- Email: [email protected]
Correlation between room temperature luminescence and energy-transfer in Er–Au co-implanted silica
Nuclear Instruments and Methods in Physics Research B 362 (2015) 68–71
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research B journal homepage: www.elsevier.com/locate/nimb
Correlation between room temperature luminescence and energytransfer in Er–Au co-implanted silica T. Cesca a,⇑, B. Kalinic a, C. Maurizio a, C. Scian a, E. Trave b, G. Battaglin b, P. Mazzoldi a, G. Mattei a a b
Department of Physics and Astronomy and CNISM, via Marzolo 8, I-35131 Padova, University of Padova, Italy Department of Molecular Sciences and Nanosystems, Dorsoduro 2137, I-30123 Venice, Ca’ Foscari University of Venice, Italy
a r t i c l e
i n f o
Article history: Received 29 August 2013 Received in revised form 9 December 2014 Accepted 11 September 2015 Available online 20 September 2015 Keywords: Sequential ion implantation Metal nanoclusters Energy-transfer Luminescence
a b s t r a c t We report on the room temperature photoluminescence characterization in the visible and near-infrared range of Er–Au co-implanted silica systems as a function of the annealing temperature. Besides the characteristic Er3+ emission at 1540 nm, the samples exhibit luminescence bands in the wavelength region 600–1400 nm related to the formation of ultra-small AuN aggregates with a number of atoms N less than 50 atoms. In particular, the correlation between such AuN-related luminescence and the enhancement of the Er3+ emission was investigated and an anti-correlation between the Er3+ luminescence at 1540 nm and an AuN-related band at 980 nm was revealed that represents a possible path for the energytransfer from AuN nanoclusters to Er3+ ions, giving rise to the Er3+ sensitized emission. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Due to their plasmonic properties, noble metal nanoparticles (NPs) have been the object of strong interest from the scientific community during these years and they have been effectively employed for applications in many different fields as sensing [1,2], catalysis [3] and photovoltaics [4]. Au nanoparticles in particular can be considered as a sort of prototype when dealing with the physical properties of metallic nanostructures [5] and since two decades they are strongly investigated for their linear and nonlinear optical properties [6–8]. Nonetheless, other very interesting properties arise when the nanoparticle size is reduced down to the sub-nanometer range, that is in a pre-plasmonic regime that corresponds to nanoclusters made just by very few atoms (up to some tens). From the point of view of the optical properties, indeed, such ultra-small nanoclusters are characterized by discrete electronic states and the opening of a HOMO–LUMO gap that exceeds thermal energy already at room temperature, allowing for efficient radiative relaxation and strongly enhancing the luminescence quantum yield [9,10]. As an example, an increase of the near-infrared luminescence yield of more than 5 orders of magnitude has been reported for 1.7 nm gold nanoclusters with respect of bulk gold [11]. Furthermore, in a recent work [12] our group demonstrated another important property of ultrasmall Au NPs obtained by ion implantation in silica: they act as very efficient nanoantennas for the energy-transfer to Er ions, strongly ⇑ Corresponding author. E-mail address: [email protected] (T. Cesca). http://dx.doi.org/10.1016/j.nimb.2015.09.040 0168-583X/Ó 2015 Elsevier B.V. All rights reserved.
enhancing the Er-related luminescence efficiency (at 1540 nm). This result has a remarkable technological relevance since many research efforts have been devoted in these years to determine the best solutions to increase the small cross-section for Er excitation that represents the main limitation to the realization of efficient optoelectronic devices [13]. In Ref. [12] ion implantation has been used to incorporate in silica matrices both the dopant species (Er and Au). For gold in particular this is of paramount importance since by properly combining implantation schemes and postimplantation thermal treatments it is possible to control the nucleation and growth of the ultra-small Au nanoclusters necessary for the energy-transfer to Er ions to efficiently occur [12]. Nonetheless, different aspects are involved in the energytransfer process from ultra-small Au nanoclusters and Er ions and concur in determining its efficiency. Aiming at elucidating some of them, in this work we investigated the interplay between the Au-related room temperature photoluminescence in the visible and near-infrared (NIR) range, and the energy-transfer to Er ions in Er–Au co-implanted silica systems. Our preliminary results show a quenching of the Au-related NIR luminescence correlated to the corresponding enhancement of the Er emission, thus revealing a possible path for the transfer of energy from ultra-small Au nanoclusters to Er ions. 2. Experimental Er-doped silica slides were produced by sequential implantations of Er ions at three energies, 180 + 100 + 50 keV, with a total
69
fluence of 7.2 1014 Er+/cm2 and annealing in nitrogen atmosphere at 800 °C for 1 h. Sequential Au implantations were then performed on the Er-doped silica slabs with an analogous triple energy scheme. The implantation energies were 190 + 110 + 60 keV and the total Au fluence was 7.2 1015 Au+/cm2. The Er and Au implantation parameters were chosen to obtain almost flat implantation profiles (about 70 nm thick) and their overlap in the Er–Au co-implanted samples. Both Er and Au implantation fluences were estimated by Rutherford Backscattering Spectrometry (RBS). After Au implantation, the samples were then heated for 1 h in nitrogen atmosphere at temperatures in the range 300–800 °C to induce Au clustering and implantation damage recovery of the silica matrix. In the following the samples will be named according to the [Au]/[Er] concentration ratio ([Au]/[Er] = 10) as Er1Au10. The Er-implanted reference sample (prior to Au implantation) will be referred as Er800. To determine the Au nanocluster size and its evolution upon annealing, Extended X-ray Absorption Fine Structure (EXAFS) experiments were performed at the Au L3-edge in fluorescence mode at the Italian beamline GILDA of the European Synchrotron Radiation Facility (Grenoble, France). The monochromator was equipped with a couple of (1 1 1) Si crystals and harmonic rejection was achieved with a couple of Pd-coated mirrors, working at an incidence angle of 3 mrad. The Au fluorescence signal was collected by a 13-element high-purity Ge detector. The samples were cooled down at 80 K to avoid radiation damage and to reduce as much as possible thermal vibrations. The analysis of the EXAFS spectra was performed by the FEFF8-FEFFIT 2.98 package [14,15]. Photoluminescence (PL) measurements were performed at room temperature by pumping the samples with the 488 nm line of a cw Ar laser. The pumping wavelength matches the Er absorption transition 4I15/2 ? 4F7/2. The pump power was 22 mW and the laser beam was mechanically chopped at 6 Hz. The PL signal was spectrally resolved in the range 600–1700 nm by a single grating monochromator and detected by a N2-cooled photomultiplier tube (Hamamatsu R5509-72). The spectra were recorded with a lock-in amplifier using the chopper frequency as reference signal.
