Effects of Eu:Ag codoping on structural, magnetic and mechanical properties of lead tellurite glass ceramics

Journal of Non-Crystalline Solids 408 (2015) 18–25

Contents lists available at ScienceDirect

Journal of Non-Crystalline Solids journal homepage: www.elsevier.com/ locate/ jnoncrysol

Effects of Eu:Ag codoping on structural, magnetic and mechanical properties of lead tellurite glass ceramics Eugen Nicolae Culea a,b, Petru Pascuta b,⁎, Marius Pustan b, Daniela Roxana Tamas-Gavrea b, Lidia Pop b, Ioan Vida-Simiti b a b

SC Emerson, 4 Emerson street, 400641 Cluj-Napoca, Romania Technical University of Cluj-Napoca, 103-10 Muncii Avenue, 400641 Cluj-Napoca, Romania

a r t i c l e

i n f o

Article history: Received 5 June 2014 Received in revised form 26 September 2014 Accepted 7 October 2014 Available online xxxx Keywords: Lead tellurite glass ceramics; Europium ions; Silver oxide; Silver nanoparticles; XRD; EPR spectroscopy; Magnetic susceptibility; Density; Elastic modulus; Hardness

a b s t r a c t Structural, magnetic and mechanical properties of lead tellurite glass ceramics doped with variable amounts of Eu2O3 (0–10 mol%) and codoped with fixed amounts of Ag2O (0.50 mol%) or Ag metallic nanoparticles, AgNPs, (0.33 mol%) have been studied. The investigation of the samples was performed by X-ray diffraction (XRD), EPR spectroscopy, magnetic susceptibility and elastic modulus measurements. XRD investigation shows that the studied samples were glass ceramics. XRD data were used to determine the nature and the quantitative ratio of the crystalline phases present in the samples. EPR spectroscopy was used to obtain information on the presence of europium and silver paramagnetic centers (Eu2+, Ag0, and Ag2+). No EPR signals assignable to these paramagnetic centers were registered. Magnetic susceptibility data show that the magnetic behavior of the studied glass ceramics is due to the europium ions present in the host matrix as Eu3+ and Eu2+ ions that show a very important clustering tendency. The fact that the paramagnetic Eu2+ ions are involved in aggregates explains why these ions were not detected by EPR in spite of the fact that Eu2+ ions are present in appreciable amounts. The level of the europium doping and the nature of the silver codopants (Ag2O or AgNPs) influence the structural, magnetic and mechanical behavior of the samples. In general, the structural effect of the codoping with Ag2O (that provides Ag+ ions capable to enter the host matrix) is more important than that of the AgNPs. © 2014 Elsevier B.V. All rights reserved.

1. Introduction Glasses and glass ceramics doped with rare earth (RE) ions show interesting properties being desirable for important applications in the field of solid-state lasers, fiber amplifiers, infrared to visible convertors, phosphors, field emission displays, biosensors, solar cells, etc. [1,2]. Between the RE ions, europium is considered extremely interesting since it may confer important spectroscopic properties such as a fluorescence very sensitive to the local environment of the Eu3+ ions [3], a persistent spectral hole burning that can be performed at room temperature (potential applications in high density optical data storage) [4], etc. The properties of glasses doped with noble metal ions or nanoparticles have been extensively studied in the last decade because of their unique optical properties (large third-order nonlinear susceptibility, ultrafast response, etc.) [5] that make them appropriate for important applications in optics, electronics and telecommunications [6,7]. In such cases, the controllable valence state of noble metal ions and the controllable size of noble metal nanoparticles are essential. The noble metal-RE codoping of glasses and glass ceramics may produce important structural and behavioral changes to the host material ⁎ Corresponding author. E-mail address: [email protected] (P. Pascuta).

http://dx.doi.org/10.1016/j.jnoncrysol.2014.10.002 0022-3093/© 2014 Elsevier B.V. All rights reserved.

(i.e., up-conversion or important enhancement of light emission), important for different applications [8,9]. A very interesting but yet controverted effect observed in the case of such materials was the enhancement of the RE ion emission, assigned to surface plasmonic resonance effects or to an energy transfer from non-plasmonic noble metal ion aggregates to the RE ions [10]. Obviously, an appropriate host glass or glass ceramic of the codopants is very important in determining the potential applications. In this view, TeO2-based glasses are very promising host matrices for optical switching and broadband amplification [11]. For practical applications of optical functional glasses and glass ceramics, it is of importance to improve their elastic and mechanical properties. Although numerous papers on optical properties of TeO2 based glasses have been reported so far, elastic and mechanical properties of such glasses have been seldom studied [12–14]. Due to the mentioned reasons, as part of an ongoing research, recently we reported aspects on the structure and luminescence enhancement of some Eu:Ag codoped lead–tellurite glasses and glass ceramics where silver was added in the form of ions (Ag+) or of nanoparticles (AgNPs) [15]. Herein, we report on structural, magnetic and mechanic behavior of these glasses. The investigation was performed by using X-ray diffraction (XRD), electron paramagnetic resonance (EPR) spectroscopy, magnetic susceptibility, density, elastic modulus and hardness

E.N. Culea et al. / Journal of Non-Crystalline Solids 408 (2015) 18–25

measurements. The aim of this study was to observe the effect of Eu:Ag codoping on the structural and behavioral properties of the TeO2–PbO glasses. EPR investigation may provide important information concerning the presence of paramagnetic silver and europium ion species and agglomerates in the host matrix. Magnetic susceptibility measurements may provide important information concerning the clustering process of magnetic europium and silver species that are extremely important for potential applications. Data concerning magnetic, density and mechanical properties (elastic modulus, hardness) of the studied samples are discussed in relation with structural data.

