The effect of thionation of the carbonyl group on the photophysics of compact spiro rhodamine-naphthalimide electron donor–acceptor dyads: intersystem crossing, charge separation, and electron spin dynamics

Abstract

Herein, a spiro rhodamine (Rho)-thionated naphthalimide (NIS) electron donor–acceptor orthogonal dyad (Rho-NIS) was prepared to study the formation of a long-lived charge separation (CS) state via the electron spin control approach. The transient absorption (TA) spectra of Rho-NIS indicated that the intersystem crossing (ISC) occurs within 7–42 ps to produce the ³NIS state via the spin orbit coupling ISC (SOC-ISC). The energy order of ³CS (2.01 eV in n-hexane, HEX) and ³LE states (1.68 eV in HEX) depended on the solvent polarity. The ³NIS state having n–π* character and a lifetime of 0.38 μs was observed for Rho-NIS in toluene (TOL). Alternatively, in acetonitrile (ACN), the long-lived ³CS state (0.21 μs) with a high CS state quantum yield (Φ_CS, 97%) was produced with the ³NIS state as the precursor and the CS took 134 ps. On the contrary, in the case of the reference Rho-naphthalimide (NI) Rho-NI dyad without thionation of its carbonyl group, a long-lived CS state (0.94 μs) with a high energy level (E_CS = 2.12 eV) was generated even in HEX with a lower Φ_CS (49%). In the presence of an acid, the Rho unit in the Rho-NIS adopted an open form (Rho-o) and the ³NIS state was produced within 24–47 ps with the ¹Rho-o state as the precursor. Subsequently, slow intramolecular triplet–triplet energy transfer (TTET, 0.11–0.60 μs) produced the ³Rho-o state (9.4–13.6 μs). According to the time-resolved electron paramagnetic resonance (TREPR) spectra of NIS-NH₂, the zero-field splitting (ZFS) parameter |D| and E of the triplet state were determined to be 6165 MHz and −1233 MHz, respectively, indicating that its triplet state has significant nπ* character, which was supported by its short triplet state lifetime (6.1 μs).

---

Introduction

Charge separation (CS) is important in fundamental photochemistry studies,^1–15 as well as in photocatalysis,^4,6,16 photovoltaics,^17–20 photodynamic therapy,²¹ thermally activated delayed fluorescence (TADF) emitters for application in OLEDs,^22–26etc. One of the major goals in this area is to prolong the CS state lifetime, i.e., to obtain a long-lived CS state, where the charge recombination (CR) should be thousands of time slower than the CS process.^2,3,27,28 The electron transfer in organic electron donor–acceptor dyads can be described by the Marcus equation:^29–31In the equation, h and k_B represent the Planck and Boltzmann constants, respectively. ΔG⁰_ET is the Gibbs free energy change of the electron transfer, λ is the total reorganization energy and V is the electronic matrix element. According to this equation, a few strategies have been developed to prolong the CS state lifetime. For example, reducing the electronic coupling matrix elements between the initial state and final state of the CS process, usually by introducing long and saturated linkers between the donor and acceptor. However, this method suffers from the drawback of challenging synthesis and slow charge transfer kinetics, which may be overcome by other relaxation processes of the ¹LE excited state (LE: locally excited), such as the direct decay of the S₁ state to the ground state (S₀). Alternatively, the Marcus inverted region effect was also exploited to prolong the CS state lifetime, although this effect is often less profound than theoretically predicted.

Recently, a new method to prolong the lifetime of the CS state was developed, i.e., using the electron spin control effect.^3,28,32–36 The CS and CR processes are electron spin selective, i.e., given that the CS has a singlet ¹LE precursor, the formation of the ¹CS state is dominant over the formation of the triplet ³CS state, considering that the electron exchange energy (J) of the unpaired electrons is large.³ Under this condition, the formation of the radical ion pair or the spin-correlated radical pair (SCRP) is unlikely.^37–39 Alternatively, an ³LE precursor will dominantly give the ³CS state, where the CR from the ³CS state to the S₀ state is slower than the CR of the ¹CS state to produce the S₀ state because the ³CS → S₀ process is electron spin forbidden. Thus, it is highly desirable to produce the ³CS state rather than the ¹CS state. Therefore, efficient intersystem crossing (ISC) of the electron donor or acceptor will be beneficial for the formation of a long-lived ³CS state.^3,11,40

Concerning this, transition metal complexes have been used for the construction of electron donor–acceptor dyads to obtain the long-lived ³CS state because the ISC in these metal complexes is fast and efficient, usually producing the ³MLCT state (MLCT: metal-to-ligand-charge-transfer).^{32,33,35,41–44} The fast ISC is due to the strong spin orbit coupling (SOC) effect of the transition metal ion. However, these precious metal complexes are expensive and the strong SOC may shorten the CS state lifetime.³⁵ Thus, heavy atom-free organic molecular systems showing efficient ISC are highly desirable for the preparation of electron donor–acceptor dyads to attain long-lived CS states by exploiting the spin control effect. However, this is challenging because very few heavy atom-free organic molecular systems show efficient ISC, and only limited examples of this approach have been reported.^3,45

Recently, we used the spin–orbit charge transfer ISC (SOCT-ISC) for this purpose,⁴⁰ which requires an orthogonal geometry between the donor and acceptor units in the molecular structure of the dyad.^5,46–48 However, the ISC is slow for our previously reported dyad.^11,40 Thus, it is clear that more methods for the formation of the ³CS state via the electron spin control effect should be studied. We noted that the thionation of the carbonyl group of a chromophore may induce efficient ISC in the otherwise strongly fluorescent chromophores.^49–55 In this case, the molecular derivatization involves the replacement of only one atom. Another advantage of thionation is the enhanced electron-accepting ability of the resulting thionated chromophore.⁵⁶ Inspired by the previously reported spiro donor–acceptor dyad based on a rhodamine (Rho) and naphthalimide (NI) dyad molecular structure to obtain the long-lived ³CS state via the electron spin control method and to enhance the ISC to populated the ³LE state,⁴⁰ herein we performed further derivatization, i.e., thionation of the NI unit, to enhance the ISC ability and the electron-accepting ability of the NI unit (Rho-NIS, Scheme 1). The electron transfer and ISC processes of the compounds were studied via steady state UV-vis absorption and fluorescence spectroscopy, femtosecond and nanosecond transient absorption spectroscopic methods, timed-resolved electron paramagnetic resonance (TREPR) spectroscopy, and theoretical computations.

Scheme 1 Molecular structures of rhodamine-thionated naphthalimide derivatives. (a) Na₂Cr₂O₇·2H₂O, AcOH, N₂, reflux, 5 h, and yield: 55%. (b) n-butylamine, EtOH, N₂, reflux, 2 h, and yield: 54%. (c) SnCl₂·2H₂O, Con. HCl, EtOH, N₂, RT, 15 min, and yield: 81%. (d) Lawesson's reagent, dry toluene, N₂, reflux, 21 h, and yield: 28%. (e) Acetic anhydride, acetic acid, 130 °C, 5 h; and yield: 26%. (f) POCl₃, dry dichloromethane (DCM), reflux, 5 h; n-butylamine, dry acetonitrile (ACN), Et₃N, reflux, 25 h; and yield: 42%. (g) POCl₃, dry DCM, reflux, 5 h; NIS-NH₂, dry ACN, Et₃N, reflux, 25 h; and yield: 50%.

Experimental

General method

All chemicals used in the synthesis were analytically pure. The solvents for the synthesis were dried before use. UV-vis spectra were recorded using a UV-2550 spectrophotometer (Shimadzu Ltd, Japan). Fluorescence emission spectra were recorded using an FS5 spectrofluorometer (Edinburgh Instruments Ltd, UK). Fluorescence lifetimes were recorded using an OB920 luminescence lifetime spectrometer (Edinburgh Instrument Ltd, UK), and a picosecond EPL laser was used for excitation. Fluorescence quantum yields (Φ_F) were measured using an absolute photoluminescence quantum yield spectrometer (Quantaurus-QY Plus C13534-11, Hamamatsu Ltd, Japan).

Synthesis of Rho-NIS

Under an N₂ atmosphere, rhodamine B (160.0 mg, 0.33 mmol) was dissolved in dry dichloromethane (DCM, 5 mL), and then POCl₃ (0.2 mL) was added dropwise. The mixture was refluxed and stirred for 5 h. The solvent was evaporated under reduced pressure, and then the crude product was dissolved in dry CH₃CN (8 mL). The mixture was added dropwise to a solution of compound NIS-NH₂ (60.1 mg, 0.20 mmol) and Et₃N (0.4 mL) in CH₃CN (7 mL), and then the mixture was refluxed and stirred for 22 h. After the reaction was finished, the solvent was evaporated under reduced pressure, and the crude product was purified by column chromatography (silica gel, DCM) to give the product as a brown solid (72.4 mg, yield: 50%). M.p.: 227.6–228.8 °C. ¹H NMR (CDCl₃, 400 MHz): δ 8.75 (d, J = 7.5 Hz, 1H), 8.59 (d, J = 8.3 Hz, 1H), 8.08 (d, J = 7.4 Hz, 1H), 7.73–7.61 (m, 3H), 7.35–7.27 (m, 2H), 6.72–6.68 (m, 2H), 648–6.42 (m, 2H), 6.26–6.20 (m, 2H), 5.93 (s, 1H), 5.30–5.25 (m, 2H), 3.42–3.36 (m, 4H), 3.19–3.15 (m, 4H), 1.89–1.82 (m, 2H), 1.47–1.41 (m, 2H), 1.21 (t, J = 7.0 Hz, 6H), 1.03–0.96 (m, 9H). ¹³C NMR (CDCl₃, 125 MHz): δ 191.01, 190.61, 166.93, 154.27, 153.60, 151.63, 149.29, 148.92, 137.92, 137.48, 133.31, 131.86, 130.57, 129.49, 129.32, 129.07, 128.75, 127.61, 126.65, 124.63, 124.13, 123.78, 108.39, 107.59, 106.51, 105.69, 97.84, 97.61, 54.87, 44.49, 44.32, 31.94, 31.60, 29.67, 27.35, 22.66, 20.02, 14.13, 13.76, 12.55, 12.39. HRMS-ESI ([C₄₄H₄₄N₄O₂S₂ + H]⁺), m/z: calcd: 725.2984; found: 725.2961.

