Silicon-Induced UV Transparency in Phosphate Glasses and Its

Subscriber access provided by UB + Fachbibliothek Chemie | (FU-Bibliothekssystem)

Article

Silicon-induced UV Transparency in Phosphate Glasses and its Application to the Enhancement of the UV Type B Emission of Gd

3+

José A. Jiménez ACS Appl. Mater. Interfaces, Just Accepted Manuscript • Publication Date (Web): 24 Apr 2017 Downloaded from http://pubs.acs.org on April 25, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Silicon-induced UV Transparency in Phosphate Glasses and its Application to the Enhancement of the UV Type B Emission of Gd3+ José A. Jiménez*, 1 Department of Chemistry, University of North Florida, Jacksonville, FL 32224, USA

*E-mail: [email protected] 1

Present address: Functional Films Lab, BASF Corporation, 2655 Route 22 West, Union, NJ 07083, USA

KEYWORDS: glasses; luminescence; optical materials; phototherapy lamps; UV transparency

ACS Paragon Plus Environment

1

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 24

ABSTRACT The silicon route to improve the ultraviolet (UV) transparency in phosphate glasses is investigated, and further exploited to enhance the UV type B (280-320 nm) emission of gadolinium(III) relevant for biomedical applications. The glasses were synthesized with a barium-phosphate composition by melt-quenching in ambient atmosphere, and the optical properties investigated by optical absorption and photoluminescence (PL) spectroscopy including emission decay kinetics. An improvement in the UV transparency was gradually developed for the glasses melted merely with increasing amounts of Si powder. A particular PL in the visible was also exhibited for such glasses under excitation at 275 nm, consistent with the presence of Si-induced defects. For Si-Gd co-doped glasses, the UV transparency was likewise manifested, while the UV emission from Gd3+ around 312 nm was enhanced with the increase in Si concentration (up to ~6.7 times). Moreover, along with the Gd3+ PL intensity enhancement, a linear correlation was revealed between the increase in decay times for the Gd3+ 6P7/2 emitting state and the amount of silicon. It is then suggested that the improved PL properties of gadolinium(III) originate from the increased UV transparency of the host and the consequent precluding of a non-radiative energy transfer from Gd3+ to the matrix. Accordingly, a role of Si as PL quenching inhibitor is supported. The demonstrated efficacy of the Si-Gd co-doping concept realized by a facile glass synthesis procedure may appeal to the application of the UVemitting glasses for phototherapy lamps.


2

Page 3 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION The synthesis of novel gadolinium(III)-activated inorganic compounds and glasses is currently of interest for producing efficient sources of ultraviolet (UV) light for phototherapy lamps valuable in the treatment of skin diseases.1-5 Hence, innovative approaches are currently sought for enhancing the UV type B (UVB) emission (280-320 nm) of Gd3+ ions suitable for the biomedical applications.6,7 For instance, co-doping of phosphors with Pr3+ or Pb2+ has been reported to improve the UVB emission intensity of Gd3+.1,6 In glasses, the use of co-dopants such as Ag+,5 and matrix modification, e.g. by Al2O3 addition,7 have been also indicated to boost the UV emission stemming from the 6P7/2 state in Gd3+. On the other hand, silicon powder has been recently proposed for producing plasmonic Ag nanocomposite phosphate glasses also possessing improved transparency in the UV.8 Indeed, the value of silicon has been demonstrated for producing diverse glasses with interesting optical properties.9-12 However, the sole use of silicon powder to produce the increase in UV light transmission has not been investigated in detail to the best of the author’s knowledge. Moreover, the impact of the Si-induced UV transparency on the UVB emission of Gd3+ is still to be unveiled. Accordingly, this work explored the silicon route to produce enhanced UV transparency in phosphate glasses, and further investigated the impact of silicon on the UVB emission of Gd3+ ions. Si doped and Si-Gd co-doped glasses were prepared with a bariumphosphate matrix desirable for photonic applications13 by the melt-quenching technique. Material characterization was carried out by optical absorption and photoluminescence (PL) spectroscopy including an emission decay dynamics assessment. The data shows consistently the improvement in UV transparency by silicon, and in the case of co-doping with gadolinium(III), a concurrent


3


Page 4 of 24

enhancement of the UVB emission. Thus, the co-doping approach shows potential for application of the glass material in phototherapy lamps. Further, physical insights regarding the origin of the enhancement are obtained by the decay kinetics assessment.