19

Magnetic susceptibility measurements were performed on a Faraday type balance in the 80–300 K temperature range. The sensitivity of the equipment was 10− 8 emu/g and measurements were performed on samples having a mass of 0.200–0.300 g. The density of the samples was measured using Archimedes' method with deionized water as the immersion liquid. Mass was measured using a digital balance with the sensitivity of 0.1 mg. Nanoindentation experiments were performed using an AFM XE 70 atomic force microscope equipped with a nanoindentation module. A three-sided pyramid diamond indenter (Berkovich type) attached by a cantilever with high stiffness (144 N/m) was used.

2. Experimental Samples from the TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO–AgNPs– Eu2O3 systems were prepared by the melt quenching technique using reagent grade compounds (Alfa Aesar, Germany). Thus, TeO2, PbO, Eu2O3 and Ag2O or metallic Ag nanoparticles (20–40 nm particle size) were mixed in suitable quantities according to the chemical compositions listed in Table 1. Note that the amount of Eu2O3 increases in the samples up to 10 mol%, while silver is added in constant amounts, 0.5 mol% Ag2O or 0.33 mol% AgNPs. The mixtures were milled in an agate ball mil for 30 min and after that were melted in air, in corundum crucibles, at 1000 °C for 15–25 min in an electric furnace. The melts were cooled to room temperature by pouring onto stainless-steel plates. In general, as resulting from the literature, the method used to produce AgNPs in glasses or glass ceramics consists in the addition of Ag2O followed by a thermal coprecipitation [16]. Alternatively, the direct codoping with AgNPs is also possible. This method permits a better control of the distribution and size of AgNPs and the removal of some undesirable effects of thermal treatment (i.e., crystallization of host material) but shows also an important disadvantage, namely the fact that the tuning of the AgNPs size is not possible. In this work, we performed a comparative study of some effects produced by the codoping with Ag2O or AgNPs on structural, magnetic and mechanical properties of TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO– AgNPs–Eu2O3 compounds. In this view, we realized, one hand, a codoping with Ag2O that ensures the presence of Ag+ ions in the host matrix (samples S in Table 1) and, on the other hand, a direct codoping with AgNPs (samples S′ in Table 1). The comparison between the two series of samples offers the possibility to make a clear separation of the potential structural and behavioral effects due to the nature of the codopant (Ag+ ions or AgNPs). The X-ray diffraction measurements of the studied samples were made by a XRD-6000 Shimadzu diffractometer, with a monochromator of graphite for Cu-Kα radiation (λ = 1.54 Å) at room temperature. The EPR measurements of powder samples were carried out in the X-band (9.4 GHz) at room temperature using an ADANI EPR spectrometer using equal quantities of samples.

3. Results and discussion 3.1. XRD data Some XRD investigation data of the TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO–AgNPs–Eu2O3 systems were reported in our earlier work [15]. The analysis of the XRD data shows that samples S1–S4 and S1′– S4′ contain large amounts of amorphous phase where a small quantity of crystalline phase is dispersed. This crystalline phase was identified as the Pb2Te3O7 [15]. In addition, in samples S5–S7 and S5′–S7′ a second crystalline phase appears, too. This crystalline phase was identified as the Eu2Te4O11. Note that the crystalline Eu2Te4O11 appears in appreciable amounts only in the samples codoped with Ag2O (S6 and S7), while in the corresponding samples containing AgNPs it was not observed (S6′) or appears in a very small amount (S7′). The degree of crystallinity of the samples, Xc, was evaluated following the procedure described by authors [17]. To assess Xc, the Reflex computer program (part of Material Studio software suit) was used [18]. The compositional evolution of Xc is presented in Fig. 1. The analysis of the data from Fig. 1 shows that increasing the europium oxide content of the samples increases the degree of crystallinity for both series of samples (codoped with Ag2O and respectively with AgNPs). Note the important increase of Xc that occurs for the 1 mol% → 3 mol% Eu2O3 content transition for both series of samples (S4 → S5 and S4′ → S5′). The addition of the codopant, Ag2O or AgNPs, enhances the crystallization process, too. Xc values are higher for the samples codoped with Ag2O (S3–S7) in comparison with those codoped with AgNPs (S3′–S7′). The fact that the addition of Ag2O produces a more important devitrification than that produced by AgNPs

Table 1 Chemical composition and elastic properties of TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO– AgNPs–Eu2O3 glass ceramics. υcalc

Sample no.

Composition (mol%) TeO2

PbO

Ag2O

AgNPs

Eu2O3

S1 S2 S3 S3′ S4 S4′ S5 S6 S6′ S7 S7′

80 80 80 80 80 80 80 80 80 80 80

20 19 19.50 19.67 18.50 18.67 16.50 14.50 19.67 9.50 9.67

0 0 0.50 – 0.50 – 0.50 0.50 – 0.50 –

– – – 0.33 – 0.33 – – 0.33 – 0.33

0 1 0 0 1 1 3 5 5 10 10

0.221 0.228 0.225 0.228 0.229 0.226 0.218 0.226 0.225 0.232 0.228

Ecalc

EIT

HIT

(GPa)

(GPa)

(GPa)

48.09 49.55 48.83 49.15 49.89 49.66 48.64 59.70 50.48 53.51 52.67

42.23 – 37.51 41.85 35.82 30.12 – 37.59 46.35 – –

18.73 – 2.58 8.31 3.68 7.48 – 4.16 0.54 – –

Fig. 1. Compositional dependence of the crystallinity of TeO2–PbO–Ag2O–Eu2O3 and TeO2– PbO–AgNPs–Eu2O3 glass ceramics (lines are only a guide for the eyes).