3. Results and discussion In order to characterize the luminescence properties of the Er–Au co-implanted samples investigated in this work we performed room temperature PL measurements in the wavelength range 600–1700 nm. As an example, in Fig. 1 we reported the room temperature PL spectra of the Er1Au10 samples annealed in nitrogen atmosphere at 300 °C, 550 °C and 700 °C. The measurements have been taken upon resonant Er3+ excitation at 488 nm. The PL spectrum of the reference sample Er800 (prior to Au implantations) is also reported as a continuous line. The spectra are vertically shifted to improve the visualization. All the samples exhibit a distinctive PL feature at 1540 nm that is characteristic of the Er3+ luminescent emission related to the 4I13/2 ? 4I15/2 transition [13]. The intensity of this band varies as a function of the annealing temperature and its maximum is reached for thermal treatments at 550 °C: in this case the sample exhibits an intense PL peak at 1540 nm, about a factor of two higher with respect to the Er800 reference sample. By studying analogous Er–Au co-implanted systems, it has been demonstrated by our group that such enhancement of the Er3+ luminescence is induced by the ultra-small Au nanoclusters formed in the samples upon the annealing treatments that act as sensitizers for the Er emission [12]. Furthermore, interesting luminescent features appear looking at the spectra in the wavelength range 600–1400 nm whose intensity also depends on the temperature of the thermal process. In particular, for low annealing temperatures the spectra of the
PL intensity (arb. u.)
T. Cesca et al. / Nuclear Instruments and Methods in Physics Research B 362 (2015) 68–71
38 36 Er1Xe12-600°C 34 32 30 28 Er800 26 24 22 20 700°C 18 16 14 12 10 550°C 8 6 4 2 300 °C 0 600 800 1000 1200 1400 1600
wavelength (nm) Fig. 1. Room temperature PL spectra of the Er–Au co-implanted silica samples (Er1Au10) annealed in nitrogen atmosphere at the indicated temperatures. For comparison the PL spectrum of the reference sample Er800 and the Er1Xe12 sample annealed in nitrogen atmosphere at 600 °C are also shown. The spectra are vertically shifted of an arbitrary amount to improve the visualization. All the samples were excited by the 488 nm line of a cw Ar laser.
Er1Au10 samples are characterized by the convolution of three main bands peaked at about 750 nm, 980 nm and 1150 nm. Concerning the physical origin of such luminescence bands we have shown very recently that they are due to the presence in the Er–Au co-implanted samples of the ultra-small Au aggregates responsible for the sensitized Er3+ luminescence [16,17]. Structural characterizations performed by EXAFS measurements on some selected samples, indeed, revealed the formation in the samples of ultra-small Au nanoclusters whose size increases by increasing the annealing temperature. The results (number of Au atoms per cluster, N, and cluster diameter, D) are reported in Table 1. The cluster diameter is calculated according to the relation reported pffiffiffiffi in Ref. [12]: D ¼ 2Ra 3 N , where Ra is the average radius of an Au atom in the cluster (Ra = 1.505 Å). Au nanoclusters with less than 50 atoms and D < 1 nm (measured in the sample annealed at 800 °C) are formed. Owing to their ultra-small dimensions such Au nanoclusters have a molecule-like character with an electronic structure characterized by discrete energy levels whose energy separation at room temperature is larger than thermal energy, making radiative transitions competitive with non-raditative phonon mediated relaxation processes [9,10]. On the other hand, no evidence of damage-related luminescence (possibly due to Er and/or Au implantations) is observed in the wavelength range explored, as confirmed by looking at the spectra of the Er800 Table 1 Photoluminescence intensity (measured at the two indicated wavelengths) and EXAFS results as a function of the annealing temperature. N is the number of Au atoms per cluster and D is the Au cluster diameter calculated according to the relation reported in the text. Sample
Annealing T (°C) PL intensity (arb. u)
EXAFS
@ 980 nm @ 1540 nm N (atoms) D (nm) Er1Au10 As-imp 300 400 500 550 700 800
– 4.9 4.9 6.0 3.6 0.8 0.6
– 3.5 5.4 11.2 22.6 3.7 1.9
3±1 – 4±1 – 20 ± 10 – 50 ± 30
0.45 ± 0.13 – 0.47 ± 0.12 – 0.87 ± 0.04 – 1.14 ± 0.02
T. Cesca et al. / Nuclear Instruments and Methods in Physics Research B 362 (2015) 68–71