EXPERIMENTAL The phosphate glasses were prepared with a 50P2O5:50BaO (mol%) composition from high purity Alfa Aesar chemicals (P2O5, ≥ 98% and BaCO3, 99.8%) by a facile procedure via the melt-quenching technique.5,8 Batch materials (about 25 g batches) were thoroughly mixed and melted in porcelain crucibles at 1150 °C between 15 and 25 min under normal atmospheric conditions and immediately quenched by pouring the molten glass on a steel plate. Such melting conditions have been previously found effective for preparing Si-containing glasses.8,10,12 Silicon and/or gadolinium doping was done by adding Si powder (Alfa Aesar, crystalline, -140 mesh, 98%) and/or Gd2O3 (Alfa Aesar, 99.9%) quantities in mol% in relation to network former P2O5. A set of glasses were prepared for this study with just Si powder added as part of the batch materials as 0.1, 0.3 and 0.5 mol% for glasses labeled 0.1Si, 0.3Si and 0.5Si, respectively. An undoped glass host was also prepared for reference purposes. The region of the glass transition temperature, Tg, was observed from differential scanning calorimetry (DSC; TA Instruments; heating rate of 10°C/min) to be around 485 °C for both the host and the 0.5Si glass as most distinct case of interest (see Supporting Information, Fig. S1). Gadolinium concentration was held constant in all Gd-doped samples at 2.0 mol%. In this case, a set of glasses were prepared having the fixed 2.0 mol% Gd2O3 with Si added as 0.2 and 0.5 mol%, for glasses referred to as 0.2Si-Gd and 0.5Si-Gd, respectively. The Tg region for the 0.5Si-Gd glass as that with the highest concentration of dopants appeared around 490 °C (Supporting Information, Fig. S1). This slight shift to higher temperature relative to the 0.5Si glass is likely connected to the effect


4

Page 5 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


of gadolinium.14 Another glass containing no silicon but only the 2.0% Gd2O3 was also prepared as reference, labeled Gd glass. All glasses presented transparent colorless appearance upon steelquenching. The glasses were cut and polished in order to produce glass slabs with final thicknesses of about 1.0 mm for spectroscopic measurements. Optical absorption measurements were performed with a Perkin-Elmer 35 double-beam spectrophotometer; the spectra were recorded with air as reference. PL spectra and emission decay curves were measured with a Photon Technology International QuantaMaster 30 spectrofluorometer equipped with a Xe flash lamp having a pulse width of about 2 µs (L4633, Hamamatsu Photonics K.K.) and a photomultiplier tube (R1527P, Hamamatsu Photonics K.K.). The flash lamp was kept operating at a frequency of 125 Hz with the total period of data collection set to 8 ms. The step size used for the spectral acquisitions was 1 nm. All PL measurements were recorded with samples mounted in a solid sample holder at an angle of 40° with particular attention given to keep conditions constant during experiments. All measurements were carried out at room temperature.

RESULTS AND DISCUSSION Effect of Si on glass optical properties. Considered first in this section is the influence of silicon on the optical properties of the barium-phosphate glasses. Figure 1 shows the optical absorption spectra obtained for the 0.1Si-0.5Si glasses along with the glass host as reference. A significant decrease in the UV absorption of the glasses is noticed as a result of adding increasing amounts of silicon to batch materials. Attempts to increase the amount of silicon added into the batch materials beyond 0.5 mol% did not provide further improvements in the UV transparency, which in fact started to deteriorate (see Supporting Information, Fig. S2). Thus, it was considered that the melting conditions employed were optimum for the 0.5Si glass. A shift toward high


