Reactions: Questioning energy from molecular motion and talking about where we work

Letters to the editor

Energy from molecular motion?

The story “The Motion of Molecules Makes Electricity,” dated Oct. 23, 2023 (page 7), highlighted a recent paper published inAPL Materials by Luan et al. that purports to demonstrate the harvesting of thermal energy. Though thermal harvesters of various kinds are well known, all are based on a thermal gradient—tapping into the energy flow from a hot reservoir to a cold one. The new work supposedly extracts energy, in the form of electrical power, from thermal noise in the environment when all parts of the device are at the same (room) temperature. Such a feat contradicts the second law of thermodynamics, as it would result in a net decrease of entropy in the universe.

The current authors are definitely not the first to report such unlikely feats. Every few months, I encounter some work reporting similar results; invariably, the energy “produced” is minuscule—in the present case, on the picowatt level—but the authors are always optimistic about scaling up the technology. If they ever attempt to do so, one can only surmise that the attempts fail, as the researchers never publish the expected follow-up.

While it is difficult to determine the source of these authors’ error, typical sources include electrochemical currents resulting from contact between dissimilar metals (in wires and electrodes); inadvertent rectification of electrical noise within measuring instruments; electromagnetic interference (for example, radio waves), and the rectification of mechanical vibrations. When authors attempt to scale up, the effect fails to increase, or it disappears altogether (for example, when one measuring instrument is replaced with another), but the erroneous report remains forever in the body of scientific literature.

Unfortunately, reports on such “free energy” devices keep being accepted by reviewers and editors, owing to a lackluster awareness and understanding of the laws of thermodynamics. I call on reviewers, editors, and scientific communicators to exercise more care both in accepting and in highlighting such reports. At the very least, all involved should recognize the extraordinary nature of the results published and demand equally extraordinary evidence in favor of them.

Ofer Kedem
Wadsworth, Illinois

The following is a response from one of the authors of the paper referenced above:

With very small currents, the most common way to exclude noise signals is to use polarization reversal and superposition. This method has been used for more than 10 years, and we have used it in our experiments. When the positive and negative polarizations of the device are reversed, the signal is reversed. The signal can also be reversed by superposition. Nanowires fixed at one end in a liquid undergo Brownian motion and thus are able to deform. This phenomenon has been proved both in theory and experiments. If a nanomaterial is piezoelectric, its deformation can give rise to electrical polarization. The induced polarization can be collected using a Schottky heterostructure to generate an electric current. These kinds of devices are called nanogenerators, which have been proved by a huge number of experiments. We are conducting further experiments that will be published after our new patent is filed. For our ongoing work, the voltage and current of the devices have been significantly increased with the use of better piezoelectric materials and new device architectures.

Yucheng Luan
Shanghai

Where we work

Your issue “Where We Work”( Oct. 2, 2023) prompts me to recount my experience as a medicinal chemist in the design of a new laboratory for the Montreal branch of the American Merck. In a few words, it was an excellent experience. We scientists (medicinal chemists, pharmacologists, pharmacists) were asked for our input for the construction of a new research laboratory. The architects by and large incorporated our suggestions, so we were quite satisfied with the new laboratory. For the historical record, these labs were built in 1968–69.

Joe Atkinson
Toronto