Research Advances on Human-Eye-Sensitive Long Persistent Luminescence Materials

Based on the actual application requirements of multicolor long persistent luminescence (LPL) materials, we highlight the recent developments in the last decade on human-eye-sensitive LPL materials and try to make a full list of known LPL compounds possessing wavelengths of 400–600 nm and a duration time longer than 10 h (>0.32 mcd/m2); these are more sensitive to the human eye's night vision and can be used throughout the night. We further emphasize our group research of novel LPL materials and the regulation of LPL color to enable a full palette. In the end, we try to summarize the challenges and perspectives of LPL materials for potential research directions based on our limited understandings. This review could offer new enlightenment for further exploration of new LPL materials in the visible light range and related applications.


INTRODUCTION
Long persistent luminescence (LPL) materials are defined as energy-saving and environmentally friendly materials that show persistent luminescence from seconds to hours after turning off the excitation sources, such as UV, visible light, X-ray, and sunlight (Zhuang et al., 2016;Xu et al., 2018;. Up to now, the most efficient visible light LPL materials remain blue CaAl 2 O 4 : Eu 2+ , Nd 3+ and green SrAl 2 O 4 : Eu 2+ , Dy 3+ (Katsumata et al., 1997(Katsumata et al., , 1998Hölsä et al., 2001), which both struggle to achieve multicolor and extensive applications. On this basis, it is the right time to provide this review to further stimulate and develop novel visible-light-emitting LPL materials and open up new applications.
As we all know, visible light can be observed and distinguished by human eyes, but the sensitivity of human eyes toward different colors is quite different. In addition, the sensitivity of the human eye also varies according to the ambient brightness (Hölsä, 2009;Poelman and Smet, 2010;Brito et al., 2012). Human eye sensitivity functions have a peak of 683 lm/W at 555 nm in photopic vision and 1,700 lm/W at 507 nm in scotopic vision (Poelman et al., 2009). That is, in scotopic vision (LPL material application conditions), human eyes are more sensitive to ∼400-600 nm light. Therefore, the development of materials in this wavelength range is very important for practical applications, especially in night vision. At the same time, compared with reviews concerning LPL materials classified by emission wavelength (UV, Red, and NIR), matrices, and activators, common features, including the history, mechanism, and application of LPL materials, have been reported on repeatedly and will not be elaborated on here (Eeckhout et al., 2010(Eeckhout et al., , 2013Smet et al., 2014;Zhuang et al., 2014;Li et al., 2016;Xu and Tanabe, 2019). In this review, we mainly focused on the following three aspects: (1) the recent development of novel LPL materials with wavelengths of 400-600 nm; (2) the approach of regulating the LPL color to achieve multicolor and discoloration on the basis of our own work; (3) the challenges and perspectives of LPL material for potential research directions. This work may be useful for the exploration of human-eye-sensitive LPL materials and the further optimization of LPL performance.

ADVANCES IN LPL MATERIALS SENSITIVE TO HUMAN EYES
As far as we know, reviews concerning the development of LPL materials with wavelengths of 400-600 nm, which play important roles in practical application, have rarely been made. In this regard, we try to make a full list of known LPL materials, possessing 400-600 nm wavelengths and a decay time longer than   10 h (>0.32 mcd/m²), which can be used throughout the whole night, as shown in Table 1    . Excited by 337 nm, the yellow emission, with a peak value of 584 nm, was detected. After excitation at 254 nm and 365 nm for 15 min, the LPL decay time exceeded 24 h. Bi 3+ ions not only act as luminescence centers but also as trap centers to effectively improve electron transfer rates. In 2018, Li et al. reported that the emission range of CaZnGe 2 O 6 : Bi 3+ is 330-500 nm, the peak value is 468 nm, and the LPL decay time can last for more than 12 h after 254 nm light excitation for 30 min (Dou et al., 2018). Furthermore, except for Bi 3+ ions, Mn 2+ ions also have been recently reported to act as activators in some LPL phosphors (Lei et al., 2006;Che et al., 2008;Takahashi et al., 2011;Xu et al., 2013;Zou et al., 2015;Zhou et al., 2018). Unfortunately, the overall LPL performance is inferior. Among them, ZnSiO 4 : Mn 2+ , Yb 3+ , reported by Zhang et al., has a glorious green LPL performance, peaking at 523 nm (Zou et al., 2015). After 15 min of exposure to 254 nm light, the LPL duration time can last ∼30 h (0.32 mcd/m 2 ). Although the past decade or so has witnessed significant advances in LPL materials with wavelengths in the range of 400-600 nm, the pattern of Eu 2+ ions acted as activator ions and lanthanide rare-earth ions serving as trapping centers or to produce defect-related trapping center is still the most effective for most excellent LPL materials. On these grounds, we mainly introduce the research of our group on ultra LPL materials doped with Eu 2+ ions in the next chapter.