5


Page 6 of 24

energy of the UV edge has been also reported for copper-doped glasses melted with up to 4 mol% Si,10 and tin(IV)-doped phosphate glasses containing up to 6 mol% Si.12 Moreover, a substantial improvement in UV transparency was consistently observed for Ag-Si co-doped barium-phosphate glasses.8 Besides the traditional assignment of reduced metallic impurities,15,16 the formation of P–O–Si bonds supported by 31P nuclear magnetic resonance spectroscopy was considered in connection with the improved light transmission in the UV.8 A model was proposed suggesting an oxygen-assisted process was linked to the incorporation of oxidized silicon into the phosphate glass network and electron trapping in non-bridging oxygens, i.e., the formation of Si–O•– and Si–O–O•– defects was postulated.8 Hence, the formation of such species in the presence case may be considered, concurrent with the incorporation of SiO4 tetrahedra expected from oxidized silicon acting as network former. Similarly, an improvement in UV transparency was recently reported for glasses melted together with multi-wall carbon nanotubes17 and graphite.18 Comprehensive spectroscopic characterizations also provided evidence in support of structural modifications of the glass matrix leading to the formation of P– O–C bonds and oxygen radicals represented in an analogous model as C–O•– and C–O–O•– defects.17,18 Noticing Fig. 1, the 0.1Si glass exposes a UV absorption peak at about 230 nm, which decreases steadily for the 0.3Si and 0.5Si glasses. As observed in the bottom inset of Fig. 1, a plot of the optical density of this peak as a function of the Si contents exhibits correlation where linear regression yielded a correlation coefficient r = – 0.984. Apart from the traditional view of reduction of metallic impurities (e.g. Fe3+),15,16,19 this behavior could be a reflect of a suppression of charge transfer transitions (O2- → Ba2+ ions in the present case) accompanying structural modifications.8,17


6

Page 7 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1. Optical absorption spectra for the host and 0.1-0.5Si glasses. The lower inset is a plot of the optical density (OD) at 230 nm of the glasses as a function of the amount of Si added; the solid line is a linear fit to the data. The top inset is a plot of (Eα)1/2 vs. hν for the host and 0.5Si glass; the solid lines are the fits to the linear regions of the absorption profiles from which the optical band gap values (Eopt) were determined (intercepts in energy axis).

An attempt to analyze the absorption data for the 0.5Si glass as most distinct case in the framework of a Tauc’s plot20 is made in the top inset of Fig. 1 in comparison to the host. In this context, the power law expression for the absorption coefficient α(ν) as a function of photon energy (E = hν) given by

α(ν ) = k

(hν − Eopt )n

(1)

hν


7


Page 8 of 24

where the exponent n is related to the nature of the transitions, k is a constant, and Eopt is the optical band gap energy.20 For glasses, the value n = 2 connected to allowed-indirect transitions may be employed for fitting the experimental results.21 So a plot of (Eα)1/2 vs. photon energy (hν) allows for estimating Eopt from extrapolation of the linear portion of the plot and determination of the intercept on the energy axis. The estimated Eopt values for the host and the 0.5Si glass are 3.39 (±0.03) eV and 4.88 (±0.33) eV, respectively. Clearly, there is larger error associated with the determination for the latter as there is also some interference from the absorption shoulder near 5.0 eV. Still, it turns out that the estimated maximum Eopt value obtained for the 0.5Si glass of 4.88 (±0.33) eV is higher than the reported for the Si co-doped glass containing 0.4 mol% silver oxide at 4.71 (±0.21) eV.8 Thus, it seems that in the latter case the presence of Ag+ ions presented some absorption in the UV as considered.8


8

Page 9 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 2. PL emission spectra for the host and 0.1-0.5Si glasses recorded under excitation at 275 nm. The inset shows the PL excitation spectrum obtained for the 0.5Si glass by monitoring emission at 445 nm.

