] [26]

12

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

[27] M. P. Hehlen, M. G. Brik, K. W. Krämer, Journal of Luminescence 136 (2013) 221-239. [28] C. K. Jayasankar, E. Rukmini, Optical Materials 8 (1997) 193-205. [29] C. Görller-Walrand, K. Binnemans, Handbook on the physics and chemistry of rare earths Vol. 25, Elsevier, Amsterdam, New York 167 (1998) 101-264. [30] S. Mahamuda, K. Swapna, M. Venkateswarlu, A. S. Rao, S. Shakya, G. V. Prakash, Journal of Luminescence 154 (2014) 410-424. [31] R. van Deun, K. Binnemans, C. Gorller-Walrand, J.-L. Adam, Proc. SPIE 3622, Rare-Earth-Doped Materials and Devices III, (1999) [32] G. Lakshminarayana, R. Yang, J.R. Qiu, M.K. Brik, G.A. Kumar, I.V. Kityk, J. Phys. D Appl. Phys. 42 (2009) 0154141–01541412. [33] H. Takebe, Y.Nageno, K. Morinaga. J. Am. Ceram. Soc., 78 (1995) 1161-1168. [34] G. Kawamura, T. Hayakawa, M. Nogami, J. Alloy. Comp, 408–412 (2006) 845– 847. [35] K. Annapurna, R. N. Dwivedi, S. Buddhudu. Part of the SPIE Conference on Design, Fabrication, and Characterization of Photonic Devices 3896 (1999) 653. [36] I. Khan, G. Rooh, R. Rajaramakrishna, N. Srisittipokakun, H.J. Kim, K. Kirdsiri, J. Kaewkhao, Acta Part A: Molecular and Biomolecular Spectroscopy 214 (2019) 1420. [37] V. Thomas, R.G.S. Sofin, M. Allen, H. Thomas, G. Jose, P.R. Biju, N.V. Unnikrishnan, Acta A 171 (2016) 144-148. [38] J. Yang, B. Zhai, X. Zhao, Z. Wang, H. Lin, Journal of Physics and Chemistry of Solids 74 (2013) 897-1062. [39] R. Lachheb, A. Herrmann, A. A. Assadi, J. Reiter, J. Körner, J. Hein, C. Rüssel, R. Maâlej, K. Damak, Journal of Luminescence 201 (2018) 245-254. [40] A. A. Assadi, A. Herrmann, R. Lachheb, K. Damak, C. Rüssel, R. Maâlej, Journal of Luminescence 176 (2016) 212-219. [41] J. A. Jiménez, Journal of Alloys and Compounds 623 (2015) 623-401. [42] S. Kuhn, A. Herrmann, C. Rüssel, Journal of Luminescence 157 (2015) 390-397. [43] M. Zekri, A. Herrmann, R. Turki, K. Damak, C. Rüssel, R. Maâlej, Journal of Luminescence 212 (2019) 354-360. [44] G. S. Ofelt, Journal of Chemical Physics 37 (1962) 511. [45] S. Kuhn, A. Herrmann, C. Rüssel,. Journal of Luminescence 158 (2015) 333-339. [46] A. G. Souza Filho, J. Mendes Filho, F. E. A. Melo, M. C. C. Custodio, R. Lebullenger, A. C. Hernandes, Journal of Physics and Chemistry of Solids 61 (2000) 1535-1542. [47] A. Herrmann, D. Ehrt, Journal of Non-Crystalline Solids 354 (2008) 916-926. [48] S. Kuhn, M. Tiegel, A. Herrmann, C. Rüssel, C. Wenisch, S. Engel, S. Gräf, F. Müller, J. Körner, R. Seifert, F. Yue, D. Klöpfel, J. Hein, M. C. Kaluza, Journal of Applied Physics 118 (2015) 1031041-1031049. [49] J. G. Edwards, Nature 212 (1966) 752–753. [50] A. M. Emaraa, M. M. Alqahtania, Y. M. Abou Deifa, E. S. Yousef, Chalcogenide Letters 14 (2017) 397-405. [51] O. Ravi, C. M. Reddy, L. Manoj, B. D. P. Raju, , Journal of Molecular Structure 1029 (2012) 53-59. [52] T. Sasikala, L. R.Moorthy, A. M. Babu, Optical and luminescent properties of Sm3+

13

ACCEPTED MANUSCRIPT

AC C

EP

TE D

M AN U

SC

RI PT

doped tellurite glasses, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy 104 (2013) 445-450. [53] S. L. Davidson, Wiley-Interscience, New York, 1981, 240 pp, Color Research & Application 7 (1982) 275-276.

14

ACCEPTED MANUSCRIPT

RI PT

Table 1 Densities, glass transition temperatures, refractive indices, band gap, Sellmeier energy gap, dispersion energy and theoretical optical basicity obtained for undoped AS glasses.

