In the late 60s of the 20th century, a small number of researchers began to use the force generated by light to propel small objects. Over the next decade, the field developed laser cooling: a powerful technique that uses Doppler shifts to slow down objects. Over time, new laser cooling studies have been explored along two parallel orbitals, ions and atoms.
In many ways, ions have an early advantage. Because they have an electric charge, they can feel a strong electromagnetic force, and they can also be trapped in an electromagnetic trap at high temperatures, which is then cooled by a UV laser. By 1981, ion trapping technology had advanced to the point where individual ions could be captured and detected, and spectroscopy could be performed with unprecedented precision.
In contrast, atoms need to slow down before they can be imprisoned by the weaker forces exerted by light and magnetic fields. Still, by 1985, Bill Phillips and his colleagues were at the National Bureau of Standards in Gaithersburg, Maryland, using light to reduce the speed of beams of sodium atoms to almost no stop, and then imprisoning them in magnetic traps. On this basis, the main challenge in the future seems to be to carry out more efficient neutral atom capture and push the cooling process to the theoretical limit. Both of these efforts have been more successful than expected, and this is closely related to Arthur Ashkin of Bell Labs.
Fig.1 Photograph taken in the late 1980s with a cloud of sodium atoms observed by the Phillips team of the National Bureau of Standards in the region of six intersecting laser beams. Phillips shared the 1997 Nobel Prize in Physics for his contributions to laser cooling research
Bell Labs
The last time we saw Ashkin was in 1970, when he had just developed the technique of "optical tweezers", for which he was awarded the Nobel Prize in Physics almost 50 years later. By the late 1970s, he was working with colleagues at Bell Labs on atomic beams. By superimposing the laser with the atomic beam, Ashkin et al. demonstrated that the atomic beam can be focused or diverged by adjusting the laser frequency. Ashkin wanted to use this effect to achieve an "all-optical" capture of atoms (i.e., without the magnetic field captivity scheme of the Phillips group). Unfortunately, because the atomic beam instruments used at the time were built with plexiglass windows, they could not maintain a sufficiently low vacuum. When these externally leaking atoms and molecules collide with the atoms in the beam, the target atom is kicked out of the potential well. After several years of disappointing results, the leadership of Bell Labs became unhappy and forced Ashkin to change the direction of his research.
It was around this time that a young researcher (who claimed to be "adept at tackling difficult experiments" entered Bell Labs. His name was Steve Chu and he was intrigued by Ashkin's ideas. Together, they built an ultra-high vacuum system suitable for atom cooling and trapping, and added a device that slows down sodium atoms by quickly scanning the laser frequency to compensate for Doppler shifts due to changes in velocity. Later, this technique became known as "chirp cooling".
Fig.2 Steven Chu (left) in the United States and Claude Cohen-Gannoudji (right) in France solved the sub-Doppler cooling problem experimentally and theoretically in the 1980s
At this point, Steven Chu suggested using three pairs of perpendicular counter-beam lasers to pre-cool the atoms, all tuned to a frequency slightly below the atomic transition frequency. This scheme provides a cooling force in three dimensions at the same time, and no matter which way the atoms move, they feel a force that is opposite to the direction of their motion. Because of the similarity to the swimmer's dilemma in a viscous liquid, Steven Chu called it an "optical sticky mass".
In 1985, the Bell Labs team collected thousands of atoms from a chirp-cooled device and demonstrated optical clays that hold atoms in the overlapping region of a beam for about a tenth of a second (an eternity in atomic physics). In the optically clay region, the atoms are constantly absorbing and emitting cooling lasers, so they look like a diffuse glowing cloud. The total amount of light emitted can be used to measure the number of atoms.
Ashkin, Steven Chu, and their collaborators also estimated the temperature of atoms. They estimated the temperature by turning it off for a short time and then turning the laser back on, measuring the proportion of atoms remaining in the optical clay. During photoshutdown, the atomic cloud expands, and some atoms flee the optically clumped region as a result of the expansion. Using this escape rate, it is possible to calculate the temperature of the atom: about 240 micro-Kelvin – exactly the theoretical expected minimum value of the laser-cooled sodium atom.
Despite being highly viscous, optical clays are not potential wells. Although it slows down the atoms, they can still escape once they drift to the edge of the laser beam. In contrast, potential wells provide a position-dependent force that pushes the atoms back into the central region.
