Stanford Engineering: 2013-2014 Year in Review by Stanford School of Engineering

the year in review 2013-2014

stanfor d Engineer ing

Building on Success With a new dean and state-of-the-art facilities, Stanford Engineering is tackling the worldâ&#x20AC;&#x2122;s toughest problems and educating students who will change the world.

s ta n f o r d e n g i n e e r i n g â&#x20AC;&#x201A;

Letter from the Dean Publisher Persis S. Drell, Dean Editor-in-Chief Laura Breyfogle Executive Editor Jamie Beckett Editor Tom Abate Managing Editor/Creative Director Rick Nobles Contributing Writers Tom Abate Amy Adams Jamie Beckett Glennda Chui Glen Martin Andrew Myers Design Rebecca Hall Lucero Copy Editor Heidi Beck School of Engineering Leadership Persis S. Drell Dean

tanford Engineering today faces an interesting challenge. We are successful in many dimensions— tremendously so. Our undergraduate enrollment is soaring. We are internationally known for our hands-on approach to learning. The accomplishments of our engineering faculty and alumni are legendary. Our current success is our challenge going forward. The great engineering school of the future certainly looks different from the Stanford School of Engineering today, and change will be essential as we evolve to our future. But from our current position of success the path forward is not always obvious. That is a challenge I embrace as your new dean. Fortunately, the school has a culture of being incredibly forward-looking and is eager to meet this challenge. Since President Hennessy named me the school’s ninth dean, I have met one-on-one with nearly all of the school’s 260 faculty members. This has been an extraordinary experience for me. And I can say that when it comes to looking to the future, our faculty are the most open-minded academics I have ever met. As I’ve talked with the faculty, two things really struck me. The first is the quality of the young faculty. Their broad

vision, civic-mindedness and raw brainpower are a bellwether for our future. They are the future of the school. The second is how committed the faculty are to finding solutions to some of the world’s most pressing problems. They might be working on something that will have an impact next year, or they might be doing the research needed to solve societal problems several decades from now. But an acceptance of our field’s responsibility to make the world a better place both now and in the future is at the heart of the school. Much of the credit for the school’s current strength goes to Jim Plummer, who stepped down as dean in August after 15 years of service. Under his leadership, the school has seen a renewal of its physical infrastructure, including completion of the Science and Engineering Quad. There has been a remarkable renewal of faculty as well—more than 50 percent of current faculty were hired during Jim’s tenure. Looking ahead, I am confident that with the help of our faculty, students and friends of the school, we will not only keep Stanford Engineering vibrantly successful but also push the boundaries of research, teaching and technology as we continue to change the world.

Jennifer Widom Sr. Assoc. Dean, Faculty and Academic Affairs Brad Osgood Sr. Assoc. Dean, Student Affairs Bernd Girod Sr. Assoc. Dean, Online Learning and Professional Development Laura Breyfogle Sr. Assoc. Dean, External Relations Scott Calvert Sr. Assoc. Dean, Finance and Administration Department Chairs Charbel Farhat Aeronautics and Astronautics Norbert Pelc Bioengineering Eric Shaqfeh Chemical Engineering Stephen Monismith Civil and Environmental Engineering Alex Aiken Computer Science Abbas El Gamal Electrical Engineering Peter Glynn Management Science and Engineering Paul McIntyre Materials Science and Engineering Kenneth E. Goodson Mechanical Engineering Margot Gerritsen Director, Institute for Computational and Mathematical Engineering

Table of Contents 2 E l e c t r o n i c f r o n t i e r s

The next generation of computing; cutting costs in the cloud; why photos go viral; and software that analyzes sentence sentiment.

6 E n g i n ee r i n g & h e a lt h

Wireless medical implants; a universal flu vaccine; a circuit board modeled on the human brain; a molecular stethoscope; better control of prosthetics; and fighting disease and bioterrorism.

12 M ee t P e r s i s D r e l l

School of Engineeringâ&#x20AC;&#x2122;s new dean discusses her life, her perspective on the school and what brought her to Stanford Engineering.

18 e n g i n ee r i n g f e at s

Using lasers to switch nerve cells on and off; the next generation of aircraft safety; and improving the structural integrity of carbon nanotubes.

22 s u s ta i n a b i l i t y

Top left: Norbert von der Groeben; Top middle: Courtesy of Austin Yee; Top right: Courtesy of the Deisseroth Lab; Bottom: Joel Simon

Microbes that convert sewage to energy; new water-recovery technology; engineering more efficient fuel cells; and improving battery performance and durability.

16 Science and Engineering Quad

28 Faculty Awards & Honors 30 Stanford Engineering Heroes 32 Financial Information and Student Statistics

16 s ta n f o r d e n g i n e e r i n g â&#x20AC;&#x201A;

Electronic frontiers

Stanford engineers overcome serious obstacles to using carbon nanotubes in computer circuits, a development that could radically increase the speed and efficiency of electronics.

first carbon nanotube computer 2

s ta n f o r d e n g i n e e r i n g

few demonstrations of complete digital systems using this exciting technology. Here is the proof.” Carbon nanotubes are long chains of carbon atoms so thin that they can be easily switched on and off. “Think of it as stepping on a garden hose,” said Wong, the Willard R. and Inez Kerr Bell Professor in the School of Engineering. “The thinner the hose, the easier it is to shut off the flow.” But CNTs also have a bedeviling array of natural imperfections that had stymied all previous efforts to build complex circuits using them. First, a fraction of CNTs will behave like metallic wires that continually conduct electricity, instead of like semiconductors that can be switched on and off. Second, CNTs do not necessarily grow in neat parallel lines, as chipmakers would like. Though researchers have devised tricks to grow 99.5 percent of CNTs in straight lines, with the potential for billions of nanotubes to be used on every chip, even a tiny degree of misalignment is a problem. The Stanford paper describes a twopronged approach to these flaws that the authors call an “imperfection-immune design.” First, to rid their circuits of metallic CNTs, Stanford engineers switched all the desirable semiconducting CNTs into the “off” state and pumped the circuit with electricity, vaporizing almost all of the undesirable metallic CNTs in a process akin

to blowing a fuse. Getting around the alignment challenge required a bit more subtlety. The Stanford researchers created a powerful design algorithm to map out a new circuit that would function whether the CNTs were aligned or not. The Stanford researchers then assembled a basic computer with 178 transistors, composed of tens of thousands of CNTs and about 2 billion carbon atoms. Their CNT computer performed tasks such as counting and number sorting. It also runs a basic operating system that allows it to swap between these processes. Though it could take years to mature, the Stanford approach points toward the tantalizing possibility of industrial-scale production of carbon nanotube electronics. “CNTs offer the potential to cut, by orders of magnitude, the power consumption of electronics,” said Professor Georges Gielen, vice rector of science, engineering and technology at Katholieke Universiteit in Leuven, Belgium. “If this materializes, this will be a major breakthrough.” n

Professors Subhasish Mitra, left, H.-S. Philip Wong, center, and doctoral student Max Shulaker, right, created a computer using carbon nanotubes instead of silicon. Opposite page: A wafer filled with carbon nanotube computers.

Left and right: Rod Searcey; middle: Joel Simon; Opposite: Norbert von der Groeben

For decades, silicon has reigned supreme as the semiconductor of choice in computer circuitry. Progress in silicon-based chipmaking has largely meant shrinking the size of each transistor to pack more transistors on a chip and achieve performance and energy efficiency at the same time. But as transistors—the tiny on-off switches at the heart of all digital electronic systems—become tinier, they waste more power and generate excess heat. These issues are at the heart of the challenges that may one day diminish silicon’s dominance. Foreseeing that demise, engineers have been in a heated race to find that next big material to replace silicon. About 15 years ago, a promising new aspirant entered the scene: carbon nanotubes (CNTs). Tiny, inexpensive and exceptionally efficient at conducting electricity, CNTs hold the potential of a new generation of electronics that run faster and more efficiently than today’s silicon-based chips. And yet no one had built a true computer using carbon nanotubes until a team at Stanford announced it had done just that in a landmark cover story in the journal Nature written by Max Shulaker and other doctoral students in Electrical Engineering, who were led by Stanford Engineering professors Subhasish Mitra and H.-S. Philip Wong. “People have been talking about a new era of carbon nanotube electronics moving beyond silicon,” said Mitra, an associate professor of electrical engineering and of computer science. “But there have been

s ta n f o r d e n g i n e e r i n g

Electronic frontiers

Speed of sharing predicts when images go viral It’s hard to guess which of the many millions of photos on Facebook will spring from obscurity and “go viral,” but Stanford researchers have found some hints by studying the “cascade” effect of photos or videos that are shared multiple times. According to Facebook, only 1 in 20 photos posted to the social network is shared even once. Just 1 in 4,000 gets more than 500 shares—a lot but hardly epidemic. “It is very hard to quantify what going viral means,” said Jure Leskovec, an assistant professor of computer science studying the phenomenon. “Anyone would say ‘Gangnam Style’ went viral, but that’s a singular event,” he said, referring to the YouTube video that has been viewed more than 2 billion times. Leskovec and team were able to determine photo popularity with some accuracy, predicting 8 out of 10 times when a photo cascade would double in shares—that is, if a photo was shared 10 times, would it be shared 20 times, or if it reached 500, would it be shared 1,000 times. The team, which included Stanford doctoral student Justin Cheng and researchers from Facebook and Cornell University, began with 150,000 Facebook photos, each of which had been shared

