- Email: [email protected]
Studies on phosphorescence and trapping effects of Mn-doped and undoped zinc germinates
Journal of Luminescence ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Contents lists available at ScienceDirect
Journal of Luminescence journal homepage: www.elsevier.com/locate/jlumin
Full Length Article
Studies on phosphorescence and trapping effects of Mn-doped and undoped zinc germinates Zhiyi He a,b, Li Ma b, Xiaojun Wang b,c,n a
Optoelectronic Institute, Guilin University of Electronic Technology, Guilin 541004, Guangxi, China Department of Physics, Georgia Southern University, Statesboro, GA 30460, USA c School of Physics, Northeast Normal University, Changchun 130024, China b
art ic l e i nf o
a b s t r a c t
Article history: Received 24 October 2014 Received in revised form 6 January 2015 Accepted 9 January 2015
Photoluminescence and phosphorescence from different recombining centers in the Mn2 þ -doped and undoped Zn2GeO4 phosphors have been observed. By UV excitation the undoped sample presents a broad band of blue–white emission from the host defects while the Mn-doped samples show both the host and Mn2 þ emissions with different phosphorescent durations. At the beginning of UV excitation after the phosphorescence has been exhausted, the fluorescent time dependence of Mn2 þ exhibits a fast decay process to a constant intensity, different from the rising or charging process as the typical behavior for the common persistent phosphors. This unusual behavior was studied using electron paramagnetic resonance (EPR) spectroscopy. A decrease of the EPR signal from Mn2 þ was found for the sample under UV irradiation, suggesting the occurrence of ionization of Mn2 þ to Mn3 þ . A slow recovering process of the ionization has also been detected, which is consistent with the observation of phosphorescence from Mn2 þ doped samples. & 2015 Elsevier B.V. All rights reserved.
Keywords: Phosphorescence Trap charging Zn2GeO4 EPR Electron–hole pair Mn2 þ
1. Introduction In recent years, persistent phosphors have been studied by many investigators in the effort to develop the phosphors with colorful emissions [1] for more applications, e.g., in vivo biomedical imaging [2,3]. The phosphorescence examination is also an effective way to analyze the charge transfer among the trapping defects [1]. Optical charging process in phosphorescence may provide important facts that reveal the trapping mechanism. In our previous work [4], we have observed a slowly rising time dependence of the photoluminescence of SrAl2O4:Eu,Dy during the excitation period, which was attributed to a trap charging process. The zinc germinate, Zn2GeO4, in nanoscale has been reported having good photonic and electronic characteristics as the photoanode to increase the open circuit voltage of CdSe solar cell [5], anodes for Li-ion batteries to improvement the capacity and cycling life [6], as well as for the UV detector with fast response [7]. The luminescent properties of undoped and Mn2 þ doped Zn2GeO4 have been studied in several reports [8–12], and the host without doping also showed a bluish emission from its native defects by UV excitation [5]. In this work, photoluminescence and
n Correspondence author at: Department of Physics, Georgia Southern University, Statesboro, GA 30458, USA. Tel.: þ1 912 478 5503; fax: þ 1 912 478 0471. E-mail address: [email protected] (X. Wang).
phosphorescence of Mn2 þ doped and undoped Zn2GeO4 have been investigated in details for better understanding the electron/ hole trapping mechanism. Phosphorescence from both the Mn2 þ ions and host defects in Zn2GeO4 has been observed upon UV excitation, while the Mn2 þ ions present a longer phosphorescent duration than the defects. The charging process has also been studied and different behaviors of Mn2 þ and defects observed. The electron paramagnetic resonance (EPR) signal of Mn2 þ has been found being decreased during the UV irradiation, indicating that the population of Mn2 þ ions decreases in the trapping state and the valence changes from Mn2 þ to Mn3 þ . A slow recovering process of the ionization has also been detected, which is consistent with the decay observation of phosphorescence from Mn2 þ doped samples.
2. Experimental 2.1. Samples preparation All samples were prepared by mixing the row materials (4N in purity) of ZnO, GeO2, and MnCO3 according to the desired ratio sintered at 900 1C in N2 þ H2 (8%) reduced atmosphere for 2 h, while both the heating-up and cooling-down processes took 4 h. The Mn doping concentrations are 0 (undoped), 0.001, 0.004, and 0.02 (in mol ratio x, the formula is Zn2 xMnxGeO4), respectively.
http://dx.doi.org/10.1016/j.jlumin.2015.01.026 0022-2313/& 2015 Elsevier B.V. All rights reserved.
Please cite this article as: Z. He, et al., J. Lumin. (2015), http://dx.doi.org/10.1016/j.jlumin.2015.01.026i
Z. He et al. / Journal of Luminescence ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
2.2. Measurements IThe measurements of photoluminescence emission, excitation, and phosphorescence spectra were performed using a SPEX Fluoro Max III and Horiba FluoroLog Spectrometers. The phosphorescent emission spectra were collected by averaging the data acquired by forward and reverse wavelength scan to eliminate the influence of the phosphorescent decay. The timing offset could be several seconds after manually blocking the excitation beam but would not significantly affect the relative line shape of the spectra. The EPR spectra were measured using a Bruker EMXplus X-band spectrometer. Photo-stimulated EPR was performed by illuminating (254 nm) the sample through the optical window of microwave chamber. All measurements were carried out at room temperature.
