Regulating the trap distribution of ZnGa₂O₄:Cr³⁺ by Li⁺/Ga³⁺ doping for upconversion-like trap energy transfer NIR persistent luminescence

Abstract

ZnGa₂O₄:Cr³⁺ persistent luminescent phosphors (PLPs) have been widely applied in bioimaging and photonics due to their ultra-long near-infrared (NIR) afterglow. However, UV and visible excitation currently in use have shallow penetration depths or harmful effects on organisms, which limit long-term bioimaging. Therefore, developing NIR PLPs excited by NIR light is urgent for bioimaging. Here, Zn_1−x(Li/Ga)_xGa₂O₄:Cr³⁺ (x = 0–1) NIR PLPs were synthesized. All the newly introduced Ga³⁺ ions occupy the tetrahedral sites. However, with increasing Li⁺/Ga³⁺ content, Li⁺ ions first occupy the tetrahedral position, then partially enter octahedral sites, and completely occupy the octahedral sites at x = 1. The incorporation of Li⁺/Ga³⁺ contributes to weakened crystal field strength, which leads to a deeper trap depth and a wider trap energy level. Complete replacement of Zn²⁺ with Li⁺/Ga³⁺ ions leads to the splitting of the trap energy level into two-divided ones, which reduces the electron transfer between deep/shallow traps and makes the deep trap energy level come close to the ²E energy level of Cr³⁺. Therefore, an enhanced NIR afterglow excited by the low-energy NIR light is found for the Li⁺/Ga³⁺ doped sample. This work provides a new category for NIR-absorptive-NIR-emissive PLPs and proposes a new phosphor for long-term bioimaging.

---

1. Introduction

In recent decades, near-infrared (NIR) persistent luminescent phosphors (PLPs) have shown broad application prospects in the field of biological imaging mainly due to their two advantages: (1) the avoidance of self-fluorescence from biological tissue due to excitation in vitro and (2) the high spatial resolution, deep penetration depth, and low optical absorption of NIR optical signals through the organism. The trivalent chromium ion (Cr³⁺) is a near-infrared red doping ion because of its emission range from 650 to 1600 nm.^1–3 Based on the above reasons, spinel long persistent luminescent materials of Cr³⁺-doped ZnGa₂O₄, Zn₃Ga₂Ge₂O₁₀, and other similar gallate hosts have become the research hotspots in the field of luminescence and biological imaging due to their near-infrared emission and long afterglow characteristics.^4–6 Subsequently, a variety of functionalized zinc gallates or zinc gallogermanate nanoparticles with NIR long afterglow were applied in biological imaging, life sciences, and biomedicine.^7–9

The research on the spinel-structure NIR persistent luminescence materials mainly focuses on two aspects. 1. Using monodisperse luminescent particles as the fluorescent probes for different biological applications, including live animal imaging, drug delivery, tumor imaging, etc. 2. Regulating the crystal structure for improved persistent luminescence, such as co-doping with Cr³⁺/Ge⁴⁺/Sn⁴⁺, Cr³⁺/Mg²⁺/Ge⁴⁺, Cr³⁺/Zn²⁺/Sn⁴⁺, Cr³⁺/Li⁺, Cr³⁺/Pr³⁺, Cr³⁺/Al³⁺/Bi³⁺, and Cr³⁺/Gd³⁺/Sn⁴⁺ in zinc gallate.^10–16 ZnGa₂O₄:Cr³⁺ belongs to the cubic spinel structure, which contains 56 atoms in F-d3̄m. In ZnGa₂O₄, Zn²⁺ ions occupy the 8a sites (CN = 4, CN: coordination number) and Ga³⁺ ions occupy the 16d sites (CN = 6).^2,10,11,17 For ZnGa₂O₄:Cr³⁺, Cr³⁺ would likely replace Ga³⁺ in the octahedron site, because Cr³⁺ and Ga³⁺ ions have similar ionic radii (R_Cr = 0.63 Å; R_Ga = 0.62 Å) in the octahedron.

Cr³⁺ ions are doped in spinel-structured gallates as emitters, and the crystal field and defect regulation are crucial for the improvement of photoluminescence and persistent luminescence. The optical absorption and luminescence of the Cr³⁺ ions can be remarkably influenced by host materials because the energy levels of the d electrons of Cr³⁺ ions are dependent on the symmetry of oxygen coordination and crystal field strength.^3,18,19 Cr³⁺ ions can exhibit a broadband emission in the ∼650–1600 nm range (the spin-allowed ⁴T₂ → ⁴A₂ transition) and a narrow-band emission at ∼700 nm (the spin forbidden ²E → ⁴A₂ transition) in a weak and strong crystal field, respectively. However, the ²E and ⁴T₂ levels are overlapped in the intermediate crystal field, leading to the coincidence of these two transitions from ⁴T₂/²E to ⁴A₂. Recently, the investigation of the influence of the crystal field on luminescence by co-doping different ions have been widely reported. However, investigations on the effect of the crystal field on electron traps and persistent luminescence are rather limited.

The afterglow behavior has been attributed to the Cr³⁺ ion and nearby anti-site defect interactions in the spinel ZnGa₂O₄ based on the currently accepted model. Research showed that traps play an important role as active centers in persistent phosphors of ZnGa₂O₄:Cr³⁺. Changing the Zn ratio or incorporating other metal cations is an effective way to change the trap distribution and depth of phosphors, and further improve their persistent luminescence. However, the exact effect of trap depths and distributions on persistent luminescence has not been defined.^14,20–22 On one hand, the slowly released electrons from the deep traps influence the afterglow of the persistent phosphor, and the deeper traps result in a slower release of electrons from traps, which contributes to a longer afterglow.^22–24 On the other hand, higher density and larger number of traps may contribute to a stronger afterglow.^22,25 Moreover, some studies indicate that the trap distribution also affects the light storage of various energies or various wavelengths.^19,26,27 In general, low-energy excitation cannot excite the electrons to the traps. However, continuously distributed traps result in up-convertible-like trap energy storage and electron transfer between deep traps (DTs) and shallow traps (STs), which facilitates biological imaging of in situ long-wavelength excitations.²⁶ However, deep traps and shallow traps are continuously distributed for most materials, which leads to rapid energy transfer between traps and accelerates the release of electrons from deep traps. Therefore, the reconstruction and segmentation of trap distribution may regulate the electron transfer between DTs and STs and thus improve the near-infrared persistent luminescence of phosphors.

Herein, a series of Zn_1−x(Li/Ga)_xGa₂O₄:Cr³⁺ (x = 0–1) NIR PLPs were synthesized using a solid-state reaction. As the Li⁺/Ga³⁺ content increases, the crystal field strength of phosphors decreases, which leads to a redshift of excitation and emission spectra. The crystal field also affects the trap depth and distribution. A complete replacement of Zn²⁺ with Li⁺/Ga³⁺ ions leads to the splitting of the trap energy level, which reduces the electron transfer between DTs and STs, but enhances the NIR afterglow by low-energy excitation. The outcome of this work provides a new category for persistent luminescence nanoparticles with a NIR-absorptive-NIR-emissive feature and a new application in biological imaging.

2. Experimental section

All the experimental methods are described in detail in the ESI.†

3. Results and discussion

3.1 Synthesis and local structure of Zn_1−x(Li/Ga)_xGa₂O₄:0.005Cr³⁺

A series of Zn_1−x(Li/Ga)_xGa₂O₄:0.005Cr³⁺ (x = 0–1) powders were synthesized and XRD analysis was performed to analyze the crystal structure and phase transition in Fig. 1. The measured XRD patterns of x = 0–0.8 indicate a crystalline structure corresponding to the reference pattern for ZnGa₂O₄ (ZGO) of the spinel structure (JCPDS file no. 38-1240). However, the LiGa₅O₈ (LGO) cubic phase (JCPDS file no. 38-1371) with new diffraction peaks appears only at a larger x value of 1. The reference patterns of LiGa₅O₈ (JCPDS file no. 38-1371) correspond to a spinel structure with the space group of P4₃32, a derivative of the Fd3̄m space group of the spinel structure. Furthermore, the (311) main diffraction of spinel gradually shifts to the bigger angle side with increasing x value, and the lattice constant of the powders decreases from a = 8.338 to 8.203 Å with the value of x increasing from 0 to 1 (Table S1†), indicating that the unit cell is shrunk by substitution of Zn²⁺ with the Li⁺/Ga³⁺ pair. For convenient description, Zn_1−x(Li/Ga)_xGa₂O₄:0.005Cr³⁺ and Zn_1−x(Li/Ga)_xGa₂O₄ are denoted as ZLGGC and ZLGGO hereafter.