Software tool cuts cost of cloud computing

s ta n f o r d e n g i n e e r i n g

Stanford doctoral student Christina Delimitrou, left, and Professor Christos Kozyrakis, right, adapted an idea from Netflix to improve data center efficiency. the movie viewing histories and ratings from one group of subscribers, for instance—to make educated guesses as to what similar viewers might like. Quasar evaluates performance requirements and server capacity instead of movie preferences and subscriber history. “Quasar recommends the minimum number of servers for each application and which applications run best together,” Delimitrou said. Kozyrakis and Delimitrou showed how collaborative filtering achieved utilization rates as high as 70 percent in a 200-server test while still meeting strict performance goals for the applications. “This is a proof of concept that could change the way we manage server clusters,” said Jason Mars, a computer science professor at the University of Michigan. n

Top left: Norbert von der Groeben; Bottom: Tom Abate

We hear a lot about the future of computing in the cloud but not much about the efficiency of the data centers that make the cloud possible. Data centers cost millions of dollars to build and operate, yet at any given moment, most of the servers in a typical data center are using only 20 percent of their capacity, keeping the excess, in theory, to ensure the data center won’t crash if demand surges. As cloud computing grows, so too will the price of this idle capacity. Sensing a need, two Stanford engineers have designed a new scheduling algorithm, called Quasar, which manages excess capacity in ways that triple server efficiency while delivering reliable service at all times—and reducing costs. Christos Kozyrakis, an associate professor of electrical engineering and of computer science, and Christina Delimitrou, a doctoral student in electrical engineering, got the idea from the software that Netflix and Amazon use to recommend movies, books and other products to their customers. It’s called collaborative filtering. Collaborative filtering uses known facts—

at least five times. The data were stripped of names and identifiers to protect privacy. The researchers calculated that at any given point in a cascade, half will double further, the other half not. The scientists then looked for variables that might help them predict the odds of share doubling better than a coin toss. Eventually they were able to accurately predict doubling events almost 80 percent of the time, and for photos shared hundreds of times, their accuracy rate approached 88 percent. The speed of sharing proved to be the best predictor of cascade growth. Simply analyzing how quickly a cascade unfolded predicted doublings 78 percent of the time. “Slow, persistent cascades don’t really double in size,” Leskovec said. Overall, however, the researchers found no simple trick to ensure widespread sharing. “Even if you have the best cat picture ever, it could work for your network but not for my boring academic friends,” Leskovec said, “You have to understand your network.” n

Software system analyzes sentence sentiment Whether the topic is politics, fashion or films, people express opinions in writing every day, rating the ideas and experiences they encounter on a scale of sentiment. With the spread of social networks, a vast reservoir of popular sentiment now exists that could, if systematically analyzed, provide clues about our collective likes and dislikes as a society. Against this backdrop, Stanford computer scientists have created a software system that analyzes sentences and gauges sentiment on a five-point scale, from strong like to strong dislike. The program, dubbed NaSent (short for Neural Analysis of Sentiment), is a new development in deep learning—machines extracting information from language without constant reference to human-made dictionaries or rules. “We wanted to get away from human experts having to create so many rules and instructions,” said computer scientist Richard Socher, who developed NaSent in collaboration with Stanford computer science professors Christopher Manning and Andrew Ng, and Christopher Potts, an associate professor of linguistics. NaSent’s multi-word approach improves on previous generations of software that analyzed single words to merely tally positive and negative words. This is no easy task. NaSent was able to understand that two sentences using different arrangements of the exact same words communicate different sentiments:

Assistant Professor Jure Leskovec explains how researchers can use the speed and pattern of photo sharing events to predict, 8 times out of 10, when a photo will go viral on Facebook.

Right: Norbert von der Groeben

“Unlike the surreal Leon, this movie is weird but likeable.” “Unlike the surreal but likeable Leon, this movie is weird.” The old system would not catch the subtle distinction, Manning said. The scientists started from roughly 12,000 movie review sentences. They split these sentences into phrases, using automated techniques to parse groups of words into grammatical units of meaning.

Professors Christopher Manning, above, and Andrew Ng, below, collaborated on an algorithm that gives machines the ability to understand how words form meaning in sentences.

The result was 214,000 phrases and sentences. Teams of three humans then read each sentence and evaluated it for intensity of like or dislike—“labeling” the data, in the researchers’ terms. Using NaSent, the computers then studied the labeled data the way a student might study a grammar text and assigned each a set of mathematical attributes. Next, with no further human rules or interventions, NaSent analyzed the labeled data and computed its own framework for predicting the sentiments that these words, phrases and sentences conveyed. “We’re still a long way from having computers ‘understand’ language the way human beings understand it,” Manning said. “But language understanding is getting better all the time.” n

s ta n f o r d e n g i n e e r i n g

E n g i n ee r i n g & h e a lt h

Electrical engineer’s breakthrough sends electromagnetic waves safely into the body, providing wireless power for a new generation of tiny pacemakers, sensors, drug delivery systems and other medical implants.

pacesetter Stanford electrical engineer Ada Poon has invented a way to wirelessly transfer power deep inside the body and use it to run tiny electronic medical gadgets such as pacemakers, nerve stimulators or new sensors and devices yet to be developed. The discoveries culminate years of efforts by Poon and her colleagues to eliminate the bulky batteries and clumsy recharging systems that prevent small medical devices from being more widely used in human medicine.

s ta n f o r d e n g i n e e r i n g

“We need to make these devices as small as possible to more easily implant them deep in the body and create new ways to treat illness and alleviate pain,” said Poon, an assistant professor of electrical engineering. Poon’s team has built a pacemaker smaller than a grain of rice. It can be powered or recharged wirelessly by holding a power source about the size of a credit card above the device— outside the body. Her lab has tested the wireless charging system in a pig and used it

Assistant Professor Ada Poon has invented an electronic implant

Courtesy of Austin Yee; Inset: Linda A. Cicero

smaller than a medicinal pill.

to power a tiny pacemaker in a rabbit. She is currently preparing the system for testing in humans. Should such tests prove successful, several years probably remain to satisfy the requirements of commercial medical devices, but that day is within sight. Poon believes this discovery will spawn a new generation of programmable microimplants—sensors to monitor vital functions deep inside the body; electrostimulators to change neural signals in the brain; and drug delivery systems to apply

medicines directly to affected areas. The breakthrough involves a new way to control electromagnetic waves inside the body. Before Poon’s discovery, there was a divide between the two main types of electromagnetic waves in everyday use: far-field and near-field waves. Far-field waves, like those from radio, either reflect off the body harmlessly or are absorbed by the skin as heat. Near-field waves can safely power wireless systems, but they transfer power only over short distances, lim-

iting their range inside the body. Poon blended the safety of nearfield waves with the reach of far-field waves into something she calls “midfield waves” by taking advantage of the fact that waves travel differently through different materials. When her special type of midfield wave moves from air to skin, it changes in such a way as to propagate, like the sound waves through a train track. With this newfound reach, it is possible to recharge batteries wirelessly through human tissue. n

s ta n f o r d e n g i n e e r i n g

E n g i n ee r i n g & h e a lt h Professor James R. Swartz’s team has made promising steps toward the

Technique to produce artificial protein is step toward universal flu vaccine Every year flu season sets off a medical guessing game with life or death consequences. Flu has many strains, and they vary by year. Public health officials must make an educated guess as to which variants they should produce vaccines against. A bad guess or an unexpected strain could send death tolls skyrocketing. Against this backdrop, Stanford researchers have made progress toward a universal flu vaccine, one that could be produced more quickly and offer broader protection than those available today. The team was led by James R. Swartz, the James H. Clark Professor in the School of Engineering.

