Synergy between structural rigidity and cluster defects in a bright near-infrared Cr³⁺-based phosphor for excellent thermal stability and long afterglow

Abstract

Near-infrared (NIR) phosphor-converted light-emitting diodes (pc-LEDs) driven by blue LEDs have emerged as promising future NIR light sources with high energy efficiency and compact structure for applications in medical diagnosis, night vision, and agriculture. However, development of thermally robust NIR light sources presents a great challenge. Herein, we report a blue light-pumped Y₃Al₂Ga₃O₁₂:Cr³⁺ phosphor with thermally stable broadband NIR emission (105% at 423 K) and a high quantum efficiency of 76%, covering the range from 630 to 950 nm. Moreover, the Y₃Al₂Ga₃O₁₂:Cr³⁺ phosphor displayed long persistent luminescence (PersL) under UV charging. Synchrotron based extended X-ray absorption fine structure (EXAFS) indicated the octahedral Cr³⁺ lattice sites and local structure. Low temperature-dependent EXAFS studies revealed rigid octahedral coordination of Cr³⁺ ions at Al/GaO₆ sites with minimal thermally activated distortion even in the second/third coordination shell. The lattice stiffness resulted in superior temperature-dependent NIR photoluminescence properties (80 to 433 K) with abnormal anti-thermal quenching desirable for pc-LEDs at 150 °C. Furthermore, the intrinsic defects were probed using positron annihilation lifetime spectroscopy and density functional theory, which suggested feasible formation of oxygen vacancies (V_O) and cluster defects (V_Al/Ga–V_O). Introduction of abundant traps with wide-ranging trap depths (∼0.7 to 1 eV) facilitated the long PersL with UV charging and defect activated thermal stability. Finally, the fabricated NIR pc-LED device exhibits efficient electroluminescence and its potential applications in imaging and night vision were demonstrated. Then, promising long PersL of the fabricated thin film under X-ray charging is presented. This work investigates the application prospects and highlights the important structural and defect related dynamics of this promising phosphor for NIR pc-LED devices.

---

1. Introduction

In recent years, near-infrared (NIR) light emitting phosphors have attracted widespread research attention due to their applications in security, night vision, surveillance technology, biomedicine, environmental monitoring, optical communication, food analysis, agriculture, etc.^1–3 The current research focus is on designing efficient broadband NIR phosphor-converted light emitting diodes (pc-LEDs) as ideal alternatives for halogen lamps, laser diodes, incandescent bulbs, and IR LEDs. These conventional sources have irreparable shortcomings such as large size, high energy consumption, shorter lifetimes, low efficiency, and high cost.^1,2 The quest for developing NIR pc-LEDs is inspired by their incomparable superiority over traditional NIR light sources through a combination of compact size, longer life, lower energy consumption and broad emission.^2–5 Therefore, development of broadband NIR light emitting phosphors with broadband emission and high thermal stability, output power, efficient photoluminescence quantum yield (PLQY), and high photoelectric conversion is preferable for NIR pc-LED applications. Adding to this, long NIR persistent luminescence (PersL) enlarges the sphere of applications including safe in vivo bioimaging, anticounterfeiting, safety signals and emergency lighting, luminous paint, night-vision surveillance, etc.⁶

Currently, the Cr³⁺ ion is the preferred choice for an ideal NIR activator among other transition metals owing to its broadband NIR emission originating from 3d–3d electronic transitions and efficient absorption in the blue region.⁷ The emission of Cr³⁺ ions is decided by crystal field strength as d orbitals are sensitive to the local structure.⁸ Hence, host selection is the critical step for incorporating Cr³⁺ as an emission center to develop broadband NIR pc-LEDs. Cr³⁺ in octahedral sites is well known to be a far red/NIR emitting center.⁹ Cr³⁺ exhibits sharp emission (∼700 nm) due to spin-forbidden ²E → ⁴A_2g transition, and the broad emission is attributed to spin-allowed ⁴T_2g → ⁴A_2g transitions with the tunable emission range of 650–1600 nm.^10,11 Importantly, the absorption band due to ⁴A_2g → ⁴T_1g transition matches aptly with blue LED chips. Therefore, the host materials offering octahedral coordination for Cr³⁺ and weak crystal field have gained significant attention in recent years. Representative phosphors with shorter and longer emission wavelengths include YGa₃(BO₃)₄:Cr³⁺,¹² BaAl₄Sb₂O₁₂:Cr³⁺,¹³ Ga₂O₃:Cr³⁺,¹⁴ Na₂CaGe₆O₁₄:Cr³⁺,¹⁵ Ga₄GeO₈:Cr³⁺,¹⁶ CaLu₂Al₄SiO₁₂,¹⁷ Li₃Mg₂NbO₆:Cr³⁺,¹⁸ SrGa₂O₄:Cr³⁺,¹⁹etc. Over recent years, substantial progress has been made in developing Cr³⁺-based long PersL phosphors and investigating the mechanism of afterglow.^1,2,15,20 Huang and co-workers²¹ designed a Cr³⁺-doped Na_0.5Gd_0.5TiO₃ NIR-PersL phosphor and investigated charging mechanisms using theoretical calculations. Taktak et al.²² reported Zn₂SnO₄:Cr³⁺ spinel phosphors showing long NIR PersL. Ding's group¹⁵ developed Na₂CaGe₆O₁₄:Cr³⁺ phosphors, which displayed broadband NIR PersL for ∼10 h.

Garnet phosphors (A₃B₂C₃O₁₂) are the hotspot matrices for Cr³⁺ doping due to the presence of multiple cation sites ([A], [B], and [C] cations in dodecahedral, octahedral, and tetrahedral coordination sites) with tunable crystal field and rigid structure.^1,23–27 A large number of reports published so far on Cr³⁺-based garnet phosphors demonstrated NIR broadband emission with high efficiency and thermal stability, which are the basic prerequisites for designing advanced NIR light sources. For example, Wang et al.²⁸ reported Cr³⁺ doped Ca₃Sc₂Ge₃O₁₂ broadband NIR phosphors with high thermal stability (∼97% at 100 °C) and an internal quantum efficiency (IQE) of 61%. Jiang et al.²³ synthesized Gd₃ScMgGa₂SiO₁₂:Cr³⁺ phosphors with a full width at half maximum (FWHM) of 150 nm, an IQE of 90% and good thermal stability (∼85.3 at 150 °C). Li et al.²⁹ reported high IQE (74%) and near-zero thermal quenching (∼97% at 423 K) for a Y₃Sc₂Al₃O₁₂ garnet phosphor. Recently, Li et al.³⁰ reported a Y₃In₂Ga₃O₁₂:Cr³⁺ phosphor with an IQE of 91.6%, a FWHM of 125 nm and zero thermal quenching (100% at 150 °C). However, the robust NIR emission was proposed due to traps but no solid proof was provided to explain the defect induced thermal stability. Other previously reported Cr³⁺ phosphors for pc-LEDs include Ca₃Sc₂Ge₃O₁₂:Cr³⁺,Ce³⁺ with good thermal stability (IQE = 22%),²⁸ Gd_2.4Lu_0.6Ga₄AlO₁₂:Cr³⁺ with a high IQE of ∼90% and low thermal stability (75% at 423 K),⁵ Ca₂YHf₂Al₃O₁₂:Cr³⁺ (90% at 423 K and IQE ∼ 75%),²⁷etc. Similarly, Y₃Al₂Ga₃O₁₂ (YAGG) is a well-known host matrix demonstrating good PLQY, thermal stability, rigid structure and accommodation of transition metals at multiple sites.^6,31–34 Based on earlier reports, Y₃Al₂Ga₃O₁₂ is the optimal composition among the solid-solutions of the Y₃Al_5−xGa_xO₁₂:Cr³⁺ (x = 0 to 5) garnet, which displayed ideal luminescence properties with intermediate crystal field strength.^2,35

Although Cr³⁺-activated garnet phosphors have been widely explored for NIR pc-LEDs, the luminescence mechanism, thermal stability, origin of defects, site-occupation and structure–property relationship of garnets lack a clear understanding. In addition, thorough investigation of structural rigidity and defects resulting in thermal stability, long afterglow, and thermally induced Cr³⁺ local structural changes in the garnet matrix is lacking. Most of these reports investigated the thermal stability in the optimum pc-LED range, PLQY, crystal field variation and thermoluminescence properties in detail. Importantly, the thermal stability is often attributed to intrinsic defects and rigidity of the host matrix without much attention to their origin. Consequently, achieving good thermal stability with desired PLQY remains a problem for fast advancement of high-power NIR pc-LEDs. Additionally, the changes in the emission peak profile at elevated temperatures pose a strong hindrance. Generally, the mechanism of long PersL is explained on the basis of trap information derived from thermoluminescence studies. Despite a handful of reports, the understanding of trap types, feasibility of formation of traps and their position in between the bandgap is unclear in the majority of reported long PersL phosphors. Understanding the thermally induced structural distortions and distribution of Cr³⁺ ions and defects in garnet materials will be crucial for development of next generation NIR phosphors for lighting and bio-medical applications. Hence, in this work, we carried out an in-depth experimental and theoretical investigation of changes in the local structure, types of traps, defect formation energies, and temperature-dependent structural distortions, which will help to unfold the mechanism of thermal quenching.