Fig. 1 XRD patterns of Zn_1−x(Li/Ga)_xGa₂O₄:0.005Cr³⁺ (x = 0–1) calcined at 1300 °C.

It is known that, in the structure of ZGO spinel, Ga³⁺ ions occupy the 16d sites, Zn²⁺ ions occupy the 8a sites, and O occupies the 32e sites. In this work, co-substitution of Li⁺/Ga³⁺ for Zn²⁺ results in a smaller lattice constant. The complete replacement of Zn²⁺ with Li⁺/Ga³⁺ ions leads to the LGO spinel structure: the Ga³⁺ ions occupy both the 8c (CN = 4) and 12d sites (CN = 6), and the Li⁺ ions occupy only the 4b sites (CN = 6). Four oxygens are surrounding the tetrahedral site, and six oxygens surround the octahedral site.^17,28,29 It is noticed that Li⁺, Zn²⁺, and Ga³⁺ ions have different radii in different coordination environments: Li⁺ (r^IV = 0.59 Å, r^VI = 0.67 Å), Zn²⁺ (r^IV = 0.60 Å, r^VI = 0.74 Å), and Ga³⁺ (r^IV = 0.47 Å, r^VI = 0.62 Å). Referring to the ionic radii, Li⁺/Ga³⁺ ions preferably occupy Zn²⁺ sites, but the specific occupations need to be determined by a variety of structural detection methods.^11,22,30

Different configurations were built for formation energy calculation to explore the site preference of Li⁺/Ga³⁺ doping in ZnGa₂O₄. There are two conditions for Li/Ga doping in spinel ZnGa₂O₄: (1) Li and Ga occupy the tetrahedral sites and (2) Li occupies the octahedral sites and Ga occupies the tetrahedral sites, which are defined as Li4–Ga4 and Li8–Ga4, respectively. Fig. 2 demonstrates the unit cell's density functional theory (DFT) total energy with different Li/Ga content doping. As shown in Fig. 2, two, seven, seven, and four crystal structure models are created for the tetrahedral positions occupied by Li and Ga (Li4–Ga4) with 25%, 50%, 75%, and 100% Li/Ga content, respectively. There are 5, 66, 106, and 22 crystal structure models created for calculating the DFT total energy of Li in the octahedron and Ga in the tetrahedron (Li4–Ga8) with 25%, 50%, 75%, and 100% Li/Ga content, respectively. The energy difference (ΔE_d) of conditions for Li/Ga doping was calculated as follows:^22,31,32

ΔE_d = E(Li4–Ga4) − E(Li8–Ga4) (1)

where E(Li4–Ga4) is the unit cell's lowest DFT total energy with Li and Ga occupying the tetrahedral sites, E(Li8–Ga4) is the unit cell's lowest DFT total energy with Li in the octahedron and Ga in the tetrahedron. All the calculated results are demonstrated in Table 1. When the content of Li⁺/Ga³⁺ is not more than 75%, the energy differences are calculated as 0.4 (25%), 0.187 (50%), and 0.074 (75%) eV, which indicates that the most likely placeholder in ZLGGO is Li and Ga both in tetrahedral sites. However, the energy difference decreases with increasing Li/Ga content. When the Li/Ga content reaches 100%, the energy difference was calculated to be −0.612 eV, which indicates that the most likely placeholder in the x = 1 sample is Li in octahedral and Ga both in octahedral and tetrahedral sites. According to the calculation results, the lowest energy configurations with different contents of Li/Ga doped in spinel ZnGa₂O₄ are shown in Fig. 3.

Fig. 2 The DFT total energy of the unit cell with different doping contents of Li/Ga: (a) 25%, (b) 50%, (c) 75%, and (d) 100%.

Fig. 3 56-Atom unit cell crystal structure with different contents of Li/Ga doped in spinel ZnGa₂O₄ under conditions of the lowest energy: (a) 25%, (b) 50%, (c) 75%, and (d) 100%.

Table 1 The calculated energy difference (ΔE_d) with various Li/Ga doping contents

Li/Ga content	E(Li4–Ga4) (eV)	E(Li8–Ga4) (eV)	ΔEf (eV)
25%	−317.043	−316.643	0.4
50%	−321.396	−321.209	0.187
75%	−325.750	−325.676	0.074
100%	−330.109	−330.721	−0.612

---

To explore the occupations of Li⁺ and Ga³⁺ ions in tetrahedral and octahedral sites, the ⁷Li and ⁷¹Ga solid-state MAS NMR spectra of ZLGGO powder were obtained (Fig. S1† and Fig. 4a). With increasing Li⁺/Ga³⁺ concentration, the peaks in ⁷Li NMR spectra exhibit a high chemical shift, which means the Li⁺ ions move from the tetrahedron to octahedron. There are interactions between electric field gradients and nucleus quadrupolar moment for ⁷¹Ga, a half-integer quadrupolar nucleus (I = 3/2), correspondingly which leads to the broadening of the NMR spectra.^10,33 Fast sample spinning frequencies and high magnetic fields are required for the resolution of the corresponding resonances in MAS spectra.¹⁰ The coordination numbers and local symmetries significantly influence the shifts of the ⁷¹Ga MAS NMR line of the central transition.^33,34 The range of Ga in an octahedral or tetrahedral environment will be from 80 to 42 ppm and 107 to 222 ppm, respectively.^10,34–36 Solid nuclear magnetic field tested at high speed will produce rotating sidebands, which is the peak caused by the rotor rotating at high speed in the probe. The sideband peaks are easy to be identified because they are symmetric relative to the main peak, and the distance from the sideband peak to the main peak is the frequency of rotation. As shown in Fig. 4a, the peak of Ga^IV overlaps with the sideband, which makes it difficult to directly quantify the content of Ga in the octahedron. The peak integral area ratio in NMR spectra is used to quantitatively analyze the content of Ga³⁺ in the tetrahedron and octahedron. The areas of ∼−110–∼15 ppm, ∼15–∼90 ppm, and ∼90–∼230 ppm are defined as Regin1 (R1), Regin2 (R2), and Regin3 (R3), respectively. The quantification of the Ga^IV/Ga^VI proportion can be determined from the three areas of signal integration using the following formula: (R3–R1)/R2 (Fig. S2†). As shown in Table 2, the quantification of Ga^IV/Ga^VI in ZGO determined from ⁷¹Ga NMR spectra were of ∼1.7%, ∼5%, ∼21%, and ∼66% for the x = 0, 0.2, 0.6, and 1 samples, respectively. The Ga^IV/Ga^VI ratios of x = 0 and x = 1 samples are 1.7% and 66%, which agree with the spinel inversion in ZGO and the site allocation in LGO reported in previous results.^10,11,28,36 The Ga^IV/Ga^VI ratio of ZLGGO measured from the MAS NMR spectrum increases from 1.7% to 66% with the Li⁺/Ga³⁺ content increasing from 0 to 100%. This indicates that more Ga³⁺ ions and Li⁺ ions occupy the tetrahedron and octahedron respectively at a higher x value.

Fig. 4 (a) ⁷¹Ga MAS NMR spectra, (b) Raman spectra of ZLGGO, and (c) schematic illustration of Li⁺/Ga³⁺ in tunable site occupations.

