1. Introduction

Nowadays the gradually scarce fossil fuels and the expected climate change force people to develop the high efficiency and environment friendly energy source, such as the solar cell, which can convert the energy of sunlight into electric energy [1]. The most widely used solar cells are based on crystalline silicon (c-Si). Si solar cells most effectively convert near infrared (NIR) photons of energy close to the semiconductor band gap (Eg ≈1.12 eV, λ ≈1000 nm). However, the incident solar spectrum is dominant in the UV-Vis region. The mismatch between the incident solar spectrum and the spectral response of Si solar cells leads to charge thermalization in the process of photovoltaic (PV) conversion. The charge thermalization is one of the major energy loss mechanism [2]. One of the methods for reducing these energy losses is to modify the solar spectrum, for example downshift, or downconversion.

Recently, many efforts have been made to develop downshift or downconverting materials, for example, RE³⁺-Yb³⁺ (RE = Tb, Pr, Tm, and Ce) co-doped powders or glasses [3–6]. In these systems, RE³⁺ ion acts as the sensitizer to absorb UV-Vis (300-500 nm) photons and transfers part of the absorbed energy to Yb³⁺ ion. Yb³⁺ ion is used as NIR emitting center, which gives the emission at about 1000 nm due to the ²F_5/2→²F_7/2 transition and perfectly matches the maximum spectral response (λ ≈1000 nm) of Si solar cells. Unfortunately, there is one significant drawback for the RE³⁺-Yb³⁺ couples mentioned above. For RE³⁺-Yb³⁺ (RE = Tb, Pr, Tm) co-doped systems, the RE³⁺ ion has only weak absorption peaks, because the f-f absorption transitions are forbidden by the parity selection rule and its optical oscillator strength is small. These features of the RE³⁺ (RE = Tb, Pr, Tm) ions mean that they cannot be efficiently excited by UV-Vis photons. For Ce³⁺-Yb³⁺ co-doped systems, compared with RE³⁺ (RE = Tb, Pr, Tm), Ce³⁺ usually has a strong excitation band, due to the allowed 4f-5d transitions. Even so, Ce³⁺ ion has a tendency to become oxidized and simultaneously Yb³⁺ has a tendency to be reduced. So the luminescence quenching of this system from the metal-to-metal charge transfer of Ce⁴⁺–Yb²⁺ is more likely to happen [7, 8]. This may greatly reduce the real quantum efficiency. All the drawbacks mentioned above have greatly limited the potential of RE³⁺-Yb³⁺ system for application in the Si solar cells.

The CTS is generally associated to an electron transfer from the ligand’s orbital to an orbital of the metal [9]. It is allowed by the spin and parity selection rules and consequently has strong and broadband absorption. The position of the CTS is determined mainly by the optical electronegativity of the ligand and the metal ion. It can generally shift towards longer wavelength with increasing covalency of the host lattice. These features make us deeply lost in thought whether CTS of Yb³⁺ ion in proper host would be an important approach to modify the solar spectrum for Si solar cell? Nowadays, CTS of Yb³⁺ ion in many compounds were experimentally reported [10, 11]. For fluorides and oxides, the position of the CTS of Yb³⁺ usually lies in high energy scales of the solar spectrum. It is obvious that it cannot effectively harvest UV-Vis (300-500 nm) photons. In order to make the position of the CTS of Yb³⁺ shifting towards longer wavelength, the host lattice should have relatively high covalency. Compared to fluorides and oxides, the CTS of Yb³⁺ in some oxysulfides having higher covalency is at relatively low energies. Therefore, it is strongly expected that the CTS of Yb³⁺ in the oxysulfides could more effectively absorb UV-Vis (300-500 nm) photons. In addition, the relatively low phonon frequency of oxysulfides may dramatically reduce the probabilities of nonradiative transition and increase the fluorescence emission efficiencies. These prompt us to explore a novel NIR emitting oxysulfide phosphor doped with Yb³⁺ suitable for solar spectral convertor.

In this paper, we report a novel NIR phosphor, CaLaGa₃S₆O:Yb³⁺. It has an efficient broad absorption band dominating at around 345 nm and extending from 250 to 500 nm, ascribed to the CTS of Yb³⁺-S^2-, and exhibits an intense NIR emission of Yb³⁺ between 920 and 1150 nm, perfectly matching the maximum spectral response of Si solar cells. The NIR emission intensity of CaLaGa₃S₆O:Yb³⁺ is 12 times as intense as that of a NIR quantum cutting phosphor Ca₂BO₃Cl:Ce³⁺, Tb³⁺, Yb³⁺ (CBC) upon 4f-5d excitation of Ce³⁺ [12]. These results demonstrate that the allowed CTS of Yb³⁺-S^2- with high absorption cross-section can be an efficient and direct sensitizer harvesting UV-blue photons and greatly enhancing the NIR emission of Yb³⁺ ion.

