Tag Archives: Riccardo Comin

Organismic learning—learning to forget

Leave a reply

This approach to mimicking the human brain differs from the memristor. (You can find several pieces about memrisors here including this August 24, 2017 post about a derivative, a neuristor).  This approach comes from scientists at Purdue University and employs a quantum material. From an Aug. 15, 2017 news item on phys.org,

A new computing technology called “organismoids” mimics some aspects of human thought by learning how to forget unimportant memories while retaining more vital ones.

“The human brain is capable of continuous lifelong learning,” said Kaushik Roy, Purdue University’s Edward G. Tiedemann Jr. Distinguished Professor of Electrical and Computer Engineering. “And it does this partially by forgetting some information that is not critical. I learn slowly, but I keep forgetting other things along the way, so there is a graceful degradation in my accuracy of detecting things that are old. What we are trying to do is mimic that behavior of the brain to a certain extent, to create computers that not only learn new information but that also learn what to forget.”

The work was performed by researchers at Purdue, Rutgers University, the Massachusetts Institute of Technology, Brookhaven National Laboratory and Argonne National Laboratory.

Central to the research is a ceramic “quantum material” called samarium nickelate, which was used to create devices called organismoids, said Shriram Ramanathan, a Purdue professor of materials engineering.

A video describing the work has been produced,

An August 14, 2017 Purdue University news release by Emil Venere, which originated the news item,  details the work,

“These devices possess certain characteristics of living beings and enable us to advance new learning algorithms that mimic some aspects of the human brain,” Roy said. “The results have far reaching implications for the fields of quantum materials as well as brain-inspired computing.”

…

When exposed to hydrogen gas, the material undergoes a massive resistance change, as its crystal lattice is “doped” by hydrogen atoms. The material is said to breathe, expanding when hydrogen is added and contracting when the hydrogen is removed.

“The main thing about the material is that when this breathes in hydrogen there is a spectacular quantum mechanical effect that allows the resistance to change by orders of magnitude,” Ramanathan said. “This is very unusual, and the effect is reversible because this dopant can be weakly attached to the lattice, so if you remove the hydrogen from the environment you can change the electrical resistance.”

…

When hydrogen is exposed to the material, it splits into a proton and an electron, and the electron attaches to the nickel, temporarily causing the material to become an insulator.

“Then, when the hydrogen comes out, this material becomes conducting again,” Ramanathan said. “What we show in this paper is the extent of conduction and insulation can be very carefully tuned.”

This changing conductance and the “decay of that conductance over time” is similar to a key animal behavior called habituation.

“Many animals, even organisms that don’t have a brain, possess this fundamental survival skill,” Roy said. “And that’s why we call this organismic behavior. If I see certain information on a regular basis, I get habituated, retaining memory of it. But if I haven’t seen such information over a long time, then it slowly starts decaying. So, the behavior of conductance going up and down in exponential fashion can be used to create a new computing model that will incrementally learn and at same time forget things in a proper way.”

The researchers have developed a “neural learning model” they have termed adaptive synaptic plasticity.

“This could be really important because it’s one of the first examples of using quantum materials directly for solving a major problem in neural learning,” Ramanathan said.

The researchers used the organismoids to implement the new model for synaptic plasticity.

“Using this effect we are able to model something that is a real problem in neuromorphic computing,” Roy said. “For example, if I have learned your facial features I can still go out and learn someone else’s features without really forgetting yours. However, this is difficult for computing models to do. When learning your features, they can forget the features of the original person, a problem called catastrophic forgetting.”

Neuromorphic computing is not intended to replace conventional general-purpose computer hardware, based on complementary metal-oxide-semiconductor transistors, or CMOS. Instead, it is expected to work in conjunction with CMOS-based computing. Whereas CMOS technology is especially adept at performing complex mathematical computations, neuromorphic computing might be able to perform roles such as facial recognition, reasoning and human-like decision making.

