Introduction

In biological systems, information transfer occurs through diverse pathways, with the interplay between energetic and chiral signal transmission playing a crucial role in complex processes such as recognition, replication, and catalytic activity1,2,3,4. A prime example is the light-harvesting systems of higher plants and green algae (Fig. 1a), where proteins create chiral environments around chlorophyll chromophores, enabling energy transfer that is essential for catalytic function5,6,7. Inspired by these natural systems, chemists have developed artificial chiral light-harvesting systems by incorporating donor–acceptor chromophores into chiral scaffolds provided by supramolecular gels8,9 and films10,11,12,13 (Type I-mimetic systems in Fig. 1b). This approach potentially brings dual chirality and energy transfer and thereby generate circularly polarized luminescence (CPL) in the excited state, holding significant promise in 3D displays14,15,16, optical sensors17,18,19,20, and photoelectric materials21,22,23. However, while these artificial systems can achieve efficient energy transfer through an optimized donor-to-acceptor ratio, they often face challenges in transferring chirality information from scaffolds to donor–acceptor chromophores24,25. This limitation arises because artificial scaffolds lack the precise chromophore positioning inherent of natural chlorophyll–protein complexes, thereby impairing stereo-communication between chiral scaffolds and achiral chromophores26.

Fig. 1: Schematic illustration for simultaneous chirality and energy transfer of donor–acceptor chromophores via bio-inspired supramolecular light-harvesting.
figure 1

a Two prototypes of natural light-harvesting systems. The protein-scaffolding systems are reproduced with permission from ref. 7 b Chiral scaffold-based artificial light-harvesting systems as inspired by the natural protein-scaffolding systems in higher plants and green algae. c Schematic illustration of an artificial light-harvesting system designed to mimic chlorosomes, which facilitates simultaneous chirality and energy transfer in a CTAB-confined aqueous environment. The chemical structures of the platinated donors (R)-1–(R)-2, the acceptor 3, the control compounds 4, and (R)−5 are also included.

Full size image

To address this issue, another natural light-harvesting system, chlorosomes (Fig. 1a) from green photosynthetic bacteria, offers a promising strategy27. In chlorosomes, bacteriochlorophyll chromophores are encapsulated within a hydrophobic environment formed by amphiphilic glycolipids and phospholipids28. These chromophores stack directly to form helical nanostructures, notably without requiring chiral protein scaffolds29. The chiral supramolecular organization of bacteriochlorophylls induces strong exciton coupling between neighboring chromophores, resulting in high light-harvesting efficiency30. Inspired by this natural design, we propose that helical stacking of artificial donor and acceptor chromophores could not only enable efficient energy transfer but also promote chirality transfer, compared to scaffold-supported artificial systems. This is due to the oriented donor/acceptor spacing and enhanced stereo-communication within supramolecular nanostructures. The dual transfer of chirality and energy could significantly improve CPL performance in artificial systems.

To validate this approach, herein we constructed artificial light-harvesting systems using σ-platinated donor–acceptor chromophores (Type II-mimetic systems in Fig. 1c). In this design, the anthracene-substituted compound (R)−1 or the tetrafluorobenzene-substituted (R)−2 serves as the donor chromophores. These chromophores densely pack into helical supramolecular aggregates due to the two-fold hydrogen bonding interactions between neighboring monomers. Additionally, the surfactant CTAB (cetyl trimethyl ammonium bromide) mimics the role of glycolipids and phospholipids in natural chlorosomes by providing a confined hydrophobic space that encapsulates the stacked chromophores. Introducing a small amount of the benzoselenadiazole-substituted acceptor 3 into this CTAB-confined environment facilitates direct donor–acceptor stacking, enabling efficient energy transfer and resulting in an antenna efficiency as high as 23.9 in water. Meanwhile, chirality transfer in the excited state leads to the emergence of CPL signals in the acceptor’s red-colored emission region. In contrast, when using the control compound 4 (Fig. 1c) as the acceptor, which lacks amide units and therefore cannot face-to-face stack with the donor chromophore (R)-1 or (R)-2, CPL signals are absent due to inefficient chirality transfer, despite comparable energy transfer efficiency. Furthermore, we aimed to determine whether energy transfer and chirality transfer occur simultaneously or sequentially by employing femtosecond fluorescence up-conversion (fs-FUC) spectroscopy and femtosecond time-resolved circularly polarized luminescence (fs-TRCPL) spectroscopy. A detailed analysis of the time-resolved CPL kinetics revealed a gradual increase in CPL signal intensity, occurring within the same timescale as the energy transfer lifetime. Furthermore, the stereochemical compatibility between the donor and acceptor was identified as critical for influencing the acceptor’s CPL signal. These findings unambiguously demonstrate the simultaneous transfer of chirality and energy from donors to acceptors, achieved through the chlorosome-mimetic design.

