Recent Trends in Elpasolite Single Crystal Scintillators for Radiation Detection

Abstract

Scintillation detection has attracted great interest in nuclear medicine, nuclear radiation detection, high-energy physics, and non-destructive inspection. The elpasolite crystals with Ce³⁺ dopants are promising for these endeavors due to their modest light yield and extremely good proportionality when excited by the gamma ray. Moreover, the ⁶Li and ³⁵Cl isotopes in elpasolite crystals endow them with excellent neutron detection capability. These features allow not only a high energy resolution but also a high detection sensitivity. The elpasolite scintillators also enable the precisely dual detection of gamma/neutron signals through pulse height discrimination (PHD) or pulse shape discrimination (PSD). In this work, we review recent investigations on using the typical elpasolite scintillators, including Ce³⁺-doped Cs₂LiYCl₆ (CLYC), Cs₂LiLaCl₆ (CLLC), and Cs₂LiLaBr₆ (CLLB), for the monitoring of gamma rays and neutrons. The scintillation properties, detection mechanism, and elpasolite crystal structure are also discussed with the aim of improving high-energy ray detection ability.

1. Introduction

Scintillation detection plays an important role in a wide range of applications, such as nuclear medicine, nuclear radiation detection, high-energy physics, and border control, due to its ability to precisely monitor multiple rays [1,2]. A scintillation detector is composed of two parts: scintillation crystal, and photoelectric converter for converting the (ultraviolet) UV–visible lights into electric signals. In these detectors, the scintillation crystals act as conversion media, which can convert high-energy radiations into detectable low-energy photons. The dielectric or semiconductor wide bandgap materials are able to accomplish this purpose by absorbing high-energy radiation energy and converting into UV or visible photons through three stages: conversion, transport, and luminescence emission, shown in Figure 1. In the conversion stage, hot electrons and holes are produced through multistep interaction between scintillator lattice and ionizing radiation. Then, the produced electrons and holes are gradually thermalized and transferred to the bottom of the conduction band and the top of the valence band, respectively. For the transport process, electrons and holes migrates from the conduction valence bands of host materials to the emission centers. Finally, the recombination of electrons and holes emits detectable photons in the luminescence emission stage. In addition, the core-to-valence luminescence (CVL) could also be detected in some elpasolite scintillators. The electrons in the core band are excited to the conduction band by ionizing radiation. The holes generated in the core band immediately recombine with electrons in the valence and emit short-wavelength (220 nm), fast emission (~1 ns) fluorescence, favoring their use in gamma/neutron discrimination.

The solid scintillators, including organic and inorganic scintillation materials, have successfully been manufactured into scintillation detectors [3]. Among them, the inorganic scintillators have become the most widely used materials nowadays due to the outstanding scintillation characteristics [4]. The first inorganic scintillator Ba[Pt(CN)₄] powder, emitting visible photons under X-ray irradiation, was discovered by Edison [5]. In the late 1940s, the alkali halides single crystals, such as NaI:Tl [6] and CsI:Tl [7], were developed as the better scintillators. Indeed, the NaI:Tl is still one of the most widely used inorganic scintillators due to its modest scintillation properties, relatively low cost, and availability of large-size single crystals [8]. The NaI:Tl crystal is often employed as the benchmark to estimate the performance of newly explored scintillation materials. Subsequently, the Lus₂SiO₅ crystal was also exploited as standard material for positron emission tomography (PET) equipment (one of the most typical applications of scintillators) [9]. The LaBr₃ crystal was also demonstrated to be the one of the best inorganic materials for the use of scintillation detectors [10]. However, these scintillators described above are severely limited by the inferior energy resolution for NaI:Tl and internal radioactive contaminants for LaBr₃ [11].

