Tikalon Blog by Dev Gualtieri

Optical Rectenna

July 5, 2021

One of my first encounters with electrical technology as a child was at the local A&P supermarket. which had an automatic door operated by a phototube, The phototube at one side of the exitway was activated by a focused light source at the other side. The light source was likely modulated to allow a reliable detection with interfering light sources. It was no wonder that one of my first electrical circuits was a light-operated relay. The circuit was just the small relay connected to an AC wall outlet in a series connection with a selenium photoconductive cell. Fortunately, electrical experiments by children do not use such dangerous voltages today.

An RCA 918 phototube. Phototubes are extremely simple devices. photons eject electrons from the surface of a photocathode, and these create a current with a nearby anode. The current is generally just a few microamperes, so amplification is needed in a device such as the door-opener described above. This particular phototube, which is ancient enough to have been used in this door-opener, had a peak response to light at about 850 nanometers, but it also had half that ssensitivity to red light, and a quarter that sensitivity in green light. It comes as no surprise that energetic ultraviolet photons produce a much stronger signal than the 850 nm infrared. (Modified Wikimedia Commons image by Grinevitski.)

Phototubes and selenium cells function by photoconductivity, but a more useful photo- phenomenon is the photovoltaic effect in which light energy is converted directly into electrical energy. Solar energy generation is accomplished today through the use of photovoltaic cells, but development of the inexpensive and efficient photovoltaic cells we use today was nearly two centuries in the making.

The photovoltaic effect was discovered in 1839 by French physicist, Edmond Becquerel (1820-1891), who was just 19 at the time. Doing experiments in his father's laboratory, Becquerel discovered the photovoltaic effect in the silver halides, silver chloride and silver bromide, deposited on a platinum electrode. Becquerel also employed such halides in the early development of chemical photography. The state of photoconductive art was advanced in 1866 when English electrical engineer, Willoughby Smith (1828-1891), accidentally discovered the photoconductivity of gray selenium while using it as a semiconductor for testing submarine cable.[1] As Smith explained,

I was induced to experiment with, bars of selenium, a known metal of very high resistance. I obtained several bars varying in length from 5 to 10 centimetres, and of a diameter from 1 to 1-1/2 millimetres. Each bar was hermetically sealed in a glass tube, and a platinum wire projected from each end for the purpose of connection.[1]

Much later, in 1877, Kings College London professor, William Grylls Adams (1836-1915), and his student, Richard Day, observed the photovoltaic effect in selenium bars provided by Willoughby Smith.[2] A candle placed an inch from a selenium bar produced a voltage, and the experimenters confirmed that this was a light effect, and it was not due to the heat of the candle.[2] A few years later, in 1883, Charles Fritts (1850-1903), a New York inventor, coated selenium with a thin layer of gold to create the first solar cell.[3-4] The thin gold layer was somewhat transparent, so it allowed light to reach the selenium surface, but it was electrically conductive. This first solar cell had an energy conversion efficiency of 1-2%.[2-4] Solar cells of crystalline silicon were produced with 6% efficiency at Bell Labs in 1954,[2] and such expensive solar cells enabled communications satellites, such as Telstar.

The photovoltaics of choice for today's solar panels are the inexpensive and efficient thin-film photovoltaics based on cadmium telluride and copper indium gallium selenide. These function by photons causing electrons to jump across the bandgap of a semiconductor junction. Although not yet commercialized, perovskite solar cells are a possibly viable technology if their robustness can be enhanced. I am personally not that excited about perovskites.

Energy conversion efficiencies of CIGS solar cells as a function of time (NREL data)

Energy conversion efficiencies of the best copper indium gallium selenide (CIGS) solar cells worldwide from 1976. (Data derived from a Wikimedia Commons image by Nikos Kopidakis of the National Renewable Energy Laboratory (NREL). Click for larger image.)

Light is an electromagnetic wave, so harvesting light energy with an antenna is also possible. The problems with this approach are that the wavelength of visible light is about 500 nanometers, so your antenna needs to be small; and, the minuscule AC voltage obtained must be rectified to produce a direct current (DC). A team of engineers from the Department of Electrical, Computer and Energy Engineering at the University of Colorado (Boulder, Colorado), has recently published its research on a rectifying antenna they call a rectenna for harvesting energy from the infrared spectrum of light.[5-6] While this device is targeted for harvesting heat energy, the same principle can be used for solar energy harvesting.

Since light-harvesting antennas must be small, they have high resistance. This means a mismatch to the impedance of free space and low efficiency. As for rectification of the voltage, traditional Schottky diodes work well at low frequencies, but their high resistance results in a cut-off frequency of about 40 GHz, far less than light frequency of about 30 terahertz (THz).[5-6] The solution to the rectification problem is the use a quantum mechanical effect called resonant tunneling by which electrons can traverse an insulating gap without losing any energy.[6] Previous resonant tunnel junctions of metal-insulator-metal were limited to voltages an order of magnitude too large for the 100 microvolts required for light-harvesting.[5] The junctions of the present study are a 100 times more efficient.[6]

Bow tie rectenna and its I-V characteristic.

Left, a scanning electron microscope image of the optical rectenna bow tie shape. Right, the current-voltage characteristic of the metal-double insulator-metal rectifier, with the black line showing the simulated response for a 4 nm NiO insulating layer. The reason you need 250,000 antennas is that a microamp and 10 millivolts is just 10 nanowatts. (Left image by the Moddel lab of the University of Colorado. Right image, fig. 2a of ref. 5, licensed under the Creative CommonsAttribution 4.0 International License.)

The rectifier junction in the present device has two insulators, not just one, and this creates a deep quantum well. Electrons of a proper energy can tunnel through the two insulators with no electrical resistance.[6] Says lead author of the study, Amina Belkadi, "If you choose your materials right and get them at the right thickness, then it creates this sort of energy level where electrons see no resistance... They just go zooming through."[6] The resonant tunnel junction was constructed as Ni/NiO/Al₂O₃/Cr/Au, and it had a resistance of 13 k-ohm and a responsivity of 0.5 amps/watt.[5]

The test device was an array of about 250,000 bowtie rectennas, and it was tested using a laboratory hot plate as an infrared source.[6] This device captured less than 1% of the hot plate heat, but there's always room for improvement.[6] Says Belkadi, "If we use different materials or change our insulators, then we may be able to make that well deeper,... The deeper the well is, the more electrons can pass all the way through.[6] One application aside from harvesting waste heat is to capture Earth's radiation into the cold of outer space.[6] I wrote about another method to do this in an earlier article (Energy-Harvesting the Earth's Heat, March 10, 2014).

Optical Rectenna

References: