Quantitative energetic and kinetic model for self-regulated valence and geometry in the co-assembly of block-copolymer micelles and hard nanoparticles

The discovery that spherical polystyrene-block-poly(2-vinylpyridine) micelles and unvarnished hard nanoparticles (silica, Au, polydopamine, ZIF-8) co-assemble through reversible hydrogen-bonding and coordination interactions into 'colloidal molecules' with self-regulated valence remains mechanistically qualitative: the Janus intermediate, the balance between hydrogen-bond adhesion and corona steric repulsion, and the preference for triangular and tetrahedral cluster shapes are described only in probabilistic language, and no quantitative model predicts saturated valence, cluster geometry, or pathway selectivity from particle size, corona dimensions, feeding ratio, temperature, or solvent. The problem remains open: the closest follow-up establishes a qualitative kinetic-competition framework in a related electrostatic system, theory work covers single-polymer binding kinetics, and existing simulations address designed rather than emergent valence. The open task is a quantitative energetic–kinetic model, grounded in the underlying reversible corona–surface interactions, validated against the published valence, morphology, and kinetics observations.

Background

Colloidal molecules are small clusters of nanoparticles with molecule-like defined geometries (dimers, trimers, tetrahedra, chains), sought as building blocks for metamaterials, plasmonics, and catalysis. The standard route fabricates patchy particles — particles carrying discrete attractive surface regions that act as directional bonds — but regioselective surface functionalization at the nanoscale is extremely difficult. An alternative demonstrated by Cui, Zhu, Cai and Qiu (Nature Communications 12, 5682; 2021) uses two ordinary ingredients: soft micelles and hard nanoparticles. A diblock copolymer $S_mV_n$ (polystyrene-block-poly(2-vinylpyridine), with degree-of-polymerization $m$ of the polystyrene block and $n$ of the P2VP block) self-assembles in ethanol into a spherical micelle — denoted $A$ — with a glassy polystyrene core and a 'hairy' corona of P2VP chains; the hard nanoparticle (silica, denoted $B$, or Au, silica-coated Au, polydopamine, or ZIF-8 — a zeolitic imidazolate framework, a metal–organic framework of Zn ions bridged by imidazolate linkers) is denoted by its chemistry. The pyridyl groups of the P2VP corona form multiple reversible hydrogen bonds to silanol groups on silica (or coordinate to Au and Zn sites), so the corona acts as a soft, multifunctional adhesive shell. On mixing, the particles co-assemble into colloidal molecules and alternating 'colloidal copolymers' whose valence — the number of micelles bound around one hard nanoparticle — is not pre-programmed but self-regulated by size and stoichiometry: 50 nm silica saturates at valence 2, 75 nm at 3 (triangular $BA_3$ tetramers), and 90 nm at 4 (tetrahedral $BA_4$ pentamers), while longer P2VP coronas lower the saturated valence, and the number feeding ratio of micelles to nanoparticles selects among silica-capped trimers, linear and branched chains, 2D networks, and stacked 3D monoliths, in close analogy to alternating condensation polymerization. The proposed mechanism is that a first micelle covers roughly one hemisphere of a silica nanoparticle, creating a Janus intermediate (a cluster with two chemically distinct faces) whose two poles then grow independently. Crucially, this mechanism is stated only qualitatively and with probabilistic hedges ('probably'): the balance between hydrogen-bond adhesion and inter-corona repulsive/dispersive forces, and the energetic origin of the triangular and tetrahedral preferences, are not computed, and no energetic or kinetic model predicts the observed valences, geometries, or morphology map. Kinetic control is likewise empirical — extended networks and monoliths required assembly at 80 °C to weaken hydrogen bonding and retard rapid aggregation — and on-surface sequential growth of colloidal oligomers stalled because surface-bound $BA_2$ trimers distort into 'v' shapes. Subsequent experimental work (reviews of colloidal-atom engineering, macroscopic composites, dialysis-controlled and electrostatically engineered colloidal molecules) has extended the paradigm but not supplied the missing quantitative description, while quantitative simulation and theory exist mainly for systems in which valence is pre-programmed by designed patches, DNA encoding, or external fields.

