Economist – Silicon photodetectors are inflexible, not particularly cheap, not that sensitive and absorb only 10-20% of the light that falls on to them. For years, therefore, engineers have been on the lookout for a cheap, bendable, sensitive photodetector. Such a device could have many novel applications—wearable electronics, for example. With a little clever engineering, graphene seems to fit the bill. By itself, graphene is worse than silicon at absorbing light. According to Dr Koppens only 2.7% of the photons falling on it are captured. But he and his colleague Gerasimos Konstantatos have managed to increase this to more than 50% by spraying tiny crystals of lead sulphide onto the surface of the graphene.
These lead sulphide crystals are so small (three to ten nanometres across, a nanometre being a billionth of a metre) that they are known as quantum dots, because at dimensions measured in nanometres the weird effects of quantum mechanics start to manifest themselves. One such is that the size of a quantum dot affects the colour of the light it best absorbs. The larger the dot, the redder that light; the smaller, conversely, the bluer. This allows Dr Koppens and Dr Konstantatos to span all wavelengths from ultraviolet to infra-red, greatly increasing the utility of any photodetector that might emerge. Infra-red, for example, is important in telecoms and night-vision applications. Visible wavelengths, by contrast, are needed for cameras and solar cells.
Hybrid graphene–quantum dot phototransistor.
According to Dr Koppens, the interaction between the dots and the graphene works because graphene has so many mobile electrons in its structure. (This is the reason it is such a good conductor of both heat and electricity.) This abundance of free electrons makes it particularly sensitive to the changes induced in a quantum dot when it absorbs a photon of light: each incident photon mobilises about 100m electrons. In the jargon of electronic engineering, therefore, the quantum dot-graphene hybrid has enormously high “gain”. And that means the material might have even wider applications than snazzy cameras and smart clothing. For what Dr Koppens and Dr Konstantatos have actually done is to create the guts of a transistor that is regulated by light.
Ordinary transistors are switches in which one electric current (usually a weak one) is used to regulate the passage of another (usually much stronger). Any signal carried by the weak current is thus amplified into one carried by the strong current—a high-gain system.
Such transistors are the workhorses of conventional electronics. But optoelectronic transistors, particularly those with high gain, are much harder to make. Which is a pity, for they are greatly in demand in the world’s telecoms networks, in which signals are processed locally as electrons but are transmitted long-distance as light.
At the moment Dr Koppens and his colleagues say their goal is to create “the thinnest and most flexible detector in the world”. It is notable, however, that they actually deposited their experimental quantum dot-graphene photodetector onto a piece of silicon. Their purpose in doing so was to show that the technology meshes with the standard silicon-processing techniques used to make computer chips.
Graphene is an attractive material for optoelectronics1 and photodetection applications because it offers a broad spectral bandwidth and fast response times. However, weak light absorption and the absence of a gain mechanism that can generate multiple charge carriers from one incident photon have limited the responsivity of graphene-based photodetectors to ~10^−2 A W−1. Here, we demonstrate a gain of ~10^8 electrons per photon and a responsivity of ~10^7 A W−1 in a hybrid photodetector that consists of monolayer or bilayer graphene covered with a thin film of colloidal quantum dots. Strong and tunable light absorption in the quantum-dot layer creates electric charges that are transferred to the graphene, where they recirculate many times due to the high charge mobility of graphene and long trapped-charge lifetimes in the quantum-dot layer. The device, with a specific detectivity of 7 × 10^13 Jones, benefits from gate-tunable sensitivity and speed, spectral selectivity from the short-wavelength infrared to the visible, and compatibility with current circuit technologies.