sample (continuous line) and of the Er1Xe12 sample annealed at 600 °C (white diamonds) reported in Fig. 1 that do not show any spectral feature but the Er3+ related luminescence at 1540 nm. The Er1Xe12-600 °C sample in particular belongs to a set of Er– Xe co-implanted samples made with the specific aim of investigating the possible effects of implantation damage on the Er3+ luminescence properties [18]. In that case Xe was used to reproduce the same amount of damage as Au (with a proper choice of the implantation parameters) and it was demonstrated that independently of Xe fluence and annealing temperature no sensitization effect of the Er emission is induced by Xe clusters (bubbles) in spite of their morphological similarity with Au aggregates. Analogously, no luminescent feature but the Er emission at 1540 nm can be observed in the wavelength range 600–1700 nm [16,17]. The luminescence bands detected from the samples investigated in the present work (Er1Au10) in the wavelength range 600–1400 nm are therefore unambiguously related to the presence in the samples of the ultra-small Au nanoclusters (AuN-related luminescence). In order to highlight the correlation between the AuN-related luminescence and the Er emission we have reported in Fig. 2 the evolution, as a function of the annealing temperature, of the PL intensity measured from the Er1Au10 samples at 750 nm, 980 nm and 1150 nm (AuN-related bands), and at 1540 nm (Er emission). For the Er emission (black dots) the trend is characterized by a maximum for annealing at 550 °C at which the PL intensity is about double the value measured in the Er800 reference sample (horizontal dashed line in Fig. 2), as previously underlined, due to the sensitization effect induced by the Au nanoclusters. The progressive increase of the Er3+ PL for T < 550 °C is related to the nucleation/aggregation of Au atoms in ultra-small clusters accompanied by a progressive recovery of the silica matrix from radiation damage. Conversely, the intensity decrease observed at higher annealing temperatures is due to the further Au cluster growth (see Table 1) that makes the Au nanoclusters less efficient as Er sensitizers, acting instead as possible non-radiative recombination centers for the Er emission. Concerning the AuN luminescence bands, on the other hand, the data in Fig. 2 reveal a slight increase of intensity for increasing annealing temperatures at T 6 500 °C, likely related to the radiation damage recovery, followed by a net decrease at higher temperatures. In particular, the 980 nm component (half-white circles) exhibits an anti-correlated trend with respect to the Er emission in the temperature range close to the
PL intensity (arb. u.)
AuN
Er3+
@ 750 nm @ 980 nm @ 1150 nm
20
@ 1540 nm
15 Er800
10
4. Conclusions In summary, the room temperature photoluminescence properties of Er–Au co-implanted silica samples have been characterized
12 10
50
8
40
6
30
4
20
2
10
0 600
5
60
Er1Au10 ann. 550 °C Er1Au5 ann. 620 °C Er1Au1 ann. 720 °C
PL intensity (arb. u.)
25
maximum of the Er3+ luminescence: indeed at T = 550 °C the Er emission at 1540 nm is double with respect to the value measured in the sample annealed at 500 °C, while the 980 nm component is almost halved (see Table 1). This behavior indicates a correlation between the 980 nm component of the AuN luminescence and the Er emission, suggesting at the same time a possible path for the energy-transfer from Au nanoclusters to Er ions. This correlation is further highlighted in Fig. 3 where we have compared the room temperature PL spectrum of the Er1Au10 sample annealed at 550 °C with the spectra of analogous Er–Au co-implanted samples in which Au has been implanted at lower fluences corresponding to [Au]/[Er] concentration ratios of [Au]/[Er] = 5 (Er1Au5) and [Au]/[Er] = 1 (Er1Au1). These two samples have been annealed in nitrogen atmosphere at 620 °C and 720 °C, respectively. To improve the visualization of all the spectral features, different vertical scales have been used for wavelengths shorter (left-hand scale) and longer (right-hand scale) than 1400 nm, as indicated by the vertical dashed line at 1400 nm in the figure. We have found that the annealing temperature necessary to get the maximum Er emission shifts towards higher values for decreasing Au fluence. For each sample series in Fig. 3, therefore, the chosen temperatures represent the thermal treatment conditions to get the maximum Er3+ luminescence at 1540 nm. The dashed gray arrows in the figure highlight the anticorrelated trend exhibited by the PL emission at 1540 nm and at 980 nm: the progressive increase of the Er3+ luminescence at 1540 nm is related to the corresponding decrease of the AuN luminescence at 980 nm. On the other hand, this behavior suggests a possible mechanism for the sensitization of the Er emission. Er3+ ions in silica indeed have the characteristic absorption transition 4 I15/2 ? 4I1/2 (980 nm) that is resonant with the corresponding AuN-related component at 980 nm and thus this may represent a channel for the transfer of energy from the ultra-small AuN nanoclusters acting as sensitizers and the Er3+ ions, giving rise to the enhancement of the Er3+ emission.
