INNOVATIONS

VIDEO: Cornell University researchers Hod Lipson and Michael Schmidt discuss their findings.

http://www.nsf.gov/news/news_videos.jsp?cntn_id=114495&media_id=64865&org=NSF

Using the digital mind that guides their self-repairing robot, researchers at Cornell University have created a computer program that uses raw observational data to tease out fundamental physical laws. The breakthrough may aid the discovery of new scientific truths, particularly for biological systems, that have until now eluded detection.

Reporting in the April 3, 2009, issue of Science, Cornell University Mechanical Engineering professor Hod Lipson and his doctoral student Michael Schmidt report that their algorithm can distill fundamental natural laws from mere observations of a swinging double pendulum and other simple systems.

Without any prior instruction about the laws of physics, geometry or kinematics, the algorithm driving the computer's number crunching was able to determine that the swinging, bouncing and oscillating of the devices arose from specific fundamental processes. 

The algorithm deciphered in hours the same Laws of Motion and other properties that took Isaac Newton and his successors centuries to realize.

The new breakthrough is not far removed from Lipson's earlier NSF CAREER award work to develop Starfish, a robot with a "self-image" that could repair itself when damaged.

"The way the robot managed to recover from damage was to create a dynamical model, a self-image," said Lipson.  "It then used that model to make predictions about itself."

A dynamical model is a mathematical representation of the way in which a system's components influence each other over time. Lipson and Schmidt realized that if a robot can create dynamical models from data about itself, why not attempt to model the surrounding world as well?

When Lipson and Schmidt experimented with that approach, they learned their algorithm was re-discovering laws that were well known to scientists and engineers, suggesting the algorithm should be able to help uncover new laws for data sets that are less well understood.

"What is fascinating is that in the same way a robot created a dynamical model of itself using robot pieces, we now can create models not from motors and joints, but from components of mathematical objects, like variables, symbols like + and -, and other mathematical operators and functions," said Lipson.

While the algorithm can work with almost any data set, for this experiment Lipson and Schmidt used motion-capture data of pendulums and oscillators--similar to the motion capture techniques used for movie special-effects. The researchers then fed the data to a computer running the new algorithm, a process modeled on the one driving their Starfish robot.

The computer began its analysis with a broad suite of mathematical building blocks, expressions that the computer could combine to recreate patterns in the data set. Using a computational process called symbolic regression, a process inspired by biological evolution, the computer then took the assemblage of expressions and competed them against each other to find matches that reflected the data. The goal was to find those aspects of the data that were invariant, that did not change from one observation to the next.

"When you look at a pendulum, for example, some things go up, some go down," said Lipson. "But to recognize that when something goes up another specific thing always go down to keep the total sum constant, this is a key to understanding the observations in a deeper sense--such as recognizing the laws of conservation."

The computer retained the mathematical expressions that were invariant and abandoned those that were not, leaving a set of expressions that matched the data set and predicted future behavior. Because such a process could find patterns that are merely coincidental, the new algorithm also contains a critical step that compares subcomponent expressions, evaluating invariant equations to show that they are meaningful and represent actual natural laws, proof that the results are truly predictive.

Ultimately, a human still has to take the final list of a dozen or so expressions and figure out what they reflect in reality--for example, which expressions are describing a motion or energy-conservation law, or something totally new. Humans are still critical to the process: the computer serves as a data miner to find the laws, but a human must interpret them and give them meaning.

"Physicists like Newton and Kepler could have used a computer running this algorithm to figure out the laws that explain a falling apple or the motion of the planets with just a few hours of computation," said Schmidt, "but a human still needs to pick the appropriate building-blocks and framework, as well as give words and interpretation to laws found by the computer."

In the future, Lipson and Schmidt plan to use the new approach for biological systems. Biology is notoriously complicated to model, and finding fundamental laws for such systems can be difficult. With the new algorithm, the enormous data sets researchers collect about biological systems may yield invariants, unchanging aspects that may reveal underlying fundamental laws.

For this study, Michael Schmidt was supported by an NSF GRFP fellowship. Hod Lipson was supported by NSF CAREER Award 0547376 and NSF Creative-IT grant 0757478.