s ta n f o r d e n g i n e e r i n g

Their approach arises from a better understanding of the structure of a key protein on the surface of the flu virus. A flu virus is made up of different proteins. Protruding from the surface are copies of the protein hemagglutinin (HA). HA resembles a mushroom, with a head and a stem. The structure of the head of HA is key to the virulence of a given strain of flu. Today’s vaccines are based on inactivated viruses that contain the heads of HA proteins. The immune system, however, still sees the HA heads as infections and creates antibodies to fight them. Primed to destroy the inactive viruses,

the immune system will also kill active viruses. The key is to be vaccinated before contracting the active virus. Swartz and colleagues focused on the stem of the HA protein, which remains more constant, rather than the head, which varies from year to year. Thus, a vaccine based on the stem might be more broadly protective, perhaps even avoiding the need for yearly injections. It could also reduce pandemic threats. The researchers started with the DNA sequence that defines the entire HA protein, both head and stem, for the H1N1 flu variant, the same one that caused the pandemic of 1918 and recurred in milder form in 2009. Subtracting the coding for the head, the researchers created a strand that contains only the instructions for making the stem for their new experimental vaccine. Producing this complex artificial protein required multiple iterative advances. It took years to iron out and perfect the technique. “It has been a tough project,” said Swartz, a chemical engineer and bioengineer. “Many labs have been trying to develop an HA stem vaccine.” The approach remains experimental and has not yet been tested on patients. Many steps remain before it’s clear whether this approach yields a better flu vaccine. What’s more, the vaccine must not only be broadly effective against different strains of flu but also cheap to produce so it can be widely distributed. Additional advances from the Swartz lab offer promise with this challenge as well. The stakes are high: Recent estimates put the worldwide death toll from flurelated illnesses at between 250,000 and 500,000 each year. “This is an important project for world health,” Swartz said. “These are big challenges but we are committed to the effort.” n

Saul Bromberger and Sandra Hoover

creation of a universal flu vaccine.

Neurogrid circuit board is modeled on the human brain tic connections—all while using only the power requirements of a tablet computer. Boahen now stands ready to reduce costs and create compiler software that would enable engineers and computer scientists to use Neurogrid to solve problems such as controlling humanoid robots or prosthetic limbs. “Right now, you have to know how the brain works to program one of these,” Boahen said, gesturing at his $40,000 prototype. “We want to create a neurocompiler so that you would not need to know anything about synapses and neurons to use one.” In keeping with Boahen’s goal of a system affordable enough to be widely used in research, Neurogrid is—despite its current high cost—the most cost-effective way to simulate neurons. He estimates that by switching manufacturing processes and fabricating chips in volume he can create a Neurogrid for $400. At that price, with better compiler software, the device

could find numerous applications. Boahen can envision, for instance, implanting in a paralyzed person’s brain a chip that can interpret thoughts and translate them to commands for prosthetic limbs. Already, Neurogrid is being used to control a small prosthetic arm in his lab in real time. Price aside, there is room for improvement on the efficiency front as well. “The human brain, with 80,000 times more neurons, consumes only three times as much power as Neurogrid,” Boahen said. “Energy efficiency remains the ultimate challenge.” n

Professor Kwabena Boahen’s Neurocore chips mimic brain functions in an energy-efficient way that could make them useful in controlling prosthetic limbs.

Tom Abate

For all their sophistication, computers pale in comparison to the biological brain. The modest cortex of the mouse, for instance, operates 9,000 times faster than a personal computer mimicking its function. Yet the computer consumes 40,000 times more power than the mouse brain. In this context, Kwabena Boahen, a professor of bioengineering at Stanford, and colleagues have developed a new circuit board, called Neurogrid, modeled on the human brain. The device could open new frontiers in robotics and computing. Boahen’s Neurogrid is an efficient circuit board of 16 custom-designed Neurocore chips that allow certain synapses to share hardware circuits. Though Neurogrid is about the size of an iPad, it can simulate orders of magnitude more neurons and synapses than other brain mimics thus far invented— 1 million neurons and billions of synap-

s ta n f o r d e n g i n e e r i n g

E n g i n ee r i n g & h e a lt h

Brain research could improve prosthetic limbs

Recent research has shown that tiny fragments of DNA circulating in the blood can be used to monitor cancer growth and even get a sneak peek into a developing fetus’s gene sequences. But the DNA offers little insight into how our bodies generate the dizzying array of cells, tissues and biological processes that define our bodies and our lives. Now researchers at Stanford University have generated a much more specific picture by monitoring another genetic material—RNA—in the blood. It’s the biological difference between a still photo and a video when it comes to figuring out what the body is doing and why. “We think of this technique as a kind of molecular stethoscope,” said Stephen Quake, the Lee Otterson Professor in the School of Engineering, “and it’s broadly useful for any tissue you care to analyze. We could potentially use it to look for things going wrong in pregnancy. And we hope to use it to track general health issues in various organs.” DNA provides information about the blueprint of the entire organism, and every cell contains the same blueprint. However, a particular form of RNA, called messenger RNA, reveals more about each of the different cell types or tissues in the organism because each cell type expresses distinct messages from different parts of the blueprint. Studying messenger RNA can also provide evidence of whether or not

s ta n f o r d e n g i n e e r i n g

Professor Stephen Quake’s team has discovered how to use RNA tests to understand organ health and development. cells are healthy—providing Quake and his colleagues a new way to zero in on the status of clearly defined tissues rather than the entire organism. In this study, the researchers used new techniques to identify which circulating RNA molecules in a pregnant woman were likely to have come from her fetus and which from her own organs. Ultimately, they were able to trace the development of specific tissues, including the fetal brain and liver, during pregnancy simply by analyzing RNA in the blood samples of pregnant women. Quake and colleagues believe this technique could be broadly useful as a diagnostic tool and could detect distress signals from diseased organs, perhaps even before any clinical symptoms are apparent. “We’ve moved beyond just detecting gene sequences to really analyzing and understanding patterns of gene activity,” said Quake, a professor of bioengineering and of applied physics. “Analyzing the RNA enables a much broader perspective of what’s going on in the body at any particular time.” n

Professor Krishna Shenoy’s team has determined what mechanism allows two brain regions to communicate. Top: Norbert von der Groeben; Bottom: Joel Simon

Molecular stethoscope tracks fetal development and disease

The human brain is a remarkable organ whose important mysteries are just beginning to be understood. Engineers are working with their colleagues in medical science, psychology and biology to unravel the intricate chemical and electrical system that is the human brain. This year, two studies led by Stanford Electrical Engineering Professor Krishna Shenoy made significant strides in expanding our knowledge of the inner workings of the brain. Shenoy is designing brain-computer interfaces to create next-generation prosthetics. In the course of this work, he and his team revealed a previously unknown process that helps two brain regions cooperate when joint action is required to perform a task. “The serendipitous interplay between basic science and engineering never ceases to amaze me,” Shenoy said. “Some of the best ideas in prosthetics come from basic neuroscience research, and some of the deepest scientific insights come from engineering measurement and medical systems.” The many discrete regions of the brain often work independently, relying on neurons inside that region to do their work. At other times, two regions must cooperate. The riddle is this: What mechanism allows two brain regions to communicate when they need to, yet avoid interfering with each other when they must work alone? The researchers derived their findings by studying trained monkeys taught to pause briefly before

Norbert von der Groeben

an arm movement, thus letting their brains prepare. The researchers took electrical readings from the arm muscles and from each of the two motor cortical regions in the brain known to control arm movements. Each motor cortex has more than 20 million neurons, far too many to probe en masse, so the team sampled about 100 to 200 individual neurons in each region. Researchers studied the activities of each neuron individually and how the group behaved as a whole. The key findings emerged from understanding how individual neurons worked together to drive the muscles. In the preparatory phase when the arm is still, neurons in both cortices showed big changes in activity, all without driving movement in the arm. The researchers now believe that the brain carefully balances the neuronal activity—some neurons firing faster, others slower—to broadcast a constant message to the muscles. At the moment of action, the group readings change again in a way that can be correlated with flexing of the muscles. This change at the group level is what differentiates preparation from action. “This is among the first mechanisms reported for letting brain areas process information continuously but only communicate what they need to,” said Matthew T. Kaufman, who was a postdoctoral scholar in Shenoy’s lab when the research took place. In related research, Shenoy refined his earlier studies about how the brain initiates physical motion. In anticipation of a planned physical act—as we reach for a set of keys, for instance—neurons adopt a state of readiness, like sprinters in a crouch. At other times—when someone unexpectedly tosses keys at us—neurons react with no preparation. Shenoy and Katherine Cora Ames, a doctoral student in his lab, studied the brain activity of monkeys in three variations of an experiment in which the monkeys were trained to touch a target on a display screen. In all instances, the first information to reach the neurons was awareness of the target. A splitsecond later, differences in the data appear. The neurons of monkeys awaiting a “go” command show a prepare-and-hold state—“ready” and “set.” In the two cases testing unplanned motion, the neurons did not demonstrate this state. Instead, a change in neuronal activity signaled the command to touch the target with no apparent preparation between perception and action. In these instances, the neurons just say “go.” n