The elpasolites materials as a new generation of halide scintillators have achieved high light output and excellent proportionality, and more outstanding scintillators are exploited from the elpasolite crystal family. Benefiting from the high proportionality, the excellent energy resolution of 3.6% at 662 keV was achieved in the Cs₂LiYCl₆ crystal [12]. Cs₂LiLaBr₆ has been found to be the brightest elpasolite scintillator, and it achieves 60,000 photons/MeV light yield and 2.9% energy resolution at 662 KeV [13]. A great number of experiments and theoretical studies on existing elpasolite crystals have been implemented with the aim of developing unprecedented scintillators for high-efficiency detection of gamma rays [14,15,16,17]. The neutron detection capabilities of elpasolite crystal scintillators were also further exploited [18,19,20,21]. The Li-containing elpasolites can provide extraordinary capability to monitor thermal neutrons by the large cross section of ⁶Li to thermal neutrons [22,23]. Elpasolite scintillators with special compositions, such as Cs₂LiYCl₆ and Cs₂LiLaBr₆ crystals, exhibit ideally simultaneous detection ability for gamma rays and neutrons through pulse shape discrimination [24]. These above investigations highlight the significant value of halide elpasolite scintillators for high-energy ray discrimination.

In this review, we summarize recent works on the exploration of various elpasolite crystals for sensing gamma rays and neutrons. This mainly includes the typical Cs₂LiYCl₆ (CLYC), Cs₂LiLaCl₆ (CLLC), and Cs₂LiLaBr₆ (CLLB) crystals and the other elpasolite crystals. We also briefly summarize the scintillation properties, detection mechanism, and structure of these elpasolite materials.

2. Scintillator Properties

The light yield (LY) and energy resolution (ER) are considered as the most important scintillation characteristics of scintillation materials. LY refers to the total number of emitting photons after the scintillator, absorbing radiation energy within a certain detection time. In a certain band gap range, relatively small band gap enables the higher light output. The total light yield also depends on the energy of excitation source and this dependence relation is called nonproportionality. The luminescence quenching, a nonlinear relationship with excitation density, is considered to be the root cause of nonproportionality. For practical detection applications, it is generally required that the emission light of the scintillator is located at the visible region, favoring the photomultiplier tube or silicon photodiode on the detector to monitor these photons. The energy resolution of scintillators is related to the LY and nonproportionality to a great extent [25,26,27,28]. Indeed, the scintillators with sufficiently high energy resolution have this ability to exactly distinguish the radiation species from the close interfere signals. The physical and scintillation properties of various scintillation crystals are also given in

Table 1. The physical and scintillation characteristics of various scintillator crystals. M.P.: melting point; E.R.: energy resolution; HY: hygroscopicity.

Chemical Component	Density (g/cm3)	M. P. (K)	Gamma-ray LY (ph/MeV)	E. R. At 662 keV (%)	Decay Times (ns)	Neutron Detection	HY	Price
NaI (Tl)	3.67	924	40,000	6.5	230	Yes	Yes	Low
CsI (Tl)	4.51	894	18,000	4.3	1250	No	Slight	Low
SrI2 (Eu)	4.55	675	80,000	3	1.2103	No	Yes	Med
LaCl3 (Ce)	3.9	1148	50,000	3.1	24	No	Yes	High
LaBr3 (Ce)	5.63	1056	70,000	2.4	16	No	Yes	High
LSO (Ce)	7.41	2320	30,000	9.0	40	No	No	High
YSO (Ce)	4.45	2200	18,500	9.3	42	No	No	High
BGO	7.13	1323	3600	20	300	No	No	Low
Cs2LiYCl6	3.31	913	20,000	3.6	3, 50	Yes	Yes	High
NaI (Tl)	3.67	924	40,000	6.5	230	Yes	Yes	Low

.

