Public Release: 

Rudimentary Atom Laser Created At MIT

American Institute of Physics

What Is The News?
--Physicists at MIT have built an elementary version of a laser that yields beams of atoms rather than beams of light.
--The MIT researchers have verified that their atom beam has an important propterty known as "coherence," which is a key attribute of optical laser beams.
--In addition to promising dramatic improvements in precision measurements at the quantum level, the "atom laser" is likely to lead to significant advances in nanotechnology, the fabrication of structures and machines at the single-atom scale.

College Park, Maryland---January 23, 1997--No one could have foreseen all of the laser's applications when the first working model was built in 1960, but it went on to revolutionize telecommunications, medicine, information storage, and many other areas of science and technology. Get ready for a new revolution. Physicists at MIT have created a rudimentary version of an "atom laser," a device that does for matter what ordinary lasers do for light. In a pair of upcoming papers, the MIT researchers describe their device and verify a key property, known as coherence, essential to demonstrating that they have a bonafide atom-laser beam. In addition to dramatically improving measurements with atoms, the atom laser may lead to major innovations in nanotechnology, the manipulation of matter at the atomic level.

An ordinary laser produces an extremely special form of light vastly different from that which emerges from a light bulb or even the Sun. Unlike all other light sources, the laser creates photons (particles of light) that are in exactly the same quantum state. Laser light is also coherent, meaning that its wavefront varies predictably in time and space. Light from a lightbulb, in contrast, is incoherent: later wavefronts have no predictable relationship to earlier ones. In most conventional laser designs, light builds up inside the laser by reflecting many times from mirrors at either end of the laser. Extracting the light is possible because one of the mirrors is only partially silvered and lets some of the light escape to produce an output beam. The beam travels in a single, well-defined direction, unlike sunlight or lamplight, which shines in all directions. This combination of unique properties makes a laser beam more intense than an equivalent stream of light emanating from the Sun.

Since 1995, scientists have been making the main ingredient for an atom laser: a Bose-Einstein condensate, a collection of gas atoms with temperatures just billionths of a degree above absolute zero, colder than anything observed before. In the 1920s, Albert Einstein and the Indian physicist Satyendra Nath Bose showed that when a sufficiently densely packed group of particles becomes very cold they would, under certain conditions, collectively enter a single quantum state and act as a single, coherent wave, central requirements for a laserlike beam of atoms. Although previous experiments provided clear evidence for the formation of a Bose-Einstein condensate, they could not directly verify that the condensate formed a single coherent wave. In addition, they were unable to extract a beam of atoms from the Bose-Einstein condensate, which exists as a fragile clump of atoms trapped by magnetic fields. In a paper to be published in the 27 January issue of Physical Review Letters, Wolfgang Ketterle and his colleagues at MIT have created an "output coupler" which allows them to pluck a controlled fraction of atoms from a BEC of sodium atoms to produce a beam that falls in the direction of gravity. Meanwhile, in a paper to appear in the 31 January issue of Science, the same group demonstrates that the beam has coherence properties analogous to those of a laser light beam. Demonstrating coherence was "the most important step," says Ketterle, "to show that a Bose condensate with an output coupler acts as an atom laser."

The MIT device delivers coherent bursts of atoms that form a single quantum state--a feat never even remotely approached in previous atomic beams. Whereas the atoms in conventional beams are distributed over a wide range of quantum states, those in an atom laser occupy a single state. And whereas most atom beams can only be considered as a collection of particles, the MIT atom beam can act as a single "matter wave" which can be manipulated and controlled in important ways just like light waves from a laser. Indeed, these matter waves can be split up into wavelets and recombined to display interference patterns which provide highly precise information on the atoms. Like an optical laser, the MIT atom beam is bright, in the special sense that it delivers a powerful, directional stream of atoms in a single quantum state, rather than a beam of atoms that are dispersed over many different quantum states.

In the upcoming Physical Review Letters paper, Ketterle and his colleagues describe their "output coupler," which allows them to extract a controlled fraction of atoms from a Bose-Einstein condensate of sodium atoms. The researchers extract atoms by applying radiofrequency (rf) radiation to the BEC, which is held in a magnetic trap. (A figure showing how the output coupler works will be available at Each of the atoms acts as a tiny magnet in the presence of the trap's magnetic fields. The atom possesses a property known as "spin"; the value of spin describes how it will respond to an external magnetic field. Initially, the atoms all have the same spin value, corresponding to a state in which they are all pushed towards the center of the trap. But the rf radiation, which contains magnetic fields of its own, will make some of the atoms "flip" their spin. This reverses the magnetic forces on them, and these atoms are expelled from the trap, forming a beam that falls in the direction of gravity.