In Fig. 2, the PL of the 0.1-0.5Si glasses is examined under excitation at 275 nm in comparison to the host as reference. The excitation wavelength is selected as it has been shown effective for the excitation of Si-induced luminescent defects in SnO2 co-doped glasses.12 Further, it proved effective as well for observing the distinct PL evolution resulting from melting the glasses with multi-wall carbon nanotubes17 and graphite.18 These studies have suggested that the presence of analogous defects is produced by oxygen radicals created in the melt by the highly reductive environment in the presence of oxygen irrespective of the element (C or Si).12,17,18 As observed in Fig. 2, although the host presents a weak structureless trace, an emission is consistently observed for the 0.1-0.5Si glasses. The 0.1Si glass shows a broad


9


Page 10 of 24

emission band with a maximum at about 465 nm. However, the 0.3Si glass shows a more intense PL band centered around 450 nm. Ultimately, the PL of the 0.5Si glass presents a more intense shoulder around 430 nm. Evidently, the PL evolution appears concurrently with the enhanced UV transparency (Fig. 1), similar to the cases where the glasses were melted together with the different sources of carbon. 17,18 However, a comparable PL was not observed for the Ag-Si codoped glass system containing a relatively small amount of silicon (0.2 mol%).8 It appeared in such instance that the concentration of luminescent centers was significantly decreased as a consequence of the electron transfer reactions that lead to the reduction of Ag+ ions and subsequently to the Ag nanoparticle precipitation.8 Also, the efficient emission of Cu+ ions was detected in the Cu-Si co-doped glasses even after the thermal processing that produced the plasmonic Cu nanocomposites.10 Nevertheless, it was recently shown that an emission resembling that of the carbon-induced defects was also produced when using silicon as reducing agent at relatively high concentrations for producing blue-emitting tin-doped phosphate glasses.12 The PL behavior examined through time-resolved measurements was consistent with the models invoking the reactive oxygen radicals accompanying either P–O–Si12 or P–O–C bonds.17 It was then considered that the Si–O–O•– defect was responsible for the PL towards high-energy, whereas the Si–O•– one was linked to the lower-energy PL.12 Thus, it could be that in the case concerning the data in Fig. 2 the PL observed for the 0.1Si glass arises as a consequence of the Si–O•– defect, while the increase in silicon may result in the formation of the Si–O–O•– defects. The latter are expected for relatively high reductant concentrations.12 Further, by monitoring the emission of the 0.5Si glass as most distinct case at 445 nm, the excitation spectrum shown in the inset of Fig. 2 was obtained. It displays a broad band centered around 275 nm which bears a resemblance to the observed for the reductant-induced PL excitation bands


10

Page 11 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


produced in silicon-tin co-doped glasses12 and also those melted with carbon.

17,18

The broad

character and asymmetry of the excitation band toward longer wavelengths suggests it likely encompasses the two aforementioned Si-related luminescent defects.

Figure 3. Emission decay curve obtained for the 0.5Si glass under excitation at 275 nm by monitoring emission at 445 nm; the solid line is a bi-exponential fit. The inset depicts the proposed Si-related defect centers presumably associated to the distinct luminescence features (designated in text).

Additional information on the nature of the Si-induced emission centers is obtained from the luminescence decay kinetics. Presented in Fig. 3 is the emission decay curve recorded for the 0.5Si glass as most distinct case, under the effective excitation at 275 nm with the emission being monitored at 445 nm. The decay clearly shows contrasting fast and slow components. An


11


Page 12 of 24

analogous decay behavior was recently reported for the tin-containing glasses melted with a relatively high concentration of Si, namely at 6.0 mol%.12 Such emission decay kinetics turned out to be a similar to the two-component decay observed for the carbon-induced PL in the barium-phosphate glasses melted together with multi-wall carbon nanotubes.17 In that case, a biexponential fit yielded lifetimes of 33.8 (±0.6) µs and 3050 (±10) µs for the fast and slow components presumably associated with the formation of C–O•– and C–O–O•– defects. Accordingly, it was considered in Ref. 12 that apart from divalent tin emission, the analogous Si–O•– and Si–O–O•– defects proposed to be connected to the improved UV transparency8 were associated to the distinct PL. Hence, the tentatively assumed presence of Si–O•– and Si–O–O•– defects at relatively high contents of silicon is supported herein by the data in Fig. 3. Thus, a biexponential fit was performed to the decay curve in Fig. 3 in accord with the bi-exponential decay apparently linked to emission from the two types of oxygen radicals proposed to accompany the P–O–Si bonds.12 The fit yielded lifetimes of 37.9 (±0.1) µs and 2467 (±40) µs, considered in connection with the Si–O•– and Si–O–O•– defects (depicted in Fig. 3, inset), respectively. For the tin and silicon co-doped glass evaluated in Ref. 12, the corresponding lifetimes obtained were 31.5 (±0.1) µs and 1771 (±19) µs. Although the fast decay components appear similar, the slow component in Ref. 12 has a considerably shorter lifetime than the estimated here in association with the Si–O–O•– defect. Further, the slow decay component in association to C–O–O•– defects in glasses melted with multi-wall carbon nanotubes was estimated to have an even longer lifetime of 3050 (±10) µs.17 It then seems that the more electropositive character of silicon contributes to the increased decay rate through a stronger Coulomb interaction, and the presence of tin(II) in the matrix further enhances this decay rate. However, the physical origin of the observed differences is not clear at present and additional