In this study, we synthesized a series of Y₃Al₂Ga₃O₁₂:xCr³⁺ (x = 0.02, 0.04, 0.06, 0.10 and 0.30 mol) NIR garnet phosphors using a solid state route. The Cr³⁺ doped Y₃Al₂Ga₃O₁₂ phosphor demonstrated high thermal stability (∼105% at 423 K) and an IQE of ∼76%, which are important parameters for NIR pc-LEDs. Furthermore, low temperature dependent synchrotron-based extended X-ray absorption fine structure (EXAFS) investigation provided useful insights into rigid coordination of Cr³⁺ ions in the YAGG matrix, which is desirable for thermally robust luminescence. The temperature-dependent structural distortions were evaluated using temperature-dependent EXAFS studies. The low and high temperature-dependent photoluminescence (PL) studies showed increase in FWHM and overall integrated intensity without much change in peak positions, which revealed thermal population of ⁴T₂ states. Moreover, the explanation of thermal stability is discussed based on thermoluminescence (TL), density functional theory (DFT) calculations, and positron annihilation lifetime spectroscopy (PALS) studies. The association between Cr³⁺ local structure, crystal field, NIR PL properties, and mechanism of thermal stability and PersL is well established benefiting from advanced EXAFS, PALS, and DFT calculations. Furthermore, we fabricated a YAGG:Cr³⁺ phosphor based NIR LED device with efficient broadband electroluminescence (EL) in the range of 630–950 nm and investigated power-dependent EL spectra of pc-LEDs. Additionally, Y₃Al₂Ga₃O₁₂:Cr³⁺ phosphors/PMMA thin films were fabricated to demonstrate the moisture-stable PL and persistent luminescence. The results obtained in this study highlight key structural and luminescence properties, which will accelerate the current development pace of NIR pc-LEDs.

2. Results and discussion

2.1 Crystal structure, morphology and oxidation state

Fig. 1(a) shows the X-ray diffraction (XRD) patterns of YAGG:xCr³⁺ (x = 0.02, 0.04, 0.06, 0.10 and 0.30 mol) samples. The diffraction peaks observed in the XRD patterns of pure YAGG and Cr³⁺-doped YAGG samples match well with the standard data (ICSD code: 280107). This suggested the formation of pure phase upon doping Cr³⁺ ions into the YAGG crystal lattice without the formation of any impurity phase. The crystal structure of Y₃Al₂Ga₃O₁₂ and coordination of each cation site are presented in Fig. 1(b), where the Y³⁺ ions occupy dodecahedral sites and Al³⁺ and Ga³⁺ occupy the octahedral and tetrahedral sites, with a higher feasibility of Ga³⁺ ions occupying the tetrahedral sites. Laguta et al.³⁶ experimentally determined the distribution of Al³⁺ and Ga³⁺ ions in Y₃Al₂Ga₃O₁₂ and reported that 72% of the Ga³⁺ ions has tetrahedral coordination and 28% occupy octahedral sites. The Cr³⁺ ion can occupy the Al³⁺/Ga³⁺ octahedral sites as the ionic radii of Cr³⁺ ions (0.615 Å) are comparable with ionic radii of Al³⁺ (0.535 Å) and Ga³⁺ (0.62 Å) ions.³⁷

Fig. 1 (a) XRD patterns of YAGG:xCr³⁺ (x = 0.02, 0.04, 0.06, 0.10 and 0.30 mol) samples. (b) Crystal structure of Y₃Al₂Ga₃O₁₂, (c) FTIR spectra. (d)–(f) SEM, TEM, and HRTEM images of the YAGG:0.04Cr³⁺ phosphor, respectively.

Fig. 1(c) shows the FTIR spectra of Cr³⁺ doped YAGG samples and no significant changes in the vibrational frequencies were observed upon incorporation of Cr³⁺ ions. The vibrational bands observed between 570 and 830 cm⁻¹ can be assigned to Ga–O and Al–O stretching frequencies in the YAGG lattice.^38,39 The scanning electron microscopy (SEM) image of YAGG:0.04Cr³⁺ confirmed the formation of micron size YAGG particles with irregular morphology (Fig. 1(d)). The EDS spectra reveal the presence of Y, Al, Ga, Cr, and O elements and successful incorporation of Cr³⁺ ions (Fig. S1 in ESI†). Fig. 1(e) and (f) show the TEM and HRTEM images of the YAGG:0.04Cr³⁺ sample. The d-spacing of ∼ 0.279 nm observed in the HRTEM image is due to the (420) crystallographic plane of YAGG.

2.2 Oxidation state of Cr and local structure

X-ray absorption near edge structure (XANES) measurements have been carried out at three absorption sites namely Y, Ga and Cr K-edges. The normalised XANES spectra of Cr, Y, and Ga along with their respective standards are shown in Fig. 2(a) and Fig. S2a, b (ESI†). It is evident that both probe elements (Ga and Y) exhibit trivalent states in all the samples, as confirmed by the alignment of their absorption edge positions with the respective trivalent standards. The absorption edge position (at the half absorption jump) of Cr doped YAGG samples appears at a relatively higher energy side; however, their white line peak positions around 6011 eV coincide. Farges et al.⁴⁰ concluded that determining the chromium valence based on the position of the edge jump region was not straightforward, as the main jump energy could vary by up to 2.6 eV in five model compounds with trivalent chromium octahedrally coordinated to oxygen ligands.⁴⁰ Consequently, an alternative approach has been suggested, which involves analysing the pre-edge region, specifically around 5994 eV. The position and intensity of localized transitions observed at the K pre-edge are recognized indicators of the chromium oxidation state and the geometry of its surrounding environment.⁴⁰ In the case of trivalent Cr₂O₃ doublet pre-peaks are observed, particularly well-resolved when chromium is in a trivalent state. The inset of Fig. 2(a) shows that the position of the pre-edge in Cr doped YAGG samples coincides with the pre-edge peaks of Cr₂O₃, confirming the trivalent state in the samples. This doublet typically arises from quadrupolar 1s to 3d transitions resulting from the hybridization of Cr 3d levels in the presence of neighbouring oxygen ligands.^41–43

Fig. 2 (a) Normalised XANES spectra at the Cr K-edge along with that of Cr foil and Cr₂O₃ standards. (b) EXAFS oscillations in wave vector space at the Y K-edge, Cr K-edge and Ga K-edge, respectively. (c) Fourier transform EXAFS spectrum (scatter points) and the respective fittings (solid line) of Cr doped YAGG at the Cr K-edge. The spectra are vertically shifted for better visualisation. (d) Local structure of Cr³⁺ ions. (e) Octahedral coordination of Cr³⁺ ions. (f) The second coordination environment of Cr³⁺ ions surrounded by YO₈ units. (g) The third coordination environment of Cr³⁺ ions surrounded by Al/GaO₄ units.

The low pre-edge peak intensity of Cr doped YAGG samples reveals octahedral coordination, indicating that the Cr sites have centrosymmetry and prohibit the hybridization. Local structural information has been extracted by analysing the EXAFS region of the absorption spectra. The EXAFS oscillations, depicted in Fig. 2(b) at the Y, Ga, and Cr K-edges after removing the bare atom background, are presented for examination. These oscillations can serve as valuable tools for the quantitative analysis of the spectra. The EXAFS oscillations at the Cr K-edge (Fig. 2(b)) exhibit a notable similarity to the Ga K-edge oscillations when compared to the Y K-edge EXAFS oscillations. As we discussed above, Ga can exist in both tetrahedral and octahedral coordination sites. The EXAFS oscillations depicted in Fig. 2(b) suggested a combination of both. Based on the obtained coordination (close to 4 instead of 6), we concluded that Ga is likely to prefer tetrahedral sites. Based on the obtained Cr–O coordination number (close to 6), we predicted them to be present at octahedral sites. We would also like to clarify that while the EXAFS oscillations of Cr are not identical to those of Ga, they are similar. The coordination number primarily affects the amplitude of the oscillations when bond lengths are quite similar. The resemblance in EXAFS oscillations between Cr and Ga can be attributed to their comparable bond lengths (1.98 Å for Cr and 1.85 Å for Ga, instead of the larger 2.32 Å). It is reasonable to expect the oscillations to be similar to Ga. Additionally, the contributions of other coordination shells to the oscillations should be similar to Ga rather than Y. This observation suggests the substitution of Cr ions at octahedral Ga/Al crystallographic sites.

Fig. 3 (a) PLE spectra of YAGG:xCr³⁺ (x = 0.02–0.30 mol) phosphors at room temperature. (b) PL emission spectra and the integral intensities as a function of Cr³⁺ concentrations (inset). (c) Normalized PL emission spectra and emission photograph of the YAGG:0.04Cr phosphor under UV light (inset). (d) FWHM as a function of Cr³⁺ concentrations. (e) Schematic of the Tanabe–Sugano (T–S) energy level diagram for octahedrally coordinated Cr³⁺ ions. (f) Variation of I(⁴T₂)/I(²E) with Cr³⁺ doping content and emission bands for ²E and ⁴T₂ levels (inset).

