Tag Archives: Lehigh University

Artificial intelligence and metaphors

This is a different approach to artificial intelligence. From a June 27, 2017 news item on ScienceDaily,

Ask Siri to find a math tutor to help you “grasp” calculus and she’s likely to respond that your request is beyond her abilities. That’s because metaphors like “grasp” are difficult for Apple’s voice-controlled personal assistant to, well, grasp.

But new UC Berkeley research suggests that Siri and other digital helpers could someday learn the algorithms that humans have used for centuries to create and understand metaphorical language.

Mapping 1,100 years of metaphoric English language, researchers at UC Berkeley and Lehigh University in Pennsylvania have detected patterns in how English speakers have added figurative word meanings to their vocabulary.

The results, published in the journal Cognitive Psychology, demonstrate how throughout history humans have used language that originally described palpable experiences such as “grasping an object” to describe more intangible concepts such as “grasping an idea.”

Unfortunately, this image is not the best quality,

Scientists have created historical maps showing the evolution of metaphoric language. (Image courtesy of Mahesh Srinivasan)

A June 27, 2017 University of California at Berkeley (or UC Berkeley) news release by Yasmin Anwar, which originated the news item,

“The use of concrete language to talk about abstract ideas may unlock mysteries about how we are able to communicate and conceptualize things we can never see or touch,” said study senior author Mahesh Srinivasan, an assistant professor of psychology at UC Berkeley. “Our results may also pave the way for future advances in artificial intelligence.”

The findings provide the first large-scale evidence that the creation of new metaphorical word meanings is systematic, researchers said. They can also inform efforts to design natural language processing systems like Siri to help them understand creativity in human language.

“Although such systems are capable of understanding many words, they are often tripped up by creative uses of words that go beyond their existing, pre-programmed vocabularies,” said study lead author Yang Xu, a postdoctoral researcher in linguistics and cognitive science at UC Berkeley.

“This work brings opportunities toward modeling metaphorical words at a broad scale, ultimately allowing the construction of artificial intelligence systems that are capable of creating and comprehending metaphorical language,” he added.

Srinivasan and Xu conducted the study with Lehigh University psychology professor Barbara Malt.

Using the Metaphor Map of English database, researchers examined more than 5,000 examples from the past millennium in which word meanings from one semantic domain, such as “water,” were extended to another semantic domain, such as “mind.”

Researchers called the original semantic domain the “source domain” and the domain that the metaphorical meaning was extended to, the “target domain.”

More than 1,400 online participants were recruited to rate semantic domains such as “water” or “mind” according to the degree to which they were related to the external world (light, plants), animate things (humans, animals), or intense emotions (excitement, fear).

These ratings were fed into computational models that the researchers had developed to predict which semantic domains had been the sources or targets of metaphorical extension.

In comparing their computational predictions against the actual historical record provided by the Metaphor Map of English, researchers found that their models correctly forecast about 75 percent of recorded metaphorical language mappings over the past millennium.

Furthermore, they found that the degree to which a domain is tied to experience in the external world, such as “grasping a rope,” was the primary predictor of how a word would take on a new metaphorical meaning such as “grasping an idea.”

For example, time and again, researchers found that words associated with textiles, digestive organs, wetness, solidity and plants were more likely to provide sources for metaphorical extension, while mental and emotional states, such as excitement, pride and fear were more likely to be the targets of metaphorical extension.

Scientists have created historical maps showing the evolution of metaphoric language. (Image courtesy of Mahesh Srinivasan)

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

Evolution of word meanings through metaphorical mapping: Systematicity over the past millennium by Yang Xu, Barbara C. Malt, Mahesh Srinivasan. Cognitive Psychology Volume 96, August 2017, Pages 41–53 DOI: https://doi.org/10.1016/j.cogpsych.2017.05.005

The early web version of this paper is behind a paywall.