3. Results and discussion 3.1. Photoluminescence spectra Upon the UV excitation at 260 nm, the updoped Zn2GeO4 sample presents a blue–white emission of the broad band extends from the near-UV to orange–red region (Fig. 1, dashed line), arising from the host crystal defects, which were identified by the EPR experiments as VO or Zni for the donor and VGE or VZn the acceptor in Kröger–Vink notations [8]. The photoluminescence of the Mn2 þ doped sample exhibits a green emission within the 3d5 configuration (4T1-6A1) of Mn2 þ accompanied by the host emission (Fig. 1, solid line). The peak wavelength of the blue–white emission from the host defects shows a little different from that previously reported [8,9], due to the condition dependence of the defect properties on the sample preparation. The intensities of the two emission spectra in Fig. 1 are not in the same scale. Under the same measurement condition, the integrated emission intensity of the undoped sample is calculated to be one order of magnitude lower than that of the doped sample. The low efficiency of the host luminescence can be ascribed to the strong phonon coupling of the transition with the host lattices [13], resulting in the occurrence of nonradiative process. Its broad emission band extended from 380 nm to 600 nm also indicates such a strong electron–phonon coupling in the radiative transition. In general the host luminescence is not as efficient as that of the properly doped activators. Nevertheless it provides an optical approach to examine the intrinsic properties of the host defects for comparison with the doped activators. The excitation spectrum of Zn2GeO4:Mn2 þ shows two bands with the peak wavelengths at 260 nm and 295 nm, respectively, while the undoped sample only presents a strong excitation band around 260 nm, as shown in Fig. 2. The bandgap of Zn2GeO4 was reported at values of 4.5 eV [5] to 4.68 eV [7] corresponding to
Fig. 1. Photoluminescence spectra of Mn2 þ doped (0.4%, solid line) and undoped (dashed line) Zn2GeO4, excited at 254 nm.
the light wavelength around 270 nm, where in Ref. [7] the photocurrent was detected upon 254 nm UV excitation but not 365 nm. The 260 nm excitation band is due to the interband absorption of the host lattice and the one at 295 nm was attributed to charge transfer transition [9]. Mn2 þ can be selectively excited at a longer wavelength, e.g., 330 nm, to avoid the interband absorption, yielding a single green band of Mn2 þ emission. 3.2. Phosphorescent properties After the UV excitation, both the Mn2 þ -doped and undoped samples show the long-lasting phosphorescence with the same emission band shapes and positions as the photoluminescence (peaked at 455 and 533 nm). There were several investigators who reported that the lasting time of Zn2GeO4:Mn2 þ varied from several tens of seconds to more than 1 h [11,12] since its luminescent property is strongly dependent on the preparation conditions including the undoped samples [8,9]. For the doped sample Zn2GeO4:0.004Mn2 þ the phosphorescent emission from Mn2 þ is also dominant over that from the host. To have a clear comparison of their change with time, the samples doped with different Mn concentrations have also been examined. As the Mn concentration increases, the blue–white host emission decreases dramatically. The host emission is hard to be detected at 0.02 mol ratio and almost all the emission shifts to the green band of Mn2 þ . Fig. 3 compares the phosphorescent spectra of the samples doped with 0.001 and 0.004 of Mn2 þ in mol ratio at different time after excitation. For the photoluminescence, the two excitation bands peaked at 295 nm and 260 nm have the same scale of efficiencies. But for phosphorescence, only the 260 nm interband absorption is efficient in producing the afterglow. For excitation at the longer wavelength region above 300 nm, the charge transfer transition corresponding to the excitation at longer wavelength region above 300 nm takes place with the electrons transferring among the localized ions. These electrons do not ionize the trapping process and therefore do not yield phosphorescent emission, referring to the photocurrent behavior under 254 and 365 nm excitations [7]. For the sample with low doping concentration Zn2GeO4:0. 001Mn2 þ , phosphorescence from the host defects has the same scale as that of the Mn2 þ ions (Fig. 3a) but decreases significantly in the sample Zn2GeO4:0.004Mn2 þ doped with a higher concentration (Fig. 3b). Moreover, considering their large intensity difference of photoluminescence under the interband excitation as shown in Fig. 1, it is evident that the Mn2 þ ions are much more competitive in capturing the electrons/holes as the radiative recombination centers. Fig. 4 depicts the phosphorescent decay processes for the doped and undoped samples with their initial intensity normalized, lasting 3 h and 1 h, respectively.
Fig. 2. Excitation spectra of the Mn2 þ doped (0.4%, solid line) and undoped (dashed line) Zn2GeO4, monitored at 530 and 455 nm, respectively.
Please cite this article as: Z. He, et al., J. Lumin. (2015), http://dx.doi.org/10.1016/j.jlumin.2015.01.026i
Z. He et al. / Journal of Luminescence ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Fig. 3. Phosphorescent spectra of (a) Zn2GeO4:0.001Mn2 þ and (b) Zn2GeO4:0.004Mn2 þ at 10, 30, and 100 s after switching off 254 nm excitation.