MIT's Wolfgang Ketterle explains some of the differences between a laser producing beams of atoms and a laser producing beams of photons. "Photons can be created, but not atoms," he says. "The number of atoms in an atom laser is not amplified. What is amplified is the number of atoms in the lowest-energy quantum state, while the number of atoms in other states decreases." In addition, he says, "Atoms interact with each other--this creates additional spreading of the atom-laser beam. Unlike light, an atom-laser beam cannot travel far through air." Unlike photons, which are massless, "atoms have mass," Ketterle points out. "They are therefore accelerated by gravity. An atom-laser beam will fall like a beam of ordinary atoms."

The foundation of the atom laser rests on an astonishing discovery that French physicist Louis de Broglie (1892-1987) made as a Ph.D. student. We usually think of atoms as solid objects having definite locations in space. But in his 1923 Ph.D. thesis, de Broglie hypothesized that atoms--and all matter in general--can also act as waves that spread out in space and combine with other waves to produce interference patterns and exhibit other wavelike phenomena. De Broglie introduced a famous formula which states that the wavelength of a particle is inversely proportional to the product of its mass and its speed. De Broglie's equation and the wavelike nature of matter have been confirmed in countless physics experiments that have followed.

In the 1950s, Charles Townes of Columbia University and Arthur Schawlow, then at Bell Laboratories, build the first maser, a laser for microwave light. The maser was the precursor to the optical laser. Theodore Maiman of Hughes Aircraft Corporation created the first working optical laser in 1960. At the time of its invention, most physicists could not foresee any practical applications for the laser. Yet, optical lasers now make up a multibillion dollar industry, having profoundly changed telecommunications, data storage, and numerous areas of medicine and surgery.

In July, 1995, Eric Cornell, Carl Wieman, and their coworkers at the National Institute of Standards and Technology and the University of Colorado created the first Bose-Einstein condensate in a dilute gas of atoms. Shortly thereafter, Randall Hulet and his coworkers at Rice University, and Wolfgang Ketterle and his colleagues at MIT produced Bose-Einstein condensates of their own. In the present work, Ketterle and his colleagues have created a technique for extracting a beam of atoms from the Bose-Einstein condensate and for verifying that their beam has properties directly analogous to an optical laser beam.

Ketterle and his colleagues have shown that their atom beam possesses important wavelike properties. The atoms in the beam are so cold that the ordinarily miniscule wavelengths associated with atoms increase to the point at which their wavelike properties become potentially detectable. A true atom-laser beam would by definition be "coherent"--the individual waves of the atoms would combine in a consistent, predictable fashion to form a single intense wave. An atom wave is a quantum-mechanical wave which can never be observed directly, but only when it combines or interferes with similar quantum-mechanical waves to produce a pattern of light and dark fringes. When two atom waves interfere, an atom making up one wave can cancel out an atom from the other wave. "When matter waves interfere destructively," explains Ketterle, "it is as if one atom plus one atom gives zero atoms. Of course, the matter is not destroyed, and appears elsewhere. Nevertheless, the interference of streams of atoms from separate sources is a dramatic phenomenon."

In their upcoming Science paper, the MIT researchers showed that the Bose-Einstein condensate--from which their atom beam is extracted--is coherent, something that has never demonstrated before. The researchers observed coherence by creating two Bose condensates in a special trap which uses magnetic and optical forces and has two separate pockets. When the trap is switched off, the two condensates fell down, expanded, and finally overlapped. The overlapping produced an interference pattern of light and dark fringes--like a pattern of a zebra stripes--which was detected with an electronic camera. Such a high-contrast pattern is possible only if the atoms in each condensate formed intense single waves--with the overall atom wave of one condensate interfering with the atom wave of the other condensate to produce a fringe pattern. The MIT team determined that the atom wave associated with each Bose-Einstein condensate had a wavelength of 30 microns, a million times larger than the wavelengths associated with room-temperature atoms.

