Sunlight driven E–Z isomerization of liquid crystals based on hexahydroxytriphenylene nano-templates for enhanced solid-state solar thermal energy storage

Abstract

Solar thermal fuels (STFs) are increasingly pivotal in addressing global energy demands, yet their widespread adoption is hindered by challenges such as low energy density, short half-life, and inadequate sunlight photoconversion efficiency. To enable large-scale utilization of STFs, innovative material design is imperative for efficient light-to-heat conversion. This study presents a new approach for achieving high energy density STFs by employing a hexahydroxytriphenylene liquid crystal (LC) core as a nano-template, tethering six tetra-ortho-substituted azobenzene units. The resulting oligomers exhibit discotic nematic mesophases, demonstrating excellent photostability, photocyclability, and prolonged half-lives of metastable Z states. High E to Z isomer conversion of up to 79.4% and 75.2% is achieved under direct sunlight with and without a bandpass filter, respectively, with high solar conversion efficiencies of up to 1.87%. Upon discharging, infrared (IR) thermal imaging reveals remarkable heat release values of up to 9.64 °C. The present study introduces an innovative method for developing high-performance STFs using LC nano-templates, surpassing prior approaches utilizing carbon nanomaterials.

---

1. Introduction

The quest for efficient and sustainable energy storage solutions has spurred considerable research into materials capable of harnessing and storing solar energy effectively. One promising avenue in this pursuit is the development of Solar Thermal Fuels (STFs), which can capture and release solar energy for various applications. However, the design and synthesis of STF materials that combine high energy density, stability, cyclability, long-term energy storage capability, affordability, and ease of handling remain critical challenges.^1,2 Several molecular structures have been explored in the quest for advanced solar energy storage materials. Norbornadiene-based STFs, for instance, have demonstrated a high energy density of 89 kJ mol⁻¹.³ However, their poor cyclability limits their practicality. Fulvalene-tetracarbonyl-diruthenium offers reasonable energy density and excellent cyclability but suffers from the drawback of being expensive due to the use of ruthenium.⁴ These limitations underscore the need for alternative molecular structures that can overcome these challenges. Among the various photoresponsive molecules investigated, azobenzene stands out as a promising candidate due to its ease of synthesis, functionalization, and stability.⁵ Despite its high cyclability, the development of advanced STFs based on azobenzene is impeded by its low energy density.⁶ Overcoming this limitation is crucial for realizing the full potential of azobenzene-based STFs.

To address this issue, several strategies have been explored such as structural modifications in azobenzene molecules by introducing substituents which can increase the intra- and intermolecular interactions to enhance their energy storage capabilities while maintaining their favourable properties such as cyclability and stability.⁷ Another approach is utilizing nanostructuring techniques such as combining azobenzene with other materials to create hybrid systems that synergistically enhance energy storage performance. For example, molecule-nanostructure hybrid STFs consisting of photoswitches attached covalently or non-covalently on rigid nanostructures such as carbon nanotubes or graphene have been developed which exhibited increased energy densities due to template-enforced steric strain effects.^8–12 Apart from the hybrid approach, preparation of dimers also resulted in enhanced energy densities of STFs.^13–15 For example, m-bisazobenzene derivatives with various alkoxy substituents exhibited twice isomerization enthalpy as compared to monoazobenzene derivatives.^16,17 Additionally, employing a dendritic approach i.e., grafting azobenzene chromophores onto multifunctional dendrimers also enabled controlled switching with solar energy storage.¹⁸ These fifth-generation dendrimers exhibited a 5.2 °C temperature increase on discharging. New design chemistries such as molecular rings formed using photoswitchable molecules connected by different linkers have also been investigated where STF properties can be tuned by using different photoswitches, linkers, ring sizes etc.¹⁹ Despite improved energy densities, these synthetically complex systems suffer from the uncontrollable morphological effects during charging and discharging cycles etc., limiting their chargeability.^20,21 All these above-mentioned efforts have only been made on the conventional azobenzene which undergoes E to Z isomer conversion in UV light, limiting the practical applicability of these systems for STFs.²² Moreover, these systems required heating for photoswitching and suffer from poor light penetration and non-uniform heat release on reverse isomerization.²³