2. Experimental section

All samples CaLa_1-xGa₃S₆O:xYb³⁺ (CLGSO:xYb³⁺), (x = 0, 0.001, 0.002, 0.004, 0.006, 0.008, 0.010) were prepared by a two-step solid-state reaction [13]. The first step was to prepare the starting sulfide materials. The starting sulfide materials β-La₂S₃, Ga₂S₃ and ε-Yb₂S₃ were prepared from La₂O₃(99.99%), Ga₂O₃(99.99%) and Yb₂O₃(99.9%) under CS₂ reducing atmosphere at 800 °C, 950 °C and 1250 °C for 3 h in horizontal tube furnaces, respectively. The second step was as follows: The stoichiometric amounts of materials CaO(A.R.), β-La₂S₃, Ga₂S₃ and ε-Yb₂S₃ were thoroughly mixed in agate mortar with a pestle, and then the mixture was transferred to a small corundum crucible and sintered at 950 °C for 2 h in Ar atmosphere in horizontal tube furnaces. Ca₂BO₃Cl:Ce³⁺, Tb³⁺, Yb³⁺ (CBC) was prepared according to Ref 12.

Powder X-ray diffraction was performed on Bruker D8 advance X-ray diffractometer (XRD) with CuKa (λ=1.5405 Å) radiation at 40 kV and 40 mA. High quality XRD data for Rietveld refinement was collected over a 2θ range from 10° to 100° at an interval of 0.02°. Structural refinement of XRD data was performed using the TOPAS-Academic [14] program.

The photoluminescence excitation (PLE) and emission (PL) spectra were measured by FSP920-combined Time Resolved and Steady State Fluorescence Spectrometers (Edinburgh Instruments) equipped with a 450 W Xe lamp, TM300 excitation monochromator and double TM300 emission monochromators, Red sensitive PMT and R5509-72 NIR-PMT in a liquid nitrogen cooled housing (Hamamatsu Photonics K.K). The spectral resolution for the steady measurements is about 0.05 nm in UV-VIS and about 0.075-0.01 nm in NIR.

3. Results and discussion

3.1 Powder X-ray diffraction

The XRD measurement was carried out at room temperature. Figure 1 shows the Rietveld refinement of the XRD pattern of the CLGSO:0.01Yb³⁺ using the $P\overline{4}{2}_{1}m$ structure model reported by C. L. Teske [15], which converged to R_wp = 4.73% and R_B = 2.43%. All of the observed peaks are consistent with the lattice constants and the re〉ection condition, indicating the formation of a single phase with no impurities. Table 1 shows the final refined structural parameters for the CLGSO:0.01Yb³⁺. The refined cell parameters for CLGSO:0.01Yb³⁺, a = b = 9.2880(1) Å, c = 6.0407(1) Å, are smaller than those for the parent material synthesized in our study (a = b = 9.2949(6) Å, c = 6.0413(5) Å), which is consistent with the substitution of smaller Yb³⁺(98.5ppm) for La³⁺(116 ppm).

 

Fig. 1 The experimental (crosses) and calculated (red solid line) XRD patterns and their difference (blue solid line) for CLGS):0.01Yb³⁺.

Download Full Size | PDF

Table 1. Final Refined Structural Parameters for CLGSO:0.01Yb^{3+ a}

View Table

3.2 Luminescence of CLGSO:Yb³⁺

Figure 2 presents the PLE and PL of CLGSO:xYb³⁺ (x = 0, 0.001, 0.002, 0.004, 0.006, 0.008, 0.010). The PLE spectrum, monitored at both 978 nm and 1020 nm, presents a similar broad band centering at around 345 nm except for the difference in intensity (shown in Fig. 2 a). In general, the energy level structure of Yb³⁺ is very simple because of its simple 4f¹³ electronic structure. The ²F spectral term is split by the spin-orbit interaction in two energy manifolds, ²F_7/2 and ²F_5/2. The ²F_5/2 level, the only excited state of the 4f levels of Yb³⁺, is situated near 1000 nm. The next nearest excited level is the CTS of Yb³⁺ and the 4f¹²5d¹ configuration of Yb³⁺ lying in higher energy (around 70000 cm⁻¹, i.e., ~140 nm) [16]. In addition, for CLGSO, the host absorption edge is around 250 nm [13]. Therefore it is reasonable to assign the prominent broadband at 345 nm to the CTS of Yb³⁺ rather than 4f→5d transition of Yb³⁺. In CLGSO, there is only one cation site available for Yb³⁺, i.e., La³⁺. La³⁺ ion are surrounded octahedrally by seven S^2- and one O^2- [15]. Consequently, the prominent broadband at 345 nm is probably ascribed to the CTS of Yb³⁺-S^2- and/or Yb³⁺-O^2-, involving an electron transfer from the surrounding 3p⁶ (2p⁶) orbital of S^2- (or O^2-) to the 4f¹³ orbital of Yb³⁺. In general, the position of the CTS of Yb³⁺-O^2- usually lies in the range of 194-262 nm [10]. So the broad absorption band dominating at around 345 nm mainly ascribes to the CTS of Yb³⁺-S^2-. Compared with the reported RE³⁺-Yb³⁺ (RE = Tb, Pr, Tm) co-doped systems, the allowed CTS of Yb³⁺-S^2- has stronger absorption in the UV-Vis region. Whereas, the RE³⁺ ions (RE = Tb, Pr, Tm) have only weak absorption peaks because the 4f–4f absorption transitions are forbidden by the parity selection rule. Therefore, rather than the 4f–4f absorption transitions, the allowed CTS of Yb³⁺-S^2- can effectively absorb UV-Vis (300-500 nm) photons of solar spectrum. In addition, the phosphor CLGSO:Yb³⁺ does not have the codoped Ce³⁺ ion. The possibility of the luminescence quenching from the metal-to-metal charge transfer of Ce⁴⁺–Yb²⁺ has also been excluded. Under 345 nm excitation, CLGSO:xYb³⁺ shows intense NIR emissions centering at 978 nm and 1020 nm, ascribed to the ²F_5/2→²F_7/2 transitions of Yb³⁺(shown in Fig. 2 c). Figure 2 d exhibits the concentration dependence of the integrated emission intensity of Yb³⁺. With the increase of Yb³⁺ concentration, it is obviously seen that the PL intensity of NIR emission from Yb³⁺ increase first, then reaches a maximum value x = 0.006, and finally shows a conspicuous decrease, due to the concentration quenching of Yb³⁺ ions. Compared with RE³⁺-Yb³⁺ (RE = Tb, Pr, Tm, and Ce) co-doped systems, the phosphor CLGSO:Yb³⁺ has higher emission efficiencies because of stronger absorption of the CTS of Yb³⁺-S^2-. At the same time, the relatively low phonon frequency of oxysulfides may dramatically reduce the probabilities of nonradiative transition and further increase the fluorescence emission efficiencies.