3
case for the germinate phosphors. Fig. 5 shows the time response of the photoluminescence from the host defects and Mn2 þ by shuttering the 254 nm excitation on and off. Before the measurement the samples are kept in dark for sufficiently long time to ensure the phosphorescent exhaustion and empty trapping states. When the shutter is open, the photoluminescence was found to jump to the maximum intensity in a fast response, even at very weak excitation (Fig. 5b). While for the host defect emission, the intensity keeps constant except for the noise fluctuation during the excitation. A surprising result is that the Mn2 þ luminescence intensity quickly reaches to a maximum then decreases to a stable level during the excitation, in contrary to a slow rising or charging process as observed in SrAl2O4:Eu,Dy [4]. By excitation at a longer wavelength ( 4330 nm) corresponding to the charge transfer transition, the luminescence is found to remain at the same level without such a decreasing process. Furthermore, the initial maximum intensity gradually decreases after each on–off cycle, as shown in Fig. 5(b). The initial peaks of the photoluminescence after the first cycle may reach the same intensity after sufficiently long phosphorescent release in the dark. Therefore, the photoluminescence decreasing process after the interband excitation (254 nm) can be associated with the trap charging mechanism or some recovering process, which will be discussed in next section. In the case of host excitation, the luminescent centers should capture the generated electrons and holes for light emission, viz. recombination via the energy levels of the radiative transition. If the traps exist in the host crystal, they can intercept either the electrons or holes (depending on the trap type) in advance of the luminescent centers until the traps are filled. Thus a gradually rising curve of the photoluminescence time dependence during the charge process is one of the most common behaviors of persistent phosphors. As we observed, the charge curve of SrAl2O4:Eu,Dy increases from nearly zero intensity at the excitation beginning. This can be understood
Fig. 4. Phosphorescent decay processes of the 0.004Mn-doped (monitored at 533 nm) and undoped (monitored at 455 nm) samples with their initial intensity normalized for comparison (after 20 min irradiation under 254 nm).
Fig. 3 also illustrates that the phosphorescence from the host defects presents a faster decay process than the Mn2 þ . This difference results from their different recombining rate and lifetime of the trapped carriers on the defects. It is presumed that the luminescent defects and Mn2 þ ions are the hole trapping centers and they share the same electrons trapped at other sites as identified as VO and Zni defects [8]. The hole trapping centers as electron acceptors VGE and VZn, as also identified in the same work, are adjacent to the electron traps. Therefore, the holes are filled by the electrons faster than those trapped at the Mn2 þ ions that are spatially separated far away from the electron traps. 3.3. Time dependence of photoluminescence If the excitation light is turned after exhaustion of the phosphorescence when all the electron/hole traps are empty, the time dependence of the photoluminescence reflects the optical charging process of the traps. Slowly rising photoluminescence has been observed for the persistent phosphor SrAl2O4:Eu,Dy and its charging behavior analyzed [4]. The charging curve is typical and reasonable for many phosphors capable of storing photon energy. By the same approach, it is expected to observe a similar rising curve of the time dependence in this work. However, this is not the
Fig. 5. Response of the photoluminescence from (a) the host defects monitored at 455 nm and (b) Mn2 þ ions of Zn2GeO4:0.004Mn2 þ at 533 nm by shuttering the 254 nm excitation on and off.
Please cite this article as: Z. He, et al., J. Lumin. (2015), http://dx.doi.org/10.1016/j.jlumin.2015.01.026i
4
Z. He et al. / Journal of Luminescence ∎ (∎∎∎∎) ∎∎∎–∎∎∎
considering that each of the luminescent centers has to get both the electron and hole spatially coincided at its site for the radiative recombination. For trapping, they can be independently captured by their corresponding type of the traps (electron or hole traps) in random distribution. Hence, trapping should be much more probable than the luminescent recombination unless the traps are filled. For explanation of the contradictory behavior of Zn2GeO4:Mn2 þ , it may be considered that the trapping capacity is too small to observe a rising charge curve by an intensive excitation. This consideration can be ruled out by a similar time response observed at very weak excitation. A reasonable explanation is that most electrons and holes are formed in mobile pairs (e–h pairs) after the interband excitation, which would not be separated and captured by the traps on their way to the luminescent recombining centers. Yet a small amount of independent electrons and holes thermally unpaired are captured by the traps and luminescent centers in spatial separation for the subsequent phosphorescence. The photo-generated e–h pairs in the persistent phosphors have also been reported [1,14]. Except for the electron or hole (single carrier) traps, e–h pair trap may be formed by a donor–acceptor defect complex in some host crystals [1]. In the Zn2GeO4 host, it is more likely that only the single type of carrier traps exists, allowing the e–h pairs to recombine directly on the luminescence centers without being trapped. 3.4. Trapping mechanism study by EPR measurement 2þ
In Fig. 5b, the decreasing process of the Mn photoluminescence may provide some guidance and hints for more detailed examination of the trapping mechanism. The initial maximum intensity becomes lower and lower when on–off cycle is repeated at delayed time. EPR experiment has been conducted to examine the change of Mn2 þ ions in the trapping state. For undoped sample, similar EPR results have been observed to that reported in the GeO2–B2O3– ZnO glasses [10] and they are very reversible with UV (254 nm) exposure on and off. For Mn2 þ -doped Zn2GeO4 samples, EPR signals are measured to examine the population change of Mn2 þ ions in the trapping process. Mn2 þ ions are located in tetrahedral sites in the host with high-spin states (S ¼5/2, Kramer’s system) [15]. The hyperfine structure (nuclear spin I¼ 5/2) of the EPR spectra measured at this high concentration is unable to be resolved, however, the broad EPR signal at g ¼2.001 can be ascribed to the introduction of Mn2 þ . Fig. 6 depicts the EPR spectra of the sample measured in dark (curve b) and under UV irradiation (curve a). Under the UV irradiation, a small but repeatedly steady decrease of the EPR signal is observed in comparison with the sample in dark. The decrease can be clearly seen at the peak position as zoomed in Fig. 6 (curves a and b). The difference of subtracting curve a from curve b is shown in curve c with 30 magnification. Curve d shows that difference becomes smaller at the second cycle after 100 s recovering in dark from first cycle. The decreases of Mn2 þ EPR signal and the photoluminescence in the trapping state after UV excitation suggest a population decrease of the Mn2 þ ions that may be ionized to Mn3 þ (non-Kramer’s system). Once the valence of Mn2 þ ions is changed in the current host lattice, they lose the role as the luminescent centers, leading to the decrease of photoluminescence that is consistent with the results in Fig. 5b. In addition, the ionization may just occur for the Mn2 þ ions in the surface, which has been observed for Mn2 þ in other hosts [16], depleting the limited population of Mn2 þ at surface even upon weak excitation. The initial maximum emission peaks gradually decrease after each on–off cycle since the recovering process of Mn2 þ from Mn3 þ is slow, as illustrated in curve d (Fig. 6), where just part of Mn3 þ ions returns to Mn2 þ after 100 s in dark. The Mn3 þ ions may also serve as hole traps (Mn3 þ ¼holeþMn2 þ ), yielding phosphorescence. The recovering time is in the same scale of phosphorescence in Mn2 þ
Fig. 6. EPR spectra and their zoomed peaks of Zn2GeO4:0.004Mn2 þ measured by irradiation of a 254 nm mercury lamp (a) and in the dark (b). Curves c and d represents the difference of 30 (b–a), corresponding to the first and second UV on–off cycles in Fig. 5, respectively.
