Lead silicate glass SiO₂–PbF₂ doped with luminescent Ag nanoclusters of a fixed site

Abstract

Bulk SiO₂–PbF₂ glass doped homogeneously with luminescent Ag nanoclusters has been synthesized using a melt-quenching technique. A broad range of luminescence from 400 nm to 900 nm can be excited with a broad range of UV and some visible excitation wavelengths. The luminescence spectra have been found to be invariant with excitation wavelength at room temperature, indicating a negligible distribution of Ag cluster sites. The UV excitation efficiency of Ag nanoclusters has been determined by shifting the absorption edge further into the UV using an Al₂O₃ glass-network modifier. The prepared glass is proposed as a low-cost glass phosphor for Hg-free environmentally-friendly white light generation in UV/blue-driven light emitting diodes, in flexible monitors and as down-shifting layers for enhanced solar cells.

---

1. Introduction

Silver nanoclusters, consisting of a few atoms,^1,2 are a novel class of luminescent particles. These nanoclusters show broad band emission covering the range form Blue to NIR with quantum yields from 10 to 70%.^3,4 Ag nanoclusters can be encapsulated in organic scaffolds,² zeolites^4,5 and glasses.^6,7 Potential applications of the luminescent Ag nanoclusters include bio-compatible luminescent labels,² novel UV/blue driven white light^8–10 phosphors for light emitting diodes (LEDs), down conversion layers for enhanced solar cells^7,11 and optical storage.^12–14

Oxyfluoride glasses have been invented as transparent host for luminescent Ag nanoclusters preserving its luminescent properties from degradation due to an influence of environment, e.g. in.^6–8 Nowadays, conventional oxyfluoride host for Ag nanoclusters consists of many components, for example SiO₂–AlO_1.5–PbF₂–CdF₂–ZnF₂, e.g. in.⁶ Growth mechanism of the Ag nanoclusters in this oxyfluoride glass and their luminescence have been investigated experimentally^6,7 and by quantum chemistry methods.^15,16

In this paper we synthesize a much more simple chemical formulation lead silicate glass SiO₂–PbF₂ doped with luminescent Ag nanoclusters by still the simple melt quenching technique.

2. Experimental

The bulk oxyfluoride glasses doped with luminescent Ag nanoclusters were prepared using conventional melt-quenching method as described elsewhere.^6,7 Briefly, puratronic grade powders SiO₂, PbF₂, and, in some cases, of Al₂O₃ were mixed in batch of 5 g and melted in platinum crucible at 1000 °C for about 10 minutes. After melting in the air these glasses were cast into an aluminium mold hold at the room temperature. The chemical compositions of the glasses were: (i) undoped 55(SiO₂) : 45(PbF₂), mol%, (OF1 glass) (ii) 55(SiO₂) : 45(PbF₂), mol% doped with 1 wt% AgNO₃, (OF2 glass) (iii) 55(SiO₂) : 45(PbF₂), mol%, doped with 5 wt% AgNO₃, (OF3 glass), and (iv) 53(SiO₂) : 45(PbF₂) : 5(AlO_1.5), mol%, doped with 1 wt% AgNO₃ (OF4 glass). Further in the text, these glasses will be called OF1, OF2, OF3 and OF4, respectively. Addition of AgNO₃ to the batch resulted in doping of the glass hosts with homogeneous dispersed luminescent Ag nanoclusters, and this was found very surprisingly indeed, as it was not expected that Ag nanoclusters could disperse homogeneously in this glass host. This simple glass composition has such disadvantages as low glass transition temperature, not large vitreous domain comparing to the system with CdF₂ and possible loss of fluorite component in form of SiF₄.¹⁷ The last one may result in limited solubility of Ag nanoclusters comparing to the previous system. Such glass compositions were selected due to their known good glass-formation.^17,18 The resulting pieces of glass (4 × 1 × 0.3 cm³) were polished and cut for optical measurements.

Absorbance spectra were recorded with a Bruker Vertex 80V Fourier Transform spectrometer. The steady state emission and excitation spectra were obtained by exciting the samples with light from a 300 W Xe arc lamp dispersed by a monochromator. The luminescence was detected with an electron-multiplied CCD camera attached to a Shamrock spectrometer. Low temperature measurements of luminescence have been carried out in a helium flow optical cryostat from the room temperature down to 5 K. Quantum yield of samples was measured using an integrating sphere and set-up previously described in,¹⁹ under excitation at 380 nm.