Engineer helps curb disease and fight bioterrorism The high-stakes logistical challenges of World War II gave birth to operations research—engineering focused on optimizing the manufacture, deployment and use of personnel and materiel to support U.S. military efforts. Today, operations research is part of the broader field known as management science and engineering, but the principles remain: efficiency and optimization in all things. Margaret Brandeau, the Coleman F. Fung Professor in the School of Engineering, is applying the rigorous modeling and mathematics of operations research to challenging problems in public health and security. “The complex systems models that have streamlined operations and manufacturing are equally adept at helping us determine the most efficient course of action in public health,” she said. In a perfect world, all public health decisions would be driven by data from randomized clinical trials. But such trials can be costly, impractical or, in some cases, unethical. Modeling fills the void. In one case, Brandeau developed recommendations to combat hepatitis B infections in Asian populations. Her work has led to a dramatic shift in policy in China; that country now provides free catch-up vaccination to all children under age 15—as many as 140 million children. Although Brandeau’s research showed that the program could cost as much as $500 million, it also indicated that the shots would probably prevent nearly 8 million acute infections, 400,000 chronic infections and almost 70,000 deaths—all while saving $1.4 billion in future health care costs. In the area of HIV prevention, Brandeau modeled the daily use of an expensive but effective antiretroviral drug known as Truvada by men who have sex with men. Brandeau’s analysis showed that Truvada could be cost effective, but only if administered to the 20 percent of people at highest risk of infection.

Professor Margaret Brandeau works internationally to champion public health and effective disease control initiatives. Brandeau also applied her expertise after the 2001 anthrax attacks that killed five people, when bioterrorism became a very real threat. She and collaborators examined the supply chain for antibiotics that would be necessary to combat a largescale anthrax attack. Surprisingly, her models showed that the limiting factor in a response is likely to be dispensing capability and not antibiotic supplies—pointing out the need for cities to develop improved plans for dispensing the antibiotics. Her work also showed that stockpiles should be amassed in local hospitals but only in the areas of greatest risk—metropolitan communities with high population density where an attack would be most likely to prove lethal. After that, Brandeau was tapped by the U.S. Department of Health and Human Services to sit on a scientific advisory board that oversees the preparedness efforts of the Centers for Disease Control and Prevention, including the $5 billion U.S. Strategic National Stockpile. “This is a war chest for any public health emergency, but we lacked a basic understanding of what exactly the stockpile should be as our nation’s needs evolve,” Brandeau said. “Modeling can help in planning the stockpile.” n

s ta n f o r d e n g i n e e r i n g

Persis S. Drell speaks at the Stanford Photonics Research Centerâ&#x20AC;&#x2122;s Women in Science Seminar in October.

s ta n f o r d e n g i n e e r i n g

ersis S. Drell, former director of SLAC National Accelerator Laboratory, became the ninth dean of the School of Engineering on Sept. 1.

Drell led the 1,600‐employee U.S. Department of Energy SLAC National

Accelerator Laboratory at Stanford from 2007 through 2012. During her tenure as director, SLAC transitioned from being a laboratory dedicated primarily to research in high‐energy physics to one that is now seen as a leader in a number of scientific disciplines. Drell is the first woman to hold the post of dean at the school. She succeeds Jim Plummer, who stepped down after 15 years, making him the longest-serving dean in Stanford Engineering history. Drell is the Frederick Emmons Terman Dean of the School of Engineering and the James and Anna Marie Spilker Professor of Materials Science and Engineering and of Physics at Stanford University. She recently sat down with Jamie Beckett, Stanford Engineering director of communications and alumni relations, to discuss her life, her perspective on the school and what brought her to Stanford Engineering.

Meet Persis Drell Saul Bromberger and Sandra Hoover

The school’s new dean was raised on The Farm, made her mark in physics and is focused on the future.

What attracted you to the School of Engineering? The School of Engineering is made up of an extraordinary group of people. That became especially clear to me when I met members of the search committee—they were the best recruiting tool imaginable. That was the moment when, for me, it got very serious.

What is your vision for the school? How would you like to see it evolve? I am extremely ambitious for the school. Engineering at Stanford, in partnership with Silicon Valley, has changed the world. I have no intention of backing away from similar ambition for our future. The biggest challenge we have is that we are so

s ta n f o r d e n g i n e e r i n g

When you stepped down as director of SLAC you cited a strong desire to return to teaching and research. But here you are— dean of the School of Engineering. What changed? This was a great example of Yogi Berra’s famous advice “When you come to a fork in the road, take it!” I was very happy doing research and teaching when this opportunity came along—and I took it! Most deans continue teaching and continue to engage in research. I’ll teach one quarter a year as Jim [Plummer] has done. In the winter quarter, I’ll again be teaching a companion course to Physics 41 [Mechanics], the largest class in the physics department with some 600 students. My course, 41A [Mechanics Concepts, Calculations and Context], is for students who did not come with great preparation

s ta n f o r d e n g i n e e r i n g

“The biggest challenge we have is that we are so successful. And that is a challenge. We must avoid becoming complacent and continue to move forward.” — Persis Drell from high school. I love that class. They are terrific students. My research is with the X-ray laser up at SLAC, where there is a very wellestablished infrastructure with a team of superb scientists. It’s a team I can be part of, rather than leading an individual lab. I think that will be manageable. In fact, I think that will help keep me grounded. How did you become interested in science? I got “tracked low” in math in seventh grade. And I didn’t like that. I started working very hard in math, and they moved me up, and that got me focused on math.

I had a horrendous physics class in high school. But then I went to Wellesley, and I had a spectacular physics professor, Phyllis Fleming, and she made me a physicist. You’re the daughter of Stanford faculty member Sidney Drell. How did having a father who was a prominent physicist affect your career choice? In the late 1960s and 1970s when my father was head of the theory group at SLAC, and later on deputy director, SLAC was the center of the universe for particle physics. People would come by the house, and there would be conversations late into the night. I wasn’t interested in the physics, but the people were fascinating—Hans Bethe, Richard Feynman, T.D. Lee—you name it. However, having a father in the field was also a challenge. When I took physics in high school and wanted help with my homework, my father would want to explain everything to me. I just wanted the answer to Problem Three. What are the key influences that have shaped your life and career so far? In addition to my family, there were two teachers who were phenomenally influential. One was my Latin teacher for six years at Terman [Middle School] and Gunn High School—Marian McNamara. And the other was Phyllis Fleming at Wellesley.

Saul Bromberger and Sandra Hoover

successful. And that is a challenge. We must avoid becoming complacent and continue to move forward. I don’t know what the school will be like in 15 years; I just know it will be different. The great thing is that the school is incredibly forward-looking. The people here are not afraid of change. In not a single conversation has someone said that we have to do things a certain way because we’ve always done them that way. Here, you get measured by what you get done, and I like that. Everyone is focused on making an impact.

They were strong personalities, and they were absolutely dedicated teachers. I learned so much from them. The Latin teacher didn’t just teach us Latin. She taught us cultural history. It was everything about the Roman Empire and Roman culture, so I’m a total Italophile. Another thing that made her so gifted —and I think it’s a lesson I carried with me—is that she judged each of her students as individuals. We were a diverse class in terms of abilities, and she set the expectations for each student depending upon what she thought that student could deliver. She demanded a lot from all of us, but we felt we were working toward clear expectations that were realistic for us. The students responded amazingly well. Which of your accomplishments has been the most satisfying? I’m proud of raising three great kids. My youngest daughter recently turned 20, so I’ve had a teenager, sometimes three teenagers, in the house for 14 years. And finally, the youngest one is no longer a teenager. It’s been wonderful to see them emerge from the teenage years as terrific young adults. Another thing is the continuing success of SLAC under the new leadership of ChiChang Kao. That is deeply, deeply satisfying to me because it’s a great institution with great people. Something else that’s been satisfying is the feeling of being able to make my own choices along the way to live life in the way that I wanted to lead it. There have been a lot of things that I’ve had to choose not to do. But I felt like the choices were under my control—that I had choices that were good choices. What do you consider to be the qualities of an effective leader? The greatest leaders that I have worked for have been able to really inspire me to take risks. But then they were also able to protect me from my weaknesses. There is a great man I worked for at Cornell— Maury Tigner—who knew how to inspire

me to do things beyond what I thought I was capable of, but who at the same time protected me by not asking me to do things that I wasn’t well constituted to do. What role do engineers have in helping to solve the challenges facing the world? I think engineers will contribute to solving health care problems, they’ll contribute to energy solutions and they’ll contribute to climate change solutions. They’ll contribute to—and are contributing to—one of the great challenges we face, which is sustainable water. My sense of the school is this: The faculty, students and staff in this school want to make a significant impact and change the world, making it a better place. And that’s hard to do, but it’s what we ought to be striving for. Is there something about you that most people don’t know? I’ve played chamber music for decades but last year I started taking cello lessons again. My teacher is Chris Costanza of the St. Lawrence String Quartet—the quartet in residence at Stanford. I will work very hard as the dean, but I will not give up practicing my cello and taking cello lessons. My husband [Jim Welch, a SLAC accelerator physicist] plays viola and I’m a cellist, and we’re always looking for good violinists [to make a string quartet]. Chamber music is an enormous source of pleasure for us. Another thing most people don’t know is that I grew up on the Stanford campus. My [old] house is sitting nearby. They picked it up and moved it to make way for the Munger dorms. And it’s now the Sexual Harassment Policy Office and the Housing Office. It was one of the original 12 houses that Sen. (Leland) Stanford built for his faculty. The hitching post is still out front. My dad bought the house, sight unseen, when nobody wanted it. My parents moved here when I was 6 months old. So I’m about as Stanford as you can get. n