THE APPROACH TO ACHIEVE MULTICOLOR LPL BASED ON OUR OWN WORK Exploiting Novel LPL Materials With Rare Colors
We know that yellow is often used to warn of danger or draw attention, such as yellow lights on traffic signs, large construction machinery, raincoats, etc. If the material itself continues to emit yellow light in dark conditions, this will further improve security and reliability. In addition, to date, only scattered yellow LPL materials have been reported, and the main reason is that the crystal field strength should be strong enough to reduce the 5d energy level of Eu 2+ and thus produce yellow luminescence (Lakshminarasimhan and Varadaraju, 2008;Sun et al., 2008;Li et al., 2013Li et al., , 2014. Here, we mainly present two excellent yellow materials: Ca 2 BO 3 Cl: Eu 2+ , Dy 3+ and Ca 6 BaP 4 O 17 : Eu 2+ , Ho 3+ . These have been exploited by our research group (Zeng et al., 2013;Guo et al., 2015). First of all, in the Ca 2 BO 3 Cl borate matrix, the photo-luminescence (PL) and LPL spectra of Ca 2 BO 3 Cl: Eu 2+ , Dy 3+ have the same profile and both present an asymmetric wide emission band at 580 nm, indicating that Eu 2+ ions occupying both Ca 2+ lattice sites can act as a luminescent center in PL and LPL processes (Figure 1). For the representative sample of Ca 2 BO 3 Cl: 0.002Eu 2+ , 0.002Dy 3+ , after 10 min excitation of 254 nm and 365 nm lights, the initial LPL intensity can achieve 0.01 cd/m 2 and its LPL can sustain more than 12 h above 0.32 mcd/m 2 . As the excitation time is further extended from 10 min to 5 h, the thermoluminescence (TL) intensity, LPL initial brightness, and decay time are significantly improved and prolonged. When the excitation time is up to 5 h, the charging procedure approaches saturation, and the LPL decay time reaches 48 h. This large energy storage capacity also provide benefits of optical storage and other potential applications as well (Zhuang et al., 2018).
Furthermore, as shown in Figure 2, in the Ca 6 BaP 4 O 17 phosphate matrix, both PL and LPL spectra of Ca 6 BaP 4 O 17 : 0.02Eu 2+ , 0.015Ho 3+ only have an asymmetric broad emission band peaking at 553 nm, resulting in a bright yellow PL and LPL. Incorporation of Ho 3+ ions can largely extend the TL characteristics and evidently elevate the LPL performance of Ca 6 BaP 4 O 17 : Eu 2+ , Ho 3+ . For Ca 6 BaP 4 O 17 : 0.02Eu 2+ , we just observe three very weak TL peaks, which indicate that the electron concentration trapped at the intrinsic defects is very low. Whereas, codoping Ho 3+ ions makes the TL peak at 335 K greatly enhanced and largely improves the defect levels. After 15 min excitation of 254 nm and 365 nm lights, for Ca 6 BaP 4 O 17 : 0.02Eu 2+ , 0.015Ho 3+ , the initial LPL brightness can reach about 0.13 cd/m 2 and LPL can last more than 47 h above 0.32 mcd/m 2 .
Looking from the other side, in the dark environment (10 −6 ∼10 −2 cd/m 2 ), the maximum visual sensitivity of the human eye is 507 nm. That is to say, the human eye is most sensitive to cyan light in the dark. Because of this, our research team was striving to exploit cyan LPL materials. We first identified Ba 2 Zr 2 Si 3 O 12 : Eu 2+ , Nd 3+ cyan LPL material . Under 340 nm excitation, Ba 2 Zr 2 Si 3 O 12 : 0.01Eu 2+ , 0.01Nd 3+ possess a broad asymmetric cyan PL band at ∼501 nm. The LPL initial intensity is 155.5 mcd/m 2 , and the LPL duration time is near 25 h before decaying to 0.32 mcd/m 2 . On this basis, we tried to adjust the bandgap of Ba 2 Zr 2 Si 3 O 12 by replacing Zr 4+ with Hf 4+ ions due to the conduction band (CB) bottom of Ba 2 Zr 2 Si 3 O 12 is mainly composed of Zr 4d electron states, and thus the energy gap of the excited state of Eu 2+ ions and trap center relative to CB can also be regulated. As a result, with the increase of Hf 4+ concentration, the bottom of the CB moves upward, making the bandgap vary from 4.288 to 4.301 eV and the TL peak shift monotonically from 51 to 91 • C, corresponding to the trap depth moves from 0.66 to 0.76 eV, as presented in Figure 3. For the optimal sample (x = 0.5), the initial brightness of LPL can reach 131.3 mcd/m 2 , and the LPL decay time can last for 31 h.
Moreover, a series of (Ba, Li) (Si, Ge, P) 2 O 5 : Eu 2+ , Pr 3+ cyan LPL materials were designed by adopting a solid solution strategy . After 10 min of excitation at 254 nm, BaSi 2 O 5 : 0.008Eu 2+ , 0.01Pr 3+ presents a strong cyan LPL located at 514 nm and the LPL duration time reaches about 38 h, which is ascribed to the increased TL intensity at 350 K by codoping Pr 3+ ions and thus generating a number of traps suitable for LPL at room temperature (Figure 4). Furthermore, by means of solid solution, the LPL decay time of (Ba, Li) (Si, Ge, P) 2 O 5 : Eu 2+ , Pr 3+ can be prolonged from 38 to 56 h.
For the abovelisted yellow and cyan LPL materials, the LPL properties have been greatly improved. The only disadvantage, same with the current common problem of LPL materials, is that they cannot be effectively excited by sunlight. This situation limits the wide application utilizing solar energy outdoors. However, emergencies, energy conservation, and multi-color requirements can still be realized through combining with UV chip and short-term charging.

Realizing Energy Transfer
On the basis of the above work, we can easily find that it is difficult to realize the diversified LPL colors in the same material. In particular, for Eu 2+ ions, the LPL color in most matrices is blue-green. In a matrix, when the emission spectrum of one luminescent center overlapped with the excitation spectrum of another luminescent center, it was possible to realize nonradiative energy transfer between two luminescent centers, such as Eu 2+ → Mn 2+ , Ce 3+ → Mn 2+ , and Ce 3+ → Tb 3+ (Jia et al., 2002;Wang et al., 2003;Xu et al., 2009aXu et al., , 2011Dai, 2014). Multicolor LPL could be achieved by the energy transfer between two luminescent centers based on the existence of a large number of trap centers. As shown in Figure 5, in SrAl 2 O 4 : Ce 3+ blue LPL material (Xu et al., 2011), with the increase of Mn 2+ ions concentration, the emission of Mn 2+ ions located at 515 nm increases, and the emission band of Ce 3+ ions at 374 nm, decreases, proving that there is an effective energy transfer from Ce 3+ to Mn 2+ ions. We improved the energy transfer efficiency by adjusting the concentration of Mn 2+ ions and realized the adjustable PL color from blue light to green light. When the excitation source is removed, the energy transfer from Ce 3+ to Mn 2+ ions continues. However, due to the energy loss in the energy transfer process and the different decay rates of the two ions, it is difficult to ensure the uniformity of LPL color.
Among them, the inevitable energy loss in the LPL energy transfer process is also one of the main reasons for the initial LPL brightness reduction.
In addition, in CaAl 2 O 4 :Eu 2+ , Nd 3+ blue LPL material, we realized the energy transfer between Eu 2+ and Mn 2+ ions through codoping different Mn 2+ concentrations, and achieved the regulation of PL color from blue to green, as shown in Figure 6 (Xu et al., 2009a). Unfortunately, it can be seen from the LPL spectrum that the energy transfer efficiency of Eu 2+ and Mn 2+ ions is lower in the LPL decay process so that the LPL color is still blue. Even worse, the incorporation of Mn 2+ ions leads to poor LPL performances. There are two main reasons. First, the incorporation of Mn 2+ ions significantly reduced the concentration of the effective trap center, which can be observed from the TL spectra. Second, in the LPL decay process, the energy transfer efficiency between Eu 2+ and Mn 2+ ions is so low leading to inevitable energy loss. Therefore, the LPL color regulation can not be easily realized through energy transfer in the same substrate, and the LPL performance can also deteriorate. In order to overcome the abovementioned problems of LPL color regulation through energy transfer in the same matrix, our group for the first time adopts a method of separating the trap center and luminescence center in two different matrices, utilizing the superior energy storage capacity of CaAl 2 O 4 : Eu 2+ , Nd 3+ (CA), the high quantum efficiency of Y 3 Al 5 O 12 : Ce 3+ (YAG), and the great overlap of the LPL spectrum of CA and the PLE spectrum of YAG, to construct CA/YAG composites, as shown in Figure 7 (Chen et al., 2016). The radiation energy transfer from CA to YAG achieves the LPL color adjustable characteristics from blue to white, prolongs the LPL duration time of CA/YAG composite with B:Y = 10:10-48 h compared with 19 h of CA, and heightens the initial LPL brightness to 3,200 mcd/m 2 , which is 2.7 times higher than pure CA. The main reason for the dramatic improvement of LPL performances of CA/YAG composite is that the human visual efficiency of yellow light is much higher than that of blue light although the energy transfer efficiency reduces the overall radiation. The white LPL performances achieved by this method overcame the problem through composite tricolor LPL materials, which required similar decay characteristics to ensure color uniformity in the LPL decay process.