3. Results and discussion

Fig. 1 shows the absorption spectra of OF1, OF2, OF3 and OF4 glasses. Undoped glass host (OF1) does not absorb in visible range and its ultraviolet absorption edge arises around 345 nm. Absorption spectra of Ag-nanoclusters doped glasses (OF2, OF3 and OF4) display red-shift of UV absorption edge (from 345 nm to 360 nm) and minor broad plasmon absorption band around 460 nm. This broad band is known to be an absorption band of amorphous Ag-nanoparticles of size 1–2 nm.²⁰ Addition of alumina (Al₂O₃) to Ag doped glass host results in considerable blue-shift of UV absorption edge of glass host. UV absorption of glass host may arise from the presence of non-bridging oxygen centres, e.g. in [ref. 21 and refs therein]. Alumina acts as a glass network former and may bridging broken bonds on neighbour oxygens resulting in blue shifting of the absorption edge, while Ag dopants may add additional non-bridging oxygens to the glass network, resulting in the red shift of UV absorption edge.

Fig. 1 Room temperature absorption spectra of undoped (OF1) and Ag-nanoclusters doped (OF2, OF3 and OF4) glasses. Red shift of absorption spectrum from OF2 to OF3 is due to larger UV absorption coefficient by Ag nanoclusters.

Under UV-blue excitation a broad 400–900 nm single emission band is detected. The shape and spectral position of this emission and excitation spectra remain independent of the excitation wavelength at room temperature indicative of fixed site for Ag nanoclusters in this glass host. Lowering the experimental temperature down to 10 K results in appearance of a shoulder on the shorter-wavelength side emission and excitation spectra. We show that addition of a small percentage of a glass network modifier, such as Al₂O₃ results in increased transparency of this glass in UV, making it more suitable for excitation at 365 nm wavelength of a commercial LED due to lower attenuation of the glass host. The glass host itself does not produce any luminescence at the excitation wavelengths used for excitation of Ag nanoclusters in this work. The proposed glass can be used in environmental friendly, Hg-free, light sources, flexible monitors and for down-shifting of UV-blue part of the solar spectrum in enhanced photovoltaics.

Room temperature excitation and emission spectra of Ag-doped oxyfluoride glasses are shown in Fig. 2. The excitation spectra of the glasses correspond to the absorption spectra shown in Fig. 1, since the Ag nanoclusters absorbs in the broad range from 300 to 450 nm. An excitation band around 460 nm was not detected for the glasses because the band does not relate to Ag nanoclusters. The band is due to absorption of amorphous Ag nanoparticles which do not emit any light.²⁰ These spectra do match the earlier reported excitation and emission spectra of Ag-nanoclusters dispersed in oxyfluoride glasses. It was reported in [ref. 7, 15 and 16 and refs therein] by means of optical spectroscopy, transmission electron microscopy (TEM), electron spin resonance techniques (ESR) and time dependent differential functional theory (TD-DFT) that Ag-nanoclusters, presumably in the form of Ag₄²⁺ tetramers, are dispersed in the fluorite type lattice of the oxyfluoride glass host and are responsible for these broad excitation and emission bands. Also it was observed in previous investigation of luminescence of Ag nanolcusters that shifting of the detection wavelength of luminescence and excitation wavelength from red to the blue results in blue-shifting of the respective excitation and emission spectra. This result was explained, similar to the case of quantum dots, by dispersion of Ag nanoclusters in Pb_xCd_1−xF₂ solid state solution with varying surroundings and nanoclusters size, i.e. with varying an Ag nanoclusters site.

Fig. 2 (a) Room temperature excitation and emission spectra of three samples: OF2, OF3 and OF4 detected at 600 nm (the emission maximum) and excited at 390 nm (the excitation maximum). (b) Excitation and emission spectra of OF3 sample. Excitation and detection wavelength are post-signed.