Results

Monomer design, synthesis, and structural characterization

In the designed light-harvesting systems, compounds (R)-1 and (R)-2 act as the chiral donors, while compound 3 serves as the achiral acceptor. To ensure effective energy transfer, different (hetero)acenes are used in 1–3 to achieve optimal spectral overlap (Supplementary Fig. 22). Additionally, σ-platination of these compounds prevents severe stacking of (hetero)acenes, maintaining sufficient emission intensity31,32. The stereogenic (R)-methyl groups in 1–2 serve as chiral induction units, transferring chirality to the inner σ-platinated (hetero)acenes and inducing single-handed helicity at the supramolecular level. CTAB creates a confined hydrophobic space that encapsulates the designed compounds, mimicking the role of glycolipids and phospholipids in chlorosomes and thereby facilitating their assembly in aqueous environments33,34,35,36.

Supramolecular stacking of the designed compounds is driven by two-fold hydrogen bonds between neighboring amides (Supplementary Fig. 16). The structural similarity between (R)-1 [or (R)-2] and 3 facilitates energy and chirality transfer upon direct co-stacking. As a comparison, compound 4, which lacks the peripheral amides present in 3, is employed as an alternative acceptor. Due to this difference, 4 prefers to be doped into the supramolecular donor stacks rather than forming face-to-face stacks. To assess the efficiency of dual transfer of chirality and energy, we compare the CPL signals from the direct donor–acceptor co-stacks of (R)-1/3 with those from the non-covalent doping mode of (R)-1/4.

The targeted compounds 1–3 were synthesized through copper(I)-catalyzed coupling reactions between the iodide-substituted σ-platinated (hetero)acenes and the alkyne intermediates (Supplementary Figs. 1, 2). The designed compounds were characterized by 1H NMR, 13C NMR, and MALDI–TOF MS (Supplementary Figs. 42–57). For example, during the formation of (R)-1, the alkyne 13C NMR resonances shifted from 83.3 to 130 ppm, accompanied by the disappearance of the 1H NMR signals at 3.06 ppm for the terminal alkyne intermediate (Supplementary Figs. 40, 41).

CTAB-confined stacking of the σ-platinated compounds in water

To ensure supramolecular stacking of compounds 1–3 in water, we injected 200 μL of a tetrahydrofuran solution containing the designed compounds (c = 2.00 mM) into 10 mL of an aqueous CTAB solution (cCTAB = 1.00 mM, which is above the critical micelle concentration). Using compound (R)-1 as an example, the resulting mixture was stirred for five hours at 353 K, allowing the organic solvent to evaporate, and resuspended by centrifugation in a high-speed centrifuge to give a transparent aqueous solution. Transmission electron microscopy revealed the formation of near-spherical nanoparticles with an average size of 120 nm [hereafter referred to as (R)-1NP, Fig. 2a and Supplementary Fig. 4], which was comparable to the hydrodynamic diameter obtained from dynamic light scattering analysis (108 nm, Supplementary Fig. 5). Energy-dispersive X-ray spectroscopy confirmed the presence of platinum and phosphorus elements within the nanoparticles, indicating the successful encapsulation of (R)-1 by CTAB (Supplementary Fig. 6). In contrast, directly injecting the tetrahydrofuran solution of (R)-1 into water without CTAB resulted in an opaque solution, highlighting the crucial role of CTAB in achieving homogeneous aqueous solutions (Supplementary Fig. 7).

Fig. 2: Supramolecular stacking behaviors in CTAB-confined environment.
figure 2

a Transmission electron microscopy (TEM) image of (R)−1NP by drop-casting onto a copper grid. b UV/Vis spectra of (R)-1mono in tetrahydrofuran and (R)-1NP in CTAB aqueous solution with tetrahydrofuran as the pre-dissolving solvent. c CD spectra of (R)-1mono in tetrahydrofuran, (R)-1NP, and (S)-1NP in CTAB aqueous solution with tetrahydrofuran as the pre-dissolving solvent. d LD spectra of (R)-1NP and (S)-1NP. e CD spectra of (R)-1NF and (S)-1NF in methylcyclohexane, together with (R)-1NP and (S)-1NP in CTAB aqueous solution with methylcyclohexane as the pre-dissolving solvent. f Temperature-dependent CD intensities at 482 nm of (R)-1NF in methylcyclohexane (red colors) and (R)-1NP in CTAB aqueous solution with methylcyclohexane as the pre-dissolving solvent (blue colors), each bar represents a single measurement. For all measurements, the concentrations of 1 and CTAB were maintained at 40 μM and 1 mM, respectively.

