Spintronics

The field of spintronics uses the electron spin, rather than its charge, as the primary means of processing information, and has already revolutionised the way in which we store data. The discovery of giant magnetoresistance earned Albert Fert and Peter Grünberg the Nobel Prize in Physics in 2007, and has enabled highly sensitive read-heads for high-density magnetic disk drives, as well as magnetic random access memories (MRAMs). It also offers the potential for a dramatic improvement in the energy efficiency of information processing: by transmitting information not through electrical currents but via pure spin currents (which involve a flow of spin angular momentum with no net flow of charge), it may be possible to significantly reduce Ohmic power dissipation.

The main promise of using organic semiconductors in spintronics lies in the long spin lifetimes that can be achieved in materials composed primarily of carbon and hydrogen. A spin’s orientation and coherence often survive for several microseconds in molecular or polymer semiconductors, which is a time period more than a thousand times longer than in many inorganic metals. This provides more time to manipulate a spin and potentially allows it to travel further. Long spin lifetimes arise from the light-element composition of these mainly carbon-based materials: there is little perturbation from nuclear spins because the abundant carbon-12 isotope has zero nuclear spin. In addition, the radial electric fields of the nuclei are small, resulting in weak spin-orbit interaction, which turns out to be both a blessing and a curse. It is highly desirable for spin-transmission layers; however, most spintronic devices (such as MRAM) require the ability to create spin polarisations from electrical currents or to convert spin currents into electrical voltages, and these processes require strong spin-orbit coupling. Therefore, it would be useful to be able to tune the strength of spin-orbit coupling.

In 2014, we were awarded a six-year Synergy project by the European Research Council (ERC) to investigate the fundamental spin physics of these organic systems and to explore their use in spintronic devices. The project involves a collaboration between the Cavendish Organic Semiconductor Group, the Hitachi Cambridge Laboratory (Jörg Wunderlich – spintronic devices), the University of Mainz (Jairo Sinova – theory), and Imperial College (Iain McCulloch – chemistry). The capital equipment budget of the project enabled us to install a highly sensitive electron spin resonance (ESR) spectrometer in the Cavendish.

We first aimed to understand how much spin-orbit coupling can be tuned in molecular materials through chemical design. To achieve this, we trapped spins on isolated molecules by dispersing them in solution at sufficiently low concentrations to minimise interactions between them, and studied their ESR response. Previously, it had been assumed that the incorporation of heavier elements such as sulphur or selenium into a molecule was the key to increasing spin-orbit coupling; however, we found that changes to the molecular geometry could be just as effective. In fact, by introducing a curvature into the backbone of a well-known molecule, we observed that the resulting shifts in the charge distribution drastically reduced spin-orbit coupling and increased spin lifetimes from just above 1 µs to 100 µs. Conversely, when aiming to increase spin-orbit coupling, it is necessary to strategically position heavier elements at sites where their interaction with the spin is maximised.

Figure 1: (a) Field-induced ESR spectra as a function of gate voltage in a high-mobility, conjugated-polymer transistor architecture; (b) temperature dependence of the spin relaxation times T1 and T2; (c) schematic diagram of different magnetic field environments encountered when charges hop from site to site.

Next, we investigated how structural changes affect spins moving through a stack of molecules in the solid state. At low and intermediate temperatures, we observed that spin lifetimes were closely linked to the mobility of charge carriers: the motion of charges through an environment of nuclear spins and spatially varying spin-orbit coupling randomises the spin orientation (Figure 1). However, when approaching room temperature, charge motion and spin relaxation appeared to become decoupled. The onset and magnitude of this effect could only be explained by vibrations of the polymer chain causing local fluctuations in the charge density. Spin-orbit coupling links these fluctuations to the spins, eventually causing them to flip. These results have practical implications for polymer design: by suppressing the relevant vibrational modes, it may be possible to engineer high-mobility polymers with both strong spin-orbit interactions and long spin lifetimes.

Figure 2: (a) Schematic showing the generation, propagation, and detection of a spin current within an organic semiconductor film using a lateral spin-pumping architecture; (b) measured voltage in the Pt spin detector in the vicinity of ferromagnetic resonance. The bottom inset shows a photograph of an actual device.

Armed with this fundamental understanding of spin relaxation mechanisms, we also investigated spin transport in device structures. One such structure (Figure 2) is based on spin-current injection from a ferromagnetic electrode via spin pumping, which involves microwave excitation of ferromagnetic resonance precession of the magnetisation. The injected spin current diffuses laterally through an organic semiconductor film and is detected by a platinum electrode with strong spin-orbit coupling, where the spin current is converted into a measurable electrical voltage. The magnitude of this voltage decays exponentially with distance from the detector, allowing the determination of the spin diffusion length. We have obtained evidence for spin-diffusion lengths of several micrometres in certain organic semiconductors. Further measurements are currently underway to validate these findings unambiguously.

Our ERC project is entering its final phase, and we are currently pursuing a number of related research directions, including achieving efficient spin-charge conversion and realising molecular spin qubits in our materials. Whether organic semiconductors will ultimately prove useful for spintronic applications remains to be seen; however, the knowledge gained through this fundamental research is already providing new insight into spin physics. The microscopic understanding of the interplay between molecular structure, charge transport, and spin dynamics that we have achieved has wide-ranging implications for the physical phenomena that can be observed in organic systems.

References

1. Ando, K, et al. Solution-processed organic spin–charge converter. Nature Materials12, 622 (2013).
2. Sinova, J, et al. Spin hall effects. Reviews of Modern Physics87, 1213 (2015).
3. Schott, S, et al. Tuning the effective spin-orbit coupling in molecular semiconductors. Nature Communications8, 15200 (2017).
4. Schott, S., et al. Polaron spin dynamics in high-mobility polymeric semiconductors. Nature Physics 15, 814-822 (2019).
5. Wang, S., et al. Long spin diffusion lengths in doped conjugated polymers due to enhanced exchange coupling. Nature Electronics2,98-107 (2019).

In collaboration with: