November 2003
http://www.coe.berkeley.edu/labnotes/1103
Commuter-friendly version

Professor Stanley Prussin has been on the UC Berkeley faculty since 1966.

Each year, nearly seven million shipping containers pass through US ports. With tight time constraints allowing just two percent of the containers to be inspected, there is a very real fear that one of these twenty to forty foot long containers could be a Trojan horse hiding the key ingredient in a nuclear weapon. The challenge is that detecting a baseball-sized bit of highly-enriched uranium or plutonium buried inside 27 metric tons of fruit, furniture, or computers is like looking for a needle in a haystack without knowing if the needle is even there. To detect the clandestine transport of nuclear weapons materials, UC Berkeley nuclear engineering professor Stanley G. Prussin and Eric B. Norman, a senior nuclear scientist at Lawrence Berkeley National Laboratory (LBNL), are working with scientists at the Lawrence Livermore National Laboratory (LLNL) to develop a nuclear detection method that may be 10,000 times more sensitive under some conditions than other approaches currently being tested.

A ship carrying some of the 7 million shipping containers that pass through US ports each year. (courtesy Dennis Slaughter/LLNL)

Prussin and Norman's method involves bombarding a shipping container with a beam of neutrons that will induce a safe fission reaction if uranium or plutonium is inside the box. The researchers can then detect gamma rays produced by the reaction. The idea came to Prussin and Norman while they observed the cargo container screening effort at LLNL. The LLNL scientists, Prussin explains, also employed a beam of neutrons to induce the fission reaction. If fission occurs, the LLNL researchers hoped they could count "delayed neutrons," the radiation slowly emitted by a fission reaction after the initial burst of neutrons is released.

The difficulty, Prussin says, is that neutrons are likely to get absorbed by materials containing large quantities of hydrogen before they even make it out of the shipping container.

"We import fruits and vegetables, filled with water," Prussin says. Computers? They're made of plastic, predominantly composed of carbon and hydrogen. Hydrogen is all over the place."

This testbed at LLNL enables researchers to experiment with nuclear detection methods on a real cargo container. (courtesy Dennis Slaughter/LLNL)

It occurred to Prussin and Norman that there are other delayed radiations that may be less prone to absorption in hydrogenous materials. High-energy gamma rays, he says, "can penetrate through such matter a heck of a lot better than neutrons can."

After a few simple calculations, Prussin determined that the relative intensity of delayed high-energy gamma rays is approximately 10 times larger than delayed neutrons. Gamma rays also slip through hydrogenous materials 100 to 1000 times more easily than neutrons. Norman then dove into the data and discovered that many of these gamma rays had energies well above any normal background noise that may be present during the screening process.

"We believe that under the right conditions, this method could provide the unequivocal signature of gamma radiation indicating that fission has occurred inside the container," Prussin says.

Already, Prussin and Norman have proven out their method in controlled laboratory experiments at LBNL. The next step is to determine the practicality of the idea by conducting experiments using a full-size container at Livermore that's packed with mock cargo and a sample of depleted uranium.

"I have no idea if we will be able to produce a practical system that can be deployed easily in ports with sufficient sensitivity and high enough throughput to satisfy all our needs," Prussin says. "What I can say is that I think this method has a much higher probability than anything that's been suggested so far. I'm optimistic, but it's no slam dunk."

UC Berkeley professor Arun Majumdar and colleagues are designing and building nanofluidic transistor from glass tubes just 100 nanometers in length. (Peg Skorpinski photo)

Traditional transistors are essentially valves that control the flow of electricity to perform calculations. But what if, instead of voltages, a transistor could manipulate the flow of biological molecules like proteins and DNA? Berkeley researchers have developed the world's first device that does just that. Eventually, this nanofluidic transistor could detect cancer in a drop of blood much smaller than the period at the end of this sentence.

Mechanical engineering professor Arun Majumdar and College of Chemistry professor Peidong Yang, in collaboration with their graduate students and visiting professor Hirofumi Daiguji, are designing and building the nanofluidic transistor from glass tubes just 100 nanometers in length. (A nanometer is one-billionth of a meter.)

"Every basic college textbook on fluid dynamics talks about the pressure and velocity of water flowing through a glass pipe," says Majumdar. "We wanted to know if anything unusual happens if the pipe is really small."