Full size image

We further studied the UV/Vis spectroscopic properties of (R)-1NP in the CTAB-containing aqueous solution. It exhibited an intense band ranging from 310 to 375 nm (λmax = 320 nm, Fig. 2b), originating from the π–π* intra-ligand transitions of the platinum acetylide (–C≡C–Pt–C≡C–) segment37. Additionally, moderately intense bands appeared between 400 and 525 nm (λmax: 448 nm and 478 nm), attributed to the π–π* transitions of the anthracene’s 1La bands. The 1La bands displayed hypochromic effects (Fig. 2b) compared to the molecularly dissolved state [hereafter referred to as (R)-1mono], indicating supramolecular stacking of (R)-1 in the CTAB-confined environment. CD spectroscopy supported this conclusion. While no chiroptical signals were observed for (R)-1mono, (R)-1NP showed Cotton effects with a negative maximum at 482 nm and a positive one at 438 nm (Δε: –75 mdeg and 13.3 mdeg, Fig. 2c). The absence of linear dichroism (LD) under the same conditions excluded artifacts in the CD measurements (Fig. 2d). Accordingly, these results confirm chirality transfer from the stereogenic (R)-methyl groups to the inner σ-platinated dialkynylanthracenes upon CTAB-confined stacking.

We then examined the effect of pre-dissolving solvents on the CTAB-confined stacking behaviors. The solvents include tetrahydrofuran, 1,4-dioxane, chloroform, dichloromethane, 1,1,2,2-tetrachloroethane, cyclohexane, methylcyclohexane and n-decane (Supplementary Fig. 8). The first five solvents serve as “good” solvents, rendering (R)-1 in a molecularly dissolved state. In contrast, the latter three act as “bad” solvents, facilitating pre-stacking before dispersing into the CTAB aqueous solution. Due to the immiscibility between the chlorinated or alkane solvents and water, additional ultrasound was required to promote the dispersion of (R)-1. Supramolecular chirality signals emerged for nanoparticles prepared from tetrahydrofuran, 1,4-dioxane, chloroform, cyclohexane, methylcyclohexane and n-decane (Supplementary Fig. 9). This indicates that pre-stacking of monomers is not necessary for CTAB-confined stacking in water. Nevertheless, we failed to fabricate clear solutions of (R)-1NP using dichloromethane and 1,1,2,2-tetrachloroethane as the pre-dissolving solvents (Supplementary Fig. 8). Dichloromethane, with its low boiling point, quickly volatilized during sonication, leading to precipitation. Conversely, the high boiling point of 1,1,2,2-tetrachloroethane made it difficult to remove from water, resulting in phase separation.

The CTAB confining effect on the chiroptical signals of (R)-1 was further investigated. When dissolved in apolar methylcyclohexane, (R)-1 tends to stack into one-dimensional nanofibers driven by directional hydrogen bonding and π–π stacking interactions [Supplementary Fig. 10, hereafter referred to as (R)-1NF]. This represents self-stacking nanostructures in non-confined environments. To compare the chiroptical signals of (R)-1 under unconfined and confined conditions, we measured the CD intensities of (R)-1NF and (R)-1NP at a monomer concentration of 40 μM. The CD intensities of (R)-1NP were 1.75 times higher than those of (R)-1NF (gabs: –2.1 × 10-3 versus –1.2 × 10−3 at 482 nm, Fig. 2e, Supplementary Figs. 11, 12). Similar trends were observed at lower concentrations (Supplementary Fig. 11b): (R)-1NP exhibited a CD intensity of –17.3 mdeg at 482 nm at a concentration of 10 μM, in stark contrast to the negligible CD signal of (R)-1NF. This difference can be explained by the amphiphilic nature of CTAB, which shields the (R)-1 stacks from water38. The decreased tendency for solute–solvent interaction increases solute–solute complexation in the CTAB-confined environment, leading to the resistance of supramolecular stacks upon dilution.

Additionally, the CTAB-confined environment endows (R)-1NP with enhanced thermal stability. Upon elevating the temperature, the Cotton signal of (R)-1NF showed a non-sigmoidal melting curve, leading to the complete disappearance of the Cotton effect at 337 K (Fig. 2f and Supplementary Fig. 13). In contrast, the CD signals for (R)-1NP were maintained even at 363 K, albeit with a slight decrease in CD intensity compared to that at 298 K (gabs at 482 nm: –2.0 × 10−3 versus –2.1 × 10−3, Fig. 2f). The strengthened chiroptical signals toward thermal elevation were also present in compound (R)-2NP (Supplementary Figs. 14, 15), validating the impact of confined space endowed by CTAB to enhance supramolecular chirality signals.