Single crystal X-ray diffraction

Single crystals of Rho-NIS and NIS-NHAc were obtained via the slow diffusion of a layer of n-hexane (HEX) into the DCM solution of the compounds. The X-ray diffraction data of the single crystals were measured and collected on a Bruker D8 Venture CMOS-based diffractometer with graphite monochromatized Mo Kα radiation (λ = 0.71073 Å) at 120 K, using the SMART and SAINT programs. The X-ray diffraction data were analyzed with the ShelXT structure solution program using Intrinsic Phasing and refined with the ShelXL refinement package using least squares minimization implemented in Olex2.⁵⁷ CCDC numbers are 2294224 (Rho-NIS) and 2294222 (NIS-NHAc), respectively.†

Electrochemical studies

Cyclic voltammetry curves were recorded using a CHI610D electrochemical workstation (CHI instruments, Inc., Shanghai, China). A platinum electrode was used as the counter electrode and a glassy carbon electrode as the working electrode. The ferrocenium/ferrocene (Fc⁺/Fc) redox couple was used as the internal reference. Spectroelectrochemistry was performed using a 0.1 cm path length quartz electrochemical cell. Gauze platinum acted as the working electrode and platinum wire as the counter electrode. The potential was regulated using a CHI610D electrochemical workstation and the spectra were recorded using an Agilent 8453 UV-vis spectroscopy system (Agilent Technologies Inc., USA). In both cases, Bu₄N[PF₆] was used as the supporting electrolyte and the Ag/AgNO₃ (0.1 M in ACN) couple was used as the reference electrode. The samples were deaerated with N₂ for ca. 15 min before measurement, and the N₂ atmosphere was maintained during the measurement.

Nanosecond transient absorption spectroscopy

Nanosecond transient absorption spectra were studied using an LP980 laser flash photolysis spectrometer (Edinburgh Instruments Ltd, UK), and the signal was digitized with a Tektronix TDS 3012B oscilloscope. The collinear configuration of the pump and probe beams was used for the measurements. The sample solutions were purged with N₂ for 15 min before measurement and excited with a nanosecond pulsed laser (OpoletteTM 355II + UV nanosecond pulsed laser. OPOTEK, USA). The typical laser power was 5 mJ per pulse. The data were analyzed using L900 software.

Femtosecond transient absorption spectroscopy

The fs-TA experiments were performed using a Ti:sapphire laser system with an ∼100 fs pulse duration and 1 kHz repetition rate (Astrella, Coherent) and a commercial ultrafast transient absorption spectrometer (Ultrafast System, Helios). The excitation wavelength was determined from the steady-state UV-vis absorption spectra and generated in an optical parametric amplifier (TOPAS, Light Conversion). The magic angle between the probe and the pump beam polarization direction was used. The Surface Xplorer and Glotaran software were used for processing the experimental data based on the data after chirp correction.^58,59

Time-resolved electron paramagnetic resonance spectra

Samples were studied in a frozen solution of TOL/2-methyltetrahydrofuran (2-MeTHF, 1/1, v/v) with a concentration of 1 × 10⁻⁴ M and transferred to quartz tubes (outer diameter: 4.5 mm and inner diameter: 3.0 mm). The time-resolved continuous-wave (CW) EPR measurements were performed on an X-band and Q-band EPR Elexsys E-580 spectrometer (Bruker) with a dielectric ring X-Band ER 4118X-MD5-W1 resonator at a temperature of 80 K. The oxygen was removed with five freeze–pump–thaw cycles. Optical excitation was carried out using an Nd:YAG pulse laser (LQ629 Solar LS) at a wavelength of 355 nm with a pulse energy of 1 mJ and frequency of 100 Hz. The spectra were simulated using the EasySpin package based on Matlab.⁶⁰

DFT calculations

The geometries of the compounds were optimized using density functional theory (DFT) with the B3LYP functional and 6-31G(d) basis set. The spin density surfaces at the optimized triplet state geometries were visualized. There were no imaginary frequencies for all optimized structures. The triplet excited state energy of the compounds was calculated by time-dependent DFT (TDDFT) with the B3LYP functional and 6-31G(d) basis set based on the optimized ground-state geometry. Natural transition orbital (NTO) analysis was performed using the Multiwfn program.⁶¹ All calculations were performed using Gaussian 09.⁶²

Results and discussion

Molecular structure design rationale

Previously, we studied a spiro electron donor–acceptor dyad based on rhodamine and NI units to realize the long-lived ³CS state via electron spin control, which was achieved by the SOCT-ISC mechanism to produce the ³LE precursor in the CS process.⁴⁰ However, the ISC of this heavy atom-free dyad is slow (¹CS → ³NI takes ca. 8 ns), which is not beneficial for efficient secondary CS to obtain the long-lived ³CS state. Thus, to improve the ISC kinetics, and consequently the CS yield, herein we performed the thionation reaction of the carbonyl group of the NI electron acceptor in the dyad to prepare a new dyad (Rho-NIS, Scheme 1). The advantage of this one-atom replacement derivatization includes a red-shifted absorption,⁵¹ increased electron-accepting ability,⁵⁶ and accelerated ISC process. Previously, it was reported that a thionyl carbonyl-containing chromophore showed ultrafast ISC (0.6 ps).^49,63 The perturbation in the molecular structure is negligible because only two atoms of the previously reported dyads are replaced (Scheme 1). The compounds NIS and NIS-NH₂ were used in this study as reference compounds (Scheme 1). The synthesis of the compounds is based on the routine derivatization methods of rhodamine and NI chromophores,^40,64 the yields were satisfactory, and the molecular structures were fully characterized (see Experimental).

The molecular structures of Rho-NIS and NIS-NHAc were determined by single-crystal X-ray diffraction (Fig. 1). Each unit cell of Rho-NIS and NIS-NHAc contains four molecules and has a triclinic crystal structure. X-ray diffraction showed that the centroid-to-centroid distance between xanthene and the NIS units in Rho-NIS is 5.3 Å and the dihedral angle between the xanthene and NIS units is 53.0° (Fig. 1). In the case of the previously reported Rho-NI dyad, the dihedral angle between xanthene and the NI moiety was 70.8°.⁴⁰ The xanthene moiety adopts a planar geometry, which was the same as observed for Rho-NIS (Fig. 1a). We also obtained a single crystal of the reference compound NIS-NHAc (Fig. 1b), where its C=S bond length is 1.64 Å, which is close to that in Rho-NIS (1.64 Å). Moreover, the C–N bond length at the 4-position of NI in NIS-NHAc is 1.42 Å, while that in Rho-NIS is 1.43 Å. It should be noted that the C=S bond length is longer than the corresponding C=O bond length (1.24 Å),⁴⁰ and the C–N bond length at the linkage between xanthene and the NI moiety is 1.43 Å.⁴⁰ These results show that the perturbation of the sulfur substitution has a negligible impact on the geometry of the dyad.

Fig. 1 The ORTEP view of the molecular structures of (a) Rho-NIS and (b) NIS-NHAc determined by single-crystal X-ray diffraction (50% probability thermal ellipsoids). Hydrogen atoms are omitted for clarity.

Steady-state UV-vis absorption and luminescence spectra

The UV-vis absorption of the compounds was studied (Fig. 2a). NIS showed an absorption band centered at 438 nm, which is redshifted by 110 nm compared to that of the native NI chromophore.^65–67 In the case of NIS-NH₂, a further red shift to 526 nm was observed, which is due to the attachment of the amino group at the 4-position of the NI moiety.⁵¹ Interestingly, Rho-NIS showed a major absorption band centered at 453 nm, which is very close to that of NIS. This result was rationalized by the acetylation of the amino group and the restricted conformation, which reduced the electron-donating ability and π-conjugation of the lone electron pairs on the N atom with the NI chromophore.^68,69 It should be noted that the previously reported analogue (Rho-NI, without thionation of the carbonyl group) showed absorption in the range of <400 nm,⁴⁰ whereas that of Rho-NIS was >400 nm, which is beneficial for harvesting visible light.

Fig. 2 (a) UV-vis absorption spectra of the compounds in TOL. (b) The evolution of the UV-vis absorption spectra of Rho-NIS with the incremental addition of trifluoroacetic acid (TFA) and (c) with the addition of TFA (c = 40 μM) or triethylamine (TEA, neat, 10 μL) in MeOH. c = 1.0 × 10⁻⁵ M. 25 °C.

Rhodamine is known to have two structures, i.e., opened and closed structures, and the transformation between these two structures can be realized in the presence of acid and base.^70–75 Interestingly, the attachment of the NIS moiety to the rhodamine chromophore does not change the opening and closing reversible transformation of the rhodamine moiety in the presence of acid and base (Fig. 2b and c, respectively). In the presence of trifluoroacetic acid (TFA), the Rho unit in Rho-NIS transforms from the closed form to the opening form, and accordingly a strong, broad absorption in the visible spectral region of 500–650 nm was observed (Fig. 2b). The absorption band of the open form of Rho disappeared with the addition of a base such as triethylamine (TEA) (Fig. 2c). This property may endow the compound with additional functionality, such as acid-activated PDT capability. Moreover, the evolution of the UV-vis absorption spectra of Rho-NIS upon the ring-opening reaction indicated that the blue-shifted absorption of the NIS-NH₂ unit in Rho-NIS is mainly due to the acetylation reaction, not the sterically encumbered microenvironment of the N atom connected to the spiro C atom. This is because for Rho-NIS with the Rho unit in the open form (no conformation restriction), the absorption band centered at 453 nm did not change to 526 nm (that of NIS-NH₂).

Fig. 3a shows that the fluorescence of the closed-ring structure of the Rho unit is quenched in Rho-NIS, where the possible reason for this is electron transfer or Förster resonance energy transfer (FRET) to the NIS moiety. Based on the femtosecond transient absorption (fs-TA) spectral studies (see later section), charge transfer from Rho to NIS was not observed in TOL upon photoexcitation. Therefore, FRET is the most probable reason for the fluorescence quenching of the closed ring structure of Rho. Fig. 3b shows that the fluorescence of NIS-NH₂ (Φ_F = 0.8%) was quenched in Rho-NIS (Φ_F = 0.3%), and we attribute this quenching to the acetylation of the amino group in Rho-NIS, which was also proven by the weaker fluorescence of the reference compound NIS-NHAc (Φ_F = 0.2%).