Recently, we have reported STF systems based on liquid crystals (LCs).^24–26 The fact that these LC systems are dynamic self-assembling materials with ordered nanostructures that align directively to transfer charge or energy, mechanical flexibility, defect self-healing, and easier processability makes them advantageous for STFs.²⁷ Due to this, we observed an enhanced heat release at room-temperature of up to 6.5 °C resulting from efficient photoconversion in mesophases. Continuing to innovate in the design and synthesis of LC STF materials with enhanced heat release, in the present study, we have employed a discotic LC core i.e., a hexahydroxy triphenylene unit as a nano-template platform to attach six tetra-ortho-substituted azobenzene units. We envisaged that this approach can lead to improved STF performance due to increased energy density (resulting from six azobenzene units) as well as efficient photoconversion in visible-light (due to the dynamic nature of LCs) and thus can overcome the disadvantages resulting from the carbon nanomaterial templates (Fig. 1).

Fig. 1 The concept of triphenylene discotic LC nano-templates to enhance the energy density of STFs and their advantages over carbon nanomaterials.

2. Results and discussion

2.1. Synthesis and mesomorphic properties

Target compounds 1–3 were prepared as shown in Scheme 1. Synthetic procedures for compounds 5 and 7 are provided in the ESI.† For the synthesis of final compounds, respective compound 7 was reacted with compound 5 through the Williamson etherification reaction to obtain the final compounds 1–3 (see the ESI† for details). The produced compounds' structural characterization was done via elemental analysis, mass spectrometry, ¹H and ¹³C NMR and UV-vis (ESI, Fig. S1–S6†). Wide-angle X-ray scattering (WAXS), polarized optical microscopy (POM), differential scanning calorimetry (DSC), and thermogravimetric analysis were used to examine the compounds' thermal behavior (Fig. 2 and ESI, Fig. S7–S11†).

Scheme 1 Synthetic pathway to obtain desired mesogens 1–3. Reagents and conditions: (i) 1,10-dibromodecane, K₂CO₃, KI, TOAB, butanone, reflux 18 h, 31.7%; (ii) NaNO₂, HCl, NaOH, 3,5-difluorophenol/3,5-dichlorophenol, 0 °C-RT, 2 h, 70%; (iii) K₂CO₃, KI, butanone, reflux, 18 h, 45–72%.

Fig. 2 (a) Cross-polarized optical microscopy images of compound 1 and (b) 1D profile of the WAXS pattern in the mesophase (2D profile in the inset) upon cooling to 25 °C from the isotropic phase.

Compound 1 was solid, whereas compounds 2 and 3 were found to be LCs at room-temperature (25 °C). All three compounds showed similar mesophase to isotropic transition temperatures of 60 °C, whereas the appearance of a mesophase on cooling from isotropic melt was observed at 40 °C for all these compounds under POM. Interestingly, the transition temperatures of the crystalline state to the mesophase and vice versa as well as the associated enthalpies were found to vary according to the azobenzene ortho-substitution (ESI, Fig. S8 and Table S1†). Though the peaks for crystalline to LC transitions were observed in DSC due to high molecular ordering changes, peaks for LC to the isotropic transition were not observed due to considerable low molecular ordering changes.^28,29 In the mesophase, all of the compounds displayed birefringent textures (Fig. 2a and ESI, Fig. S9†). We used the aligned samples for WAXS studies to examine the mesophases of 1–3 (ESI, Fig. S10†). Compounds 1–3 demonstrated a broad peak in the wide-angle region and a weak peak in the small-angle region, supporting the presence of discotic nematic mesophase and the lack of any high order phase (Fig. 2b and ESI, Fig. S11†).

2.2. Photophysical studies

We further studied photoswitching behavior of compounds 1–3 by carrying out UV-vis analysis with regulated exposure to visible light. 10 μM solutions of 1–3 in either DMSO or toluene were employed for these studies. E (π → π*) and Z (n → π*) isomer peak intensities were monitored with different irradiation wavelengths for all the compounds (ESI, Fig. S12†).³⁰ Irradiation with a 430 nm LED for 10 minutes led to maximum E isomer formation for compounds 1 and 3, whereas, irradiation with 530 nm for 15 minutes led to the maximum Z isomer (Fig. 3a and ESI, Fig. S13†). Likewise, compound 2 yielded the maximum E isomer after five minutes of consecutive exposure to a 430 nm LED and irradiation with 530 nm for 10 minutes generated the maximum Z isomer. We further determined the proportion of photoisomers in each of the photostationary states (PSSs) by ¹H NMR (Table 1). Compound 1 was made up of 68% Z and 73.5% E isomer, compound 2 exhibited 74.6% E and 65% Z and compound 3 consisted of 78.7% E and 74.4% Z isomer in their respective PSSs in 7 mM solution in CDCl₃ (Fig. 3b and ESI, Fig. S14†). Cyclability experiments were also conducted for 20 cycles after alternate irradiations with LEDs to obtain Z and E PSSs (Fig. 3c and ESI, Fig. S15†).³¹ These studies confirm the stability, reliability, and suitability of the compounds for STF applications. Additionally, compounds 1–3 were subjected to 12-hour continuous 530 nm LED irradiation. The UV-vis spectra were recorded every 15 minutes to assess the compounds' photostability (Fig. 3d and ESI, Fig. S16†). These investigations demonstrated that every compound had exceptional photostability. Besides the optical trigger, the Z isomer can also spontaneously convert back to E over time and half of this time is referred to as half-life time. This half-life time for STFs directly relates to the molecules' thermal stability in their charged state as well as their long-term energy storage capability.³² To investigate this property, we studied the thermal Z to E isomerization kinetics for compounds 1–3 at different temperatures (ESI, Fig. S17†). Half-lives of Z isomers of 1–3 in 10 μM solutions at 25 °C as calculated from Arrhenius plots were found to be 33.47, 5.45 and 8.95 days, respectively (ESI, Fig. S18† and Table 2).³³ These values indicate excellent thermal stability and long-term energy storage capability of these STFs at room-temperature. Other thermodynamic parameters associated with the thermal reverse isomerization of the compounds were also calculated from Eyring–Polanyi plots (ESI, Fig. S19† and Table 2).³⁴

Fig. 3 (a) UV-vis absorption spectra of compound 1 (10 μM in DMSO) recorded without any irradiation and after irradiation with LEDs as mentioned in the inset. (b) ¹H NMR spectra of compound 1 after irradiation with LEDs (as mentioned) to achieve the respective E and Z PSSs showing the ratio of E and Z isomers. (c) Cyclic photoisomerization of compound 1 (10 μM in DMSO) after alternating irradiation to obtain Z and E PSSs and recording absorbance at λ_max (325 nm) at 25 °C. (d) Photostability measurements of compound 1 (10 μM in DMSO) carried out by 530 nm LED irradiation at 25 °C and recording absorbance at λ_max over 12 hours.

Table 1 E and Z isomer percentages obtained at PSSs for the synthesized compounds

Compound	Solution PSS^a		Film PSS^b
	E (%)	Z (%)	LED^c		Sunlight
					With an SY filter	Without a filter
			E (%)	Z (%)	Z (%)	Z (%)
a For 7 mM solution of compounds in CDCl3. To produce E and Z PSSs, 430 nm and 530 nm LEDs were used respectively, to irradiate all compounds. b For a thick film of 12.8 μm. c E and Z PSSs were obtained by irradiating films of compounds 1 and 3 with 430 nm and 530 nm LEDs and compound 2 with 395 nm and 625 nm LEDs.
1	73.5	68.0	79.6	76.6	78.1	70.1
2	74.6	65.0	76.3	85.0	73.0	69.3
3	78.7	74.4	82.6	77.4	79.1	74.7