20

E.N. Culea et al. / Journal of Non-Crystalline Solids 408 (2015) 18–25

is due to the silver ions that enter the host glass ceramic matrix modifying its structure which is not the case for the AgNPs.

3.2. EPR spectroscopy EPR measurements may provide important information concerning the Ag and Eu paramagnetic ions and aggregates. In the studied TeO2– PbO–Ag2O–Eu2O3 and TeO2–PbO–AgNPs–Eu2O3 glass ceramics, due to the starting raw materials and to the preparation conditions, the presence of Eu3+ and Ag+ ions is expected but these are not paramagnetic species. However, as it was previously reported, both europium and silver may appear in oxide glasses in multiple valence states and aggregates some of them being paramagnetic [19–28]. Then, these paramagnetic species, including the Ag2+ and Eu2+ ions as well as some silver and europium aggregates, can be identified by EPR spectroscopy at room temperature. EPR spectra of the TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO–AgNPs– Eu2O3 glass ceramics are presented in Fig. 2. These spectra were obtained using the maximum signal amplification available for our EPR equipment. The EPR spectra of the host matrix, free of Ag and Eu (80%TeO2–20%PbO, S1), as well as of the samples free of Ag (80%TeO2–19%PbO–1%Eu2O3, S2) and free of Eu (80%TeO2 –19.5%PbO0.5%Ag 2O, S3, and 80%TeO 2 – 19.67%PbO–0.33%AgNPs, S3′) are presented for comparison. Examination of Fig. 2 shows that these EPR spectra are not sensible to the doping with europium and silver since all the resonance signals appear starting with the host glass and remain almost unchanged after the doping. Then, note that the very weak EPR signals are not related to paramagnetic Ag and/or Eu species being generated probably by

some small amounts of paramagnetic impurities and/or paramagnetic structural defects. As mentioned before, glasses doped with Ag2O may present a variety of silver isolated and agglutinated ion species [19–25]. It was shown that in aluminoborate glasses, under oxidizing or reducing conditions, the Ag+ ions initially present in the melt evolve Ag2+ or Ag0 [21]. The Ag+ ions act as electron trapping centers generating Ag0 atoms. Since the Ag0 atoms are not stable, interaction with the Ag+ ions takes place clusters. Modifications of the silver species leading to more stable Agx+ y present in the glass may take place not only during the melting but also after glass elaboration as a result of heating processes [21–23]. Thus, appropriate annealing may produce the excitation of electrons from Ag0 atoms to form Ag+ ions [23]. After that, the high mobility of silver and the interactions between Ag0 and Ag+ contribute to the formation of silver clusters. Irradiation of silver doped glasses leads to similar results. IR laser irradiation of aluminoborate glasses doped with silver leads to the silver clusters [21]. Gamma irradiation of localized formation of Agx+ y silver zinc phosphate glasses also produces several silver ion species and aggregates [22]. Some of the mentioned silver species are paramagnetic: Ag0 (with the 4d105s1electronic configuration), Ag++ (with the 4d9 electronic configuration) as well as some silver nanoclusters [19–25]. Thus, at room temperature, EPR signals attributed to Ag0 atoms occur at g ~ 2.2 and g ~ 1.8 [21], while signals attributed to Ag2 + occur at g⊥ ~ 2.05 and g∥~2.35 [20]. Concerning the paramagnetic silver clusters we men2+ that gention Ag+ 2 that generates the EPR signal at g ~ 1.994 and Ag3 erates the absorption at g ~ 1.980 [26–28]. However, as shown in Fig. 2, no EPR signal assignable to silver was detected in the studied TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO– AgNPs–Eu2O3 glass ceramics. Thus, the analysis of these EPR spectra declines the presence of paramagnetic Ag species (Ag0, Ag++ or paramagnetic nanoclusters) [19–27]. Then, EPR spectroscopy suggests that the reduction of the Ag+ ions to Ag0 atoms takes no place in the case of the studied TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO–AgNPs–Eu2O3 glass ceramics. This is explained by the relatively short melting times used to prepare our samples. After that, since appreciable amounts of Ag0 were not observed, the possibility of the apparition of silver clusters is negligible, too. This fact agrees our expectations that the samples codoped with Ag2O contain only Ag+ ions and no other silver ionic or agglutinated species. Concerning the presence of europium paramagnetic species in the studied glass ceramics, note that previous reports have shown that EPR signals located at g ~ 2.0, g ~ 2.8 and g ~ 6.0 (the so-called characteristic “U” spectrum) and at g ~ 4.6 may appear due to the Eu2+ ions [25, 27–31]. The negative EPR result consisting in the absence of resonance signals assignable to paramagnetic europium species suggests the absence of paramagnetic Eu2 + ions in the studied TeO2–PbO–Ag2O– Eu2O3 and TeO2–PbO–AgNPs–Eu2O3 glass ceramics. To be more precise, if Eu2+ ions are still present in these glass ceramics, their amounts are beyond the EPR detection limit. 3.3. Magnetic susceptibility data

Fig. 2. EPR spectra of TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO–AgNPs–Eu2O3 glass ceramics.