The χ(R) versus R plots, derived from the Fourier transform of χ(k) spectra, are illustrated in Fig. S3 (ESI†) at the Y, Ga, and Cr K-edges. The displayed Fourier transform spectra are not phase-corrected, leading to the coordination peak in these spectra appearing at slightly lower inter-atomic distances (R) in comparison to the actual bond length. Upon visual inspection of the Fourier transform spectra at the Ga, Y, and Cr K-edges, it is apparent that the Cr spectra exhibit a notable similarity to the Ga K-edge spectra. The outcomes of the fitting process are presented in Table 1 and Tables S1, S2 (ESI†), and the optimal fitting spectra are illustrated in Fig. S4 and S5 (ESI†). Although Y³⁺ ions typically occupy dodecahedral sites with eight-coordinate geometry, this coordination is spread across two adjacent coordination shells. The Y–O coordination is represented by two nearby distances of 2.32 Å and 2.45 Å, as shown in Table S1 (ESI†). The noteworthy findings are evident in the Cr K-edge results, as displayed in Fig. 2(c).

Table 1 Bond length, coordination number and disorder factor obtained by EXAFS fitting at the Cr K-edge

Path	Parameters	YAGG:2% Cr	YAGG:3% Cr	YAGG:5% Cr	YAGG:15% Cr
Cr–O	R (Å)	1.98 ± 0.01	1.99 ± 0.02	1.98 ± 0.02	2.00 ± 0.01
	N	5.77 ± 0.32	6.02 ± 0.52	5.77 ± 0.25	5.53 ± 0.21
	σ ^2 (Å^2)	0.0022 ± 0.0009	0.0020 ± 0.0012	0.0021 ± 0.0013	0.0020 ± 0.0015
Cr–Y	R (Å)	3.49 ± 0.01	3.47 ± 0.02	3.47 ± 0.02	3.49 ± 0.02
	N	5.76 ± 0.25	6.02 ± 0.35	5.79 ± 0.39	5.53 ± 0.35
	σ ^2 (Å^2)	0.0187 ± 0017	0.0110 ± 0014	0.0172 ± 0011	0.0159 ± 0017
Cr–Al/Ga	R (Å)	3.49 ± 0.02	3.47 ± 0.02	3.47 ± 0.02	3.49 ± 0.02
	N	5.76 ± 0.25	6.02 ± 0.35	5.79 ± 0.39	5.53 ± 0.35
	σ ^2 (Å^2)	0.0015 ± 0.0016	0.0018 ± 0.0016	0.0021 ± 0.0016	0.0037 ± 0.0011

---

The Cr ions situated at substitutional sites exhibit octahedral coordination, a conclusion substantiated by the discernible features observed in XANES measurements. The EXAFS oscillations in K-space discussed earlier signify the presence of Cr at Al/Ga sites, with the subsequent EXAFS results identifying an octahedral site preferred by Cr ions. The local structure of Cr³⁺ ions with the second and third coordination shells is presented in Fig. 2(d)–(g). The determined coordination number decisively precludes the possibility of Cr occupying tetrahedral sites, yielding a compelling result. This observation is particularly intriguing. The preference for the octahedral site can be rationalized by considering the analogous ionic radii of trivalent Cr in octahedral coordination (0.615 Å) with trivalent Ga ions in octahedral coordination (0.62 Å), in contrast to tetrahedral coordination (0.47 Å).⁴⁴

2.3 Photoluminescence studies

The PLE spectra of Cr³⁺ doped YAGG phosphors monitored at 690 nm emission are shown in Fig. 3(a), in which three excitation bands were observed at 280 nm, 440 nm, and 610 nm, which can be ascribed to the spin-allowed ⁴A₂ → ⁴T₁ (⁴P), ⁴A₂ → ⁴T₁ (⁴F) and ⁴A₂ → ⁴T₂ (⁴F) transitions of Cr³⁺ ions, respectively.^45,46Fig. 3(b) displays the concentration-dependent PL emission spectra of Cr³⁺ doped YAGG phosphors upon blue light excitation at 440 nm. The ²E → ⁴A₂ transition results in a sharp-line emission peak at 690 nm (R-line), and the intense broadband NIR emission in the range of 630–950 nm can be attributed to ⁴T₂ → ⁴A₂ transition.^45,46 The comparison of PL emission spectra of the YAGG:0.04Cr phosphor recorded at 80 K and room temperature is shown in Fig. S6 (ESI†). At low-temperature, the sharp-line emission from ²E states is dominant. The Stokes and anti-Stokes phonon sidebands of the R-line can be seen in the PL emission spectrum over a broad emission band at 710 and 725 nm at 80 and 300 K.¹³ At room temperature, the overlap of the broadband ⁴T₂ emission band and phonon sidebands at 710 nm and 725 nm is observed with increment in ⁴T₂ emission while the sharp ²E emission decreased sharply.

The maximum integrated emission intensity was observed at x = 0.04 (inset of Fig. 3(b)), and further increment in Cr³⁺ doping content (x ≥ 0.06) led to decreased PL intensity due to concentration quenching. The IQE of YAGG:0.04Cr³⁺ is 76% (Fig. S7, ESI†), which confirmed high luminous efficiency of Cr³⁺-doped YAGG phosphors. The emission photograph of the YAGG:0.04Cr phosphor under UV light is provided in the inset of Fig. 3(c). It is well reported that the luminescence of Cr³⁺ is significantly influenced by the crystal field.⁴⁷ The presence of emission bands due to both ²E/⁴T₂ → ⁴A₂ transitions revealed the incorporation of Cr³⁺ ions in octahedral sites with intermediate crystal field strength. We evaluated the crystal field strength (Dq) and Racah parameter (B) of the Cr³⁺ sites in the YAGG:0.04Cr³⁺ sample by using the following equations:⁴⁸

Dq/B = 15(x − 8)/x² − 10x (1)

Dq = E(⁴A₂ − ⁴T₁)/10 (2)

x = E(⁴A₂ → ⁴T₁) − E(⁴A₂ → ⁴T₂)/Dq (3)

where the x value was calculated by using the average peak energies of the ⁴A₂ → ⁴T₁/⁴T₂ transitions of Cr³⁺ ions.¹⁴Fig. 3(e) displays the Tanabe–Sugano (T–S) diagram with located crystal field strength and energy level positions of Cr³⁺ ions in octahedral coordinated Al/GaO₆ sites in the YAGG matrix. Accordingly, the Dq/B, Dq, and B values were determined as 2.37, 1523.9 cm⁻¹, and 642 cm⁻¹, respectively. Based on the T–S diagram, the ⁴T₂ energy level is slightly higher in energy than ²E due to which the dominant sharp-line emission along with broad NIR emission is observed for all Cr³⁺-doped YAGG phosphors. Two regions were selected to evaluate the integrated intensity of ⁴T₂ and ²E emission bands¹⁷ (see the inset of Fig. 3(f)). An increasing trend was observed for the I(⁴T₂)/I(²E) values calculated as a function of Cr³⁺ doping content in Fig. 3(f).

No changes were noticed in the peak positions of excitation and emission bands in concentration-dependent PLE and PL emission spectra of Cr³⁺ doped YAGG phosphors. This revealed that increasing content of Cr³⁺ has a negligible effect on the crystal field of Cr³⁺ ions. This was supported by intact Cr–O bonds of Cr³⁺ ions in Al/GaO₆ octahedral sites, as no significant changes were observed in bond lengths and disorder values (σ²) in the EXAFS analysis (Table 1). The normalized PL emission as a function of Cr³⁺ doping concentration is shown in Fig. 3(c). It is noteworthy that the PL emission became broader (Fig. 3(c)) and the FWHM of emission band increased from 44 to 72 nm (see Fig. 3(d)) with increase in doping concentration of Cr³⁺ ions (x = 0.02 to 0.30 mol). Here, the increased FWHM cannot be explained by changes in the crystal field generally accounted in the previous reports.¹ This may be attributed to the increasing electron–phonon coupling (EPC) effect in the YAGG lattice.⁴⁹ EPR is a hypersensitive technique used to gain useful information about the local structure of Cr³⁺ ions. The EPR spectra of all the Cr³⁺ doped YAGG samples were recorded at 100 K to gain insights into the local structure of Cr³⁺ ions in the YAGG lattice (see Fig. S8, ESI†). Two EPR signals were observed at g ∼ 3.75 and g ∼ 1.98, which can be attributed to isolated Cr³⁺ ions in octahedral coordination and Cr³⁺–Cr³⁺ pairs, respectively, based on the available literature.^30,49 The formation of Cr³⁺–Cr³⁺ ion pairs is more feasible at higher Cr³⁺ doping content, and therefore, these pairs contribute significantly to the increased tail-like emission at higher wavelengths (>800 nm).¹⁷ The probable electron migration from ²E to ⁴T₂ energy levels between neighbouring Cr³⁺ ions is expected to increase with increment in Cr³⁺ concentration, which results in thermal population of the ⁴T₂ energy state.