The MIT scientists report that they have also demonstrated coherence in the beams of atoms that are produced by their output coupler. The researchers again create two independent Bose-Einstein condensates, then produce a beam of atoms from each condensate. The two beams fall down, expand and create interference patterns recorded by a camera.

Work on the atom laser has just begun. Ketterle and his colleagues are already thinking of improvements to their device. For instance, the atom laser currently only produces a beam that falls in the direction of gravity. Future steps are to make the atom beam travel in other directions by combining the output coupler with "atom mirrors" that use optical or magnetic forces to direct the atom. In addition, the MIT atom beam diffracts, or spreads out, somewhat as it emerges from the trap; upcoming designs may reduce such diffraction effects. Finally, the current design produces only bursts of atoms; future designs may produce continuous beams.

Just as the laser has greatly improved optical experiments, the atom laser promises to increase the precision of many atomic beam experiments. For instance, it may be possible to improve dramatically the already impressive precision of atomic clocks. It may enable more powerful tabletop tests aiming to determine the relationships between the fundamental forces in nature. It is likely to increase the precision in measurements of fundamental physical constants. In particular, it promises to improve the technique known as atom interferometry in which an atom from a conventional beam is split into wavelets and recombined to form interference patterns which provide precise information about the atom. Using atoms from atom-laser beams will greatly improve the technique, which is already rivaling conventional optical interferometers for measuring Earth's rotation and testing relativity.

Finally, the atom laser holds exciting possibilities for nanotechnology, the manipulation of matter at the atomic level. With an atom laser beam, it would be possible to deposit atoms onto surfaces with unprecedented precision, potentially allowing scientists to create more sophisticated nanostructures than ever before. Ketterle points out, however, that these first atom lasers will only be able to make nanostructures at a very slow rate. According to Ketterle, the fluxes of atoms emerging from an atom laser are currently too small to lead to a practical nanofabrication scheme in which nanostructures could be mass produced. And he notes that the atom laser must operate in extreme vacuum conditions, unlike ordinary lasers whose light can be used in all types of environments. Nonetheless, the atom laser has the potential to become a tool with unexpected and widespread consequences.

# # #

For graphical illustrations associated with this work, figures will be available starting on Friday, January 24 at the following private Web site:

List of Experts
Wolfgang Ketterle, MIT--617-253-6815;
Michael Andrews, MIT--617-253-2518
Eric Cornell, NIST--303-492-6281
Carl Wieman, NIST--303-492-6963
Dan Kleppner, MIT--617-253-6811
David Pritchard, MIT--617-253-6812
Randall Hulet, Rice University--713-527-6087
William Phillips, NIST--301-975-6554
Keith Burnett, Oxford University--011-44-865-272377
Robert Ballagh, Otago University, New Zealand--011-64-03-4797793,

SPECIAL NOTE: Wolfgang Ketterle is currently out of the country, but will return to his office on Monday, January 27.

Journal Articles:
M.-O. Mewes, M.R. Andrews, D.M. Kurn, D.S. Durfee, C.G. Townsend, and W. Ketterle, "An Output Coupler for Bose Condensed Atoms," Physical Review Letters 78, 582, 27 January 1997.
M.R. Andrews, C.G. Townsend, H.-J. Miesner, D.S. Durfee, D.M. Kurn, and W. Ketterle, "Observation of Interference between two Bose Condensates," Science, 31 January 1997.

Earlier articles of interest:
M. R. Andrews, M.-O. Mewes, N. J. van Druten, D. S. Durfee, D. M. Kurn, W. Ketterle, Science, "Direct, Nondestructive Observation of a Bose Condensate," Science 273, 84, 5 July 1996
M.H. Anderson, J.R. Ensher, M.R. Matthews, C.E. Wieman, and E.A. Cornell, "Observation of Bose-Einstein Condensation in a Dilute Atomic Vapor," Science 269, 198 (1995).

Magazine articles

Some Previous Accounts of MIT Group's Work
New Scientist, 1 June 1996
Physics Today, March 1996, August 1996
Physics World, October 1996
Science, 3 November 1995; 14 June 1996
Science News, 25 May 1996

Previous News Articles on Bose-Einstein Condensation
Physics Today, August 1995
Physics World, August 1995
Scientific American, August 1995
Science, 8 July 1994, 14 July 1995

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.