Cationic telluroviologen derivatives as type‐I photosensitizers for tumor photodynamic theranostics

The hypoxia of the tumor microenvironment (TME) seriously restricts the photodynamic therapy (PDT) effect of conventional type‐II photosensitizers, which are highly dependent on O2. In this work, a new type‐I photosensitizer (TPE‐TeV‐PPh3) consisting of a tetraphenylethylene group (TPE) as a bioimaging moiety, triphenyl‐phosphine (PPh3) as a mitochondria‐targeting group, and telluroviologen (TeV2+) as a reactive oxygen species (O2•−, •OH) generating moiety is developed. The luminescence intensity of TPE‐TeV‐PPh3 increased significantly after specific oxidation by excess H2O2 in the TME without responding to normal tissues via the formation of Te═O bond, which can be used for monitoring abnormal H2O2, positioning, and imaging of tumors. TPE‐TeV‐PPh3 with highly reactive radicals generation and stronger hypoxia tolerance realizes efficient cancer cell killing under hypoxic conditions, achieving 88% tumor growth inhibition. Therefore, TPE‐TeV‐PPh3 with low phototoxicity in normal tissue achieves tumor imaging and effective PDT toward solid tumors in response to high concentrations of H2O2 in the TME, which provides a new strategy for the development of type‐I photosensitizers.


INTRODUCTION
Cancer is a major threat to human health and development. [1] Traditional cancer therapies, including surgery, chemotherapy, and radiotherapy, have been used clinically for decades, but they have inherent shortcomings. For example, surgery requires chemotherapy and radiotherapy for adjuvant treatment after tumor removal to prevent recurrence and deterioration. [2] However, radiotherapy and chemotherapy are always accompanied by serious side effects and resistance during treatment. [3] Compared with traditional therapy methods, photodynamic therapy (PDT) has received increasing attention due to its low systemic toxicity, high selectivity, and reduced trauma without drug/radiation resistance. [4] The main mechanism of action is to employ a light source to excite the photosensitizer to produce reactive oxygen Qi Sun and Qi Su contributed equally to this work.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2022 The Authors. Aggregate published by SCUT, AIEI, and John Wiley & Sons Australia, Ltd. species (ROS), which induce cytotoxicity and kill tumor cells. [5] Because cancer cells proliferate rapidly and consume a large amount of O 2 , the tumor microenvironment (TME) is always hypoxic. [6] From a mechanistic point of view, compared with widely used type-II photosensitizers (energy transfer), which need high oxygen dependence and the high oxygen consumption by 1 O 2 species, type-I photosensitizers can react with O 2 to generate ROS (O 2 •− , •OH) efficiently via electron transfer mechanism, in which O 2 can be supplemented through the subsequent reaction process, thus realizing the cyclic supplementation of O 2 . Thus, type-I photosensitizers are more suitable for the hypoxic environment (oxygen pressure < 5 mm Hg) of tumors. [7] Although lots of type-I photosensitizers have been reported, macromolecules with complex components such as organic nanocomposites, [8] organometallic complexes, [9] and metal-organic frameworks, [10] are suffering from worrisome biosafety, poor reproducibility, and complex pharmacokinetics. Organic small molecules have greatly weakened fluorescence emission and generation of ROS due to their aggregation in living organisms. [11] These are far from ideal for clinical application. Therefore, it would be of great significance to exploit photosensitizers with good biosafety, structural diversity, flexible preparation, and aggregation-induced emission (AIE) characteristics.
Viologen is a cationic compound with good redox properties, and it is easy to obtain an electron to generate cationic free radicals, [12] which also makes it widely used in medicine, pesticides, and other fields. [13] Although the cationic structure of viologen makes it easy to interact with negatively charged bacteria or cancer cells, [14] viologen derivatives with a wide bandgap and poor light absorption are difficult to apply in the field of PDT. Previous research has shown that introducing chalcogen elements into the viologen skeleton is an effective strategy for the development of new viologen photosensitizers. For example, chalcogenoviologen has a high degree of conjugation and a low energy gap and can achieve photodynamic killing of Gram-negative and Gram-positive bacteria and promote skin wound repair at low concentrations. [14a] However, chalcogenoviologens are still weak in producing 1 O 2 , which also greatly limits their application as type-II in photodynamic antibacterials. Nevertheless, chalcogenoviologens, especially telluroviologen, will produce a large number of highly stable radical cations with prolonged lifetime under light irradiation, [12a,15] which could react with O 2 to form ROS (O 2 •− , •OH) efficiently via electron transfer mechanism (type-I mechanism). [7c,16] Obviously, telluroviologen is very suitable as a type-I photosensitizer to achieve photodynamic killing of cancer cells under hypoxia due to the generated viologen radicals being more stable in a hypoxic environment. [17] However, telluroviologen is nonemissive due to the heavy atom effect, making it impossible to realize the imaging of cancer cells. Taking into account that the tellurium-containing compounds, such as the tellurium atoms in HBPTe1900, [18] 2-Me TeR, [19] are easily oxidized by H 2 O 2 (similar to the high concentration of H 2 O 2 in cancer cells [20] ), the fluorescence effect of the oxidized products is often greatly improved. It is not difficult to imagine that the oxidized telluroviologens can not only generate cationic radicals but also have good luminescence properties. Combined with the corresponding electron-donating groups and suitable targeting groups, it is possible to develop a new type-I photosensitizer for comprehensive diagnosis and treatment based on telluroviologens.
Based on these considerations, this study combined the tetraphenylethylene (TPE) group with AIE characteristics, [14b,21] the classic triphenylphosphine mitochondrial targeting group, [22] and telluroviologen to prepare a new type-I of photosensitizer (TPE-TeV-PPh 3 ). The mechanism of PDT with TPE-TeV-PPh 3 is shown in Scheme S1b. As the product of H 2 O 2 oxidation in the tumor, TPE-TeOV-PPh 3 had both excellent luminescence and cationic free radical generation performance, thus realizing the integration of efficient tumor imaging and treatment ( Figure 1). Meanwhile, TPE-TeV-PPh 3 has less phototoxicity in normal tissues since the normal tissue environment will reduce the formation of free radicals and reduce the phototoxicity in normal tissues, such as Chlorin e6 (Ce6). [23] Both experimental and theoretical results proved that TPE-TeV-PPh 3 showed good imaging and inhibitory effects on tumor cells in hypoxic environments in vitro and in vivo. This research not only provides new ideas for the design of AIE active type-I photosensitizers, but also lays a foundation for the development of PDT.

Synthesis and characterization
To obtain a desired type-I photosensitizer, compounds 1, 2, and 4 were synthesized according to the literature. [12a,24] 1 and 2 reacted in DMF at a ratio of 3 to 1 to obtain 3 (TPE-TeV) as an orange solid in 65% yield. TPE-TeV-Me and TPE-TeV-PPh 3 were obtained by 3 reacting with MeI and 4 in DMSO, respectively, followed by using excess NH 4 PF 6 for anion exchange, and the yields were 80% and 25%, respectively (Scheme 1). The detailed synthesis processes and structural characterization using high resolution mass spectrometry, 1 H NMR,and 13 C NMR are shown in the Supporting Information.

Photophysical and electrochemical properties
The maximum absorption wavelengths of TPE-TeV-PPh 3 and TPE-TeV-Me were 430 and 425 nm, respectively, in DMSO, indicating that TPE-TeV-PPh 3 and TPE-TeV-Me are suitable for white light-based therapy ( Figures S1 and S2). According to cyclic voltammetry (CV, Figure S3), the electrochemical properties of TPE-TeV-Me and TPE-TeV-PPh 3 were tested in the scanning range from −1.5 to −0.75 V. Both results showed two reversible one-electron reductions from the viologen skeleton, consistent with the previous results. [12a,25] Considering the oxidation reaction of tellurium-containing compounds, TPE-TeV-PPh 3 was reacted with H 2 O 2 to generate oxidized TPE-TeOV-PPh 3 . The optical properties of TPE-TeV-PPh 3 and TPE-TeOV-PPh 3 were assessed in DMSO. As shown in Figure 2A and Figure S4, the TPE-TeOV-PPh 3 solution was brighter under 365 nm light. Comparing the PL spectra, the emission intensity of TPE-TeOV-PPh 3 was greatly enhanced, the maximum emission peak position shifted from 568 to 510 nm (Figure 2A), and the fluorescence lifetimes of TPE-TeV-PPh 3 and TPE-TeOV-PPh 3 did not change significantly ( Figure S6). However, their fluorescence quantum yield improved to 1.66% from approaching 0% ( Figures S7 and S8). The result indicated that the combination of oxygen atoms weakens the heavy atom effect of tellurium atoms in TPE-TeV-PPh 3 . [19] This in turn means that we can detect oxidizing environments. A significantly high concentration of ROS (equivalent to approximately 100 µM H 2 O 2 ) has been reported in tumor tissue. [26] When TPE-TeV-PPh 3 was oxidized, the fluorescence intensity increased. The fluorescence changes can be detected by the imaging system, and the tumor site can be imaged. At the same time, corresponding changes in the UV−Vis absorption spectrum were also observed, and the maximum absorption peak appeared blueshifted at approximately 450 nm ( Figure S5). As shown in Figure 2B, the spectrum of TPE-TeOV-PPh 3 showed a new peak at F I G U R E 1 Schematic illustration of TPE-TeV-PPh 3 for PDT and imaging in response to the tumor microenvironment

