Tag Archives: Li-Dong Zhao

Hacking the human brain with a junction-based artificial synaptic device

Earlier today I published a piece featuring Dr. Wei Lu’s work on memristors and the movement to create an artificial brain (my June 28, 2017 posting: Dr. Wei Lu and bio-inspired ‘memristor’ chips). For this posting I’m featuring a non-memristor (if I’ve properly understood the technology) type of artificial synapse. From a June 28, 2017 news item on Nanowerk,

One of the greatest challenges facing artificial intelligence development is understanding the human brain and figuring out how to mimic it.

Now, one group reports in ACS Nano (“Emulating Bilingual Synaptic Response Using a Junction-Based Artificial Synaptic Device”) that they have developed an artificial synapse capable of simulating a fundamental function of our nervous system — the release of inhibitory and stimulatory signals from the same “pre-synaptic” terminal.

Unfortunately, the American Chemical Society news release on EurekAlert, which originated the news item, doesn’t provide too much more detail,

The human nervous system is made up of over 100 trillion synapses, structures that allow neurons to pass electrical and chemical signals to one another. In mammals, these synapses can initiate and inhibit biological messages. Many synapses just relay one type of signal, whereas others can convey both types simultaneously or can switch between the two. To develop artificial intelligence systems that better mimic human learning, cognition and image recognition, researchers are imitating synapses in the lab with electronic components. Most current artificial synapses, however, are only capable of delivering one type of signal. So, Han Wang, Jing Guo and colleagues sought to create an artificial synapse that can reconfigurably send stimulatory and inhibitory signals.

The researchers developed a synaptic device that can reconfigure itself based on voltages applied at the input terminal of the device. A junction made of black phosphorus and tin selenide enables switching between the excitatory and inhibitory signals. This new device is flexible and versatile, which is highly desirable in artificial neural networks. In addition, the artificial synapses may simplify the design and functions of nervous system simulations.

Here’s how I concluded that this is not a memristor-type device (from the paper [first paragraph, final sentence]; a link and citation will follow; Note: Links have been removed)),

The conventional memristor-type [emphasis mine](14-20) and transistor-type(21-25) artificial synapses can realize synaptic functions in a single semiconductor device but lacks the ability [emphasis mine] to dynamically reconfigure between excitatory and inhibitory responses without the addition of a modulating terminal.

Here’s a link to and a citation for the paper,

Emulating Bilingual Synaptic Response Using a Junction-Based Artificial Synaptic Device by
He Tian, Xi Cao, Yujun Xie, Xiaodong Yan, Andrew Kostelec, Don DiMarzio, Cheng Chang, Li-Dong Zhao, Wei Wu, Jesse Tice, Judy J. Cha, Jing Guo, and Han Wang. ACS Nano, Article ASAP DOI: 10.1021/acsnano.7b03033 Publication Date (Web): June 28, 2017

Copyright © 2017 American Chemical Society

This paper is behind a paywall.

The reason the findings in a popular thermoelectricity paper can’t be replicated

It seems to me that over the last few years there’s been a lot more discussion about errors in science. There have always been scandals but this public interest in reproducibility of scientific results seems relatively new. In any event, a Nov. 17, 2016 news item on Nanowerk highlights research that explains why scientists have been unable to reproduce results of an influential 2014 paper (Note: A link has been removed),

A team of physicists in Clemson University’s College of Science and Academia Sinica in Taiwan has determined why other scientists have been unable to replicate a highly influential thermoelectricity study published in a prestigious, peer-reviewed journal.

In the April 2014 issue of the journal Nature (“Ultralow thermal conductivity and high thermoelectric figure of merit in SnSe crystals”), a group of scientists described an emerging crystalline material made of tin selenide that provided the highest efficiency ever recorded for thermoelectricity, the process of capturing wasted energy which is released as heat and making it available again as electricity. The paper has been viewed 45,000 times and its findings have been referenced in 600 subsequent studies, according to Google Scholar.

