Graphene chip 200 times more sensitive than silicon and will be lower cost

Researchers from the National University of Singapore (NUS) have developed a new material that can potentially replace the silicon widely used in sensor chips in appliances, paving the way for smaller and cheaper electronic products. A new graphene-based sensor chip is more than 200 times more sensitive than current chips available in the market that use silicon or indium antimonide.

Supported by boron nitride substrates, the graphene-based sensor chip forms an interface whereby electrons can move very fast in the material.

“These electrons can thus respond to magnetic fields with greater sensitivity,” said the professor, who is from NUS’ Department of Electrical and Computer Engineering. The material is in the process of getting patented.

In addition, the graphene material is also flexible “like paper” — which makes it suitable to be incorporated into wearable technology — and demonstrates stable performance across temperature changes. Current silicon-based sensor chips will break when bent, and change in properties between room temperature and 127°C — the maximum temperature at which most electronic products operate — affecting their reliability.

With a graphene-based chip, the NUS researchers found a gain in sensitivity of more than eight-fold at 127°C. This makes it a suitable chip for environments with a higher temperature.

Said Prof Yang, “This can axe the need for the current temperature correction mechanism in car sensors, for example. There are about 30 to 40 magnetic sensors … in a car alone.”

Researchers from NUS developed a new hybrid magnetic sensor that has been shown to be more than 200 times more sensitive than commercially available sensors. (Photo: NUS)

Nature Communications – Extremely large magnetoresistance in few-layer graphene/boron–nitride heterostructures

Cheaper than Silicon

The cost to harvest graphene is also much lower compared to silicon.

“By principle, sand is very cheap, but the numerous processes to get silicon from it cost a fortune,” he explained. “Graphene, on the other hand, can be easily mined from graphite (a crystalline form of carbon). Carbon is everywhere in cloth, plants or even insects.”

Following the findings, the researchers at NUS plan to scale their current 2cm-wide sensor up at least 10 times for industry use, said Prof Yang. A common wafer size for silicon is 30cm.

“The industry wants big wafers so they can cut and shape from it to cut costs. The challenge is in keeping the graphene sensor material stable and consistent in bigger wafers,” said Prof Yang.

The sensor chip industry is expected to grow to US$2.9 billion (S$4 billion) in five years, according to NUS. The industry was estimated to be worth US$1.8 billion last year.

Abstract – Extremely large magnetoresistance in few-layer graphene/boron–nitride heterostructures

Understanding magnetoresistance, the change in electrical resistance under an external magnetic field, at the atomic level is of great interest both fundamentally and technologically. Graphene and other two-dimensional layered materials provide an unprecedented opportunity to explore magnetoresistance at its nascent stage of structural formation. Here we report an extremely large local magnetoresistance of~2,000% at 400 K and a non-local magnetoresistance of >90,000% in an applied magnetic field of 9 T at 300 K in few-layer graphene/boron–nitride heterostructures. The local magnetoresistance is understood to arise from large differential transport parameters, such as the carrier mobility, across various layers of few-layer graphene upon a normal magnetic field, whereas the non-local magnetoresistance is due to the magnetic field induced Ettingshausen–Nernst effect. Non-local magnetoresistance suggests the possibility of a graphene-based gate tunable thermal switch. In addition, our results demonstrate that graphene heterostructures may be promising for magnetic field sensing applications.