---

Table 2 Half-life and other photophysical and thermodynamic parameters at 298 K for compounds 1–3

Compound	t ½	E a	A	ΔH^d	ΔS^e	Φ	ED^g	η (%)
								SY	SL	LED
a Half-life of the Z isomer at 298 K for 10 μM solution (in days) and in the solid-state (values indicated in brackets, in hours). b Activation energy (in kJ mol^−1). c Arrhenius constant (in s^−1). d Enthalpy of activation (in kJ mol^−1). e Entropy of activation (in J mol^−1 K^−1). f Quantum yield (in fraction) for E to Z isomerization. g Gravimetric energy density in kJ mol^−1. h Solar conversion efficiency (%) for E to Z conversion under an SY filter (SY), direct sunlight (SL) and a 530 nm LED (LED).
1	33.47 (5.36)	62.62 (85.13)	4.41 × 10^8 (3.13 × 10^10)	62.19 (82.14)	−98.67 (−46.16)	0.27	74.2	1.09	0.38	1.39
2	5.45 (14.16)	41.93 (57.33)	2.09 × 10^5 (3.04 × 10^7)	47.99 (55.74)	−79.49 (−49.21)	0.54	82.2	0.92	0.54	1.93
3	8.95 (7.45)	45.17 (77.12)	3.49 × 10^5 (4.23 × 10^9)	51.96 (74.39)	−66.12 (−65.76)	0.59	70.4	1.87	0.47	1.77

---

In addition, we also estimated half-lives of compounds 1–3 in solid-states by analyzing their thermal Z to E isomerization kinetics at different temperatures (ESI, Fig. S20†). The half-lives and other thermodynamic parameters in solid-state as observed from Arrhenius and Erying–Polanyi plots are listed in Table 2 (ESI, Fig. S21 and S22†).

2.3. Charging and discharging properties

When exposed to visible light, the energy contained in the synthesized photoactive molecules' metastable Z forms can be released (Scheme 2).³⁵ We first investigated % Z conversion of compounds 1–3 in both direct and filtered sunlight and when exposed to LED light. As in our earlier research, we used bandpass filters from Shopee, Edmund Optics and Neewer to obtain filtered sunlight (ESI, Fig. S23†).^24,25 In these investigations, 1.5 mg of sample was drop cast onto a 1.4 cm² glass substrate to create thin films of compounds 1–3 (Scheme 2 and ESI, Fig. S24, S25†). An estimate of the film's thickness was around 12.8 μm (Fig. 4a). First, E isomer rich states were obtained for all prepared films by irradiating them for 30 minutes at room temperature with LEDs (430 nm for 1 and 3 and 395 nm for 2), achieving an E isomer % of 79.6, 76.3, and 82.6, respectively. For five hours, these E-rich films of compounds 1–3 with various filters on them were stored in a glass-walled greenhouse (Fig. 4b and ESI, Fig. S26†). Compound 1 yielded a maximum Z% of 79.4 with the SY filter; compound 2 yielded a 68.5% conversion with the NR filter; and compound 3 yielded a 79.1% Z isomer with the most efficient filter, SY. It's interesting to note that the films stored in direct sunlight likewise displayed extraordinarily high conversion rates of up to 75.2%. Using the SY filter, we looked into the compounds' time-dependent chargeability in more detail (ESI, Fig. S27†). Over the course of five hours, the percentage of the Z isomer rose exponentially for each of the compounds. Compound 1 attained the Z PSS after two hours, whereas compounds 2 and 3 mostly photoconverted within 30 minutes with little increase after continued exposure (Table 1). The average intensity and temperature inside the greenhouse were recorded to be 598 W m⁻² and 51 °C, respectively. Time-dependent experiments were also conducted under direct sunlight over a period of 5 hours (Fig. 5 and ESI, Fig. S28†). Compound 1 attained a Z PSS of 68.3% after 2 hours and compound 2 attained a Z PSS of 68% after an hour, whereas compound 3 achieved a Z PSS of 73.4% after 2 hours. We further scrutinized the chargeability of compounds 1–3 using LEDs at room temperature (ESI, Fig. S29†). After irradiating with a 530 nm LED for 30 minutes, compound 1 obtained a Z PSS of approximately 75.1%. In contrast, compound 2 obtained a Z PSS of 85% conversion after one hour of continuous irradiation, and compound 3 obtained a Z PSS of 74.8% after 90 minutes of irradiation with the 530 nm LED. A comparison of time-dependent charging of compounds 1–3 under the SY filter, direct sunlight and LEDs is shown in Fig. 5.

