Tag Archives: measurements

New approach to brain-inspired (neuromorphic) computing: measuring information transfer

An April 8, 2024 news item on Nanowerk announces a new approach to neuromorphic computing that involves measurement, Note: Links have been removed,

The biological brain, especially the human brain, is a desirable computing system that consumes little energy and runs at high efficiency. To build a computing system just as good, many neuromorphic scientists focus on designing hardware components intended to mimic the elusive learning mechanism of the brain. Recently, a research team has approached the goal from a different angle, focusing on measuring information transfer instead.

Their method went through biological and simulation experiments and then proved effective in an electronic neuromorphic system. It was published in Intelligent Computing (“Information Transfer in Neuronal Circuits: From Biological Neurons to Neuromorphic Electronics”).

An April 8, 2024 Intelligent Computing news release on EurekAlert delves further into the topic,

Although electronic systems have not fully replicated the complex information transfer between synapses and neurons, the team has demonstrated that it is possible to transform biological circuits into electronic circuits while maintaining the amount of information transferred. “This represents a key step toward brain-inspired low-power artificial systems,” the authors note.

To evaluate the efficiency of information transfer, the team drew inspiration from information theory. They quantified the amount of information conveyed by synapses in single neurons, then measured the quantity using mutual information, the analysis of which reveals the relationship between input stimuli and neuron responses.

First, the team conducted experiments with biological neurons. They used brain slices from rats, recording and analyzing the biological circuits in cerebellar granule cells. Then they evaluated the information transmitted at the synapses from mossy fiber neurons to the cerebellar granule cells. The mossy fibers were periodically stimulated with electrical spikes to induce synaptic plasticity, a fundamental biological feature where the information transfer at the synapses is constantly strengthened or weakened with repeated neuronal activity.

The results show that the changes in mutual information values are largely consistent with the changes in biological information transfer induced by synaptic plasticity. The findings from simulation and electronic neuromorphic experiments mirrored the biological results.

Second, the team conducted experiments with simulated neurons. They applied a spiking neural network model, which was developed by the same research group. Spiking neural networks were inspired by the functioning of biological neurons and are considered a promising approach for achieving efficient neuromorphic computing.

In the model, four mossy fibers are connected to one cerebellar granule cell, and each connection is given a random weight, which affects the information transfer efficiency like synaptic plasticity does in biological circuits. In the experiments, the team applied eight stimulation patterns to all mossy fibers and recorded the responses to evaluate the information transfer in the artificial neural network.

Third, the team conducted experiments with electronic neurons. A setup similar to those in the biological and simulation experiments was used. A previously developed semiconductor device functioned as a neuron, and four specialized memristors functioned as synapses. The team applied 20 spike sequences to decrease resistance values, then applied another 20 to increase them. The changes in resistance values were investigated to assess the information transfer efficiency within the neuromorphic system.

In addition to verifying the quantity of information transferred in biological, simulated and electronic neurons, the team also highlighted the importance of spike timing, which as they observed is closely related to information transfer. This observation could influence the development of neuromorphic computing, given that most devices are designed with spike-frequency-based algorithms.

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

Information Transfer in Neuronal Circuits: From Biological Neurons to Neuromorphic Electronics by Daniela Gandolfi, Lorenzo Benatti, Tommaso Zanotti, Giulia M. Boiani, Albertino Bigiani, Francesco M. Puglisi, and Jonathan Mapell. Intelligent Computing 1 Feb 2024 Vol 3 Article ID: 0059 DOI: 10.34133/icomputing.0059

This paper is open access.

Metallic nanoparticles: measuring their discrete quantum states

I tend to forget how new nanotechnology is and unconsciously take for granted stunning feats such as measuring a metallic nanoparticle’s electronic properties. A June 15, 2015 news item on Nanowerk provides a reminder with its description of the difficulties and a new technique to make it easier (Note:  A link has been removed),

How do you measure the electronic properties of individual nanoparticles or molecules that are only a few nanometers in size? Conventional methods using electron transport spectroscopy rely on contacting a material with two contacts, a source and a drain electrode. By applying a small potential difference over the electrodes and monitoring the resulting current, valuable information about the electronic properties are extracted. For example if a material is metallic or semiconducting.
But this becomes quite a challenge if the material is only a few nm in size. Even the most sophisticated fabrication tools such as electron-beam lithography have a resolution of about 10 nm at best, which is not precise enough. Scientists have developed workarounds such as creating small gaps in narrow metallic wires in which a nanoparticle can be trapped if it matches the gap size. However, even though there have been some notable successes using this approach, this method has a low yield and is not very reproducible.