PL intensity (arb. u.)
70
0 800
1000
1200
1400
1600
wavelength (nm) 0
300
400
500
600
700
800
Annealing temperature (°C) Fig. 2. Integrated PL intensity measured at the different indicated wavelengths as a function of the annealing temperature of the Er1Au10 samples. The emission at 1540 nm represents the Er3+ luminescence, while the other three wavelengths correspond to the maxima of the main detected AuN-related PL bands. The Er emission intensity measured in the Er800 reference sample is also indicated in the graph.
Fig. 3. Comparison between room temperature PL spectra of the Er1Au10 sample annealed at 550 °C (black squares), the Er1Au5 sample annealed at 620 °C (black continuous line) and the Er1Au1 sample annealed at 720 °C (white dots). The samples have been thermally treated in nitrogen atmosphere. The annealing temperatures represent the thermal treatment conditions to get the maximum Er3+ PL emission at 1540 nm. Different vertical scales have to be considered for the two wavelength regions separated by the vertical dashed line at 1400 nm: the left-hand scale for shorter wavelengths, while the right-hand scale for longer wavelengths. The gray dashed arrows show the anti-correlated trend of the PL emission at 980 nm and 1540 nm.
T. Cesca et al. / Nuclear Instruments and Methods in Physics Research B 362 (2015) 68–71
in the wavelength range 600–1700 nm as a function of the annealing temperature under Er3+ resonant excitation at 488 nm by a cw Ar laser. Besides the characteristic Er3+ PL emission at 1540 nm, the spectra revealed the occurrence of three main luminescence bands at about 750 nm, 980 nm and 1150 nm related to the formation in the samples of ultra-small Au aggregates with less than 50 atoms (AuN luminescence). The relative intensity of these bands depends on the annealing temperature and an anticorrelated behavior has been show between the Er3+ luminescence at 1540 nm and the 980 nm AuN-related component that, being resonant with the Er3+ absorption transition 4I15/2 ? 4I1/2 at 980 nm, may represent a possible path for the energy-transfer from AuN nanoclusters to Er3+ ions giving rise to the Er3+ sensitized emission. Acknowledgment This work has been partially supported by the Progetto di Ateneo CPDA101587 of the University of Padova, Italy. References [1] J.N. Anker, W.P. Hall, O. Lyandres, N.C. Shah, J. Zhao, R.P. Van Duyne, Nat. Mater. 7 (2008) 442–453.
71
[2] G. Mattei, G. Mattei, P. Mazzoldi, M.L. Post, D. Buso, M. Guglielmi, A. Martucci, Adv. Mater. 19 (2007) 561–564. [3] M.S. Chen, D.W. Goodman, Science 306 (2004) 252–255. [4] H.A. Atwater, A. Polman, Nat. Mater. 9 (2010) 205–213. [5] M.-C. Daniel, D. Astruc, Chem. Rev. 104 (2004) 293–346. [6] G. Mattei, P. Mazzoldi, H. Bernas, Metal nanoclusters for optical properties, in: Materials Science with Ion Beams, in: Topics in Advanced Science, vol. 116, Springer, 2010, pp. 287–316. [7] T. Cesca, P. Calvelli, G. Battaglin, P. Mazzoldi, G. Mattei, Opt. Express 20 (2012) 4537. [8] A.L. Stepanov, Rev. Adv. Mater. Sci 27 (2011) 115. [9] J.P. Wilcoxon, J.E. Martin, F. Parsapour, B. Wiedenman, D.F. Kelley, J. Chem. Phys. 108 (1998) 9137. [10] S. Link, A. Beeby, S. FitzGerald, M.A. El-Sayed, T.G. Schaaff, R.L. Whetten, J. Phys. Chem. B 106 (2002) 3410. [11] T.P. Bigioni, R.L. Whetten, Ö. Dag, J. Phys. Chem. B 104 (2000) 6983. [12] C. Maurizio, E. Trave, G. Perotto, V. Bello, D. Pasqualini, P. Mazzoldi, G. Battaglin, T. Cesca, C. Scian, G. Mattei, Phys. Rev. B 83 (2011) 195430. [13] A. Polman, J. Appl. Phys. 82 (1997) 1. [14] A. Ankudinov, B. Ravel, J.J. Rehr, M. Newville, FEFFIT Manual Within the FEFF Project, University of Washington, Seattle, USA, 1992–1999. [15] A.L. Ankudinov, B. Ravel, J.J. Rehr, S.D. Conradson, Phys. Rev. B 58 (1998) 7565. [16] T. Cesca, C. Maurizio, B. Kalinic, C. Scian, E. Trave, G. Battaglin, P. Mazzoldi, G. Mattei, Nucl. Instr. Phys. Res. B 326 (2014) 7. [17] T. Cesca, B. Kalinic, C. Maurizio, C. Scian, G. Battaglin, P. Mazzoldi, G. Mattei, Nanoscale 6 (2014) 1716. [18] T. Cesca, C. Maurizio, B. Kalinic, G. Perotto, P. Mazzoldi, E. Trave, G. Battaglin, G. Mattei, Opt. Express 20 (2012) 16639.