Circularly polarized luminescence of (R)-1 and (R)-2 generated upon CTAB confinement

The designed σ-platinated compounds exhibit fascinating emission properties. For example, (R)-1NP showed the 1La emission band centered at 508 nm, along with a vibronic peak at 538 nm in the CTAB-confined environment (Fig. 3a). In contrast, the control species (R)−5NP (Fig. 1c) displayed very weak emission under the same conditions (Fig. 3a and Supplementary Fig. 17a). The difference is ascribed to the σ-platinization effect in (R)−1, which prevents severe aggregation-induced quenching (Supplementary Fig. 17a). Additionally, the emission color can be tuned by modifying the inner (hetero)acene core. For instance, replacing the anthracene core in (R)-1 with benzoselenadiazole shifted the emission wavelength from 508 nm in (R)-1NP to 620 nm in 3NP (Supplementary Fig. 17b), due to the intramolecular charge transfer effect induced by benzoselenadiazole39,40,41.

Fig. 3: Characterization of excited-state supramolecular donor stacks.
figure 3

a Emission spectra of (R)-1NP and (R)-5NP in CTAB-confined environments with tetrahydrofuran as the pre-dissolving solvent (λex = 420 nm). b Emission spectra of (R)−2NP and (R)-2NF without nitrogen bubbling (λex = 365 nm). c CPL spectra of (S)−1NP and (R)-1NP in CTAB-confined environments with tetrahydrofuran as the pre-dissolving solvent, compared with (R)-1mono in tetrahydrofuran. d CPL spectra of (S)−2NP and (R)-2NP in CTAB-confined environments with tetrahydrofuran as the pre-dissolving solvent, compared with (R)-2mono in tetrahydrofuran. For all measurements, the concentrations of the (hetero)acene compounds and CTAB were maintained at 40 μM and 1 mM, respectively.

Full size image

We further examined the emission properties of (R)−2, which incorporates tetrafluorobenzene unit instead of the anthracene unit in (R)-1. This compound exhibited an intriguing dual-emissive character in methylcyclohexane (Fig. 3b and Supplementary Fig. 18): blue emission (λmax: 425 nm) under ambient conditions and yellow-green emission upon nitrogen bubbling (λmax: 540 nm; τ: 36.2 μs). The presence of electronegative fluorine atoms in (R)-2 narrows the singlet–triplet energy gap, enabling triplet emission42. This assertion is supported by DFT and TD-DFT calculations (Supplementary Fig. 19), which show an energy gap of 0.002 eV between the S1 and T2 states. In comparison, the nearest singlet-to-triplet energy gap of (R)-1 is 0.24 eV, preventing triplet emission and resulting in only fluorescent emission. Interestingly, triplet emission is enhanced in a CTAB-confined environment, leading to predominant phosphorescence emission of (R)-2NP even without nitrogen bubbling (Fig. 3b). This phenomenon is due to the CTAB surfactant shielding the (R)-2 stacks, excluding oxygen and preventing the quenching of the triplet emitter43.

Combining supramolecular chirality and emission properties, we investigated the CPL properties of compounds (R)-1 and (R)-2 in the excited state. (R)-1NP demonstrated fluorescent CPL centered at 530 nm (Fig. 3c), although a complete signal could not be obtained due to the small Stokes shift (λex at 420 nm versus λem at 508 nm). In comparison, the CPL signal was absent for (R)-1mono (Fig. 3c), while the glum value of (R)-1NF in unconfined environment exhibited half of that of (R)-1NP (glum: –8 × 10−4 versus –1.5 × 10−3, Supplementary Fig. 20). Additionally, (R)-2NP demonstrated phosphorescent CPL centered at 540 nm (glum: –0.5 × 10−3, Fig. 3d), sharply contrasting with the negligible signals observed in both (R)-2mono (Fig. 3d) and (R)-2NF (Supplementary Fig. 21). Furthermore, both the enantiomeric species (S)-1NP and (S)-2NP displayed mirror-image CPL signals under identical conditions (Fig. 3c, d). Overall, the strengthened CPL signals originate from the enhanced transfer of chirality from the molecular to the supramolecular level within CTAB-confined environments.