S C H E M E 1 Synthesis of TPE-TeV-Me and TPE-TeV-PPh 3
600 cm −1 , which was considered to indicate the formation of Te═O. [18] As shown in Figures S9 and S10, the fluorescence intensity increased, which was accompanied by the continuous oxidation of TPE-TeV-PPh 3 . The detection limit of the TPE-TeV-PPh 3 H 2 O 2 response was also tested by fluorescence intensity. The H 2 O 2 could also be detected even if the concentration of TPE-TeV-PPh 3 was as low as 10 −6 M. Early cancer is often accompanied by abnormal hydrogen peroxide, the high sensitivity of TPE-TeV-PPh 3 can provide a reference for early cancer monitoring. In the process of biological imaging, aggregation-caused quenching (ACQ) often reduces the imaging effect. [27] TPE-TeOV-PPh 3 showed weak fluorescence in DMSO. However, the fluorescence intensity of TPE-TeOV-PPh 3 increased significantly when phosphate buffered saline (PBS) was added as a poor solvent, demonstrating the typical AIE of TPE-TeOV-PPh 3 ( Figure 2C and  Figure S11a). The nanoparticles size distribution of TPE-TeOV-PPh 3 in mixed solvents was also measured. As shown in Figure S11b, the nanoparticles size distribution of nanoaggregates was ∼160 nm, which facilitates precise biological imaging in the next step. [28]

Light-triggered ROS generation
To verify our design strategy, the type-I ROS generation capacities of TPE-TeV-Me and TPE-TeV-PPh 3 were assessed. To exclude the contribution of 1 O 2 to ROS generation, anthracene-9,10-dipropionicacid (ADPA) was selected as the probe for testing 1 O 2 , and methylene blue (MB) was employed as the reference. [29] As depicted in Figure 2D, under white light irradiation for different times, the absorption peak at 378 nm decreased to different degrees, revealing that TPE-TeV-Me, TPE-TeV-PPh 3 , and TPE-TeOV-PPh 3 have lower 1 O 2 yields. To further verify the free-radical ROS generation of TPE-TeV-Me and TPE-TeV-PPh 3 , the photosensitizers were treated with Zn powder [30] or light radiation [7a,15b] without the addition of free radical trap and the free-radicals by electron paramagnetic resonance (EPR) spectroscopy. [16a,31] Broad signals were observed in Figure 2E,F indicating the delocalization of the radical over a large portion of the molecular scaffold. In Figure 2G, the color of the solution becomes darkened after light irradiation.
The results showed that TPE-TeV-Me and TPE-TeV-PPh 3 had a strong ability to generate intramolecular free radicals under white light irradiation. The viologen cation radicals showed a longer lifetime ( Figure S12). Moreover, it could still generate free radicals under light irradiation when TPE-TeV-PPh 3 was oxidized to TPE-TeOV-PPh 3 ( Figure 2H). The viologen radical cation can transfer electrons to O 2 , yielding ROS species such as O 2 •− . Then we further confirmed the O 2 •− production by O 2 •− probe dihydrorhodamine 123 (DHR123), [32] which is non-fluorescent but can react with O 2 •− to emit strong green fluorescence centered at 540 nm.
As shown in Figure 2I, the fluorescence of DHR123 did not increase significantly under light conditions. The TPE-TeV-Me, TPE-TeV-PPh 3 , and TPE-TeOV-PPh 3 slightly increased the fluorescence intensity of DHR123 under normoxia. However, they distinctly increased the fluorescence intensity of DHR123 under hypoxic conditions. The guess is that O 2 limits the formation of viologen free radicals to a certain extent, thereby limiting the production of O 2 •− . In addition, to clarify that TPE-TeV-Me and TPE-TeV-PPh 3 can also generate free radicals when energized, electrochromic devices and UV-Vis spectra were employed. As shown in Figure S13, when a voltage of −0.8 V was applied to the device, the color of the device became darker, and exhibited intense absorption in an entire interval of the visible region, which is from the accumulation of radical states. [12a,33] This provides a feasible reference for the weak current promoting the release of free radicals from materials.