For anyone interested in the ‘Metaphor Map of English’ database mentioned in the news release, you find it here on the University of Glasgow website. By the way, it also seems to be known as ‘Mapping Metaphor with the Historical Thesaurus‘.

Discovering why nanoscale gold has catalytic properties

Gold’s glitter may have inspired poets and triggered wars, but its catalytic prowess has helped make chemical reactions greener and more efficient. (Image courtesy of iStock/sbayram) [downloaded from http://www1.lehigh.edu/news/scientists-uncover-secret-gold%E2%80%99s-catalytic-powers

Gold’s glitter may have inspired poets and triggered wars, but its catalytic prowess has helped make chemical reactions greener and more efficient. (Image courtesy of iStock/sbayram) [downloaded from http://www1.lehigh.edu/news/scientists-uncover-secret-gold%E2%80%99s-catalytic-powers

A Sept. 27, 2016 news item on phys.org describes a discovery made by scientists at Lehigh University (US),

Settling a decades-long debate, new research conclusively shows that a hierarchy of active species exists in gold on iron oxide catalysis designed for low temperature carbon monoxide oxidation; Nanoparticles, sub-nanometer clusters and dispersed atoms—as well as how the material is prepared—are all important for determining catalytic activity.

A Sept. 27, 2016 Lehigh University news release by Lori Friedman, which originated the news item, provides more information about the discovery that gold nanoparticles can be used in catalysis and about the discovery of why that’s possible,

Christopher J. Kiely calls the 1982 discovery by Masatake Haruta that gold (Au) possessed a high level of catalytic activity for carbon monoxide (CO) oxidation when deposited on a metal-oxide “a remarkable turn of events in nanotechnology”—remarkable because gold had long been assumed to be inert for catalysis.

Haruta showed that gold dispersed on iron oxide effectively catalyzed the conversion of harmful carbon monoxide into more benign carbon dioxide (CO2) at room temperatures—a reaction that is critical for the construction of fire fighters’ breathing masks and for removal of CO from hydrogen feeds for fuel cells. In fact, today gold catalysts are being exploited in a major way for the greening of many important reactions in the chemical industry, because they can lead to cleaner, more efficient reactions with fewer by-products.

Haruta and Graham J. Hutchings, who co-discovered the use of gold as a catalyst for different reactions, are noted as Thompson Reuters Citation Laureates and appear annually on the ScienceWatch Nobel Prize prediction list. Their pioneering work opened up a new area of scientific inquiry and kicked off a decades-long debate about which type of supported gold species are most effective for the CO oxidation reaction.

In 2008, using electron microscopy technology that was not yet available in the 1980s and ’90 s, Hutchings, the director of the Cardiff Catalysis Institute at Cardiff University worked with Kiely, the Harold B. Chambers Senior Professor Materials Science and Engineering at Lehigh, examined the structure of supported gold at the nanoscale. One nanometer (nm) is equal to one one-billionth of a meter or about the diameter of five atoms.

Using what was then a rare piece of equipment—Lehigh’s aberration-corrected JEOL 2200 FS scanning transmission electron microscope (STEM)—the team identified the co-existence of three distinct gold species: facetted nanoparticles larger than one nanometer in size, sub-clusters containing less than 20 atoms and individual gold atoms strewn over the support. Because only the larger gold nanoparticles had previously been detected, this created debate as to which of these species were responsible for the good catalytic behavior.

Haruta, professor of applied chemistry at Tokyo Metropolitan University, Hutchings and Kiely have been working collaboratively on this problem over recent years and are now the first to demonstrate conclusively that it is not the particles or the individual atoms or the clusters which are solely responsible for the catalysis—but that they all contribute to different degrees. Their results have been published in an article in Nature Communications titled: “Population and hierarchy of active species in gold iron oxide catalysts for carbon monoxide oxidation.”

“All of the species tend to co-exist in conventionally prepared catalysts and show some level of activity,” says Kiely. “They all do something—but some less efficiently than others.”