Now an international collaboration including researchers in Japan, the university [sic] of Cambridge and the LCN [London Centre for Nanotechnology] in the UK have approached this in a different way as described in a paper in Nature’s Scientific Reports (“Radio-frequency capacitance spectroscopy of metallic nanoparticles”). Their method only requires a single electrode to be in direct contact with a nanoparticle or molecule, thus significantly simplifying fabrication.

A June 15, 2015 (?) LCN press release, which originated the news item, describes the achievement,

The researchers demonstrated the potential of the radio-frequency reflectometry technique by measurements on Au nanoparticles of only 2.7 nm in diameter. For such small particles, the electronic spectrum is discrete which was indeed observed in the measurements and in very good agreement with theoretical models. The researchers now plan to extend these measurements to other nanoparticles and molecules with applications in a range of areas such as biomedicine, spintronics and quantum information processing.

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

Radio-frequency capacitance spectroscopy of metallic nanoparticles by James C. Frake, Shinya Kano, Chiara Ciccarelli, Jonathan Griffiths, Masanori Sakamoto,  Toshiharu Teranishi, Yutaka Majima, Charles G. Smith & Mark R. Buitelaar. Scientific RepoRts 5:10858 DOi: 10.1038/srep10858 Published June 4, 2015

This is an open access paper.

New method of Rayleigh scattering for better semiconductors

Rayliegh scattering provides a scientific explanation first devised in the 19th century for why the sky is blue during the day and why it turns red in the evening. A March 4, 2014 news item on Nanowerk describes some research into measuring semiconductor nanowires with a new Rayleigh scattering technique,

A new twist on a very old physics technique could have a profound impact on one of the most buzzed-about aspects of nanotechnology.

Researchers at the University of Cincinnati [UC] have found that their unique method of light-matter interaction analysis appears to be a good way of helping make better semiconductor nanowires.

The March 4, 2014 University of Cincinnati news release, which originated the news item, has the researcher describing his work in further detail (Note: Links have been removed),

“Semiconductor nanowires are one of the hottest topics in the nanoscience research field in the recent decade,” says Yuda Wang, a UC doctoral student. “Due to the unique geometry compared to conventional bulk semiconductors, nanowires have already shown many advantageous properties, particularly in novel applications in such fields as nanoelectronics, nanophotonics, nanobiochemistry and nanoenergy.”

Wang will present the team’s research “Transient Rayleigh Scattering Spectroscopy Measurement of Carrier Dynamics in Zincblende and Wurtzite Indium Phosphide Nanowires” at the American Physical Society (APS) meeting to be held March 3-7 [2014] in Denver. …

Key to this research is UC’s new method of Rayleigh scattering, a phenomenon first described in 1871 and the scientific explanation for why the sky is blue in the daytime and turns red at sunset. The researchers’ Rayleigh scattering technique probes the band structures and electron-hole dynamics inside a single indium phosphide nanowire, allowing them to observe the response with a time resolution in the femtosecond range – or one quadrillionth of a second.

“Basically, we can generate a live picture of how the electrons and holes are excited and slowly return to their original states, and the mechanism behind that can be analyzed and understood,” says Wang, of UC’s Department of Physics. “It’s all critical in characterizing the optical or electronic properties of a semiconducting nanowire.”

Semiconductors are at the center of modern electronics. Computers, TVs and cellphones have them. They’re made from the crystalline form of elements that have scientifically beneficial electrical conductivity properties.

Wang says the burgeoning range of semiconductor nanowire applications – such as smaller, more energy-efficient electronics – has brought rapid improvement to nanowire fabrication techniques. He says his team’s research could offer makers of nanotechnology a new and highly effective option for measuring the physics inside nanowires.