Fig. 3 Fluorescence emission spectra of the compounds in TOL. (a) λ_ex = 310 nm and (b) λ_ex = 420 nm. Optically matched solutions were used (all solutions show the same absorbance at the excitation wavelength, A_ex = 0.1). Asterisks in (a) and (b) stand for the Raman scattering peak of the solvent. (c) Fluorescence emission spectra of Rho-NIS with the addition of TFA (c = 40 μM) or TEA (neat, 10 μL) in MeOH. (d) The comparison of fluorescence emission spectra of Rho and Rho-NIS with the addition of TFA (c = 40 μM) in MeOH. λ_ex = 540 nm. c = 1.0 × 10⁻⁵ M. 25 °C.

In the case of Rho-NIS, in the presence of trifluoroacetic acid (TFA), the fluorescence was enhanced and peaked at 583 nm (Fig. 3c), which is attributed to the Rho moiety in the open form. Interestingly, the fluorescence of Rho-NIS (Φ_F = 0.5%, Table 1) was quenched compared to Rho under the same condition (Φ_F = 44.2%, Fig. 3d). This is probably due to the charge transfer from the Rho moiety in the open form to the NIS moiety. In this case, no FRET to the NIS moiety occurs. The fluorescence decay trace at 583 nm of Rho-NIS under this condition has a biexponential feature and the lifetime is 0.64 ns (population rate: 70%)/2.3 ns (30%) (Fig. S17, ESI†). The component with the lifetime of 2.3 ns is attributed to the open form of the Rho moiety in Rho-NIS, which is the same with that of Rho with the addition of TFA, showing mono-exponential decay with the lifetime of 2.3 ns (Fig. S17, ESI†). The short-lived component with the lifetime of 0.64 ns further indicates that there is a non-radiative transition process from the singlet state of the Rho moiety in the open form.

Table 1 Photophysical properties of the compounds^a

Compounds	λ abs	ε	λ F	Φ F	τ F	Φ Δ	λ P	τ P
a In TOL. b Maximal UV-vis absorption wavelength. c = 1.0 × 10^−5 M in nm. c Molar absorption coefficient, in 10^4 M^−1 cm^−1. d Maximal fluorescence emission wavelength in nm. e Fluorescence quantum yields in %. f Fluorescence lifetime. c = 1.0 × 10^−5 M in ns. g Singlet oxygen quantum yields. h Maximum phosphorescence emission wavelength in MeTHF in nm at 77 K. i Phosphorescence lifetimes (λex = 445 nm), c = 1.0 × 10^−4 M, 77 K. j Not observed. k In MeOH. l Not measured. m Literatures values, ref. 76. n λ ex = 430 nm. o λ ex = 560 nm.
NIS	438	1.94	—^j	0.4	—^j	0.46; 0.04^k	725	1.6 ms
NIS-NH2	526	0.96	506	0.8	8.4 ns	0.95; 0.58^k	—^j	—^j
NIS-NHAc	463	1.80	494	0.2	4.8 ns	1.0; 0.46^k	—^l	—^l
Rho	313	1.32	391	1.1	1.7 ns (73%)	0.27; 0^k	—^j	—^j
					4.7 ns (27%)
Rho + TFA^k	557^m	6.01^m	578	44.2	2.3 ns	—^j	—^j	—^j
Rho-NIS	453	1.88	495	0.3	—^j	0.72; 0^k	740	7.0 μs (89%)
								16.2 μs (11%)
Rho-NIS + TFA^k	559	6.73	583	0.5	0.64 ns (70%)	0.87^n	—^l	—^l
					2.3 ns (30%)	0.43^o

---

To determine the energy of the ³NIS state of the compounds, their phosphorescence emission spectra in frozen 2-methyltetrahydrofuran (2-MeTHF) at 77 K were studied (Fig. S18, ESI†). The phosphorescence emission bands were observed in the range of 700–825 nm under these conditions. Based on the similar spectra observed for NIS and Rho-NIS, the phosphorescence bands are attributed to the ³NIS state. According to the on-set of the phosphorescence band on the high-energy side (can be approximated as the vibrational 00 transition of T₁ → S₀ radiative relaxation), the ³NIS state energy was approximated to be 1.71 eV for NIS and 1.68 eV for Rho-NIS (Table 1). These values are close to the ³NIS state energy (1.65 eV) calculated using the ZINDO/S semiempirical quantum chemical method and slightly higher than that estimated by DFT calculation (1.57 eV).^51,77

As an approximation of the ISC efficiency, the singlet oxygen quantum yields (Φ_Δ) of the compounds were measured (Table 1). In the case of NIS, the Φ_Δ in TOL and MeOH are 46% and 4%, respectively, which are close to the reported Φ_Δ of NIS in TOL (32%).⁵¹ Interestingly, the reported NIS analogue, thionated perinone with a single thionated carbonyl group, has the Φ_Δ of 100%.⁵²NIS-NH₂ has higher Φ_Δ (95% in TOL and 58% in MeOH) compared to NIS. We propose that although the ISC quantum yields in all these thionated compounds may be close to unity, the singlet oxygen quantum yields are dependent on the triplet state lifetime of the compounds (see later section). A short triplet state lifetime may reduce the singlet oxygen quantum yield, although for Rho-NIS, Φ_Δ is 72% in TOL and no singlet oxygen photosensitizing was observed in MeOH upon excitation at the NIS moiety (λ_ex = 430 nm). In the presence of acid, the Φ_Δ of Rho-NIS in MeOH increased to 87% upon excitation at the NIS moiety (430 nm). Also, the Φ_Δ was 43% upon excitation at the rhodamine moiety in the open form (560 nm). In the case of Rho in the open form, no singlet oxygen photosensitizing was observed.

Electrochemistry study

To calculate the Gibbs free energy changes (ΔG⁰_CS) in the photo-induced CS and determine the energy of the CS states, the redox potentials of the compounds were studied via cyclovoltammetry (Fig. 4a and b). In the case of the Rho reference compound, two reversible oxidation waves were observed at +0.51 V and +0.64 V (vs. Fc/Fc⁺), respectively. No reduction wave was observed in the potential range applied in this study. Alternatively, in the case of NIS, one reversible reduction wave was observed at −1.09 V, and this wave was anodically shifted by 0.66 V compared to the native NI (unthionated, −1.75 V). This result indicates that NIS is a stronger electron acceptor than the un-thionated NI. With the attachment of the amino substituent, NIS-NH₂ showed a cathodically shifted reduction wave at −1.29 V, indicating that NIS-NH₂ is a poorer electron acceptor than NIS. This is reasonable because that the –NH₂ in NIS-NH₂ is an electron-donating group. In the case of Rho-NIS, two irreversible oxidation waves were observed at +0.52 V and +0.87 V, as well as a pseudo reversible reduction wave at −1.07 V. It should be noted that the aminoNIS moiety in Rho-NIS is similar to NIS, not NIS-NH₂. This result is in agreement with the UV-vis absorption spectral studies.

Fig. 4 (a) and (b) Cyclic voltammograms of the compounds in deaerated ACN. Ferrocene (Fc/Fc⁺) was used as the internal reference (set as 0 V in the cyclic voltammograms). 0.10 M Bu₄N[PF₆] as the supporting electrolyte and Ag/AgNO₃ as the reference electrode. Scan rate: 100 mV s⁻¹. c = 1.0 × 10⁻³ M. UV-vis absorption spectra of (c) Rho-NIS and (d) NIS chemically reduced with tetrabutylammonium fluoride (TBAF) to generate NIS˙⁻ in deaerated DMF. c = 3.0 × 10⁻⁵ M. 25 °C.

Based on the electrochemical data, the driving force for the photo-induced CS, i.e. ΔG⁰_CS, was calculated using the Rehm–Weller equation (eqn (S2)–(S5), refer to the ESI,† for details), as follows:^78–80

ΔG⁰_CS = e[E_OX − E_RED] – E₀₀ + ΔG_S (2)

ΔG⁰_CR = −(ΔG⁰_CS – E₀₀) (4)

E_CSS = e[E_OX − E_RED] + ΔG_S (5)

The data showed that CS may occur for Rho-NIS upon photoexcitation in a polar solvent, such as ACN (Table 2). However, this does not mean that the CS state will be observed experimentally because the lowest state must be a CS state not ³LE state, and the CS process must outcompete the other relaxation processes of the ¹LE state. The ³NIS state energy of Rho-NIS was estimated by 00 transition of the phosphorescence emission spectra and determined to be 1.68 eV.⁵¹ It should be noted this value is much lower than that of the native NI chromophore (2.25 eV).⁸¹ The calculation showed that the energy of the CS state (2.01 eV in HEX and 1.90 eV in TOL) is higher than the ³NIS state (Table 2) in non-polar and low-polar solvents. In polar solvents such as ACN, the energy of the CS state (1.52 eV) was lower than the ³NIS state (1.68 eV). It should be noted that the ³LE state energy level was not significantly affected by a variation in the solvent polarity.

Table 2 Electrochemical redox potentials, Gibbs free energy changes of the charge separation (ΔG⁰_CS) and the energy of the charge-separated state (E_CSS) of the compounds in different solvents^a

	E OX (V)	E RED (V)	ΔG^0CS (eV)/ECSS (eV)
			HEX	TOL	THF	ACN
a Cyclic voltammetry in N2-saturated solvent containing 0.10 M Bu4N[PF6]. Pt electrode as the counter electrode, glassy carbon as the electrode working electrode, ferrocene (Fc/Fc^+) as the internal reference (set as 0 V in the cyclic voltammograms), and Ag/AgNO3 couple as the reference electrode. In ACN. E00 = 1.68 eV. E00 is the T1 state energy of NIS estimated by 00 transition based on phosphorescence emission spectra. THF stands for tetrahydrofuran. b Not observed. c Not applicable.
Rho	+0.51, +0.64	—^b	—^c	—^c	—^c	—^c
NI	—^b	−1.75	—^c	—^c	—^c	—^c
NIS	—^b	−1.09	—^c	—^c	—^c	—^c
NIS-NH2	+0.19	−1.29	—^c	—^c	—^c	—^c
NIS-NHAc	—^b	−1.10	—^c	—^c	—^c	—^c
Rho-NIS	+0.52, +0.87	−1.07	0.33/2.01	0.22/1.90	−0.05/1.63	−0.16/1.52