The easiest way to construct a potential well is to use a strongly focused laser beam, similar to the optical tweezers developed by Ashkin for capturing microscopic objects. Although the volume of the laser focal area is only a small fraction of the volume of the optical clay, there are still a large number of atoms that can be aggregated in such a potential well by the random diffusion of atoms in the clay. When they added a separate capture laser beam to the clime, the results were encouraging: a small bright spot appeared in the diffuse optical clay cloud, representing hundreds of atoms being captured. However, further development comes with technical challenges. The movement of the atomic energy level caused by the laser makes the trap trap possible on the one hand, and hinders the cooling process on the other: when the trapped laser moves the energy of the atom's ground state downward, it changes the equivalent detuning of the cooled laser. Using another laser or alternating between cooling and trapping lasers can increase the number of atoms captured, but at the cost of increasing the complexity of the system. To make further progress, physicists need colder atoms or better potential wells.
Proposal from France
Claude Cohen-Tannoudji and his team at the École Normale Supérieure in Paris are mainly working on the problem of laser cooling from a theoretical point of view. Jean Dalibard, who was a new PhD to the group at the time, remembers that they had studied the theoretical analysis of Ashkin and Jim Gordon ("a brilliant paper") and the achievable cooling limit temperature of laser cooling that was determined in 1977 by the Soviet duo (Vladilen Letokhov and Vladimir Minogin, and Boris D Pavlik).
This limiting temperature is called the Doppler cooling limit, and it stems from the random "kicking" that occurs when atoms absorb the cooled light and then re-emit photons. Dalibard was curious about how strict this "limit" really was, and looked for ways to keep the atom "in the dark" as much as possible. To do this, he used atomic properties that are not considered in standard Doppler cooling theory: the true atomic state is not a single energy state, but a collection of many sub-levels with the same energy but different angular momentum. These different sub-levels or momentum states change the energy in the presence of a magnetic field (Zeeman effect). Another complicating factor is that the polarization of the laser determines which sub-levels will absorb photons, with one polarization increasing the angular momentum of the atom while the other decreasing.
In this theory, Dalibard combines these sub-levels with a magnetic field, and laser equivalent detuning depends on the position of the atoms. As a result, an atom can only absorb a particular laser at a specific location where the combination of detuning, Doppler shift, and Zeeman shift is just right. In this way, Dalibard hopes to limit the ability of atoms to absorb light, believing that this may lower their extreme temperatures. After he calculated the negative result, he abandoned the idea. "I see that this is a potential well, but I am not looking for a potential well, but for a sub-Doppler cooling," he explained.
If it weren't for the visit of MIT physicist Dave Pritchard to Paris in 1986, the matter might have ended there. During the visit, he gave a presentation on the idea of producing larger volume potential wells, and finally said that other better proposals were welcome.
"I went to Dave and I said, 'Okay, I have an idea, but I'm not quite sure if it's better, but it does be different from yours,'" Dalibard recalled. Pritchard brought Dalibard's idea back to the United States, and in 1987 he and Steven Chu built the first magneto-optical trap (MOT). Dalibard was invited to be a co-author of the final paper, but he only happily accepted the acknowledgment in the acknowledgments.
MOT is of great importance for the development of laser cooling. It is a relatively simple device that requires only a single laser frequency and a relatively weak magnetic field to produce a strong trap. More important is its capacity, Steven Chu and Ashkin's first all-optical potential well contained hundreds of atoms, Phillips's first magnetic potential well contained thousands of atoms, but the first magneto-optical trap contained tens of millions of atoms. With the introduction of inexpensive diode lasers at the University of Colorado, the advent of MOTs triggered a rapid growth in the number of groups studying laser cooling around the world, and the pace of research accelerated considerably.
An unexpected discovery in Gaithersburg
When Pritchard and Steven Chu were building the first MOT, Phillips and his colleagues in Gaithersburg, Maryland, encountered a highly unusual problem in the study of optical cleats. Since the optical clay was so effective, the Phillips team decided to conduct a more systematic study, including the measurement of the temperature of the atomic cluster. The "release-recapture" method developed by Bell Labs' team had relatively high uncertainties, so the Phillips team tried a new approach. They place a detection light near the optical clay, and when the optical clay is turned off, the atoms fly away. The time it takes for them to reach the detection area can reflect their velocity, which measures their temperature.