Researchers can monitor drug distribution and perform medical diagnostics rapidly using a new 3D imaging technique

Microscopes have revolutionized the practice of science, especially in the fields of biology and medicine. Just a few hundred years ago, gaining the ability to study what was previously unobservable opened up an entirely new world. Today, imaging techniques remain indispensable to clinicians and researchers who regularly diagnose medical conditions and work to develop new treatments.

Test results can often take hours or even days because cells or tissues must be subjected to lengthy fixation and labeling processes, sometimes called staining, in order to visualize and distinguish cellular components. In addition to long processing times, staining procedures often include harsh treatments or conditions that alter the tissues themselves, making interpretation of results difficult.

A newly developed label-free imaging technique called stimulated Raman scattering (SRS) will likely revolutionize biomedical imaging in research and diagnostic laboratories.  A team lead by Sunney Xie at Harvard University reported this new technique in the December 19 issue of Science.

"It is a big step forward in terms of biology," said Xie. "SRS is a powerful imaging modality with widespread applications on many fronts of biology and medicine. This work compliments an earlier technique we developed with funding from the National Science Foundation, adding a new imaging modality to the vibrational microscopy field."

The key to this new chemical imaging technique is the use of two lasers with different frequencies. Researchers visualize samples by tuning the laser frequencies to match the vibrational frequency of a specific chemical bond. Each type of molecule within a sample, including nutrients or drugs, is detectable at a unique frequency. By combining sample data collected at numerous frequencies, researchers can produce a high-resolution 3D image of the sample. SRS microscopy represents a big gain in biomedical imaging because it avoids labor-intensive sample preparation and autofluorescence, or "background noise", associated with traditional fluorescence microscopy.

Xie is enthusiastic about the ways in which SRS imaging will facilitate progress in many fields. "Applications of SRS imaging range from mapping distribution of small metabolite and drug molecules in cells and tissues to medical diagnosis of cancer. Neuroimaging is another exciting area of application."

-NSF-

The National Science Foundation (NSF) is an independent federal agency that supports fundamental research and education across all fields of science and engineering, with an annual budget of $6.06 billion. NSF funds reach all 50 states through grants to over 1,900 universities and institutions. Each year, NSF receives about 45,000 competitive requests for funding, and makes over 11,500 new funding awards. NSF also awards over $400 million in professional and service contracts yearly.

 October 23, 2008-An international team of scientists recently performed the ultimate miniaturization of computer memory: storing information at the nucleus of an atom. The breakthrough is a key step in bringing to life quantum computers, devices based on the theory of quantum mechanics.

In the quantum world, objects such as atoms can exist simultaneously in multiple states--that is, they could literally be in two places at once, or possess a number of other seemingly mutually exclusive properties. Quantum computing is seen as a holy grail of computing because each individual piece of information, or 'bit', can have more than one value at once.

A bit is a fundamental unit of information, represented as a 0 or 1 in a normal digital computer. Putting bits together creates a code, which generates or processes information. However, a quantum bit, or qubit, could be both 1 and 0 at the same time. That means a single qubit has twice the power of a normal bit, and once qubits start interacting with each other, the processing power increases exponentially.

How to maneuver and control quantum bits of information has been a major focus of experimentation. Researchers have been testing ways to isolate a quantum bit from a noisy environment, protecting its delicate quantum information, while allowing it to interact with the outside world so that it can be manipulated and measured.

Supported in part by the National Science Foundation, a team of scientists reported a solution in this week's issue of the journal Nature. The scientists were from Princeton University in New Jersey, Oxford University in the United Kingdom and the Department of Energy's Lawrence Berkeley National Laboratory in California.

The team described a system that used both the electron and nucleus of a phosphorous atom embedded in a silicon crystal. Both the electron and nucleus behaved as tiny quantum magnets capable of storing quantum information.

Inside the silicon crystal, the electron cloud was more than a million times bigger than the atom's nucleus, with a magnetic field a thousand times stronger. The size of the electron cloud made it well-suited for manipulation and measurement, but not so good for storing information because of electron instability. To overcome the problem, researchers moved the information into the nucleus where it survived much longer.