Simultaneous transfer of chirality and energy from (R)-1 [or (R)-2] to 3 in water

Given the yellow CPL of chiral donors (R)-1NP and (R)-2NP in CTAB-confined environments and the red emission of the achiral acceptor 3, we aimed to investigate the chirality and energy transfer upon donor–acceptor co-stacking. To achieve this, (R)-1 [or (R)-2] was pre-mixed with 10 mol% of 3 in tetrahydrofuran to ensure thorough mixing, followed by injection into a CTAB-containing aqueous solution. In the resulting species [(R)-1/30.1]NP, a red emission signal at 620 nm strengthened, accompanied by the attenuation of the yellow emission from (R)-1 (Fig. 4a and Supplementary Figs. 22–27). The energy transfer efficiency (ΦET) was determined to be 91%, with an antenna efficiency (AE) of 16.4, corresponding to an energy transfer rate (kET) of 15.3× 109 s−1. This demonstrates effective energy transfer from (R)-1 to 3. Concurrently, the CPL signals of (R)-1 at 530 nm disappeared, and new signals corresponding to the red emission of 3 emerged (glum at 620 nm: –1.7 × 10−3, Fig. 4b and Supplementary Fig. 28a). Given the achiral nature of compound 3, this result confirms the successful transfer of both chirality and energy in the CTAB-confined environment.

Fig. 4: Dual transfer of chirality and energy from (R)−1 [or (R)−2] to 3 in CTAB-confined environment.
figure 4

a Emission spectra of (R)−1NP, [(R)−1/30.1]NP and 3NP. The excitation wavelength was 420 nm for (R)−1NP and [(R)−1/30.1]NP, and 508 nm for 3NP. b CPL spectra of [(R)−1/30.1]NP, [(S)−1/30.1]NP, [(R)−1/40.1]NP and [(S)−1/40.1]NP in CTAB-confined environments. c Emission lifetime decay at 620 nm for [(R)−1/30.1]NP. d Emission spectra of (R)-2NP, [(R)-2/30.1]NP and 3NP. The excitation wavelength was 365 nm for (R)-2NP and [(R)-2/30.1]NP, and 508 nm for 3NP. e CPL spectra of [(S)-2/30.1]NP and [(R)-2/30.1]NP. f Emission lifetime decay at 620 nm for [(R)-2/30.1]NP. For all measurements, the concentrations of donors 1 or 2, acceptor 3 or 4, and CTAB were 40 μM, 4 μM and 1 mM, respectively.

Full size image

We further investigated the fluorescence up-conversion kinetics of [(R)-1/3]NP using fs-FUC spectroscopy combined with fs-TRCPL spectroscopy (Fig. 5a) to determine whether energy transfer and chirality transfer occur simultaneously or sequentially. For the species [(R)-1/30.1]NP, a clear build-up process indicative of energy transfer process was observed in the fluorescence up-conversion kinetics (see the upper panel of Fig. 5b). Fitting the data to a multi-exponential decay function revealed that the first lifetime, corresponding to the build-up process, was 14 ps. Notably, the TRCPL signal (see the bottom panel of Fig. 5b) showed a gradual increase that occurred on a timescale consistent with the build-up process observed in the fluorescence kinetics. This finding confirms the simultaneous transfer of energy and chirality from donor (R)-1 to acceptor 3. Upon increasing the concentration of acceptor 3, the accumulation process accelerated, with lifetimes of 9 ps and 4.2 ps for the species [(R)-1/30.5]NP and [(R)-1/31]NP, respectively. This was accompanied by a gradual increase in CPL signal intensity, occurring on the same timescale as the energy transfer lifetime (Fig. 5c, d, Supplementary Figs. 29, 30 and Supplementary Table 1). These results further confirm that the chiral emission is transferred synchronously with energy transfer from the chiral donor to the achiral acceptor.

Fig. 5: Characterization of fluorescence up-conversion kinetics.
figure 5

a Scheme of the fs-TRCPL experimental setup. (b–d) Fs-FUC (upper panel) and fs-TRCPL kinetic trace (bottom panel) of b [(R)−1/30.1]NP, c [(R)−1/30.5]NP, and d [(R)−1/31]NP. All tests were probed at 680 nm under 400 nm excitation, and the concentrations of (R)−1 and CTAB was 120 μM and 3 mM, respectively. The concentrations of 3 were 12 μM, 60 μM, and 120 μM in the case of (b–d), respectively.