12

Page 13 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


investigations are needed for clarification. Still, the enhanced UV transparency produced by melting the glasses with silicon suggests favorable conditions can be attained for improving the UV emission of Gd3+ ions as considered next.

Figure 4. Optical absorption spectra for the Gd, 0.2Si-Gd and 0.5Si-Gd glasses; the absorption spectrum for the 0.5Si glass is also shown for comparison. The inset shows the UV spectral region magnified.

Photoluminescence properties of Gd3+ ions in Si co-doped glass. Let us then consider the optical properties of the Si and Gd co-doped glasses. Presented in Fig. 4 are the optical absorption spectra obtained for the Gd-containing glasses, i.e. the Gd, 0.2Si-Gd and 0.5Si-Gd glasses, alongside the 0.5Si glass for comparison. While the Gd glass shows an


13


Page 14 of 24

absorption profile in resemblance to the host (Fig. 1), the 0.2Si-Gd and 0.5Si-Gd glasses exhibit consistently the Si-induced improved UV transparency with the absorption feature around 230 nm. Moreover, the reproducibility in the glass synthesis process is evidenced by the similarity in the absorption curves obtained for the 0.5Si-Gd and 0.5Si glasses, both prepared with 0.5 mol% Si. In addition, as shown magnified in the inset of Fig. 4, the 0.5Si-Gd glass displays visibly the absorption features characteristic of Gd3+ ions around 275 nm.1-5 In Fig. 5, the PL of the Gd, 0.2Si-Gd and 0.5Si-Gd glasses is examined under excitation at 275 nm which is resonant with 8S7/2 → 6IJ transitions in Gd3+ ions1,5 (excitation spectra shown as Supporting Information, Fig. S3). Interestingly, the UV type B emission of Gd3+ around 312 nm due to 6P7/2 → 8S7/2 transitions increases noticeably with increasing concentration of silicon added to the 0.2Si-Gd and 0.5Si-Gd glasses. The maximum enhancement is estimated to be approximately 6.7× for the 0.5Si-Gd glass relative to the one merely containing Gd. This is at least two times the reported enhancement of Gd3+ UVB emission obtained by excitation energy transfer from Ag+ ions in the Ag-Gd co-doped phosphate matrix.5 The degree of enhancement is also greater than those reported by Kalpana et al.7 and Gandhi et al.22 of about four times for Gd3+-containing BaO-B2O3-P2O5 and Li2O-PbO-P2O5 glasses when mixed with 3.0 mol% of Al2O3 and 3.0 mol% of SnO2, respectively. The top inset of Fig. 5 shows a plot of the Gd3+ UV emission intensity as a function of the Si contents, where the linear regression performed yielded a correlation coefficient r = 0.970. The inset in the bottom of Fig. 5 shows magnified the region of the spectra for the glasses which is relevant for the Si-induced defects. The emission spectrum obtained for the 0.5Si glass under the same excitation wavelength of 275 nm is shown for comparison. The Gd glass shows no distinct PL in the spectral range relevant to the defects. However, the 0.2Si-Gd and 0.5Si-Gd glasses show evidence of the progressive generation of the


14

Page 15 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Si-related centers considered (vide supra). For instance, the 0.5Si-Gd glass shows an increased emission around 430 nm relative to the 0.2Si-Gd glass, analogous to the case of the 0.3Si and 0.5Si glasses as discussed (Fig. 2). Moreover, the emission of the 0.5Si glass overlaid appears similar to the one for the 0.5Si-Gd glass. This is consistent with the fact that the two glasses displayed similar degree of UV absorption (apart from the Gd3+ features) and also sustains the reproducibility of the material synthesis procedure.