Contrary, in Fig. 2(a) it is not observed any shifting of the respective excitation and emission spectra of Ag nanoclusters with varying doping level and glass host composition. Likewise, a shifting of detection and excitation wavelength of luminescence does not result in any shitting/altering of excitation and emission spectra of samples, Fig. 2(b). Comparing the previous results, and the results presenting in this paper we may conclude that here it is observed only excitation and emission of Ag nanoclusters occupying mostly the PbF₂ lattice (which proves in fact that Ag nanoclusters tend to disperse on fluoride component of oxy-fluoride glasses, rather than in their oxide component, in agreement with previous argumentation^6,15,16). Subsequently, the mechanism of Ag nanocluster formation may be very similar to reported in.^6,15,16 In this case, the intrinsic F⁻ – vacancies, which is known to exist in fluorite lattice, play role of nucleation centers for growth of silver nanoclusters as they attract Ag⁺ ions. The Ag⁺ cations substitute for Pb²⁺ ions as described by the following equations:

M²⁺ = 2Ag⁺; 2Ag + e = Ag₂⁺ (1)

According to charge compensation mechanism, pair of Ag⁺ ions substitutes simultaneously for one Pb²⁺ ion. In this case, one Ag⁺ occupies site of Pb²⁺ ion and another occupies the neighbour cation hole site. Absence of Cd²⁺ ions in the glass composition may result in worse solubility of Ag nanoclusters in fluorite lattice because Cd²⁺ and Ag⁺ ionic radii are more close to each other than radii of Ag⁺ and Pb²⁺. Further, the neighbouring paramagnetic Ag₂⁺ dimers combine in diamagnetic Ag₄²⁺ tetramers through electronic coupling. The rest possible Ag nanoclusters, such as Ag₂⁰, Ag₂⁺, Ag₃⁰ and etc., may be ruled out because the glasses did not exhibit any paramagnetic signal in electron spin resonance measurements and the respective calculated excitation spectra of the Ag nanoclusters did not match to experimental excitation spectra.¹⁶ Strictly speaking, the luminescent Ag nanoclusters differ from Ag₄²⁺ unit. Although this structural Ag unit may be a basis motif of larger Ag nanoclusters, such as hexamers, octamers and etc., in oxyfluoride glasses resulting in red shift of excitation and emission spectra comparing to Ag₄²⁺ tetramer. Indeed, the excitation and emission spectra of the glasses prepared in this paper are red shifted comparing to the results reported in (ref. 6).

The temperature dependence of luminescence of OF2 and OF4 glasses is represented in Fig. 3. Decreasing the temperature down to 5 K results in blue-shifting of maxima of luminescence spectra, as may be related to some shrinking of host lattice and Ag nanoclusters, and suppressing of energy transfer from the blue to green and red emitting Ag nanoclusters.²⁵ For the same reason, a minor UV-blue band arises in range from 400 to 450 nm. Intensity of this UV-blue band depends on concentration of alumina in glass composition indicating that alumina supports Ag nanoclusters sites which emit in the blue.²⁵ Low temperature excitation spectra of OF2 and OF4 glasses are shown in Fig. 3(b and d). Decrease of temperature down to 25 K results in arising of new “bluish” excitation band centred around 320 nm comparing to the room temperature excitation spectra. Also the “red” shoulder still has been observed in the low temperature excitation spectra. This shoulder corresponds to absorption of Ag nanoclusters at room temperature. The observation also indicates that blue emitting clusters do mostly absorb at low temperature without transferring energy to lower energy placed clusters, perhaps due to thermal barriers between these clusters.^15,25

Fig. 3 Temperature dependence of luminescence and excitation spectra of OF2 (a and b) and OF4 (c and d) glasses. Excitation wavelength 380 nm. Detection wavelength and temperature are post-signed.

Earlier similar temperature dependence for absorption/excitation and luminescence spectra have been reported for Ag⁻ or Ag₂²⁻ centres in alkali halide crystals.^22,23 These centres may occupy F⁻ vacancies in the fluorite-type lattice. These vacancies are intrinsic defects²⁴ and known to nucleate the growth of Ag nanoclusters.⁶

The CIE chromaticity diagram shows emission colour of OF1 and OF3 glasses (Fig. 4). The change of excitation wavelength from 340 to 420 nm does not affect the colour of emission at the room temperature. Contrary, at 5 K the change of excitation wavelength in range from 300 to 420 nm results in considerable alternation of emission color, in agreement with arguments from previous paragraph. Under excitation at 380 nm, quantum yield of OF2, OF3 and OF4 samples was found to be 4.8, 5.1 and 4.8%, respectively.