Full size image

When the control compound 4, which retains the σ-platinated benzoselenadiazole unit but lacks amide groups, was used instead of 3 in a confined environment, the resulting species [(R)-1/40.1]NP exhibited energy transfer properties comparable to those of [(R)-1/30.1]NP, with a ΦET of 89% and an AE of 18.4 (Supplementary Fig. 31). However, despite the effective energy transfer, the chirality was not transferred to the acceptor in the excited state, as evidenced by the absence of CPL signals across the spectrum (Fig. 4b and Supplementary Fig. 28b). This failure in chirality transfer is attributed to the absence of amide groups in acceptor 4, which prevents the formation of hydrogen bonds with the donor (R)-1. The absence of hydrogen bonding causes the acceptor to be loosely integrated into the donor’s stacks, without the face-to-face orientation of σ-platinated (hetero)acenes. Consequently, hydrogen bonding between (R)-1 and 3 is crucial for achieving the tight stacking of donor–acceptor chromophores, which is essential for successful chirality transfer in the excited state.

In addition to exploring singlet-to-singlet energy transfer from (R)-1 to 3, we also investigated the possibility of triplet-to-singlet energy transfer by using the triplet emitter (R)-2 as the donor. In the resulting species [(R)-2/30.1]NP, energy transfer was confirmed by the shortened emission lifetime of the donor (R)-2, which decreased from 36.2 μs to 11.8 μs (Supplementary Fig. 32). Meanwhile, the emission lifetime of the acceptor 3 in [(R)-2/30.1]NP was determined to be 234 ns (Fig. 4f), significantly longer than in [(R)-1/30.1]NP under the same conditions (τ = 4.0 ns, Fig. 4c). Notably, triplet-to-singlet energy transfer resulted in a 1.5-fold increase in antenna efficiency for [(R)-2/30.1]NP compared to [(R)-1/30.1]NP (AE: 23.9 versus 16.4, Fig. 4d). Chirality transfer also occurred in [(R)-2/30.1]NP alongside energy transfer, as evidenced by the quenching of yellow CPL at 540 nm and the emergence of red CPL at 620 nm (Fig. 4e). The glum value of 3 upon triplet-to-singlet energy transfer was –1 × 10−3, higher than that of the individual donor (R)-2 under the same conditions (–0.5 × 10−3).

Stereo-dependent chirality and energy transfer behaviors

Next, we aimed to determine whether introducing stereogenic units on the acceptor would influence chirality and energy transfer behaviors. To investigate this, we designed and synthesized a pair of enantiomers, (R)-6 and (S)-6 (Fig. 6 and Supplementary Fig. 3), by incorporating two stereogenic methyl units into 3. When the donor (R)-1 and acceptor (R)-6, both with the same stereocenters, were mixed in CTAB-confined environments, a CPL signal emerged in the red-emissive region for the resulting species [(R)-1/(R)-60.1]NP (Fig. 7a and Supplementary Fig. 33). This signal exhibited a glum value of –1.6 × 10−3 at 620 nm, comparable to the value observed for [(R)-1/30.1]NP (glum: –1.7 × 10−3). Interestingly, the individual compound (R)-6 failed to generate CPL in CTAB-confined environments, despite showing ground-state chiroptical and red fluorescent signals (Supplementary Fig. 34). This suggests that (R)-6NP forms less-ordered homo-stacks in the excited state, likely due to dipole–dipole interactions between neighboring benzoselenadiazole units44. Co-stacking (R)-1 and (R)-6 reduces the likelihood of benzoselenadiazole self-stacking, leading to the emergence of red-colored CPL in (R)-6 through energy and chirality transfer from the donor to the acceptor. In stark contrast, when the donor (R)-1 and acceptor (S)-6, which have opposite stereocenters, were mixed in CTAB-confined environments, negligible CPL signal emerged in the emission region of (S)-6 (Fig. 7a and Supplementary Fig. 33). These results demonstrate stereo-dependent differences in chirality and energy transfer behaviors between the donor and acceptor molecules, leading to variations in CPL signals.

Fig. 6: Schematic illustration of stereo-dependent chirality and energy transfer in CTAB-confined environments.
figure 6

The diagram shows the transfer from the chiral donor (R)−1 to either the chiral acceptor (R)-6 or (S)-6. The chemical structures of the acceptors (R)-6 and (S)-6 are depicted within the frame.

Full size image
Fig. 7: Stereo-dependent chirality and energy transfer behaviors.
figure 7

a CPL spectra in CTAB-confined environments, comparing the complexes [(R)−1/(R)-60.1]NP, [(S)−1/(S)-60.1]NP, [(R)−1/(S)-60.1]NP and [(S)−1/(R)-60.1]NP. The concentrations of donor 1, acceptor 6 and CTAB were 40 μM, 4 μM and 1 mM, respectively. b Melting curves of [(R)−1/(R)-6]NF and [(R)−1/(S)-6]NF in methylcyclohexane. The concentrations of (R)−1 and (R)-6 [or (S)-6] were 40 μM and 80 μM, respectively. c, d Femtosecond fluorescence up-conversion kinetics (upper panel) and Time-resolved CPL kinetic trace (bottom panel) of c [(R)−1/(R)-61]NP and d [(R)−1/(S)-61]NP probed at 680 nm under 400 nm excitation. The concentrations of (R)−1, (R)−6 [or (S)−6] and CTAB were 120 μM, 120 μM and 3 mM, respectively.