doped samples. This also explains the results for undoped sample shown in Fig. 5a, where the initial peaks are absent or no ionization occurs when UV light is on. Finally, the EPR signal maintains the same if irradiating the doped sample with UV at longer wavelengths (4300 nm), which is consistent with the observation of photoluminescence, i.e., no initial peaks when UV is on for all the samples, as discussed in previous section.
4. Conclusions The phosphorescence from both the Mn2 þ ions and host defects in Zn2GeO4 was observed by the UV interband excitation. The Mn2 þ ions present a longer phosphorescent duration than the defects, due to the different recombination rates of their trapped holes with the detrapped electrons. In the phosphorescent charging process under UV excitation, the electron–hole pairs are formed and transport across the single carrier traps to the radiative recombining centers, leading to an immediate response to the maximum intensity for the Mn2 þ and host defect photoluminescence. Nevertheless some of the pairs are thermally separated as independent electrons and holes and then trapped for the subsequent phosphorescence. During the charging period the host defect emission maintains a constant intensity, while Mn2 þ emission shows a decay curve to the lower intensity due to the ionization of Mn2 þ to Mn3 þ . This is verified by EPR and UV stimulated EPR measurements. A slow recovering process of the ionization is also detected, which is consistent with the observation of phosphorescence from Mn2 þ doped samples.
Acknowledgments This work was supported by the National Natural Science Foundation of China (Grant no. 61077036), the Natural Science Foundation of Guangxi (Grant no. 2011GXNSFA018157), and the Research Foundation of Georgia Southern University.
Please cite this article as: Z. He, et al., J. Lumin. (2015), http://dx.doi.org/10.1016/j.jlumin.2015.01.026i
Z. He et al. / Journal of Luminescence ∎ (∎∎∎∎) ∎∎∎–∎∎∎
References [1] D. Gourier, A. Bessière, S.K. Sharma, L. Binet, B. Viana, N. Basavaraju, K.R. Priolkar, J. Phys. Chem. Solids 75 (2014) 826. [2] T. Maldiney, B. Viana, A. Bessière, D. Gourier, M. Bessodes, D. Scherman, C. Richard, Opt. Mater. 35 (2013) 1852. [3] N. Basavaraju, S. Sharma, A. Bessi‘ ere, B. Viana, D. Gourier, K.R. Priolkar, J. Phys. D: Appl. Phys. 46 (2013) 375401. [4] Z.Y. He, X.J. Wang, W.M. Yen, J. Lumin. 119-120 (2006) 309. [5] L.Y. Lu, J.J. Chen, W.Y. Wang, Appl. Phys. Lett. 103 (12013) 123902. [6] R. Yi, J.K. Feng, D.P. Lv, M.L. Gordin, S. Chen, D. Choi, D.H. Wang, Nano Energy 2 (2013) 498. [7] C.Y. Yan, N.D. Singh, P.S. Lee, Appl. Phys. Lett. 96 (2010) 053108.
5
[8] Z.S. Liu, X.P. Jing, L.X. Wang, J. Electrochem. 154 (2007) H500. [9] M.M. Shang, G.G. Li, D.M. Yang, X.J. Kang, C. Peng, Z.Y. Cheng, J. Lin, Dalton Trans. 40 (2011) 9379. [10] O. Arellano-Tanori, R. Melendrez, M. Pedroza- Montero, B. Castaneda, V. Chernov, W.M. Yen, M. Barboza-Flores, J. Lumin. 128 (2008) 173. [11] Y. Takahashi, M. Ando, R. Ihara, T. Fujiwara, Opt. Mater. Express 1 (2011) 372. [12] M.H. Wan, Y.H. Wang, X.H. Wang, H. Zhao, Z.F. Hu, Opt. Mater. 36 (2014) 650. [13] C.W. Struck, W.H. Fonger, J. Lumin. 10 (1975) 1. [14] J.B. Qiu, H. Igarashi, A. Makishima, Sci. Technol. Adv. Mater. 6 (2005) 431. [15] D.W. Feldman, J. Chem. Phys. 25 (1973) 363. [16] Z.V. Saponjic, N.M. Dimitrijevic, O.G. Poluektov, L.X. Chen, E. Wasinger, U. Welp, D.M. Tiede, X.B. Zuo, T. Rajh, J. Phys. Chem. 110 (2006) 25441.