Roy’s team performed the research work on the plasticity model, and other collaborators concentrated on the physics of how to explain the process of doping-driven change in conductance central to the paper. The multidisciplinary team includes experts in materials, electrical engineering, physics, and algorithms.

“It’s not often that a materials science person can talk to a circuits person like professor Roy and come up with something meaningful,” Ramanathan said.

Organismoids might have applications in the emerging field of spintronics. Conventional computers use the presence and absence of an electric charge to represent ones and zeroes in a binary code needed to carry out computations. Spintronics, however, uses the “spin state” of electrons to represent ones and zeros.

It could bring circuits that resemble biological neurons and synapses in a compact design not possible with CMOS circuits. Whereas it would take many CMOS devices to mimic a neuron or synapse, it might take only a single spintronic device.

In future work, the researchers may demonstrate how to achieve habituation in an integrated circuit instead of exposing the material to hydrogen gas.

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

Habituation based synaptic plasticity and organismic learning in a quantum perovskite by Fan Zuo, Priyadarshini Panda, Michele Kotiuga, Jiarui Li, Mingu Kang, Claudio Mazzoli, Hua Zhou, Andi Barbour, Stuart Wilkins, Badri Narayanan, Mathew Cherukara, Zhen Zhang, Subramanian K. R. S. Sankaranarayanan, Riccardo Comin, Karin M. Rabe, Kaushik Roy, & Shriram Ramanathan. Nature Communications 8, Article number: 240 (2017) doi:10.1038/s41467-017-00248-6 Published online: 14 August 2017

This paper is open access.

Split some water molecules and save solar and wind (energy) for a future day

Leave a reply

Professor Ted Sargent’s research team at the University of Toronto has a developed a new technique for saving the energy harvested by sun and wind farms according to a March 28, 2016 news item on Nanotechnology Now,

We can’t control when the wind blows and when the sun shines, so finding efficient ways to store energy from alternative sources remains an urgent research problem. Now, a group of researchers led by Professor Ted Sargent at the University of Toronto’s Faculty of Applied Science & Engineering may have a solution inspired by nature.

The team has designed the most efficient catalyst for storing energy in chemical form, by splitting water into hydrogen and oxygen, just like plants do during photosynthesis. Oxygen is released harmlessly into the atmosphere, and hydrogen, as H2, can be converted back into energy using hydrogen fuel cells.

Discovering a better way of storing energy from solar and wind farms is “one of the grand challenges in this field,” Ted Sargent says (photo above by Megan Rosenbloom via flickr) Courtesy: University of Toronto

Discovering a better way of storing energy from solar and wind farms is “one of the grand challenges in this field,” Ted Sargent says (photo above by Megan Rosenbloom via flickr) Courtesy: University of Toronto

A March 24, 2016 University of Toronto news release by Marit Mitchell, which originated the news item, expands on the theme,

“Today on a solar farm or a wind farm, storage is typically provided with batteries. But batteries are expensive, and can typically only store a fixed amount of energy,” says Sargent. “That’s why discovering a more efficient and highly scalable means of storing energy generated by renewables is one of the grand challenges in this field.”

You may have seen the popular high-school science demonstration where the teacher splits water into its component elements, hydrogen and oxygen, by running electricity through it. Today this requires so much electrical input that it’s impractical to store energy this way — too great proportion of the energy generated is lost in the process of storing it.

This new catalyst facilitates the oxygen-evolution portion of the chemical reaction, making the conversion from H2O into O2 and H2 more energy-efficient than ever before. The intrinsic efficiency of the new catalyst material is over three times more efficient than the best state-of-the-art catalyst.

Details are offered in the news release,

The new catalyst is made of abundant and low-cost metals tungsten, iron and cobalt, which are much less expensive than state-of-the-art catalysts based on precious metals. It showed no signs of degradation over more than 500 hours of continuous activity, unlike other efficient but short-lived catalysts. …

“With the aid of theoretical predictions, we became convinced that including tungsten could lead to a better oxygen-evolving catalyst. Unfortunately, prior work did not show how to mix tungsten homogeneously with the active metals such as iron and cobalt,” says one of the study’s lead authors, Dr. Bo Zhang … .