Table 2 The quantification of Ga^IV/Ga^VI proportion

x	R1	R2	R3	Ga^IV/Ga^VI
0	6.09 × 10^9	1.36 × 10^10	6.52 × 10^9	3.2%
0.2	8.41 × 10^9	1.44 × 10^10	9.14 × 10^9	5.1%
0.6	3.17 × 10^9	1.57 × 10^10	6.48 × 10^9	21.1%
1	3.61 × 10^8	5.48 × 10^9	3.98 × 10^9	66.1%

---

Raman spectroscopy was performed to further reveal the changes in the structure in Fig. 4b. Cubic spinel contains 42 vibrational modes in its group theory. The following modes of spinels are predicted by group theory analysis at the Γ point:^18,37–40

where most modes are silent or acoustic modes except for the indicated infrared-active and Raman-active modes. There are only four infrared active modes and five Raman active modes in normal non-defective spinels. In general, the mode frequencies follow the sequence: T_2g < E_g < T_2g < T_2g < A_1g. The spinel ZnGa₂O₄ first-order Raman active mode is due to the tetrahedral site Zn²⁺ ions, but has nothing to do with the octahedral site cations, suggesting that the ZnO₄ group is the only contributor to the first-order Raman mode. In AB₂O₄, the highest frequency T_2g Raman-active mode is assigned to the AO₄ tetrahedron symmetric stretching vibration, the remaining T_2g mode is assigned to the BO₆ octahedron translation against the A cation, and the E_g and A_1g Raman-active modes correspond to the symmetric bending motion of the AO₄ unit oxygens and the asymmetric bending motion of the A bonded oxygens, respectively.³⁷ For the x = 0 sample, the Raman spectrum peaks at ∼453, ∼609, and ∼713 cm⁻¹, correspond to T_2g, T_2g, and A_1g modes (k = 0), respectively. The number of Raman-active modes is related to the material's crystal symmetry. When the crystal transforms to a lower symmetry, the splitting of degenerate vibration modes or a new Raman mode can be seen. The peaks at ∼713 and ∼609 cm⁻¹ correspond to the asymmetric bending motion of the Zn bonded oxygens in ZnGa₂O₄ and the symmetric stretching vibration of the ZnO₄ tetrahedron, respectively (Fig. 4b). They gradually weaken and then disappear with increasing Li⁺/Ga³⁺ concentration. Incorporating Li⁺/Ga³⁺ in the lattice reduces the intensity of Raman bands, hinting at the onset of disorder of cations among the octahedral and tetrahedral sites. Meanwhile, Li⁺/Ga³⁺ doping leads to the splitting of T_2g and A_1g vibration modes at ∼648 and ∼734 cm⁻¹, indicating that the structure transforms into a lower symmetry. Because the low-frequency Raman-active mode of T_2g is the GaO₆ octahedron against the incorporation of Li⁺ cations, it becomes stronger with increasing Li⁺/Ga³⁺ concentration. For the x = 1 sample, a band near 655 cm⁻¹ corresponds to the bending and stretching vibrations of GaO₄. The translational motion of the GaO₄ tetrahedron generates a band in the lower wavenumber region.^39,41 The FWHM of XRD patterns and Raman spectrum are demonstrated in Fig. S3,† which can prove the orderliness of samples. The x = 1 sample shows a narrow FWHM (full width at half maximum) of Raman spectra, which demonstrates it owns an ordered structure. The DFT calculation results, NMR spectra, and Raman spectra show that the newly introduced Ga³⁺ ions tend to occupy the tetrahedral sites, but the Li⁺ ion occupation is dependent on the content. The tetrahedral sites are occupied by Li⁺ ions at a low content with x ≤ 0.4, and they partially enter octahedral sites at a higher content with x = 0.6. When x reaches 1, Li ions completely occupy the octahedral sites. The increase of the x value from 0 to 1 makes the crystal structure experience a process from order to disorder and back to order (Fig. 4c).

Rietveld refinement of the XRD pattern was conducted for the Zn_1−x(Li/Ga)_xGa₂O₄ powders based on the above structural analysis to investigate the effect of Li⁺/Ga³⁺ substitution on the crystal structure and phase composition of the spinel phase (Fig. 5a–d). The consistency between diffraction peaks and standard data card suggests that these samples are pure phase and have a structure evolution. The low values of the residual factors (R_p, R_wp, and χ²) of the samples suggest that the results are credible (Tables S2–S6†). Incorporation of Li⁺–Ga³⁺ to replace Zn²⁺–Zn²⁺ in ZnGa₂O₄ induces a shrink of the unit cell due to the smaller ionic radii sum of the Li⁺–Ga³⁺ pair than that of Zn²⁺–Zn²⁺ pair in ZnGa₂O₄, further resulting in the shift of the diffraction peaks to the higher angle side (Fig. 1). The cell volumes, lattice parameters a = b = c, as well as M–O average bond length (M = Li/Zn/Ga) of MO₄ tetrahedra and MO₆ octahedra for ZGO samples obtained from Rietveld refinement manifest an obvious linear decrease (Table S1† and Fig. 5e), further implying that the unit cell becomes smaller by substitution of Zn²⁺–Zn²⁺ with the Li⁺–Ga³⁺ pair. According to the above analysis, the crystal structure of the polyhedron is drawn in Fig. 5f: the mismatched ionic radius caused by co-doping Li⁺–Ga³⁺ leads to the tetrahedral and octahedral tilting, which could further influence the crystal field of Cr³⁺ in ZLGGC phosphors.^42,43

Fig. 5 The results of Rietveld refinement for XRD patterns of the ZLGGO host with different x values: (a) x = 0, (b) x = 0.2, (c) x = 0.6, and (d) x = 0.1, respectively. (e) Average bond length for the MO₄ tetrahedron and MO₆ octahedron of ZLGGO (x = 0–1). (f) Structural distortion trend with different doping contents.

The morphologies of ZLGGO samples obtained by TEM are shown in Fig. S4a–d,† which are irregular with dispersed micron and submicron particles, similar to those of high temperature solid-state reaction produced samples. The particle size is not affected by co-doping. High-resolution TEM (HR-TEM; Fig. S4e–h†) shows the clear lattice fringe spaces of 2.76, 2.67, 4.89, and 2.47 Å, which is assigned to the (220), (311), (111), and (311) planes for x = 0, 0.2, 0.6 and 1 samples, respectively. Selected-area electron diffraction (SAED) exhibits a set of diffraction spots (Fig. S4i–l†), while the calculated d-spacings of 3.05, 3.02, 2.91 and 2.90 Å can be assigned to the (220) plane of cubic ZLGGO spinel. The results also indicate that the unit cell becomes smaller by substitution of Zn²⁺–Zn²⁺ with the Li⁺–Ga³⁺ pair.

3.2 Photoluminescence and persistent luminescence of ZLGGC

UV-vis reflection spectra can be used to measure the optical absorption bands of Zn_1−x(Li/Ga)_xGa₂O₄:0.005Cr³⁺ phosphors. In the diffuse reflection spectra (Fig. 6a), all the samples show three common absorption broadbands, among which the band located at 200–320 nm corresponds to the ⁴A₂ → ⁴T₁ (⁴P) transition of Cr³⁺ and the band-gap transition (VB → CB), and the bands at ∼410 and ∼570 nm correspond to the ⁴A₂ → ⁴T₁ (⁴F) and ⁴A₂ → ⁴T₂ (⁴F) transitions of Cr³⁺.^11,22 It is worth noting that the ⁴A₂ → ⁴T₂ (⁴F) transition of Cr³⁺ has an obvious redshift after Li⁺/Ga³⁺ doping, illustrating that Li⁺/Ga³⁺ doping gives rise to a weaker crystal field. The diffuse reflectance (R_∞) of ZLGGO was obtained (Fig. S5†) and converted by the Kubelka–Munk function, and it can be used to characterize the band structure in detail:^14,30

F(R_∞) = (1 − R_∞)²/2R_∞ (3)

and the [F(R_∞)hν]²verses hν plots were obtained based on the following function:

[F(R_∞)hν]² = A(hν − E_g) (4)

where E_g and hν are the bandgap energy and incident photon energy, respectively, and A is a constant. The [F(R_∞)hν]² − hν plot of ZLGGO shown in Fig. 6b was used to determine the bandgap energy, and from the intercept of a fitted straight line, the bandgap energy was determined to increase from 4.47 to 4.99 eV with the increasing x value from 0 to 1 (Fig. 6b).

Fig. 6 The band structures of ZLGGO: (a) UV-vis reflection spectra of ZLGGC and (b) [F(R_∞)hν]² − hν plot of ZLGGO. (c–g) Total and partial density of states of ZLGGO: (c) 0%, (d) 25%, (e) 50%, (f) 75%, and (g) 100%. The Fermi energy is taken as the zero energy. (h–l) The electronic band structure of ZLGGO: (h) 0%, (i) 25%, (j) 50%, (k) 75%, and (l) 100%.