A thermoelectricity (TE) module captures waste energy, released as heat, converts it to electricity and returns it to a device. Image Credit: Thomas Masservy, Clemson University

There appears to have been a mistake in the original research. A Nov. 17, 2016 Clemson University news release, which originated the news item, expands on the theme (Note: A link has been removed),

“If it were true, basically, they would have found a crown jewel,” said Apparao Rao, the Robert A. Bowen professor of Physics and the director of the Clemson Nanomaterials Institute.

On Nov. 3, 2016, Nature ran a brief communication by the Clemson-Sinica team explaining why the 2014 data could not be replicated.

Thermoelectricity could provide enormous monetary and environmental savings because it is sustainable; instead of requiring fuel it continually captures wasted heat energy and puts it to use. And there’s a lot of wasted energy; about 70 percent in most machines, including cars.

“When your laptop gets hot, energy is released as waste heat because it doesn’t use all the supplied electricity. Machines have limited efficiency,” according to Ramakrishna Podila, assistant professor of physics and astronomy at Clemson who co-authored the paper solving the mystery.

But, so far, the perfect material for capturing and creating thermoelectricity has proven elusive.

Heat and electrical current can flow through any material when heat is applied to one side. But to efficiently harness thermoelectricity, the material has to trap heat on one side while letting the current flow. The difference in temperature, from one side to the other, generates energy.

Imagine cookware. Expensive pots and pans are copper or they have copper cores. Copper is a great heat-conducting material: it quickly and evenly spreads heat so food cooks evenly. Copper makes for good cookware, but poor thermoelectric material.

In an ideal thermoelectric material, current-carrying electrons should flow unimpeded from the hot side to the cold side, but heat-carrying phonons, which are atomic vibrations, must be blocked, either by large atoms or defects where the material is of lower density.

Rao; Podila; Sriparna Bhattacharya, a research assistant professor in astronomy and physics; and Jian He, an associate professor in physics and astronomy at Clemson and a thermoelectrics expert, performed their own study on tin selenide in collaboration with Academia Sinica’s Institute of Physics in Taipei.

Right away, Bhattacharya noticed a problem. “The most puzzling thing was that when we measured our own tin-selenide material, we observed the same electrical flow as reported in the 2014 article, but the heat carried by the phonons was relatively higher,” Bhattacharya said.

The original research group “made tin-selenide crystal that was not fully dense,” Bhattacharya said. Ideally, a crystalline material matches its “theoretical density,” meaning it’s as dense as it can be expected to get.

“Instead of reaching 100 percent theoretical density, it reached 89 percent. A 10 percent difference might not seem like much,” she said, but it can have a huge implication on the electron and phonon flow.

The Clemson-Taiwan collaborators are now focusing on their own assessment of thermoelectricity in tin-selenide. They expect to publish soon.

Here’s a link to and a citation to the 2014 thermoelectricity paper and a link to and a citation for the 2016 paper critiquing it,

Ultralow thermal conductivity and high thermoelectric figure of merit in SnSe crystals by Li-Dong Zhao, Shih-Han Lo, Yongsheng Zhang, Hui Sun, Gangjian Tan, Ctirad Uher, C. Wolverton, Vinayak P. Dravid, & Mercouri G. Kanatzidis. Nature 508, 373–377 (17 April 2014) doi:10.1038/nature13184 Published online 16 April 2014

The intrinsic thermal conductivity of SnSe by Pai-Chun Wei, S. Bhattacharya, J. He, S. Neeleshwar, R. Podila, Y. Y. Chen, & A. M. Rao. Nature 539, E1–E2 (03 November 2016) doi:10.1038/nature19832 Published online 02 November 2016

Both papers are behind a paywall.

One final observation, scientists may mistakes as do we all. The point after all is to contribute and the mistakes can be as useful as the successes.