“The key to a good optimization process is an excellent feedback, or a characterization method,” Wang says. “Rayleigh scattering appears to be an exceptional way to measure several nanowire properties simultaneously in a non-invasive and high-quality manner.”

Additional contributors to this research are UC alumnus Mohammad Montazeri; UC physics professors Howard Jackson and Leigh Smith and adjunct associate professor Jan Yarrison-Rice, all of the McMicken College of Arts and Sciences; and Tim Burgess, Suriati Paiman, Hoe Tan, Qiang Gao and Chennupati Jagadish of Australian National University.

You can get more information about the American Physical Society March 3 – 7, 2014 meeting in Denver, Colorado here.

Adventures in time, mass, and topological insulators

Nano at a billionth (of a second, or a metre, or some other measure) is not the smallest unit of measurement, despite how we often talk about nano ‘anything’. But, as we continue to explore matter at ever more subtle levels, we need ever smaller units of measure and there are some ready for use.

I have a few excerpts from a Sept. 18, 2012  article (Explained: Femtoseconds and attoseconds) by David Chandler at the Massachusetts Institute of Technology (MIT) describing some of these smaller units of measure and how they were devised,

Back in the first half of the 20th century, when MIT’s famed Harold “Doc” Edgerton was perfecting his system for capturing fast-moving events on film, the ability to observe changes unfolding at a scale of microseconds — millionths of a second — was considered a remarkable achievement. This led to now-famous images such as one of a bullet piercing an apple, captured in midflight.

Nowadays, microsecond-resolution imagery is almost ho-hum. The cutting edge of research passed through nanoseconds (billionths of a second) and picoseconds (trillionths) in the 1970s and 1980s. Today, researchers can easily reach into the realm of femtoseconds — quadrillionths (or millionths of a billionth) of a second, the timescale of motions within molecules.

Femtosecond laser research led to the development, in 2000, of a system that revolutionized the measurement of optical frequencies and enabled optical clocks. Continuing the progress, today’s top-shelf technologies are beginning to make it possible to observe events that last less than 100 attoseconds, or quintillionths of a second.

Those prefixes — micro, nano, pico, femto and atto — are part of an internationally agreed-upon system called SI units (from the French Système International d’Unités, or International System of Units). The system was officially adopted in 1960, and has been updated periodically, most recently in 1991. It encompasses a total of 20 prefixes, 10 of them for decimal amounts, and 10 more for large multiples of the basic units (mega, giga, tera and so on).

As Chandler points out in more detail than I have, there’s a reason for developing these units of measure,

The ability to observe events on such timescales is important for basic physics — to understand how atoms move within molecules — as well as for engineering semiconductor devices, and for understanding basic biological processes at the molecular level.

But physicists and engineers are interested in pushing these limits ever further. To understand the movements of electrons, and eventually those of subatomic particles, requires attaining the attosecond and ultimately zeptosecond (sextillionths of a second) range, Kaertner says. Achieving that requires pushing technology to produce pulses using higher-wavelength sources, and also producing pulses that encompass a wider range of frequencies — a more broadband source.

I finally managed to conceptualize the nanoscale a few years ago but it appears I have more work to do. Chandler offers some suggestions for imagining the femtoscale,

So, just how short is a femtosecond? One way to think of it, Kaertner [Franz Kaertner, MIT adjunct professor of electrical engineering] says, is in terms of how far light can move in a given amount of time. Light travels about 300,000 kilometers (or 186,000 miles) in one second. That means it goes about 30 centimeters — about one foot — in one nanosecond. In one femtosecond, light travels just 300 nanometers — about the size of the biggest particle that can pass through a HEPA filter, and just slightly larger than the smallest bacteria.

Another way of thinking about the length of a femtosecond is this: One femtosecond is to one second as one second is to about 32 million years.

Chandler discuses in another MIT article (Watching electrons move at high speed) also posted on Sept. 18, 2012, a new electronic material, a topological insulator, and the importance of viewing the behaviour of electrons present in such an insulator,

Topological insulators are exotic materials, discovered just a few years ago, that hold great promise for new kinds of electronic devices. The unusual behavior of electrons within them has been very difficult to study, but new techniques developed by a team of researchers at MIT could help unlock the mysteries of exactly how electrons move and react in these materials, opening up new possibilities for harnessing them.