The band structure was further evaluated by the calculation of partial and total density of states using density functional theory calculations for ZLGGO (Fig. 6c–g). The results show that only the O orbital levels contribute to the valence band maximum, whereas both the O and Ga orbital levels contribute significantly to the conduction band minimum, consistent with previous reports.^22,30 The calculated bandgap of a perfect ZGO crystal (2.27 eV) is smaller than the experimental value (4.47 eV) (Fig. 6c). Because the generalized gradient approximation (GGA) underestimates the bandgap size, all the calculated band structures using the DFT approach are smaller than the experimental values (Fig. 6c–l).^22,25,31 All the results from DFT calculations indicate that the bandgaps increase (from 2.274 to 2.482 eV) as the content of Li⁺/Ga³⁺ increases (Fig. 6c–l), which matches well with the local structure and UV-vis analysis results.

In chromium doped spinels, lattice disorder and perturbation arising from cation inversion and incorporation of new ions would give rise to varied luminescence of Cr³⁺. The Cr³⁺ ion energy levels are sensitive to coordination and dependent on the strength of the crystal field with 3d³ electronic configuration.^3,44,45 Obtained by monitoring the Cr³⁺ emission peak at 708 nm, the photoluminescence excitation (PLE) spectra (Fig. 7a and Fig. S6a†) demonstrate three strong bands corresponding to the band-gap transition (VB → CB) and the ⁴A₂ → ⁴T₁ (⁴P) transition with a peak at ∼260 nm (the strongest), the ⁴A₂ → ⁴T₁ (⁴F) transition with a peak at ∼409 nm, and the ⁴A₂ → ⁴T₂ (⁴F) transition with a peak at ∼554 nm. The PL spectra of the ZLGGC phosphors recorded under UV-light are shown in Fig. 7b and Fig. S6b.† The intense emission bands in the wavelength range of 650–800 nm locate in the NIR area, which are assigned to the ²E → ⁴A₂ transition of Cr³⁺.^1–4,10,13 The sample Cr³⁺ luminescence is excited preferentially by UV range energies, in which the band-gap transition at about 250 nm corresponds to the transition of electrons from oxygen orbitals to gallium orbitals (VB → CB) (Fig. 7a).^14,46 It can be found that with increasing Li⁺/Ga³⁺ content, the excitation peaks of VB → CB shift to the short-wavelength side, consistent with band gap and band structure analysis. In the PLE spectra, the blue shift in the VB → CB is due to the enlarged E_g for the Li⁺/Ga³⁺ incorporated samples.^17,47,48 However, a different observation was found for the 3d–3d transitions of Cr³⁺. The 3d–3d transitions shift to the long-wavelength side with increasing Li⁺/Ga³⁺ content, which is correlated with the crystal field of host materials.^3,45,49 Under the 260 nm excitation, the samples output strong deep red and NIR emissions (²E → ⁴A₂ transition of Cr³⁺ at ∼700 nm) as demonstrated in Fig. 7b. For ZnGa₂O₄:0.005Cr³⁺, the PL spectra display several narrow lines that indicate that the Cr³⁺ ion is in octahedral coordination with a strong field. Similar to the reported results, the emissions centered at 688 nm (∼1.81 eV) correspond to the R1 and R2 zero phonon lines for the ²E → ⁴A₂ transition of Cr³⁺.^2,50 As previously reported, N2-lines are dependent on the host lattice structure, and they arise from Cr³⁺ ions in the distorted octahedron. In addition, all N-lines are spectroscopic analogs of the R-line and can be interpreted as zero-phonon lines of different luminescence centers. In Fig. 7b, the peak at ∼695 nm corresponds to another type of Cr³⁺ ion with perturbed short-range crystalline order different from the ideal octahedral coordination of the normal spinel, and the peak at 708 nm (14 124 cm⁻¹) has ∼400 cm⁻¹ lower energy than the R line, showing a strong emission.^2,50 According to the report of Mikenda et al., in ZnGa₂O₄, the structure-dependent line N4 for Cr³⁺ lies at ∼400 cm⁻¹ lower energy than R1, so it is quite possible that the 708 nm line is attributed to the N4 line of Cr³⁺.⁵¹ Here, the N2 and N4 lines of Cr³⁺ in ZnGa₂O₄ spinel originate from the Cr³⁺ ions in two kinds of distorted environments, which are around the antisite defects and the defects arising from Li occupying the Zn site, respectively. The R lines peak in the range of 650–750 nm and are accompanied by their anti-Stokes (AS) and Stokes (S) phonon side bands (PSBs). The PSBs of N2 and N4 lines, which also exist at 650–750 nm, are broader while exhibiting less intense features than PSB of R lines. Increasing Li⁺/Ga³⁺ results in reduced intensities of zero phonon R lines and decreased vibrionic PSB, as well as a red shift of emission peaks. When x reaches 1, the emission peak has an obvious change: the emission peak centers at ∼718 nm, corresponding to the Cr³⁺ zero-phonon emission from the ²E to ⁴A₂ level. The emission band at 650–750 nm corresponds to the ⁴T₂ → ⁴A₂ transition, which contains both the anti-Stokes and Stokes PSBs.

Fig. 7 (a) PLE and (b) PL spectra of ZLGGC at room temperature. (c) Mechanism diagram of luminescence in ZLGGC phosphors. (d) Tanabe–Sugano energy-level diagram of Cr³⁺ in x = 0 and x = 1 samples.

According to the reports, the Racah parameters B and C were estimated for mean peak energies of ⁴A₂ → ⁴T₂ and ⁴A₂ → ⁴T₁ bands, and the crystal field strength was evaluated using the crystal field parameter (D_q) from the mean peak energy of the ⁴A₂ → ⁴T₂ transition, based on the PLE data at room temperature:^3,44–46,52

10D_q = E(⁴T₂) − E(⁴A₂) (5)

D_q/B = 15(x − 8)/(x² − 10x) (6)

x = [E(⁴T₂) − E(⁴A₂)]/D_q (7)

C ≅ E(⁴T₂) − E(⁴A₂) (8)

and the calculated values are listed in Table 3. The estimated values of B = 652.8 cm⁻¹, D_q = 1818.2 cm⁻¹, and D_q/B = 2.78 for the x = 0 sample clearly agree with the existing reports showing that by following the calculated energy level diagram for Cr³⁺ in octahedral symmetry, the Cr³⁺ embedded in the zinc gallate host is subjected to a strong crystal field approximation.⁴⁵ Meanwhile, the ratios of D_q/B are determined to be 2.77 (x = 0.2), 2.57 (x = 0.4), 2.29 (x = 0.6), 2.20 (x = 0.8) and 2.08 (x = 1), which indicate that the crystal field strength decreases as Li⁺/Ga³⁺ doping content increases. It can be seen that the energy levels of ⁴T₁ and ⁴T₂ are closely correlated with the crystal field strength. The energy gap between the ground state (⁴A₂) and the excitation state (⁴T₁ and ⁴T₂) can be narrowed by weaker crystal field strength, leading to a red shift of the 3d Cr³⁺ transitions. It is worth considering that the crystal field strength is closely correlated with the parameters affecting outer orbits. To figure out the real reasons, a deeper understanding of the parameters that affect outer orbits is needed. Brik et al. introduced a new parameter of the nephelauxetic effect β as follows to explore the relationship between the covalence of the metal–ligand chemical bonds and the energies of the lowest energy spin-forbidden transition ²E_g → ⁴A_2g:^44,53

β = [(B/B₀)² + (C/C₀)²]^1/2 (9)

where B₀, C₀ (C₀ = 3850 cm⁻¹ and B₀ = 918 cm⁻¹) and B, C are the Racah parameters of Cr³⁺ in free and crystal states, respectively. β increases from 1.09 to 1.12 with increasing x from 0 to 1, indicating an increased bonding of the Cr³⁺-ligand. The peak of the ²E → ⁴A₂ transition in the emission spectra only shows a slight red shift compared to the large shift observed in the excitation spectra since the energy gap between the ground state and the ²E level is insensitive to the changes of crystal field strength.⁵⁴ The PLE/PL spectra of ZLGGC at a low temperature demonstrate a similar pattern as shown in Fig. S7.†Fig. 7c shows the mechanism diagram of luminescence for ZLGGC phosphors based on the above results. The electrons of Cr³⁺ in ZLGGC are pumped from ⁴A₂ (ground state level) to ⁴T₂ (⁴F), ⁴T₁ (⁴F), ⁴T₁ (⁴P) (excited state levels), and conduction band, then transferred to the ²E and ⁴T₂ level through non-radiative relaxation, and eventually changed to the ⁴A₂ state, which has a good NIR light output. With the increasing value of x, the PLE and PL spectra exhibit a distinct redshift. This phenomenon can be explained by utilizing Tanabe–Sugano energy-level diagram shown in Fig. 7d.