Full size image

We hypothesized that the observed differences between [(R)-1/(R)-60.1]NP and [(R)-1/(S)-60.1]NP arises from distinct donor–acceptor stacking modes. As previously documented, melting curves are a powerful tool for probing such arrangements45. However, the enhanced temperature tolerance in CTAB-confined environments (Fig. 2f) prevented us from obtaining melting curves under these conditions. To address this, we instead acquired the melting curves of [(R)-1/(R)-6]NF and [(R)-1/(S)-6]NF in methylcyclohexane (Fig. 7b and Supplementary Figs. 35, 36). In these unconfined environments, only one transition point was observed when the donor (R)-1 and acceptor (R)-6 were mixed in methylcyclohexane (Ttp = 47.3 oC), coinciding with that of the individual compound (R)-1 (Supplementary Fig. 36). This result suggests that (R)-1 undergoes pre-nucleation, followed by the heterogeneous elongation of (R)-6, forming randomly mixed co-stacks46,47. In contrast, when (R)-1 and (S)-6 were mixed, two distinct melting transition points were observed (Ttp1 = 47.0 oC, Ttp2 = 27.8 oC), indicating separate nucleation events. This finding supports a narcissistic arrangement of (R)-1 and (S)-6 in unconfined environments.

The narcissistic stacking of (R)-1/(S)-6 suggests that the donor and acceptor do not engage in direct face-to-face co-stacking, which explains the absence of a CPL signal in [(R)-1/(S)-60.1]NP. In contrast, the randomly mixed co-stacks of (R)-1/(R)-6 facilitate face-to-face co-stacking, resulting in the generation of CPL in the acceptor’s emission region. This conclusion is further supported by fluorescence up-conversion kinetics studies combining fs-FUC and fs-TRCPL spectroscopy. The donor (R)-1 and acceptor (R)-6 [or (S)-6] were mixed in an equivalent ratio to ensure the highest energy transfer efficiency (Supplementary Fig. 37). Specifically, for the species [(R)-1/(R)-61]NP, fitting the data to a multi-exponential decay function revealed that the first lifetime, corresponding to the build-up process, was 3.9 ps (Fig. 7c and Supplementary Table 2). This is attributed to the face-to-face stacking of (R)-1/(R)-6, which results in a helical stacking mode similar to that of (R)-1/3. Both species exhibit comparable kET values (20.8 × 1010 s−1 of [(R)-1/(R)-61]NP versus 23.4 × 1010 s−1 of [(R)-1/31]NP), leading to consistent chirality transfer results. In stark contrast, the first lifetime corresponding to the build-up process was 10 ps for the species [(R)-1/(S)-61]NP (Fig. 7d and Supplementary Table 2). This is rationalized by the narcissistic stacking mode of (R)-1/(S)-6, which significantly reduces its kET (9.25 × 1010 s−1) and consequently diminishes the ability to transfer chirality information.

Discussion

In summary, we successfully achieved simultaneous chirality and energy transfer by developing a chlorosome-mimetic design principle. Similar to the bacteriochlorophyll chromophores in chlorosomes, the donor (R)-1 [or (R)-2] and acceptor 3 adopt a face-to-face stacking arrangement, as facilitated by the presence of amide units on these chromophores to promote intermolecular hydrogen bonding. Simultaneously, CTAB acts like the glycolipids and phospholipids in chlorosomes, providing a confined hydrophobic environment that shields the donor/acceptor co-stacks from water. This setup enabled efficient singlet-to-singlet energy transfer from (R)-1 to 3, and triplet-to-singlet energy transfer from (R)-2 to 3, both with high antenna efficiency. More importantly, the simultaneous chirality transfer from chiral donors to the achiral acceptor led to the emergence of CPL.