 

Fig. 2 The PLE (a) of CLGSO:0.006Yb³⁺ (λ_em = 978 nm and 1020 nm); the PLE (b) and PL (c) of CLGSO:xYb³⁺ (x = 0, 0.001, 0.002, 0.004, 0.006, 0.008, 0.010); the concentration dependence (d) of the integrated emission intensity of Yb³⁺..

Download Full Size | PDF

As an ideal NIR material, it should downconvert the UV-to-blue (300-500 nm) part of the solar spectrum to ~1000 nm photons, perfectly matching the maximum spectral response of Si solar cells. As discussed above, CLGSO:0.006Yb³⁺ shows promising NIR emission performance. In order to explore the possibility of it as solar spectral convertor, Fig. 3 presents the solar spectrum, the spectral response of c-Si solar cells and spectra of CLGSO:0.006Yb³⁺ and Ca₂BO₃Cl:Ce³⁺_0.002, Tb³⁺_0.02, Yb³⁺_0.01(CBC). Ca₂BO₃Cl:Ce³⁺, Tb³⁺, Yb³⁺, recently demonstrated by us, is a known NIR quantum cutting phosphor promising as a solar spectral convertor. we demonstrated that Ce³⁺ ion was an efficient sensitizer harvesting UV photons and greatly enhancing the NIR emission of Yb³⁺ ion through efficient energy feeding by allowed 4f-5d absorption of Ce³⁺ ion with high oscillator strength. Even so, CBC does not have prominent absorption beyond 315 nm, as shown in Fig. 3(a). In other words, it is not perfect one to effectively harvest UV-Vis (300-500 nm) photons of the solar spectrum. Comparatively, CLGSO:Yb³⁺ has a broader absorption band extending from 250 to 500 nm, of which the maximum absorption wavelength is relatively situated at longer wavelength, i.e., 345 nm (shown in Fig. 3(a)). Under 345 nm excitation, the NIR emission intensity of CLGSO:0.006Yb³⁺ is 12 times that of Ca₂BO₃Cl:Ce³⁺_0.002, Tb³⁺_0.02, Yb³⁺_0.01 (λ_ex = 369 nm) upon 4f-5d excitation of Ce³⁺. These results demonstrate that CLGSO:Yb³⁺ can effectively harvest UV-blue photons of the incident solar spectrum and convert to ~1000 nm photons, perfectly matching the maximum spectral response of Si solar cells. It is a promising luminescent material used as solar spectral convertor.

 

Fig. 3 Solar spectrum, spectral response of C-Si, spectrum of CLGSO:0.006Yb³⁺ and Ca₂BO₃Cl:Ce³⁺_0.002, Tb³⁺_0.02, Yb³⁺_0.01(CBC).

Download Full Size | PDF

4. Conclusions

To summarize, an intense NIR phosphor, CaLaGa₃S₆O:Yb³⁺, was investigated. It has an efficient broad absorption band around the 345 nm, ascribing to the CTS of Yb³⁺-S^2-, and shows a broad NIR emission band, perfectly matching the maximum spectral response of Si solar cells. The NIR emission intensity of CaLaGa₃S₆O:Yb³⁺ is 12 times as intense as that of Ca₂BO₃Cl:Ce³⁺, Tb³⁺, Yb³⁺ (CBC) upon 4f-5d excitation of Ce³⁺. We believe the phosphor CaLaGa₃S₆O:Yb³⁺ could be a promising luminescent material used as solar spectral convertor.