Room temperature luminescence decay profiles were recorded for Cr³⁺-doped YAGG phosphors under 440 nm excitation, monitoring emissions at 690, 710, and 750 nm (Fig. 4(a) and (b) and Fig. S9, ESI†). The average lifetimes (τ_avg) were calculated by fitting the excited state decay curves with a bi-exponential decay function:

I(t) = I₀ + A₁e^−t/τ₁ + A₂e^−t/τ₂ (4)

where I₀ and I(t) represent the emission intensities at zero offset and at ‘t’ time, respectively, τ₁ and τ₂ are the two lifetime components, and A₁ and A₂ are constants. The lifetime values for different emission peak positions of Cr³⁺-doped YAGG phosphors are summarized in Table S3 (ESI†). The average lifetimes of the 690 nm emission peak decreased significantly from 477.6 to 263.5 μs upon increasing the Cr³⁺ doping concentrations (x = 0.02 to 0.30) (Fig. 4(a)). Similarly, the lifetime values of 750 nm peaks decreased from 445.4 to 257.8 μs (Fig. 4(b)). This decreasing lifetime trend indicated a higher probability of non-radiative relaxation processes at higher Cr³⁺ doping due to concentration quenching. For a better understanding of decay kinetics, the time-resolved emission spectrum (TRES) is shown in Fig. 4(c). The sharp-line emission (²E) can be seen for longer delay times (∼3 ms); however, the broadband emission (⁴T₂) vanished at shorter delay times. Hence, the best fit of decay curves into the bi-exponential model and microsecond decay time suggested strong spin–orbit coupling (SOC) between ⁴T₂ and ²E excited states of Cr³⁺ ions.⁵⁰

Fig. 4 Decay curves of YAGG:xCr³⁺ (x = 0.02–0.30 mol) phosphors monitored at (a) 690 nm and (b) 750 nm, upon 440 nm excitation, and (c) TRES spectra of the YAGG:0.04Cr sample under 440 nm excitation.

2.4 Temperature-dependent PL studies

To evaluate thermal stability of the Cr³⁺-doped YAGG phosphor, the temperature-dependent PL properties were investigated in detail. Fig. 5(a) displays the temperature-dependent PL emission spectra of the YAGG:0.04Cr³⁺ phosphor. The PL intensity of ²E → ⁴A₂ (R-line) and phonon bands decreased gradually at elevated temperature; however, the PL emission band became broader at higher temperatures due to enhanced emission from the ⁴T₂ level (Fig. 5(b)). The total integral intensity (²E + ⁴T₂ → ⁴A₂) remains almost constant at elevated temperatures and nearly zero-thermal quenching was observed in the temperature range of 298 to 433 K (see Fig. 5(c) and (d)). The YAGG:0.04Cr³⁺ phosphor exhibits excellent thermal stability and maintained 105% of the initial emission intensity at 150 °C. The peak intensities at 690 (²E → ⁴A₂) and 750 nm (dominant ⁴T₂ → ⁴A₂) as a function of temperature are shown in Fig. 5(e). The sharp-line emission intensity from the ²E state decreased significantly (∼67% at 423 K), while the intensity at 750 nm increased to 113% at 423 K. Hence, the increased broadband emission originating from ⁴T₂ → ⁴A₂ compensates for the thermal quenching of PL intensity. Consequently, FWHM of the emission band increased significantly with increase in temperature (Fig. 5(f)).

Fig. 5 (a) Temperature-dependent PL emission spectra of YAGG:0.04Cr³⁺ in the range of 298 to 433 K. (b) Normalised PL emission variation with temperature. (c) Contour plot of PL emission spectra at different temperatures. (d) Integral intensity at different temperatures. (e) Peak intensity for ²E (690 nm) and ⁴T₂ (750 nm) as a function of temperature. (f) FWHM as a function of temperature.

We evaluated the activation energy for thermal quenching of the ²E → ⁴A₂ emission band using the Arrhenius equation:⁵¹

I_T = I₀/(1 + A exp(−ΔE/kT)) (5)

where I₀ and I_T are the integrated emission intensities at the initial (T₀) and given temperatures (T), respectively, ΔE denotes the activation energy for thermal quenching, k is the Boltzmann constant, and A is a constant. The E_a values were calculated by linear fit of experimental data in the ln[(I₀/I_T) − 1] versus 1/kT plot shown in Fig. 6(a). Two sections were observed with distinct ΔE values (0.38 eV and 0.166 eV) on different temperature scales. This revealed the prevalence of different thermal processes in different temperature ranges. At lower temperature, the energy barrier for thermal quenching is higher. However, with increase in temperature, a considerable lowering of the activation energy was noticed.

Fig. 6 (a) ln[(I₀/I_T) − 1] versus 1/kT plot. (b) Configuration coordinate diagram of the thermal quenching mechanism of the Cr³⁺-doped YAGG matrix.

The configurational coordinate diagram of Cr³⁺ can be referred to get a better understanding of the thermal quenching behavior (Fig. 6(b)). The sharp line emission from the ²E state is shown by ²E → ⁴A₂ transition (process 1). In general, the thermal quenching mechanism of broadband emission originating from ⁴T₂ → ⁴A₂ transition (process 2) is due to increased non-radiative loss at elevated temperatures. The electrons in the ⁴T₂ level may cross the intersection point of ⁴T₂ (excited state) and ⁴A₂ (ground state) with the energy barrier (ΔE₂) at higher temperatures. Consequently, the electron returns to ⁴A₂ states by nonradiative transitions that result in decreased intensity of ⁴T₂ → ⁴A₂ emission (process 5).

Here, the thermal quenching mechanism of the sharp-line emission from the ²E state (690 nm) is distinct from ⁴T₂, which can be observed by intensity changes at 750 nm. The different thermal quenching behaviour of the ²E state from the ⁴T₂ state can be explained by the intersection between ²E and ⁴T states with lower activation energy due to intermediate crystal field strength. The thermal depopulation of the ⁴T₂ state is realized through a cross-over process from the ⁴T₂ to  the ⁴A₂ state (process 4).⁵² At high temperatures, the electrons can easily reach the intersection point of the ²E and ⁴T₂ excited levels from the ²E state in the case of intermediate crystal field strength, and this process (process 3) has an energy barrier denoted as ΔE₁ (where ΔE₁ < ΔE₂ for the intermediate crystal field). Following this, the electron can undergo radiative transition or non-radiative transition from ⁴T₂ excited levels to the ground state. Because of the thermal population of ⁴T₂ from the ²E state, the broad ⁴T₂ → ⁴A₂ emission can be enhanced at elevated temperatures. The above results can be further supported by the temperature-dependent lifetime study of ⁴T₂ → ⁴A₂ and ²E → ⁴A₂ transitions discussed in Section 2.6. The decrease of ²E → ⁴A₂ lifetimes explains the abnormal anti-thermal quenching process of the broadband ⁴T₂ → ⁴A₂ emission observed in this work. The weakening of the sharp-line emission (²E) and the strengthened broad ⁴T₂ → ⁴A₂ emission (Fig. 5(a) and (b)) revealed that thermal population of electrons in ⁴T₂ from the ²E state may compensate for the non-radiative loss at higher temperatures. However, the enhanced ⁴T₂ → ⁴A₂ broadband emission cannot be explained solely by thermal population. Therefore, the role of defects has been investigated thoroughly in the next section.

2.5 Role of defects in thermal stability and NIR PersL: experimental and theoretical investigation

We employed the PALS technique to understand the nature and density of defects in Cr³⁺-doped YAGG phosphors. The fitting of positron annihilation lifetime spectra yielded three lifetimes in all the cases as shown in the typical fitting in Fig. S10a (ESI†). The longest lifetime was ∼1.5 ns and with an intensity of <0.5% but the inclusion of this component was necessary for best fit with χ² close to 1. This component is due to positronium formation on the surface of the powder particles and is well known and hence it is not discussed further. The other two positron lifetimes denoted as τ₁ and τ₂ in the increasing order of their magnitudes and the fractional intensity (I₁) corresponding to τ₁ (I₂ ∼100 − I₁) are shown in Fig. S10b and c (ESI†). The intensity weighted average of the two positron lifetimes (τ₁) is also shown in Fig. S10b (ESI†). The first positron lifetime is from the delocalised bulk with some contribution from shallow positron traps like oxygen vacancies, while the second positron lifetime is from defects and surface annihilation of positrons. The first positron lifetime in undoped YAG is about 180 ps and is in good agreement with reported positron lifetimes of YAG based samples.^53,54 With increase in Cr³⁺ doping, the first positron lifetime reduced and reached minima at about 5% doping and it increased for the 15% doped sample. The second positron lifetimes are more than 400 ps for the undoped sample and reduced with Cr³⁺ doping, suggesting that the pre-existing large vacancy clusters near the surface of the particles anneal out in favour of isolated vacancies. The average positron lifetime decreased with increase in Cr³⁺ doping up to 5% doping and later increased. The intensity of positrons annihilating from delocalised states (I₁) showed minima at about 2% doping and showed drastic reduction in the 15% doped sample. The results obtained upon comparison with theoretically calculated positron lifetimes for various combinations of possible vacancies^55,56 suggest the formation of Al/Ga vacancies or aluminium/gallium vacancies associated with oxygen vacancies at higher concentration of Cr³⁺ doping.