“We invented a new way to distribute the catalyst homogenously in a gel, and as a result built a device that works incredibly efficiently and robustly.”

This research united engineers, chemists, materials scientists, mathematicians, physicists, and computer scientists across three countries. A chief partner in this joint theoretical-experimental studies was a leading team of theorists at Stanford University and SLAC National Accelerator Laboratory under the leadership of Dr. Aleksandra Vojvodic. The international collaboration included researchers at East China University of Science & Technology, Tianjin University, Brookhaven National Laboratory, Canadian Light Source and the Beijing Synchrotron Radiation Facility.

“The team developed a new materials synthesis strategy to mix multiple metals homogeneously — thereby overcoming the propensity of multi-metal mixtures to separate into distinct phases,” said Jeffrey C. Grossman, the Morton and Claire Goulder and Family Professor in Environmental Systems at Massachusetts Institute of Technology. “This work impressively highlights the power of tightly coupled computational materials science with advanced experimental techniques, and sets a high bar for such a combined approach. It opens new avenues to speed progress in efficient materials for energy conversion and storage.”

“This work demonstrates the utility of using theory to guide the development of improved water-oxidation catalysts for further advances in the field of solar fuels,” said Gary Brudvig, a professor in the Department of Chemistry at Yale University and director of the Yale Energy Sciences Institute.

“The intensive research by the Sargent group in the University of Toronto led to the discovery of oxy-hydroxide materials that exhibit electrochemically induced oxygen evolution at the lowest overpotential and show no degradation,” said University Professor Gabor A. Somorjai of the University of California, Berkeley, a leader in this field. “The authors should be complimented on the combined experimental and theoretical studies that led to this very important finding.”

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

Homogeneously dispersed, multimetal oxygen-evolving catalysts by Bo Zhang, Xueli Zheng, Oleksandr Voznyy, Riccardo Comin, Michal Bajdich, Max García-Melchor, Lili Han, Jixian Xu, Min Liu, Lirong Zheng, F. Pelayo García de Arquer, Cao Thang Dinh, Fengjia Fan, Mingjian Yuan, Emre Yassitepe, Ning Chen, Tom Regier, Pengfei Liu, Yuhang Li, Phil De Luna, Alyf Janmohamed, Huolin L. Xin, Huagui Yang, Aleksandra Vojvodic, Edward H. Sargent. Science  24 Mar 2016: DOI: 10.1126/science.aaf1525

This paper is behind a paywall.

University of Toronto researchers combine 2 different materials for new hyper-efficient, light-emitting, hybrid crystal

1 Reply

The Sargent Group at the University of Toronto has been quite active with regard to LEDs (light-emitting diodes) and with quantum dots. Their latest work is announced in a July 16, 2015 news item on Nanotechnology Now (Note: I had to include the ‘oatmeal cookie and chocolate chips’ analogy in the first paragraph as it’s referred to subsequently),

It’s snack time: you have a plain oatmeal cookie, and a pile of chocolate chips. Both are delicious on their own, but if you can find a way to combine them smoothly, you get the best of both worlds.

Researchers in The Edward S. Rogers Sr. Department of Electrical & Computer Engineering [University of Toronto] used this insight to invent something totally new: they’ve combined two promising solar cell materials together for the first time, creating a new platform for LED technology.

The team designed a way to embed strongly luminescent nanoparticles called colloidal quantum dots (the chocolate chips) into perovskite (the oatmeal cookie). Perovskites are a family of materials that can be easily manufactured from solution, and that allow electrons to move swiftly through them with minimal loss or capture by defects.