Scheme 2 Method of preparation of films to study solid-state solar energy storage and release, with the charging and discharging mechanisms.

Fig. 4 (a) SEM image of a cross-section of the thin film of compound 1 showing the average thickness of the film. (b) Percentage of the Z isomer obtained after five hours of exposure to sunlight with and without different filters. Average temperature inside the greenhouse during the solar experiments was 50.2 °C and the average intensity of sunlight was 635 W m⁻².

Fig. 5 Percentage of the Z isomer (as measured by ¹H NMR in CDCl₃) obtained in thin films of compounds 1–3 upon (a) sunlight irradiation through an SY filter, (b) direct sunlight irradiation and (c) LED irradiation (530 nm for 1 & 3, 625 nm for 2), showing exponentially increasing Z isomer population for compounds over the time.

We further measured the quantum yields (Φ) for E to Z isomerization for compounds 1–3 using previous methods in the literature (ESI, Fig. S30 and S31†).^35–37Φ values were estimated to be around 0.27, 0.54 and 0.59 for compounds 1–3, respectively (Table 2). The solar energy conversion efficiencies (η) were also estimated in an AM 1.5 solar irradiation spectrum as per a previously reported method (see the ESI† for details).^38,39 The calculated η values are listed in Table 2. Compound 3 exhibited a maximum η value of 1.87% under the SY filter. This value is considerably higher than those of the other reported azobenzene systems with values mostly between 0.25 and 1.28%.³⁵ Interestingly it also surpasses those of previously reported nano-templated azobenzene derivatives on carbon nanomaterials.^22,40,41

For STFs, it's equally important to investigate the dischargeability as well as its chargeability.⁷ When we look at practical aspects of STFs, the discharging process should take place at room temperature. To investigate these properties, we have irradiated the charged, Z rich films with 430 and 395 nm LEDs until it reaches E PSSs and NMR spectra are recorded after regular intervals to investigate the conversion. All of the final compounds showed an exponential pattern in the E isomer conversion further reaching its PSSs with more than 65% E isomer population at room temperature (Fig. 6 and ESI, Fig. S32†). Among the final compounds, 3 exhibited rapid Z to E isomerization, reaching its PSS with more than 70% E isomer within 2 minutes of light exposure. Compound 1 showed comparatively delayed reverse isomerization among the final compounds, reaching its PSS with 76% E isomer after 15 minutes. Dischargeability of 2 was found to be the least among the final compounds with 64% E isomer population after 10 minutes at room temperature (on irradiation with a 395 nm LED).

Fig. 6 Percentage of the E isomer (as measured by ¹H NMR in CDCl₃) in thin films of compounds 1–3 upon irradiation with LEDs (430 nm for 1 & 3 and 395 nm for 2) showing an exponential increase of the E isomer until PSSs.