For DFT calculations, we have considered cubic crystal structure with the space group Ia3̄d (space group number: 230) for Y₃Al₂Ga₃O₁₂ (Fig. 1(b)), where Ga³⁺, Al³⁺ and Y³⁺ ions occupy the tetrahedral, octahedral and dodecahedral sites, respectively. The calculated lattice parameter is found to be 12.30 Å, which is close to the previously calculated and experimental values.^57–59 The calculated band gap is found to be 6.96 eV, which is in excellent agreement with the experimental value of 7.01 eV.⁶⁰ The density of states (DOS) plot for Y₃Al₂Ga₃O₁₂ in the absence of any defect is shown in Fig. 7(a). The valence band minimum (VBM) is mainly contributed by O (p) states with minor contribution from Y (p, d) and Ga (s, p, d). The bottom of the conduction band (CB) is mainly composed of Y (d) with minor Y (p), Ga (s, p), and O (p) states. The calculated defect formation energies of the defect containing Y₃Al₂Ga₃O₁₂ are provided in Table S4 (ESI†). It can be observed that formation of single vacancy follows the order, V_O < V_Ga < V_Al < V_Y. We have also considered vacancy clusters, containing cation–anion and cation–cation vacancies. The formation energy follows the order: (V_Ga, V_O) < (V_Al, V_O) < (V_Y, V_O) < (V_Ga, V_Al). Now, we proceed to discuss the influence of different types of defects on the electronic structure of Y₃Al₂Ga₃O₁₂. The DOS plots of Y₃Al₂Ga₃O₁₂ in the presence of different types of defects are shown in Fig. 7(b)–(f) and Fig. S11, S12 (ESI†). Fig. S13 (ESI†) shows the summary of the electronic structure description of all the defect containing systems. It can be seen from the DOS plots that the discrete impurity states in the forbidden region appear in the presence of all the vacancy defects, resulting in the formation of intermediate defect states between the valence band (VB) and CB.

Fig. 7 (a) DOS of Y₃Al₂Ga₃O₁₂ and DOS of Y₃Al₂Ga₃O₁₂ in the presence of (b) one neutral oxygen vacancy, (c) neutral gallium and oxygen vacancies, (d) neutral aluminium and oxygen vacancies, (e) one neutral aluminium vacancy, and (f) one neutral gallium vacancy. The vertical dashed line indicates the Fermi level.

Based on the above DFT and PALS results, we can conclude that the presence of oxygen vacancies (V_O) and oxygen vacancy–cation vacancy clusters in the YAGG matrix assisted in achieving excellent PL thermal stability. The DFT calculations suggested feasibility of V_O, V_Ga–V_O and V_Al–V_O cluster defect formation due to lowest defect formation energies. This is in good correlation with PALS lifetime results, which indicated the formation of cluster defects along with V_O formation due to high temperature sintering.

The Phonopy package⁶¹ has been used for processing the data from VASP calculations to calculate the phonon spectra. Fig. S14 (ESI†) shows the phonon dispersion plot of Y₃Al₂Ga₃O₁₂. It can be observed that there exists no significant imaginary mode, indicating the dynamic stability of Y₃Al₂Ga₃O₁₂. This is expected since the garnet materials have been studied experimentally and reported to have high structural rigidity and thermal stability.¹

To further characterize the trap depths and distribution, the TL studies were performed for the YAGG:0.04Cr³⁺ phosphor and the TL glow curve after charging with blue light (440 nm) for 100 s is displayed in Fig. 8(a). In the TL spectrum, the YAGG:0.04Cr³⁺ phosphor exhibited a broad TL band, which indicated a continuous distribution of traps in the range of 320 to 500 K. The deconvolution of the broad TL band provided three Gaussian peaks at 358, 394, and 458 K, respectively. The trap depth (E) can be evaluated using the following equation:

E = T_m/500 (6)

where T_m represents the peak temperature (in K). The calculated trap depth values of the YAGG:0.04Cr³⁺ phosphor are 0.72, 0.82, and 0.91 eV. The three trap depths revealed the presence of both shallow and deep trap levels, which are located in between the valence band (VB) and conduction band (CB) of the YAGG:0.04Cr³⁺ phosphor. To verify the presence of V_O, the XPS study of the YAGG:0.04Cr³⁺ sample was carried out and the survey scan confirmed the presence of Y, Al, Ga, Cr and O elements (Fig. S15, ESI†). The XPS spectrum of O 1s is illustrated in Fig. 8(b). The Gaussian deconvolution of O 1s resulted in two peaks at ∼530.2 and ∼531.8 eV, which correspond to lattice oxygen of the sample and oxygen vacancies, respectively.^6,15

Fig. 8 (a) TL spectrum of the YAGG:0.04Cr³⁺ phosphor at room temperature upon charging with 440 nm light for 100 s, (b) O 1s XPS spectra of the YAGG:0.04Cr³⁺ phosphor, and (c) PersL decay curve under UV lamp (254 nm) charging for 2 min.

The PersL decay curve was acquired after charging the YAGG:0.04Cr³⁺ phosphor with a 254 nm UV lamp for 2 min (Fig. 8(c)). The long PersL decay was observed and afterglow intensity was much higher than background even after 1 hour of decay, demonstrating the potential of YAGG:Cr³⁺ phosphors as efficient PersL materials.

Compiling the above results, possible mechanisms for defect induced thermal stability and PersL are proposed and illustrated in the schematic diagrams in Fig. 9(a) and (b), respectively. Fig. 9(c) and (d) demonstrate the schematic representation of V_O and (V_O–V_Al/Ga) cluster defects in the YAGG matrix, respectively. In summary, the thermal activation of NIR emission involving defect states can be divided into six parts based on earlier reports.^15,60 Firstly, the electrons of Cr³⁺ ions are excited to the ⁴T₁ (⁴F) state by blue light excitation (440 nm) from the ground state, ⁴A₂ (process 1). It is noteworthy that electrons will not be reaching the CB upon blue light excitation, and thus, shallow traps will not be occupied during low energy excitation. Then, the electrons can relax back to lower energy excited states, ⁴T₂ (⁴F) and ²E by non-radiative pathways (process 2). This can be followed by processes 3 and 4, resulting in sharp-line and broadband NIR emission upon radiative transitions to the ground state. However, some of the electrons in excited states (⁴T₂ (⁴F) and ²E) can be captured by deep traps such as V_O and cluster defects via tunnelling processes, as shown in process 5. The electrons cannot escape these deep traps (E > 0.8 eV) at room temperature; however, thermal stimulation will provide activation energy required for detrapping of electrons from deep traps to Cr³⁺ excited energy levels (process 6). Thereafter, final transition to the ground state will result in enhanced emission intensities at higher temperatures.

Fig. 9 Schematic representation of the mechanism of (a) defect induced thermal stability and (b) PersL. (c) and (d) Schematic representation of V_O and (V_O–V_Al/Ga) cluster defects in the YAGG matrix, respectively.

Furthermore, using UV excitation to obtain PersL (Fig. 9(b)) will populate ⁴T₁ states and filling of traps through the conduction band is feasible (process 7). With time, the detrapping processes (process 8) and tunnelling of electrons (process 6) from deep traps will result in PersL of Cr³⁺ for a longer time.

Here, we could achieve thermally robust luminescence of Cr³⁺ ions and explain the role of structural rigidity and defects in the YAGG matrix. The quantum efficiency of the YAGG:Cr³⁺ matrix is quite high, ∼76%. In order to balance the both, optimization of defect concentration is crucial for achieving high quantum efficiency and maintaining the loss of luminescence at elevated temperatures (not more than 3–4%). Defect engineering will be performed in the future to investigate the effect of defect density on the PLQY and achieve higher quantum yield for the thermally robust YAGG:Cr phosphor reported in this work.

2.6 Low temperature-dependent PL study and EXAFS investigation

The investigation of photophysical properties of Cr³⁺ ions at low temperatures can provide useful insights into the different radiative and non-radiative processes and temperature-dependent population of excited states (⁴T₂/²E). In order to explore the luminescence properties of Cr³⁺ at cryogenic temperatures, we recorded low temperature dependent PL emission spectra of the YAGG:0.04Cr³⁺ sample in the range of 80 to 300 K as shown in Fig. 10(a). The emission intensities as a function of temperature could be visualised in Fig. 10(b). The normalised PL emission spectrum at 80 and 300 K revealed considerable broadening of the emission band with diminished ²E emission at room temperature (Fig. 10(c)). In this work, the Dq/B value of 2.37 falls into the regime of the intermediate crystal field, where ⁴T₂ and ²E excited states lie within the reach of the intersection point between the ⁴T₂/²E excited states, with a very small ΔE₁ (Fig. 6(b)). The emergence of dominant sharp emission (²E) at low temperature is in agreement with the thermal population of ⁴T₂ excited states from ²E at elevated temperatures. Therefore, strong ²E → ⁴A₂ emission is observed at low temperature and the broadband ⁴T₂ → ⁴A₂ emission is attainable at higher temperatures. The normalised intensity PL emission spectra revealed similar increment in the FWHM of the emission band as observed at high temperatures (Fig. 10(d)). No prominent peak shifts were observed for the R-line and phonon bands. Fig. 10(e) depicts the variation in total integrated and peak intensities with temperature. The overall integrated intensity does not vary much with increase in temperature; however, the intensity of phonon sidebands decreased with temperature. The emission loss is compensated by increase in the broadband ⁴T₂ → ⁴A₂ emission. The same can be supported by the decreased lifetime of the ²E excited state at elevated temperatures. The decay curves under 440 nm excitation and monitoring 690 nm emission as a function of temperature are shown in Fig. 10(f). Decay curves monitored at 710 and 725 nm are presented in Fig. S16 (ESI†) and lifetime values are tabulated in Table S5 (ESI†), where a similar decrease in the lifetimes was observed.

Fig. 10 (a) Low temperature-dependent PL emission spectra of YAGG:0.04Cr³⁺ in the range of 80 to 300 K. (b) Contour plot of PL emission spectra as a function of temperature. (c) Normalised PL emission at 80 and 300 K. (d) Normalised PL emission spectra at different temperatures. (e) Integral intensity at different temperatures and peak intensities at 690, 708, and 750 nm as a function of temperature. (f) Luminescence decay profiles in the temperature range of 80 to 300 K monitored at 690 nm under 440 nm excitation.

