Thin layers of two non-magnetic metals – copper and manganese – become magnets when they are in contact with buckminsterfullerene molecules. This discovery has been made by physicists in the UK, US and Switzerland, and could lead to new types of practical electronic devices and even quantum computers.
Ferromagnets – such as familiar fridge magnets – are materials that have permanent magnetic moments. There are only three metals that are ferromagnetic at room temperature – iron, nickel and cobalt – and this is explained in terms of the “Stoner criterion”, which was first derived in 1938 at the University of Leeds by Edmund Stoner.
Stoner knew that magnetism in metals is a property of the conduction electrons. These electrons are subject to the exchange interaction that allows them to reduce their energy by aligning their spin magnetic moments in the same direction – thus creating a ferromagnetic metal. However, having spins that point in the same direction increases the overall kinetic energy of the electrons. Stoner realized that ferromagnetism will only occur when the reduction in energy caused by exchange is greater than the gain in kinetic energy. Quantitatively, he showed that this occurs when the product of the electron density of states (DOS) – the number of energy states available to the electrons – and the strength of the exchange interaction (denoted by U) is greater than one.
Giving U a boost
U is called the Stoner criterion, and it is greater than one for iron, nickel and cobalt but not for their neighbours in the periodic table – manganese and copper. Now, an international team including Fatma Al Ma’Mari and Tim Moorsom of the University of Leeds in the UK has found a way to boost the DOS and exchange interaction in copper and manganese so that they are ferromagnetic at room temperature.
The team made its samples by depositing several alternating layers of C60 and copper (or manganese) onto a substrate. The copper layers were about 2.5 nm thick and the C60 layers about 15 nm thick. C60 is used because it has a large electron affinity, which means that each molecule will take up to three conduction electrons from the copper. This is expected to increase both the DOS and the strength of the exchange interaction in copper.
The team then measured the magnetization of the layered samples and found them to be ferromagnetic materials. The researchers also looked at samples in which the copper and C60 layers were separated by layers of aluminium and found no evidence of magnetism, which suggests that ferromagnetism occurs at the interfaces between the copper and C60. This was backed up by experiments using muons, which are depth-sensitive and showed that the ferromagnetism occurs in the copper near to the C60 interface. The team also found room-temperature ferromagnetism in C60/manganese layers, but with a weaker magnetization.
Critical field
Surprisingly, when the researchers calculated U for their copper samples, they found it to be less than one. In other words, the samples should not have been ferromagnetic according to the Stoner criterion. However, further theoretical investigations suggest that the samples should become ferromagnet when exposed to a relatively small magnetic field – something that would have happened during the preparation of the samples. This suggests that other non-magnetic metals could be made ferromagnetic by boosting U but not necessarily all the way to one.
Although further work is needed to increase the strength of the copper and manganese magnets, the research could result in the development of new types of tiny magnetic components. These could find use in spintronic devices, which use the spin of the electron to store and process information, or even in quantum computers in which electron spins are used as quantum bits of information.
The research is described in Nature.