3. Structure of Elpasolite Scintillators

The absolute multitude of investigations on elpasolite materials have been focused on crystals with formula A₂BMX₆. Figure 2 shows the typical double-perovskite structure diagram of an elpasolite crystal. Three cations (A, B, and M) and one anion (X) constitute the A₂BMX₆ elpasolite crystal, which can be regarded as the linking of octahedra BX₆ and MX₆. Most of them possess cubic or near-cubic structure. The most prominent feature of perovskite-related architecture is the structure distortions accompanied by phase transitions [29]. Flerov, et al. demonstrated the lattice distortion in the phase transformation process of elpasolite crystal [30]. The Goldschmidt’s tolerance factor (t) has been used to determine stability range of elpasolite structures at room temperature. The cubic phase is stable in the range of 0.8 < t < 1 (for varied cations and anions, t value is slightly different). The smaller or larger t value leads to a distorted variant of the cubic structure. When the t value is close to the boundary of stability phase, a phase transition from P4/nbm tetragonal to cubic is also observed in the Rb₂NaTmCl₆ crystal. The grown crystals with large size may crack during cooling after growth, which has been the main obstacle for growing the large LaBr₃ crystals with low cost. Understanding phase transformation from crystallization temperature to room temperature is useful for avoiding crystal cracking during post-growth cooling and fabricating devices.

The distortion metric predicting the stable structure of crystals at room temperature is defined by analyzing embedded-atom method (EIM) database and molecular statics study of crystal stability in Ref. [31]. A number of A₂BMX₆ crystals have been studied and their corresponding distortion metric versus rank number sorted by this metric, shown in Figure 3, which successfully screens out a number of cubic elpasolite compositions. This favors guiding the growth of large crystals with scintillation performance. The anions also influence crystal structure. The elpasolite chlorides with smaller anion radius have stable cubic structure at room temperature. In contrast, the elpasolite iodides with larger anion exhibit non-cubic structure, which hampers the production of high-quality scintillators.

4. Cs₂LiYCl₆ (CLYC)

The CLYC:Ce³⁺ is one of the most typical scintillators in the elpasolite crystal family, and it was developed as an effective thermal neutron detector in 1999 by Combes, et al. [32]. Subsequently, William M. Higgins, et al. further studied the CLYC:Ce³⁺ crystals and finally fabricated the commercial dual-detection scintillators [33]. The initially reported energy resolution of CLYC:Ce³⁺ crystals is 7.3% at 662 keV of ¹³⁷Cs radiation, as shown in Figure 4. Benefiting from the improvement in vertical Bridgman technology and purification method, the energy resolution of one inch CLYC:Ce³⁺ crystal has been elevated to 3.6% at 662 keV in 2012 [12]. It is worth noting that the energy resolution value of CLYC:Ce³⁺ outperforms many commercial NaI:Tl products, which stems from exceptionally good energy proportionality. In addition, the shaping time, as another important parameter controlled by photomultipliers (PMT), can also influence the measured energy resolution of CLYC crystals. For instance, the energy resolution measured with 4 us shaping time was significantly superior to that obtained with 0.5 us shaping time. The virtually absent direct capture of electrons and holes by Ce³⁺ activators results in the slow decay time of CLYC scintillation crystals, which may explain the shaping time dependence of energy resolution.

The emission spectra of CLYC scintillation crystals were also extensively studied. A broad emission band ranging from 240 to 460 nm was observed in the X-ray-excited emission spectrum of pure CLYC, which is attributed to the self-trapped exciton (STE) emission shown in Figure 5 [32]. The spontaneous formation of STE, the luminescent defect itself, plays a significant role in the luminescent process of elpasolite crystals. As a result, the decay time of elpasolite crystals is relatively long compared to other commercially available scintillators. The very fast decay component was also found in the emission spectrum, which can be ascribed to the intrinsic core–valence luminescence (CVL) [24]. CVL appears at 310 nm with a fast decay time of several ns. It has been demonstrated that the smaller energy gap between the core band and the top of valence band than band gap favors the intense CVL. The slow decay process of the CLYC scintillation crystal is found to be a significant drawback, leading to the limited applications [35]. To address this problem, the Ce³⁺ activators were introduced into the CLYC lattice. For a Ce³⁺-doped CLYC system, the emission spectrum obtained from X-ray excitation shows the typical doublets emission peak, which is caused by transition from the lowest 5d level to the ²F_5/2 and ²F_7/2 levels of Ce³⁺ activators, shown in Figure 5. The Bessiere group reported that the regulation of the Ce³⁺ concentration can improve the scintillation performance, and the optimal doping concentration was determined to be less than 1% [36]. As shown in the literature, 0.1% Ce³⁺-doped CLYC exhibits a 700 photons/MeV LY with 8% energy resolution at 662 keV of ¹³⁷Cs radiation. The best energy resolution value of 7% was achieved in 0.02% Ce³⁺-doped CLYC. The most optimum doping concentration varies slightly with different crystal growth processes, and 0.5% Ce³⁺ concentration is considered as the standard doping level nowadays [37]. The 0.5% Ce³⁺-doped CLYC outputs the typical Ce emissions with 40 ns decay time. The slow decay component still accounts for most of the light yield due to the localized band structure of CLYC:Ce³⁺. Developing new elpasolite compositions with delocalized band structure may be a valid way to reduce the decay time.

Growing of CLYC:Ce³⁺ scintillation crystals has been attempted in the last ten years. In 2010, a 3 inch inclusion-free single crystal was successfully grown by using the vertical Bridgman technique at the Radiation Monitoring Devices (RMD) company for the first time [33]. Figure 6a presents the photo of 2 inch inclusion-free CLYC:Ce³⁺ crystals grown by the Saint-Gobain company. Figure 6b illustrates the photo of a 2 inch inclusion-free single crystal grown by RMD with energy resolution of 4.12%. However, the key growth technology was not provided by both of them. As we know, the complex components of a CLYC:Ce³⁺ crystal, as well as the incongruent melting characteristic, is not conducive for the growth of the single crystals. This leads to the generation of defects such as bubbles, inclusions, and cracks in single crystals. The CLYC:Ce³⁺ phase diagram based on Cs₂YCl₅ and LiCl was studied to decrease the defect and acquire the optimized melting characteristic for a single crystal [38]. The 1 × 1 × 0.5 cm CLYC crystal was fabricated into a portable scintillation detector with a solid-state photomultiplier (1 cm × 1 cm) in ref. [39]. The neutron counts of fabricated CLYC:Ce³⁺ detector were compared with the 3-He tubes under ²⁴¹Am/Be neutron source (4 MeV of average neutron energy). The result indicates that when the distance between neutron source and detector is greater than 6 mm, the CLYC:Ce³⁺ detector could outperform traditional 3-He tubes in count rates and sensitivity. In addition, the fabricated detector also shows the excellent detection ability for gamma signals. However, since the elpasolite scintillators are hygroscopic, the packaging technique was employed to protect the crystal and avoid the degradation of light output. The performance of CLYC:Ce³⁺ detectors can be also influenced by packing processes. In the future, the improved packaging techniques are expected to assist the CLYC:Ce³⁺ detector for radiation detection.