---

Furthermore, to facilitate the assignment of the CS state in the transient absorption spectra, chemical reduction of the compounds was performed to obtain the UV-vis absorption of their anion (Fig. 4c and d). In the case of Rho-NIS, upon reduction with tetrabutylammonium fluoride (TBAF), the absorption band centered at 453 nm decreased, and a new dominant absorption band centered at 462 nm and weaker bands centered at 668 nm and 794 nm developed, which are attributed to the NIS˙⁻ radical anion species (Fig. 4c). Similar results were observed for the reference compound NIS (Fig. 4d). The UV-vis spectra of NIS˙⁻ for NIS were also computed by the TDDFT method (Fig. S52, ESI†), which supports the experimental results. With the addition of TBAF, NIS-NH₂ showed absorption bands centered at 390 nm and 600 nm (Fig. S19, ESI†), which are attributed to [NIS-NH₂]˙⁻ and different from the NIS˙⁻ of NIS and Rho-NIS. These results are in agreement with the UV-vis absorption spectral studies and further indicate that the acetylation of the amino group and the conformation restriction of Rho-NIS reduced the electron-donating ability of the N atom and the π-conjugation of the lone electron pairs on the N atom with the NI chromophore. The absorption of the NI radical anion (NI˙⁻) without thionation of the carbonyl group was observed at 420 nm, 491 nm and 830 nm, which is different from that of the NIS˙⁻ of NIS and Rho-NIS.^82–84

Femtosecond transient absorption (fs-TA) spectra

To study the excited-state dynamics of the compounds upon photoexcitation, the fs-TA spectra of the dyads were recorded in solvents with different polarity (Fig. 5). The data (Fig. 5a–c) were analyzed with singular value decomposition (SVD) and global analysis using a linear unidirectional decay scheme. The evolution-associated difference spectra (EADS) for each kinetic constant were obtained by global analysis (Fig. 5d–f). In the case of NIS, the fs-TA spectra in TOL are shown in Fig. 5c. Also, the result of a target analysis of the fs-TA data with a sequential model was obtained (Fig. 5f). The ground-state bleaching (GSB) band was observed in the range of 420–500 nm, which is in agreement with the steady-state absorption spectrum (Fig. 2). The S₁ state with an absorption band centered at 505 nm and 740 nm was populated after photoexcitation. Then, the ³NIS state with the absorption at 510 nm was generated by the fast ISC process within 10.6 ps from the ¹NIS state, which was attributed to the spin orbit coupling ISC (SOC-ISC) mechanism. Similar results were observed for NIS in ACN with the ISC process of 1.4 ps (Fig. S20, ESI†). It should be noted that the first species with the lifetime of 277 fs in Fig. 5f is the signal of the TOL solvent, which cannot be neglected in the global fitting. A similar result was observed in Fig. 5d. The ³NIS state did not decay completely within the available time window of the fs-TA spectrometer, which is supported by the ns-TA spectra (see later section). In the case of NIS-NH₂, its ESA bands were different from that of NIS, and the ISC from ¹NIS-NH₂ → ³NIS-NH₂ takes 31 ps (Fig. S21, ESI†). Thus, the ISC kinetics is in good agreement with the theoretically predicted ISC rate constants of ∼10 ps.⁸⁵ Interestingly, the ISC kinetics of NIS is similar to that of the unsubstituted NI (∼10–20 ps).⁸⁶ It should be noted for the 4-amino-substituted NI, the ISC is negligible.^66,87,88 The ISC kinetics of the thionated NI is similar to that observed for the thionated perylene bisimide (PBI).^49,63

Fig. 5 Femtosecond transient absorption (fs-TA) spectra of Rho-NIS in (a) TOL and (b) ACN. The related evolution associated difference spectra (EADS) in (d) TOL and (e) ACN for Rho-NIS. (c) fs-TA spectra and (f) the related EADS of NIS in TOL. EADS were obtained from target analysis with the sequential model. λ_ex = 355 nm. c = 1.0 × 10⁻⁴ M, 25 °C.

The fs-TA spectra of Rho-NIS in different solvents upon excitation are shown in Fig. 5a and b, together with the EADS obtained from global analysis. In TOL, two species were required for the satisfactory fitting of the spectra (independent of the solvent, Fig. 5d). The first species can be assigned as a locally excited singlet state of ¹NIS by comparison with the spectra of NIS reported in Fig. 5f. After 42.1 ps, the second species was formed with absorption bands at 530 nm and 602 nm, which is attributed to the ³NIS state of Rho-NIS and the same in the ns TA spectra (see later section). Thus, the process of Rho-NIS upon photo-excitation in TOL can be summarized as ¹NIS → ³NIS by SOC-ISC, taking 42.1 ps, which is slightly slower than the ISC kinetics of NIS (10.6 ps). No CS state was observed. In ACN, initial EADS showed the appearance of ESA bands peaked at 500 nm and 760 nm, which can be assigned to the ¹NIS state by comparison with the spectrum of NIS (Fig. 5e). After 6.5 ps, the ESA band at 500 nm was red shifted to 525 nm and the ESA band at 760 nm disappeared. This species is attributed to the ³NIS state of Rho-NIS, which is similar to the fs-TA signal of Rho-NIS and NIS observed in TOL on a long-timescale. After 133.6 ps, the transient spectrum evolved significantly, with the appearance of two positive bands peaked at 485 nm and 780 nm, respectively. The comparison with the absorption spectra of NIS˙⁻ generated by chemical reduction (Fig. 4) allowed us to assign these bands to the NIS˙⁻ radical anion, and this species is attributed to the CS state (i.e., Rho˙⁺-NIS˙⁻). The final spectral component lived well beyond the time explored with the fs-TA measurement (3 ns), in accordance with the ns-TA spectra result. Thus, the photophysical process for Rho-NIS in ACN is ¹NIS → ³NIS → CS state. Therefore, electron spin control was realized in formation of the long-lived CS state. This is similar to the previously reported Rho-NI without thionated NI.⁴⁰ The difference is that ISC is achieved with the thionated NI moiety not by the SOCT-ISC. It should be noted that the CS state energy level is dependent on the solvent polarity, whereas the ³LE state is much less dependent on it.⁸⁹

The fs-TA spectra of Rho-NIS in the presence of acid were also studied (Fig. S23 and S24, ESI†). In TOL, based on the EADS analysis, the first species showed the ESA band peaked at 440 nm and GSB band peaked at 560 nm (Fig. S23, ESI†). Moreover, the strong negative bands in the range of 570–590 nm are attributed to both the GSB and stimulated emission (SE) bands due to the fact that the fluorescence of the open form of rhodamine has maximum at 575 nm. An SE band in the range of 600–680 nm was also observed. Therefore, the first species is attributed to the singlet state of the open form of rhodamine, i.e., ¹Rho-o. After 23.7 ps, ESA bands centred at 530 nm and beyond 590 nm were observed. It should be noted that this positive band may cover the range of 480–700 nm, but this band is distorted by the GSB band. It shares the same characteristics with Rho-³NIS observed in TOL without the addition of acid. Consequently, this species is assigned as ³NIS and it did not decay to the baseline in the time detection window of fs-TA (3 ns), which is same as the that observed in the ns-TA spectra initially. In MeOH, similar results were observed (Fig. S24, ESI†). The lifetime of ¹Rho-o was 46.5 ps. The mechanism by which the energy transferred from the ¹Rho-o to ³NIS state is unclear. It is known that the ISC process of the Rho-o unit is not efficient,⁷⁰ and therefore it is unlikely that the mechanism involves the formation of the T₁ state of the Rho-o unit.⁹⁰ This is probably a singlet–triplet energy transfer (STET) or ³NIS state generated by charge recombination (CR) from the CS state (i.e., Rho-o˙⁺-NIS˙⁻); however, the CS state could not be observed due to the short lifetime. Based on the electrochemical results, charge transfer from Rho-o to NIS with S₁ state of Rho-o (E_S = 2.19 eV) as a precursor is a thermodynamics allowed process (ΔG⁰_CS = −0.15 eV, E_CS = 2.04 eV in TOL and ΔG⁰_CS = −0.53 eV, E_CS = 1.66 eV in MeOH). Also, the energy of the ³NIS state (E_T = 1.68 eV) is close or lower than that of the CS state.

Nanosecond transient absorption (ns-TA) spectra

To study the long-lived transient species formed in Rho-NIS upon photoexcitation, the ns-TA spectra of the compounds were studied (Fig. 6). In the case of Rho-NIS upon photoexcitation in TOL, a GSB band centered at 450 nm was observed, as well as a positive absorption band in the range beyond 500 nm (Fig. 6a). It should be noted that this positive band may also cover the range of 350–550 nm, but it is probably distorted by the GSB band. The reference compound NIS was also studied, and a similar spectral feature was observed (Fig. 6c); thus, the positive absorption band of Rho-NIS is attributed to the ³NIS state not a CS state. Interestingly, the lifetime of the ³NIS state of Rho-NIS was determined to be 0.38 μs (Fig. 6e), which is slightly longer than that of the reference NIS of 0.20 μs in TOL (Fig. 6g). The slightly longer ³LE state lifetime of Rho-NIS compared to NIS is attributed to the acetylation of the amino group in Rho-NIS, which is supported by the slightly longer triplet state lifetime of NIS-NHAc (0.57 μs) than NIS (Fig. S38, ESI†). It should be noted that these ³NIS state lifetimes are much shorter than the triplet state lifetime of the native NI moiety (unthionated, 67 μs).^66,91 In ACN, Rho-NIS showed ESA bands at 480 nm and 780 nm (Fig. 6b), which are different from that of the spectra observed in TOL, but the results are similar to the absorption spectra of NIS˙⁻ generated by chemical reduction (Fig. 4), demonstrating the formation of a CS state in this case. The lifetime of the CS state was determined to be 0.21 μs (Fig. 6f). It should be noted that the CS state we observed with ns-TA spectra is most likely a ³CS state because the fluorescence of the Rho-NIS state was quenched, and no emission was observed from ¹CS, given that its lifetime was too short and could not be determined using the ns-TA spectrometer. It should be noted that the ¹CS state and ³CS state may share similar optical absorption character.^92–95 The ³CS state of Rho-NIS is generated from ³NIS by electron spin control based on the fs-TA spectral studies, i.e., the photophysical process for Rho-NIS in ACN is ¹NIS → ³NIS → ³CS state. The quantum yield of the charge separation (Φ_CS) was determined using a relative method with anthracene as a standard (for details, refer to the ESI†). In ACN, the Φ_CS of Rho-NIS was determined to be 97%, which is higher than that of the previously reported Rho-NI (49% in HEX).⁴⁰ We propose that the thionation of the carbonyl group in Rho-NIS induces efficient ISC, and further generates the CS state efficiently via the electron spin control effect. Alternatively, the triplet state lifetime of NIS in ACN was shortened to 0.06 μs (Fig. S35, ESI†). Interestingly, NIS-NH₂ showed different ESA bands and longer triplet state lifetime (ca. 6 μs, Fig. 6d and h) compared with NIS and Rho-NIS, and it was less dependent on the solvent polarity (Fig. S37, ESI†).