While the interactions that occur at the interior surface of glass tubes are of minor importance at the macroscale, the researchers quickly realized that the surface becomes the most important locale for interactions at the nanoscale. Glass tubes are naturally negatively charged, meaning that the material is coated with atoms that have lost an electron. Negatively charged tubes attract positive ions, Majumdar explains. If you fill a tube from one end with a liquid, the negatively charged ions in the liquid are pumped out leaving just the positively charged ions inside. At the macroscale, this effect isn't particularly useful. But with nanotubes of internal diameters of just five to fifty nanometers, the ability to separate the negative and positive ions could enable the glass pipe to become a key component in a nanofluidic transistor, also called a unipolar ionic field-effect transistor.

The silica nanotube in the transmission electron micrograph image has an internal diameter of about 10 nanometers. (courtesy the researchers)

The next step was to fashion a valve, the equivalent of a transistor's gate, to control the ionic current passing through the tube. The researchers placed a fine metal wire on top of the tube to act as an electrode. Applying a voltage to the wire controls the flow of the ions.

"Of course, the gate is too slow to manipulate information like a transistor, but you can push biological molecules through because things like proteins and DNA are all electrically charged," Majumdar says.

Once the researchers fine-tune their control of the ionic flow, they'll line the inside of the tube with biological molecules called antigens. The human body's immune system recognizes antigens as threats, and forms specific antibodies that bind to the foreign molecules. It's this biochemical reaction that signals the immune system to launch an attack on a disease.

To use the nanofluidic transistor as a disease detector, a tiny drop of blood will be pushed through the glass tube. If the blood contains the antibodies characteristic of a particular kind of disease, those antibodies will bind to the antigens inside the tube just as they do in the body. Those bound antibodies will partially clog the tube, blocking the current flowing through it. The resulting drop in the current flow will indicate that the blood contains signs of the disease.

Majumdar, also an investigator with the Center for Information Technology Research in the Interest of Society (CITRIS), and his colleagues are currently working to demonstrate the ability of their nanofluidic transistor to detect prostate cancer, an effort sponsored by the National Cancer Institute. The transistor's diminutive size also means that an array of the tubes could be integrated into one small device to quickly screen for a variety of diseases without the burden of expensive and bulky laboratory equipment.

"I call it picoliter biology because the volume of a human cell is on the order of one picoliter," Majumdar says. "The question at the end of the day is can we use this technology to analyze a single human cell?"

UC Berkeley researcher Mark Stacey calls the entire San Francisco Bay his "laboratory."

UC Berkeley researcher Mark Stacey calls the entire San Francisco Bay his "laboratory." The professor of civil and environmental engineering analyzes the physics of the 1,600 square mile waterway between the Pacific Ocean and the Sacramento-San Joaquin Delta. By studying the estuary, from centimeter scale turbulence to the seasonal transport of salt between the ocean and the Bay, Stacey's research could impact everything from the preservation of delicate ecosystems to the quality of our drinking water.

"To understand how everything from nutrients and salt to contaminants and invasive species move through the Bay, you have to understand the basic physics of water flow," Stacey says.

Stacey's research is focused on ocean-estuary exchanges, the flow between freshwater from upstream river systems and fluxes from the open sea. In the San Francisco Bay, Stacey explains, the relationship between the coastal ocean and the estuary is largely unknown. But through a combination of boat trips around the Bay and computer simulations, Stacey and his students are developing predictive models of how water and, more importantly, the various materials in the mix, is transported through the estuary system.

The largest harbor on the US Pacific Coast, the San Francisco Bay-Delta Estuary contains over 90 percent of the state's coastal wetlands and supports 750 species of fish, animals, and birds.

For example, one of Stacey's Bay-based research efforts is concentrated on the movement of salt in the water. The Bay's salt field, or seawater front, shifts within the Bay in response to myriad factors like tidal dynamics and freshwater inflows from rivers. The salt field, Stacey says, is partially at the mercy of the state's water managers. Opening reservoir valves upstream pushes the salt field out of the Bay while scaling down the flow from the reservoirs causes the salt to flow back into the Bay.

"Where the salt field sits during different seasons of the year has implications for the health of the Bay's ecosystem," Stacey says. " "Certain species need it to be pushed downstream to the ocean during the spring or back into the Bay during the fall."

Still, the water managers must cater to the needs of the Bay's human population as well. A vast majority of the state's population drinks at least some water from the Delta, and too much salinity in the Bay can jeopardize drinking water supplies.

