Energy harvesting is all the rage at the moment. I’d like devices that convert waste energy into electricity. If I could, I would fill my house with all sorts of high-tech Rube Goldberg-esque machines for getting an extra few watts back. But Wi-Fi energy harvesting has never impressed me as a potential source of wattage, even though the latest research shows that it might well be getting somewhere.
Most houses and public spaces abound with Wi-Fi signals, but harvesting the energy in those signals is difficult. The first challenge is to collect the radiation.
Advanced bunny-ear dynamics
Collecting the radiation requires some near-impossible antennas. The antennas must be omni-directional, meaning they absorb radiation from all directions equally well. The antennas should also absorb all the radiation that hits them, which ain’t an easy task either.
To get this to happen, the spatial configuration of the antennas has to match the spatial configuration of the Wi-Fi radiation. That latter configuration is not just unknown, but it changes with time.
Nevertheless, there has been a lot of progress here. Antennas that can absorb well over 90 percent of incoming energy have been demonstrated. But these don’t operate on all Wi-Fi bands and/or don’t absorb radiation from every direction.
Assuming that these issues could be solved, the next problem is to turn the absorbed radiation into a useful source of electrons. You may have noticed that most of your low-power gadgets—your phone, tablet, and watch—run on direct current (DC). The antennas that absorb Wi-Fi radiation provide you with alternating current (AC). And, unlike the current from a wall socket, this is hard to convert to DC.
Depending on where you are in the world, grid power has a frequency of 50 or 60Hz. This is converted to DC by a network of diodes that only allow current to flow in one direction. Diodes, however, have to switch on and off, which takes time. This is where the problem lies. For a 60Hz signal, switching in a microsecond is adequate. For Wi-Fi signals, which are at 2.4 GHz and up, you probably want to switch in a picosecond (10-12s).
The diodes need to switch fast, but in practice, their switching speed depends on the size of the AC voltage. Applying 12V to a diode will cause it to switch much faster than applying a few millivolts. Yet our Wi-Fi antenna is going to be supplying millivolts at best. Most likely, the diode is going to be asked to switch microvolts or nanovolts, which drops the speed considerably. As a result, the power-conversion efficiency falls off a cliff.
The antenna might absorb 90 percent of the Wi-Fi signal, but the diode promptly loses between half and 95 percent of that, depending on the Wi-Fi signal strength.
Making faster diodes
I honestly didn’t see much prospect for improvement either, because these diodes are critical components in Wi-Fi transceivers. Engineers have been making them as efficient as possible for many years, and improvements have slowed considerably. Essentially, we’ve pushed the limits of the materials that we’re using (silicon and germanium), which determine how fast the diode can switch.
The solution, then, is to change materials. A team of researchers created diodes based on a 2D material, called molybdenum disulfide (MoS2). Molybdenum disulfide changes from a semiconductor to a conductor with a bit of chemistry.
A diode based on a combination of metallic electrodes, a tiny slice of semiconducting molybdenum disulfide and conducting molybdenum disulfide was created. This diode switches very fast, even when the applied voltage is very low. That’s because diode switching is fast if the diode cannot store charge, and a 2D material has no room to store charge.
The researchers built flexible Wi-Fi antennas that also contained the diodes and went in search of waste energy to recover. The diodes work, but at signal levels that might reasonably be expected, the efficiency was below 10 percent. That number, however, is not the end of the story. In this case, the researchers really focused on making a good diode, while the antenna design was as simple as possible. Combined with a good antenna, the researchers might double their efficiency.
The actual achievement of this research is substantial. The antenna plus diode was just a start; the researchers also demonstrated an integrated mixer (a key component in radio transceivers) built into a flexible antenna. This opens up the possibility of nice form-matching Wi-Fi systems with more efficient transceivers. I like that.
Is Wi-Fi energy harvesting going to be a thing, though? Probably not. A lot of tiny, low-power devices out there might be Wi-Fi powered. But the amount of energy harvested by a Wi-Fi antenna is directly proportional to the area of the antenna. Those tiny devices may not be so tiny once the energy harvester is added. So, my skepticism remains.