About the school of engineering The Stanford School of Engineering has been at the forefront of innovation for nearly a century, turning big ideas into solutions that improve people’s lives. Our mission is to solve important global problems and educate leaders who will make the world a better place by using engineering principles, techniques and systems. We educate engineers who possess not only deep technical excellence but also the creativity, cultural awareness and entrepreneurial skills that come from exposure to the liberal arts, business, medicine and other disciplines that are an integral part of the Stanford experience. Collaborating with colleagues across disciplines, Stanford engineers strive to create efficient energy sources, diagnose and treat diseases, ensure clean water, enhance global communication and unleash human creativity. Stanford Engineering at a Glance •

• • •

4,894 students 1,444 undergraduates 3,450 graduates 260 faculty members Three No. 1 department rankings; all departments in the top six More than 80 labs, centers and affiliate programs involving students in research

The School’s Nine Departments • • • • • • • • •

Aeronautics and Astronautics Bioengineering Chemical Engineering Civil and Environmental Engineering Computer Science Electrical Engineering Management Science and Engineering Materials Science and Engineering Mechanical Engineering

s ta n f o r d e n g i n e e r i n g

Moved in and moving forward When the Shriram Center for Bioengineering and Chemical Engineering opened last fall, it completed the Science and Engineering Quad. The research and teaching carried out in this 8.2-acre complex exemplifies the increasingly porous boundaries among disciplines, helping to advance Stanford Engineeringâ&#x20AC;&#x2122;s efforts to solve the worldâ&#x20AC;&#x2122;s most complex and pressing problems. Clockwise from the left foreground: the Shriram Center, the James and Anna Marie Spilker Engineering and Applied Sciences Building, the Jen-Hsun Huang Engineering Center, and the Jerry Yang and Akiko Yamazaki Environment and Energy Building.

s ta n f o r d e n g i n e e r i n g

JoelSimon Joel Simon

s ta n f o r d e n g i n e e r i n g â&#x20AC;&#x201A;

E n g i n ee r i n g f e at s

In a brain-exploration technique called optogenetics, molecules called “microbial opsins” are targeted to specific nerve cells such as the neuron shown here, with the green showing

In 2005, Stanford bioengineer Karl Deisseroth, a practicing psychiatrist, pioneered a brain-exploration technique known as optogenetics that involves selectively introducing lightsensitive molecules to very specific nerve cells within a living animal’s brain and then carefully positioning near those nerve cells the tip of a lengthy, ultra-thin optical fiber. The fiber, in turn, is connected to a laser outside the brain. Using optogenetics, Deisseroth, the D.H. Chen Professor, can remotely stimulate or inhibit the modified brain circuits—and only those circuits—at the flip of a light switch. The animal remains alive, free to move around. In the years since, optogenetics has moved beyond psychiatry alone and is having far-reaching effects in other fields as well. Scott Delp, the James H. Clark Pro-

s ta n f o r d e n g i n e e r i n g

Professor Karl Deisseroth is a pioneer in optogenetics, which uses light to control brain activity.

fessor in the School of Engineering and a professor of bioengineering and of mechanical engineering, has used optogenetics to relieve pain in mice. The optogenetically modified cells are not in the brain but in the mice’s paws. Using one color of light, Delp can increase pain in the mice; with another he can decrease it. “This is an entirely new approach to

studying a huge public health issue,” Delp said. “It’s a completely new tool that is now available to neuroscientists everywhere.” Meanwhile, Deisseroth continues to hone his own technique. Exciting as it was at the time, early optogenetics was flawed in that it was good at switching cells on but less so at turning them off. This year, Deisseroth’s team re-engineered its light-sensitive proteins to switch cells off far more efficiently than before. At its most basic level, optogenetic light pulses cause a modified protein in the cell to open a “channel” in the cell membrane through which positive ions flow continually, like water through a garden hose. The positive ions are like gasoline, fueling cell activity. In inhibitory functions, the proteins move negative ions rather than positive, but they do not produce a contin-

Top: Courtesy of the Deisseroth Lab; Left: Saul Bromberger and Sandra Hoover

opsin distribution in the cell.

Using lasers to turn animal nerve cells on and off, bioengineers can open and close circuits, relieve pain and even affect behavior.

Steve Fisch

light switch ual flow. Instead, they act more like a pump than a channel, moving one ion for every incoming photon—a squirt gun versus a garden hose. Deisseroth wanted an inhibitory protein to function more like a continuous channel, so he modified 10 of the protein’s 300 amino acid building blocks. The result is a vastly more light-sensitive inhibitory channel than ever before and a greater ability for researchers to turn off certain cell behaviors as efficiently as they can turn others on. “This creates a powerful tool that allows neuroscientists to apply a brake in any specific circuit with millisecond precision, beyond the power of any existing technology,” said Thomas Insel, director of the National Institute of Mental Health, which funded the study. Deisseroth also uses optogenetics to advance scientific knowledge of how the brain works. In that respect,

Professor Scott Delp has used optogenetically modified cells to control pain in mice.

this year he was able to instantly increase a mouse’s appetite for getting to know a strange mouse and likewise to inhibit the mouse’s drive to socialize. The new findings may shed light on psychiatric disorders marked by impaired social interaction such as autism, social anxiety, schizophrenia and depression. “People with autism often have an

outright aversion to social interaction,” Deisseroth said. “Every behavior presumably arises from a pattern of activity in the brain, and every behavioral malfunction arises from malfunctioning circuitry.” Deisseroth was able to manipulate the activity in a specific region known as the ventral tegmental area (VTA), which transmits signals to other centers in the brain to produce dopamine, creating pleasurable brain activity. The VTA is linked to drug abuse and depression, but much less is known about its role in social behavior. “The ability, for the first time, to pinpoint a particular connection from one part of the brain to another that is involved in the social behavior of a living, moving animal will greatly enhance our ability to understand how social behavior operates and how it can go wrong,” Deisseroth said. n

s ta n f o r d e n g i n e e r i n g

E n g i n ee r i n g f e at s

Stanford professor shapes safety specs for next-generation aircraft

s ta n f o r d e n g i n e e r i n g

sors can continuously keep tabs on performance while planes are in operation, benefiting air travelers by keeping more planes safely in flight. Chang explains that a key structural health monitoring technology is based on piezoelectricity, in which squeezing or bending a solid object generates electricity. The reverse is also true: Electricity can cause materials to change shape. Piezoelectricity works in structural health monitoring in passive and active ways: Sensors built into carbon composites can detect strains in these structural elements in a passive way. At the same time, actuators can be used to create active diagnostic tests. Much like a doctor striking your knee with a rubber mallet, aircraft designers are effectively “tapping” components with jolts of electricity to ascertain their structural health. Chang said technology is only part of the story. Structural health monitoring has the potential to change the mindset of

Professor Fu-Kuo Chang has developed carbon-fiber panels containing sensors that monitor the structural health of aircraft.

aircraft maintenance. Today, structural maintenance is scheduled. After a certain number of flight hours, aircraft are taken out of service for inspection. Flaws that develop after such an inspection could be dangerous. Meanwhile, structurally sound aircraft are needlessly taken out of the flight rotation. The built-in structural health monitoring systems could provide feedback from operating aircraft, spot trouble earlier and keep sound aircraft in operation longer. “Structural health monitoring allows us to shift from scheduled maintenance to condition-based maintenance,” Chang said. n

Rod Searcey

A new generation of aircraft is taking flight on wings made from carbon composites, a class of materials less expensive to manufacture than aluminum with the added benefit of embedded safety sensors in the frame and skin. As these new materials have entered commercial use, researchers are working to understand how carbon composites withstand the rigors of flight. Fu-Kuo Chang, a professor of aeronautics and astronautics, recently led an international working group in publishing the world’s first comprehensive guidelines for using structural health monitoring to improve commercial aircraft worldwide. Experts say structural health monitoring will reduce the unnecessary downtime and last-minute delays for evaluation and maintenance. The carbon composites are woven from filaments that allow manufacturers to include sensors that act as a sort of central nervous system to monitor the plane’s structural health. These sen-

