Organic dyes absorb visible light strongly, can be synthesized and modified with precision, and are attractive building blocks for photocatalysis and artificial photosynthesis.[1] Yet they often face two persistent problems: they degrade under irradiation and, when they aggregate, their excited states are usually quenched before they can do useful chemistry. For this reason, aggregation is often treated as something to avoid in molecular photocatalysis. In our work, we asked whether this assumption could be reversed. Could aggregation itself become the feature that activates an organic dye for photocatalysis?
The idea is conceptually simple. Some organic dyes lose absorbed energy through intramolecular motion: twisting, rotation or structural relaxation in the excited state. If these motions are restricted, more excited states survive long enough to react. This principle is well known in aggregation-induced emission,[2,3] where molecules become brighter when they assemble. We wondered whether the same rigidification could also make them better photocatalysts (Figure 1a).
This project began from a serendipitous observation. Recently, we reported dye-based aggregates in which aggregation enhanced fluorescence and could be exploited for photocatalysis.[4,5] These results suggested that emissive aggregation might be a powerful handle for light-driven chemistry, but they also revealed a limitation: both systems depended on rather specific packing motifs, making it difficult to generalize the concept. We therefore moved from serendipitous discovery to a deliberately designed system, choosing distyrylanthracene (DSA), as a tunable amphiphilic dye platform that self-assembles in water and whose aggregation could be controlled through counterions, concentration and salt addition (Figure 1b). By changing these parameters, we could tune whether the molecules remained dissolved or formed supramolecular nanostructures. The key observation was that the most emissive aggregates were also the most photocatalytically active. Fluorescence thus became a reporter of a deeper process: aggregation was increasing the population of reactive excited states (Figure 2).
The work was initiated by an undergraduate researcher (Luca Vaccarin) and then took over by two PhD students (Marianna Barbieri and David Cappelletti). A key aspect of the work was the collaboration needed to connect molecular design, supramolecular structure, excited-state dynamics and photocatalytic function. Synthetic chemistry and photocatalysis provided the first clues, but the full picture required microscopy, three-dimensional electron diffraction, transient absorption spectroscopy, and mechanistic photochemical experiments. Contributions from colleagues in Padova, Bologna, Modena and Rigaku Europe allowed us to move beyond the simple observation that the aggregates were brighter, and to understand why they became more reactive under irradiation.
The assembled dyes produced reactive oxygen species under light irradiation. Mechanistic experiments showed that oxygen was required and identified superoxide as a key intermediate. In the absence of a probe molecule, this reactivity could be channelled into hydrogen peroxide formation. The same aggregates could also support hydrogen evolution in the presence of Pt nanoparticles and a sacrificial donor. In the dissolved state, the dye was essentially inactive; upon salt-induced aggregation, hydrogen production turned on (Figure 2).
One of the most interesting parts of the study was that the “best” aggregates were not the most thermodynamically ordered ones. Kinetically trapped aggregates outperformed annealed, thermodynamic structures in both reactive oxygen species generation and hydrogen evolution. This was surprising because many supramolecular photocatalysts are designed to maximize extended order, exciton migration and charge separation. Here, the opposite lesson emerged: local excited-state confinement and pathway-dependent aggregation can be more important than long-range structural perfection. This point was important for us because it changes how one might design soft organic photocatalysts. Instead of asking only how to obtain the most crystalline or electronically coupled assembly, we can ask how to trap a packing arrangement that best preserves reactive excited states. In this sense, supramolecular pathway control becomes a functional photocatalytic parameter.
We also wanted to know whether the concept was limited to one dye scaffold. We therefore tested a structurally different tetraphenylethylene (TPE) derivative, a classic aggregation-induced emission motif. Even though this chromophore forms amorphous rather than crystalline aggregates, it also became fluorescent and photocatalytically active upon aggregation. This supported the broader idea that aggregation-induced rigidification can activate different organic chromophores for light-driven chemistry.
More broadly, this work suggests that aggregation does not need to be the enemy of molecular photocatalysis. When properly controlled, it can suppress unproductive decay, improve photostability, enable recovery of the active material and retain some of the molecular precision of homogeneous dyes while gaining features usually associated with heterogeneous photocatalysts.
Do you want to know more? Take a look at our article in Nature Chemistry:
Supramolecular dye polymers for aggregation-induced photocatalysis
https://doi.org/10.1038/s41557-026-02151-4. I hope you enjoy it!
References
[1] Romero, N. A. & Nicewicz, D. A. Organic photoredox catalysis. Chem. Rev. 116, 10075–10166 (2016).
[2] Würthner, F. Aggregation-induced emission (AIE): a historical perspective. Angew. Chem. Int. Ed. 59, 14192–14196 (2020).
[3] Mei, J., Leung, N. L. C., Kwok, R. T. K., Lam, J. W. Y. & Tang, B. Z. Aggregation-induced emission: together we shine, united we soar! Chem. Rev. 115, 11718–11940 (2015).
[4] Đorđević, L. et al. Mechanical and light activation of materials for chemical production. Adv. Mater. 37, 2418137 (2025). https://doi.org/10.1002/adma.202418137
[5] Barbieri, M. et al. Controlled aggregation of pyrene-based supramolecular nanostructures for light-driven switchable H2 or H2O2 production. Adv. Funct. Mater. 36, 2505835 (2026). https://doi.org/10.1002/adfm.202505835