Fig. 6 Nanosecond transient absorption spectra of Rho-NIS in deaerated (a) TOL and (b) ACN and (c) NIS and (d) NIS-NH₂ in deaerated TOL. Decay traces of Rho-NIS in deaerated (e) TOL at 530 nm and (f) ACN at 480 nm and (g) NIS at 520 nm and (h) NIS-NH₂ at 630 nm in deaerated TOL. Excited with nanosecond pulsed laser. λ_ex = 355 nm for Rho-NIS and NIS. λ_ex = 530 nm for NIS-NH₂. c[Rho-NIS] = 4.0 × 10⁻⁵ M. c[NIS, NIS-NH₂] = 2.0 × 10⁻⁵ M. 25 °C.

Concerning the reported Rho-NI dyad,⁴⁰ without thionation of the carbonyl group of the NI unit, the NI moiety showed a higher triplet state energy (E_T = 2.21 eV) and weaker electron-accepting ability (E_red = −1.75 V) compared with the NIS moiety (E_T = 1.68 eV, E_red = −1.09 V). This led to the higher energy of the CS state (2.12 eV) and longer lifetime of the CS state (0.94 μs in HEX) by electron spin control strategy for the previously reported Rho-NI dyad.⁴⁰ The photophysical processes of Rho-NI in HEX upon photoexcitation are summarized as ¹NI → ¹CS → ³NI → ³CS. The lifetime of the CS state of Rho-NI was 0.62 μs in TOL and no CS state was observed in ACN. In contrast, for Rho-NIS, the relative energy order of the CS and ³LE states is dependent on the solvent polarity. In the low-polarity TOL solvent, the ³NIS state (τ_T = 0.38 μs) as the lowest triplet state was observed, in the polar solvent ACN and the CS state with the lifetime of 0.21 μs was generated by electron spin control, i.e., the photophysical process of Rho-NIS in ACN is ¹NIS → ³NIS → ³CS.

Given that the transient absorption spectra show that the lowest triplet state of NIS has ³n-π* character,⁵¹ the ns-TA spectra of NIS were also measured in the protic polar solvent isopropanol (IPA) (Fig. 7). In the short delay time range, the ³NIS state showing an ESA band centered at 510 nm was observed in deaerated IPA (Fig. 7a). At a longer delay time, an ESA band peaking at 430 nm appeared and increased gradually as the ESA band of the ³NIS state decreased (Fig. 7a and b). This is different from the feature of NIS˙⁻ absorption (485 nm and 780 nm) and can be assigned as NISH˙ upon hydrogen abstraction by ³NIS from IPA. The evolution of the kinetic trace at 430 nm was monitored, which showed a raised component with the time of 87 ns (Fig. 7e), which is attributed to the accumulation of the NISH˙. Also, the lifetime of NISH˙ was determined to be 284.3 μs (Fig. 7f), which is much longer than that of the ³NIS state (0.2 μs, Fig. 6). To differentiate the species in the transient absorption spectra, EADS by global fitting to the ns-TA spectra with the sequential model were obtained (Fig. 7c). Two species were identified, where the first species is the ³NIS state with the lifetime of 124 ns, which is close to that monitored at 520 nm (135 ns, Fig. 7d). The second species is the long-lived NISH˙ (284.3 μs). The result of hydrogen abstraction between the photoexcited state of NIS in IPA indicates the n–π* character of the lowest-lying triplet state of NIS.⁹⁶ Under an air atmosphere (aerated IPA), only the ³NIS state was observed and its lifetime was much shorter (0.07 μs, Fig. S43, ESI†). This short excited-state lifetime makes the hydrogen abstraction process unlikely.^97,98 Similar results were observed for Rho-NIS and NIS-NHAc in deaerated IPA upon photoexcitation (Fig. S42 and S45, respectively, ESI†), indicating the n–π* character of the triplet state. Long-lived NISH˙ (25.9 μs for Rho-NIS and 349.2 μs for NIS-NHAc) was generated together with the decay of ³NIS. However, NIS-NH₂ showed the same ns-TA spectra in IPA and other aprotic solvents, proving that it has ³π–π* character or weak ³n-π* character (Fig. S44, ESI†). This is supported by its longer triplet state lifetime (Fig. 6h, 6.1 μs). NI was also observed to undergo hydrogen abstraction by the ³NI state in IPA (Fig. S46, ESI†), resulting in the formation of NIH˙ (329.7 μs), which absorbs at 410 nm, in agreement with the reported ns-TA spectra of NIH˙ (Fig. S46, ESI†).^98,99 The slow rate of the formation of NIH˙ (19.1 μs) is probable due to the weak n–π* character in ³NI, leading to the non-efficient hydrogen abstraction from IPA. Photo-induced hydrogen abstraction, and thus formation of a radical, is potentially useful for the development of new photoinitiators for photopolymerization.^100,101

Fig. 7 Nanosecond transient absorption spectra of NIS in deaerated isopropanol in (a) short and (b) long delay time range to study the hydrogen abstraction. (c) Related evolution-associated difference spectrum (EADS) obtained from the target analysis with the sequential model for NIS. (d) Decay traces of NIS at 520 nm and at 430 nm in (e) short and (f) long timescale, excited with nanosecond pulsed laser. λ_ex = 355 nm. c = 3.0 × 10⁻⁵ M, 25 °C.

In the case of Rho-NIS, with the addition of acid, the transformation of the closed form of the rhodamine moiety to the open form occurred, and its ns-TA spectra were studied upon photoexcitation. In TOL, in the presence of TFA, the ³NIS state was observed initially (Fig. S33, ESI†). The ESA band in the range of 530–600 nm overlaps with the GSB band originating from the absorption of the rhodamine unit in the open form. As the delay time increased, the ESA band centered at 530 nm disappeared and the new ESA band in the range of 400–500 nm increased. To identify these different species, global fitting was performed and EADS was obtained (Fig. S33b, ESI†). The first species is assigned as the ³NIS state. In the following 460 ns, the second species was formed, which is tentatively attributed to the triplet state of the rhodamine unit in the open form, i.e., ³Rho-o, similar to the reported ns-TA spectra of ³Rho-o.⁷⁴ Therefore, triplet–triplet energy transfer (TTET) occurred from ³NIS to ³Rho-o within 460 ns. The decay trace monitored at 470 nm has a raised component with a time of 0.60 μs, which is close to the time for TTET determined from EADS (0.46 μs). The decay trace at 560 nm has a biexponential feature and the lifetime is 0.62 μs (32%)/9.4 μs (68%). The component with the lifetime of 0.62 μs is attributed to the decay of ³NIS, i.e., TTET process. The component with the lifetime of 9.4 μs is attributed to the decay of ³Rho-o. Notably, the ³Rho-o state energy is reported to be 1.73 eV (estimated by TDDFT calculations),^77,90 which is slightly higher than the energy of the ³NIS state (1.68 eV). This is probably because the TDDFT calculation overestimates the energy of the ³Rho-o state. In MeOH, in the presence of TFA, a similar TTET process from ³NIS to ³Rho-o was observed within 0.11 μs and the lifetime of ³Rho-o was 13.6 μs (Fig. S30, ESI†). It should be noted that in the absence of TFA, the CS state of Rho-NIS with the lifetime of 0.19 μs was observed in MeOH (Fig. S27, ESI†). These results demonstrate that the Rho-NIS dyad with the Rho unit in the open form shows different photophysical properties compared to Rho-NIS with the Rho unit in the close form. This external stimuli-responsive photophysical property may be used to develop functional materials.^102,103

Time-resolved electron paramagnetic resonance (TREPR) spectroscopy

To study the transient paramagnetic species of the compounds formed upon photoexcitation, a TREPR spectral study was carried out, which is useful to discriminate between the ³CS and ³LE states and the ³CS state and the spin-correlated radical pair.^{12,35,104–107} In the case of NIS and Rho-NIS, no TREPR spectra were observed in frozen solution at 80 K, even at 10 K, with an X- and Q-band spectrometer. This is probably the result of the short triplet state lifetime (0.20 μs for NIS and 0.38 μs for Rho-NIS) and the fast spin relaxation. The triplet-state TREPR spectra at the Q-band of NIS-NH₂ was observed with the electron spin polarization (ESP) phase pattern of (e, e, e, a, a, a) (Fig. 8a and b). The spectrum at Q-band was well described by the EasySpin package (Fig. 8b). The triplet spectra exhibit a forbidden transition in the X-band (Fig. S49, ESI†).

Fig. 8 TREPR spectra of NIS-NH₂ measured with Q-band EPR spectrometer excited with 532 nm laser with energy 10 mJ per pulse (a) at different delay time and (b) at 0.3 μs in TOL/2-MeTHF (1 : 3, v/v) at 80 K. The simulation (red curve in b) was performed with EasySpin package. c = 1.0 × 10⁻⁴ M.

An isotropic g-tensor of 2.0070 was used to simulate the TREPR spectrum of NIS-NH₂ (Fig. 8b). The results of the simulations are presented in Table 3. The zero-field splitting (ZFS) parameters |D| of 6165 MHz and E of −1233 MHz were obtained by simulation of the TREPR spectra of NIS-NH₂, which are both larger than that of NI with |D| of 2475 MHz and |E| of 135 MHz, and its analogue without thionation as that in NIS-NH₂, i.e., NI-NH-Br, gave |D| of 2314 MHz and |E| of 193 MHz.^66,88 The large ZFS parameter |D| of NIS-NH₂ indicates that its triplet state has mixed n–π* and π–π* character.¹⁰⁸ This led to the shorter triplet state lifetime of NIS-NH₂ (6.1 μs) compared to compound 4 (17.5 μs, obtained by TTET, Fig. S47, ESI†). The relative population rates of the three sublevels of the triplet state of NIS-NH₂ at zero magnetic field are P_x : P_y : P_z = 0.5 : 0.5 : 0.