Table 3 Energy states, crystal field parameters and calculated values of ZLGGC

Simple	^4A2–^4T1 (cm^−1)	^4A2–^4T2 (cm^−1)	^4A2–^2E (cm^−1)	D q	D q/B	B	C	β
x = 0	24 875	18 182	14 388	1818	2.78	652.8	3163.4	1.09
x = 0.2	24 814	18 116	14 144	1812	2.77	654.1	3081.2	1.07
x = 0.4	24 752	17 794	14 144	1779	2.57	690.8	3004.9	1.08
x = 0.6	24 450	17 153	14 124	1715	2.29	746.2	2887.7	1.10
x = 0.8	24 331	16 920	13 986	1692	2.20	767.0	2802.0	1.11
x = 1	24 155	16 584	13 926	1658	2.08	798.7	2721.9	1.12

---

3.4 Trap energy up-conversion-like NIR persistent luminescence in bioimaging

Fig. S8a† shows the afterglow decay curves of ZLGGC phosphors monitored at 695–718 nm after irradiation using 254 nm UV light for 10 min. The results showed that in the first 500 s, the afterglow intensity of all phosphors rapidly drops, followed by a slow decrease, which is correlated with the release speed of electrons from the traps. Afterglow decay curves and time-dependent afterglows of ZLGGC phosphors (Fig. S8a† and Fig. 8a) demonstrate a similar phenomenon that with increasing value of x from 0 to 0.8, the afterglow intensity increases, but with further increase of the Li⁺/Ga³⁺ content (x > 0.8), it drops. Fig. S8b† shows the persistent luminescence spectra with x = 0.8 phosphor as an example. It is shown that the spectra have a band with a peak at ∼708 nm, indicating that the phosphors output excellent NIR afterglow, which can last more than 2 h. The TL curve of ZLGGC samples was measured at a temperature of 300–450 K to explore the trap depth and distribution (Fig. 8b, and Fig. S9†). The trap density n and trap depth E are estimated according to the equations:^22,55,56

E = T_m/500 (10)

n = ωI_m/{β × [2.52 + 10.2 × (μ_g − 0.42)]} (11)

Fig. 8 (a) Time-dependent NIR afterglows of ZLGGC after 254 nm UV light illumination for 10 min. (b) TL curves of ZLGGC after 254 nm UV light illumination for 10 min. (c) T_m and FWHM of the TL bands for different x values. (d) Schematics illustrating the multi-trap energy level diagram with increasing x.

All statistics are demonstrated in Table S7.† The x = 0.8 sample outputs the best persistent luminescence because it has the deepest trap depth and the highest electron trap density. It is known that most of the TL bands have asymmetric shapes. The trap energy is assumed to have a “quasi-continuous” distribution, in which case a well-defined TL peak is not observed, but a broad peak composed of several closely spaced TL peaks superimposed. In this study, with the increase of the x value (higher Li⁺/Ga³⁺ content), the TL peaks gradually moved to high temperature, the full-width at half maximum (FWHM) gradually increased and split into two peaks when x = 1 (Fig. 8b). It's fantastic that two broad bands appear in the TL curve of the x = 1 sample, and the centers at 405 K and 544 K, respectively, which indicates that the x = 1 sample owns two kinds of traps, including deep traps (DTs) and shallow traps (STs). Fig. 8c additionally shows the FWHM of the TL band and the peak temperatures of the TL bands (T_m), as a function of the x value. It is seen that the FWHM and peak temperatures of the TL bands are 139/396, 141/398, 144/401, 153/402 and 173/415 K for x = 0, 0.2, 0.4, 0.6 and 0.8, respectively. This phenomenon is closely related to the crystal structure and crystal field strength: it may be that the increase of Li⁺/Ga³⁺ content leads to the decrease of crystal field, which further deepens the trap depth and widens the energy level, similar to the centroid shift of Ce³⁺ ion luminescence, which has also been observed in previous studies.^22,57 The multi-trap energy level diagram (Fig. 8d) was drawn as shown in Fig. 8b and c with the increase of the x value, there are two effects (1) the trap energy level center moved down and (2) the trap energy level coverage area gradually increased. When x ≤ 0.8, there are many lattice defects in the sample, which are in the disordered state. The crystal field strength is greater than 2.10, thus Cr³⁺ ions are in a strong crystal field with ²E → ⁴A₂ transition dominating, which makes the continuously distributed trap energy level gradually move down and become wider. When x reaches 1, the crystal field strength is less than 2.10. Therefore, Cr³⁺ ions are in a weak crystal field, leading to the trap level splitting into two divided traps including deep traps (DTs) and shallow traps (STs). This further makes the deep trap energy level and ²E energy level prone to electron transfer by tunneling. The splitting of two kinds of traps in the x = 1 sample means that the rate of electron transfer between DTs and STs may decrease, which favors the DTs to preserve the charged electrons and sets the stage for the subsequent prolongation of the low-energy-excitation afterglow.^26,55,56,58

To further explore the spatial distributions of trap energy levels following optical excitation, TL spectra of the x = 1 sample were detected at 0 min, 30 min, and 24 h after ceasing the UV excitation (Fig. 9a, and Fig. S10†). The low-temperature TL band shifts from 405 to 476 K when the interval time following UV irradiation increases from 0 s to 30 min and 24 h, while little change was observed in the high-temperature band intensity peaked at 544 K, indicative of carrier exhaustion in STs and little release of the carriers in DTs. These results suggest that the trapping process consists of two steps, the shallow trapping step and DT capture step (Fig. 9b): first, following the pumping of high-energy photons, the charge carriers are introduced into shallow traps, second, the carriers are further filled into deeper traps via nonradiative relaxation. In addition to the conventional capture of carriers from STs to DTs, the mechanism also involves an up-conversion-like electron transition from DTs to STs. Based on previous reports, ZGC has a NIR persistent luminescence repeatedly activated by red light (∼650 nm). NIR excitation (longer wavelength) maybe more beneficial for bioimaging because it can reduce tissue scattering, thereby increasing the effective in vivo penetration depth.²⁶ To investigate the existence of electron transition from DTs to STs in the system, ∼650 nm red (25 W) and ∼740 nm NIR LEDs (25 W) were used to estimate the ability to re-excitation. Time-dependent afterglows and afterglow decay curves of ZLGGC phosphors are demonstrated in Fig. S11 and S12.† The results indicate that samples exhibit a better afterglow after the ∼650 nm red LED excitation (there's not much difference in intensity.) and the x = 1 sample exhibits the best afterglow after low energy excitation, which is mainly because that trap splitting reduces the rate of electron transfer from the deep trap to the shallow trap. However, under the same thickness of pork (5 mm), the x = 1 sample particles demonstrate a better afterglow after the ∼740 nm NIR LED excitation than the ∼650 nm red LED excitation since longer-wavelength NIR excitation can decrease tissue scattering (Fig. S13†). Therefore, a large irradiation area NIR LED (∼740 nm) was utilized as the light source in the subsequent tests (Fig. S14†). With a pork thickness of 2 cm, the x = 1 sample particles covered with pork still maintained ∼30% emission intensity compared to the x = 1 sample particles not covered with pork under the 254 nm-UV light (Fig. S15†). After ceasing the ∼740 nm NIR LED excitation, the afterglow decay curve in Fig. S16† clearly shows that NIR irradiation leads to the afterglow at 718 nm lasting more than 1500 s.