"We'd like to manage our freshwater resources in a way that the position of the salt field in the Bay doesn't compromise either drinking water quality or the safety of the ecosystem," he says.

Through field studies, Mark Stacey (left) and his graduate students also hope to gain an understanding of where sediments dredged from the Bay--to clear the way for boats, for instance-- eventually settle. This data could help determine the availability of sediment necessary for marshland restoration in the South Bay.

To gather their data, Stacey and his team frequently embark on Bay cruises aboard a small boat outfitted with a variety of sensing instruments. An Acoustic Doppler Current Profiler (ADCP) converts the echoes of audio waves into a three-dimensional representation of the current. Meanwhile, other sensors immersed in the water keep track of temperature and salinity while also measuring chlorophyll concentration, indicative of what's living in the water.

The key to gathering useful data is measuring flows and currents on wide time scales, from "turbulent scales" lasting only a few seconds to 12-hour tidal scales, to lunar and annual cycles. Minor variations in the small scale dynamics can have a big impact, Stacey says. Ignoring how the various size and time scales are intertwined makes it impossible to adequately represent the physical properties in computer models and test various hypotheses about the ocean-Bay exchanges.

"The thing I like about field studies is that we're not idealizing," he says. "We capture the real physics and then go into a computer modeling stage."

Most recently, Stacey and his team began unraveling a long-standing mystery of ocean-estuary exchange. As part of the Flux Observations at the Golden Gate (FOGG) project, the researchers are not only studying how much salt and other materials move between the Bay and the ocean, but why this happens at all. The researchers compared their measurements of the flux of salt coming and out of the Bay against theoretical amounts predicted based on the flow of water from the rivers upstream.

"Measurements have never been made this way before with such a high level of detail," he says. "Finally, we can really peel back the layers of these mechanisms to determine which ones are most important.

The goal now is to identify the specific physical mechanisms--wind or tides, for example--that are creating the flux.

"There are a vast number of unanswered but interrelated questions," Stacey says. "It's very detailed work, but I get excited thinking about how it may affect ecosystem dynamics and California water policy."

Professor Rhonda Righter, an alumnus of UC Berkeley's IEOR department, returned as a faculty member in July of this year.

What do a sandwich shop, a bank and an automobile plant have in common? They're examples of places that could benefit from research in UC Berkeley's Department of Industrial Engineering and Operations Research (IEOR). IEOR professors Rhonda Righter and Hyun-soo Ahn are teasing out general "rules-of-thumb" from the esoteric mathematics of IEOR to help managers train employees and efficiently distribute a company's workload, thereby amping up workplace productivity.

Perhaps the most ubiquitous workflow approach is for a single employee to follow one task from beginning to end. While this may seem like a logical approach, Righter and Ahn's research reveals that it may not be the most efficient. One method they're investigating is known as "last-buffer, first-served" (LBFS). The idea is simple: employees should prioritize projects that are closest to being completed even it had been another employee's responsibility.

"If you have people coming and going in a company or getting sick during the course of a large project, there are often many unfinished tasks," Righter explains. "Under certain circumstances, it's better for someone else to pick up where something was left off rather than waiting for the first person to come back and finish it."

LBFS may seem like common sense, but it's often underutilized, Righter says. Take the process a bank uses to approve a loan. Typically, one employee follows a loan from start to finish. In reality though, there may be ways to dramatically speed things along by assigning various components to a number of employees. Perhaps, Righter says, certain individuals may be trained to float through the office, helping those who are bogged down at certain stages in the approval process.

Professor Hyun-soo Ahn also conducts research on supply chain and service operation management.

"We're especially interested in situations where there could be random factors," Righter says. "In the case of bank loans, the approval process could be held up because more data is needed. We'd like to figure out how to model these kinds of uncertainties to develop better policies."

To mathematically prove their rules-of-thumb theories, Righter and Ahn employ stochastic models — mathematical models that contain random variables representing random factors. Specifically, they look at so-called Markov chains, models of a sequence of events where the probability of one event depends on whether the preceding event occurred or not.

"By using these models to prove theorems, we can determine that a particular workflow policy is optimal, the best you can do in that situation," Righter says.