Please cite this article as: Z. He, et al., J. Lumin. (2015), http://dx.doi.org/10.1016/j.jlumin.2015.01.026i
Contents lists available at ScienceDirect
Journal of Luminescence journal homepage: www.elsevier.com/locate/jlumin
Full Length Article
Studies on phosphorescence and trapping effects of Mn-doped and undoped zinc germinates Zhiyi He a,b, Li Ma b, Xiaojun Wang b,c,n a
Optoelectronic Institute, Guilin University of Electronic Technology, Guilin 541004, Guangxi, China Department of Physics, Georgia Southern University, Statesboro, GA 30460, USA c School of Physics, Northeast Normal University, Changchun 130024, China b
art ic l e i nf o
a b s t r a c t
Article history: Received 24 October 2014 Received in revised form 6 January 2015 Accepted 9 January 2015
Photoluminescence and phosphorescence from different recombining centers in the Mn2 þ -doped and undoped Zn2GeO4 phosphors have been observed. By UV excitation the undoped sample presents a broad band of blue–white emission from the host defects while the Mn-doped samples show both the host and Mn2 þ emissions with different phosphorescent durations. At the beginning of UV excitation after the phosphorescence has been exhausted, the fluorescent time dependence of Mn2 þ exhibits a fast decay process to a constant intensity, different from the rising or charging process as the typical behavior for the common persistent phosphors. This unusual behavior was studied using electron paramagnetic resonance (EPR) spectroscopy. A decrease of the EPR signal from Mn2 þ was found for the sample under UV irradiation, suggesting the occurrence of ionization of Mn2 þ to Mn3 þ . A slow recovering process of the ionization has also been detected, which is consistent with the observation of phosphorescence from Mn2 þ doped samples. & 2015 Elsevier B.V. All rights reserved.
Keywords: Phosphorescence Trap charging Zn2GeO4 EPR Electron–hole pair Mn2 þ
1. Introduction In recent years, persistent phosphors have been studied by many investigators in the effort to develop the phosphors with colorful emissions [1] for more applications, e.g., in vivo biomedical imaging [2,3]. The phosphorescence examination is also an effective way to analyze the charge transfer among the trapping defects [1]. Optical charging process in phosphorescence may provide important facts that reveal the trapping mechanism. In our previous work [4], we have observed a slowly rising time dependence of the photoluminescence of SrAl2O4:Eu,Dy during the excitation period, which was attributed to a trap charging process. The zinc germinate, Zn2GeO4, in nanoscale has been reported having good photonic and electronic characteristics as the photoanode to increase the open circuit voltage of CdSe solar cell [5], anodes for Li-ion batteries to improvement the capacity and cycling life [6], as well as for the UV detector with fast response [7]. The luminescent properties of undoped and Mn2 þ doped Zn2GeO4 have been studied in several reports [8–12], and the host without doping also showed a bluish emission from its native defects by UV excitation [5]. In this work, photoluminescence and
n Correspondence author at: Department of Physics, Georgia Southern University, Statesboro, GA 30458, USA. Tel.: þ1 912 478 5503; fax: þ 1 912 478 0471. E-mail address: [email protected] (X. Wang).
phosphorescence of Mn2 þ doped and undoped Zn2GeO4 have been investigated in details for better understanding the electron/ hole trapping mechanism. Phosphorescence from both the Mn2 þ ions and host defects in Zn2GeO4 has been observed upon UV excitation, while the Mn2 þ ions present a longer phosphorescent duration than the defects. The charging process has also been studied and different behaviors of Mn2 þ and defects observed. The electron paramagnetic resonance (EPR) signal of Mn2 þ has been found being decreased during the UV irradiation, indicating that the population of Mn2 þ ions decreases in the trapping state and the valence changes from Mn2 þ to Mn3 þ . A slow recovering process of the ionization has also been detected, which is consistent with the decay observation of phosphorescence from Mn2 þ doped samples.
2. Experimental 2.1. Samples preparation All samples were prepared by mixing the row materials (4N in purity) of ZnO, GeO2, and MnCO3 according to the desired ratio sintered at 900 1C in N2 þ H2 (8%) reduced atmosphere for 2 h, while both the heating-up and cooling-down processes took 4 h. The Mn doping concentrations are 0 (undoped), 0.001, 0.004, and 0.02 (in mol ratio x, the formula is Zn2 xMnxGeO4), respectively.
http://dx.doi.org/10.1016/j.jlumin.2015.01.026 0022-2313/& 2015 Elsevier B.V. All rights reserved.
Please cite this article as: Z. He, et al., J. Lumin. (2015), http://dx.doi.org/10.1016/j.jlumin.2015.01.026i
Z. He et al. / Journal of Luminescence ∎ (∎∎∎∎) ∎∎∎–∎∎∎
2
2.2. Measurements IThe measurements of photoluminescence emission, excitation, and phosphorescence spectra were performed using a SPEX Fluoro Max III and Horiba FluoroLog Spectrometers. The phosphorescent emission spectra were collected by averaging the data acquired by forward and reverse wavelength scan to eliminate the influence of the phosphorescent decay. The timing offset could be several seconds after manually blocking the excitation beam but would not significantly affect the relative line shape of the spectra. The EPR spectra were measured using a Bruker EMXplus X-band spectrometer. Photo-stimulated EPR was performed by illuminating (254 nm) the sample through the optical window of microwave chamber. All measurements were carried out at room temperature.