Magnetic behavior of the TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO– AgNPs–Eu2O3 glass ceramics was assumed to be due to the presence of the europium ions since magnetic susceptibility measurements showed that the host matrices 80%TeO 2–20%PbO (sample S1), 80%TeO2 – 19.5%PbO0.5%Ag2O (sample S3) and 80%TeO2–19.67%PbO–0.33%AgNPs (sample S3′) were diamagnetic. Thus, for these samples the χ = 4.33 × 10− 7 (S1), χ = 5.62 × 10− 7 (S3) and χ = 4.27 × 10− 7 (S3′) magnetic susceptibility values were found. Consequently, the experimental magnetic susceptibility values obtained for the samples doped with europium were corrected taking into account the diamagnetic contribution of the host matrices. Fig. 3 presents the temperature dependence of the inverse magnetic susceptibility for the studied TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO–

E.N. Culea et al. / Journal of Non-Crystalline Solids 408 (2015) 18–25

21

Table 2 Magnetic parameters (paramagnetic Curie temperature — θp, molar Curie constant — CM and effective magnetic moment per RE ion, μeff) of TeO2–PbO–Ag2O–Eu2O3 and TeO2– PbO–AgNPs–Eu2O3 glass ceramics.

Fig. 3. Temperature dependence of the inverse magnetic susceptibility of the TeO2–PbO– Ag2O–Eu2O3 and TeO2–PbO–AgNPs–Eu2O3 glass ceramics.

AgNPs–Eu2O3 glass ceramics. The data collapse to straight lines indicating a Curie–Weiss type behavior described by the −1

χ


 ¼ T−θp =C

ð1Þ

equation where C is the Curie constant, θp is the paramagnetic Curie temperature and T is the temperature. By fitting the experimental magnetic susceptibility data according to Eq. (1), we determined the C and θp values correspond to the samples. After that, the effective magnetic moment per europium ion, μeff, was calculated using the μ eff ¼ 2:827

rffiffiffiffiffiffiffi CM 2x

ð2Þ

where CM represents the molar Curie constant and x represents the molar fraction of europium ions in the samples [32]. The obtained values are shown in Table 2. In general, for the oxide glasses containing RE ions (including europium ions), the temperature dependence of the magnetic susceptibility follows a Curie type behavior, magnetic behavior for low RE oxide contents (θp = 0 up to 3–5 mol% RE oxide content depending on the nature of the RE ions and of the host glass) and a Curie–Weiss type one for higher RE contents (θp ≠ 0) [33–37]. In the case of the studied TeO2–

Sample

−θp [K]

CM [emu/mol]

μ [μB/atom]

S4 S6 S7 S4′ S6′ S7′

80 340 365 155 263 350

0.029 0.443 0.821 0.037 0.488 1.192

3.40 5.96 5.73 3.83 6.25 6.90

PbO–Ag2O–Eu2O3 and TeO2–PbO–AgNPs–Eu2O3 glass ceramics, magnetic susceptibility data obey the Curie–Weiss law even for the lowest Eu2O3 content (1 mol%). Note also that the values of the paramagnetic Curie temperatures, θp, shown in Table 2 are higher than those previously reported for other glasses and glass ceramics [33–37] suggesting intense magnetic interaction between the magnetic europium ion species. In order to discuss the compositional variation of θp, note that θp is a rough indicator of the magnetic interactions between the magnetic ions from the studied glasses. Thus, in the frame of the molecular field model [32], the molecular field constant is given by J ≈ 2NcJij/Ng2μ2B ≈ θp/C where N is the total number of magnetic ions, Nc is the number of exchange coupled magnetic ions, g is the spectroscopic splitting factor, μB is the Bohr magneton and Jij is the exchange interaction between the magnetic ions. In this view, the zero values for θp suggest that magnetic RE ions appear in the host matrix as isolated species, while the values different from zero suggest that magnetic ions appear as both isolated and coupled species. Table 2 shows that the values measured for θp are different from zero and negative for all the composition range. The θp values different from zero even for very low Eu2O3 contents suggest a much accentuated clustering tendency of the europium ions in the studied samples. The negative θp values suggest the antiferromagnetic nature of the interactions between the magnetic europium ions from the studied glass ceramics. Previous reports stated that the magnetic RE ions from vitreous matrices are coupled via RE–O–RE superexchange interactions [33–36]. As was mentioned before, the θp values determined for the studied glass ceramics are higher than those reported previously for other glasses and glass ceramics. This behavior may be related to the partial crystallization of the studied samples that ensures shorter mean distances between the magnetic europium ions than those available in amorphous samples. To analyze the effective magnetic moment values, μeff, determined for the TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO–AgNPs–Eu2O3 glass ceramics, note that, as was previously reported, europium ions may appear in oxide glasses in multiple valence states, usually 3 + and 2 + [33–36]. Note that both Eu3 + and Eu2 + are magnetic species, Eu3 + with the 3.4μB effective magnetic moment per europium ion [35] and Eu2 + with the 7.94 μB theoretical magnetic moment of the free ion [37]. The μeff values shown in Table 2 suggest that in the studied TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO–AgNPs–Eu2O3 glass ceramics europium ions appear as both Eu2+ and Eu3+ since the calculated μeff values are between those corresponding to these valence states. Using the calculated μeff values from Table 2 and a procedure previously reported [38], we determined the fraction f = Eu2+/Eutotal (where Eu2+ represents the amount of europium ions in their 2+ valence state and Eutotal represents the total amount of europium ions present in the host matrix). In this view the used equations: 2