Due to space constraints, the initial detection light was placed slightly above the clump area. For atoms moving at the Doppler limiting velocity, the signal should be visible. But when they tried this experiment, no atoms reached the detector. Eventually, they moved the detector position below the goomass, and that's when they saw a nice signal. The temperature was much cooler than expected. This begs the question: the Doppler cooling limit is 240 microKelvin, but this "time-of-flight" method measurement shows a temperature of 40 microkelvin.
This result seems to violate Murphy's Law, which states that "everything that can go wrong must go wrong," so they are reluctant to accept it right away. They remeasured the temperature using several different techniques, including an improved release-recapture technique, but they kept getting the same result: the atoms were much cooler than the theoretical assumptions might be.
In early 1988, Phillips contacted other groups that were researching laser cooling, asking them to check the temperatures they had experimented with. Steven Chu and Wieman were quick to confirm this surprising result: optical clays not only cooled the atoms, but also worked better than the theories predicted.
Over the hills
The Paris group did not conduct experiments, but Dalibard and Cohen-Tannoudji solved the problem theoretically by tracing the characteristics that Dalibard used in his early development of the MOT theory, namely multiple intraatomic states. The ground state of sodium has 5 sub-levels with the same energy, and the distribution of atoms between these states depends on the intensity and polarization of the light. The combination of 6 backpropagation beams creates a complex polarization distribution because the beams are combined in different ways at different locations within the optical clime. The atoms are constantly optically pumped into different distributions, prolonging the cooling process and allowing for lower temperatures.
By the summer of 1988, Dalibard and Cohen-Tannoudji had devised a graceful model to explain the cooling of Yadoppler. (Steven Chu independently came up with a similar result, which he recalls on the train between two meetings in Europe.) They considered a simplified atom with only two ground-state sublevels, traditionally labeled -1/2 and +1/2, illuminated by propagating beams of linearly polarized light perpendicular to each other, creating periodic optical shifts in space. When the -1/2 state atom moves towards a high-energy region, the atom is slowed down, like a ball rolling up a mountain. When the "top of the mountain" is reached, the light pump changes the atom to the +1/2 state, and is at the bottom of the potential energy, and the atom continues to climb the "hill" to lose energy, and the speed slows down, and so on.
This process of losing energy by constantly climbing the "mountain" has a vivid name: Dalibard and Cohen-Tannoudji call it the "Sisyphus cooling", derived from the king of Greek mythology, who was sentenced to forever push a boulder up the mountain, only to have the rock slip and leave and return to the bottom. The atoms in the optical clay are in a similar predicament, always climbing mountains and losing energy, only to optically pump them to the bottom and force them to start over.
Sisyphusian cooling theory makes specific predictions about minimum temperatures and how they rely on laser detuning and magnetic fields. These predictions were quickly confirmed by laboratories around the world. In the fall of 1989, the Journal of the Optical Society of America B published a special issue on laser cooling, which included the results of the experiments of the Phillips group, the Sisyphus theory in Paris, and the combination of experiments and theory by the Steven Chu group. Over the next decade, the special issue was recognized by students as the definitive explanation for laser cooling, with Cohen-Tannoudji, Steven Chu and Phillips sharing the 1997 Nobel Prize in Physics.
In the extreme case, the Sisyphus effect can cool the atoms to such an extent that they no longer have enough energy to climb a "mountain" and are confined to a tiny area of a single polarization. This limit is as strict as trapping ions, and also makes the two branches of laser cooling correspond to each other. By the beginning of the 90s of the 20th century, both trapped ions and neutral atoms could be cooled to a state where their quantum properties became significant: individual ions in potential wells, or atoms in "wells" created by Sisyphean cooling, could only exist in certain discrete energy states. These discrete states were soon measured in both systems. Today, they are an essential part of atom- and ion-based quantum computing.
Another interesting direction of research involves the potential well itself. These potential wells are formed when the beam interferes and naturally occurs periodically in large arrays at half-wavelength spacing of the laser. The periodic nature of these so-called "optical lattices" mimics the microstructure of solid matter, in which atoms play the role of electrons in the crystal lattice. This similarity makes the trapped ultracold atoms a useful platform for exploring condensed matter physics phenomena such as superconductivity.
However, to truly explore the superconductivity of cold atoms, the crystal lattice must be loaded with atoms with higher densities and lower temperatures than Sisyphus cooling can achieve. Achieving this will require another new set of tools and technologies, which will not only create analogues of known systems, but also completely new states of matter, thus opening a new chapter.
(浙江大学 颜波 编译自Chad Orzel. Physics World，2023，(11)：30)
#来点儿干货#