Figure 5. PL emission spectra for the Gd, 0.2Si-Gd and 0.5Si-Gd glasses recorded under excitation at 275 nm. The lower inset shows magnified the region relevant to emission from Siinduced defects where the emission spectrum for the 0.5Si glass is also shown for comparison. The top inset is a plot of the maximum Gd3+ UV emission intensity in the glasses as a function of the amount of Si added; the solid line is a linear fit to the data.


15


Page 16 of 24

It could be considered that an absorption competition occurs between Gd3+ ions and the barium-phosphate matrix at the excitation wavelength of 275 nm prior to the Si addition. However, further insights on the nature of the enhancement in the UV type B emission of Gd3+ in the glasses melted with silicon can be obtained from analyzing the decay dynamics of the 6P7/2 emitting state. Figure 6 shows the emission decay curves recorded for the Gd, 0.2Si-Gd and 0.5Si-Gd glasses under excitation of 8S7/2 → 6IJ transitions at 275 nm with the 6P7/2 emission monitored at 315 nm. The curves clearly show the first-order decay behavior typical of 6P7/2 → 8

S7/2 transitions in Gd3+ ions.23 Thus, single-exponential fits were performed, which yielded

lifetimes of 3069 (±4) µs, 3164 (±6) µs and 3329 (±7) µs, for the Gd, 0.2Si-Gd and 0.5Si-Gd glasses, respectively. The inset in Fig. 6 is a plot of the determined 6P7/2 emitting state lifetimes as a function of the Si contents. The linear regression performed yielded a correlation coefficient r = 0.999, thus pointing out to a major involvement of the 6P7/2 emitting state in the process. Such lifetime correlation suggests that the enhanced UVB emission from Gd3+ has an energy transfer process associated to its origin. It may be then proposed that apart from an absorption competition, the increase in Si prevents an unfavorable energy transfer from the 6P7/2 state in Gd3+ ions to the glass matrix as illustrated in the schematic in Fig. 7. If present, such transfer possibly to metallic impurities15,16 and/or to charge transfer states (O2- → Ba2+) in the host8,12,17 can provide a non-radiative channel for depopulating the 6P7/2 emitting state. This may be the cause for the shorter 6P7/2 lifetimes observed for the Gd and 0.2Si-Gd glasses. Conversely, a Gd3+ de-clustering effect has been invoked by Kalpana et al.7 and Gandhi et al.22 to account, at least partially, for the enhanced UV emission of Gd3+ ions in phosphate-based glasses. However, it appears in the present case that the improved UV transparency that comes with the increase in Si contents is responsible for the enhanced UVB emission from Gd3+ and its preventing of the Gd3+


16

Page 17 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


→ host energy transfer depicted in Fig. 7. In other words, a role of Si as a PL quenching inhibitor is supported. Other than that, it does not seem that the Si-induced defects have a direct interaction with the Gd3+ ions, as evidenced by the fact that the emission ascribed to these is similar to that in the glass matrix merely containing Si (e.g. Fig. 5). As a final remark, although the luminescence dynamics data points to a key role of the 6P7/2 emitting state in the Gd3+ ions, a direct transfer from the 6IJ levels to the host cannot be wholly excluded at present.

Figure 6. Emission decay curves obtained for the Gd, 0.2Si-Gd and 0.5Si-Gd glasses under excitation at 275 nm by monitoring emission at 315 nm; the solid lines are single exponential fits. The inset is a plot of the estimated Gd3+ 6P7/2 lifetimes as a function of the amount of Si added to the glasses; the solid line is a linear fit to the data.