2

2

x  μ eff ¼ x1  μ Eu2þ þ x2  μ Eu3þ

ð3Þ

and x ¼ x1 þ x2

ð4Þ

22

E.N. Culea et al. / Journal of Non-Crystalline Solids 408 (2015) 18–25

where x = molar content of europium ions, x1 = molar content of Eu2+ ions, x2 = molar content of Eu3+ ions, μ Eu2þ ¼ 7:9μ B (the theoretical magnetic moment of the free Eu2+ ions) and μ Eu3þ ¼ 3:4μ B(the effective magnetic moment of Eu3+ ions). The compositional evolution of f fraction is shown in Fig. 4. This plot permits to follow the evolution of the redox equilibrium between the valence states of europium ions in the melts of the studied samples. The data from Fig. 4 show that f has higher values in comparison with those previously reported for other oxide glasses doped with europium [33–36]. In the case of samples codoped with AgNPs, the f values increase with increasing the Eu2O3 content for all the compositional ranges while for the samples codoped with Ag2O, the f values increase up to x = 0.05 and after that remains relatively constant. The f values are higher for the samples codoped with AgNPs in comparison with those codoped with Ag2O suggesting that the reduction process is more advanced in the first case. Note also that the high f = Eu2+/Eutotal fractions determined based on magnetic data seem to be in contradiction with the fact that EPR data did not reveal the presence of Eu2 + ions. To explain this controversy, we have to take into account that, as shown by the magnetic data, the europium ions present in the studied samples have a much accentuated clustering tendency appearing as aggregates even for very low Eu2O3 contents. Under such conditions, it is possible that the Eu2+ ions involved in aggregates will not be detected by EPR spectroscopy even if they are present in appreciable amounts.

3.4. Density data Density measurements may offer interesting information related to the structural changes that occur in glasses and glass ceramics. In general, the density of glasses and glass ceramics can be explained in terms of a competition between the size and compactness of the various structural groups present in the sample. Accordingly, density is related to how tightly the ions and ionic groups are packed together in the structure. Fig. 5 shows the density–Eu2O3 content relation for the TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO–AgNPs–Eu2O3 samples. A simple survey of this figure shows that the host matrix is sensible to the nature and level of dopants, Eu2O3, Ag2O and AgNPs. Increasing the Eu2O3 content of samples produces a non-linear variation of density. This fact suggests that the addition of europium ions modifies the structural units present in the host matrix and is in agreement with previous reports [34]. As shown in Fig. 5, the addition of Ag2O exerts a more important effect on the density that addition of AgNPs. Thus, increasing of AgNP

Fig. 5. Compositional dependence of density of TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO– AgNPs–Eu2O3 glass ceramics (lines are only a guide for the eyes).

content of the samples generates a relatively monotonic decrease of density, while the addition of Ag2O determines an irregular variation of density. This effect may be explained by the fact that the added AgNPs are simply inserted to the host matrix, while the doping with Ag2O provides Ag ions that, as well as europium ions, enter this matrix and modify its structural units playing a network modifier role. 3.5. Nanoindentation and elastic properties Data concerning elastic properties of RE doped glasses and glass ceramics, namely Young's modulus (E) and Poisson's ratio (ν), are important in view of their potential applications. Makishima and Mackenzie [39,40] proposed the most used theoretical model to calculate E and ν in terms of chemical composition of glasses. Thus, E and ν can be calculated using the packing density (Vt) and the dissociation energy per unit volume (Gt) of glasses and glass ceramics, as follows: E ¼ 2Vt  Gt

ν ¼ 0:5−

1 : 7:2  Vt

ð5Þ

ð6Þ

For multi-component glasses Makishima and Mackenzie reported the following relations for calculating Vt and Gt: Vt ¼

Gt ¼

ρX V x M i i i X

Gi  xi

ð7Þ

ð8Þ

i

Fig. 4. Compositional dependence of the f = Eu2+/Eutotal fraction of TeO2–PbO–Ag2O– Eu2O3 and TeO2–PbO–AgNPs–Eu2O3 glass ceramics (lines are only a guide for the eyes).

where ρ = density, M = effective molecular weight of the samples, xi = mole fraction, Vi = packing factor and Gi = dissociation energy per unit volume of the component oxide i. We calculated Vt and Gt for the TeO2–PbO–Ag2O–Eu2O3 and TeO2– PbO–AgNPs–Eu2O3 glass ceramics by using Eqs. (7) and (8). The values of the parameters required in order to calculate Vt and Gt were taken from literature [41–44]. The calculated values obtained for Poisson's ratio and Young's modulus of the studied samples, υcalc and Ecalc, are shown in Table 1.

E.N. Culea et al. / Journal of Non-Crystalline Solids 408 (2015) 18–25

23

Fig. 6. Contact stiffness S = dF/dh (a) and indentation depth hc (b) (after Ref. [45]).