5. Cs₂LiLaCl₆ (CLLC)

CLLC:Ce³⁺ single crystals were also exploited as scintillation material for monitoring gamma rays. The luminescent emitting properties of the CLLC:Ce³⁺ crystals were measured by vacuum ultraviolet (VUV) synchrotron radiation [41]. The luminescence spectra contain two emission bands: core–valence luminescence (CVL) emitting band peaking at 4.4 eV with decay time of 1.4 ns and Ce³⁺ emitting band peaking at 3.3 eV with decay time of 36 ns, shown in Figure 7. The first CLLC:Ce³⁺ crystal based on Bridgman growth technique was reported by Glodo, et al. and its size reached 1 inch in diameter [42]. However, the as-grown crystal was opaque because it had a great number of inclusions. Compared with CLYC:Ce³⁺, the main advantage of CLLC:Ce³⁺ possesses a higher light yield. Their light yields reach 35,000 photons/MeV and 110,000 photons/neutron under irradiation from ¹³⁷Cs (662 keV) and ²⁴¹Am/Be (thermalized neutrons) sources, respectively. A 3.4% energy resolution at 662 KeV was achieved by gamma ray excitation, which is superior to that of CLYC:Ce³⁺. The 36 ns decay time of typical Ce³⁺ luminescence (372 nm and 400 nm) in 1% Ce³⁺-doped CLLC was obtained, which also outperforms the CLYC:Ce³⁺ crystals. Despite the excellent scintillation performance, there is no report on the successful growth of a CLLC single crystal. This is attributed to the Cs₃LaCl₆ s phase existing in the melt during crystal growth arising from the incongruently melting characteristic of the CLLC crystal. The inclusions were thus formed in the grown crystal, resulting in opacity of the grown crystal. In order to solve this problem, Hebing Zhu designed and employed the rich LiCl melt with high axial temperature gradient to produce the CLLC:Ce³⁺ single crystal [43]. The rich LiCl melts favor the suppression of Cs₃LaCl₆ s phase. In addition, the high axial temperature gradient enables the sufficient impurity transport, benefiting the production of transparent CLLC crystals. The 12 mm crystal wafer cut from the transparent section of the as-grown crystal exhibits better optical transmittance than that grown by Glodo, et al. However, the low energy resolution of 7.1% implies a poor crystal quality.

6. Cs₂LiLaBr₆ (CLLB)

CLLB:Ce³⁺, as another elpasolite crystal, is exploited for the purpose of dual detection of gamma ray and neutron. The Ouspenski group demonstrated that the light yield of CLLB:Ce³⁺ crystal reaches 60,000 photons/MeV, which is brighter than CLYC:Ce³⁺ and CLLC:Ce³⁺ [44]. As a result, an excellent energy resolution of 2.9% at 662 keV is achieved. Moreover, they found that the light yield increases and the decay time decreases with increasing temperature, which can be ascribed to the more effective energy transfer from the CLLB to the Ce³⁺ activators. Shirwadkar, et al. studied the CLLB doped with varied Ce³⁺ concentration under ¹³⁷Cs excitation [45]. The low light yield was observed when the Ce³⁺ concentration was less than 0.2%. However, the light yield indicated a significant increase as the Ce³⁺ doping concentration exceeded 0.5%. This phenomenon is distinctly different from that of the CLYC crystal. The energy resolution of the CLYC crystal reduced significantly at high-concentration Ce³⁺-doping. X-ray induced emission spectra of CLLB are presented in Figure 8. The pure CLLB has a broad response, whereas Ce³⁺-doped CLLB exhibits typically dual bands peaking at 390 and 420 nm. With increasing Ce³⁺ concentration, the intensity ratio of 420 nm/390 nm emissions shows an increased trend. This may be attributed to the absorption of Ce³⁺ activator to 390 nm light and re-emission of 420 nm photons. Under the excitation of gamma ray, the Ce³⁺-doped CLLB showed a relatively long decay time of 55 ns. This can be attributed to the improved efficiency of direct capture of energy by Ce³⁺ activators. With regard to La-containing scintillators, such as LaBr₃:Ce³⁺ and CLLB:Ce³⁺, the alpha contamination degrading the energy resolution is a thorny problem. The relationship between the internal alpha contamination and the energy resolution was studied and presented in ref. [46].

Similar to the growth of the CLLC crystal, the growth of the CLLB crystal mainly faces two major obstacles: (I) the raw materials are extremely sensitive to air and moisture, requiring vacuum seal in a quartz crucible during growth; (II) the incongruent melting property always lead to bubbles and cracks in as-grown crystals, degrading the scintillation performance. Lin, et al. demonstrated the CLLB:Ce³⁺ crystal can be grown by using the Bridgeman growth technique [47]. The transparent 1 inch crystal with energy resolution of 4.09% at 662 keV has been successfully grown by Saint-Gobain Crystals, shown in Figure 9. The CLLB:Ce³⁺ scintillator can be fabricated into portable gamma/neutron dual detectors using silicon photomultipliers by Saint-Gobain Crystals [48]. The parameters, including Ce³⁺ concentration, shape, size, silicon photomultiplier (SiPM) model, and SiPM placement on detector, are systematically researched. The energy resolution of detecting gamma ray reaches 6.0% at 662 keV and achieves a PSD figure of merit (PSD FOM) of 1.9 under²⁴¹Am/Be source (4 MeV of average neutron energy).