Engineers improve the structural integrity of carbon nanotubes

Top: John Todd; Bottom: Goodson Lab

When engineers design devices, they must often join two materials that expand and contract as temperatures change. The problem is that the two materials often do so at different rates, compromising the physical integrity of, say, a circuit and raising the potential for failure. These concerns intensify as devices advance into and beyond the nanoscale and more electricity is pushed through ever-smaller circuits. “Think about the heat sink for a microprocessor,” said Kenneth Goodson, the Bosch Mechanical Engineering Department Chairman and Davies Family Provostial Professor at Stanford. “It is exposed to repeated instances of heating and cooling.” Addressing these concerns, Stanford engineers have created strong and supple carbon nanotube structures that could improve structural integrity at these critical thermal interfaces. Goodson and his team have spent five years exploring singlewalled carbon nanotubes (CNTs) that are a triple threat structurally speaking—displaying remarkable strength, flexibility and heat conductivity. To get here, the team has learned a lot about CNTs and how to assemble them with a different structural characteristic than they would otherwise display in nature. Left to nature, the carbon atoms that form CNTs will create structures that are notoriously hard to work with and—if we

5 um

could see them—would resemble a bowl of spaghetti. The CNTs created by Goodson’s team grow like grasses, upward and relatively straight. These longer, less-dense CNTs seemed to have the best combination of flexibility, heat conductivity and strength for the kinds of applications Goodson anticipates. To fully understand what was going on at the molecular scale, Wei Cai, an associate professor of mechanical engineering, helped build a computer simulation of the CNT assembly process with an eye toward understanding why the CNTs normally become bent and entangled. In particular, the Stanford engineers wanted to understand the role of what are known as van der Waals forces, named for the Dutch physicist who first described the weak attractions that exist between molecules. Cai said that while van der Waals forces may be too weak to affect larger structures, carbon nanotubes are so thin that such minute forces are enough to cause bending and entangling. “When you hear about nanotechnology, it’s usually superlative—the strongest this, the thinnest that,” Goodson said. “But we think deeper answers lie in using such knowledge to invent bulk materials that have unusual combinations of properties that never exist together in nature.” n

2 um

Professor Kenneth Goodson, top, and Associate Professor Wei Cai, above, have shown how to optimize carbon nanotube arrays. Below, zooming in on experimental carbon nanotube structures shows the entanglement that impedes their performance.

500 nm

s ta n f o r d e n g i n e e r i n g

s u s ta i n a b i l i t y

Special microbes produce electricity by digesting plant and animal waste dissolved in sewage.

wired microbes microbes 22

s ta n f o r d e n g i n e e r i n g

Inside a murky vial, clinging to a piece of carbon-like barnacles to a ship’s hull, an unusual type of bacteria—“wired microbes”—feast on sewage filled with particles of organic plant and animal waste. Grim as the scene sounds, however, the naturally occurring microbes’ byproduct is a very valuable commodity: electricity. The invention’s creators—Yi Cui, a materials scientist; Craig Criddle, an environmental engineer; and doctoral student Xing Xie—call their device a microbial battery. One day they hope it will be used in sewage treatment plants, or to break down organic pollutants in “dead zones” of lakes and coastal wa-

These exoelectrogenic microbes’ wire-like tendrils attach to carbon filaments to produce electricity. More than 100 of these microbes

Xing Xie

could fit side by side across the diameter of a human hair.

ters where fertilizer runoff and other organic waste can deplete oxygen levels and suffocate marine life. “We call it fishing for electrons,” said Criddle, a professor of civil and environmental engineering. Scientists have long known of what they call exoelectrogenic microbes— organisms that evolved in airless environments. To get their requisite, lifesustaining oxygen, they react with oxide minerals rather than breathe air. For a dozen years or so, research groups have tried to harness these microbes as bio-generators, but it has proved challenging. What is new about the Stanford

microbial battery is its simple yet efficient design. At the battery’s negative electrode, colonies of wired microbes cling to carbon filaments that serve as efficient electrical conductors. “The microbes make nanowires to dump off their excess electrons,” Criddle said. About 100 of these microbes could fit, side by side, in the width of a human hair. As these microbes ingest organic matter and convert it into biological fuel, their excess electrons flow into the carbon filaments and across to the positive electrode, which is made of silver oxide, a material that attracts electrons.

The electrons flowing to the positive node gradually reduce the silver oxide to silver, storing the spare electrons in the process. According to Xie, after a day or so the positive electrode has absorbed a full load of electrons and has largely been converted into silver. At that point it is removed from the battery and re-oxidized back to silver oxide, releasing the stored electrons. The Stanford engineers estimate that the microbial battery can extract about 30 percent of the potential energy locked in wastewater, roughly the same efficiency at which the best commercially available solar cells convert sunlight into electricity today. n

s ta n f o r d e n g i n e e r i n g

s u s ta i n a b i l i t y

New Stanford facility will test water-recovery technologies and director of the multi-institution engineering research center Re-inventing the Nation’s Urban Water Infrastructure. Researchers will also test innovations such as using by-products of water purification, such as methane, to power treatment plants. Stanford postdoctoral researcher Yaniv Scherson is piloting an effort to “turbocharge” methane combustion by adding nitrous oxide created from the ammonia in wastewater. Though probably years away from implementation on the Stanford campus, the technologies could eventually lead to recycled water of an acceptable quality to serve as an alternative non-potable water supply and play a role in Stanford’s longterm sustainable water management plans. Before Stanford can reuse wastewater, however, the reliability and safety of new technologies will be perfected to deal with pathogens, pharmaceuticals and personal hygiene products, among other contaminants. That’s where the research of faculty members Craig Criddle and Perry McCarty comes into play. Criddle, who will direct

the Codiga Center, is a professor of civil and environmental engineering who has worked with colleagues to develop technologies for recovery of energy from wastewater. McCarty, a professor emeritus of civil and environmental engineering, is recognized for developing economical anaerobic treatment systems that rely on naturally occurring beneficial microbes. He collaborated with colleagues at Inha University in South Korea to create a new anaerobic technology to recover clean water and energy from wastewater. “These technologies could revolutionize wastewater treatment,” Criddle said. “Many of these treatment systems do not exist at full scale anywhere in the world.” n

Professor Richard Luthy; Tom Zigterman, associate director for Water Services and Civil Infrastructure; and Professor Craig Criddle paved the way for the Codiga Resource Recovery Center at Stanford.

Norbert von der Groeben

Water purveyors and wastewater utilities nationwide struggle with the dual challenges of replacing our aging wastewater infrastructure while coping with water shortages. Against that backdrop, the university recently broke ground on the William and Cloy Codiga Resource Recovery Center at Stanford that will test new methods to turn wastewater into clean water and energy. The facility’s mission is to test these technologies at a scale large enough to demonstrate effectiveness and to stimulate investment for full-scale implementation. The new technologies may make it possible for wastewater, which is now transported to centralized sewage treatment plants, to be purified locally and recycled for irrigation, restoration of ecosystems and other purposes. “This effort will showcase how the campus can meet its future water supply needs through innovative approaches that produce non-potable water locally where it’s needed and in ways that save energy and money,” said Richard Luthy, the Silas H. Palmer Professor of Civil Engineering

s ta n f o r d e n g i n e e r i n g

Top: Chueh lab; right, Matt Beardsley/SLAC

X-rays illuminate path to more efficient fuel cells Solar power, wind power and other renewable energies don’t produce energy as predictably as plants fueled by oil, coal or natural gas. Ideally, alternative energy sources would include massive systems to store and dispense power at times when the sun isn’t shining or the wind isn’t brisk—think batteries on steroids. Fuel cells use oxygen and hydrogen as fuel to create electricity; if the process were reversed, the fuel cells could be used to store electricity as well. But until now those catalytic processes weren’t well understood. William Chueh, an assistant professor of materials science and engineering, recently led a team that for the first time observed the hydrogenoxygen reaction in a specific type of fuel cell to illuminate how it works. The knowledge could lead to more efficient fuel cells and make utility-scale alternative energy systems practical. In a fuel cell, a catalytic reaction at the cathode produces negatively charged oxygen ions. The ions then make their way across the cell to the anode, where they

react with stored hydrogen molecules to produce electricity. In a typical fuel cell, a gas-tight membrane separates the anode and cathode. While the oxygen ions pass easily through the membrane, electrons cannot. They must instead circumvent it through an electrical circuit that can be harnessed to run anything from cars to power plants. While electrons are thought to be the critical component of fuel cells, the lessunderstood ion flow is just as important, Chueh said. He and his colleagues from SLAC National Accelerator Laboratory and elsewhere applied high-brilliance X-rays to illuminate the routes the oxygen ions take through the best of fuel cell catalysts—cerium oxide—to create “snapshots” revealing just why it is such a superb material: In short, cerium oxide works because it is defective. “In this context, cerium oxide is missing oxygen atoms,” Chueh said. “For a fuel cell catalyst, that’s highly desirable.” Such “vacancies” allow for higher reactivity and quicker ion transport, translating

