Search

Machine learning is a valuable tool in the development of chemical technologies but its applications into supramolecular chemistry have been limited. Here, the utility of kernel-based support vector machine learning using density functional theory calculations as training data is evaluated when used to predict equilibrium binding coefficients of small molecules with cucurbit[7]uril (CB[7]). We find that utilising SVMs may confer some predictive ability. This algorithm was then used to predict the binding of drugs TAK-580 and selumetinib. The algorithm did predict strong binding for TAK-580 and poor binding for selumetinib, and these results were experimentally validated. It was discovered that the larger homologue cucurbit[8]uril (CB[8]) is partial to selumetinib, suggesting an opportunity for tunable release by introducing different concentrations of CB[7] or CB[8] into a hydrogel depot. We qualitatively demonstrated that these drugs may have utility in combination against gliomas. Finally, mass transfer simulations show CB[7] can independently tune the release of TAK-580 without affecting selumetinib. This work gives specific evidence that a machine learning approach to recognition of small molecules by macrocycles has merit and reinforces the view that machine learning may prove valuable in the development of drug delivery systems and supramolecular chemistry more broadly.

External-stimuli controlled reversible formation of radical species is of great interest for synthetic and supramolecular chemistry, molecular machinery, as well as emerging technologies ranging from (photo)catalysis and photovoltaics to nanomedicine. Here we show a novel hybrid colloidal system for light-driven reversible reduction of chemical species that, on their own, do not respond to light. This is achieved by the unique combination of photo-sensitive plasmonic aggregates and temperature-responsive inorganic species generating radicals that can be finally accepted and stabilised by non-photo-responsive organic molecules. In this system Au nanoparticles (NPs) self-assembled via sub-nm precise molecular spacers (cucurbit[n]urils) interact strongly with visible light to locally accelerate the decomposition of dithionite species (S2O42-) close to the NP interfaces. This light-driven process leads to the generation of inorganic radicals whose electrons can then be reversibly picked up by small organic acceptors, such as the methyl viologen molecules (MV2+) used here. During light-triggered plasmon- and heat-assisted generation of radicals, the S2O42- species work as a chemical 'fuel' linking photo-induced processes at the NP interfaces with redox chemistry in the surrounding water environment. By incorporating MV2+ as a Raman-active reporter molecule, the resulting optically-controlled redox processes can be followed in real-time.

External-stimuli controlled reversible formation of radical species is of great interest for synthetic and supramolecular chemistry, molecular machinery, as well as emerging technologies ranging from (photo)catalysis and photovoltaics to nanomedicine. Here we show a novel hybrid colloidal system for light-driven reversible reduction of chemical species that, on their own, do not respond to light. This is achieved by the unique combination of photo-sensitive plasmonic aggregates and temperature-responsive inorganic species generating radicals that can be finally accepted and stabilised by non-photo-responsive organic molecules. In this system Au nanoparticles (NPs) self-assembled via sub-nm precise molecular spacers (cucurbit[n]urils) interact strongly with visible light to locally accelerate the decomposition of dithionite species (S(2)O(4)(2–)) close to the NP interfaces. This light-driven process leads to the generation of inorganic radicals whose electrons can then be reversibly picked up by small organic acceptors, such as the methyl viologen molecules (MV(2+)) used here. During light-triggered plasmon- and heat-assisted generation of radicals, the S(2)O(4)(2–) species work as a chemical ‘fuel’ linking photo-induced processes at the NP interfaces with redox chemistry in the surrounding water environment. By incorporating MV(2+) as a Raman-active reporter molecule, the resulting optically-controlled redox processes can be followed in real-time.

Plasmonic nanostructures can focus light far below the diffraction limit, and the nearly thousandfold field enhancements obtained routinely enable few- and single-molecule detection. However, for processes happening on the molecular scale to be tracked with any relevant time resolution, the emission strengths need to be well beyond what current plasmonic devices provide. Here, we develop hybrid nanostructures incorporating both refractive and plasmonic optics, by creating SiO 2 nanospheres fused to plasmonic nanojunctions. Drastic improvements in Raman efficiencies are consistently achieved, with (single-wavelength) emissions reaching 10 7 counts⋅mW −1 ⋅s −1 and 5 × 10 5 counts∙mW −1 ∙s −1 ∙molecule −1 , for enhancement factors >10 11 . We demonstrate that such high efficiencies indeed enable tracking of single gold atoms and molecules with 17-µs time resolution, more than a thousandfold improvement over conventional high-performance plasmonic devices. Moreover, the obtained (integrated) megahertz count rates rival (even exceed) those of luminescent sources such as single-dye molecules and quantum dots, without bleaching or blinking.