17


Page 18 of 24

Figure 7. Simplified schematic of energy transfer (ET) from 6P7/2 emitting state in Gd3+ ions to the glass gloss that results in the PL quenching which is inhibited by the Si co-doping. Vertical solid-straight and solid-curved arrows represent radiative and non-radiative transitions, respectively. The competing absorption of the 275 nm excitation wavelength by the glass host is also illustrated.


18

Page 19 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


CONCLUSIONS The synthesis of phosphate glasses was pursued by melting the glasses in ambient atmosphere merely using Si powder for producing an enhanced UV transparency. Together with the improved UV transparency achieved with increasing silicon contents from 0.1 to 0.5 mol% Si, a distinct PL emission was observed upon UV excitation at 275 nm. The luminescence was interpreted as arising from the presence of Si–O•– and Si–O–O•– defects which have been also linked to the improved UV transparency and may occur at relatively high Si concentrations. The assignment was further supported by a bi-exponential decay possessing fast and slow components attributed to the oxygen radicals associated with P–O–Si bonds in the matrix. In addition, the optical properties of Si-Gd co-doped glasses were evaluated. The Si-induced UV transparency and defects luminescence was consistently developed for such glasses, upholding the reproducibility of the material preparation approach. Further, upon exciting 8S7/2 → 6IJ transitions in Gd3+ at 275 nm, the UV type B emission of Gd3+ around 312 nm was progressively enhanced with the increase in Si concentration. Herein, a maximum intensification of about a factor of 6.7 was estimated by use of 0.5 mol% Si together with 2 mol% Gd2O3. Moreover, a linear correlation was observed between the increase in decay times for the Gd3+ 6P7/2 state and the Si contents, providing insights on the physical origin of the enhancement. Accordingly, the data suggested that the improved PL properties of gadolinium(III) originate from the increased UV transparency of the host and the consequent hindering of the detrimental non-radiative energy transfer from Gd3+ to the matrix. Thus, a role of Si as UV PL quenching inhibitor is indicated. The present results may be of use for the development of glasses as luminescent materials for UV type B phototherapy lamps.


19


Page 20 of 24

ASSOCIATED CONTENT Supporting Information Thermal analysis curves, and absorption and photoluminescence excitation spectra are provided as supplementary data. REFERENCES (1) Mokoena, P.P.; Nagpure, I.M.; Kumar, V.; Kroon, R.E.; Olivier, E.J.; Neethling, J.H.; Swart, H.C.; Ntwaeaborwa, O.M. Enhanced UVB Emission and Analysis of Chemical States of Ca5(PO4)3OH:Gd3+,Pr3+ Phosphor Prepared by Co-Precipitation. J. Phys. Chem. Solids 2014, 75, 998-1003. (2) Ramteke, D.D.; Gedam, R.S. Luminescence Properties of Gd3+ Containing Glasses for Ultra-violet (UV) Light. J. Rare Earths 2014, 32, 389-393. (3) Singh, V.; Sivaramaiah, G.; Rao, J.L.; Kim, S.H. Luminescence and EPR Studies of Gd3+Activated Strong UV-Emitting CaZrO3 Phosphors Prepared via Solution Combustion Method. J. Electron. Mater. 2014, 143, 3486-3492. (4) V.V. Shinde, R.G. Kunghatkar, S.J. Dhoble, UVB-Emitting Gd3+-Activated M2O2S (where M = La, Y) for Phototherapy Lamp Phosphors. Luminescence 2015, 30, 1257-1262. (5) Jiménez, J.A. Enhanced UV Emission of Gd3+ in Glass by Ag+ Co-Doping. Mater. Lett. 2015, 159, 193-196. (6) Chauhan, A.O.; Gawande, A.B.; Omanwar, S.K. A Novel Gd3+–Pb2+ Doped LiSrPO4 Phosphor for Phototherapy Lamp Applications. J. Inorg. Organomet. Polym. 2016, 26, 1023-1027.


