Silicon chips are being turned into atomic-scale laboratories where individual impurity atoms act as artificial molecules, solids, and transistors. This research tackles a fundamental gap: single-atom devices in silicon have shown enormous potential for quantum computing and spintronics, but no one has yet built them at a scale that matters. The team will use deterministic doping—placing individual atoms with atomic precision—to create wafer-scale arrays of these structures. They will also add magnetic dopants and germanium layers, expanding the range of possible devices. If successful, the work could produce stable, scalable quantum processors and simulators that do not require the elaborate laser-cooling and vacuum systems needed for cold-atom experiments. These “frozen atom chips” would be inherently scalable, electrically contactable, and compatible with existing semiconductor manufacturing. That compatibility matters because any new quantum technology must integrate with conventional electronics to be practical. This is primarily fundamental science. The immediate output will be new understanding of how overlapping impurity wavefunctions behave in silicon. But the long-term payoff—a manufacturable platform for quantum information—could transform computing, secure communications, and metrology, much as earlier silicon research underpinned the entire digital age.
View original technical description
The aim of the ADDRFSS Programme is to exploit deterministic doping to explore the fundamental physics and processing requirements of alternative, disruptive silicon-based semiconductor device paradigms for quantum information technologies, spintronics, optically integrated electronics and metrology. Specifically, we will produce a great variety of "molecule" and "lattice" structures, and exploit them for new physics and new devices with a major focus on scale-up and manufacturability. Single defects in semiconductors, placed with atomic-scale precision, have suggested enormous potential for new quantum and classical devices, termed "solotronics", but what is required now is practical implementation of such device concepts. Silicon offers the exciting possibility of using wavefunctions built up by arrangement of overlapping impurity wavefunctions as atomic-scale functional device components and has the crucial advantage of a huge knowledge base. Although high-volume, low-cost CMOS research is not a UK priority, any new quantum technologies must be compatible with it to enable process integration with existing IC technology. By advancing our deterministic doping capabilities for high throughput doping, and broadening our chemical specificity to allow use of magnetic dopants and thin germanium doped layers, we will produce devices of wafer-scale dimensions and diverse functionality. One example of the classical architectures we will develop will incorporate the smallest possible silicon devices with form of an n(+)-n(++)-n transistor, where there are three donors of different ionization potential (e.g. P-Sb-Bi etc). Current flow from left to right is blocked unless the potential of the central donor is lowered to be between that of the ends. At a more fundamental level, artificial solids in cold-atom lattices are generating great excitement due to their ability to transfer quantum information along the lattice, and to entangle multiple atoms. The aim is to realize large scale quantum computers and quantum simulators in and out of equilibrium (for modelling e.g. phase transitions in high-Tc superconductors etc that are particularly difficult to model with classical computers). The silicon "molecules" and "solids" we propose to build are also attractive for these purposes but with significant benefits: the impurities can be trapped inside a Si "vacuum" permanently and direct images of wavefunctions can be obtained with scanning tunnelling microscopy, and there is no need for elaborate optical and gas-handling systems - the resulting "frozen" atom chips are stable and inherently scalable and their costs will inevitably fall as they always have done for the semiconductor technology. We have established that the main disadvantage (non-radiative relaxation) is surmountable and we aim to mirror the atom-trap developments with a system that is scalable and electrically contactable, both with lithographically patterned wires and through scanning probe tips.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know