3. Results and discussion 3.1. Photoluminescence spectra Upon the UV excitation at 260 nm, the updoped Zn2GeO4 sample presents a blue–white emission of the broad band extends from the near-UV to orange–red region (Fig. 1, dashed line), arising from the host crystal defects, which were identified by the EPR experiments as VO or Zni for the donor and VGE or VZn the acceptor in Kröger–Vink notations [8]. The photoluminescence of the Mn2 þ doped sample exhibits a green emission within the 3d5 configuration (4T1-6A1) of Mn2 þ accompanied by the host emission (Fig. 1, solid line). The peak wavelength of the blue–white emission from the host defects shows a little different from that previously reported [8,9], due to the condition dependence of the defect properties on the sample preparation. The intensities of the two emission spectra in Fig. 1 are not in the same scale. Under the same measurement condition, the integrated emission intensity of the undoped sample is calculated to be one order of magnitude lower than that of the doped sample. The low efficiency of the host luminescence can be ascribed to the strong phonon coupling of the transition with the host lattices [13], resulting in the occurrence of nonradiative process. Its broad emission band extended from 380 nm to 600 nm also indicates such a strong electron–phonon coupling in the radiative transition. In general the host luminescence is not as efficient as that of the properly doped activators. Nevertheless it provides an optical approach to examine the intrinsic properties of the host defects for comparison with the doped activators. The excitation spectrum of Zn2GeO4:Mn2 þ shows two bands with the peak wavelengths at 260 nm and 295 nm, respectively, while the undoped sample only presents a strong excitation band around 260 nm, as shown in Fig. 2. The bandgap of Zn2GeO4 was reported at values of 4.5 eV [5] to 4.68 eV [7] corresponding to
Fig. 1. Photoluminescence spectra of Mn2 þ doped (0.4%, solid line) and undoped (dashed line) Zn2GeO4, excited at 254 nm.
the light wavelength around 270 nm, where in Ref. [7] the photocurrent was detected upon 254 nm UV excitation but not 365 nm. The 260 nm excitation band is due to the interband absorption of the host lattice and the one at 295 nm was attributed to charge transfer transition [9]. Mn2 þ can be selectively excited at a longer wavelength, e.g., 330 nm, to avoid the interband absorption, yielding a single green band of Mn2 þ emission. 3.2. Phosphorescent properties After the UV excitation, both the Mn2 þ -doped and undoped samples show the long-lasting phosphorescence with the same emission band shapes and positions as the photoluminescence (peaked at 455 and 533 nm). There were several investigators who reported that the lasting time of Zn2GeO4:Mn2 þ varied from several tens of seconds to more than 1 h [11,12] since its luminescent property is strongly dependent on the preparation conditions including the undoped samples [8,9]. For the doped sample Zn2GeO4:0.004Mn2 þ the phosphorescent emission from Mn2 þ is also dominant over that from the host. To have a clear comparison of their change with time, the samples doped with different Mn concentrations have also been examined. As the Mn concentration increases, the blue–white host emission decreases dramatically. The host emission is hard to be detected at 0.02 mol ratio and almost all the emission shifts to the green band of Mn2 þ . Fig. 3 compares the phosphorescent spectra of the samples doped with 0.001 and 0.004 of Mn2 þ in mol ratio at different time after excitation. For the photoluminescence, the two excitation bands peaked at 295 nm and 260 nm have the same scale of efficiencies. But for phosphorescence, only the 260 nm interband absorption is efficient in producing the afterglow. For excitation at the longer wavelength region above 300 nm, the charge transfer transition corresponding to the excitation at longer wavelength region above 300 nm takes place with the electrons transferring among the localized ions. These electrons do not ionize the trapping process and therefore do not yield phosphorescent emission, referring to the photocurrent behavior under 254 and 365 nm excitations [7]. For the sample with low doping concentration Zn2GeO4:0. 001Mn2 þ , phosphorescence from the host defects has the same scale as that of the Mn2 þ ions (Fig. 3a) but decreases significantly in the sample Zn2GeO4:0.004Mn2 þ doped with a higher concentration (Fig. 3b). Moreover, considering their large intensity difference of photoluminescence under the interband excitation as shown in Fig. 1, it is evident that the Mn2 þ ions are much more competitive in capturing the electrons/holes as the radiative recombination centers. Fig. 4 depicts the phosphorescent decay processes for the doped and undoped samples with their initial intensity normalized, lasting 3 h and 1 h, respectively.
Fig. 2. Excitation spectra of the Mn2 þ doped (0.4%, solid line) and undoped (dashed line) Zn2GeO4, monitored at 530 and 455 nm, respectively.
Please cite this article as: Z. He, et al., J. Lumin. (2015), http://dx.doi.org/10.1016/j.jlumin.2015.01.026i
Z. He et al. / Journal of Luminescence ∎ (∎∎∎∎) ∎∎∎–∎∎∎
Fig. 3. Phosphorescent spectra of (a) Zn2GeO4:0.001Mn2 þ and (b) Zn2GeO4:0.004Mn2 þ at 10, 30, and 100 s after switching off 254 nm excitation.