Contents lists available at ScienceDirect
Nuclear Instruments and Methods in Physics Research B journal homepage: www.elsevier.com/locate/nimb
Correlation between room temperature luminescence and energytransfer in Er–Au co-implanted silica T. Cesca a,⇑, B. Kalinic a, C. Maurizio a, C. Scian a, E. Trave b, G. Battaglin b, P. Mazzoldi a, G. Mattei a a b
Department of Physics and Astronomy and CNISM, via Marzolo 8, I-35131 Padova, University of Padova, Italy Department of Molecular Sciences and Nanosystems, Dorsoduro 2137, I-30123 Venice, Ca’ Foscari University of Venice, Italy
a r t i c l e
i n f o
Article history: Received 29 August 2013 Received in revised form 9 December 2014 Accepted 11 September 2015 Available online 20 September 2015 Keywords: Sequential ion implantation Metal nanoclusters Energy-transfer Luminescence
a b s t r a c t We report on the room temperature photoluminescence characterization in the visible and near-infrared range of Er–Au co-implanted silica systems as a function of the annealing temperature. Besides the characteristic Er3+ emission at 1540 nm, the samples exhibit luminescence bands in the wavelength region 600–1400 nm related to the formation of ultra-small AuN aggregates with a number of atoms N less than 50 atoms. In particular, the correlation between such AuN-related luminescence and the enhancement of the Er3+ emission was investigated and an anti-correlation between the Er3+ luminescence at 1540 nm and an AuN-related band at 980 nm was revealed that represents a possible path for the energytransfer from AuN nanoclusters to Er3+ ions, giving rise to the Er3+ sensitized emission. Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction Due to their plasmonic properties, noble metal nanoparticles (NPs) have been the object of strong interest from the scientific community during these years and they have been effectively employed for applications in many different fields as sensing [1,2], catalysis [3] and photovoltaics [4]. Au nanoparticles in particular can be considered as a sort of prototype when dealing with the physical properties of metallic nanostructures [5] and since two decades they are strongly investigated for their linear and nonlinear optical properties [6–8]. Nonetheless, other very interesting properties arise when the nanoparticle size is reduced down to the sub-nanometer range, that is in a pre-plasmonic regime that corresponds to nanoclusters made just by very few atoms (up to some tens). From the point of view of the optical properties, indeed, such ultra-small nanoclusters are characterized by discrete electronic states and the opening of a HOMO–LUMO gap that exceeds thermal energy already at room temperature, allowing for efficient radiative relaxation and strongly enhancing the luminescence quantum yield [9,10]. As an example, an increase of the near-infrared luminescence yield of more than 5 orders of magnitude has been reported for 1.7 nm gold nanoclusters with respect of bulk gold [11]. Furthermore, in a recent work [12] our group demonstrated another important property of ultrasmall Au NPs obtained by ion implantation in silica: they act as very efficient nanoantennas for the energy-transfer to Er ions, strongly ⇑ Corresponding author. E-mail address: [email protected] (T. Cesca). http://dx.doi.org/10.1016/j.nimb.2015.09.040 0168-583X/Ó 2015 Elsevier B.V. All rights reserved.
enhancing the Er-related luminescence efficiency (at 1540 nm). This result has a remarkable technological relevance since many research efforts have been devoted in these years to determine the best solutions to increase the small cross-section for Er excitation that represents the main limitation to the realization of efficient optoelectronic devices [13]. In Ref. [12] ion implantation has been used to incorporate in silica matrices both the dopant species (Er and Au). For gold in particular this is of paramount importance since by properly combining implantation schemes and postimplantation thermal treatments it is possible to control the nucleation and growth of the ultra-small Au nanoclusters necessary for the energy-transfer to Er ions to efficiently occur [12]. Nonetheless, different aspects are involved in the energytransfer process from ultra-small Au nanoclusters and Er ions and concur in determining its efficiency. Aiming at elucidating some of them, in this work we investigated the interplay between the Au-related room temperature photoluminescence in the visible and near-infrared (NIR) range, and the energy-transfer to Er ions in Er–Au co-implanted silica systems. Our preliminary results show a quenching of the Au-related NIR luminescence correlated to the corresponding enhancement of the Er emission, thus revealing a possible path for the transfer of energy from ultra-small Au nanoclusters to Er ions. 2. Experimental Er-doped silica slides were produced by sequential implantations of Er ions at three energies, 180 + 100 + 50 keV, with a total
69
fluence of 7.2 1014 Er+/cm2 and annealing in nitrogen atmosphere at 800 °C for 1 h. Sequential Au implantations were then performed on the Er-doped silica slabs with an analogous triple energy scheme. The implantation energies were 190 + 110 + 60 keV and the total Au fluence was 7.2 1015 Au+/cm2. The Er and Au implantation parameters were chosen to obtain almost flat implantation profiles (about 70 nm thick) and their overlap in the Er–Au co-implanted samples. Both Er and Au implantation fluences were estimated by Rutherford Backscattering Spectrometry (RBS). After Au implantation, the samples were then heated for 1 h in nitrogen atmosphere at temperatures in the range 300–800 °C to induce Au clustering and implantation damage recovery of the silica matrix. In the following the samples will be named according to the [Au]/[Er] concentration ratio ([Au]/[Er] = 10) as Er1Au10. The Er-implanted reference sample (prior to Au implantation) will be referred as Er800. To determine the Au nanocluster size and its evolution upon annealing, Extended X-ray Absorption Fine Structure (EXAFS) experiments were performed at the Au L3-edge in fluorescence mode at the Italian beamline GILDA of the European Synchrotron Radiation Facility (Grenoble, France). The monochromator was equipped with a couple of (1 1 1) Si crystals and harmonic rejection was achieved with a couple of Pd-coated mirrors, working at an incidence angle of 3 mrad. The Au fluorescence signal was collected by a 13-element high-purity Ge detector. The samples were cooled down at 80 K to avoid radiation damage and to reduce as much as possible thermal vibrations. The analysis of the EXAFS spectra was performed by the FEFF8-FEFFIT 2.98 package [14,15]. Photoluminescence (PL) measurements were performed at room temperature by pumping the samples with the 488 nm line of a cw Ar laser. The pumping wavelength matches the Er absorption transition 4I15/2 ? 4F7/2. The pump power was 22 mW and the laser beam was mechanically chopped at 6 Hz. The PL signal was spectrally resolved in the range 600–1700 nm by a single grating monochromator and detected by a N2-cooled photomultiplier tube (Hamamatsu R5509-72). The spectra were recorded with a lock-in amplifier using the chopper frequency as reference signal.