"Nobody really knew how long a nucleus might hold quantum information in this system," said Steve Lyon, leader of the Princeton team. "With crystals painstakingly grown by the Berkeley team and very careful measurements, we were delighted to see memory times exceeding the threshold."

The international team has demonstrated that information stored in the nucleus has a lifetime of about 1 3/4 seconds. This is significant because before this technique was developed, the longest researchers could preserve quantum information in silicon was less than one-tenth of a second. Other researchers studying quantum computing recently calculated that if a quantum system could store information for at least one second, error correction techniques could then protect that data for an indefinite period of time.

"The electron acts as a middle-man between the nucleus and the outside world," said John Morton, a research fellow at Oxford's St. John's College. "It gives us a way to have our cake and eat it--fast processing speeds from the electron, and long memory times from the nucleus."

-NSF-

The National Science Foundation (NSF) is an independent federal agency that supports fundamental research and education across all fields of science and engineering, with an annual budget of $6.06 billion. NSF funds reach all 50 states through grants to over 1,900 universities and institutions. Each year, NSF receives about 45,000 competitive requests for funding, and makes over 11,500 new funding awards. NSF also awards over $400 million in professional and service contracts yearly. NSF Home Page: http://www.nsf.gov

PASADENA, Calif.--Researchers at the California Institute of Technology have turned science fiction into reality with their development of a super-compact high-resolution microscope, small enough to fit on a finger tip. This "microscopic microscope" operates without lenses but has the magnifying power of a top-quality optical microscope, can be used in the field to analyze blood samples for malaria or check water supplies for giardia and other pathogens, and can be mass-produced for around $10.

"The whole thing is truly compact--it could be put in a cell phone--and it can use just sunlight for illumination, which makes it very appealing for Third-World applications," says Changhuei Yang, assistant professor of electrical engineering and bioengineering at Caltech, who developed the device, dubbed an optofluidic microscope, along with his colleagues at Caltech.

The new instrument combines traditional computer-chip technology with microfluidics--the channeling of fluid flow at incredibly small scales. An entire optofluidic microscope chip is about the size of a quarter, although the part of the device that images objects is only the size of Washington's nose on that quarter.

"Our research is motivated by the fact that microscopes have been around since the 16th century, and yet their basic design has undergone very little change and has proven prohibitively expensive to miniaturize. Our new design operates on a different principle and allows us to do away with lenses and bulky optical elements," says Yang.

The fabrication of the microscopic chip is disarmingly simple. A layer of metal is coated onto a grid of charge-coupled device (CCD) sensor (the same sensors that are used in digital cameras). Then, a line of tiny holes, less than one-millionth of a meter in diameter, is punched into the metal, spaced five micrometers apart. Each hole corresponds to one pixel on the sensor array. A microfluidic channel, through which the liquid containing the sample to be analyzed will flow, is added on top of the metal and sensor array. The entire chip is illuminated from above; sunlight is sufficient.

When the sample is added, it flows--either by the simple force of gravity or drawn by an electric charge--horizontally across the line of holes in the metal. As cells or small organisms cross over the holes, one hole after another, the objects block the passage of light from above onto the sensor below. This produces a series of images, consisting of light and shadow, akin to the output of a pinhole camera.

Because the holes are slightly skewed, so that they create a diagonal line with respect to the direction of flow, the images overlap slightly. All of the images are then pieced together to create a surprisingly precise two-dimensional picture of the object.

Yang is now in discussion with biotech companies to mass-produce the chip. The platform into which the chip is integrated can vary depending upon the needs of the user. For example, health workers in rural areas could carry cheap, compact models to test individuals for malaria, and disposable versions could be carried into the battlefield. "We could build hundreds or thousands of optofluidic microscopes onto a single chip, which would allow many organisms to be imaged and analyzed at once," says Xiquan Cui, the lead graduate student on the project.

In the future, the microscope chips could be incorporated into devices that are implanted into the human body. "An implantable microscope analysis system can autonomously screen for and isolate rogue cancer cells in blood circulation, thus, providing important diagnostic information and helping arrest the spread of cancer," says Yang.