Adjusting the Occupancy Rate of the Luminescent Center in Multi-Sites
In addition to the emergency lighting mentioned above, LPL materials have gradually shown more excellent performance in anti-counterfeiting and other aspects (Jiang et al., 2016;Sandhyarani et al., 2017). At present, the relatively mature fluorescent anti-counterfeiting is to realize visible light output and achieve anti-counterfeiting recognition using ultraviolet excitation. Compared with fluorescent anti-counterfeiting, the LPL anti-counterfeiting is still sustainable after turning off the excitation light source. On this basis, if the color of LPL after turning off the light source is obviously different from that of PL excited by the light source, it will greatly improve the level and force of anti-counterfeiting and provide a more effective guarantee for the safety and reliability of commodities, bills, etc. Here, we achieved this purpose mainly by regulating the Eu 2+ ions occupation rate of multiple sites in the matrix. Figure 8A presents the total and partial density of states of BaSc 2 Si 3 O 10 calculated according to the density function theory . It can be seen that the valence band (VB) is derived from 2p electronic state of O and the CB is mainly composed of 3d electronic state of Sc. Figure 8B displays that the PL spectrum of BaSc 2 Si 3 O 10 : 0.01Eu 2+ exhibits a wide asymmetric blue emission band peaking at 443 nm, which can be fitted well into two Gaussian peaks at 443 nm and 508 nm, ascribed to the 5d-4f transitions of Eu 2+ ion occupying Ba and Sc sites, respectively. As shown in Figures 8C,D, the luminescence decay curves of BaSc 2 Si 3 O 10 : 0.01Eu 2+ and BaSc 2 Si 3 O 10 : 0.01Eu 2+ , 0.01Nd 3+ prove that the green LPL at 508 nm is due to Eu 2+ emission, not some defect level. Based on the above analysis of the electronic structure and luminescence performances, it is believed that only Eu 2+ ions occupying the Sc sites participate in the LPL process, while Eu 2+ ions occupying the Ba sites are mainly involved in PL, producing an interesting phenomenon of blue PL and green LPL. Analogously, in CaAl 2 O 4 : Eu 2+ , Nd 3+ LPL material, the green emission peak at 550 nm was enhanced while the blue emission peak at 440 nm was weakened by the partial replacement of Ca 2+ by Sr 2+ ions, as shown in Figure 9 (Xu et al., 2009b). When the doping concentration of Sr 2+ ions is 0.4 mol, the two emission peaks make the PL color appear as white light. When the excitation light source is removed, the LPL color is yellowish-green on account of the different decay characteristics of Eu 2+ ions occupying different lattice sites.

SUMMARY AND PROSPECT
In this review, we provided the development of LPL materials with wavelengths in the 400-600 nm range, mainly introduced our group's research on developing ultra LPL materials based on Eu 2+ ions, achieved a multicolor LPL phenomenon through energy transfer, and realized the distinct color of PL and LPL performances through adjusting the Eu 2+ ions occupancy rate to improve the anti-counterfeiting level. However, there are still many areas for further research and development, including but not limited to the following: (a) unearthing novel host lattices and activator ions; (b) controlling excitation and emission wavelengths; (c) structuring a unified test standard; and (d) opening up novel multifunctional application fields.

AUTHOR CONTRIBUTIONS
YW was responsible for the framework design, thinking arrangement, and text calibration of the paper. HG was responsible for the paper writing and the investigation of research progress at home and abroad. Both authors contributed to the article and approved the submitted version.