In ferromagnetic/non-magnetic bilayers of nickel/titanium, the precession of magnetization in the nickel layer can inject an orbital current into the titanium layer through orbital pumping, the orbital counterpart of spin pumping.

A membrane of polymer fibres containing semi-embedded liquid metal particles can be selectively ruptured with a patterned stamp to produce stretchable circuits with high resolution and interfacial adhesion between the liquid metal and the polymer.

An individual-particle passivation strategy that reduces ion migration in perovskite nanocrystal film can be used to make high-refresh-rate active-matrix displays with microsecond response times reduced by three orders of magnitude compared with typical perovskite light-emitting diodes.

Probabilistic-bit-based Ising machines implemented on field-programmable gate arrays can be used to train artificial intelligence networks with the same performance as software-based approaches while using fewer model parameters.

A neuromorphic computational model based on multigate silicon nanowire transistors can perform dendritic computation by integrating synaptic organization with dendritic tree-like morphology and can be used to develop a multilayer network system that emulates three-dimensional spatial motion perception in the retina.

A yttrium-doped metallic two-dimensional buffer layer can be used to improve charge carrier transport between the metal contacts and semiconductor channel in molybdenum-disulfide-based transistors.

Heterogeneous integration of chips in three-dimensional systems will be needed to meet increasing global demands for computing power.

The three-dimensional integration of electronic and photonic integrated circuits could solve critical input/output limitations in existing computing chips, and create larger, more complex chips for application in future data centres and high-performance systems.

Three-dimensional electronics is our 2024 technology of the year.

Two-dimensional (2D) semiconductors could be used to build advanced 3D chips based on monolithic 3D integration. But challenges related to growing single-crystalline materials at low temperatures — as well as enhancing the performance of 2D transistors — need to be addressed first.

Three-dimensional technology — which can offer enhanced integration density and improved data communication — will be required to build large-scale artificial computing systems inspired by the brain.