To probe the local structure changes in the octahedral coordination of Cr³⁺ ions, low-temperature dependent EXAFS studies were carried out at the Ga K-edge as Cr³⁺ ions substitute Ga/Al octahedral sites. The temperature-dependent EXAFS measurements were conducted using a closed-cycle He cryostat, spanning a range from 16 to 300 K at the Ga K-edge of the YAGG:0.04Cr³⁺ sample in the transmission mode. Fig. 11(a) presents the XANES spectra from these temperature-dependent measurements, revealing a consistent oxidation state across the temperature spectrum. However, notable alterations are evident in the Fourier-transform EXAFS spectra, as depicted in Fig. 11(b). Specifically, the second and third coordination peaks (observed at 2.9 and 3.3 Å, respectively) exhibit an augmentation in peak amplitude with decreasing temperature. This temperature-dependent variation appears to be a conventional thermal effect, devoid of any anomalous behavior across the temperature range.

Fig. 11 (a) Temperature-dependent XANES spectra at the Ga K-edge. (b) Fourier transform EXAFS fittings of undoped and Cr³⁺ doped YAGG at the Ga K-edge for temperature-dependent measurements. (c) Variation of peak amplitude of three coordination peaks. (d) Variation of Debye–Waller factors of the corresponding coordination peaks with temperature.

An intriguing observation emerges regarding the nearly constant amplitude of the first coordination peak throughout the temperature range. This phenomenon could be attributed to the minimal vibrational activity induced by temperature changes in the octahedral coordination of Ga sites (GaO₆), particularly in proximity to neighboring atoms. The observed differences could be attributed to varying coupling levels. The robust coupling interaction between gallium (Ga) and oxygen (O) atoms may render the Ga–O coordination less susceptible to external influences. Conversely, the coordination with other atoms is characterized by greater distances from the Ga atoms, resulting in weaker coupling interactions. Consequently, the next-nearest–neighbor coordination and third-nearest–neighbor coordination are more prone to temperature-induced alterations due to their comparatively smaller coupling interactions. Fig. 11(c) illustrates the amplitude of all three peaks, corresponding to Ga–O and Ga–Y/Al/Ga coordinations, displaying a characteristic temperature-dependent pattern. Furthermore, the Debye–Waller factors obtained from the EXAFS fitting are shown in Fig. 11(d). Despite the presence of sizable error bars, a systematic trend is discernible with respect to temperature, indicative of underlying structural variations similar to the correlated Debye model.

The above results revealed rigidity of GaO₆ octahedral sites as the bond lengths remained intact without significant increase in the disorder values. This suggested that the substitution of rigid Al/GaO₆ sites by Cr³⁺ ions will result in less structural distortion as a function of temperature and thus assist in attaining the abnormal thermal quenching of NIR luminescence.

2.7 Applications

2.7.1 NIR LED fabrication. The NIR pc-LED device based on YAGG:0.04Cr³⁺ was fabricated using the 420 nm blue LED chip with the YAGG:0.04Cr phosphor. Fig. 12(a) shows the electroluminescence (EL) spectra of the YAGG:0.04Cr³⁺ phosphor and the images of a 420 nm LED and NIR pc-LED device emission (without/with a blue filter) in the inset. In order to demonstrate the potential of our device for night vision applications, the imaging performance was evaluated by capturing the photos of BARC logo and flowers in daylight (left panel) and under NIR pc-LED light (right panel) with a camera (Fig. 12(b)).

Fig. 12 (a) Electroluminescence (EL) spectra of the YAGG:0.04Cr³⁺ phosphor and the images of a 420 nm LED and NIR pc-LED device emission (without/with a blue filter), (b) photos of objects under daylight (left panel) and NIR light captured by a camera (right panel). (c) Power-dependent EL spectra of the fabricated YAGG:0.04Cr³⁺ NIR pc-LED. (d) Contour maps for intensity as a function of power. (e) Logarithmic plots of I versus P.

The power-dependent EL spectra of the fabricated NIR pc-LED are shown in Fig. 12(c) and (d). The EL emission intensity increased when the power is increased from 600 mW to 1200 mW (Fig. 12(e)). The slope of 1 obtained for logarithmic plot of intensity versus power revealed linear relation of EL intensity with power. Hence, the NIR pc-LED fabricated in this work displays superior optoelectronic properties, which can be utilized for night vision and other potential applications.

2.7.2 Thin film for NIR persistent luminescence. Thin film of YAGG:0.04Cr³⁺ phosphor/PMMA was fabricated and uniform dispersion of the material in PMMA was confirmed by elemental mapping (Fig. S17, ESI†). Fig. 13(a) shows the PL emission spectra of the thin film upon 440 nm blue light excitation. The inset displays the emission image of the thin film, and photographs of the thin film kept upon letters to show transparency of the fabricated thin film. The thin film fabrication imparted moisture stability to YAGG:0.04Cr³⁺ phosphor emission that was recorded after keeping the film in water for a day (Fig. S18, ESI†). Benefitting from large trap distribution observed in the TL spectrum, we investigated the persistent (PersL) luminescence of the thin film of the YAGG:0.04Cr³⁺ phosphor upon charging by X-rays for 1 minute. Fig. 13(b) presents the persistent luminescence decay curve of the YAGG:0.04Cr³⁺ phosphor and a NIR PersL for 2.5 hours was attained in this work. The NIR PersL can be utilized for night vision and imaging applications.

Fig. 13 (a) PL emission spectra of YAGG:0.04Cr³⁺/PMMA film with the emission image and film photographs in the inset. (b) PersL decay curve of YAGG:0.04Cr³⁺ thin film under X-ray charging for 1 minute.

3. Experimental section

3.1 Synthesis

A series of YAGG:xCr³⁺ (x = 0.02, 0.04, 0.06, 0.10 and 0.30 mol) phosphors were synthesized by a high-temperature two step solid-state method. The raw materials, Y₂O₃ (99.99%), Al₂O₃ (AR), Ga₂O₃ (AR), and Cr₂O₃ (AR), were weighed in stoichiometric amounts and mixed homogeneously by grinding for 20 min using an agate mortar and pestle. Then, the fine mixture was calcined at 1200 °C for 15 h in a tubular furnace in air. Then, the powder was annealed at 1400 °C for 15 h in air after intermittent grinding for 15 min. After cooling, all the samples were ground well into fine powder for further measurements.

3.2 Fabrication of thin film

1.5 g of poly(methyl methacrylate) (PMMA) polymer (average M.W. ∼15 000) was dissolved in toluene solution. Then, the YAGG:0.04Cr³⁺ powder was dispersed in solution and kept stirring overnight. The thin films were fabricated on a glass substrate by a drop casting method and left to dry at room temperature.

The characterization techniques, EXAFS experimental details, and computational details are provided in Sections S1, S2 and S3 in the ESI.†

4. Conclusions

In this work, we synthesized an efficient Y₃Al₂Ga₃O₁₂:Cr³⁺ NIR garnet phosphor, which exhibits broadband emission covering a range of 630–950 nm. The Y₃Al₂Ga₃O₁₂:Cr³⁺ phosphor displays thermally stable broadband NIR emission and at 150 °C, it maintains 105% of the intensity at room temperature. This material possesses a high internal quantum efficiency of 76%. EXAFS studies confirmed the octahedral coordination of Cr³⁺ ions and no significant changes were observed in bond lengths at higher Cr³⁺ content. EPR studies correlated well with EXAFS and an intense signal at g ∼ 3.75 confirmed isolated Cr³⁺ ions in octahedral sites. Temperature dependent studies probed the structural changes of GaO₆ sites with temperature and revealed rigid octahedral sites. Furthermore, TL studies revealed large trap formation with a wide range of trap depths (∼0.7 to 1 eV). The formation of oxygen vacancies (V_O) and cluster defects (V_Al/Ga–V_O) was affirmed by XPS and PALS studies, which facilitated the observed long PersL with UV charging (254 nm). DFT calculations revealed the lower formation energies of V_O and V_O–V_Ga/Al cluster defects. The mechanism of defect activated thermal stability and PersL was established based on the experimental and theoretical calculations. Finally, an NIR pc-LED was fabricated, which displays efficient electroluminescence and can be utilized for potential applications in imaging and night vision. The thin film was fabricated using the YAGG:Cr³⁺ phosphor, which showed long PersL under X-ray charging.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

A. Balhara and Santosh K. Gupta would like to thank Dr Biplab Ghosh, RRCAT for allotment of the EXAFS beamline and Dr Deepak Tyagi, BARC for help with XPS measurements. We would also like to acknowledge Dr Mohit Tyagi, BARC, for performing X-ray PersL measurements, Dr Manoj Kumar Gupta, AMPRI, for SEM analysis, and Dr Praveen Kumar, IACS, Kolkata for help with TEM measurements. This research is funded by Goverment of India through Department of Atomic Energy.