3. Results and discussion In order to characterize the luminescence properties of the Er–Au co-implanted samples investigated in this work we performed room temperature PL measurements in the wavelength range 600–1700 nm. As an example, in Fig. 1 we reported the room temperature PL spectra of the Er1Au10 samples annealed in nitrogen atmosphere at 300 °C, 550 °C and 700 °C. The measurements have been taken upon resonant Er3+ excitation at 488 nm. The PL spectrum of the reference sample Er800 (prior to Au implantations) is also reported as a continuous line. The spectra are vertically shifted to improve the visualization. All the samples exhibit a distinctive PL feature at 1540 nm that is characteristic of the Er3+ luminescent emission related to the 4I13/2 ? 4I15/2 transition [13]. The intensity of this band varies as a function of the annealing temperature and its maximum is reached for thermal treatments at 550 °C: in this case the sample exhibits an intense PL peak at 1540 nm, about a factor of two higher with respect to the Er800 reference sample. By studying analogous Er–Au co-implanted systems, it has been demonstrated by our group that such enhancement of the Er3+ luminescence is induced by the ultra-small Au nanoclusters formed in the samples upon the annealing treatments that act as sensitizers for the Er emission [12]. Furthermore, interesting luminescent features appear looking at the spectra in the wavelength range 600–1400 nm whose intensity also depends on the temperature of the thermal process. In particular, for low annealing temperatures the spectra of the
PL intensity (arb. u.)
T. Cesca et al. / Nuclear Instruments and Methods in Physics Research B 362 (2015) 68–71
38 36 Er1Xe12-600°C 34 32 30 28 Er800 26 24 22 20 700°C 18 16 14 12 10 550°C 8 6 4 2 300 °C 0 600 800 1000 1200 1400 1600
wavelength (nm) Fig. 1. Room temperature PL spectra of the Er–Au co-implanted silica samples (Er1Au10) annealed in nitrogen atmosphere at the indicated temperatures. For comparison the PL spectrum of the reference sample Er800 and the Er1Xe12 sample annealed in nitrogen atmosphere at 600 °C are also shown. The spectra are vertically shifted of an arbitrary amount to improve the visualization. All the samples were excited by the 488 nm line of a cw Ar laser.
Er1Au10 samples are characterized by the convolution of three main bands peaked at about 750 nm, 980 nm and 1150 nm. Concerning the physical origin of such luminescence bands we have shown very recently that they are due to the presence in the Er–Au co-implanted samples of the ultra-small Au aggregates responsible for the sensitized Er3+ luminescence [16,17]. Structural characterizations performed by EXAFS measurements on some selected samples, indeed, revealed the formation in the samples of ultra-small Au nanoclusters whose size increases by increasing the annealing temperature. The results (number of Au atoms per cluster, N, and cluster diameter, D) are reported in Table 1. The cluster diameter is calculated according to the relation reported pffiffiffiffi in Ref. [12]: D ¼ 2Ra 3 N , where Ra is the average radius of an Au atom in the cluster (Ra = 1.505 Å). Au nanoclusters with less than 50 atoms and D < 1 nm (measured in the sample annealed at 800 °C) are formed. Owing to their ultra-small dimensions such Au nanoclusters have a molecule-like character with an electronic structure characterized by discrete energy levels whose energy separation at room temperature is larger than thermal energy, making radiative transitions competitive with non-raditative phonon mediated relaxation processes [9,10]. On the other hand, no evidence of damage-related luminescence (possibly due to Er and/or Au implantations) is observed in the wavelength range explored, as confirmed by looking at the spectra of the Er800 Table 1 Photoluminescence intensity (measured at the two indicated wavelengths) and EXAFS results as a function of the annealing temperature. N is the number of Au atoms per cluster and D is the Au cluster diameter calculated according to the relation reported in the text. Sample
Annealing T (°C) PL intensity (arb. u)
EXAFS
@ 980 nm @ 1540 nm N (atoms) D (nm) Er1Au10 As-imp 300 400 500 550 700 800
– 4.9 4.9 6.0 3.6 0.8 0.6
– 3.5 5.4 11.2 22.6 3.7 1.9
3±1 – 4±1 – 20 ± 10 – 50 ± 30
0.45 ± 0.13 – 0.47 ± 0.12 – 0.87 ± 0.04 – 1.14 ± 0.02
T. Cesca et al. / Nuclear Instruments and Methods in Physics Research B 362 (2015) 68–71