Table 3 Zero-field splitting parameters (|D| and E) and relative populations P_x,y,z of the spin states at zero magnetic field of the compounds^a

	|D| (MHz)	|E| (MHz)	P x	P y	P z	g-Factor
a Obtained from simulations of the triplet-state TREPR spectra of the indicated molecules at 80 K. Pi is the relative population of the i-th ZFS state. b Literature values, see ref. 66. c Not reported. d Literature values, see ref. 88.
NIS-NH2	6165	1233	0.50	0.50	0	2.0070
NI^b	2745	135	0.16	0.84	0	—^c
NI-NH-Br^d	2314	193	0	0.38	0.62	2.0050

---

It is known that 9,10-anthraquinone (AQ) has a T₁ state with n–π* character, ZFS D of −9000 MHz, and E of 160 MHz.¹⁰⁸ In the case of 9,10-diazaphenanthrene (DAP), the T₁ state was assigned as the n–π* state in the single crystal of biphenyl, and the ZFS D and E parameters were determined to be −4011 MHz and 2667 MHz, respectively.¹⁰⁹ In polar solvents, such as 2,2,2-trifluoroethanol, the T₁ state of DAP has π–π* character, and the D and E values are −2788 MHz and 1199 MHz, respectively. Furthermore, the T₁ state of pyridazine has n–π* character and its ZFS D and E parameters were determined to be −4137 MHz and 3478 MHz, respectively.^110,111 These results demonstrate that the compounds with a T₁ state having n–π* character may give a large ZFS parameter D than the corresponding ³n–π* state.

Theoretical computations

To rationalize the photophysical property of the compounds, theoretical computations were performed. The ground state geometry of the compounds was optimized by DFT calculations (Fig. S50, ESI†). In the case of Rho-NIS, the dihedral angle between the π-conjugation planes of the rhodamine xanthene moiety and NIS is 61.6°, which is slightly different from the molecular structure determined by the single-crystal X-ray diffraction experiments (53.0°), and the difference probably results from the packing effect in the single crystal. The centroid-to-centroid distance between xanthene and the NIS moieties is 5.6 Å, which is close to the previously reported dyads without thionation of the NI unit.⁴⁰ The frontier molecular orbitals (MOs) of the dyads were studied (Fig. 9). The highest occupied molecular orbital (HOMO) of Rho-NIS is confined on the Rho moiety and the lowest unoccupied molecular orbital (LUMO) is confined on the NIS moiety. There is negligible delocalization on these two moieties for both MOs, which is in agreement with the molecular structure design rationale to obtain the CS state upon photoexcitation.

Fig. 9 Selected frontier molecular orbitals contributing to the T₁ states of the compounds (isovalue = 0.04). The orbital transition probability and energy of T₁ states are also shown. Calculation was performed by DFT at the B3LYP/6-31G(d) level with Gaussian 09.

The frontier molecular orbitals involved in the T₁ states of the compounds were examined (Fig. 9). In the gas phase, the orbitals involved in the transition of the T₁ state Rho-NIS and NIS and NIS-NHAc mainly have n–π* character. This is supported by the short triplet excited state of these compounds, as well as the hydrogen abstraction properties observed in their ns-TA spectra. The MOs of the T₁ state of NIS have n–π* character and are similar to the T₁ state of 9,10-diazaphenanthrene (DAP), which has a pure n–π* T₁ state.¹⁰⁹ However, for NIS-NH₂, the orbital transition of the T₁ state is a mixture of n–π* and π–π* in nearly equal proportion (49%). This is in agreement with the longer triplet excited state lifetime of 6.1 μs (Fig. 6h).

The T₁ states of the compounds were also analysed by the natural transition orbital (NTO, Fig. S54, ESI†). For all these compounds, the contribution from the NTO pair with the largest eigenvalue is over 95%. In the gas phase, the hole involved in the T₁ state of NIS spreads mainly over the two sulphur atoms and the contribution of the sulphur atoms to the hole is over 70%, whereas the particle is spread over the whole molecule (Fig. S54 and S55, ESI†), which indicates that the T₁ state of NIS mainly has n–π* character. Similar results were observed for Rho-NIS and NIS-NHAc, which have similar T₁ states with n–π* character. However, in the case of NIS-NH₂, the hole involved in the T₁ state spread over the two sulphur atoms as well as the aromatic framework, and the contribution of the two sulphur atoms to the hole is 27.0% and 15.3%, respectively, and the sum of their contribution is less than 50% (Fig. S54 and S55, ESI†). The particle is spread over the whole molecular structure. Therefore, in the case of NIS-NH₂, the T₁ state is an admixture of n–π* and π–π* character based on the NTO analysis, which is in accordance with the analysis based on frontier molecular orbitals involved in the T₁ states and the ns-TA spectral results.

The electron spin density surface of the T₁ state of Rho-NIS in the gas phase, TOL and ACN was studied (Fig. 10) to corroborate the ns-TA spectral observation. In the gas phase and TOL, the electron spin density was confined on the thionated NI unit, and the sulfur atoms contributed significantly, and thus we conclude that it is a triplet state with significant n–π* character, not a pure ³CS state (with the closed ring structure of rhodamine as the electron donor). This is in agreement with the ns-TA spectral results. However, in ACN, the electron spin density of the T₁ state was delocalized on both the thionated NI and rhodamine units. Notably, there was no ³NIS and ³Rho excited-state equilibrium due to the large energy gap between the triplet state of NIS (1.68 eV) and rhodamine moieties (3.36 eV).⁷⁷ This result suggests that a ³CS state may be formed in ACN upon photoexcitation, which is in agreement with the ns-TA spectral studies (Fig. 6).

Fig. 10 Isosurfaces of spin density at the optimized triplet state geometries of Rho-NIS in the gas phase, TOL and ACN (isovalue = 0.0004), respectively. Calculation was performed by DFT method at the UB3LYP/6-31G(d) level with Gaussian 09.

Based on the above-mentioned results, the energy diagram of Rho-NIS was constructed (Scheme 2). In the absence of acid, upon the photoexcitation of the NIS moiety in both TOL and ACN, the singlet excited state of Rho-NIS (i.e., ¹NIS state) is populated. Then, in TOL, fast ISC (42 ps) produces the triplet state localized on the NIS moiety (³NIS) with the lifetime of 0.38 μs. The ³NIS state has the n–π* character, leading to hydrogen abstraction from the protic solvent to generate the long-lived NISH˙ (25.9 μs). In ACN, ³NIS was produced by fast ISC (7 ps) from the ¹NIS state, and then the charge separation (CS) from the ³NIS state takes 134 ps to generate the ³CS state. The ³CS state lifetime is 0.21 μs. The photophysical process of Rho-NIS in ACN upon photoexcitation is ¹NIS → ³NIS → ³CS. In the presence of acid (Scheme 2b), upon photoexcitation, the singlet of rhodamine in the open form (i.e., ¹Rho-o state) was populated, and then the ³NIS state was formed within 24 ps by singlet–triplet energy transfer or rapid charge recombination from the CS state (i.e., Rho-o˙⁺-NIS˙⁻, generated by charge transfer from ¹Rho-o state). Subsequently, TTET from the ³NIS state takes 0.6 μs, leading to the formation of the ³Rho-o state with the lifetime of 9.4 μs.

Scheme 2 Simplified Jablonski diagram illustrating the photophysical processes involved in Rho-NIS in the (a) absence and (b) presence of acid. The energy of the excited singlet states is estimated by UV-vis absorption spectra and fluorescence spectra. The energy of the triplet state is estimated by the 00 transition based on the phosphorescence emission spectra and the energy of CT states is obtained by electrochemical calculation. The number of the superscript denotes the spin multiplicity. IPA stands for isopropanol.

Conclusions

In conclusion, using the rhodamine (Rho) moiety as an electron donor, having the ability to undergo a reversible transformation of spirolactam ↔ open form amide structure, and thionated naphthalimide (NIS) as an electron acceptor, we prepared a compact, orthogonal electron donor–acceptor dyad Rho-NIS. Our aim was to access the long-lived triplet charge separation (³CS) state based on the electron spin control approach, and the purpose of thionation was to enhance the ISC to produce the ³NIS state. Rho (lactam form) was attached to NIS at the 4-position via a single C–N bond. The transient absorption (TA) spectra showed that intersystem crossing (ISC) of ¹NIS → ³NIS occurs rapidly in different solvents (7–42 ps) upon photoexcitation, which was assigned to the spin orbit coupling ISC (SOC-ISC) mechanism due to the ³n-π* character of the T₁ state of the NIS moiety. In TOL, a triplet state localized on the NIS moiety (³NIS) was observed for Rho-NIS with the lifetime of 0.38 μs. Also, the ³NIS state has the ability to undergo hydrogen abstraction in protic solvents due to its n–π* character and generates the long-lived NISH˙ (25.9 μs). Alternatively, the long-lived CS state was observed in ACN, with the lifetime of 0.21 μs and the quantum yield of the CS state (Φ_CS) was high (97%). The relative energy of the ³CS and ³LE states changed in solvents with different polarity. The long-lived CS state of the dyad was based on the electron spin control effect, that is, the generation of the final ³CS state with a triplet precursor (³NIS state). For the previously reported analogue Rho-naphthalimide (NI) dyad, without thionation of the carbonyl group of the NI unit, the relative energy of the CS and ³LE states was not inverted by a variation in the solvent polarity. The long-lived CS state generated by the electron spin control strategy was observed in both n-hexane (HEX, 0.94 μs; Φ_CS, 49%) and TOL (0.62 μs), which are longer than the CS state lifetime of Rho-NIS in ACN (0.21 μs). However, no CS state was observed for Rho-NI in ACN. The current thionation strategy makes the formation of the long-lived ³CS state in ACN and improvement of the Φ_CS possible. Furthermore, the effect of the open/closed form of Rho on the photophysical process was also studied. In the presence of acid, the Rho unit in Rho-NIS transformed to the open form (Rho-o), which showed a broadband absorption in the visible spectral region (500–650 nm). The ³NIS state was formed from the ¹Rho-o state within 24–47 ps. Then, ³Rho-o state (9.4–13.9 μs) was formed by triplet–triplet energy transfer (TTET) from the ³NIS state within 0.11–0.60 μs. In the time-resolved electron paramagnetic resonance (TREPR) spectra, the triplet state signal of NIS-NH₂ was observed and the zero-field splitting (ZFS) parameters |D| and E of NIS-NH₂ were determined to be 6165 MHz and −1233 MHz, respectively. Compared to the NI compound with a long triplet state lifetime (67.2 μs), |D| of 2475 MHz and |E| of 135 MHz, the large |D| value of NIS-NH₂ demonstrates that its triplet state has the mixed character of n–π* and π–π*, leading to its short triplet state lifetime (6.1 μs). Thus, these results indicate that thionation may significantly shorten the triplet state lifetime and this strategy should be carefully used to design heavy atom-free triplet photosensitizers to realize the desirable long triplet state lifetime. These results are also useful for understanding the charge transfer and access to the long-lived CS state in compact electron donor/acceptor dyads.