References

1. P. Dang, Y. Wei, D. Liu, G. Li and J. Lin, Recent advances in chromium-doped near-infrared luminescent materials: fundamentals, optimization strategies, and applications, Adv. Opt. Mater., 2023, 11(3), 2201739.
2. V. Rajendran, W.-T. Huang, K.-C. Chen, H. Chang and R.-S. Liu, Energy-saving chromium-activated garnet-structured phosphor-converted near-infrared light-emitting diodes, J. Mater. Chem. C, 2022, 10(39), 14367–14378.
3. L. Zhang, D. Wang, Z. Hao, X. Zhang, G. H. Pan, H. Wu and J. Zhang, Cr³⁺-doped broadband NIR garnet phosphor with enhanced luminescence and its application in NIR spectroscopy, Adv. Opt. Mater., 2019, 7(12), 1900185.
4. H. Jiang, L. Chen, X. Wu, Z. Luo, R. Li, Y. Liu, Z. Liu, P. Sun, W. You and J. Jiang, Spectral Regulation and Efficiency Optimization in Cr³⁺-Doped Gadolinium Aluminum Gallium Garnet Near-Infrared Ceramic Phosphors via Crystal-Field Engineering, Adv. Mater. Technol., 2022, 7(12), 2200519.
5. X. Zou, X. Wang, H. Zhang, Y. Kang, X. Yang, X. Zhang, M. S. Molokeev and B. Lei, A highly efficient and suitable spectral profile Cr³⁺-doped garnet near-infrared emitting phosphor for regulating photomorphogenesis of plants, Chem. Eng. J., 2022, 428, 132003.
6. T. Yang, H. Jiang, O. Hai, Y. Dong, S. Liu and S. Gao, Effect of Oxygen Vacancies on the Persistent Luminescence of Y₃Al₂Ga₃O₁₂:Ce³⁺,Yb³⁺ Phosphors, Inorg. Chem., 2021, 60(23), 17797–17809.
7. G. Liu and Z. Xia, Modulation of Thermally Stable Photoluminescence in Cr³⁺-Based Near-Infrared Phosphors, J. Phys. Chem. Lett., 2022, 13(22), 5001–5008.
8. M.-H. Chan, W.-T. Huang, K.-C. Chen, T.-Y. Su, Y.-C. Chan, M. Hsiao and R.-S. Liu, The optical research progress of nanophosphors composed of transition elements in the fourth period of near-infrared windows I and II for deep-tissue theranostics, Nanoscale, 2022, 14(19), 7123–7136.
9. H. Xiao, J. Zhang, L. Zhang, H. Wu, H. Wu, G. Pan, F. Liu and J. Zhang, Cr³⁺ activated garnet phosphor with efficient blue to far-red conversion for pc-LED, Adv. Opt. Mater., 2021, 9(20), 2101134.
10. D. Huang, S. Liang, D. Chen, J. Hu, K. Xu and H. Zhu, An efficient garnet-structured Na₃Al₂Li₃F1₂:Cr³⁺ phosphor with excellent photoluminescence thermal stability for near-infrared LEDs, Chem. Eng. J., 2021, 426, 131332.
11. H. Lin, G. Bai, T. Yu, M. K. Tsang, Q. Zhang and J. Hao, Site occupancy and near-infrared luminescence in Ca₃Ga₂Ge₃O₁₂:Cr³⁺ persistent phosphor, Adv. Opt. Mater., 2017, 5(18), 1700227.
12. Y. Zou, C. Hu, S. Lv, Y. Shao, B. Teng, F. You, H. Xu and D. Zhong, Realization of Broadband Near-Infrared Emission with High Thermal Stability in YGa₃(BO₃)₄:Cr³⁺ Borate Phosphor, Inorg. Chem., 2023, 62(48), 19507–19515.
13. J. Wang, Z. Lyu, G. Peng, D. Sun, S. Shen, Z. Lu, L. Wang, H. Zhao, S. Wei and H. You, Cr³⁺-Doped NIR Phosphor with Well-Defined Stokes/Anti-Stokes Phonon Sidebands for a Ratiometric Thermometer and Thermally Stable NIR LEDs, Chem. Mater., 2023, 35(10), 3968–3975.
14. D. Chen, Z. Wan, Y. Zhou, X. Zhou, Y. Yu, J. Zhong, M. Ding and Z. Ji, Dual-Phase Glass Ceramic: Structure, Dual-Modal Luminescence, and Temperature Sensing Behaviors, ACS Appl. Mater. Interfaces, 2015, 7(34), 19484–19493.
15. S. Ding, P. Feng, J. Cao, X. Ma and Y. Wang, Multiple coordination of chromium ion luminescence: a strategy for designing ultra-broadband NIR long persistent luminescent materials, ACS Appl. Mater. Interfaces, 2022, 14(39), 44622–44631.
16. L. Yao, Q. Shao, M. Shi, T. Shang, Y. Dong, C. Liang, J. He and J. Jiang, Efficient ultra-broadband Ga₄GeO₈:Cr³⁺ phosphors with tunable peak wavelengths from 835 to 980 nm for NIR pc-LED application, Adv. Opt. Mater., 2022, 10(4), 2102229.
17. Q. Zhang, X. Wei, J. Zhou, B. Milićević, L. Lin, J. Huo, J. Li, H. Ni and Z. Xia, Thermal stability improvement of Cr³⁺-activated broadband near-infrared phosphors via state population optimization, Adv. Opt. Mater., 2023, 11(14), 2300310.
18. F. Xiao, C. Xie, S. Xie, R. Yi, H. Yuan and H. Zhou, Multi-site Cr³⁺ occupation-related broadband NIR luminescence in Cr³⁺-doped Li₃Mg₂NbO₆, CrystEngComm, 2021, 23(33), 5585–5594.
19. K. Sudarshan, A. Balhara, M. S. Kambli, R. T. Parayil, S. M. Patil and S. K. Gupta, Multiple local site Cr³⁺ emission leading to broadband NIR light in novel SrGa₂O₄ phosphors, Mater. Res. Bull., 2024, 169, 112537.
20. F. Zhao, Z. Song and Q. Liu, Advances in chromium-activated phosphors for near-infrared light sources, Laser Photonics Rev., 2022, 16(11), 2200380.
21. L. Huang, L. Lin, W. Xie, Z. Qiu, H. Ni, H. Liang, Q. Tang, L. Cao, J.-X. Meng and F. Li, Near-infrared persistent luminescence in a Cr³⁺-doped perovskite for low-irradiance imaging, Chem. Mater., 2020, 32(13), 5579–5588.
22. O. Taktak, H. Souissi and S. Kammoun, Optical properties of the phosphors Zn₂SnO₄:Cr³⁺ with near-infrared long-persistence phosphorescence for bio-imaging applications, J. Lumin., 2020, 228, 117563.
23. L. Jiang, X. Jiang, C. Wang, P. Liu, Y. Zhang, G. Lv, T. Lookman and Y. Su, Rapid discovery of efficient long-wavelength emission garnet: Cr NIR phosphors via multi-objective optimization, ACS Appl. Mater. Interfaces, 2022, 14(46), 52124–52133.
24. Y. Wang, Z. Wang, G. Wei, Y. Yang, S. He, J. Li, Y. Shi, R. Li, J. Zhang and P. Li, Ultra-Broadband and high efficiency Near-Infrared Gd₃Zn_xGa_5−2xGe_xO₁₂:Cr³⁺ (x = 0–2.0) garnet phosphors via crystal field engineering, Chem. Eng. J., 2022, 437, 135346.
25. C. Cai, S. Liu, F. Zhao, H. Cai, Z. Song and Q. Liu, Inducing octahedral distortion to enhance NIR emission in Cr-doped garnet Ca₃(Al,Sc)₂Ge₃O₁₂, J. Mater. Chem. C, 2023, 11(25), 8462–8469.
26. Z. Jia, C. Yuan, Y. Liu, X.-J. Wang, P. Sun, L. Wang, H. Jiang and J. Jiang, Strategies to approach high performance in Cr³⁺-doped phosphors for high-power NIR-LED light sources, Light: Sci. Appl., 2020, 9(1), 86.
27. Q. Zhang, G. Li, P. Dang, D. Liu, D. Huang, H. Lian and J. Lin, Enhancing and tuning broadband near-infrared (NIR) photoluminescence properties in Cr³⁺-doped Ca₂YHf₂Al₃O₁₂ garnet phosphors via Ce³⁺/Yb³⁺-codoping for LED applications, J. Mater. Chem. C, 2021, 9(14), 4815–4824.
28. Y. Wang, M. Shang, Y. Sun, M. Deng, X. Xing, P. Dang and J. Lin, Highly efficient and thermally stable broadband near-infrared emitting garnet Ca₃Sc₂Ge₃O₁₂:Cr³⁺,Ce3+ phosphors for multiple pc-LED applications, J. Mater. Chem. C, 2023, 11(27), 9291–9299.
29. H. Li, J. Jiao, X. Xiang, J. Wu, W. Hu, J. Xie, S. Huang, H. Zhang and J. Zhu, Directly Identifying Multiple Cr³⁺ Emitting Centers for Broad Near-Infrared Emission in an Efficient and Near-Zero Thermal Quenching Garnet-Type Phosphor, Adv. Opt. Mater., 2024, 2302391.
30. C. Li and J. Zhong, Highly efficient broadband near-infrared luminescence with zero-thermal-quenching in garnet Y₃In₂Ga₃O₁₂:Cr³⁺ phosphors, Chem. Mater., 2022, 34(18), 8418–8426.
31. V. Boiko, Z. Dai, M. Markowska, C. Leonelli, C. Mortalò, F. Armetta, F. Ursi, G. Nasillo, M. L. Saladino and D. Hreniak, Particle size-related limitations of persistent phosphors based on the doped Y₃Al₂Ga₃O₁₂ system, Sci. Rep., 2021, 11(1), 141.