MgAS

CaAS

SrAS

BaAS

Density (g/cm3)

2.64

2.75

3.25

3.67

Tg (°C)

831

813

816

790

Eg (eV)

3.25

3.40

3.40

3.20

Es (eV)

12.22

12.09

12.04

11.75

M AN U

SC

Physical property

17.05

17.99

18.06

18.51

0.567

0.605

0.622

0.630

435.8

1.5728

1.6048

1.6089

1.6350

Refractive index n,

480.0

1.5682

1.5998

1.6039

1.6297

for different

546.1

1.5634

1.5944

1.5986

1.6240

589.3

1.5611

1.5922

1.5961

1.6212

643.8

1.5588

1.5896

1.5934

1.6184

Ed (eV)

AC C

EP

wavelengths (nm)

TE D

Optical basicity Λ

15

ACCEPTED MANUSCRIPT

Table 2 Absorption band wavenumbers, experimental and calculated electric dipole line strengths of Sm3+ doped aluminosilicate glasses.

(pm2)

-

H13/2

H15/2+6F1/2+6F3/2+6F5/2

Sedmeas Sedcal

(cm-1) (pm2) (pm2) 5215 0.167 0.121

meas

SrAS

ν-1 Sed

Sedcal

ν-1

Sedmeas

Sedcal

(cm-1) (pm2)

(pm2)

(cm-1) (pm2)

(pm2)

5221 0.154

0.096

5244

0.121

0.112

6952 2.377

2.377

6963 3.568 3.568

6969 2.837

2.838

6962

2.867

2.867

8160 1.601 1.675

8161 1.332

1.419

8159

1.364

1.425

6

F7/2

8155 1.141

1.167

6

F9/2

9306 0.616

0.614

9313 0.777 0.735

9314 0.674

0.597

9315 0.711

0.667

F11/2

10679 0.067

0. 843

10657 0.108 0.095

10692 0.098

0.075

10650 0.995

0.088

17828 0.009 0.004

17756 0.073 0.042

17799 0.006

0.003

21060 0.265 0.160

21097 0.222 0.129

21063 0.230 0.144

24721 1.019 0.957

24714 0.879

0.835

26711 0.192

0.132

6

4

-

G5/2

20965 0.218

0.133

TE D

G7/2+4M15/2+4I9/2+4I11/2+4I13/2 4

P3/2

24694 0.578

0.584

24758 1.125 1.071

4

P7/2

26717 0.196

0.125

26750 0.234 0.143

± 0.0533

EP

δ rms (pm2)

AC C

4

(cm-1) (pm2)

ν-1

M AN U

6

Sedcal

CaAS

SC

6

Sedmeas

ν-1

from 6H5/2 to

MgAS

RI PT

BaAS Transition

17

± 0.0730

-

±0.0737

±0.0558

ACCEPTED MANUSCRIPT

Table 3 The Judd-Ofelt intensity parameters (Ωλ ×10-20 cm2) of the Sm3+ doped AS glasses. J-O parameters Glass composition Ω2

Ω4

Ω6

BaAS

2.63

3.31

1.59

SrAS

2.08

4.96

1.63

CaAS

1.19

5.68

1.40

MgAS

2.38

6.36

ZBLA (Fluoride)

1.66

SiO2-B2O3-Al2O3-LiO-GdF3

0.94

Silicates

2.23-2.89

Phosphates

2.49-3.48

Borates

2.59-3.22

SiO2-Al2O3-MgO-Li2O

SiO2-Na2O-Bi2O3

This paper

Ω4>Ω6>Ω2

This paper

1.77

Ω4>Ω2>Ω6

This paper

1.69

1.24

Ω4>Ω2>Ω6

[29]

2.46

1.18

Ω4>Ω6>Ω2

[32]

1.69-4.44

0.82-2.07

-

[33]

2.90-4.00

1.78-2.27

-

[33]

2.03-4.39

1.52-2.59

-

[33]

M AN U

SC

Ω4>Ω2>Ω6

2.09

1.56

Ω4>Ω2>Ω6

[34]

1.94

2.63

1.62

Ω4>Ω2>Ω6

[35]

2.25

4.33

1.50

Ω4>Ω2>Ω6

[36]

2.46

3.46

2.09

Ω4>Ω2>Ω6

[37]

2.87

3.34

1.86

Ω4>Ω2>Ω6

[38]

AC C

SiO2-Al2O3-CdO

This paper

1.90

EP

SiO2-Li2O-BaO-Gd2O3

Ω4>Ω2>Ω6

TE D

SiO2-Al2O3-TiO2-ZrO2

References

RI PT

Trends

18

ACCEPTED MANUSCRIPT

BaAS

SrAS

CaAS

MgAS

AR

AR

AR

AR

24.74

25.06

137.17

146.24

G5/2⟶6H5/2

23.60

24.53

4

G5/2⟶6H7/2

105.75

130.57

4

G5/2⟶6H9/2

92.45

G5/2⟶6H11/2

26.53

AT

283.27

M AN U

4

SC

Transition

4

RI PT

Table 4 Transition probabilities AR (s-1), total transition probability AT (s-1), and radiative branching ratios βR (%) for the prominent emission transitions of Sm3+ ions in the prepared AS glasses.

