Sign In

Communications of the ACM

ACM News

Boosting Photodiode Efficiency to 220%

View as: Print Mobile App Share:
The photodiode.

Eindhoven University of Technology’s super-sensitive photodiode will be used in sensors for medical monitors, wearables, light communication, health surveillance systems, and machine vision.

Credit: Eindhoven University of Technology

The physics of thermodynamics dictates that a photodiode's "energy efficiency" must be less than 100%, defined as "the ratio between the useful output and input of an energy conversion process." However, in sensor applications, it is the quantum efficiency that is important—the ratio between the outputs (electrons) and inputs (photons), and as such is not limited to 100%.

Until now, the highest quantum efficiency of a photodiode used as a sensor, according to its designers at Eindhoven University of Technology (EUT, Netherlands), was 70%. However, the EUT researchers say their newest invention is a tandem-like photodiode (similar to tandem-layer solar cells) that have a hitherto impossibly high quantum efficiency of 220%—since they output 22 electrons for every 10 photons input.

"The tandem-like architecture developed by this group successfully solves several key challenges in the field of near-infrared detection," said photodiode expert professor Fei Huang at the South China University of Technology, who was not involved in the project. "EUT demonstrated devices with a simple but reliable configuration, achieving excellent narrowband detection, very promising operational stability, as well as very low signal-to-noise ratio."

All photodiodes convert photons into electrons—for uses ranging from the near-infrared (NIR) pulse oximeter sensor your doctor clips on your fingertip, to the visible-light solar panels on your roof. As sensors, photodiodes are optimized for superb sensitivity in reflected NIR detection of heart rate, its variability, and blood oxygen levels, whereas photodiodes for solar cells are optimized for energy conversion efficiency from sunlight to electricity. As sensors, photodiodes are usually tuned to different frequencies, too—typically NIR for medical sensors, instead of the visible light spectrum from the Sun for solar cells.

Tandem photovoltaic cells, on the one hand, harvest photonic energy from two different bands of light coming from the Sun, then combine the resulting two separate electron streams, boosting the overall performance of the tandem solar cells to achieve 20% to 40% energy efficiency.

EUT's tandem-like photodiode, on the other hand, achieves greater than 100% quantum efficiency by separately filling a "reservoir" of extra electrons with self-generated photons from an LED shining on the tandem layer. The electrons harvested there are then "gated" into the output electron stream by the incident NIR photons, thus enhancing quantum efficiency far beyond the highest energy efficiency of any solar cell.

"The efficiency that we are talking about is the quantum efficiency—it counts the number of charges (electrons) that pass a circuit per incident photon. This is not really related to the energy efficiency," said Rene Janssen, professor and leader of the interdepartmental research group called Molecular Materials and Nanosystems at EUT. "For our photodiodes, the quantum efficiency is what counts. For a photovoltaic solar cell, it is the energy efficiency that counts. These are related, but the working principle is entirely different—we cannot claim that the effect that we demonstrate here would boost the efficiency of photovoltaic solar cells."

In more detail, tandem solar cells harvest more than one band of the solar spectrum using multi-junction devices to increase their energy output. The tandem-like photodiode architecture designed by EUT, on the other hand, harvests electrons produced by a single band of light (the 830 nanometer wavelength—NIR) which first passes through a donor layer prefilled with electrons harvested from a second green band of light (540nm wavelength) that shines directly on the sensor from a green LED. The two layers, first the 540nm wavelength (inorganic perovskite layer) then the second 830 nm wavelength (organic heterojunction layer), work together to boost quantum efficiency. The principle that differs from the dual-channel solar cells is that the electrons in ETU's donor layer are trapped in there until a "gate" is opened between the two layers by the sensor's detected NIR.

"The 540nm light is mainly absorbed by the perovskite layer. It generates electrons and holes in this layer, but there is a barrier for the extraction of the electrons at the interface between the perovskite layer and the organic layer," said Janssen. "The 830nm light lowers the barrier for electrons at the perovskite/organic interface, which allows the electrons that are created by the green 540nm light to also be collected. One could argue that the 830nm light acts as a means to open the gate between the perovskite and organic layer to make the electrons that are present [in the first perovskite layer] collectible too."

The electrons from the inorganic and the organic layers combine when the gate is opened, boosting the quantum efficiency of the tandem-like photodiode to 220%—as well as reducing the overall power used from 40 milliwatts (mw) to just .2mw.

"So in practice, our tandem-layer photodiode is beneficial from an energy point of view too, but this is not the prime issue," said Janssen.


The researchers call their new tunable super-sensitive biosensor a "vitality monitor" and experimentally demonstrated how it can detect the heart's pulse rate, its regularity, respiration health, and blood oxygenation levels from distances of more than four feet (130 centimeters) away. This outperforms the traditional oximeter used in doctor offices today, which must clip directly to your fingertip and consumes 200 times more power.

The tandem-like photodiode developed at EUT works in a tunable narrow band, allowing it to also reject noise better than today's silicon photodiodes that use optical filters to narrow their bandwidth. In addition, according to the researchers, its ability to work at a distance of four feet away makes possible vitality surveillance applications, like checking the health of people without touching them, similar to the way a terahertz detector checks for weapons at the airport.

Working with the FORSEE project, in collaboration with the Catharina Hospital in Eindhoven, the researchers are developing a camera that can monitor patients' heart and respiration rates without the need to attach monitoring wires. It would reduce the workload of clinical staff who manually attach such wires today, and could automatically monitor and hopefully anticipate deterioration in a patient's condition to catch adverse cardiac events for earlier interventions before they become life-threatening.


R. Colin Johnson is a Kyoto Prize Fellow who ​​has worked as a technology journalist ​for two decades.


No entries found