A local minimum in the conduction band of a semiconductor, representing a distinct electron state.In semiconductor physics, a valley is a local minimum in the conduction band (or maximum in the valence band) in momentum space. Electrons in a crystal can occupy states near these minima, and each valley represents a distinct set of electron states with specific momentum and effective mass. In materials like silicon or graphene, multiple valleys exist, and the valley degree of freedom — which valley an electron occupies — can be used as an information carrier, similar to electron spin. This has led to the field of valleytronics, which aims to encode and manipulate information using valley states.
The concept of valleys in semiconductors dates back to the 1950s, with early work on silicon's multi-valley band structure by Conyers Herring and others. The term 'valleytronics' was coined in the 2000s as researchers began exploring the valley degree of freedom for information processing.
The band structure of a semiconductor arises from the periodic potential of the crystal lattice. Electrons near a valley behave like free particles but with an effective mass that depends on the curvature of the band at that point. By applying external fields — electric, magnetic, or optical — one can selectively populate or read out electrons in different valleys. In materials like transition metal dichalcogenides (e.g., tungsten diselenide), the valleys are coupled to light polarization, allowing optical control of the valley population.
Valleytronics is being explored for next-generation electronic and optoelectronic devices, such as valley-based transistors and valley-polarized light emitters, which could offer faster and more energy-efficient computing.
In graphene, electrons in different valleys behave like they have opposite 'pseudo-spin', and this valley degree of freedom is so robust that it can be used to create topologically protected states that are immune to scattering.