Problem Statement

Construct and validate a quantitative energetic and kinetic model of the self-regulated co-assembly of hairy block-copolymer micelles and unvarnished hard nanoparticles — the system of Cui, Zhu, Cai & Qiu (Nature Communications 12, 5682; 2021, https://doi.org/10.1038/s41467-021-25995-5) — that predicts, from the experimentally controllable parameters (hard-nanoparticle diameter; micelle core size and the diblock copolymer lengths $m$ of the core-forming polystyrene block and $n$ of the corona-forming P2VP block; micelle-to-nanoparticle number feeding ratio; temperature; solvent), the three experimentally observed classes of outcome: (i) the saturated valence of the hard nanoparticle (observed: 2, 3 and 4 for 50, 75 and 90 nm silica against $S_{2090}V_{380}$ micelles, decreasing for longer P2VP coronas); (ii) the equilibrium geometry of the saturated valence clusters (micelles-capped $BA_2$ trimers, regular triangular $BA_3$ tetramers, and tetrahedral $BA_4$ pentamers), which the source rationalizes only qualitatively; and (iii) the pathway and kinetic selectivity of assembly — Janus-intermediate-mediated growth, alternating-condensation-polymerization behavior, and the dynamically trapped architectures of the feeding-ratio-dependent morphology map (silica-capped trimers, linear and branched colloidal copolymers, 2D networks, and 3D stacked monoliths) — including consistency with the reported kinetic control whereby assembly at 80 °C weakens hydrogen bonding and retards rapid association. The model must be grounded in the underlying physical interactions — cooperative multiple hydrogen bonds between corona pyridyl groups and surface silanol (or metal/imidazolate) sites, corona–corona steric repulsion and dispersion, and corona–solvent interactions — with parameters that are independently measurable or transferable rather than refit to each observation, and it must be validated against the observations published in that paper and its Supplementary Information: the saturated valences 2, 3 and 4 for 50, 75 and 90 nm silica against $S_{2090}V_{380}$; the decrease of saturated valence for longer P2VP coronas; the regular triangular $BA_3$ and tetrahedral $BA_4$ geometry of saturated clusters; the feeding-ratio-dependent morphology map from capped trimers through branched colloidal copolymers to 2D networks and 3D monoliths; and the 80 °C kinetic-control effect. Resolution criterion: with the model's parameters fixed before comparison with these observations — each parameter either independently measured or transferred from an established model of the same materials, and none fitted to the target observations — the task is resolved when the model reproduces the integer saturated valences 2, 3 and 4 for 50, 75 and 90 nm silica and the regular triangular $BA_3$ and tetrahedral $BA_4$ geometries of the saturated clusters, and reproduces as qualitative trends the decrease of saturated valence for longer P2VP coronas, the feeding-ratio-dependent morphology map, and the retarded association at 80 °C.