2.4. Optically triggered heat release from films

We further evaluated the heat release on optically triggered Z to E isomerization for compounds 1–3. For this, we prepared Z rich films and irradiated them with corresponding LEDs until they reached their E isomer PSSs and using an infrared (IR) thermal camera, the temperature increase on the film surface was observed. The temperature of the films increased noticeably due to optically triggered photoisomerization (Fig. 7 and Table 3). Among the final compounds, 3 showed maximum heat release with around a 9.64 °C temperature increase within two minutes of light exposure. Compounds 1 and 2 showed an average heat release of around 7.88 and 1.86 °C, respectively. The low heat release values for 1 and 2 as compared to 3 could be due to the slow discharging of fuels. We further evaluated the fatigue resistance of these fuels by performing charging and discharging up to five cycles where all the compounds exhibited excellent cyclability (ESI, Fig. S33†). The energy densities of compounds 1–3 were also estimated by their DSC analysis in Z PSSs and were estimated to be around 74.2, 82.2 and 70.4 kJ mol⁻¹, respectively (ESI, Fig. S34† and Table 2). It is to be noted that though the energy density of 3 is the least, it showed the highest heat release due to its efficient photoconversion. The thicker film of 3 didn't much affect the heat release though the rate of heat release was slowed down (ESI, Fig. S35†).²⁶ We also evaluated the heat release from the E rich film of 3 on irradiation with a 430 nm LED. The results showed no significant heat release reinforcing that heat release from Z rich films is solely due to the photothermal process associated with Z–E isomerization and not simply a result of light exposure (ESI, Fig. S35†).

Fig. 7 Heat release from Z rich films of compounds (a) 1, (b) 2, and (c) 3 as depicted in IR thermal camera images.

Table 3 Average heat release observed from the film surface after five consecutive charging and discharging cycles

Compound	Average temperature rise (°C)
1	7.88 (±0.06)
2	1.86 (±0.09)
3	9.64 (±0.13)

---

Earlier, Luo and co-workers observed a heat release of 10 °C upon optically triggered photoisomerization from carbon nanotube and azobenzene based hybrid STFs.⁹ These systems required 12 hours for charging and 8–10 min for discharging, whereas the half-life of the Z isomer was found to be 16 hours. Similarly, Feng et al. reported covalently functionalized azobenzene onto reduced graphene oxide which required 65 hours of charging time.⁴² Grossman and co-workers developed polymer based STFs composed of alkyl chain backbones and azobenzene side chains exhibiting a 10 °C temperature increase.⁴³ These systems needed 16 hours of charging time and 3–4 min for discharging with a half-life of 55 hours for the Z isomer. In search of enhancing STF performance, Zhang et al. demonstrated cation–π interaction in phase changeable azobenzene STFs.⁴⁴ These systems exhibited a heat release of 8 °C with a charging and discharging time of 20 and 5 minutes, respectively. Wang and coworkers developed azobenzene based dendrimers which exhibited a heat release of 5.2 °C.¹¹ These STFs possessed a half-life of 40 hours and a charging and discharging time of 2 hours and 5 minutes, respectively. In all these earlier reported studies, attempts to increase energy density in STFs have only been made using azobenzene units that employ UV light for charging. This limits their practical applicability since the UV light component in the solar spectrum is only up to 5%. Moreover, most of these systems need heating for charging to provide configurational freedom for photoisomerization. Bringing visible-light responsive LCs into play for STFs resolves all these issues due to their intriguing properties. We have earlier reported a heat release of 5.4 °C and 6.5 °C, at room-temperature from chiral nematic and discotic nematic LC STFs which was much higher than that of their non-LC counter parts.^24,25 In the present study, we achieved an enhanced heat release of around 9.64 °C resulting from increased energy density due to introduction of six azobenzene groups on triphenylene as well as the intermolecular interaction between triphenylene cores.

3. Conclusions

In conclusion, this study introduces a novel method for achieving high energy density, visible-light responsive STFs utilizing LC nano-templates. For these STFs, the dynamic nature of LCs has made it easier to efficiently capture, convert, and release solar energy. This system's unique characteristics may have an impact on lowering carbon footprints and dependency on fossil fuels. Notably, these STFs exhibit functionality in regions with lower sunlight intensity, promoting the broader adoption of renewable energy and a consequent reduction in greenhouse gas emissions. These STF materials' versatility is highlighted by their application as deicing and solar blanket coatings for solid-state devices. This research lays the groundwork for developing solar energy storage systems based on LC nano-templates optimized for diverse temperature ranges through molecular interactions.