Nanoindentation experiments were carried out on some of the studied samples for evaluation of nanoindentation Young's modulus, EIT, and hardness, HIT. Typical load/unload vs. displacement curve is shown in Fig. 6a. Here, Fmax is the maximum indentation force and hmax is the depth beneath the material free surface. The contact depth hc and the slope of the elastic unloading dF/dh allow to calculate EIT and HIT. Fig. 6b presents a schematic illustration of the unloading process showing the parameters characterizing the contact geometry [45,46]. To analyze the load/unload–displacement curves in order to obtain EIT and HIT values, we used the method given by Oliver and Pharr [45]. EIT of the studied samples was calculated by using equation 1 1−υ2 1−υ2i ¼ þ EIT Ei Eeff

ð9Þ

where Eeff is the effective elastic modulus of the samples, ν is the Poisson's ratio of the studied samples, while Ei and υi are Young's modulus and Poisson's ratio for the indenter. In this study, a Berkovich diamond nanoindenter was used. The elastic parameters for diamond are Ei = 1140 GPa and νi = 0.07 [47]. Eeff was found from nanoindentation test data and according to Oliver and Pharr's method is given by the equation: Eeff ¼

pffiffiffi π S  pffiffiffiffi 2β A

ð10Þ

where S = dF/dh is the contact stiffness, as shown in Fig. 6a, β is a constant depending on the geometry of the indenter used, for the Berkovich type indenter being β = 1.034. A is the projected contact area of the indentation at the maximum load Fmax. This parameter is measured as a function of the indentation depth h c , as shown in Fig. 6b. For the Berkovich indenter, the projected contact area is A ≈ 24.5 ⋅ h2c . The nanoindentation hardness for the studied samples

was calculated according to Oliver and Pharr's method using the equation: HIT ¼

F max : A

ð11Þ

As an example, Fig. 7 shows the load/unload displacement curve obtained at room temperature in air for sample S6. The values determined for EIT and HIT for the studied TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO– AgNPs–Eu2O3 glass ceramics are shown in Table 1. Note that calculated and experimental data are of the same order of magnitude. To discuss the compositional evolution of Young's modulus EIT and hardness HIT of the studied TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO– AgNPs–Eu2O3 glass ceramics, their dependences versus the Eu2O3 content of the samples were plotted and are shown in Figs. 8 and 9. Comparing the data from Table 1 and Fig. 8, we observe that increasing Eu2O3 content of the samples determines an increase of calculated Ecalc values while the experimental EIT values first decrease for sample with 1 mol% Eu2O3 and after that increase for sample with 5 mol% Eu2O3. The differences are due to the fact that the compositional evolution of theoretical values simulates the case of a higher mass dopant added to the host matrix considered not sensitive to the dopant. Actually, experimental data (i.e., IR data) suggest that structural modifications produced by dopants occur in the host matrix. The compositional evolution of the Young's modulus suggests that the dopants may improve the elastic properties of the host matrix, but this depends on the nature of the dopant and on the structural modifications produced to the host matrix. Having in mind that structural modifications induced by the addition of dopants/codopants influence the elastic properties of the samples, the comparison of Figs. 1 and 8 suggests that the higher elasticity of samples codoped with AgNPs (except S4′) may be related to their lower degree of crystallinity.

24

E.N. Culea et al. / Journal of Non-Crystalline Solids 408 (2015) 18–25

Fig. 7. Load/unload displacement curve at room temperature for sample S6.

Fig. 9 shows a very different effect of the Ag codoping on the compositional evolution of hardness for the studied TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO–AgNPs–Eu2O3 glass ceramics, depending on the nature of codopant. Thus, increasing Eu2O3 content of the samples leads to the increase of HIT for the samples codoped with Ag2O but to a decrease of HIT for the samples codoped with AgNPs. This may be explained by the fact that the addition of Ag2O provides silver ions that enter the host matrix and produce structural modifications that are, in general,

as shown previously, more important than those produced by the addition of AgNPs.

Fig. 8. Compositional dependence of elastic modulus of TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO–AgNPs–Eu2O3 glass ceramics (lines are only a guide for the eyes).

Fig. 9. Compositional dependence of hardness of TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO– AgNPs–Eu2O3 glass ceramics (lines are only a guide for the eyes).

4. Conclusions Glass ceramics of the TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO– AgNPs–Eu2O3 systems were prepared by the melt quenching technique.