3
case for the germinate phosphors. Fig. 5 shows the time response of the photoluminescence from the host defects and Mn2 þ by shuttering the 254 nm excitation on and off. Before the measurement the samples are kept in dark for sufficiently long time to ensure the phosphorescent exhaustion and empty trapping states. When the shutter is open, the photoluminescence was found to jump to the maximum intensity in a fast response, even at very weak excitation (Fig. 5b). While for the host defect emission, the intensity keeps constant except for the noise fluctuation during the excitation. A surprising result is that the Mn2 þ luminescence intensity quickly reaches to a maximum then decreases to a stable level during the excitation, in contrary to a slow rising or charging process as observed in SrAl2O4:Eu,Dy [4]. By excitation at a longer wavelength ( 4330 nm) corresponding to the charge transfer transition, the luminescence is found to remain at the same level without such a decreasing process. Furthermore, the initial maximum intensity gradually decreases after each on–off cycle, as shown in Fig. 5(b). The initial peaks of the photoluminescence after the first cycle may reach the same intensity after sufficiently long phosphorescent release in the dark. Therefore, the photoluminescence decreasing process after the interband excitation (254 nm) can be associated with the trap charging mechanism or some recovering process, which will be discussed in next section. In the case of host excitation, the luminescent centers should capture the generated electrons and holes for light emission, viz. recombination via the energy levels of the radiative transition. If the traps exist in the host crystal, they can intercept either the electrons or holes (depending on the trap type) in advance of the luminescent centers until the traps are filled. Thus a gradually rising curve of the photoluminescence time dependence during the charge process is one of the most common behaviors of persistent phosphors. As we observed, the charge curve of SrAl2O4:Eu,Dy increases from nearly zero intensity at the excitation beginning. This can be understood
Fig. 4. Phosphorescent decay processes of the 0.004Mn-doped (monitored at 533 nm) and undoped (monitored at 455 nm) samples with their initial intensity normalized for comparison (after 20 min irradiation under 254 nm).
Fig. 3 also illustrates that the phosphorescence from the host defects presents a faster decay process than the Mn2 þ . This difference results from their different recombining rate and lifetime of the trapped carriers on the defects. It is presumed that the luminescent defects and Mn2 þ ions are the hole trapping centers and they share the same electrons trapped at other sites as identified as VO and Zni defects [8]. The hole trapping centers as electron acceptors VGE and VZn, as also identified in the same work, are adjacent to the electron traps. Therefore, the holes are filled by the electrons faster than those trapped at the Mn2 þ ions that are spatially separated far away from the electron traps. 3.3. Time dependence of photoluminescence If the excitation light is turned after exhaustion of the phosphorescence when all the electron/hole traps are empty, the time dependence of the photoluminescence reflects the optical charging process of the traps. Slowly rising photoluminescence has been observed for the persistent phosphor SrAl2O4:Eu,Dy and its charging behavior analyzed [4]. The charging curve is typical and reasonable for many phosphors capable of storing photon energy. By the same approach, it is expected to observe a similar rising curve of the time dependence in this work. However, this is not the
Fig. 5. Response of the photoluminescence from (a) the host defects monitored at 455 nm and (b) Mn2 þ ions of Zn2GeO4:0.004Mn2 þ at 533 nm by shuttering the 254 nm excitation on and off.
Please cite this article as: Z. He, et al., J. Lumin. (2015), http://dx.doi.org/10.1016/j.jlumin.2015.01.026i
4
Z. He et al. / Journal of Luminescence ∎ (∎∎∎∎) ∎∎∎–∎∎∎
considering that each of the luminescent centers has to get both the electron and hole spatially coincided at its site for the radiative recombination. For trapping, they can be independently captured by their corresponding type of the traps (electron or hole traps) in random distribution. Hence, trapping should be much more probable than the luminescent recombination unless the traps are filled. For explanation of the contradictory behavior of Zn2GeO4:Mn2 þ , it may be considered that the trapping capacity is too small to observe a rising charge curve by an intensive excitation. This consideration can be ruled out by a similar time response observed at very weak excitation. A reasonable explanation is that most electrons and holes are formed in mobile pairs (e–h pairs) after the interband excitation, which would not be separated and captured by the traps on their way to the luminescent recombining centers. Yet a small amount of independent electrons and holes thermally unpaired are captured by the traps and luminescent centers in spatial separation for the subsequent phosphorescence. The photo-generated e–h pairs in the persistent phosphors have also been reported [1,14]. Except for the electron or hole (single carrier) traps, e–h pair trap may be formed by a donor–acceptor defect complex in some host crystals [1]. In the Zn2GeO4 host, it is more likely that only the single type of carrier traps exists, allowing the e–h pairs to recombine directly on the luminescence centers without being trapped. 3.4. Trapping mechanism study by EPR measurement 2þ
In Fig. 5b, the decreasing process of the Mn photoluminescence may provide some guidance and hints for more detailed examination of the trapping mechanism. The initial maximum intensity becomes lower and lower when on–off cycle is repeated at delayed time. EPR experiment has been conducted to examine the change of Mn2 þ ions in the trapping state. For undoped sample, similar EPR results have been observed to that reported in the GeO2–B2O3– ZnO glasses [10] and they are very reversible with UV (254 nm) exposure on and off. For Mn2 þ -doped Zn2GeO4 samples, EPR signals are measured to examine the population change of Mn2 þ ions in the trapping process. Mn2 þ ions are located in tetrahedral sites in the host with high-spin states (S ¼5/2, Kramer’s system) [15]. The hyperfine structure (nuclear spin I¼ 5/2) of the EPR spectra measured at this high concentration is unable to be resolved, however, the broad EPR signal at g ¼2.001 can be ascribed to the introduction of Mn2 þ . Fig. 6 depicts the EPR spectra of the sample measured in dark (curve b) and under UV irradiation (curve a). Under the UV irradiation, a small but repeatedly steady decrease of the EPR signal is observed in comparison with the sample in dark. The decrease can be clearly seen at the peak position as zoomed in Fig. 6 (curves a and b). The difference of subtracting curve a from curve b is shown in curve c with 30 magnification. Curve d shows that difference becomes smaller at the second cycle after 100 s recovering in dark from first cycle. The decreases of Mn2 þ EPR signal and the photoluminescence in the trapping state after UV excitation suggest a population decrease of the Mn2 þ ions that may be ionized to Mn3 þ (non-Kramer’s system). Once the valence of Mn2 þ ions is changed in the current host lattice, they lose the role as the luminescent centers, leading to the decrease of photoluminescence that is consistent with the results in Fig. 5b. In addition, the ionization may just occur for the Mn2 þ ions in the surface, which has been observed for Mn2 þ in other hosts [16], depleting the limited population of Mn2 þ at surface even upon weak excitation. The initial maximum emission peaks gradually decrease after each on–off cycle since the recovering process of Mn2 þ from Mn3 þ is slow, as illustrated in curve d (Fig. 6), where just part of Mn3 þ ions returns to Mn2 þ after 100 s in dark. The Mn3 þ ions may also serve as hole traps (Mn3 þ ¼holeþMn2 þ ), yielding phosphorescence. The recovering time is in the same scale of phosphorescence in Mn2 þ
Fig. 6. EPR spectra and their zoomed peaks of Zn2GeO4:0.004Mn2 þ measured by irradiation of a 254 nm mercury lamp (a) and in the dark (b). Curves c and d represents the difference of 30 (b–a), corresponding to the first and second UV on–off cycles in Fig. 5, respectively.