Fig. 9 (a) TL spectra of the x = 1 sample after ceasing the UV excitation (excitation wavelength: 254 nm; exposure time: 10 min; interval time: 30 s, 30 min, and 24 h). (b) Schematics illustrating the normal trapping mode of deep traps (DTs). (c) TL curves of the x = 1 sample were measured using the ∼740 nm NIR LED (25 W) with different excitation times (exposure time: 200, 600, and 1200 s). (d) Schematics illustrating the up-conversion trapping process from DTs to STs under the low energy excitation. (e) In vivo NIR afterglow images of a nude mouse (BALB/cA-nu adult mouse) after intramuscular injection of LGC-NH₂ solution (1 mL, 2 mg mL⁻¹, 10 min exposure to a 254 nm UV light before injection) and recharging NIR afterglow decay images after 20 min secondary excitation with a ∼740 NIR LED lamp.

To investigate the existence of electron transition from DTs to STs in this system, the TL curves of the x = 1 sample were measured using the ∼740 nm NIR LED as the excitation source with different excitation times. After a 200 s irradiation with a NIR LED (∼740 nm), a distinct TL peak was observed at 530 K in the TL spectra (Fig. 9c, and Fig. S17†). With time increasing from 200 to 1200 s, the TL peak shifted from 540 to 507 K, and a new low-temperature TL peak was observed at 410 K, suggesting that the exposure time can affect the distribution of carriers in DTs and STs. Meanwhile, the afterglow has been extended with the exposure time changing from 200 to 1200 s, which further indicates the transfer of electrons from DTs to STs (Fig. S18†). All results imply that the trapping process and NIR excited persistent luminescence are as follows (Fig. 9d): under NIR excitation, the electrons are pumped from ²E to ⁴A₂, and the excited electrons are captured by the DTs through tunneling; under NIR irradiation, some electrons are stored in the DTs and some are pumped into the STs like up-conversion energy transfer, which leads to NIR emission; on removing the NIR light source, the stored electrons are released from DTs to STs and then to the excited energy level of Cr³⁺ ions, which finally contributes to the NIR afterglow. The energy storage based on the up-conversion-like trapping process also requires reproducibility and photostability under irradiation of NIR, which are essential for reliable biological imaging.^4,5 Therefore, the attenuation spectra after four ∼740 nm NIR LED irradiation are shown in Fig. S19.† The initial intensity and decay duration remained essentially unchanged even following more than eight on/off cycles (Fig. S20†). In order to realize the near infrared light renewable charging in biological imaging, the nanoparticles of the x = 1 sample with good dispersion and size within 200 nm were collected by sieving (Fig. S21†). Before the in vivo imaging experiment, the Cell Counting Kit-8 assay was performed on RAW264, A549-1, and HEK293T cells to assess the cytotoxicity of LGC-NH₂ (Fig. S22†), which confirms the low toxicity of LGC-NH₂ to the cells. The nanoparticles–NH₂ glucose solution (1 mg mL⁻¹) was exposed to 254 nm UV light for 10 minutes for in vivo imaging. After the irradiation process, 500 μL of the irradiated aqueous solution was injected subcutaneously into a nude adult mouse on the belly and the NIR afterglow signal was tested for bioimaging analysis.^16,59 The intense signal of NIR afterglow could be detected during the test time throughout the mouse body with no further excitation treatment, although signal loss took place over time. The afterglow luminescence signal lasted for at least 2 h (Fig. 9e) under such NIR excitation. To assess the repeatability under such NIR excitation, the nude adult mice were subjected to a 20 min in situ excitation using an NIR LED lamp (∼740 nm). Following red-light excitation, a repeatable NIR afterglow signal appeared, which could last for more than 1 h (Fig. 9e). The above results suggest that the prepared Cr³⁺-doped gallate probes are effective fluorescent dyes for NIR light recovery afterglow to enhance bioimaging by connecting the electron transport channel from DTs to STs.

4. Conclusion

In this study, a series of Zn_1−x(Li/Ga)_xGa₂O₄:Cr³⁺ (x = 0–1) NIR persistent luminescent phosphors were synthesized using a solid-state reaction. The characterization of the samples was achieved by Raman spectroscopy, DFT calculations, XRD, UV-Vis-NIR spectroscopy, NMR spectroscopy, Rietveld refinement, TL, PLE/PL spectroscopy, and persistent luminescence decay analysis. With the increase of the Li⁺/Ga³⁺ content in ZnGa₂O₄, all the newly introduced Ga³⁺ ions occupy the tetrahedral sites, but Li⁺ ions firstly occupy the tetrahedral sites, then partially enter the octahedral sites, and completely occupy the octahedral sites when x reaches 1. These contribute to the contraction of the tetrahedron and the expansion of the octahedron, as well as changing the structure from order to disorder and back to order. Incorporation of Li⁺/Ga³⁺ results in the decrease of the lattice constant and the increase of the band gap. The crystal field strength of phosphors decreases as the Li⁺/Ga³⁺ content increases, which leads to a redshift of the emission and excitation spectra. The crystal field also affects the trap depth and distribution. When the crystal field becomes weaker, the trap depth becomes deeper and the trap energy level becomes wider and tends to split into two-divided energy levels. This makes the deep trap energy level come close to the ²E energy level of Cr³⁺. A complete replacement of Zn²⁺ with Li⁺/Ga³⁺ ions leads to the splitting of the trap energy level, which reduces the electron transfer between DTs and STs, but enhances the NIR afterglow of the samples, which are excited by the low-energy NIR light. This work provides a new category for the NIR-absorptive-NIR-emissive phosphors and a new application for long-term biological imaging.

Author contributions

Qi Zhu and Ji-Guang Li conceived the project; Junqing Xiahou, Minghui Jin, and Fan Li carried out the experiments; Junqing Xiahou carried out the data analysis; Lin Zhu carried out the DFT calculations; Qi Zhu and Junqing Xiahou drafted the manuscript and performed the analysis. All the authors were involved in the discussion of results, and have read and approved the final manuscript.

Conflicts of interest

The authors declare that they have no conflict of interest.

Acknowledgements

This work was supported in part by the Natural Science Foundation of Liaoning Province (Grant 2020-MS-081) and the National Natural Science Foundation of China (Grants 51302032, U21A2045, and 52172112).