References

1. Cui, Y., Zhu, H., Cai, J. & Qiu, H. Self-regulated co-assembly of soft and hard nanoparticles. Nature Communications 12, 5682 (2021). DOI: 10.1038/s41467-021-25995-5, https://doi.org/10.1038/s41467-021-25995-5
2. He, H., Yao, C., Shen, X., Tao, J., Zheng, D., Dai, L., Qiao, R., Sang, Y. & Nie, Z. Highly scalable, dialysis-controlled synthesis of colloidal molecules via regulated self-assembly of polymer-grafted gold nanoparticles. Nature Communications 17, 9042 (2026). DOI: 10.1038/s41467-026-75712-3, https://doi.org/10.1038/s41467-026-75712-3
3. Cui, Y., Xing, Y., Hou, J., Zhang, H. & Qiu, H. Co-Assembly of Soft and Hard Nanoparticles into Macroscopic Colloidal Composites with Tailored Mechanical Property and Processability. Small 20, 2401432 (2024). DOI: 10.1002/smll.202401432, https://doi.org/10.1002/smll.202401432
4. Cui, Y., Wang, J., Liang, J. & Qiu, H. Molecular Engineering of Colloidal Atoms. Small 19, 2207609 (2023). DOI: 10.1002/smll.202207609, https://doi.org/10.1002/smll.202207609
5. Tang, Q., Huang, Y. & Müller, M. Predicting protracted binding kinetics of polymers: Integral of first-passage times. Physical Review E 110, 044502 (2024). DOI: 10.1103/PhysRevE.110.044502, https://doi.org/10.1103/PhysRevE.110.044502
6. Zhang, N.-N., Shen, X., Liu, K., Nie, Z. & Kumacheva, E. Polymer-Tethered Nanoparticles: From Surface Engineering to Directional Self-Assembly. Accounts of Chemical Research 55, 1503–1513 (2022). DOI: 10.1021/acs.accounts.2c00066, https://doi.org/10.1021/acs.accounts.2c00066
7. Shen, X., He, H., Zheng, D., Cao, W., Sang, Y. & Nie, Z. Engineering the Electrostatic Interactions between Oppositely Charged Polymer-Grafted Nanoparticles for Constructing Colloid Molecules on Substrates. ACS Nano 18, 20999–21008 (2024). DOI: 10.1021/acsnano.4c01891, https://doi.org/10.1021/acsnano.4c01891
8. Qiu, X., Tang, H., Zhang, L. & Wang, R. Directional Self-Assembly of Programmable Atom-like Nanoparticles into Colloidal Molecules. Journal of Physical Chemistry Letters 16, 3141–3148 (2025). DOI: 10.1021/acs.jpclett.5c00056, https://doi.org/10.1021/acs.jpclett.5c00056
9. Zhang, Y. & Wu, N. Numerical simulations of electric-field-driven assembly of binary colloids into colloidal molecules and crystals. Journal of Chemical Physics 163, 194901 (2025). DOI: 10.1063/5.0287245, https://doi.org/10.1063/5.0287245
10. Fonseca, J., Jiao, L., Broto-Ribas, A., Singh, G. & Dong, A. Co-assembly of nanometer- and submicrometer-sized colloidal particles into multi-component ordered superstructures. Soft Matter 21, 5583–5654 (2025). DOI: 10.1039/D5SM00247H, https://doi.org/10.1039/D5SM00247H
11. Zhou, Y., Dong, J., Wang, X., Song, Z., Lu, Q., Jiang, J. & Wang, Q. Programming the Valence and Orientation of Anisotropic Nanoparticles via Three-Dimensional DNA Ligand Encoding. JACS Au 5, 2350–2358 (2025). DOI: 10.1021/jacsau.5c00380, https://doi.org/10.1021/jacsau.5c00380
12. Fan, Y., Ma, J., Li, Y., Huang, X., Feng, S. & Chen, D. Co-assembly of Synthetic Particles with Heterogenous Components. Chemistry of Materials 36, 4011–4033 (2024). DOI: 10.1021/acs.chemmater.4c00134, https://doi.org/10.1021/acs.chemmater.4c00134
13. Fielden, S. D. P., Collins, S. M., Derry, M. J., Ducati, C., Fairclough, S. M., Miller, A. J., O'Reilly, R. K. & Topham, P. D. Kinetically controlled hetero-fusion is a systems-level behaviour of polymer nanoparticle populations. Nature Communications 16, 11701 (2025). DOI: 10.1038/s41467-025-66827-0, https://doi.org/10.1038/s41467-025-66827-0
14. Ma, W., Shen, J., Zhang, J., Zhu, Y., Zhou, M., Shi, G., Qiao, X., He, Y., Zhang, W. & Pang, X. Precision Colloidal Polymerization of Hierarchical Gold Nanocluster Assemblies. ACS Nano 20, 8471–8486 (2026). DOI: 10.1021/acsnano.5c19127, https://doi.org/10.1021/acsnano.5c19127
15. Lizano-Villalobos, A., Ma, F., Tang, W., Sun, W. & Tang, X. Machine Learning-based Optimal Control for Colloidal Self-Assembly. arXiv:2512.16402 (2025). arXiv: 2512.16402, https://arxiv.org/abs/2512.16402
16. Lyu, D., Xu, W., Zhou, N., Duan, W., Wang, Z., Mu, Y., Zhou, R. & Wang, Y. Biomimetic thermoresponsive superstructures by colloidal soft-and-hard co-assembly. Science Advances 9, eadh2250 (2023). DOI: 10.1126/sciadv.adh2250, https://doi.org/10.1126/sciadv.adh2250