A bifurcated knot is synthesized via two sequential and selective imine condensation reactions. The bifurcated knot contains 3 classical crossing points and 20 bifurcated junctions. This strategy enables the tying of molecular strand termini, granting access to topologically intricate molecules with increased robustness.

An interface-regulated vapour crystallization strategy is reported that enables precise control over the spectral linewidth of solution-processed halide perovskite thin films. The demonstrated perovskite films have low inhomogeneous broadening and photoluminescence linewidths as low as 13.6 nm at 464 nm, enabling efficient perovskite light-emitting diodes spanning pure blue to pure green.

Alkyl boronates are typically involved in direct C–B bond transformations. Now the use of boronates as directing groups to enable β-selective C(sp³)–H functionalization is reported. The process uses Lewis pair chemistry between an in situ-generated alkyl BF₂ and a 2-haloheteroarene, selectively forming β-heteroarylated alkyl boronates under photoredox conditions.

Industrial self-driving laboratories need more than safe robots and instruments. They need an autonomy safety harness that defines how AI-generated intent becomes executable experiments, monitored actions and trustworthy evidence for downstream research and development (R&D).

Bis-functionalization of unactivated alkanes is synthetically challenging and typically requires multiple steps. Now, a method for the direct aziridination of unactivated alkanes is reported, using a double hydrogen atom transfer process facilitated by photo-generated triplet sulfonyl nitrenes and iodosobenzene.

Conventional electrodeposition often yields ill-defined morphologies and aggregated nanocrystals. Here, by linking growth behaviour to lattice mismatch and metal–metal bond energy, seed-assisted electrodeposition enables the controlled synthesis of structurally defined multimetallic nanocrystals, including core–shell, multilayered and hollow architectures, through programmable electrochemical pathways.

Independently tuning pore sizes and functionalities remains challenging in multi-porous framework materials. Now, triskele-shaped building units enable selective pore modulation in dual-porous hydrogen-bonded organic frameworks through independent precursor tuning. The resulting materials show exceptional acid stability and enhanced proton conductivity following the incorporation of superacids.

Conventional hydrazide-based native chemical ligation is often limited by the need for multiple steps and pH adjustments. Now a ferricyanide-mediated oxidation system is reported that enables ligation of peptide hydrazides with N-terminal cysteine peptides in a single neutral buffer, streamlining the synthesis of diverse proteins and cyclic peptides.

Alcohols are regarded as ideal C(sp³) synthons, yet efficient C–O bond activation methods remain scarce for cross-coupling. Now, a Ni-catalysed electrochemical double dehydroxylative cross-coupling for C(sp³)–C(sp³) construction is reported. This versatile method achieves late-stage methylation of bioactive molecules and simplifies drug molecule synthesis.

Owing to competitive strain-induced ring-opening and isomerization side reactions, enantioselective single-atom skeletal editing of saturated carbocycles, particularly bridged hydrocarbons, remains challenging. Now a method for enantioselective single-atom skeletal editing of bicyclo[1.1.0]butanes via Rh-catalysed carbene insertion is reported. Using triftosylhydrazones as carbene precursors, enantioenriched 1,2,2,3-tetrasubstituted bicyclo[1.1.1]pentanes can be formed.

Self-driving laboratories (SDLs) can accelerate scientific discovery, but their broader value lies in supporting scale translation for multiscale process development. SDLs should generate scale-aware candidate landscapes that link laboratory-scale experimental results to representative-device and process-development stages, distinguishing candidate regions as validated for next-scale testing, or excluded by industrial constraints, or requiring targeted exploration to reduce scale-dependent uncertainty.

Industrial self-driving laboratories need more than safe robots and instruments. They need an autonomy safety harness that defines how AI-generated intent becomes executable experiments, monitored actions and trustworthy evidence for downstream research and development (R&D).