doped samples. This also explains the results for undoped sample shown in Fig. 5a, where the initial peaks are absent or no ionization occurs when UV light is on. Finally, the EPR signal maintains the same if irradiating the doped sample with UV at longer wavelengths (4300 nm), which is consistent with the observation of photoluminescence, i.e., no initial peaks when UV is on for all the samples, as discussed in previous section.
4. Conclusions The phosphorescence from both the Mn2 þ ions and host defects in Zn2GeO4 was observed by the UV interband excitation. The Mn2 þ ions present a longer phosphorescent duration than the defects, due to the different recombination rates of their trapped holes with the detrapped electrons. In the phosphorescent charging process under UV excitation, the electron–hole pairs are formed and transport across the single carrier traps to the radiative recombining centers, leading to an immediate response to the maximum intensity for the Mn2 þ and host defect photoluminescence. Nevertheless some of the pairs are thermally separated as independent electrons and holes and then trapped for the subsequent phosphorescence. During the charging period the host defect emission maintains a constant intensity, while Mn2 þ emission shows a decay curve to the lower intensity due to the ionization of Mn2 þ to Mn3 þ . This is verified by EPR and UV stimulated EPR measurements. A slow recovering process of the ionization is also detected, which is consistent with the observation of phosphorescence from Mn2 þ doped samples.
Acknowledgments This work was supported by the National Natural Science Foundation of China (Grant no. 61077036), the Natural Science Foundation of Guangxi (Grant no. 2011GXNSFA018157), and the Research Foundation of Georgia Southern University.
Please cite this article as: Z. He, et al., J. Lumin. (2015), http://dx.doi.org/10.1016/j.jlumin.2015.01.026i
Z. He et al. / Journal of Luminescence ∎ (∎∎∎∎) ∎∎∎–∎∎∎
References [1] D. Gourier, A. Bessière, S.K. Sharma, L. Binet, B. Viana, N. Basavaraju, K.R. Priolkar, J. Phys. Chem. Solids 75 (2014) 826. [2] T. Maldiney, B. Viana, A. Bessière, D. Gourier, M. Bessodes, D. Scherman, C. Richard, Opt. Mater. 35 (2013) 1852. [3] N. Basavaraju, S. Sharma, A. Bessi‘ ere, B. Viana, D. Gourier, K.R. Priolkar, J. Phys. D: Appl. Phys. 46 (2013) 375401. [4] Z.Y. He, X.J. Wang, W.M. Yen, J. Lumin. 119-120 (2006) 309. [5] L.Y. Lu, J.J. Chen, W.Y. Wang, Appl. Phys. Lett. 103 (12013) 123902. [6] R. Yi, J.K. Feng, D.P. Lv, M.L. Gordin, S. Chen, D. Choi, D.H. Wang, Nano Energy 2 (2013) 498. [7] C.Y. Yan, N.D. Singh, P.S. Lee, Appl. Phys. Lett. 96 (2010) 053108.
5
[8] Z.S. Liu, X.P. Jing, L.X. Wang, J. Electrochem. 154 (2007) H500. [9] M.M. Shang, G.G. Li, D.M. Yang, X.J. Kang, C. Peng, Z.Y. Cheng, J. Lin, Dalton Trans. 40 (2011) 9379. [10] O. Arellano-Tanori, R. Melendrez, M. Pedroza- Montero, B. Castaneda, V. Chernov, W.M. Yen, M. Barboza-Flores, J. Lumin. 128 (2008) 173. [11] Y. Takahashi, M. Ando, R. Ihara, T. Fujiwara, Opt. Mater. Express 1 (2011) 372. [12] M.H. Wan, Y.H. Wang, X.H. Wang, H. Zhao, Z.F. Hu, Opt. Mater. 36 (2014) 650. [13] C.W. Struck, W.H. Fonger, J. Lumin. 10 (1975) 1. [14] J.B. Qiu, H. Igarashi, A. Makishima, Sci. Technol. Adv. Mater. 6 (2005) 431. [15] D.W. Feldman, J. Chem. Phys. 25 (1973) 363. [16] Z.V. Saponjic, N.M. Dimitrijevic, O.G. Poluektov, L.X. Chen, E. Wasinger, U. Welp, D.M. Tiede, X.B. Zuo, T. Rajh, J. Phys. Chem. 110 (2006) 25441.
Please cite this article as: Z. He, et al., J. Lumin. (2015), http://dx.doi.org/10.1016/j.jlumin.2015.01.026i