32. Q. Du, J. Ueda and S. Tanabe, Toward color variation of long persistent luminescence in Pr³⁺-doped garnet transparent ceramic phosphors, J. Mater. Chem. C, 2023, 11(46), 16225–16233.
33. M. Qiang, Y. Tang, S. Ye, W. Lou, J. Chao, H. Lin and D. Zhang, Novel YAGG:Cr garnet phosphor ceramic with broadband near infrared (NIR) emission and luminescence thermal stability for plant growth, Ceram. Int., 2024, 50(3), 4823–4830.
34. J. Ueda, S. Miyano and S. Tanabe, Formation of deep electron traps by Yb³⁺ codoping leads to super-long persistent luminescence in Ce³⁺-doped yttrium aluminum gallium garnet phosphors, ACS Appl. Mater. Interfaces, 2018, 10(24), 20652–20660.
35. J. Xu, J. Ueda, Y. Zhuang, B. Viana and S. Tanabe, Y₃Al_5−xGa_xO₁₂:Cr³⁺: A novel red persistent phosphor with high brightness, Appl. Phys. Express, 2015, 8(4), 042602.
36. V. Laguta, Y. Zorenko, V. Gorbenko, A. Iskaliyeva, Y. Zagorodniy, O. Sidletskiy, P. Bilski, A. Twardak and M. Nikl, Aluminum and Gallium Substitution in Yttrium and Lutetium Aluminum–Gallium Garnets: Investigation by Single-Crystal NMR and TSL Methods, J. Phys. Chem. C, 2016, 120(42), 24400–24408.
37. V. Boiko, M. L. Saladino, F. Armetta, F. Ursi, M. Markowska, K. Grzeszkiewicz, C. Mortalò, C. Leonelli and D. Hreniak, Urea Glass Route as a Way to Optimize YAGG:Ce³⁺,Cr³⁺,Pr³⁺ Nanocrystals for Persistent Luminescence Applications, Langmuir, 2022, 38(38), 11539–11549.
38. E. Arroyo, B. T. Herrero, J. M. de la Fuente, M. Ocaña and A. I. Becerro, Highly uniform Y₃Al₂Ga₃O₁₂-based nanophosphors for persistent luminescence bioimaging in the visible and NIR regions, Inorg. Chem. Front., 2022, 9(11), 2454–2461.
39. Z. Dai, V. Boiko, K. Grzeszkiewicz, M. Markowska, F. Ursi, J. Hölsä, M. L. Saladino and D. Hreniak, Effect of annealing temperature on persistent luminescence of Y₃Al₂Ga₃O₁₂:Cr³⁺ co-doped with Ce³⁺ and Pr³⁺, Opt. Mater., 2021, 111, 110522.
40. F. Farges, Chromium speciation in oxide-type compounds: application to minerals, gems, aqueous solutions and silicate glasses, Phys. Chem. Miner., 2009, 36, 463–481.
41. A. M. Beale, D. Grandjean, J. Kornatowski, P. Glatzel, F. M. de Groot and B. M. Weckhuysen, Unusual coordination behavior of Cr³⁺ in microporous aluminophosphates, J. Phys. Chem. B, 2006, 110(2), 716–722.
42. A. Pantelouris, H. Modrow, M. Pantelouris, J. Hormes and D. Reinen, The influence of coordination geometry and valency on the K-edge absorption near edge spectra of selected chromium compounds, Chem. Phys., 2004, 300(1–3), 13–22.
43. M. L. Peterson, G. E. Brown Jr and G. A. Parks, Direct XAFS evidence for heterogeneous redox reaction at the aqueous chromium/magnetite interface, Colloids Surf., A, 1996, 107, 77–88.
44. R. D. Shannon, Revised effective ionic radii and systematic studies of interatomic distances in halides and chalcogenides, Acta Crystallogr., Sect. A: Cryst. Phys., Diffr., Theor. Gen. Crystallogr., 1976, 32(5), 751–767.
45. B. B. Srivastava, S. K. Gupta and Y. Mao, Remarkable enhancement of photoluminescence and persistent luminescence of NIR emitting ZnGa₂O₄:Cr³⁺ nanoparticles, CrystEngComm, 2020, 22(14), 2491–2501.
46. B. B. Srivastava, S. K. Gupta, S. Mohan and Y. Mao, Molten-Salt-Assisted Annealing for Making Colloidal ZnGa₂O₄:Cr Nanocrystals with High Persistent Luminescence, Chem. – Eur. J., 2021, 27(44), 11398–11405.
47. Y. Zhang, S. Miao, Y. Liang, C. Liang, D. Chen, X. Shan, K. Sun and X.-J. Wang, Blue LED-pumped intense short-wave infrared luminescence based on Cr³⁺–Yb³⁺-co-doped phosphors, Light: Sci. Appl., 2022, 11(1), 136.
48. Y. Wang, Z. Wang, G. Wei, Y. Yang, S. He, J. Li, Y. Shi, R. Li, J. Zhang and P. Li, Highly efficient and stable near-infrared broadband garnet phosphor for multifunctional phosphor-converted light-emitting diodes, Adv. Opt. Mater., 2022, 10(11), 2200415.
49. W. Nie, Y. Li, J. Zuo, Y. Kong, W. Zou, G. Chen, J. Peng, F. Du, L. Han and X. Ye, Cr³⁺-activated Na₃X₂Li₃F₁₂ (X = Al, Ga, or In) garnet phosphors with broadband NIR emission and high luminescence efficiency for potential biomedical application, J. Mater. Chem. C, 2021, 9(42), 15230–15241.
50. X. Dai, X. Zou, H. Zhang, W. Chen, C. Yang, M. S. Molokeev, Z. Xia, Y. Liu, X. Zhang and M. Zheng, Novel Cr³⁺-Doped Garnet Phosphor with Broadband Efficient Far-Red Emission for Photochrome Matching Plant-Lighting, Adv. Opt. Mater., 2023, 2302380.
51. A. Balhara, S. K. Gupta, M. Abraham, B. Modak, S. Das, C. Nayak, H. V. Annadata and M. Tyagi, Trap engineering through chemical doping for ultralong X-ray persistent luminescence and anti-thermal quenching in Zn₂GeO₄, J. Mater. Chem. C, 2024, 12(5), 1728–1745.
52. K. Elzbieciak-Piecka and L. Marciniak, Optical heating and luminescence thermometry combined in a Cr³⁺-doped YAl₃(BO₃)₄, Sci. Rep., 2022, 12(1), 16364.
53. P. Husband, I. Bartošová, V. Slugeň and F. Selim, In Positron annihilation in transparent ceramics, J. Phys.: Conf. Ser., 2016, 012013 , IOP Publishing.
54. F. Selim, C. Varney, M. Tarun, M. Rowe, G. Collins and M. McCluskey, Positron lifetime measurements of hydrogen passivation of cation vacancies in yttrium aluminum oxide garnets, Phys. Rev. B: Condens. Matter Mater. Phys., 2013, 88(17), 174102.
55. A. Marinopoulos, First-principles study of the formation energies and positron lifetimes of vacancies in the Yttrium-Aluminum Garnet Y₃Al₅O₁₂, Eur. Phys. J. B, 2019, 92, 1–9.
56. F. Selim, D. Winarski, C. Varney, M. Tarun, J. Ji and M. McCluskey, Generation and characterization of point defects in SrTiO₃ and Y₃Al₅O₁₂, Results Phys., 2015, 5, 28–31.
57. Z. Li, B. Zhong, Y. Cao, S. Zhang, Y. Lv, Z. Mu, Z. Hu and Y. Hu, Energy transfer and tunable luminescence properties in Y₃Al₂Ga₃O₁₂:Tb³⁺,Eu³⁺ phosphors, J. Alloys Compd., 2019, 787, 672–682.
58. C. Linderälv, D. Åberg and P. Erhart, Luminescence quenching via deep defect states: A recombination pathway via oxygen vacancies in Ce-doped YAG, Chem. Mater., 2020, 33(1), 73–80.
59. J. H. Yoder and M. Keith, Complete substitution of aluminum for silicon: The system 3MnO· Al₂O₃·3SiO₂—3Y₂O₃·5Al₂O₃, Am. Mineral., 1951, 36(7–8), 519–533.
60. J. Xu, D. Murata, J. Ueda, B. Viana and S. Tanabe, Toward rechargeable persistent luminescence for the first and third biological windows via persistent energy transfer and electron trap redistribution, Inorg. Chem., 2018, 57(9), 5194–5203.
61. A. Togo and I. Tanaka, First principles phonon calculations in materials science, Scr. Mater., 2015, 108, 1–5.

---

Footnote

† Electronic supplementary information (ESI) available: Characterization techniques, EXAFS experimental details, computational details, EDS spectra, XANES and Fourier transform EXAFS spectra at Y and Ga K-edges, tables showing EXAFS results with bond lengths, coordination number and disorder parameter values at Y and Ga K-edges, PL emission spectra at 80 K, quantum yield spectra, EPR spectra at 100 K, emission decay curves and lifetime values, PALS lifetime plots, table showing defect formation energies, density of states plots, defect level diagram, phonon dispersion plot, XPS spectrum, and elemental mapping and PL of thin film. See DOI: https://doi.org/10.1039/d4tc01530d

---