7. The Elpasolite Scintillators for Neutron Detection

Highly efficient neutron detection is conducive to prevent proliferation of nuclear material. In recent years, dwindling supplies of ³He and growing emphasis on global nuclear threat have triggered great demands for developing appropriate neutron detection scintillators to accomplish the mission [49]. Scintillators for neutron detection can be divided into solid, liquid, and gas according to the physical state of the material, and solid scintillators mainly include plastic, glass, and crystal scintillators. In this paper, we focus on the elpasolite crystal scintillators. A great number of isotopes were exploited for monitoring neutron [50]. Figure 10 presents the neutron capture cross sections of lithium, boron, gadolinium, and hydrogen as a function of neutron energy. Among them, gadolinium isotope (yellow line) shows the highest cross section area at low energy region. However, the high-energy gamma rays were produced when the Gd isotope of high effective atomic number (Zeff) interacted with neutrons, escaping from the material and resulting in low detection efficiency. Boron isotope also exhibited high cross section value for low-energy neutron detection. The low light yield, however, severely limits its use for neutron detection. The hydrogen isotope has a high cross section in the high-energy region, favoring the detection of fast neutrons.

It is worth noting that the lithium isotope with small effective atomic number (Zeff) shows large neutron capture cross section in the high-energy region, which benefits the improvement in resolution and produces considerable light yield with respect to thermal neutron. A great number of ⁶Li-based neutron detection scintillators (⁶Li-glass, ⁶LiI:Eu, ⁶LiF/ZnS:Ag etc.) have been exploited [51,52]. However, the ⁶Li-glass shows a low light yield of 6000 ph/neutron due to severe scintillation quenching. Despite that the ⁶LiI:Eu exhibits high light yield, it cannot discriminate gamma and neutron signal reliably. The ⁶LiF/ZnS:Ag has been developed as the most widely used neutron detection scintillator with the advantages of high light yield (160,000 ph/neutron), low sensitivity to gamma rays, and good gamma/neutron discrimination ability. However, the preparation of ⁶LiF/ZnS:Ag scintillator achieved by mixing ZnS:Ag and ⁶LiF leads to the opacity due to their sharply different refractive indices, which severely limits the thickness of scintillator layer (≤0.4 mm) and results in low neutron detection efficiency. To address this problem, efforts have been devoted to the elpasolite scintillators containing Li for differentiating neutron from gamma signals. For instance, CLYC:Ce³⁺, CLLC:Ce³⁺, and CLLB:Ce³⁺ crystals containing ⁶Li achieved, respectively, 2.9, 2.6 and 1.8% energy resolution [34]. Among them, the best neutron detector was demonstrated to be the CLYC:Ce³⁺ crystal, which provided light yield as high as 70,000 photons/neutron and better gamma ray discrimination ability with maximum FOM value of 3.33 under ²⁴¹Am/Be source (4 MeV of average neutron energy). The light yield value of 70,000 ph/neutron is higher than that of ⁶Li-glass (6000 ph/neutron), ⁶LiI:Eu (50,000 ph/neutron), and LiBaF₃:Ce (3600 ph/neutron). It is only smaller than that of ⁶LiF/ZnS:Ag (160,000 ph/neutron). Notably, the relatively high price and strong hygroscopicity are the main obstacles for their widespread application.