Fig. 4 Chromaticity CIE diagram indicates an emission colour of OF1 (a) and OF4 (b) samples at different excitation wavelengths and temperatures. Excitation wavelength and temperature of measurements are post-signed. The white zone indicates a white light gamut area, and the black dot indicates the colour of black body radiation at 6667 K.

Therefore, the prepared glasses can be used as yellow-orange phosphors. Comparing to our previous results, we have prepared the oxyfluoride glass (Cd-free) with simple composition using very short melting time (∼10 minutes) and temperature as low as 1000 °C, which is very promising for production due to low energy consumption during the preparation. This simple composition is capable to dissolve Ag nanoclusters and the resulting glasses demonstrate quantum yield about 5%. Likewise, the effect of alumina on optical properties of the glasses was investigated. Addition of alumina results in shift of the absorption edge and does not alter Ag nanoclusters site or quantum yield of luminescence. It is also worth mentioning that this is the first report on the glasses containing Ag nanoclusters of fixed site. The fixed site of Ag nanoclusters results in invariance of the excitation and the emission spectra on changing the detection and the excitation wavelength. Current work is directed towards shifting the emission color of these samples closer to the centre of a white gamut, which is shown in Fig. 4. This will be realised by extensive work on glass host composition and addition of rare-earth dopants, which will affect a total color of samples luminescence; this extensive work is out of the scope of this paper. According to review,²⁶ Ag nanoclusters doped glasses are very stable against photobleaching and ageing: the optical properties remain unchanged for years. Likewise, Ag-nanoclusters doped glasses may be prepared by simple melt-quenching method. Production of the glasses is much cheaper and simpler comparing to production methods of current phosphors.²⁷ It is also worth mentioning that Ag is cheap and abundant material. Silver ions, nanoclusters and nanoparticles cause no serious risk for human health and possess strong antimicrobial properties which led to use of silver in medicine.²⁸ In our case, Ag nanoclusters are not in free state but encapsulated in the glasses insoluble in water which means that they are not dangerous for environment.

4. Conclusions

Ag luminescent nanoclusters doped oxyfluoride glasses, based on simple, trivial, 55(SiO₂) : 45(PbF₂) chemical formulation, have been prepared and investigated. The shape and spectral position of their emission and excitation spectra has been found to be independent of the excitation wavelength at room temperature, indicative that Ag nanoclusters embed mostly in PbF₂ compound of the glass host. Temperature dependence of luminescence of Ag nanoclusters indicates the transfer from blue absorbing/emitting Ag nanoclusters to lower energy placed Ag nanoclusters is suppressed at low temperature, in agreement with previous work.^15,23 Addition of alumina to the host results in the blue-shifting of UV absorption edge, making this alumina-containing glass more efficient for excitation by commercial 365 nm LED due to elimination of the host absorption at 365 nm.

Acknowledgements

We thank the Methusalem Funding of Flemish Government and the “Fonds voor Wetenschappelijk Onderzoek, FWO” for the support of this work.