References

1. Z. Pan, Y. Y. Lu and F. Liu, Sunlight-Activated Long-Persistent Luminescence in the Near-Infrared from Cr³⁺-Doped Zinc Gallogermanates, Nat. Mater., 2011, 11, 58–63.
2. A. Bessière, S. K. Sharma, N. Basavaraju, K. R. Priolkar, L. Binet, B. Viana, A. J. J. Bos, T. Maldiney, C. Richard, D. Scherman and D. Gourier, Storage of Visible Light for Long-Lasting Phosphorescence in Chromium-Doped Zinc Gallate, Chem. Mater., 2014, 26, 1365–1373.
3. 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, 5579–5588.
4. T. Maldiney, A. Bessiere, J. Seguin, E. Teston, S. K. Sharma, B. Viana, A. J. Bos, P. Dorenbos, M. Bessodes, D. Gourier, D. Scherman and C. Richard, The in Vivo Activation of Persistent Nanophosphors for Optical Imaging of Vascularization, Tumours and Grafted Cells, Nat. Mater., 2014, 13, 418–426.
5. J. Shi, X. Sun, J. Li, H. Man, J. Shen, Y. Yu and H. Zhang, Multifunctional Near Infrared-Emitting Long-Persistence Luminescent Nanoprobes for Drug Delivery and Targeted Tumor Imaging, Biomaterials, 2015, 37, 260–270.
6. F. Liu, W. Yan, Y. J. Chuang, Z. Zhen, J. Xie and Z. Pan, Photostimulated Near-Infrared Persistent Luminescence as a New Optical Read-Out from Cr³⁺-Doped LiGa₅O₈, Sci. Rep., 2013, 3, 1554.
7. Y. Lv, D. Ding, Y. Zhuang, Y. Feng, J. Shi, H. Zhang, T. L. Zhou, H. Chen and R. J. Xie, Chromium-Doped Zinc Gallogermanate@Zeolitic Imidazolate Framework-8: A Multifunctional Nanoplatform for Rechargeable In Vivo Persistent Luminescence Imaging and pH-Responsive Drug Release, ACS Appl. Mater. Interfaces, 2019, 11, 1907–1916.
8. Z. Li, Y. Zhang, X. Wu, L. Huang, D. Li, W. Fan and G. Han, Direct Aqueous-Phase Synthesis of Sub-10 nm “Luminous Pearls” with Enhanced in Vivo Renewable Near-Infrared Persistent Luminescence, J. Am. Chem. Soc., 2015, 137, 5304–5307.
9. J. Shi, X. Sun, S. Zheng, J. Li, X. Fu and H. Zhang, A New Near-Infrared Persistent Luminescence Nanoparticle as a Multifunctional Nanoplatform for Multimodal Imaging and Cancer Therapy, Biomaterials, 2018, 152, 15–23.
10. M. Allix, S. Chenu, E. Véron, T. Poumeyrol, E. A. Kouadri-Boudjelthia, S. Alahraché, F. Porcher, D. Massiot and F. Fayon, Considerable Improvement of Long-Persistent Luminescence in Germanium and Tin Substituted ZnGa₂O₄, Chem. Mater., 2013, 25, 1600–1606.
11. Q. Zhu, J. Xiahou, X. Li, X. Sun and J.-G. Li, Defect Cluster Engineering in ZnGa_2−x(Mg/Ge)_xO₄:Cr³⁺ Nanoparticles for Remarkably Improved NIR Persistent Luminescence, J. Am. Ceram. Soc., 2021, 104, 4594–4605.
12. B. Zhao, Q. Zhu, X. Sun and J.-G. Li, Co-Doping Zn²⁺/Sn⁴⁺ in ZnGa₂O₄:Cr³⁺ for Dynamic Near-Infrared Luminescence and Advanced Anti-Counterfeiting, Ceram. Int., 2021, 47, 17000–17007.
13. A. Abdukayum, J. T. Chen, Q. Zhao and X. P. Yan, Functional Near Infrared-Emitting Cr³⁺/Pr³⁺ Co-Doped Zinc Gallogermanate Persistent Luminescent Nanoparticles with Superlong Afterglow for In Vivo Targeted Bioimaging, J. Am. Chem. Soc., 2013, 135, 14125–14133.
14. Y. Zhuang, J. Ueda and S. Tanabe, Tunable Trap Depth in Zn(Ga_1−xAl_x)₂O₄:Cr,Bi Red Persistent Phosphors: Considerations of High-Temperature Persistent Luminescence and Photostimulated Persistent Luminescence, J. Mater. Chem. C, 2013, 1, 7849.
15. R. Zou, S. Gong, J. Shi, J. Jiao, K.-L. Wong, H. Zhang, J. Wang and Q. Su, Magnetic-NIR Persistent Luminescent Dual-Modal ZGOCS@MSNs@Gd₂O₃ Core-Shell Nanoprobes For In Vivo Imaging, Chem. Mater., 2017, 29, 3938–3946.
16. R. Zou, J. Huang, J. Shi, L. Huang, X. Zhang, K.-L. Wong, H. Zhang, D. Jin, J. Wang and Q. Su, Silica Shell-Assisted Synthetic Route for Mono-Disperse Persistent Nanophosphors with Enhanced In Vivo Recharged Near-Infrared Persistent Luminescence, Nano Res., 2017, 10, 2070–2082.
17. O. Q. De Clercq, L. I. D. J. Martin, K. Korthout, J. Kusakovskij, H. Vrielinck and D. Poelman, Probing the Local Structure of the Near-Infrared Emitting Persistent Phosphor LiGa₅O₈:Cr³⁺, J. Mater. Chem. C, 2017, 5, 10861–10868.
18. M. S. Relvas, M. R. N. Soares, S. O. Pereira, A. V. Girão, F. M. Costa and T. Monteiro, Trends in Cr³⁺ Red Emissions from ZnGa₂O₄ Nanostructures Produced by Pulsed Laser Ablation in a Liquid Medium, J. Phys. Chem. Solids, 2019, 129, 413–423.
19. Y. Li, Y. Li, R. Chen, K. Sharafudeen, S. Zhou, M. Gecevicius, H. Wang, G. Dong, Y. Wu, X. Qin and J. Qiu, Tailoring of the Trap Distribution and Crystal Field in Cr³⁺-Doped Non-Gallate Phosphors with Near-Infrared Long-Persistence Phosphorescence, NPG Asia Mater., 2015, 7, e180–e180.
20. Y. Zhuang, J. Ueda, S. Tanabe and P. Dorenbos, Band-Gap Variation and a Self-Redox Effect Induced by Compositional Deviation in Zn_x,Ga₂O_3+x:Cr³⁺ Persistent Phosphors, J. Mater. Chem. C, 2014, 2, 5502.
21. Y. Li, X. Du, K. Sharafudeen, C. Liao and J. Qiu, A Long Persistent Phosphor Based on Recombination Centers Originating from Zn Imperfections, Spectrochim. Acta, Part A, 2014, 123, 7–11.
22. J. Xiahou, Q. Zhu, L. Zhu, S. Li and J.-G. Li, Local Structure Regulation in Near-Infrared Persistent Phosphor of ZnGa₂O₄:Cr³⁺ to Fabricate Natural-Light Rechargeable Optical Thermometer, ACS Appl. Electron. Mater., 2021, 3, 3789–3803.
23. H. Guo, Y. Wang, W. Chen, W. Zeng, S. Han, G. Li and Y. Li, Controlling and revealing the Trap Distributions of Ca₆BaP₄O₁₇:Eu²⁺,R³⁺ (R = Dy, Tb, Ce, Gd, Nd) by Codoping Different Trivalent Lanthanides, J. Mater. Chem. C, 2015, 3, 11212–11218.
24. Y. Li, S. Zhou, Y. Li, K. Sharafudeen, Z. Ma, G. Dong, M. Peng and J. Qiu, Long Persistent and Photo-Stimulated Luminescence in Cr³⁺-Doped Zn-Ga-Sn-O Phosphors for Deep and Reproducible Tissue Imaging, J. Mater. Chem. C, 2014, 2, 2657.
25. H. Guo, W. Chen, W. Zeng, G. Li, Y. Wang, Y. Li, Y. Li and X. Ding, Structure and Luminescence Properties of a Novel Yellow Super Long-Lasting Phosphate Phosphor Ca₆BaP₄O₁₇:Eu²⁺,Ho³⁺, J. Mater. Chem. C, 2015, 3, 5844–5850.
26. X. Chen, Y. Li, K. Huang, L. Huang, X. Tian, H. Dong, R. Kang, Y. Hu, J. Nie, J. Qiu and G. Han, Trap Energy Upconversion-Like Near-Infrared to Near-Infrared Light Rejuvenateable Persistent Luminescence, Adv. Mater., 2021, 33, e2008722.
27. W. Lambrecht, Dopants and Defects in Semiconductors, Mater. Today, 2012, 15, 349.
28. P. Xiong, B. Huang, D. Peng, B. Viana, M. Peng and Z. Ma, Self–Recoverable Mechanically Induced Instant Luminescence from Cr³⁺–Doped LiGa₅O₈, Adv. Funct. Mater., 2021, 31, 2010685.
