Graphene coated with nanoparticles has been used to make wearable light sensors that measure the human pulse and blood oxygen levels from ambient light passing through tissue, offering a potential platform for health-care monitoring.

Deji Akinwande & Dmitry Kireev

The popularity of wearable technology has risen enormously, with the US market projected to be in the tens of billions of dollars by 2022 (see go.nature.com/33tcein). However, the effectiveness of the most common wearable devices is hindered by the physical specifications of their components: although the device is often embedded in a flexible soft shell, the main parts, such as the sensors and electronics, are still rigid1,2. Now, writing in Science Advances, Polat et al.3 report a class of truly flexible, transparent wearable device that is based on graphene covered with a layer of semiconducting nanoparticles known as quantum dots. Impressively, the devices measure various vital signs using only ambient light as a signal.

Materials that are just one or a few atoms thick are said to be two-dimensional. The best-known example is graphene, which consists of single sheets of carbon atoms arranged in a hexagonal lattice. 2D materials in general, and graphene in particular, have tremendous potential for the development of next-generation wearable, soft biosensors because they combine electrical conductivity, optical transparency and mechanical flexibility with outstanding biocompatibility4 and stability to biological electrolytes. Graphene-based tattoo-like devices5 have previously been used to record human health signals such as heart rhythm, skin hydration and body temperature. Their outstanding performance is associated with the subnanometre thickness of graphene, which allows it to bend and stretch with the skin, without affecting the sensor performance.

Flexible self-powered biosensors

Polat et al. have now expanded the functionality of graphene in wearable devices by depositing light-sensitive quantum dots made of the semiconductor lead(ii) sulfide (PbS) onto the graphene layer. When illuminated, the quantum dots generate pairs of charged particles: negatively charged electrons and positively charged holes (quasiparticles associated with the absence of an electron in an atomic lattice). The electrons stay trapped in the quantum dots, but the holes are transferred into the graphene layer and increase its electrical conductivity, producing a measurable electrical signal. The authors used this behaviour to construct light sensors from the quantum-dot-coated graphene.

The researchers observed that the responsivity (the electrical output per optical input) of their devices was remarkably large. The high responsivity is attributable to the fact that the holes in the graphene layer are recycled by the quantum dots, effectively increasing the number of charge carriers generated per absorbed photon in the devices — the devices are said to exhibit a photoconductive gain.

Previously reported light sensors typically do not have photoconductive gain, and therefore require an amplifier device to boost the electrical signal; this increases both power consumption and the size of the overall device6. Moreover, the amplifier must be in close proximity to the sensor, which can limit the ability of wearable devices to take on the contour of the skin. The intrinsic photoconductive gain of Polat and colleagues’ devices eliminates the need for an amplifier, solving the above problems and making the sensors particularly suitable for real-life applications.

Integration of optoelectronics into fibres enhances textiles

So how were the sensors used to measure vital signs? Light at certain wavelengths passes easily through human skin and adjacent tissue, but is absorbed strongly by blood7 — more specifically, it is absorbed by haemoglobin, the molecule that transports oxygen in red blood cells. By continuously monitoring the intensity of light passing through tissue, sensors can produce read-outs called photoplethysmograms (PPGs) that contain information about volumetric changes to blood vessels, which can be correlated to heart rate8. Polat and colleagues show that their wearable devices can, remarkably, use the ambient light that passes through tissue to measure human heart rates accurately. Moreover, the sensitivity of the devices allowed the researchers to estimate the rate of breathing by mathematically analysing the PPG data. Physical movements associated with breathing usually produce artefacts and noise in the PPG signals detected by rigid wearable devices9, but the physical unobtrusiveness and flexibility of the new devices overcome this problem.

Polat et al. report that their wearable devices can also monitor another vital health signal that is often checked by doctors: arterial oxygen saturation (SpO2), which is the percentage of haemoglobin in blood that is loaded with oxygen (Fig. 1). Low SpO2 levels can result in loss of consciousness, impaired mental functions, and respiratory and cardiac arrest. The absorption of red light and near-infrared light by oxygen-rich red blood cells is significantly different from the absorption by oxygen-free cells. The authors therefore estimated SpO2 levels by using their devices to measure light absorption at these two wavelengths.

Figure 1 | A sensor that detects vital signs using light. a, Polat et al.3 have used graphene coated with semiconducting nanoparticles (not shown) to make flexible, transparent devices that can detect the light transmitted through tissue. Ambient light passes easily through human skin and tissue, but is absorbed by the haemoglobin in red blood cells. By monitoring the modulation of ambient light as it passes through tissue, the sensors can thus monitor blood flow and heart rate. The rate of respiration can also be estimated from mathematical analysis of the blood-flow data. b, Moreover, absorption of red and near-infrared light by oxygen-rich red blood cells is significantly different from that by oxygen-poor cells. By monitoring the absorption of red and near-infrared light, the sensors can monitor the oxygen content of the blood.

Finally, Polat and colleagues’ reported a further application of their technology: the monitoring of ultraviolet light. Certain UV wavelengths can be harmful to the skin, and can potentially even cause cancer10, making it desirable to measure UV levels in the environment. The authors show that their devices can be integrated with previously fabricated chips that enable the sensors to wirelessly transfer UV measurements to a mobile phone, thus enabling continuous and convenient monitoring of the environmental UV index.

The reported sensors are all designed to communicate wirelessly to any other electronics needed for a wearable device, clearly separating the soft sensor from any rigid components. But the wireless design requires a read-out device (such as a mobile phone) to be close to the sensor, which makes it difficult to perform long-term monitoring — as might be needed for heart-rate monitoring, for example. Establishing long-term, continuous communication between the wearable flexible sensors and conventional electronics will be essential for future applications. Alternatively, it might be possible to include components that enable memory storage and simple digital processing in the flexible platform. This could be achieved in the future using 2D materials other than graphene11.

Graphene has now been used as a sensor and as a signal transducer in various prototypes for wearable and mobile health devices12. More importantly, however, graphene has paved the way for other 2D materials to be used in sensors and mobile health-monitoring devices. Thousands of such materials have been discovered, with as-yet unknown properties13. We think that the comprehensive study of those materials will be essential for the development of future biosensors that can be worn by, or even integrated into, humans.