2.4
In vitro PDT performance under normoxia and hypoxia ROS production by TPE-TeV-PPh 3 and TPE-TeV-Me in cells was measured by flow cytometry. By using 2,7-Dichlorodihydrofluoresecein diacetate (DCFH-DA) as a probe for ROS, after light irradiation, the fluorescence intensity of cells increased ( Figure S14), suggesting efficient intracellular ROS generation. [34] The PDT efficiency for cancer cells was explored in detail by the standard methyl thiazolyl tetrazolium (MTT). [35] 7402 cells and MCF-7 cells were used as model cell lines. As shown in Figure 3A, TPE-TeV-PPh 3 and TPE-TeV-Me exhibit relatively low dark toxicity ( Figures S15 and S17). The relative lower cytotoxicity of TPE-TeV-PPh 3 compared with TPE-TeV-Me was attributed to the ability of target mitochondria by PPh 3 group. Then, the PDT of cancer cells under normoxia and hypoxia was further evaluated. Under normoxia, when the concentration of TPE-TeV-PPh 3 was 30 µM with light (50 mW/cm 2 , 5 min), the cell viability was reduced to 38%, showing a good ability to kill 7402 cancer cells. Conversely, TPE-TeV-Me showed a poor PDT effect ( Figure S15a,b). TPE-TeV-PPh 3 and TPE-TeV-Me had similar effects on MCF-7 cells ( Figure  S17). It was reported that the oxygen content of tumors is approximately 0%-4%. [36] Tumor cells in a 1% oxygen environment can be cultured to simulate the hypoxic environment of tumors in vitro. Under hypoxia, TPE-TeV-PPh 3 and TPE-TeV-Me showed a better cancer cell inhibition rate ( Figure 3A and Figure S15c). The cell viability at all concentrations in the light-irradiated group decreased. When the concentration of TPE-TeV-PPh 3 was 20 µM, cell viability was reduced from 68% to 42%. By comparing the data of cell viability in the TPE-TeV-PPh 3 with light irradiation group under normoxia and hypoxia ( Figure 3B), TPE-TeV-PPh 3 showed a better ability to kill cancer cells under hypoxia, and the data were statistically significant. The cell inhibition rate of TPE-TeV-PPh 3 was much higher than that of TPE-TeV-Me under light, likely due to the mitochondriatargeting capability ( Figure 3C). [37] The photosensitizers generally showed the disadvantage of strong phototoxicity to normal tissues. The phototoxicity of TPE-TeV-PPh 3 was compared with the commonly used photosensitizer Ce6 as a reference. As shown in Figure 3D, the phototoxicity of TPE-TeV-PPh 3 was much lower than that of Ce6 due to TPE-TeV-PPh 3 can obtain electrons under light conditions to form intramolecular free radicals, which is susceptible to the influence of O 2 in tissues and will reduce free radical generation, [23c,31] thereby reducing the phototoxicity in normal tissues.