Their research revealed the sub-nanometer clusters and 1-3nm nanoparticles to be the most efficient for catalyzing this CO oxidation reaction, while larger particles were less so and the atoms even less.  Nevertheless, Kiely cautions, all the species present need to be considered to fully explain the overall measured activity of the catalyst.

Among the team’s other key findings: the measured activity of gold on iron oxide catalysts is exquisitely dependent on exactly how the material is prepared. Very small changes in synthesis parameters  influence the relative proportion and spatial distribution of these various Au species on the support material and thus have a big impact on its overall catalytic performance.

A golden opportunity

Building on their earlier work (published in a 2008 Science article), the team sought to find a robust way to quantitatively analyze the relative population distributions of nanoparticles of various sizes, sub-nm clusters and highly dispersed atoms in a given gold on iron oxide sample. By correlating this information with catalytic performance measurements, they then hoped to determine which species distribution would be optimal to produce the most efficient catalyst, in order to utilize the precious gold component in the most cost effective way.

Ultimately, it was a catalyst synthesis problem the team faced that offered them a golden opportunity to do just that.

During the collaboration, Haruta’s and Hutchings’ teams each prepared gold on iron oxide samples in their home labs in Tokyo and Cardiff. Even though both groups nominally utilized the same ‘co-precipitation’ synthesis method, it turned out that a final heat treatment step was beneficial to the catalytic performance for one set of materials but detrimental to the other. This observation provided a fascinating scientific conundrum that detailed electron microscopy studies performed by Qian He, one of Kiely’s PhD students at the time, was key to solving. Qian He is now a University Research Fellow at Cardiff University leading their electron microscopy effort.

“In the end, there were subtle differences in the order and speed in which each group added in their ingredients while preparing the material,” explains He. “When examined under the electron microscope, it was clear that the two slightly different methods produced quite different distributions of particles, clusters and dispersed atoms on the support.”

“Very small variations in the preparation route or thermal history of the sample can alter the relative balance of supported gold nanoparticles-to-clusters-to-atoms in the material and this manifests itself in the measured catalytic activity,” adds Kiely.

The group was able to compare this set of materials and correlate the Au species distributions with catalytic performance measurements, ultimately identifying the species distribution that was associated with greater catalytic efficiency.

Now that the team has identified the catalytic activity hierarchy associated with these supported gold species, the next step, says Kiely, will be to modify the synthesis method to positively influence that distribution to optimize the catalyst performance while making the most efficient use of the precious gold metal content.

“As a next stage to this study we would like to be able to observe gold on iron oxide materials in-situ within the electron microscope while the reaction is happening,” says Kiely.

Once again, it is next generation microscopy facilities that may hold the key to fulfilling gold’s promise as a pivotal player in green technology.

Despite the link to the paper already in the news release, here’s one that includes a citation,

Identification of Active Gold Nanoclusters on Iron Oxide Supports for CO Oxidation by Andrew A. Herzing, Christopher J. Kiely, Albert F. Carley, Philip Landon, Graham J. Hutchings. Science  05 Sep 2008: Vol. 321, Issue 5894, pp. 1331-1335 DOI: 10.1126/science.1159639

This paper is currently behind a paywall but, if you can wait one year, free access can be gained if you register (for free) with Science.

Nanoavalanches in glass

An Aug. 24, 2016 news item on Nanowerk takes a rather roundabout way to describe some new findings about glass (Note: A link has been removed),

The main purpose of McLaren’s exchange study in Marburg was to learn more about a complex process involving transformations in glass that occur under intense electrical and thermal conditions. New understanding of these mechanisms could lead the way to more energy-efficient glass manufacturing, and even glass supercapacitors that leapfrog the performance of batteries now used for electric cars and solar energy.

“This technology is relevant to companies seeking the next wave of portable, reliable energy,” said Himanshu Jain, McLaren’s advisor and the T. L. Diamond Distinguished Chair in Materials Science and Engineering at Lehigh and director of its International Materials Institute for New Functionality in Glass. “A breakthrough in the use of glass for power storage could unleash a torrent of innovation in the transportation and energy sectors, and even support efforts to curb global warming.”