Righter and Ahn are also studying another workflow policy called a "bucket brigade." Named for the way old-time firefighters passed containers of water down the line to quench a blaze, here products are progressively completed as they move down the line. Once the last employee puts the finishing touches on a product, the employee walks “upstream” to take over the task of the person ahead. The trick to maximizing the efficiency of a bucket brigade, Righter explains, is to ensure that the workers are sequenced so that the fastest employee is placed at the end of the line. This results in a workflow that "tunes" itself so that the brigade produces the most products possible given the employees' varying speeds.

From order-pickers in warehouses to submarine sandwich makers, a carefully orchestrated bucket brigade can dramatically improve production, Righter says. But setting one up is not so simple. In a busy photocopy and printing shop, for example, should every person be trained in every step of completing a job, from customer service to collating? If only one person knows the design software, will that create a bottleneck so severe that it makes financial sense to train other employees in the software?

Those are the kinds of questions Righter and Ahn will tackle next. They hope to develop a set of rules to determine how flexible employees should be. Increased flexibility, Righter says, means more on-the-job training.

"It's expensive to train your employees to perform every task," Righter says. "And once they know how to do many different jobs, how do you determine who is assigned what task and in what order?"

Werner Goldsmith with the models he used over the years to study adult head injuries.

From his expert testimonies during trials related to the beating of Rodney King to his campaign urging physicians and prosecutors to bring a more scientific eye to purported cases of "shaken baby syndrome," Werner Goldsmith, UC Berkeley mechanical engineering professor emeritus, was recognized as an international authority on the mechanics of collision. He died August 23 after a brief illness. He was 79.

Goldsmith literally wrote the book on impact. His 1960 monograph, Impact: The Theory and Physical Behaviour of Colliding Solids, was the first textbook to systematize the mechanics of collision. Still the essential text in the field, Impact analyzes the mechanics of colliding solids in everything from car crashes to refinery explosions.

A registered mechanical and safety engineer for the state of California, Goldsmith was sought after as a consultant for nearly 50 years in the areas of impact, vehicle collisions, head and neck injuries, and the effectiveness of protective devices such as sports and armed forces helmets.

"He testified sometimes for the defense, sometimes for the prosecution, whatever side he thought was correct," said George Leitmann, a Professor in the Graduate School at UC Berkeley and Goldsmith's second Ph.D. student. "He was a person of very strong convictions that were very firmly held and very firmly expressed."

In 2001, Goldsmith sought to bring public attention to "shaken-baby syndrome," a common allegation in cases of infant abuse. He launched new research on the biomechanics of infant head and neck trauma to help pediatricians and prosecutors differentiate between abuse and accident.

Born in Dusseldorf, Germany, Goldsmith was the only member of his family to escape the Holocaust. He emigrated to the United States in his early teens, graduating from the University of Texas with undergraduate and master's degrees in mechanical engineering. He became a US citizen in 1945 before working for two years as an engineer at Westinghouse Electric Corporation and an instructor at the universities of Pittsburgh and Pennsylvania. In 1947, Goldsmith came to UC Berkeley where he earned his Ph.D. in mechanical engineering in just two years, simultaneously holding an appointment as instructor. By 1960, he was a full professor.

In 1995, Goldsmith received UC Berkeley's prestigious Berkeley Citation and, in 2001, he was honored by the UC Berkeley College of Engineering with its Distinguished Engineering Alumni Award. Even as he made history, he devoted a great deal of time chronicling the success of his colleagues. Goldsmith's history book of the department, Mechanical Engineering at Berkeley: The First 125 Years, was published in 1997.

"Werner was a great man, with friends all over the world," says Professor Emeritus Jerome Sackman of UC Berkeley's Department of Civil & Environmental Engineering, a friend and longtime colleague. "Post docs and graduate students came from Asia, from Europe, from all over to work with him, and many of his students are now holding leading positions in government, industry and first-class universities all over the world."

Goldsmith was working on his next scientific paper until the last week of his life. "Brain injury in infants and children," co-written with John Plunkett, M.D., is scheduled for publication this year in the American Journal of Forensic Medicine and Pathology.
Werner Goldsmith is survived by his wife, Penelope Goldsmith of Oakland; daughters, Andrea Goldsmith of Menlo Park and Remy Margarethe Goldsmith of Oakland; son, Stephen of Santa Rosa; and four grandchildren.

Donations in his memory may be made to the Berkeley Engineering Annual Fund, College of Engineering, University of California, Berkeley, 208 McLaughlin Hall (1722), Berkeley, CA 94720-1722; or Bay Area Holocaust Oral History Project, P.O. Box 1597, Burlingame, CA 94011-1597