sample (continuous line) and of the Er1Xe12 sample annealed at 600 °C (white diamonds) reported in Fig. 1 that do not show any spectral feature but the Er3+ related luminescence at 1540 nm. The Er1Xe12-600 °C sample in particular belongs to a set of Er– Xe co-implanted samples made with the specific aim of investigating the possible effects of implantation damage on the Er3+ luminescence properties [18]. In that case Xe was used to reproduce the same amount of damage as Au (with a proper choice of the implantation parameters) and it was demonstrated that independently of Xe fluence and annealing temperature no sensitization effect of the Er emission is induced by Xe clusters (bubbles) in spite of their morphological similarity with Au aggregates. Analogously, no luminescent feature but the Er emission at 1540 nm can be observed in the wavelength range 600–1700 nm [16,17]. The luminescence bands detected from the samples investigated in the present work (Er1Au10) in the wavelength range 600–1400 nm are therefore unambiguously related to the presence in the samples of the ultra-small Au nanoclusters (AuN-related luminescence). In order to highlight the correlation between the AuN-related luminescence and the Er emission we have reported in Fig. 2 the evolution, as a function of the annealing temperature, of the PL intensity measured from the Er1Au10 samples at 750 nm, 980 nm and 1150 nm (AuN-related bands), and at 1540 nm (Er emission). For the Er emission (black dots) the trend is characterized by a maximum for annealing at 550 °C at which the PL intensity is about double the value measured in the Er800 reference sample (horizontal dashed line in Fig. 2), as previously underlined, due to the sensitization effect induced by the Au nanoclusters. The progressive increase of the Er3+ PL for T < 550 °C is related to the nucleation/aggregation of Au atoms in ultra-small clusters accompanied by a progressive recovery of the silica matrix from radiation damage. Conversely, the intensity decrease observed at higher annealing temperatures is due to the further Au cluster growth (see Table 1) that makes the Au nanoclusters less efficient as Er sensitizers, acting instead as possible non-radiative recombination centers for the Er emission. Concerning the AuN luminescence bands, on the other hand, the data in Fig. 2 reveal a slight increase of intensity for increasing annealing temperatures at T 6 500 °C, likely related to the radiation damage recovery, followed by a net decrease at higher temperatures. In particular, the 980 nm component (half-white circles) exhibits an anti-correlated trend with respect to the Er emission in the temperature range close to the
PL intensity (arb. u.)
AuN
Er3+
@ 750 nm @ 980 nm @ 1150 nm
20
@ 1540 nm
15 Er800
10
4. Conclusions In summary, the room temperature photoluminescence properties of Er–Au co-implanted silica samples have been characterized
12 10
50
8
40
6
30
4
20
2
10
0 600
5
60
Er1Au10 ann. 550 °C Er1Au5 ann. 620 °C Er1Au1 ann. 720 °C
PL intensity (arb. u.)
25
maximum of the Er3+ luminescence: indeed at T = 550 °C the Er emission at 1540 nm is double with respect to the value measured in the sample annealed at 500 °C, while the 980 nm component is almost halved (see Table 1). This behavior indicates a correlation between the 980 nm component of the AuN luminescence and the Er emission, suggesting at the same time a possible path for the energy-transfer from Au nanoclusters to Er ions. This correlation is further highlighted in Fig. 3 where we have compared the room temperature PL spectrum of the Er1Au10 sample annealed at 550 °C with the spectra of analogous Er–Au co-implanted samples in which Au has been implanted at lower fluences corresponding to [Au]/[Er] concentration ratios of [Au]/[Er] = 5 (Er1Au5) and [Au]/[Er] = 1 (Er1Au1). These two samples have been annealed in nitrogen atmosphere at 620 °C and 720 °C, respectively. To improve the visualization of all the spectral features, different vertical scales have been used for wavelengths shorter (left-hand scale) and longer (right-hand scale) than 1400 nm, as indicated by the vertical dashed line at 1400 nm in the figure. We have found that the annealing temperature necessary to get the maximum Er emission shifts towards higher values for decreasing Au fluence. For each sample series in Fig. 3, therefore, the chosen temperatures represent the thermal treatment conditions to get the maximum Er3+ luminescence at 1540 nm. The dashed gray arrows in the figure highlight the anticorrelated trend exhibited by the PL emission at 1540 nm and at 980 nm: the progressive increase of the Er3+ luminescence at 1540 nm is related to the corresponding decrease of the AuN luminescence at 980 nm. On the other hand, this behavior suggests a possible mechanism for the sensitization of the Er emission. Er3+ ions in silica indeed have the characteristic absorption transition 4 I15/2 ? 4I1/2 (980 nm) that is resonant with the corresponding AuN-related component at 980 nm and thus this may represent a channel for the transfer of energy from the ultra-small AuN nanoclusters acting as sensitizers and the Er3+ ions, giving rise to the enhancement of the Er3+ emission.