96.14

86.22

108.88

35.89

39.57

42.07

320.40

317.08

357.64

βR

βR

βR

8.33

7.66

7.81

7.01

37.33

40.65

43.248

40.89

TE D

βR G5/2⟶6H5/2

4

G5/2⟶6H7/2

4

G5/2⟶6H9/2

32.64

30.00

27.19

30.44

G5/2⟶6H11/2

9.36

11.20

12.48

11.76

AC C

4

EP

4

19

ACCEPTED MANUSCRIPT

Table 5 Measured (τmeas) and radiative (τR) lifetimes, and quantum efficiencies η of the Sm3+ ions in the prepared glasses. τR (μs)

τmeas (µs)

η (%)

BaAS

3530

2770

78.47

SrAS

3123

2580

82.61

CaAS

3154

2600

82.43

MgAS

2796

2620

93.70

CdAS

2830

1700

this paper this paper this paper this paper

SC 60.01

M AN U TE D EP AC C 20

Reference

RI PT

Host

[38]

ACCEPTED MANUSCRIPT

Parameter

SrAS

λp

562

563

563

∆λp

8.52

8.12

8.94

σ(λp)

1.58

1.58

1.42

σ(λp)·∆λp

1.35

1.28

σ(λp)·τR

4.40

4.97

G5/2⟶6H7/2

9.18

1.13

1.27

1.04

4.43

3.98

599

600

600

601

∆λp

12.0

11.7

11.4

12.5

σ(λp)

6.57

7.89

7.62

3.56

σ(λp) ·∆λp

7.90

9.24

8.75

4.48

18.33

24.85

23.77

12.56

646

647

647

648

∆λp

13.4

16.5

14.2

12.9

σ(λp)

7.64

4.76

6.09

4.86

σ(λp) ·∆λp

10.3

7.88

8.69

6.29

σ(λp) ·τR

21.31

14.99

19.00

17.15

λp

706

706

706

706

∆λp

29.3

27.0

23.0

20.9

σ(λp)

1.94

1.90

2.01

0.75

σ(λp)·∆λp

5.69

5.14

4.64

1.57

σ(λp)·τR

5.41

5.98

6.27

5.30

G5/2⟶6H9/2

AC C

EP

λp

4

564

λp

σ(λp) ·τR 4

BaAS

SC

G5/2⟶6H5/2

M AN U

4

CaAS

TE D

4

MgAS

RI PT

Table 6 Emission peak wavelength λP (nm), effective bandwidths ∆λP (nm), stimulated emission cross-sections σ(λp) (10-22 cm2), gain bandwidth (σ(λp) · ∆λp) (10-28 cm3) and optical gain (σ(λp)·τR) (10-25 cm2·s) for the prominent emission transitions of Sm3+ doped AS glasses.

G5/2⟶6H11/2

21

ACCEPTED MANUSCRIPT

Table 7 CIE color coordinates of the Sm3+ doped AS glasses. x

y

MgAS

0.5947

0.4002

CaAS

0.5955

0.3979

SrAS

0.5959

0.3973

BaAS

0.5934

SC

RI PT

Sample

AC C

EP

TE D

M AN U

0.3976

22

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

Fig. 1. Raman spectra of the prepared Sm3+ doped aluminosilicate glasses.

23

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Fig. 2. Absorption spectra of the prepared Sm3+ doped aluminosilicate glasses.

24

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Fig. 3. Excitation spectra of the prepared Sm3+ doped aluminosilicate glasses (Emission: 600 nm).

25

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Fig. 4. Emission spectra of the prepared Sm3+ doped aluminosilicate glasses in the visible and the NIR range under 402 nm excitation.

26

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Fig. 5. Partial energy level diagram of Sm3+.

27

TE D

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

Fig. 6. Semilogarithmic decay curves of the 4G5/2 level luminescence of Sm3+ for the prepared glass compositions.

28

M AN U

SC

RI PT

ACCEPTED MANUSCRIPT

AC C

EP

TE D

Fig. 7. CIE chromaticity diagram of the prepared Sm3+ doped aluminosilicate glasses.

29

ACCEPTED MANUSCRIPT

Highlights

RI PT

• A set of optical amplification material was synthesized by melt quenching method from peralkaline aluminosilicate glasses doped with samarium III • A complete Judd–Ofelt calculation was carried out on the Sm3+ doped aluminosilicate glasses

SC

• Spectroscopic properties of Sm3+ in the luminescent peralkaline aluminosilicate glasses were evaluated

AC C

EP

TE D

M AN U

• Structural properties were analyzed using Raman spectroscopy