Data availability

The data supporting this article have been included as part of the ESI.†

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

M. G. expresses gratitude to IGSTC for the WISER Grant (IGSTC/WISER 2023/MG/32/2023-24) and SERB for the Ramanujan Fellowship Grant (RJF/2021/000072). We thank Prof. Frank Würthner and his co-workers Dr Vladimir Stepanenko and Mr Philipp Kirchner at the Center for Nanosystems Chemistry of the University of Würzburg for hosting MG within the WISER program for four weeks and providing support with WAXS and MS measurements.

Notes and references

1. X. Xu and G. Wang, Small, 2022, 18, 2107473.
2. Y. Liu and J. C. Grossman, Nano Lett., 2014, 14, 7046–7050.
3. J. Orrego-Hernández, A. Dreos and K. Moth-Poulsen, Acc. Chem. Res., 2020, 53, 1478–1487.
4. Y. Kanai, V. Srinivasan, S. K. Meier, K. P. C. Vollhardt and J. C. Grossman, Angew. Chem., Int. Ed., 2010, 49, 8926–8929.
5. F. A. Jerca, V. V. Jerca and R. Hoogenboom, Nat. Rev. Chem., 2022, 6, 51–69.
6. L. Dong, Y. Feng, L. Wang and W. Feng, Chem. Soc. Rev., 2018, 47, 7339–7368.
7. E. N. Cho, D. Zhitomirsky, G. G. D. Han, Y. Liu and J. C. Grossman, ACS Appl. Mater. Interfaces, 2017, 9, 8679–8687.
8. A. M. Kolpak and J. C. Grossman, Nano Lett., 2011, 11, 3156–3162.
9. Y. Jiang, J. Huang, W. Feng, X. Zhao, T. Wang, C. Li and W. Luo, Sol. Energy Mater. Sol. Cells, 2019, 193, 198–205.
10. W. Feng, S. Li, M. Li, C. Qin and Y. Feng, J. Mater. Chem. A, 2016, 4, 8020–8028.
11. X. Xu, B. Wu, P. Zhang, H. Yu and G. Wang, Sol. RRL, 2020, 4, 1900422.
12. P. Porcu, A. S. Estrada-Montaño, M. Vonlanthen, F. Cuétara-Guadarrama, I. González-Méndez, K. Sorroza-Martínez, G. Zaragoza-Galán and E. Rivera, Dyes Pigm., 2022, 197, 109858.
13. M. Mansø, A. U. Petersen, Z. Wang, P. Erhart, M. B. Nielsen and K. Moth-Poulsen, Nat. Commun., 2018, 9, 1945.
14. A. Gonzalez, Q. Qiu, J. Usuba, J. Wan and G. G. D. Han, ACS Mater. Au, 2024, 4, 30–34.
15. D. Dong and T. Li, ChemPhotoChem, 2024, e202400007.
16. M. Morikawa, Y. Yamanaka, J. K. H. Hui and N. Kimizuka, RSC Adv., 2023, 13, 24031–24037.
17. M. Morikawa, Y. Yamanaka and N. Kimizuka, Chem. Lett., 2022, 51, 402–406.
18. X. Xu, P. Zhang, B. Wu, Y. Xing, K. Shi, W. Fang, H. Yu and G. Wang, ACS Appl. Mater. Interfaces, 2020, 12, 50135–50142.
19. E. Durgun and J. C. Grossman, J. Phys. Chem. Lett., 2013, 4, 854–860.
20. G. F. Martins, B. D. P. Cardoso, N. Galamba and B. J. C. Cabral, Energy Fuels, 2023, 37, 1731–1756.
21. H. Yu, J. Mater. Chem. C, 2014, 2, 3047–3054.