Author contributions

X. X., T. M. and A. A. S. contributed equally to this work.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

J. Z. thanks the NSFC (U2001222), the Research and Innovation Team Project of Dalian University of Technology (DUT2022TB10), the Fundamental Research Funds for the Central Universities (DUT22LAB610), and the State Key Laboratory of Fine Chemicals for financial support. A. A. S. and V. K. V. acknowledge financial support from the government assignment for FRC Kazan Scientific Center of RAS.

Notes and references

1. D. M. Guldi, Chem. Commun., 2000, 321–327.
2. J. W. Verhoeven, H. J. van Ramesdonk, M. M. Groeneveld, A. C. Benniston and A. Harriman, ChemPhysChem, 2005, 6, 2251–2260.
3. J. W. Verhoeven, J. Photochem. Photobiol., C, 2006, 7, 40–60.
4. S. Fukuzumi, Pure Appl. Chem., 2007, 79, 981–991.
5. D. Veldman, S. M. A. Chopin, S. C. J. Meskers and R. A. J. Janssen, J. Phys. Chem. A, 2008, 112, 8617–8632.
6. O. Kei and F. Shunichi, Bull. Chem. Soc. Jpn., 2009, 82, 303–315.
7. S. Suzuki, M. Kozaki, K. Nozaki and K. Okada, J. Photochem. Photobiol., C, 2011, 12, 269–292.
8. E. Vauthey, ChemPhysChem, 2012, 13, 2001–2011.
9. M. Delor, T. Keane, P. A. Scattergood, I. V. Sazanovich, G. M. Greetham, M. Towrie, A. J. H. M. Meijer and J. A. Weinstein, Nat. Chem., 2015, 7, 689–695.
10. G. N. Lim, C. O. Obondi and F. D'Souza, Angew. Chem., Int. Ed., 2016, 55, 11517–11521.
11. Y. Hou, X. Zhang, K. Chen, D. Liu, Z. Wang, Q. Liu, J. Zhao and A. Barbon, J. Mater. Chem. C, 2019, 7, 12048–12074.
12. N. Zarrabi, B. J. Bayard, S. Seetharaman, N. Holzer, P. Karr, S. Ciuti, A. Barbon, M. Di Valentin, A. van der Est, F. D’Souza and P. K. Poddutoori, Phys. Chem. Chem. Phys., 2021, 23, 960–970.
13. M. Lv, X. Lu, Y. Jiang, M. E. Sandoval-Salinas, D. Casanova, H. Sun, Z. Sun, J. Xu, Y. Yang and J. Chen, Angew. Chem., Int. Ed., 2022, 61, e202113190.
14. X. Niu, K. Tajima, J. Kong, M. Tao, N. Fukui, Z. Kuang, H. Shinokubo and A. Xia, Phys. Chem. Chem. Phys., 2022, 24, 14007–14015.
15. H. Imahori, Bull. Chem. Soc. Jpn., 2023, 96, 339–352.
16. S. M. Sartor, Y. M. Lattke, B. G. McCarthy, G. M. Miyake and N. H. Damrauer, J. Phys. Chem. A, 2019, 123, 4727–4736.
17. S. Suzuki, Y. Matsumoto, M. Tsubamoto, R. Sugimura, M. Kozaki, K. Kimoto, M. Iwamura, K. Nozaki, N. Senju, C. Uragami, H. Hashimoto, Y. Muramatsu, A. Konno and K. Okada, Phys. Chem. Chem. Phys., 2013, 15, 8088–8094.
18. A. N. Bartynski, M. Gruber, S. Das, S. Rangan, S. Mollinger, C. Trinh, S. E. Bradforth, K. Vandewal, A. Salleo, R. A. Bartynski, W. Bruetting and M. E. Thompson, J. Am. Chem. Soc., 2015, 137, 5397–5405.
19. E. Salvadori, N. Luke, J. Shaikh, A. Leventis, H. Bronstein, C. W. M. Kay and T. M. Clarke, J. Mater. Chem. A, 2017, 5, 24335–24343.
20. D. Lee, J. Lee, D. H. Sin, S. G. Han, H. Lee, W. Choi, H. Kim, J. Noh, J. Mun, W. Sung, S. W. Kim, B. G. Jeong, S. H. Kim, J. Rho, M. S. Jeong and K. Cho, J. Phys. Chem. C, 2022, 126, 3171–3179.
21. S. Callaghan, M. A. Filatov, H. Savoie, R. W. Boyle and M. O. Senge, Photochem. Photobiol. Sci., 2019, 18, 495–504.
22. Y. Im, S. Y. Byun, J. H. Kim, D. R. Lee, C. S. Oh, K. S. Yook and J. Y. Lee, Adv. Funct. Mater., 2017, 27, 1603007.
23. Z. Yang, Z. Mao, Z. Xie, Y. Zhang, S. Liu, J. Zhao, J. Xu, Z. Chi and M. P. Aldred, Chem. Soc. Rev., 2017, 46, 915–1016.
24. F. Ni, N. Li, L. Zhan and C. Yang, Adv. Opt. Mater., 2020, 8, 1902187.
25. Y. Xu, P. Xu, D. Hu and Y. Ma, Chem. Soc. Rev., 2021, 50, 1030–1069.
26. L. Chen, W.-C. Chen, Z. Yang, J.-H. Tan, S. Ji, H.-L. Zhang, Y. Huo and C.-S. Lee, J. Mater. Chem. C, 2021, 9, 17233–17264.
27. H. Kotani, K. Ohkubo and S. Fukuzumi, Faraday Discuss., 2012, 155, 89–102.
28. V. M. Blas-Ferrando, J. Ortiz, K. Ohkubo, S. Fukuzumi, F. Fernández-Lázaro and Á. Sastre-Santos, Chem. Sci., 2014, 5, 4785–4793.
29. R. A. Marcus, Angew. Chem., Int. Ed. Engl., 1993, 32, 1111–1121.
30. D. I. Schuster, P. Cheng, P. D. Jarowski, D. M. Guldi, C. Luo, L. Echegoyen, S. Pyo, A. R. Holzwarth, S. E. Braslavsky, R. M. Williams and G. Klihm, J. Am. Chem. Soc., 2004, 126, 7257–7270.
31. M. Kuss-Petermann and O. S. Wenger, Angew. Chem., Int. Ed., 2016, 55, 815–819.
32. B. Geiß and C. Lambert, Chem. Commun., 2009, 1670–1672.
33. J. Hankache and O. S. Wenger, Chem. Commun., 2011, 47, 10145–10147.
34. S. Fukuzumi, H. Kotani, K. Ohkubo, S. Ogo, N. V. Tkachenko and H. Lemmetyinen, J. Am. Chem. Soc., 2004, 126, 1600–1601.
35. S. Suzuki, R. Sugimura, M. Kozaki, K. Keyaki, K. Nozaki, N. Ikeda, K. Akiyama and K. Okada, J. Am. Chem. Soc., 2009, 131, 10374–10375.
36. M. Murakami, K. Ohkubo, T. Nanjo, K. Souma, N. Suzuki and S. Fukuzumi, ChemPhysChem, 2010, 11, 2594–2605.
37. Z. E. X. Dance, Q. Mi, D. W. McCamant, M. J. Ahrens, M. A. Ratner and M. R. Wasielewski, J. Phys. Chem. B, 2006, 110, 25163–25173.
38. Y. Kobori, M. Fuki and H. Murai, J. Phys. Chem. B, 2010, 114, 14621–14630.
39. R. Carmieli, A. L. Smeigh, S. M. Mickley Conron, A. K. Thazhathveetil, M. Fuki, Y. Kobori, F. D. Lewis and M. R. Wasielewski, J. Am. Chem. Soc., 2012, 134, 11251–11260.
40. D. Liu, A. M. El-Zohry, M. Taddei, C. Matt, L. Bussotti, Z. Wang, J. Zhao, O. F. Mohammed, M. Di Donato and S. Weber, Angew. Chem., Int. Ed., 2020, 59, 11591–11599.
41. S.-H. Lee, C. T.-L. Chan, K. M.-C. Wong, W. H. Lam, W.-M. Kwok and V. W.-W. Yam, J. Am. Chem. Soc., 2014, 136, 10041–10052.
42. J. E. McGarrah, Y.-J. Kim, M. Hissler and R. Eisenberg, Inorg. Chem., 2001, 40, 4510–4511.
43. G. Ajayakumar and K. R. Gopidas, Photochem. Photobiol. Sci., 2008, 7, 826–833.
44. D. R. Subedi, H. B. Gobeze, Y. E. Kandrashkin, P. K. Poddutoori, A. van der Est and F. D'Souza, Chem. Commun., 2020, 56, 6058–6061.
45. S. I. van Dijk, C. P. Groen, F. Hartl, A. M. Brouwer and J. W. Verhoeven, J. Am. Chem. Soc., 1996, 118, 8425–8432.
46. C. Trinh, K. Kirlikovali, S. Das, M. E. Ener, H. B. Gray, P. Djurovich, S. E. Bradforth and M. E. Thompson, J. Phys. Chem. C, 2014, 118, 21834–21845.
47. M. A. Filatov, Org. Biomol. Chem., 2020, 18, 10–27.
48. D. J. Gibbons, A. Farawar, P. Mazzella, S. Leroy-Lhez and R. M. Williams, Photochem. Photobiol. Sci., 2020, 19, 136–158.
49. A. J. Tilley, R. D. Pensack, T. S. Lee, B. Djukic, G. D. Scholes and D. S. Seferos, J. Phys. Chem. C, 2014, 118, 9996–10004.
50. M. Hussain, J. Zhao, W. Yang, F. Zhong, A. Karatay, H. G. Yaglioglu, E. A. Yildiz and M. Hayvali, J. Lumin., 2017, 192, 211–217.
51. V.-N. Nguyen, S. Qi, S. Kim, N. Kwon, G. Kim, Y. Yim, S. Park and J. Yoon, J. Am. Chem. Soc., 2019, 141, 16243–16248.
52. J. R. Palmer, K. A. Wells, J. E. Yarnell, J. M. Favale and F. N. Castellano, J. Phys. Chem. Lett., 2020, 11, 5092–5099.
53. J. Tang, L. Wang, A. Loredo, C. Cole and H. Xiao, Chem. Sci., 2020, 11, 6701–6708.
54. T. C. Pham, S. Heo, V.-N. Nguyen, M. W. Lee, J. Yoon and S. Lee, ACS Appl. Mater. Interfaces, 2021, 13, 13949–13957.