Cell imaging
To verify the effect of TPE-TeV-PPh 3 in the mitochondria of the cells, we further investigated mitochondrial colocalization. MitoOrange is a mitochondrial red dye that can locate mitochondria. [38] After the 7402 cells were incubated with TPE-TeV-PPh 3 and stained with MitoOrange, they were observed by confocal laser scanning microscopy (CLSM). When TPE-TeV-PPh 3 was oxidized by H 2 O 2 , its fluorescence intensity in the cells was significantly enhanced ( Figure 3E). In the colocalization map, it can be clearly seen that the colocalization effect of TPE-TeV-PPh 3 in mitochondria was enhanced, and the colocalization coefficient was higher, which verified the role of TPE-TeV-PPh 3 . Interestingly, by observing the cell images of TPE-TeV-PPh 3 and TPE-TeV-Me, it was found that when the mitochondria were destroyed, the morphology of the mitochondria changed from filamentous to massive. Then, the mitochondrial membrane potential was tested to verify mitochondrial damage. [39] The mitochondria of cells were damaged in the TPE-TeV-PPh 3 + light group ( Figure S16), which was consistent with previous expectations. To observe cell viability, we conducted a live/dead cell survival test. Under hypoxia, the corresponding groups were added compounds TPE-TeV-PPh 3 and TPE-TeV-Me, and lighted. When cells were stained by calcein propidium iodide (PI), bright green fluorescence represented surviving cells, and red fluorescence represented dead cells. [40] Figure 3F shows that the TPE-TeV-PPh 3 + light group had a better inhibitory effect on cancer cells under hypoxia. Additionally, TPE-TeV-PPh 3 + light also showed a good inhibitory effect on MCF-7 cells ( Figure S18).

In vivo tumor fluorescence imaging and PDT performance
Since TPE-TeV-PPh 3 showed a good luminescence performance improvement after oxidation, and had ideal photodynamic test performance in a hypoxic environment, small animal fluorescence imaging and solid tumor inhibition experiments of 7402 liver cancer cells were carried out. After intratumoral injection of TPE-TeV-PPh 3 , it was oxidized by H 2 O 2 in the tumor. The fluorescence at the tumor site increased gradually, and the intensity reached a plateau after 1 h. The tumor site was clearly observed by the fluorescence signal. At 4 h post injection, the tumor site still possessed strong luminescence, indicating a strong retention rate of TPE-TeV-PPh 3 in the tumor ( Figure 4A,B). After the mice were sacrificed at 4 h, fluorescence images of various organs and tumor tissues of the mice were obtained ( Figure S19). There was a strong fluorescent signal at the tumor site, but no fluorescent signal was detected in other organs such as the heart, liver, spleen, lungs, and kidneys. For PDT treatment in vivo, when the tumor volume reached approximately 100 mm 3 , TPE-TeV-PPh 3 and TPE-TeV-Me were injected intratumorally. In order to aggregate the photosensitizer sufficiently, [16b,28a] PDT treatment was performed after 24 h post injection. As shown in Figure 4C, the tumor volume of the PBS control group continued to increase during the treatment process, and increased linearly in the later stage of the experiment. In addition, the nude mice that were treated with TPE-TeV-PPh 3 and TPE-TeV-Me + light showed moderate tumor growth inhibition. Intriguingly, the TPE-TeV-PPh 3 + light group showed 88% tumor growth inhibition. The tumor growth profile confirmed the antitumor PDT effect of TPE-TeV-PPh 3 upon light irradiation ( Figure 4D), and the data were statistically significant. Importantly, during the entire PDT treatment, none of the mice showed abnormal changes in body weights ( Figure 4E). After treatment, the important organs and tumors of all groups were collected and stained with H&E. By comparing the tumor sections of each group, it was found that the number of cancer cells in the TPE-TeV-PPh 3 + light group was significantly lower than that in the other groups, proving the effect of PDT on tumor growth inhibition ( Figure 4F and Figure S20). Meanwhile, H&E-stained section images of major organs showed no obvious morphological changes. These data demonstrate that TPE-TeV-PPh 3 and TPE-TeV-Me show low systemic toxicity and can be used in in vivo systems, confirming their clinical therapeutic potential.

Cell culture
Human liver cancer cell line 7402 and human breast cancer cell line MCF-7 were purchased from Shanghai Institute of Cell Biology in the Chinese Academy of Sciences. 7402 cells were cultured in Roswell Park Memorial Institute 1640 medium supplemented with 10% fetal bovine serum (FBS) and antibiotics (100 IU/mL penicillin and 100 µg/mL streptomycin), and MCF-7 was cultured in Dulbecco's modified eagle's medium (DMEM) supplemented with the same FBS and antibiotics. Both of the cell lines were incubated in a humidified atmosphere of 5% CO 2 at 37 • C.