Assistant Professor William Chueh and his team used high-brilliance X-rays to track the process that fuel cells use to produce electricity.

into faster reaction rates and higher power. “We can now probe these vacancies to determine just how and to what degree they contribute to ion transfer,” Chueh said. “That has huge implications. We can make better catalysts and, ultimately, better fuel cells.” n

s ta n f o r d e n g i n e e r i n g

s u s ta i n a b i l i t y

Pomegranate design solves battery problem An electrode designed like a pomegranate—with silicon nanoparticles clustered like seeds in a tough carbon rind— overcomes several remaining obstacles to using silicon for a new generation of lithium-ion batteries, its inventors say. “This design brings us closer to using silicon anodes in smaller, lighter and more powerful batteries,” said Yi Cui, an associate professor of materials science and engineering who led the research. The anode, or negative electrode, is where energy is stored when a battery charges. Silicon anodes could store 10 times more charge than the graphite anodes in today’s rechargeable lithium-ion batteries, but the brittle silicon swells and falls apart during battery charging, and it reacts with the battery’s electrolyte to form gunk that degrades performance.

During the past eight years, Cui’s team has tackled the breakage problem by encasing the nanoparticles in carbon “yolk shells” that protect them as they swell and shrink during charging. In the latest development, graduate student Nian Liu and postdoctoral researcher Zhenda Lu gathered these silicon yolk shells into clusters, and coated each cluster with a second, thicker layer of carbon— like a pomegranate. Although the clusters are too small to see individually, together they form a fine black powder that can be used to form an anode. The carbon rinds hold the clusters together and provide a sturdy highway for electrical current. And since each pomegranate-like cluster has just one-tenth the surface area of the individual particles inside it, much less of the silicon is exposed to the elec-

trolyte, reducing the gunk that forms to a manageable level. “Our pomegranate-inspired anode operates at 97 percent capacity even after 1,000 cycles of charging and discharging—well within the desired range for commercial operation,” Cui said. n

Above left: Microscopic clusters form a fine black powder that can be coated on foil to create an anode. Above middle: A single cluster. Above right: A silicon nanoparticle can be seen inside its shell, with space to swell during battery charging. Below: A diagram shows how silicon nanoparticles swell during charging.

Nian Liu, Zhenda Lu and Yi Cui

After cycling

s ta n f o r d e n g i n e e r i n g

Brad Plummer/SLAC

Battery electrode heals itself for longer life Stanford researchers have made the first battery electrode that heals itself, opening a potential commercially viable path to the next generation of lithium-ion batteries for electric cars, cellphones and other devices. The secret is a stretchy polymer that coats and spontaneously heals tiny cracks in the electrodes that develop during charging and reduce battery life, says the team from Stanford Engineering and SLAC National Accelerator Laboratory. “Self-healing is very important for the survival and long lifetimes of animals and plants,” said Chao Wang, a postdoctoral researcher in the lab of chemical engineering Professor Zhenan Bao. “We want to incorporate this feature into lithium-ion batteries so they will have a long lifetime as well.” Wang added tiny nanoparticles of carbon to the polymer so it would conduct electricity. Then, to make the self-healing coating, scientists deliberately weakened some of the chemical bonds within stretchy polymers—long, chain-like molecules with many identical units. The resulting material breaks easily, but the broken ends are chemically drawn to each other and quickly re-link, mimicking the process that allows biological molecules such as DNA to break down, rearrange and reassemble. “Silicon electrodes lasted 10 times longer when coated with the self-healing polymer,” Bao said.

The development is an important step in a worldwide effort by researchers to find ways to store more energy in the negative electrodes of lithium-ion batteries, with the goal of achieving higher performance while reducing weight. One of the most promising materials is silicon; it has a high capacity for soaking up lithium ions from the electrolyte during charging. But it also has a tendency to crack. Researchers have tested many ways to keep silicon electrodes intact and improve their performance, but most involve exotic materials and fabrication techniques that would be challenging to scale up in mass production. The self-healing electrode, however, is made from silicon microparticles that are widely used and is the first solution that seems to offer a practical road forward,

said Yi Cui, an associate professor of materials science and engineering who led the research with Bao. The electrodes worked for about 100 charge-discharge cycles without significantly losing their energy storage capacity. “That’s still quite a way from the goal for cellphones and electric vehicles, but the promise is there,” Cui said. n

Above, from left: Professor Zhenan Bao, postdoctoral researcher Chao Wang and Associate Professor Yi Cui. Below: A prototype lithium-ion battery contains a silicon electrode protected with a coating of self-healing polymer.

s ta n f o r d e n g i n e e r i n g

Faculty awards & honors

Chris Gerdes (ME) • Rudolf Kalman Best Paper Award, Journal of Dynamic Systems, Measurement, and Control, American Society of Mechanical Engineers

Zhenan Bao (ChemE) • 2014 Fellow, Materials Research Society • 2014 POLY Fellow, American Chemical Society Kenneth Goodson (ME) • Fellow, American Association for the Advancement of Science • 2014 Heat Transfer Memorial Award, American Society of Engineers Jennifer Dionne (MSE) • Presidential Early Career Award for Scientists and Engineers

Stephen Boyd (EE) • Member, National Academy of Engineering

Pat Hanrahan (CS, EE) • Technical Achievement Award, Academy of Motion Picture Arts and Sciences Charbel Farhat (AA) • 2014 Gauss-Newton Medal, International Association for Computational Mechanics Sigrid Close (AA) • Presidential Early Career Award for Scientists and Engineers

s ta n f o r d e n g i n e e r i n g

Edward Feigenbaum (CS, Emeritus) • 2014 George R. Stibitz Computer and Communications Pioneer Award, American Computer and Robotics Museums

Mark Jacobson (CEE) • 2013 Policy Design Award, Global Green USA • 2013 Atmospheric Sciences Ascent Award, American Geophysical Union

Sachin Katti (CS, EE) • Rising Star Award, Association for Computing Machinery Special Interest Group on Data Communications

Michael May (MS&E, Emeritus) • 2014 Joseph A. Burton Forum Award, American Physical Society

Oussama Khatib (CS) • 2014 George Saridis Leadership Award in Robotics and Automation, IEEE Robotics & Automation Society

Mehran Sahami (CS) • Presidential Award, Association for Computing Machinery Manu Prakash (BioE) • TR35 Innovators Under 35, MIT Technology Review

Yinyu Ye (MS&E) • 2014 Optimization Prize, Society for Industrial and Applied Mathematics

Anne Kiremidjian (CEE) • Distinguished Member, American Society of Civil Engineers

Stephen Quake (BioE) • Member, American Academy of Arts and Sciences

David Lentink (ME) • Top 40 Under 40 Young Scientist, World Economic Forum

Xiaolin Zheng (ME) • TR35 Innovators Under 35, MIT Technology Review • National Geographic Emerging Explorer

Phil Levis (CS, EE) • Okawa Foundation Award Percy Liang (CS) • 2014 Microsoft Research Faculty Fellow Chris Manning (CS) • 2013 Fellow, Association for Computing Machinery

Mendel Rosenblum (CS, EE) • Member, American Academy of Arts and Sciences

s ta n f o r d e n g i n e e r i n g

2014 Stanford Engineering Heroes The Stanford Engineering Heroes program recognizes the achievements of Stanford engineers who have profoundly advanced the course of human, social and economic progress through engineering. Because engineers often work behind the scenes, the Heroes program's objective is to highlight the profound effect engineering has on our everyday lives and to inspire the next generation of engineers. Twenty-nine people—selected from among alumni and former faculty by a panel of distinguished subject-matter experts and technology historians—have been named Heroes since the program began in 2010. Kenneth Arrow shared the 1972 Nobel Memorial Prize in Economic Sciences with Sir John Hicks for pioneering contributions to general equilibrium theory and welfare theory—theories underlying the assessment of business risk and government economic and welfare policies. One of the 20th century’s most influential economists, Arrow played a major role in the School of Engineering by helping to create and serving as a seminal faculty member in the Department of Operations Research—now part of Management Science and Engineering. Arrow has written on topics that include the economics of racial discrimination, malaria drugs, climate change and innovation. In a paper written more than 50 years before health care reform, he noted that markets do not work in health care because patients lack the information to evaluate the quality of the services they are receiving. This work changed how people think about health care and launched the field of health care economics. He is also the first economist to apply the learning curve to understand the role of experience in increasing productivity. He has shown that under certain conditions an economy reaches a general equilibrium. Arrow is the Joan Kenney Professor of Economics, Emeritus, and professor emeritus of operations research at Stanford. He has served on the economics faculties of the University of Chicago, Harvard University and Stanford. He has received the American Economic Association’s John Bates Clark Medal, and is a member of the National Academy of Sciences and the Institute of Medicine. Sergey Brin co-founded web-search giant Google Inc. in 1998 with fellow Stanford graduate student Larry Page. A key innovation behind the company was their “PageRank” algorithm that calculated the relevance of a web page to the user’s query based in part on the number of other pages that linked to it. Today, Brin directs Google’s special projects, developing Glass and driverless cars. Brin has a bachelor’s degree with honors in mathematics and computer science from the University of Maryland at College Park and a master’s degree in computer science from Stanford. He is a member of the National Academy of Engineering, a fellow of the American Academy of Arts and Sciences and a recipient of a National Science Foundation Graduate Fellowship. He has received the Marconi Prize, given to those who achieve advances in communications and information technology for the social, economic and cultural development of all humanity.