It is of note that almost all neutron facilities have gamma ray background. Therefore, the reliable discrimination of neutron from gamma signals requires the gamma rejection ratio (GRR) larger than 10⁻⁶ [53]. As shown in Figure 11a, the full energy peaks of gamma equivalent energy of neutron in CLYC:Ce³⁺, CLLC:Ce³⁺, and CLLB:Ce³⁺ appear at 3.5 MeV, 3.1 MeV, and 3.2 MeV, respectively. Under ¹³⁷Cs gamma source and ²⁴¹Am/Be source irradiation, such high gamma equivalent energy can identify neutron peak from gamma ray peaks at 662 KeV by pulse height discrimination (PHD). In addition, pulse shape discrimination (PSD) has been also applied as a reliable method for neutron detection [54]. The various characteristics of pulse shapes between gamma rays and neutron signals are of importance for the PSD technique. As an important parameter, the PSD ratio is defined by the following formula:

$$\mathrm{PSD}\mathrm{Ratio}=\frac{\mathrm{A}\left[{\mathrm{w}}_2\right]}{\mathrm{A}\left[\mathrm{w}\right]}$$

(1)

where A[w₂] is the integral of the signals over head region of the trace; A[w] is the integral of the signals over the full region of the trace. Another parameter for effectively separating two peaks can be acquired from the following formula:

$$\mathrm{FOM}=\frac{{\mathit{D}}_{\mathit{n}}-{\mathit{D}}_{\mathit{g}}}{{\mathit{F}}_{\mathit{n}}{+\mathit{F}}_{\mathit{g}}}$$

(2)

where F_n is the full-width half-maximum (FWHM) of neutron peaks, F_g is the FWHM of gamma ray peaks, and D_n and D_g are the Gaussian mean of neutron and gamma peaks, respectively. FOM is figure of merit. When the FOM value is larger than one, the gamma rays and neutron signals can be clearly distinguished. The head region in Formula (1) ranging from 0 to 450 ns often contributes to great FOM value, and head region should be optimized for different materials due to various signal pulse characteristics [54]. Figure 11b presents the PSD scatter plot of CLLB:Ce³⁺, in which the obvious difference of pulse shape is observed in the tail region. The FOM value reaches 1.9 under²⁴¹Am/Be source (4 MeV of average neutron energy); therefore, the gamma rays and neutron signals can be well distinguished. Figure 11c,d give the PSD scatter plots of CLYC:Ce³⁺ and CLLC:Ce³⁺. As can be seen, their gamma rays and neutron signals can be well separated and detected, which can be attributed to the relatively large FOM values of 3.33 and 1.48 under²⁴¹Am/Be source (4 MeV of average neutron energy). It is notable that the larger FOM value from the CLYC:Ce³⁺ leads to the better ability for dual detection of gamma rays and neutron. Concerning the CLYC:Ce³⁺, the very fast core–valence luminescence (CVL) seems to provide marked variations in the scintillation temporal response. However, for crystals with large volume or high Ce³⁺ concentration, signal pulse spectrum does not exhibit CVL component due to the self-absorption effect. Thus, the CVL can only be exploited in small-volume CLYC with low Ce³⁺ concentration for the PSD method [33].

8. Other Elpasolites

Great efforts have been also devoted to the exploration of other elpasolite crystals with various scintillation characteristics, listed in

Table 2. Scintillation characteristics of various elpasolite crystals. FOM value was measured under²⁴¹Am/Be source (4 MeV of average neutron energy). Ce Con: Ce concentration; E. R.: energy resolution. For comparison, the NaI:Tl crystal is also included [8,29,57,58,59,60,61,62,63,64,65,66,67,68].