55. A. I. Wright, B. M. Kariuki and Y.-L. Wu, Eur. J. Org. Chem., 2021, 4647–4652.
56. N. Pearce, E. S. Davies, R. Horvath, C. R. Pfeiffer, X.-Z. Sun, W. Lewis, J. McMaster, M. W. George and N. R. Champness, Phys. Chem. Chem. Phys., 2018, 20, 752–764.
57. O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard and H. Puschmann, J. Appl. Crystallogr., 2009, 42, 339–341.
58. I. H. M. van Stokkum, D. S. Larsen and R. van Grondelle, Biochim. Biophys. Acta, Bioenerg., 2004, 1657, 82–104.
59. J. J. Snellenburg, S. Laptenok, R. Seger, K. M. Mullen and I. H. M. van Stokkum, J. Stat. Software, 2012, 49, 1–22.
60. S. Stoll and A. Schweiger, J. Magn. Reson., 2006, 178, 42–55.
61. T. Lu and F. Chen, J. Comput. Chem., 2012, 33, 580–592.
62. G. W. T. M. J. Frisch, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Mennucci and G. A. Petersson, et al., Gaussian 09 (Revision A.01), Gaussian, Inc., Wallingford CT, 2009.
63. Y.-L. Lee, Y.-T. Chou, B.-K. Su, C.-C. Wu, C.-H. Wang, K.-H. Chang, J.-A. A. Ho and P.-T. Chou, J. Am. Chem. Soc., 2022, 144, 17249–17260.
64. Q. Qi, L. Huang, R. Yang, J. Li, Q. Qiao, B. Xu, W. Tian, X. Liu and Z. Xu, Chem. Commun., 2019, 55, 1446–1449.
65. H. Guo, M. L. Muro-Small, S. Ji, J. Zhao and F. N. Castellano, Inorg. Chem., 2010, 49, 6802–6804.
66. X. Zhang, X. Liu, M. Taddei, L. Bussotti, I. Kurganskii, M. Li, X. Jiang, L. Xing, S. Ji, Y. Huo, J. Zhao, M. Di Donato, Y. Wan, Z. Zhao and M. V. Fedin, Chem. – Eur. J., 2022, 28, e202200510.
67. L. Cao, X. Liu, X. Zhang, J. Zhao, F. Yu and Y. Wan, Beilstein J. Org. Chem., 2023, 19, 1028–1046.
68. S. Sasaki, K. Hattori, K. Igawa and G.-I. Konishi, J. Phys. Chem. A, 2015, 119, 4898–4906.
69. M. Hu, A. A. Sukhanov, X. Zhang, A. Elmali, J. Zhao, S. Ji, A. Karatay and V. K. Voronkova, J. Phys. Chem. B, 2021, 125, 4187–4203.
70. M. H. Lee, J. H. Han, J. H. Lee, N. Park, R. Kumar, C. Kang and J. S. Kim, Angew. Chem., Int. Ed., 2013, 52, 6206–6209.
71. C. Wang and K. M.-C. Wong, Inorg. Chem., 2013, 52, 13432–13441.
72. P. Majumdar, X. Yuan, S. Li, B. Le Guennic, J. Ma, C. Zhang, D. Jacquemin and J. Zhao, J. Mater. Chem. B, 2014, 2, 2838–2854.
73. Y. Zhao, Y. Ren, H. Li, T. Han, H. Chen and W. Guo, Dyes Pigm., 2016, 132, 255–261.
74. X. Cui, J. Zhao, A. Karatay, H. G. Yaglioglu, M. Hayvali and B. Küçüköz, Eur. J. Inorg. Chem., 2016, 5079–5088.
75. Q. Qiao, W. Liu, J. Chen, X. Wu, F. Deng, X. Fang, N. Xu, W. Zhou, S. Wu, W. Yin, X. Liu and Z. Xu, Angew. Chem., Int. Ed., 2022, 61, e202202961.
76. X. Cui, J. Zhao, Z. Lou, S. Li, H. Wu and K.-L. Han, J. Org. Chem., 2015, 80, 568–581.
77. L. D. Patsenker and Y. Y. Artyukhova, J. Mol. Struct., 2003, 655, 311–320.
78. R. Ziessel, B. D. Allen, D. B. Rewinska and A. Harriman, Chem. – Eur. J., 2009, 15, 7382–7393.
79. J. Hankache and O. S. Wenger, Chem. – Eur. J., 2012, 18, 6443–6447.
80. M. A. Collini, M. B. Thomas, V. Bandi, P. A. Karr and F. D'Souza, Chem. – Eur. J., 2017, 23, 4450–4461.
81. B. Ventura, A. Bertocco, D. Braga, L. Catalano, S. d’Agostino, F. Grepioni and P. Taddei, J. Phys. Chem. C, 2014, 118, 18646–18658.
82. D. Gosztola, M. P. Niemczyk, W. Svec, A. S. Lukas and M. R. Wasielewski, J. Phys. Chem. A, 2000, 104, 6545–6551.
83. G. Tang, A. A. Sukhanov, J. Zhao, W. Yang, Z. Wang, Q. Liu, V. K. Voronkova, M. Di Donato, D. Escudero and D. Jacquemin, J. Phys. Chem. C, 2019, 123, 30171–30186.
84. X. Xiao, Y. Yan, A. A. Sukhanov, S. Doria, A. Iagatti, L. Bussotti, J. Zhao, M. Di Donato and V. K. Voronkova, J. Phys. Chem. B, 2023, 127, 6982–6998.
85. P. K. Samanta and N. J. English, J. Phys. Chem. C, 2020, 124, 8178–8185.
86. T. Gerbich, H.-C. Schmitt, I. Fischer, R. Mitrić and J. Petersen, J. Phys. Chem. A, 2016, 120, 2089–2095.
87. X. Chen, J. Pang, M. Imran, X. Li, J. Zhao and M. Li, Photochem. Photobiol. Sci., 2021, 20, 69–85.
88. W. Tian, A. A. Sukhanov, L. Bussotti, J. Pang, J. Zhao, V. K. Voronkova, M. Di Donato and M.-D. Li, J. Phys. Chem. B, 2022, 126, 4364–4378.
89. H. Imahori, M. E. El-Khouly, M. Fujitsuka, O. Ito, Y. Sakata and S. Fukuzumi, J. Phys. Chem. A, 2001, 105, 325–332.
90. K. Xu, Y. Xie, X. Cui, J. Zhao and K. D. Glusac, J. Phys. Chem. B, 2015, 119, 4175–4187.
91. M. Imran, A. A. Sukhanov, P. Maity, A. Elmali, J. Zhao, A. Karatay, O. F. Mohammed and V. K. Voronkova, J. Phys. Chem. B, 2021, 125, 9244–9259.
92. B. K. Rugg, B. T. Phelan, N. E. Horwitz, R. M. Young, M. D. Krzyaniak, M. A. Ratner and M. R. Wasielewski, J. Am. Chem. Soc., 2017, 139, 15660–15663.
93. D. Mims, J. Herpich, N. N. Lukzen, U. E. Steiner and C. Lambert, Science, 2021, 374, 1470–1474.
94. X. Zhao and J. Zhao, Chem. Commun., 2022, 58, 7666–7669.
95. X. Zhang, X. Zhao, K. Ye and J. Zhao, Chem. – Eur. J., 2023, 29, e202203737.
96. M. Shah, N. S. Allen, M. Edge, S. Navaratnam and F. Catalina, J. Appl. Polym. Sci., 1996, 62, 319–340.
97. A. Samanta, B. Ramachandram and G. Saroja, J. Photochem. Photobiol., A, 1996, 101, 29–32.
98. D. W. Cho, M. Fujitsuka, A. Sugimoto, U. C. Yoon, D. W. Cho and T. Majima, Phys. Chem. Chem. Phys., 2014, 16, 5779–5784.
99. D. W. Cho, M. Fujitsuka, U. Chan Yoon and T. Majima, Phys. Chem. Chem. Phys., 2008, 10, 4393–4399.
100. P. Xiao, W. Hong, Y. Li, F. Dumur, B. Graff, J. P. Fouassier, D. Gigmes and J. Lalevée, Polym. Chem., 2014, 5, 2293–2300.
101. F. Dumur, Eur. Polym. J., 2022, 163, 110962.
102. P. Majumdar, R. Nomula and J. Zhao, J. Mater. Chem. C, 2014, 2, 5982–5997.
103. X. Xiao, W. Tian, M. Imran, H. Cao and J. Zhao, Chem. Soc. Rev., 2021, 50, 9686–9714.
104. X. Chen, A. A. Sukhanov, Y. Yan, D. Bese, C. Bese, J. Zhao, V. K. Voronkova, A. Barbon and H. G. Yaglioglu, Angew. Chem., Int. Ed., 2022, e202203758.
105. A. Karimata, H. Kawauchi, S. Suzuki, M. Kozaki, N. Ikeda, K. Keyaki, K. Nozaki, K. Akiyama and K. Okada, Chem. Lett., 2013, 42, 794–796.
106. B. H. Drummond, N. Aizawa, Y. Zhang, W. K. Myers, Y. Xiong, M. W. Cooper, S. Barlow, Q. Gu, L. R. Weiss, A. J. Gillett, D. Credgington, Y.-J. Pu, S. R. Marder and E. W. Evans, Nat. Commun., 2021, 12, 4532.
107. X. Xiao, I. Kurganskii, P. Maity, J. Zhao, X. Jiang, O. F. Mohammed and M. Fedin, Chem. Sci., 2022, 13, 13426–13441.
108. S. Yamauchi and N. Hirota, J. Chem. Phys., 1987, 86, 5963–5970.
109. M. Terazima, S. Yamanuchi and N. Hirota, J. Phys. Chem., 1985, 89, 1220–1227.
110. M. Terazima, S. Yamauchi and N. Hirota, Chem. Phys. Lett., 1985, 120, 321–326.
111. M. Terazima, S. Yamauchi and N. Hirota, J. Chem. Phys., 1986, 84, 3679–3687.

---

Footnotes

† Electronic supplementary information (ESI) available: General experimental methods, synthesis of compounds, molecular structure characterization, computational details and additional spectra. CCDC 2294222 and 2294224. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d3cp04891h

‡ These authors contributed equally to this work.

---