A July 15, 2015 University of Toronto news release (also on EurekAlert), which originated the news item, reveals more about the research (Note: A link has been removed),

“It’s a pretty novel idea to blend together these two optoelectronic materials, both of which are gaining a lot of traction,” says Xiwen Gong, one of the study’s lead authors and a PhD candidate working with Professor Ted Sargent. “We wanted to take advantage of the benefits of both by combining them seamlessly in a solid-state matrix.”

The result is a black crystal that relies on the perovskite matrix to ‘funnel’ electrons into the quantum dots, which are extremely efficient at converting electricity to light. Hyper-efficient LED technologies could enable applications from the visible-light LED bulbs in every home, to new displays, to gesture recognition using near-infrared wavelengths.

“When you try to jam two different crystals together, they often form separate phases without blending smoothly into each other,” says Dr. Riccardo Comin, a post-doctoral fellow in the Sargent Group. “We had to design a new strategy to convince these two components to forget about their differences and to rather intermix into forming a unique crystalline entity.”

The main challenge was making the orientation of the two crystal structures line up, called heteroexpitaxy. To achieve heteroepitaxy, Gong, Comin and their team engineered a way to connect the atomic ‘ends’ of the two crystalline structures so that they aligned smoothly, without defects forming at the seams. “We started by building a nano-scale scaffolding ‘shell’ around the quantum dots in solution, then grew the perovskite crystal around that shell so the two faces aligned,” explained coauthor Dr. Zhijun Ning, who contributed to the work while a post-doctoral fellow at UofT and is now a faculty member at ShanghaiTech.

The resulting heterogeneous material is the basis for a new family of highly energy-efficient near-infrared LEDs. Infrared LEDs can be harnessed for improved night-vision technology, to better biomedical imaging, to high-speed telecommunications.

Combining the two materials in this way also solves the problem of self-absorption, which occurs when a substance partly re-absorbs the same spectrum of energy that it emits, with a net efficiency loss. “These dots in perovskite don’t suffer reabsorption, because the emission of the dots doesn’t overlap with the absorption spectrum of the perovskite,” explains Comin.

Gong, Comin and the team deliberately designed their material to be compatible with solution-processing, so it could be readily integrated with the most inexpensive and commercially practical ways of manufacturing solar film and devices. Their next step is to build and test the hardware to capitalize on the concept they have proven with this work.

“We’re going to build the LED device and try to beat the record power efficiency reported in the literature,” says Gong.

I see that Sargent’s work is still associated with and supported by Saudi Arabia, from the news release,

This work was supported by the Ontario Research Fund Research Excellence Program, the Natural Sciences and Engineering Research Council of Canada (NSERC), and the King Abdullah University of Science & Technology (KAUST).

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

Quantum-dot-in-perovskite solids by Zhijun Ning, Xiwen Gong, Riccardo Comin, Grant Walters, Fengjia Fan, Oleksandr Voznyy, Emre Yassitepe, Andrei Buin, Sjoerd Hoogland, & Edward H. Sargent. Nature 523, 324–328 (16 July 2015) doi:10.1038/nature14563 Published online 15 July 2015

This paper is behind a paywall.

Finally, the researchers have made a .gif of their hybrid crystal available.

A glowing quantum dot seamlessly integrated into a perovskite crystal matrix (Image: Ella Marushchenko). Courtesy: University of Toronto

A glowing quantum dot seamlessly integrated into a perovskite crystal matrix (Image: Ella Marushchenko). Courtesy: University of Toronto

ETA July 17, 2015:

Dexter Johnson provides some additional insight into the work in his July 16, 2015 posting on the Nanoclast blog (on the Institute for Electrical and Electronics Engineers website), Note:  Links have been removed,

Ted Sargent at the University of Toronto has built a reputation over the years as being a prominent advocate for the use of quantum dots in photovoltaics. Sargent has even penned a piece for IEEE Spectrum covering the topic, and this blog has covered his record breaking efforts at boosting the conversion efficiency of quantum dot-based photovoltaics a few times.

Earlier this year, however, Sargent started to take an interest in the hot material that has the photovoltaics community buzzing: perovskite. …