The paper, "Lensless high-resolution on-chip optofluidic microscopes for Caenorhabditis elegans and cell imaging," was published July 28 in the early online edition of the Proceedings of the National Academy of Sciences. Yang's coauthors are graduate students Xiquan Cui and Lap Man Lee; postdoctoral research associates Xin Heng and Weiwei Zhong; Paul W. Sternberg, the Thomas Hunt Morgan Professor of Biology and an Investigator with the Howard Hughes Medical Institute; and Demetri Psaltis, the Thomas G. Myers Professor of Electrical Engineering at Caltech.

The work was funded by DARPA's Center for Optofluidic Integration at Caltech, the Wallace Coulter Foundation, the National Science Foundation, and the National Institutes of Health.

Visit the Caltech Media Relations website at http://pr.caltech.edu/media.

Tailor-made for Electrophysiological Experiments -Axio Examiner Microscope System

JENA/Germany, 27.05.2008- Electrophysiological experiments can be performed with the Axio Examiner Fixed Stage Microscope from Carl Zeiss with considerably greater ease and with maximum flexibility and safety.

In the field of the neurosciences, Axio Examiner is particularly suitable for patch clamp experiments on nerve cells, examinations of brain sections and for measuring electrical signals on cells. Together with the ZEISS LSM 710 NLO Laser Scanning Microscope, it can be used as a multiphoton system. The connection of an AxioCam camera and the use of the AxioVision 4.7 software with a special physiology module make the quantitative evaluation of typical experiments very comfortable and convenient.

Axio Examiner has been designed so that complex experimental setups, e.g. with various stages, micropipettes or light sources are easy to prepare or modify and safe to use. This is ensured by the extremely stable stand design and by the slim design in the front area of the system, providing the user with optimum access to the specimens. There is also scope for configuration in the specimen area and in the selection of the contrasting method. In addition, motorized functions can also be remote-controlled via the docking station or the AxioVision software.

For configuring his or her specific Axio Examiner microscope system, the user has four upper parts, two lower parts and a large number of different components and motorization options at his or her disposal. Numerous interfaces allow flexible adaptation to the respective requirements, particularly if a system is being used by several people. The objective changer can be just as easily removed as the stage carrier and the condenser carrier. The specimen area is flexible and, depending on the system configuration, is extendable to a free working distance of up to 100 millimeters so that electrophysiological processes can be conveniently examined not only on living tissue sections and organs, but also on whole organisms. Settings can be changed without difficulty during the experiment as all relevant controls are arranged on the front of the system.

The optical design developed for Axio Examiner also offers maximum optical quality for transmitted light techniques and for advanced fluorescence applications. With the W N-ACHROPLAN and W Plan-APOCHROMAT series, water immersion objectives specially developed to meet the requirements of neuroscience are available for visible light and infrared. Various contrasting methods such as Dodt gradient contrast are easy to integrate. Depending on the specimen, this makes it possible to achieve the highest resolution, the best contrast and optimum structure recognition in deeper tissue layers.

In all motorized functions, the motors are automatically deactivated after the target position has been reached. In addition, the motors can be actively grounded. This guarantees that any existing voltage can also drain off.

The new Aquastop for condensers effectively protects the system against overflowing liquids.

Axio Examiner offers flexible possibilities for additional magnification, e.g. with the double camera tube featuring an integrated continuous zoom system or a special additional magnification turret with fixed magnification steps.

--TEAM 0.5, the world's best transmission electron microscope, has been installed at the National Center for Electron Microscopy--

BERKELEY, CA -- TEAM 0.5, the world's most powerful transmission electron microscope -- capable of producing images with half-angstrom resolution (half a ten-billionth of a meter), less than the diameter of a single hydrogen atom -- has been installed at the Department of Energy's National Center for Electron Microscopy (NCEM) at Lawrence Berkeley National Laboratory.

"We have beam down the column," announced Uli Dahmen of Berkeley Lab's Materials Sciences Division, who is head of NCEM and director of DOE's collaborative TEAM Project, when the TEAM 0.5 microscope first delivered its ultrabright electron beam at Berkeley Lab in late December.