22. T. J. Kucharski, N. Ferralis, A. M. Kolpak, J. O. Zheng, D. G. Nocera and J. C. Grossman, Nat. Chem., 2014, 6, 441–447.
23. B. Zhang, Y. Feng and W. Feng, Nano-Micro Lett., 2022, 14, 138.
24. A. Krishna KM, S. Sony, S. Dhingra and M. Gupta, ACS Mater. Lett., 2023, 5, 3248–3254.
25. M. Gupta and Ashy, Adv. Energy Mater., 2024, 2303845.
26. M. Gupta, A. Krishna KM, S. Sony and S. Dhingra, Adv. Funct. Mater., 2024, 2409245,  DOI:10.1002/adfm.202409245.
27. J. Uchida, B. Soberats, M. Gupta and T. Kato, Adv. Mater., 2022, 34, 2109063.
28. A. Concellón, M. Marcos, P. Romero, J. L. Serrano, R. Termine and A. Golemme, Angew. Chem., Int. Ed., 2017, 56, 1259–1263.
29. A. Concellón, R. Termine, A. Golemme, P. Romero, M. Marcos and J. L. Serrano, J. Mater. Chem. C, 2019, 7, 2911–2918.
30. D. Bléger, J. Schwarz, A. M. Brouwer and S. Hecht, J. Am. Chem. Soc., 2012, 134, 20597–20600.
31. X. Xu, B. Wu, P. Zhang, Y. Xing, K. Shi, W. Fang, H. Yu and G. Wang, ACS Appl. Mater. Interfaces, 2021, 13, 22655–22663.
32. C. Knie, M. Utecht, F. Zhao, H. Kulla, S. Kovalenko, A. M. Brouwer, P. Saalfrank, S. Hecht and D. Bléger, Chem.–Eur. J., 2014, 20, 16492–16501.
33. A. Miniewicz, H. Orlikowska, A. Sobolewska and S. Bartkiewicz, Phys. Chem. Chem. Phys., 2018, 20, 2904–2913.
34. H. Eyring, J. Chem. Phys., 1935, 3, 107–115.
35. Y. Shi, M. A. Gerkman, Q. Qiu, S. Zhang and G. G. D. Han, J. Mater. Chem. A, 2021, 9, 9798–9808.
36. W. Sun, Z. Shangguan, X. Zhang, T. Dang, Z.-Y. Zhang and T. Li, ChemSusChem, 2023, 16, e202300582.
37. X. Xu, B. Wu, P. Zhang, Y. Xing, K. Shi, W. Fang, H. Yu and G. Wang, ACS Appl. Mater. Interfaces, 2021, 13, 22655–22663.
38. Y. He, Z. Shangguan, Z. Zhang, M. Xie, C. Yu and T. Li, Angew. Chem., Int. Ed., 2021, 60, 16539–16546.
39. A. Dolai, S. Bhunia, K. Manna, S. Bera, S. M. Box, K. Bhattacharya, R. Saha, S. Sarkar and S. Samanta, ChemSusChem, 2024, e202301700.
40. W. Luo, Y. Feng, C. Qin, M. Li, S. Li, C. Cao, P. Long, E. Liu, W. Hu, K. Yoshino and W. Feng, Nanoscale, 2015, 7, 16214–16221.
41. Y. Jiang, J. Huang, W. Feng, X. Zhao, T. Wang, C. Li and W. Luo, Sol. Energy Mater. Sol. Cells, 2019, 193, 198–205.
42. W. Luo, Y. Feng, C. Cao, M. Li, E. Liu, S. Li, C. Qin, W. Hu and W. Feng, J. Mater. Chem. A, 2015, 3, 11787–11795.
43. D. Zhitomirsky, E. Cho and J. C. Grossman, Adv. Energy Mater., 2015, 6, 1502006.
44. T. Song, H. Lei, F. Cai, Y. Kang, H. Yu and L. Zhang, ACS Appl. Mater. Interfaces, 2022, 14, 1940–1949.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d4ta05275g

---