29. P. Xiong, M. Peng, K. Qin, F. Xu and X. Xu, Visible to Near–Infrared Persistent Luminescence and Mechanoluminescence from Pr³⁺–Doped LiGa₅O₈ for Energy Storage and Bioimaging, Adv. Opt. Mater., 2019, 7, 1901107.
30. J. Xiahou, Q. Zhu, L. Zhu, S. Huang, T. Zhang, X. Sun and J.-G. Li, Lattice-Site Engineering in ZnGa₂O₄:Cr³⁺ Through Li⁺ Doping for Dynamic Luminescence and Advanced Optical Anti-Counterfeiting, J. Mater. Chem. C, 2022, 10, 7935–7948.
31. J. P. Perdew, K. Burke and M. Ernzerhof, Generalized Gradient Approximation Made Simple, Phys. Rev. Lett., 1996, 77, 3865.
32. L. Zhu, L. Li and T.-M. Cheng, First Principle Study of the Electrochemical Properties of Li₂FeSiS₄, Comput. Mater. Sci., 2015, 106, 135–139.
33. C. Bernuy-Lopez, W. Manalastas, J. M. Lopez del Amo, A. Aguadero, F. Aguesse and J. A. Kilner, Atmosphere Controlled Processing of Ga-Substituted Garnets for High Li-Ion Conductivity Ceramics, Chem. Mater., 2014, 26, 3610–3617.
34. L. Buannic, B. Orayech, J.-M. López Del Amo, J. Carrasco, N. A. Katcho, F. Aguesse, W. Manalastas, W. Zhang, J. Kilner and A. Llordés, Dual Substitution Strategy to Enhance Li⁺ Ionic Conductivity in Li₇La₃Zr₂O₁₂ Solid Electrolyte, Chem. Mater., 2017, 29, 1769–1778.
35. D. S. Cook, J. E. Hooper, D. M. Dawson, J. M. Fisher, D. Thompsett, S. E. Ashbrook and R. I. Walton, Synthesis and Polymorphism of Mixed Aluminum-Gallium Oxides, Inorg. Chem., 2020, 59, 3805–3816.
36. S. Takahashi, A. Kan and H. Ogawa, Microwave Dielectric Properties of Spinel-Structured Li_0.5Ga_2.5O₄ Ceramics with Cation Ordering on Octahedral Sites, Jpn. J. Appl. Phys., 2016, 55, 10TE01.
37. Pressure-Induced Phase Transitions in AB2X4 Chalcogenide Compounds, ed. D. Errandonea, Springer, Berlin Heidelberg, 2014.
38. D. Li, Y. Wang, K. Xu, L. Li, Z. Hu and H. Zhao, Enhancement of Photoluminescence, Persistent Luminescence and Photocatalytic Activity in ZnGa₂O₄ Phosphors by Lithium Ion Doping, Opt. Mater., 2015, 42, 313–318.
39. P. Yadav, S. Uniyal, S. Uma and R. Nagarajan, Ordered LiGa₅O₈ Loaded with Redox Capable Cu²⁺, Cr³⁺ Ions to Manifest Interesting Optical, Magnetic, and Catalytic Properties, J. Mater. Sci., 2021, 56, 20111–20125.
40. J. H. Park, B. W. Park, N. S. Choi, Y. T. Jeong, J. S. Kim and J. S. Yang, Enhancement in Cathodoluminescent Properties of Carbon Nanotube-ZnGa₂O₄:Mn²⁺ Phosphor Composites, Electrochem. Solid-State Lett., 2008, 11, J12–J14.
41. Y. Wang, J. Jiang, Z. Song and J. Zhang, Structural Characterization &, Optical and Electrical Properties of LiGaO₂ & LiGa₅O₈ Micro-Nanoparticles-Based Photodetectors, J. Alloys Compd., 2021, 887, 161438–161447.
42. N. Basavaraju, K. R. Priolkar, D. Gourier, A. Bessiere and B. Viana, Order and Disorder around Cr³⁺ in Chromium Aoped Persistent Luminescent AB₂O₄ Spinels, Phys. Chem. Chem. Phys., 2015, 17, 10993–10999.
43. N. Basavaraju, K. R. Priolkar, D. Gourier, S. K. Sharma, A. Bessiere and B. Viana, The Importance of Inversion Disorder in the Visible Light Induced Persistent Luminescence in Cr³⁺ Doped AB₂O₄ (A = Zn or Mg and B = Ga or Al), Phys. Chem. Chem. Phys., 2015, 17, 1790–1799.
44. M. Back, E. Trave, J. Ueda and S. Tanabe, Ratiometric Optical Thermometer Based on Dual Near-Infrared Emission in Cr³⁺-Doped Bismuth-Based Gallate Host, Chem. Mater., 2016, 28, 8347–8356.
45. Y. Tanabe and S. Sugano, On the Absorption Spectra of Complex Ions II, J. Phys. Soc. Jpn., 1954, 9, 766–779.
46. Q. Bai, S. Zhao, L. Guan, Z. Wang, P. Li and Z. Xu, Design and Control of the Luminescence of Cr³⁺-Doped Phosphors in the Near-Infrared I Region by Fitting the Crystal Field, Cryst. Growth Des., 2018, 18, 3178–3186.
47. W. Huang, X. Gong, R. Cui, X. Li, L. Li, X. Wang and C. Deng, Enhanced Persistent Luminescence of LiGa₅O₈:Cr³⁺ Near-Infrared Phosphors by Codoping Sn⁴⁺, J. Mater. Sci.: Mater. Electron., 2018, 29, 10535–10541.
48. V. Singh, R. V. S. S. N. Ravikumar, G. Sivaramaiah, J. L. Rao and S. H. Kim, Investigations of the Optical and EPR Properties of LiGa₅O₈:Cr³⁺ Phosphor, Mater. Res. Bull., 2015, 61, 183–188.
49. F. H. Teherani, D. C. Look, D. J. Rogers, S. K. Sharma, A. Bessiere, D. Gourier, G. Sraiki, B. Viana, P. J. Deren, D. Rudnicka, A. Watras, N. Basavaraju, K. R. Priolkar, T. Maldiney, D. Scherman and C. Richard, Optical Properties and Storage Capabilities in AB₂O₄:Cr³⁺ (A=Zn, Mg, B=Ga, Al), Proc. SPIE, 2014, 8987, 89870J.
50. A. Bessière, S. Jacquart, K. Priolkar, A. Lecointre, B. Viana and D. Gourier, ZnGa₂O₄:Cr³⁺: a New Red Long-Lasting Phosphor with High Brightness, Opt. Express, 2011, 19, 10131–10137.
51. W. Mikenda, N-Lines, in The Luminescence Spectra of Cr³⁺-Doped Spinels (I): Identification of N-Lines, J. Lumin., 1981, 26, 53–66.
52. T. Si, Q. Zhu, J. Xiahou, X. Sun and J.-G. Li, Regulating Mn²⁺/Mn⁴⁺ Activators in ZnGa₂O₄ via Mg²⁺/Ge⁴⁺ Doping to Generate Multimode Luminescence for Advanced Anti-Counterfeiting, ACS Appl. Electron. Mater., 2021, 3, 2005–2016.
53. M. G. Brik, S. J. Camardello, A. M. Srivastava, N. M. Avram and A. Suchocki, Spin-Forbidden Transitions in the Spectra of Transition Metal Ions and Nephelauxetic Effect, ECS J. Solid State Sci. Technol., 2015, 5, R3067–R3077.
54. S. Li, Q. Zhu, X. Sun and J.-G. Li, Magical Polyhedral Twist via Chemical Unit Co-Substitution in LaAlO₃:Mn⁴⁺ to Greatly Enhance the Zero Phonon Line for High-Efficiency Plant-Growth LEDs, J. Mater. Chem. C, 2021, 9, 7163–7173.
55. R. Chen, On the Calculation of Activation Energies and Frequency Factors from Glow Curves, J. Appl. Phys., 1969, 40, 570–585.
56. R. Chen, Glow Curves with General Order Kinetics, J. Electrochem. Soc., 1969, 116, 1254–1257.
57. Q. Meng, Q. Zhu, X. Li, X. Sun and J. G. Li, New Mg²⁺/Ge⁴⁺-Stabilized Gd₃Mg_xGe_xAl_5-2x,O₁₂:Ce Garnet Phosphor with Orange-Yellow Emission for Warm-White LEDs (x = 2.0–2.5), Inorg. Chem., 2021, 60, 9773–9784.
58. Y. Li, M. Gecevicius and J. Qiu, Long Persistent Phosphors–from Fundamentals to Applications, Chem. Soc. Rev., 2016, 45, 2090–2136.
59. Q. Zhu, J. Xiahou, Y. Guo, H. Li, C. Ding, J. Wang, X. Li, X. Sun and J.-G. Li, Zn₃Ga₂Ge₂O₁₀:Cr³⁺ Uniform Microspheres: Template-Free Synthesis, Tunable Bandgap/Trap Depth, and In Vivo Rechargeable Near-Infrared-Persistent Luminescence, ACS Appl. Bio Mater., 2018, 2, 577–587.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d3qi00184a

---