The TEAM Project (TEAM stands for Transmission Electron Aberration-corrected Microscope) is led by Berkeley Lab in a collaboration with DOE's Argonne and Oak Ridge National Laboratories, the Frederick Seitz Materials Laboratory of the University of Illinois, and two private companies specializing in electron microscopy, the FEI Company headquartered in Portland, Oregon, and CEOS of Heidelberg, Germany.

Now that TEAM 0.5's basic systems are operational, additional components and facilities are being completed and tuned, including a state-of-the-art control room display that shows the sample under the microscope on a flat panel resembling a wide-screen, high-definition TV. After a long series of rigorous tests and adjustments, TEAM 0.5 will become available to outside users by October, 2008.

Where these two gold crystals meet they are joined by a complex arrangement of atoms, forming a nanobridge that accommodates their different orientations. The atoms are 2.3 angstroms apart.

Atom by Atom in 3-D

In preliminary tests at the FEI Company, before the TEAM 0.5 was shipped, NCEM's Christian Kisielowski tested the microscope's ability to resolve individual atoms and precisely locate their positions in three dimensions. He made a series of images of two gold crystals connected by a "nanobridge" only a few dozen atoms wide. From each exposure to the next, individual gold atoms could be seen changing positions.

To achieve this extraordinary resolution, TEAM 0.5 embodies technical advances that have only recently become possible, including ultra-stable electronics, improved aberration correctors, and an extremely bright electron source.

Spherical aberration degrades images, making points of light look like disks, and correcting it can make dramatic improvements to image resolution. (This was famously demonstrated in 1993, when spherical aberration in the Hubble Space Telescope's optical lenses was corrected in a special space mission.) In the case of electron microscopes, a series of multipole magnetic lenses of varying geometries shapes the electron beam.

"Correcting spherical aberration in an electron microscope has long been possible in theory," says Dahmen. "But only recently has it become practical," because today's stable electronics reduce drift and fast computers allow continuous adjustments in real time. Corrector technology has even become available commercially, says Dahmen, "but no off-the-shelf corrector can match TEAM 0.5's ability to compensate even higher-order aberrations."

Correcting spherical aberration makes it possible to use the TEAM 0.5 not only for broad-beam, "wide-angle" images but also for scanning transmission electron microscopy (STEM), in which the tightly focused electron beam is moved across the sample as a probe, capable of performing spectroscopy on one atom at a time -- an ideal way to precisely locate impurities in an otherwise homogeneous sample, such as individual dopant atoms in a semiconductor material.

Aberration correction is also essential for another advanced feature of TEAM 0.5: its ability to maintain high resolution with lower electron beam energies.

"Low-energy electrons have longer wavelengths, so they are harder to focus," Dahmen explains. "Aberration correction allows better than one-angstrom resolution with excellent contrast even at 80 kilovolts. This is important when you don't want to damage the sample with a high-energy beam -- in biological studies, for example."

It's not just high resolution that makes TEAM 0.5 the world's best microscope, Dahmen says. When all the electrons in the beam focus at the same plane, image contrast and signal-to-noise ratio improve tremendously.

"It's because the signal-to-noise ratio is so good that you can adjust focus atom by atom, with enough sensitivity to obtain information about the three-dimensional atomic structure of a single nanoparticle." Dahmen adds, "This brings us within reach of meeting the great challenge posed by the famous physicist Richard Feynman in 1959: the ability to analyze any chemical substance simply by looking to see where the atoms are."

The position of individual atoms in a structure can be determined by taking images at different angles, from which the computer reconstructs a 3-D tomograph of the sample, as in a CAT scan. To make this possible an innovative system capable of tilting and rotating the sample, and moving it up, down, or sideways under the electron beam, is also being developed at NCEM.

Much smaller than sample stages now in use, the new TEAM stage will be housed entirely inside the microscope column. Manipulating the sample by such methods as minute piezoelectric "crawlers" that change shape when electricity is applied, the new stage will be able to control and reproduce the sample's position and attitude with an accuracy of less than a billionth of a meter.

Installation of the new stage must await the next phase of the TEAM Project: the TEAM I microscope, due to be set up at NCEM early in 2009.