As part of his doctoral research, McLaren discovered that applying a direct current field across glass reduced its melting temperature. In their experiments, they placed a block of glass between a cathode and anode, and then exerted steady pressure on the glass while gradually heating it. McLaren and Jain, together with colleagues at the University of Colorado, published their discovery in Applied Physics Letters (“Electric field-induced softening of alkali silicate glasses”).

The implications for the finding were intriguing. In addition to making glass formulation viable at lower temperatures and reducing energy needs, designers using electrical current in glass manufacturing would have a tool to make precise manipulations not possible with heat alone.

“You could make a mask for the glass, for example, and apply an electrical field on a micron scale,” said Jain. “This would allow you to deform the glass with high precision, and soften it in a far more selective way than you could with heat, which gets distributed throughout the glass.”

Though McLaren and Jain had isolated the phenomenon and determined how to dial up the variables for optimal results, they did not yet fully understand the mechanisms behind it. McLaren and Jain had been following the work of Dr. Bernard Roling at the University of Marburg, who had discovered some remarkable characteristics of glass using electro-thermal poling, a technique that employs both temperature manipulation and electrical current to create a charge in normally inert glass. The process imparts useful optical and even bioactive qualities to glass.

Roling invited McLaren to spend a semester at Marburg to analyze the behavior of glass under electro-thermal poling, to see if it would reveal more about the fundamental science underlying what McLaren and Jain had observed in their Lehigh lab.

An Aug. 22, 2016 Lehigh University news release by Chris Quirk, which originated the news item, describes the latest work,

McLaren’s work in Marburg revealed a two-step process in which a thin sliver of the glass nearest the anode, called a depletion layer, becomes much more resistant to electrical current than the rest of the glass as alkali ions in the glass migrate away. This is followed by a catastrophic change in the layer, known as dielectric breakdown, which dramatically increases its conductivity. McLaren likens the process of dielectric breakdown to a high-speed avalanche, and uses spectroscopic analysis with electro-thermal poling as a way to see what is happening in slow motion.

“The results in Germany gave us a very good model for what is going on in the electric field-induced softening that we did here. It told us about the start conditions for where dielectric breakdown can begin,” said McLaren.

“Charlie’s work in Marburg has helped us see the kinetics of the process,” Jain said. “We could see it happening abruptly in our experiments here at Lehigh, but we now have a way to separate out what occurs specifically with the depletion layer.”

“The Marburg trip was incredibly useful professionally and enlightening personally,” said McLaren. “Scientifically, it’s always good to see your work from another vantage point, and see how other research groups interpret data or perform experiments. The group in Marburg was extremely hard-working, which I loved, and they were very supportive of each other. If someone submitted a paper, the whole group would have a barbecue to celebrate, and they always gave each other feedback on their work. Sometimes it was brutally honest––they didn’t hold back––but they were things you needed to hear.”

“Working in Marburg also showed me how to interact with a completely different group of people. “You see differences in your own culture best when you have the chance to see other cultures close up. It’s always a fresh perspective.”

Here are links and citations for both the papers mentioned. The first link is for the most recent paper and second link is for the earlier work,

Depletion Layer Formation in Alkali Silicate Glasses by
Electro-Thermal Poling by C. McLaren, M. Balabajew, M. Gellert, B. Roling, and H. Jain. Journal of The Electrochemical Society, 163 (9) H809-H817 (2016) H809 DOI: 10.1149/2.0881609jes Published July 19, 2016

Electric field-induced softening of alkali silicate glasses by C. McLaren, W. Heffner, R. Tessarollo, R. Raj, and H. Jain. Appl. Phys. Lett. 107, 184101 (2015); http://dx.doi.org/10.1063/1.4934945 Published online 03 November 2015

The most recent paper (first link) appears to be open access; the earlier paper (second link) is behind a paywall.