20

Page 21 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(7) Kalpana, T.; Gandhi, Y.; Sanyal, B.; Sudarsan, V.; Bragiel, P.; Piasecki, M.; Ravi Kumar, V.; Veeraiah, N. Influence of Alumina on Photoluminescence and Thermoluminescence Characteristics of Gd3+ Doped Barium Borophosphate Glasses. J. Lumin. 2016, 179, 44-49. (8) Jiménez, J.A. Ag and Si Codoped Phosphate Glasses: Plasmonic Nanocomposites with Enhanced UV Transparency. J. Am. Ceram. Soc. 2017, 100, 125-132. (9) Liang, X.; Xing, Z.; Yang, Y.; Wang, S.; Chen, G. Luminescence Properties of Eu2+/Mn2+ Codoped Borophosphate Glasses. J. Am. Ceram. Soc. 2011, 94, 849-853. (10) Jiménez, J.A. Silicon as Reducing Agent for Controlled Production of Plasmonic Copper Nanocomposite Glasses: A Spectroscopic Study. J. Electron. Mater. 2015, 44, 4418-4423. (11) Lv, T.-S.; Xu, X.-H.; Yu, X.; Qiu, J.-B. Evolution in the Oxidation Valences and Sensitization Effect of Copper Through Modifying Glass Structure and Sn2+/Si Codoping. J. Am. Ceram. Soc. 98, 2015, 2078-2085. (12) Jiménez, J.A. Blue-Emitting Phosphate Glasses Synthesized via Reduction of Tin(IV) by Silicon. Opt. Mater. 2017, 66, 179-184. (13) Yamane, M.; Asahara, Y. Glasses for Photonics; Cambridge University Press: UK, 2000. (14) Sendova, M.; Jiménez, J.A.; Honaman, C. Rare Earth-Dependent Trend of the Glass Transition Activation Energy of Doped Phosphate Glasses: Calorimetric Analysis. J. NonCryst. Solids 2016, 450, 18-22. (15) Ehrt, D.; Seeber, W. Glass for High Performance Optics and Laser Technology. J. NonCryst. Solids 1991, 129, 19-30. (16) Ehrt, D.; Carl, M.; Kittel, T.; Müller, M.; Seeber, W. High-Performance Glass for the Deep Ultraviolet Range. J. Non-Cryst. Solids 1994, 177, 405-419.


21


Page 22 of 24

(17) Jiménez, J.A.; Sendova, M.; Fachini, E.R.; Zhao, C. Enhanced UV Transparency in Phosphate Glasses via Multi-Wall Carbon Nanotubes. J. Mater. Chem. C 2016, 4, 97719778. (18) Jiménez, J.A. On the Graphite-Induced UV Transparency in Phosphate Glasses. Opt. Mater. 2016, 62, 42-46. (19) Galleani, G.; Ledemi, Y.; de Lima Filho, E.S.; Morency, S.; Delaizir, G.; Chenu, S.; Duclere, J.R.; Messaddeq, Y. UV-Transmitting Step-Index Fluorophosphate Glass Fiber Fabricated by the Crucible Technique. Opt. Mater. 2017, 64, 524-532. (20) Tauc, J.; Menth, A. States in the Gap. J. Non-Cryst. Solids 1972, 8-10, 569-585. (21) Jiménez, J.A. UV Emission of Gd3+ in the Presence of Cu2+: Towards Luminescence Quenching through Quantum Cutting? ChemPhysChem 2015, 16, 1683-1686. (22) Gandhi, Y.; Rajanikanth, P.; Sundara Rao, M.; Ravi Kumar, V.; Veeraiah, N.; Piasecki, M. Effect of Tin Ions on Enhancing the Intensity of Narrow Luminescence Line at 311 nm of Gd3+ Ions in Li2O-PbO-P2O5 Glass System. Opt. Mater. 2016, 57, 39-44. (23) Tong, Y.; Yan, Z.; Zeng, H.; Chen, G. Enhanced Blue Emission of SnO2 Doped Phosphate Glasses by Gd2O3 Co-Doping. J. Lumin. 2014, 145, 438-442.


22

Page 23 of 24 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60



23


Graphical Abstract 349x218mm (96 x 96 DPI)


Page 24 of 24

Silicon-Induced UV Transparency in Phosphate Glasses and Its

Recommend Documents