While TEAM 0.5 corrects spherical aberration in both the "probe" beam (the electron beam before it strikes the sample) and the image beam (after it exits the sample, but before it reaches the detector), TEAM I will also correct chromatic aberration in the image beam, which has never beeen accomplished before. Spherical aberration is caused by the shape of a lens; chromatic aberration results when a lens refracts light or electrons of different wavelengths (different colors or energies) at different angles.

"Correcting chromatic aberration is harder and takes more space," says Dahmen. "The chromatic aberration corrector will add two feet to the height of the TEAM I column. But the new configuration will also allow us to enlarge the gap between the pole pieces, into which the sample fits. In TEAM 0.5 this gap is only about two millimeters, so we have to use traditional outside-mounted sample stages, with limited space to manipulate the sample. In TEAM I the gap will be five millimeters; the sample stage will have much greater freedom of movement."

New Vistas In The Realm of The Small

TEAM 0.5 and TEAM I will be housed side by side at NCEM for some time, occupying the two multistory "silos" that until recently were the homes of the historic High-Voltage Electron Microscope and the Atomic Resolution Microscope, the most powerful microscopes in the world when NCEM was established in the early 1980s.

Ambitious as those microscopes were in their day, says TEAM's Project Manager, Peter Denes of the Engineering Division, "when the TEAM Project was launched in 2004, it was not quite clear if the goals could even be achieved. The electron microscopy community had never done a collaborative project like TEAM before, and certainly not with full DOE project-management rigor."

Says Denes, "Perhaps the biggest contributor to success was a series of scientific workshops that contributed to forming a converging opinion on what the next steps would be, and what would constitute success. That helped in getting everyone -- if not quite on the same page -- at least in the same book."

Dahmen agrees. "This is a big jump for the microscopy community. TEAM's success will open the door to other ambitious developments around the world."

Dahmen suggests at least two broad categories of researchers who will benefit from the powerful new electron microscopes: experts with sophisticated microscopy problems to solve, and scientists less familiar with electron microscopy but with a particular problem for which microscopy can provide the answer.

"For example, Jim Zuo at the University of Illinois is doing studies of electron diffraction from the surface of single nanoparticles," Dahmen says. "He sees evidence of surface contraction. But when we at NCEM do imaging of similar nanoparticles, we find that the surface is expanding. Jim looks forward to using the TEAM microscope because it can do diffraction and imaging of the same particle at the same time -- a grand experiment, and the only way to solve the apparent contradiction."

An example of a problem-solving nonspecialist, says Dahmen, might be a materials scientist who has created a new kind of nanostructure, such as a tetrapod semiconductor, and needs to know exactly where in this complex, three-dimensional shape the impurity atoms reside. "TEAM's ability to image the structure in 3-D through tomography and its ability to do spectroscopy with single-atom sensitivity can identify each kind of atom at each position in the structure. That has never been possible before."

The basic TEAM components of aberration correction, enhanced signal-to-noise ratio, single-atom sensitivity, and an ultrabright beam that can be used in both TEM and STEM modes -- all the while manipulating the sample in the beam -- are goals that until recently seemed at the very edge of technological daring. All are on track, and some have been solved ahead of schedule. The TEAM Project's continuing success, signaled by the installation of TEAM 0.5 at NCEM, has opened the possibility of numerous future advances in electron microscopy that were barely conceivable when TEAM was launched.

The multi-institutional TEAM project represents a new kind of distributed planning and cooperation for the electron microscope community, moving beyond the limited, incremental improvements of individual investigators and harnessing the power of collaboration. Argonne National Laboratory is leading the development of the chromatic-aberration corrector in close collaboration with CEOS in Heidelberg. The University of Illinois's Frederick Seitz Materials Laboratory is jointly developing the new piezoelectric-controlled sample stage with Berkeley Lab's NCEM, and Oak Ridge National Laboratory is helping to optimize the new probe corrector. NCEM acts as project leader to integrate the individual components into single instruments, in close collaboration with all other TEAM partners. The TEAM Project is sponsored by the U.S. Department of Energy's Office of Science. For more about the TEAM Project, visit http://ncem.lbl.gov/TEAM-project/.

###

Berkeley Lab is a U.S. Department of Energy national laboratory located in Berkeley, California. It conducts unclassified scientific research and is managed by the University of California. Visit our website at http://www.lbl.gov.