E.N. Culea et al. / Journal of Non-Crystalline Solids 408 (2015) 18–25

The doping of the samples was realized by the addition of Eu2O3 in variable amounts, from 0 to 10 mol%. Two series of samples were obtained by a codoping with constant amounts of silver, namely 0.5 mol% Ag2O or 0.33 mol% AgNPs. XRD data show that the samples with up to 1 mol% Eu2O3 contain large amounts of amorphous phase where a small quantity of crystalline phase is dispersed, while those with higher amounts of Eu2O3 contain appreciable amounts of crystalline phases, identified as Pb2Te3O7 and Eu2Te4O11. Increasing the Eu2O3 content of the samples increases their degree of crystallinity, Xc. The Xc values are higher for the samples codoped with Ag2O in comparison with those codoped with AgNPs. EPR spectra show no resonance signals assignable to silver in the studied TeO2–PbO–Ag2O–Eu2O3 and TeO2–PbO–AgNPs–Eu2O3 glass ceramics. The analysis of EPR spectra declines the presence of paramagnetic Ag species (Ag0, Ag++ or paramagnetic nanoclusters) and confirms that the samples codoped with Ag2O contain only Ag+ ions and no other silver ionic or agglutinated species while those codoped with AgNPs contain only these nanoparticles. EPR spectra show no resonance signals due to the Eu2+ ions. Magnetic susceptibility data suggest the presence of europium ions with two valence states, 3+ and 2+. Apparently, there is a controversy between the magnetic data, that reveal the presence of appreciable amounts of Eu2 + ions, and the EPR data, that did not reveal their presence. This may be solved taking into account that the Eu2 + ions are involved in aggregates that may hide the paramagnetic character of the participant ions. Density, elastic modulus and hardness strongly depend on the europium doping level as well as on the nature of codopant (Ag2O or AgNPs). In general, the structural effect of Ag2O, which provides Ag+ ions capable to enter the host matrix, is more important than that of the AgNPs. Acknowledgments One of the authors (ENC )would like to thank UEFISCDI for the financial support offered by contract ID 137070/2014 — ATRACTING. References [1] S.A. Lourenco, N.O. Dantas, E.O. Serqueira, W.E.F. Ayta, A.A. Andrade, M.C. Filadelpho, J.A. Sampaio, M.J.V. Bell, M.A. Pereira da Silva, Eu3+ photoluminescence enhancement due to thermal energy transfer in Eu2O3-doped SiO2–B2O3–PbO2 glasses system, J. Lumin. 131 (2011) 850–855. [2] Y. Tian, B. Chen, R. Hua, J. Sun, H. Zhong, Y. Zheng, T. Yu, L. Huang, H. Yu, Optical transition, electron–phonon coupling and fluorescent quenching of La2(MoO4)3: Eu3+ phosphor, J. Appl. Phys. 109 (2011) 053511. [3] R. Rodriquez-Mendoza, I.R. Martin, V.D. Rodriquez, Solids optical spectroscopy analysis of the Eu3+ ions local structure in calcium diborate glasses, J. Non-Cryst. Solids 319 (2003) 200–216. [4] W. Chung, J. Heo, Room temperature persistent spectral hole burning in X-ray irradiated Eu3+ doped borate glasses, Appl. Phys. Lett. 79 (2001) 326. [5] D. Ricard, P. Roussignol, C. Flytzanis, Surface-mediated enhancement of optical phase conjugation in metal colloids, Opt. Lett. 10 (1985) 511–516. [6] F. Hache, D. Ricard, C. Flytzanis, Optical nonlinearities of small metal particles: surface-mediated resonance and quantum-size effects, J. Opt. Soc. Am. B 3 (1986) 1647–1953. [7] X.G. Huang, M.R. Wang, Y. Tsui, C. Wu, Characterization of erasable inorganic photochromic media for optical disk data storage, J. Appl. Phys. 83 (1998) 3795. [8] L.R.P. Kassab, F.A. Bomfim, J.R. Martinelli, N.U. Wetter, J.J. Neto, B. Cid de Araujo, Energy transfer and frequency upconversion in Yb3+–Er3+-doped PbO–GeO2 glass containing silver nanoparticles, Appl. Phys. B 94 (2009) 239–242. [9] D.M. da Silva, L.R.P. Kassab, J.R. Martinelli, Cid B. de Araujo, Production and characterization of RF-sputtered of PbO–GeO2 amorphous thin films containing silver and gold nanoparticles, J. Non-Cryst. Solids 356 (2010) 2602–2605. [10] H. Guo, J. Jing Li, F. Li, H. Zhang, Origin of white luminescence in Ag–Eu co-doped oxyfluoride glasses, J. Electrochem. Soc. 158 (6) (2011) J165–J168. [11] J.S. Wang, E.M. Vogel, E. Snitzer, Tellurite glass: a new candidate for fiber devices, Opt. Mater. 3 (1994) 187. [12] K. Narita, Y. Benino, T. Fujiwara, T. Komatsu, Vickers nanoindentation hardness and deformation energy of transparent erbium tellurite nanocrystallized glasses, J. NonCryst. Solids 316 (2003) 407–412. [13] F. Torres, Y. Benino, T. Fujiwara, T. Komatsu, Evaluation of elastic/mechanical properties of some glasses and nanocrystallized glass by cube resonance and nanoindentation methods, Mater. Res. Bull. 39 (2004) 1431–1443.