s ta n f o r d e n g i n e e r i n g

Irmgard Flügge-Lotz (1903-1974), Stanford’s first female professor of engineering, was internationally renowned for her many important contributions to aerodynamics and to automatic control theory. She made significant advancements in methods for the prediction of aerodynamic pressures on bodies, wings and turbine blades, some of which were adopted as standard procedures throughout the world. In automatic control theory, she developed the first theory of discontinuous, or on-off, control systems. A professor emerita of applied mechanic s and of aeronautic s and astronautic s, she was the first woman elected as a fellow by the American Institute of Aeronautics and Astronautics and received the Achievement Award from the Society of Women Engineers. She was also a senior member of the Institute of Electrical and Electronics Engineers, a member of Sigma Xi and a member of the advisory boards of several scientific journals. Flügge-Lotz published more than 50 technical papers and wrote two books. She received a diploma of engineering and a doctor of engineering degree from Technische Hochschule in Hanover, Germany. Edward Ginzton (1915-1998), co-founder of Varian Associates, was a pioneer in development of the klystron radio tube for use in radar and linear accelerators. During World War II, Ginzton worked with a Stanford team hired to employ the klystron in radar, which played an important role in the war. Ginzton later joined brothers Sigurd and Russell Varian, who invented the klystron, to form Varian Associates, which became the world leader in medical linear accelerators and played a major role in Silicon Valley’s early development. As a Stanford professor of electrical engineering and of applied physics, Ginzton led a Stanford team that designed the world’s most powerful particle accelerator. He received the IEEE Medal of Honor, and was a member of the National Academy of Engineering and the National Academy of Sciences. He earned bachelor’s and master’s degrees in electrical engineering from the University of California-Berkeley, and a doctorate in electrical engineering from Stanford.

Larry Page is chief executive officer and co-founder of Google Inc., the world’s dominant web-search company. While pursuing a PhD at Stanford, Page and fellow graduate student Sergey Brin developed a “PageRank” algorithm that calculated the relevance of a web page to the user’s query based in part on the number of other pages that linked to it. They launched Google in 1998 with Page as the company’s first CEO. From 2001 to 2011, Page was president of products, then resumed responsibility for day-to-day operations as CEO. Page holds a bachelor’s degree in engineering from the University of MichiganAnn Arbor and a master’s degree in computer science from Stanford. He is a member of the National Academy of Engineering and a fellow of the American Academy of Arts and Sciences. He has received the Marconi Prize, given to those who achieve advances in communications and information technology for the social, economic and cultural development of all humanity.

Sally Ride (1951-2012) was the first American woman to fly in space. She became widely known for her passionate advocacy for science, technology, engineering and math (STEM) education. She served on the commissions investigating the Challenger explosion in 1986 and the Columbia disaster in 2003. Ride was a professor of physics at the University of California-San Diego and director of the California Space Institute. She founded Sally Ride Science to motivate girls and boys to study science and to explore careers in STEM. She also co-wrote seven science books for children. Ride was a member of the President’s Committee of Advisors on Science and Technology, and the National Research Council’s Space Studies Board. She was a fellow of the American Physical Society, and was inducted into the National Women’s Hall of Fame and the Astronaut Hall of Fame. She received the Presidential Medal of Freedom posthumously in 2013. Ride earned bachelor’s degrees in physics and English, and master’s and doctoral degrees in physics, all from Stanford.

s ta n f o r d e n g i n e e r i n g

Financial Information This report covers the fiscal year that began September 1, 2013, and closed August 31, 2014. During this period the School of Engineering reported direct expenses of $359,332,440 (see chart below for breakdown). In addition to these direct expenses, the school incurred $49,451,964 in indirect costs on sponsored research and $2,343,220 in infrastructure costs for a total of $411,127,624 in consolidated expenses.

Expenses by Category

Student Aid $109,488,235

Faculty Salaries $58,706,626

Other Teaching $13,679,981

Total Direct Expenses $359,332,440

Research & Admin Staff Salaries $64,013,000

Equipment & Supplies $113,444,598

Sources of Funding

General Funds $78,000,686 Other $63,071,704

Non-Federal Grants $27,381,840

Total All Sources $359,332,440

Federal Grants & Contracts $110,659,081

Endowment $54,830,510 Gifts $25,388,619

2013-14 Research Volume

Total expenditures by agency in millions Federal $150,674,287.08

Breakdown of Federal Research

Department of Defense Total Direct Expenses $186,614,928.49

$59,571,414.46

National Institutes of Health

$35,131,212.93

National Science Foundation

$26,308,771.99

Department of Energy

$13,964,581.40

Other Federal

$11,309,397.74

National Aeronautics and Space Administration

Non-Federal $35,940,641.41

$4,388,908.56

10000000 20000000 30000000 40000000 50000000 60000000

2013-14 Gifts and Affiliate Fees Total gifts/fees received in millions (rounded) Gifts $83,320,000

Breakdown of Gifts

Living Individuals Total $99,130,000 Affiliate Revenues $15,810,000

s ta n f o r d e n g i n e e r i n g

$48,464,000

Foundations & Associations

$16,816,000

Corporations

$16,086,000

Bequests

$1,954,000 0

10000000 20000000 30000000 40000000 50000000

Profile of Stanford Engineering Graduates Historically, engineering majors have constituted 20 percent of all Stanford juniors and seniors, but in recent years an increasing number of undergraduates have completed degrees in the field. During the 2013-2014 academic year 33 percent of all baccalaureate degrees awarded at Stanford went to engineers. Computer Science is now the largest major in the university, and virtually every Stanford undergraduate now takes an introductory programming course. Graduate enrollment has remained steadier because the school controls how many students are admitted. During the 2013-2014 academic year 45 percent of all master’s level degrees conferred by Stanford went to

BS Degrees Conferred July 1, 2013 - June 30, 2014

MS Degrees Conferred July 1, 2013 - June 30, 2014

PhD Degrees Conferred July 1, 2013 - June 30, 2014

engineers, and 30 percent of all Stanford PhDs were awarded through the School of Engineering. The charts below show the number of engineering degrees conferred at the BS, MS and PhD levels. These results are broken down by the program of study, and reflect the male and female students in each category. A special note about the BS and MS degrees in Engineering: these are conferred in fields such as Product Design, Aeronautics and Astronautics, Biomechanical Engineering, Bioengineering and other specialized studies in which BS or MS degrees are not awarded directly.

Degree

Male

Female

Total

Chemical Engineering

Civil and Environmental Engineering

Computer Science

168

211

Electrical Engineering

Engineering

123

Individually Designed Major in Engineering

Management Science and Engineering

Materials Science and Engineering

Mechanical Engineering

Totals

388

155

Degree

Male

Female

Total

Aeronautics and Astronautics

Bioengineering

Chemical Engineering

544

Civil and Environmental Engineering

101

188

Computer Science

137

170

Electrical Engineering

132

180

Engineering

Institute for Computational and Mathematical Engineering

Management Science and Engineering

114

160

Materials Science and Engineering

Mechanical Engineering

112

142

Totals

726

295

1,021

Degree

Male

Female

Total

Aeronautics and Astronautics

512

Bioengineering

Chemical Engineering

Civil and Environmental Engineering

Computer Science

Electrical Engineering

109

Institute for Computational and Mathematical Engineering

Management Science and Engineering

Materials Science and Engineering

17 31

Mechanical Engineering

Totals

235

296

s ta n f o r d e n g i n e e r i n g

ENGINEERING Stanford University School of Engineering 475 Via Ortega Stanford, CA 94305-4121

s ta n f o r d e n g i n e e r i n g

NONPROFIT U.S. POSTAGE PAID LAWRENCE KS PERMIT NO. 116