PL intensity (arb. u.)
70
0 800
1000
1200
1400
1600
wavelength (nm) 0
300
400
500
600
700
800
Annealing temperature (°C) Fig. 2. Integrated PL intensity measured at the different indicated wavelengths as a function of the annealing temperature of the Er1Au10 samples. The emission at 1540 nm represents the Er3+ luminescence, while the other three wavelengths correspond to the maxima of the main detected AuN-related PL bands. The Er emission intensity measured in the Er800 reference sample is also indicated in the graph.
Fig. 3. Comparison between room temperature PL spectra of the Er1Au10 sample annealed at 550 °C (black squares), the Er1Au5 sample annealed at 620 °C (black continuous line) and the Er1Au1 sample annealed at 720 °C (white dots). The samples have been thermally treated in nitrogen atmosphere. The annealing temperatures represent the thermal treatment conditions to get the maximum Er3+ PL emission at 1540 nm. Different vertical scales have to be considered for the two wavelength regions separated by the vertical dashed line at 1400 nm: the left-hand scale for shorter wavelengths, while the right-hand scale for longer wavelengths. The gray dashed arrows show the anti-correlated trend of the PL emission at 980 nm and 1540 nm.
T. Cesca et al. / Nuclear Instruments and Methods in Physics Research B 362 (2015) 68–71
in the wavelength range 600–1700 nm as a function of the annealing temperature under Er3+ resonant excitation at 488 nm by a cw Ar laser. Besides the characteristic Er3+ PL emission at 1540 nm, the spectra revealed the occurrence of three main luminescence bands at about 750 nm, 980 nm and 1150 nm related to the formation in the samples of ultra-small Au aggregates with less than 50 atoms (AuN luminescence). The relative intensity of these bands depends on the annealing temperature and an anticorrelated behavior has been show between the Er3+ luminescence at 1540 nm and the 980 nm AuN-related component that, being resonant with the Er3+ absorption transition 4I15/2 ? 4I1/2 at 980 nm, may represent a possible path for the energy-transfer from AuN nanoclusters to Er3+ ions giving rise to the Er3+ sensitized emission. Acknowledgment This work has been partially supported by the Progetto di Ateneo CPDA101587 of the University of Padova, Italy. References [1] J.N. Anker, W.P. Hall, O. Lyandres, N.C. Shah, J. Zhao, R.P. Van Duyne, Nat. Mater. 7 (2008) 442–453.
71
[2] G. Mattei, G. Mattei, P. Mazzoldi, M.L. Post, D. Buso, M. Guglielmi, A. Martucci, Adv. Mater. 19 (2007) 561–564. [3] M.S. Chen, D.W. Goodman, Science 306 (2004) 252–255. [4] H.A. Atwater, A. Polman, Nat. Mater. 9 (2010) 205–213. [5] M.-C. Daniel, D. Astruc, Chem. Rev. 104 (2004) 293–346. [6] G. Mattei, P. Mazzoldi, H. Bernas, Metal nanoclusters for optical properties, in: Materials Science with Ion Beams, in: Topics in Advanced Science, vol. 116, Springer, 2010, pp. 287–316. [7] T. Cesca, P. Calvelli, G. Battaglin, P. Mazzoldi, G. Mattei, Opt. Express 20 (2012) 4537. [8] A.L. Stepanov, Rev. Adv. Mater. Sci 27 (2011) 115. [9] J.P. Wilcoxon, J.E. Martin, F. Parsapour, B. Wiedenman, D.F. Kelley, J. Chem. Phys. 108 (1998) 9137. [10] S. Link, A. Beeby, S. FitzGerald, M.A. El-Sayed, T.G. Schaaff, R.L. Whetten, J. Phys. Chem. B 106 (2002) 3410. [11] T.P. Bigioni, R.L. Whetten, Ö. Dag, J. Phys. Chem. B 104 (2000) 6983. [12] C. Maurizio, E. Trave, G. Perotto, V. Bello, D. Pasqualini, P. Mazzoldi, G. Battaglin, T. Cesca, C. Scian, G. Mattei, Phys. Rev. B 83 (2011) 195430. [13] A. Polman, J. Appl. Phys. 82 (1997) 1. [14] A. Ankudinov, B. Ravel, J.J. Rehr, M. Newville, FEFFIT Manual Within the FEFF Project, University of Washington, Seattle, USA, 1992–1999. [15] A.L. Ankudinov, B. Ravel, J.J. Rehr, S.D. Conradson, Phys. Rev. B 58 (1998) 7565. [16] T. Cesca, C. Maurizio, B. Kalinic, C. Scian, E. Trave, G. Battaglin, P. Mazzoldi, G. Mattei, Nucl. Instr. Phys. Res. B 326 (2014) 7. [17] T. Cesca, B. Kalinic, C. Maurizio, C. Scian, G. Battaglin, P. Mazzoldi, G. Mattei, Nanoscale 6 (2014) 1716. [18] T. Cesca, C. Maurizio, B. Kalinic, G. Perotto, P. Mazzoldi, E. Trave, G. Battaglin, G. Mattei, Opt. Express 20 (2012) 16639.