Rainbows, what are we going to do with them?

The title is attention-getting initially then quickly leads to confusion for anyone not familiar with plasmonics, “Trapping a rainbow: Researchers slow broadband light waves with plasmonic structures.” I have to confess to being more interested in the use of the metaphor than I am in the science. However in deference to any readers who are more taken by the science, here’s more from the March 14, 2011 news item on Nanowerk,

A team of electrical engineers and chemists at Lehigh University have experimentally verified the “rainbow” trapping effect, demonstrating that plasmonic structures can slow down light waves over a broad range of wavelengths.

The idea that a rainbow of broadband light could be slowed down or stopped using plasmonic structures has only recently been predicted in theoretical studies of metamaterials. The Lehigh experiment employed focused ion beams to mill a series of increasingly deeper, nanosized grooves into a thin sheet of silver. By focusing light along this plasmonic structure, this series of grooves or nano-gratings slowed each wavelength of optical light, essentially capturing each individual color of the visible spectrum at different points along the grating. The findings hold promise for improved data storage, optical data processing, solar cells, bio sensors and other technologies.

While the notion of slowing light or trapping a rainbow sounds like ad speak, finding practical ways to control photons—the particles that makes up light— could significantly improve the capacity of data storage systems and speed the processing of optical data.

The research required the ability to engineer a metallic surface to produce nanoscale periodic gratings with varying groove depths. This alters the optical properties of the nanopatterned metallic surface, called Surface Dispersion Engineering. The broadband surface light waves are then trapped along this plasmonic metallic surface with each wavelength trapped at a different groove depth, resulting in a trapped rainbow of light.

You can get still more scientific detail in the item but I found a later posting, April 12, 2011 news item, also on Nanowerk, where the researcher Qiaoquiang Gan (pronounced “Chow-Chung” and “Gone”) gave this description for his work,

An electrical engineer at the University at Buffalo, who previously demonstrated experimentally the “rainbow trapping effect” [emphasis mine] — a phenomenon that could boost optical data storage and communications — is now working to capture all the colors of the rainbow.

In a paper published March 29 in the Proceedings of the National Academy of Sciences, Qiaoquiang Gan (pronounced “Chow-Chung” and “Gone”), PhD, an assistant professor of electrical engineering at the University at Buffalo’s School of Engineering and Applied Sciences, and his colleagues at Lehigh University, where he was a graduate student, described how they slowed broadband light waves using a type of material called nanoplasmonic structures.

Gan explains that the ultimate goal is to achieve a breakthrough in optical communications called multiplexed, multiwavelength communications, where optical data can potentially be tamed at different wavelengths, thus greatly increasing processing and transmission capacity.

“Light is usually very fast, but the structures I created can slow broadband light significantly,” says Gan. “It’s as though I can hold [emphasis mine] the light in my hand.”

I like the notion of ‘holding’ a rainbow better than ‘trapping’ one. (ETA April 18, 2011: The original sentence, now placed at the end of this posting, has been replaced with this: There’s a big difference between the two verbs, trapping and holding and each implies a difference relationship to the object. Which would you prefer, to be trapped or to be held? What does it mean to the one who does the trapping or the holding? Two difference relationships to the object and to the role of a scientist are implied.

It’s believed that the metaphors we use when describing science have a powerful impact on how science is viewed and practiced. One example I have at hand is a study by Kevin Dunbar mentioned in my Jan. 4, 2010 posting (scroll down) where he illustrates how scientists use metaphors to achieve scientific breakthroughs. Logically, if metaphors help us achieve breakthroughs, then they are quite capable of constraining us as well.

Meanwhile, this gives me an excuse to include this video of a Hawaiian singer, Israel Kamakawiwo’ole and his extraordinary version of Somewhere over the Rainbow. Happy Weekend!

The original (April 15, 2011) sentence:
It’s more gentle and implies a more humble attitude and I suspect it would ultimately prove more fruitful.