For the first time, the MIT team has managed to create three-dimensional “movies” of electron behavior in a topological insulator, or TI. [can be viewed here] The movies can capture vanishingly small increments of time — down to the level of a few femtoseconds, or millionths of a billionth of a second — so that they can catch the motions of electrons as they scatter in response to a very short pulse of light.

Electrons normally have mass, just like many other fundamental particles, but when moving along the surface of TIs they move as if they were massless, like light — one of the extraordinary characteristics that give these new materials such promise for new technologies. [emphases mine]

It’s the bit about mass and masslessness that caught my eye. Fascincating, non? Here’s a graphical representation of what the MIT scientists observed (I think it looks like a cup or a grail),

Three-dimensional graphical representations of the way electrons respond to an input of energy, delivered by a pulse of laser light. The horizontal axis represents the electrons’ momentum, and the vertical axis shows their energy. The time sequence runs from top left to bottom right, and the laser pulse arrives just before the second image, causing a sudden burst of higher energy levels. Images courtesy of Yihua Wang and Nuh Gedik [of MIT]

Here’s a bit more about TIs and possible future applications,

TIs are a class of materials with seemingly contradictory characteristics: The bulk of the material acts as an insulator, almost completely blocking any flow of electrons. But the surface of the material behaves as a very good conductor, like a metal, allowing electrons to travel freely. In fact, the surface is even more conductive than normal metals — allowing electrons to travel at almost the speed of light and to be unaffected by impurities in the material, which normally hinder their motion.

Because of these characteristics, TIs are seen as a promising new material for electronic circuits and data-storage devices. But developing such new devices requires a better understanding of exactly how electrons move around on and inside the TI, and how the surface electrons interact with those inside the material.

I highly recommend reading both of Chandler’s articles.

Measuring nanoparticles

When manufacturers claim to produce nanoparticles that are 30 nm in diameter, they are giving customers the average size of the nanoparticles being delivered. (From Nanowerk Spotlight’s Meaningful nanotechnology EHS research requires independent nanomaterial characterization)

“For example, it might be stated that a certain nanoparticle is being sold as 30 nm in diameter and, although ’30 nm’ might be close to the average diameter, there is usually a range of particle sizes that can extend from as much as small as 5 nm to as large as 300 nm.” (Vicki H. Grassian, a professor in the Department of Chemistry at the University of Iowa)

As I noted in my April 27, 2010 posting this size range could pose problems with Canada’s proposed plan/inventory/scheme. Happily, the US National Institute of Standards and Technology (NIST) has successfully measured and sorted nanoparticles with a device that operates like a coin sorter (separating pennies, dimes, nickels, and quarters). From the news item on physorg.com,

First introduced in March 2009 …, the device consists of a chamber with a cascading “staircase” of 30 nanofluidic channels ranging in depth from about 80 nanometers at the top to about 620 nanometers (slightly smaller than an average bacterium) at the bottom. Each of the many “steps” of the staircase provides another “tool” of a different size to manipulate nanoparticles in a method that is similar to how a coin sorter separates nickels, dimes and quarters.

In a new article in the journal Lab on a Chip, the NIST research team demonstrates that the device can successfully perform the first of a planned suite of nanoscale tasks—separating and measuring a mixture of spherical nanoparticles of different sizes (ranging from about 80 to 250 nanometers in diameter) dispersed in a solution.

Seems to me that is pretty exciting news. I wonder when this device will go into standard use. The usual answer to this question includes the number 5 as in 3-5 years, 5-7 years, or 5 years. In any case, the researchers are also hoping to use the technique to sort nanoparticles of different shapes as in tubes from spheres and that sort of thing.

For anyone interested in the researcher’s article the citation is,

S.M. Stavis, J. Geist and M. Gaitan. Separation and metrology of nanoparticles by nanofluidic size exclusion. Lab on a Chip, forthcoming, August 2010