Chemical Component	Ce Con. (mol %)	Gamma Ray LY (ph/MeV)	E. R. At 662 keV (%)	Decay Times (ns)	Thermal Neutron Detection	Fast Neutron Detection	FOM	References
Cs2LiYCl6	0.5	20,000	3.6	3, 50, 1.0103	▲	▲	3.33	[57]
Cs2LiLaCl6	1	35,000	3.4	85, 450, 1.3103	▲	▲	1.48	[58]
Cs2NaCeCl6	0	20,000	8.3	91, 3.2103		▲		[59]
K2LiCeCl6	0	21,000	16	70, 360, 2.0103	▲	▲	0.81	[60]
Tl2LiYCl6	1.5	27,000	4.0	37, 341, 947	▲	▲		[29]
Tl2LiGdCl6	10	58,000	4.6	34, 191, 1.2103	▲	▲		[61]
Cs2LiLaBr6	2	60,000	2.9	122, 661	▲		1.9	[62]
Cs2LiYBr6	1	24,000	4.1	50, 270	▲		1.23	[63]
Cs2LiGdBr6	15	37,800	6.2	62, 224, 1.0103	▲			[64]
Cs2LiLuBr6	1	25,000	19	37, 337, >103	▲			[65]
Cs2NaCeBr6	0	25,000	6.7	140, 880				[66]
Rb2LiCeBr6	0	33,000	6.3	55, 284	▲			[67]
Cs2NaLaI6	5	46,000	5.0	45, 170				[68]
NaI	4 (Tl)	40,000	6.5	230	▲			[8]

. The Ce³⁺-doping concentration and the most important characteristics, including g-ray light yield (LY), energy resolution, decay time, and FOM, are provided. For comparing the ray detection performance of these elpasolite crystals, the commonly-used NaI(Tl) is also included in Table 2. A number of Li-containing elpasolites were developed for the specific purpose of thermal neutron detection by exploiting the ⁶Li isotope with large cross section area. Among the elpasolite chloride crystals, Cs₂LiYCl₆ with 0.5% Ce³⁺ shows a larger FOM value of 3.33 under²⁴¹Am/Be source (4 MeV of average neutron energy), favoring the accurate discrimination of neutron and gamma radiations. Cs₂LiLaBr₆ with a light yield of 60,000 photons/MeV enables the most efficient gamma ray detections. The elpasolite crystals with ³⁵Cl can implement the monitoring of fast neutrons. Due to the detection ability of ⁶Li for neutrons, the elpasolite chloride crystals are endowed with the simultaneous-sensing performance of thermal and fast neutrons. The high-atomic-number Tl-based elpasolite scintillators are developed for the purpose of increasing photo-fraction of detected gamma events, such as Tl₂LiYCl₆ and Tl₂LiGdCl₆, exhibiting higher light output than Cs₂LiYCl₆. It is assumed that the elpasolite scintillators with less electronegative anions show higher light output and relatively slower decay time. This is because their less electronegative anions contribute to smaller band gap and more delocalized valence band edge states. The decay time decreases with the sequence of Cl → Br → I and the light yield of elpasolite bromides is generally higher than that of chlorides counterpart. Moreover, the narrow band gap also favors the elevation of light yield and enhancement of the stabilization of STEs by inhibiting STEs quenching at room temperature [56].

9. Conclusions

This review summarizes recent progress on the scintillation properties, detection mechanism, crystal structure, and high-energy detection applications of elpasolite scintillators. Novel detection strategies are always based on the development of materials science with new and novel properties. The typical elpasolite scintillators, including Cs₂LiYCl₆ (CLYC), Cs₂LiLaCl₆ (CLLC), and Cs₂LiLaBr₆ (CLLB) crystals, and the other elpasolite crystals indeed provide great opportunities for gamma ray and neutron detection. The optimal Ce³⁺ concentration is determined to be 0.5% in Cs₂LiYCl₆ and 2% in Cs₂LiLaBr₆, which benefits the elevation of energy resolution and light yield. Obviously, challenges still remain; for instance, the development of new elpasolite materials with less localized band structure are required for overcoming the slow decay time of existing elpasolite scintillators. Despite that the CLLC crystals exhibits better energy resolution and pulse shape discrimination properties, the poor crystal qualities must be improved greatly. In addition, the highly efficient elpasolite crystal growth technology for eliminating inclusions and elevating optical quality needs to be explored.