Cell viability assay
Cells were seeded in 96-well plates at 5000 cells per well in 180 µL of complete medium, and incubated in a 5% CO 2 atmosphere or 1% O 2 and 5% CO 2 hypoxic environment at 37 • C for 24 h. Then the culture medium was replaced with freshly prepared culture medium containing 20 µL different tellurium viologens at different concentrations. Then the mixture solutions were exposed to 50 mW/cm 2 white light for 10 min, or incubated in the dark for 10 min. The cells were further incubated for 24 h, and then the medium was replaced with fresh culture medium and the MTT solution (5 mg/mL) was added. The cells were incubated for another 4 h to allow viable cells to reduce the yellow tetrazolium salt (MTT) into dark blue formazan crystals. Finally, 100 µL of lysis buffer was added to wells and incubated for another 4 h at 37 • C. The absorbance was measured at 490 nm using a Bio-Rad 680 microplate reader. The IC50 values were calculated using GraphPad Prism software (version 8.0) based on data from five parallel experiments.

Confocal laser scanning microscopy characterization
The 7402 cells were seeded at a density of 1 × 10 4 cells on a round coverslip (diameter 12 mm) in complete RPMI 1640 culture medium. After 24 h, they were treated with different tellurium viologens at the concentration of 30 µM for 2 h and lighted. And added H 2 O 2 (100 µM) in different groups. Then we added the mitochondrial targeting group MitoOrange (100 nM) to 7402 cells for 30 min. Cells were then washed twice with cold PBS and CLSM was used to observe the co-localization of mitochondria in cells. The data was analyzed using Image J.

In vivo fluorescence imaging
Treatment of four mice in the TPE-TeV-PPh 3 group, human tumor equal concentrations of H 2 O 2 (100 µM,100 µL) are injected into the tumor of mice. After 30 min, TPE-TeV-PPh 3 (100 µL) was injected into tumor. Subsequently, the mice were imaged with a PerkinElmer IVIS Spectrum in vivo fluorescence imaging system at 0.5, 1, 2, 3, and 4 h, respectively. Parameters for TPE-TeV-PPh 3 : Excitation filter: 430 nm; Emission filter: 520 nm; Acquisition setting: 500 to 540 nm; Exposure time: Auto. The autofluorescence was removed by spectral unmixing software. The fluorescence signal intensities were quantified by software.

In vivo tumor therapy study
In order to detect the PDT efficiency of different tellurium viologens, we studied their antitumor performance in vivo by intratumor administration. All sixteen mice were divided into four groups, four mice from each group. The intratumoral administration rate was 100 s/mL, the solute in the injection preparation was photosensitizer, and the mixed solvent was 99% PBS and 1% DMSO. Group I (PBS + light group), PBS administration (120 µL) followed by light irradiation (450 nm, 200 mW/cm 2 , 10 min); group II (TPE-TeV-Me + light group), administration with TPE-TeV-Me according to nude mouse body weight (10 mg/mL, 50 mg/kg) and followed by light irradiation (450 nm, 200 mW/cm 2 , 10 min); group III (TPE-TeV-PPh 3 ), administration with TPE-TeV-PPh 3 according to nude mouse body weight (10 mg/mL, 50 mg/kg) alone; group IV (TPE-TeV-PPh 3 + light group), administration with TPE-TeV-PPh 3 according to nude mouse body weight (10 mg/mL, 50 mg/kg) and followed by light irradiation (450 nm, 200 mW/cm 2 , 10 min). The treatment of frequency was once every 2 days. During the treatment period, the tumor volume of all mice was measured every 2 days using a vernier caliper. Then, the greatest longitudinal diameter (length) and the greatest transverse diameter (width) were used to calculate the tumor volume. Tumor volume V = length × width 2 /2. After 12 days post-treatment, tumors in all groups were harvested and weighed.

A C K N O W L E D G M E N T S
The authors are thankful for the financial assistance from the Natural Science Foundation of China (22175138 and 21875180), the Key Research and Development Program of Shaanxi (2021GXLH-Z023), and the Independent Innovation Capability Improvement Project of Xi'an Jiaotong University (PY3A066). The authors thank Dr. Lu Bai and Dr. Yu Wang at the Instrument Analysis Center of Xi'an Jiaotong University for HRMS and photoluminescence measurements.

C O N F L I C T O F I N T E R E S T
The authors declare no conflict of interest.

D ATA AVA I L A B I L I T Y S TAT E M E N T
All data are reported in this article and available upon request. O R C I D Gang He https://orcid.org/0000-0002-5319-7084