It is important to emphasize that since Liu’s pioneering work in 2017, several studies have reported the transfer of CPL via donor-to-acceptor energy transfer8. The generation of CPL is understood to result from the arrangement of π-aromatic chromophores within non-aromatic helical scaffolds. These scaffolds interact with the π-aromatic moieties through electronic or magnetic dipoles, including both transition and static dipoles48. However, these artificial light-harvesting systems typically display CPL of the π-aromatic chromophores only in gel or film states. The emergence of CPL in dilute solutions is rare, primarily because solvents hinder stereo-communication between chiral scaffolds and achiral chromophores (Supplementary Fig. 38). In the current study, this issue is mitigated by the confined environment provided by CTAB, which, along with the face-to-face stacking of donor/acceptor chromophores aided by hydrogen bonds, enables the emergence of CPL in dilute aqueous solutions.

Additionally, introducing stereogenic units on the acceptor resulted in stereo-dependent chirality and energy transfer behaviors, producing an obvious red-colored CPL signal in the (R)-1/(R)-6 system but negligible one in the (R)-1/(S)-6 system. Recent studies by Salam et al49,50. proposed that the energy transfer rate between chiral donor–acceptor molecules consist of two components: a non-discriminatory part and a discriminatory part. The discriminatory component is proportional to the optical rotatory strength of the acceptor, enabling differentiation between enantiomeric acceptors. This hypothesis offers a compelling theoretical framework for enantioselective energy transfer, though experimental validation has so far been limited. Our fs-TRCPL spectroscopy of [(R)-1/(R)-61]NP and [(R)-1/(S)-61]NP provides strong experimental support for this theoretical framework (Supplementary Fig. 39 and Supplementary Table 3). Moreover, it is reasonable to postulate that the difference of kET is largely governed by the orientation factor (κ) of the transition dipole moments between the donor and acceptor moieties (Supplementary Eq. 3). The stacking modes of (R)-1 and (R)-6 [or (S)-6] significantly influence κ during energy transfer, with the face-to-face stacking mode of (R)-1/(R)-6 being more conductive to efficient energy transfer. Furthermore, the chiral helical structure profoundly affects κ during the energy transfer process, resulting in a distinct rate disparity between left- and right-circularly polarized emissions due to the single-handed chiral nanostructures (Supplementary Fig. 39 and Supplementary Table 3). This difference introduces additional emission asymmetry during energy transfer, contributing to the increased fs-TRCPL signal intensity. Overall, the chlorosome-mimetic design exemplified in this study provides avenues for achieving simultaneous chirality and energy transfer in artificial light-harvesting systems.

Methods

Measurements

1H NMR spectra were collected on a Bruker AscendTM 400 MHz spectrometer with TMS as the internal standard. 13C NMR spectra were recorded on a Bruker AscendTM 400 MHz spectrometer at 101 MHz. High-resolution mass spectrometer (MS) data were obtained by LTQ-Orbitrap XL from Thermo-Fisher. MALDI‒TOF measurements were recorded on a Bruker Autoflex Speed spectrometer with DCTB as the matrix. UV–Vis spectra were recorded on a UV-1800 Shimadzu spectrometer. Circular dichroism (CD) measurements were performed on a Jasco J-1500 circular dichroism spectrometer, equipped with a PFD-425S/15 Peltier-type temperature controller. Circular polarized luminescence (CPL) measurements were performed on a Jasco CPL-300 circularly polarized luminescence spectrophotometer. Steady-state emission spectra were recorded on FS5 spectrofluorometer from Edinburg Instruments. Emission lifetime studies were conducted with Fluorolog-3-Tau and Deltaflex (Horiba Scientific) with a semiconductor laser as the excitation source. Fourier transform-infrared (FT-IR) spectra were collected using a Nicolet 6700 FT-IR spectrometer. Atomic force microscope (AFM) measurements (tapping mode) were performed using a Bruker Dimension Icon with ScanAsyst system in air with silica cantilevers (RFESPA-75, Ohm-cm Antimony (n) doped Si) and a resonance frequency of ~ 75 kHz and a spring constant of ~ 3 N m-1. The images were analyzed using the Pico Image processing program. TEM images were performed on a Tecnai G2 Spirit BioTWIN electron microscope (acceleration voltage: 120 kV), and the EDS data were obtained on JEM-F200 (URP).

Theoretical calculations

The calculation was performed using the Gaussian 16 suite of programs. Dispersion interactions were considered, and all calculations were carried out in vacuum. The reliability of the optimized structures was checked via frequency calculations based on density functional theory (DFT, Gaussian 09 W B.01 software package). There are no imagery frequencies for the optimized geometries. The long side chains were substituted by H atoms to make the calculation feasible. The DFT functional ωB97XD was used, together with the basis set 6-31 G(d) was adopted. The molecular orbitals were visualized using the Gauss View 6.0 software. Time-dependent density functional theory (TD-DFT) calculations were performed at the same computational level without adding a solvation model.