References

1. I. Díez and R. H. A. Ras, Nanoscale, 2011, 3, 1963.
2. S. Choi, R. M. Dickson and J. Yu, Chem. Soc. Rev., 2012, 41, 1867.
3. A. S. Kuznetsov, J. J. Velázquez, V. K. Tikhomirov, J. Mendez-Ramos and V. V. Moshchalkov, Appl. Phys. Lett., 2012, 101, 251106.
4. E. Coutino-Gonzalez, M. B. J. Roeffaers, B. Dieu, G. De Cremer, S. Leyre, P. Hanselaer, W. Fyen, B. Sels and J. Hofkens, J. Phys. Chem. C, 2013, 117, 6998.
5. G. De Cremer, E. Coutiño-Gonzalez, M. B. J. Roeffaers, B. Moens, J. Ollevier, M. Van der Auweraer, R. Schoonheydt, P. A. Jacobs, F. C. De Schryver, J. Hofkens, D. E. De Vos, B. F. Sels and T. Vosch, J. Am. Chem. Soc., 2009, 131, 3049.
6. V. K. Tikhomirov, V. D. Rodríguez, A. Kuznetsov, D. Kirilenko, G. Van Tendeloo and V. V. Moshchalkov, Opt. Express, 2010, 18, 22032.
7. V. K. Tikhomirov, T. Vosch, E. Fron, V. D. Rodríguez, J. J. Velázquez, D. Kirilenko, G. Van Tendeloo, J. Hofkens, M. Van der Auweraer and V. V. Moshchalkov, RSC adv., 2012, 2, 1496.
8. H. Guo, X. Wang, J. Chen and F. Li, Opt. Express, 2010, 18, 18900.
9. M. Eichelbaum and K. Rademann, Adv. Funct. Mater., 2009, 19, 2045.
10. A. S. Kuznetsov, V. K. Tikhomirov and V. V. Moshchalkov, Mater. Lett., 2013, 92, 4.
11. M. Mattarelli, M. Montagna, K. Vishnubhatla, A. Chiasera, M. Ferrari and G. Righini, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 75, 125102.
12. K. Bourhis, A. Royon, G. Papon, L. Canioni, N. Makria, Y. Petit and T. Cardinal, J. Non-Cryst. Solids, 2013, 377, 142–145.
13. M. Bellec, A. Royon, K. Bourhis, J. Choi, B. Bousquet, M. Treguer, T. Cardinal, J.-J. Videau, M. Richardson and L. Canioni, J. Phys. Chem. C, 2010, 114, 15584.
14. G. De Cremer, B. F. Sels, J. Hotta, M. B. J. Roeffaers, E. Bartholomeeusen, E. Coutiño-Gonzalez, V. Valtchev, D. E. De Vos, T. Vosch and J. Hofkens, Adv. Mater., 2010, 22, 957.
15. J. J. Velázquez, V. K. Tikhomirov, L. F. Chibotaru, N. T. Cuong, A. S. Kuznetsov, V. D. Rodríguez, M. T. Nguyen and V. V. Moshchalkov, Opt. Express, 2012, 20, 13582.
16. N. T. Cuong, V. K. Tikhomirov, L. F. Chibotaru, A. Stesmans, V. D. Rodríguez, M. T. Nguyen and V. V. Moshchalkov, J. Chem. Phys., 2012, 136, 174108.
17. M. A. P. Silva, V. Briois, M. Poulain, Y. Messaddeq and S. J. L. Ribeiro, J. Phys. Chem. Solids, 2003, 64, 95.
18. S. Ueda and M. Maeda, Metall. Mater. Trans. B, 1999, 30, 921.
19. E. Coutino-Gonzalez, M. B. J. Roeffaers, B. Dieu, G. De Cremer, S. Leyre, P. Hanselaer, W. Fyen, B. Sels and J. Vanacken, J. Phys. Chem. C, 2013, 117, 6998–7004.
20. M. V. Shestakov, M. Meledina, S. Turner, V. K. Tikhomirov, N. Verellen, V. D. Rodríguez, J. J. Velázquez, G. Van Tendeloo and V. V. Moshchalkov, J. Appl. Phys., 2013, 114, 073102.
21. G. H. Sigel, Jr, J. Phys. Chem. Solids, 1971, 32, 2373.
22. T. Kawai, I. Akai, H. Ichida, Y. Kanematsu, K. Mizoguchi and T. Karasawa, Phys. Status Solidi B, 2009, 246, 1327.
23. H. Takezoe, I. Maruyama, A. Yoshikawa, A. Honma and R. Onaka, J. Phys. Soc. Jpn., 1971, 30, 1759.
24. S. Hull, Rep. Prog. Phys., 2004, 67, 1233.
25. A. S. Kuznetsov, V. K. Tikhomirov and V. V. Moshchalkov, Opt. Express, 2012, 20, 21576.
26. A. S. Kuznetsov, V. K. Tikhomirov, M. V. Shestakov and V. V. Moshchalkov, Nanoscale, 2013, 5, 10065–10075.
27. C. C. Li and R. S. Liu, J. Phys. Chem. Lett., 2011, 2, 1268–1277.
28. K. Mijnendonckx, N. Leys, J. Mahillon, S. Silver and R. Van Houdt, BioMetals, 2013, 26, 609–621.

---