25

[14] R. El-Mallawany, H. Afifi, Elastic moduli and crosslinking of some tellurite glass systems, Mater. Chem. Phys. 12 (2013) 1–4. [15] E. Culea, I. Vida-Simiti, Gh. Borodi, E.N. Culea, R. Stefan, P. Pascuta, Structural and spectroscopic effects of Ag–Eu3+ codoping of TeO2–PbO glass ceramics, J. Mater. Sci. 49 (13) (2014) 4620–4628, http://dx.doi.org/10.1007/s10853-014-8164-9. [16] B. Cid Araújo, L.R.P. Kassab, R.A. Kobayashi, L.P. Naranjo, P.A. Santa Cruz, Luminescence enhancement of Pb2+ ions in TeO2–PbO–GeO2 glasses containing silver nanostructures, J. Appl. Phys. 99 (2006) 123522–123524. [17] H.P. Klug, L.E. Alexander, X-ray Diffraction Procedures, 2nd ed. John Wiley & Sons Inc., 1974. 687–703. [18] Accelrys Software Inc., Materials Studio, Release 5.5, 2010. [19] E. Malchukova, B. Boizot, Reduction of Eu3+ to Eu2+ in aluminoborosilicate glasses under ionizing radiation, Mater. Res. Bull. 45 (2010) 1299–1303. [20] R. Yokota, H. Imagawa, ESR-studies of radiophotoluminescence centers in silver activated phosphate glasses, J. Phys. Soc. Jpn. 23 (1967) 1038. [21] B. Hua, J. Qiu, Y. Shimotsu, K. Fujita, K. Hirao, Photoreduction of Ag+ in alumino borate glasses induced by irradiation by a femtosecond laser, J. Mater. Res. 20 (2005) 644–648. [22] C. Maurel, T. Cardinal, M. Bellec, L. Canioni, B. Bousquet, M. Treguer, J.J. Videau, J. Choi, J. Richardson, Luminescence properties of silver zinc phosphate glasses following different irradiations, J. Lumin. 129 (2009) 1514–1518. [23] A.S. Kuznetsov, T. Ngo Cuong, V.K. Tikhomirov, M. Jivanescu, A. Stesmans, L.F. Chibotaru, J.J. Velázquez, V.D. Rodríguez, D. Kirilenko, G. Van Tendeloo, V.V. Moshchalkov, Effect of heat-treatment on luminescence and structure of Ag nanoclusters doped oxyfluoride glasses and implication for fiber drawing, Opt. Mater. 34 (2012) 616–621. [24] A.T. Brant, B.E. Kananan, M.K. Murari, J.W. McClory, J.C. Petrosky, V.T. Adamiv, Ya.V. Burak, P.A. Dowben, L.E. Halliburton, Electron and hole traps in Ag-doped lithium tetraborate (Li2B4O7) crystals, J. Appl. Phys. 110 (2011) 093719-1–093719-7. [25] Sijun Fan, Chunlei Yu, Dongbing He, Kefeng Li, Lili Hu, White light emission from γ-irradiated Ag/Eu co-doped phosphate glass under NUV light excitation, J. Alloys Compd. 518 (2012) 80–85. [26] R. Espiau de Lamaestre, H. Béa, H. Bernas, J. Belloni, J.L. Marignier, Irradiated-induced Ag nanocluster nucleation in silicate glasses: analogy with photography, Phys. Rev. B 76 (2007) 205431–205449. [27] J.R. Morton, K.F. Preston, Paramagnetic silver clusters in irradiated Ag. A molecular sieves, J. Magn. Reson. 68 (1986) 121–126. [28] C.M. Brodbeck, L.E. Iton, The EPR spectra of Gd3+ and Eu2+ in glassy systems, J. Chem. Phys. 83 (1985) 4285–4298. [29] T. Hayakawa, M. Nogami, Dynamical Faraday rotation effects of sol–gel derived Al2O3–SiO2 glass containing Eu2+ ions, Solid State Commun. 116 (2000) 77–82. [30] C.F. Zhu, Y. Yang, X. Liang, S. Yuan, G. Chen, Composition induced reducing effects on eu ions in borophosphate glasses, J. Am. Ceram. Soc. 90 (2007) 2984–2986. [31] H. Zeng, Y. Yang, Z. Lin, X. Liang, S. Yuan, G. Chen, L. Sun, The effect of B2O3 on the luminescent properties of Eu ion-doped aluminoborosilicate glasses, J. Non-Cryst. Solids 357 (2011) 2328–2331. [32] L.F. Bates, Modern Magnetism, Cambridge University Press, London, 1962. 133. [33] L. Pop, T. Ristoiu, N. Dempsey, M. Culea, E. Culea, Low-temperature study of magnetic behaviour of europium in lead–bismuthate glasses, Mod. Phys. Lett. B 20 (2006) 1–6. [34] E. Culea, Structural and magnetic behaviour of lead–bismuthate glasses containing rare earth ions, J. Non-Cryst. Solids 357 (2011) 50–54. [35] E. Burzo, I. Ardelean, I. Ursu, Magnetic properties of xEu2O3(1 − x)[3B2O3·PbO] glasses, Mater. Lett. 26 (1996) 103–105. [36] E. Culea, L. Pop, S. Simon, Spectroscopic and magnetic behaviour of xGd2 O 3 (1 − x)(Bi2O3·PbO) glasses, Mater. Sci. Eng. B 112 (1) (2004) 59–63. [37] I. Schellenberg, M. Eul, R. Pottgen, On the magnetic and mossbauer spectroscopic characterization of EuPd2Sb2, Z. Naturforsch. 65b (2010) 18–22. [38] K.N.R. Taylor, M.I. Darby, Physics of Rare Earth Solids, Chapman & Hall, London, 1972. 212. [39] A. Makishima, J.D. Mackenzie, Direct calculation of Young's moidulus of glass, J. NonCryst. Solids 12 (1973) 3–43. [40] A. Makishima, J.D. Mackenzie, Calculation of bulk modulus, shear modulus and Poisson's ratio of glass, J. Non-Cryst. Solids 17 (1975) 147–156. [41] S. Inaba, S. Oda, K. Morinaga, Heat capacity of oxide glasses at high temperature region, J. Non-Cryst. Solids 325 (1–3) (2003) 258–266. [42] A. Abd El-Moneim, Correlation between acoustical and compositional parameters of borate and tellurite glasses, Mater. Chem. Phys. 135 (2012) 653–657. [43] A. Mohajerani, V. Martin, D. Boyd, J.W. Zwanziger, On the mechanical properties of lead borate glass, J. Non-Cryst. Solids 381 (2013) 29–34. [44] S. Inaba, S. Fujino, K. Morinaga, Young's modulus and compositional parameters of oxide glasses, J. Am. Ceram. Soc. 82 (1999) 3501–3507. [45] W.C. Oliver, G.M. Pharr, Measurement of hardness and elastic modulus by instrumented indentation: advances in understanding and refinements, J. Mater. Res. 19 (2004) 3–20. [46] Q. Zhao, M. Guerette, L. Huang, Nanoindentation and Brillouin light scattering studies of elastic moduli of sodium silicate glasses, J. Non-Cryst. Solids 358 (2012) 652–657. [47] K. Shinozaki, T. Honma, T. Komatsu, Morphology and dispersion state of Ba2TiSi2O8 nanocrystals in transparent glass-ceramics and their nanoindentation behavior, J. Non-Cryst. Solids 358 (2012) 1863–1869.