As a Co-Sponsor of the Norman F. Ramsey Prize 2020 of the American Physical Society, Menlo System congratulates Philip H. Bucksbaum. In this interview the American atomic physicist talks about the fascinating development of his research field – the interaction between light and matter.

Philip H. Bucksbaum was awarded with the Norman F. Ramsey Prize in Atomic, Molecular and Optical Physics 2020 by the American Physical Society. Philip H. Bucksbaum is Marguerite Blake Wilbur Professor in Natural Science in the Departments of Physics, Applied Physics, and Photon Science at Stanford University and the SLAC National Accelerator Laboratory. He was recognized with the prize for pioneering explorations of ultrafast strong field physics from the optical to the X-ray regime. As a Co-Sponsor of the Ramsey Prize Menlo System warmly congratulates Philip H. Bucksbaum. With science journalist Thorsten Naeser Philip Bucksbaum talked about his research, his aims and his visions as well.

Menlo Systems: Congratulations on winning the Norman F. Ramsey Prize 2020.

Philip Bucksbaum: Thanks, this is a wonderful honor.

An important part of your work deals with the interaction between laser light and matter. Can you give us a short overview over the research field?

The fundamental theory of light-matter interactions has been well-established for nearly a century. Much of what we know about the structure of atoms and molecules came from experiments like photoionization that use light as a weak probe. The theory predicts what we see about as well as any physical theory in history. 

But powerful laser sources interact in ways that can’t be understood so simply, because the fields in the laser light are no longer weak. Focused light from a modern ultrafast laser can overwhelm the internal binding of electrons to the nuclei, and literally pull electrons away, a process called field ionization. 

A number of new phenomena appear when electrons are field ionized, with names like above-threshold ionization, high harmonic generation, nonsequential multiple ionization. These retain some of the familiar quantum signatures of the photoelectric effect; but they also appear in many ways like classical plasmas inside atoms and molecules.

The timescale for electron motion in these strong fields is very fast – in fact, we use the word “ultrafast” to describe it. Over the past 25 years we have taken advantage of many advances in ultrafast laser source technology. One of these advances, chirped-pulse amplification, is well-known because it earned a Nobel Prize in Physics in 2018 for the inventors, Strickland and Mourou. These steady advances in laser technology have driven the duration of a table-top laser pulse well below the picosecond (trillionth of a second) range where we started in the 1980s through the femtosecond range, and now well into the range attosecond pulse duration. And this is important, because light traverses a molecule in about one attosecond, a billionth-billionth second; and electrons in molecules typically travel across the same distance in a few hundred attoseconds. So this means we can now study that electron motion directly.

Can you give us a short overview of how ultrafast measurements are made using these short pulses?

In my field of ultrafast AMO physics, short pulses are used in two ways, called “pump” and “probe”. The pump is the “starter’s pistol” to initiate motion in the quantum system we wish to observe; the probe is a “flashbulb” to illuminate the structure of the system at a known delay after the start pulse. The pump must be short to get the molecules off to a clean start. The probe pulse must be short as well, or else the data collected will be blurred by motion.

You have been recognized with the prize for pioneering explorations of ultrafast strong field physics from the optical to the X-ray regime. What impact does the extension of ultrafast strong field physics from the optical to the X-ray regime have on laser sciences?

We just discussed why short pulses are important, but the wavelength also matters for a couple of reasons. The most important is because every type of atom has a unique fingerprint of X-ray absorption, which corresponds to the threshold X-ray photon energy to ionize a deeply bound electron near the nucleus. For carbon atoms this is around 290 eV. For nitrogen, it’s 400 eV, and for oxygen it’s 530 eV, and so on through the periodic table. Ultrafast X-ray lasers are very important as atom-specific pump or probe pulses.

A second important role for ultrafast X-ray pulses is to image the location of the atoms in a molecule that is undergoing some internal motion. We can’t do this with optical or ultraviolet light because molecules are too small to resolve with those wavelengths. But X-ray lasers with angstrom wavelengths – corresponding to photon energies of 10 keV or so – can do this job. This is the way we can now make X-ray movies of the internal motion of molecules. It’s one of the more exciting new kinds of physical measurement made possible by ultrafast X-ray lasers, such as the LCLS at SLAC, where I work.

For a time you co-held the record for the shortest wavelength coherent radiation produced in the laboratory. How short was this wavelength?

When I was a first at Bell Labs, long before current sources of ultrafast X-rays were invented, my colleague and postdoctoral mentor Jeff Bokor decided we should go for the record in short-wavelength coherent light by creating intense short pulses of amplified ultraviolet light, and then focusing that light into dense helium gas to make use of nonlinear frequency multiplication. All the stuff I spoke about just a minute ago, with new phenomena like high harmonic generation, was still off in the future. So, we spent the better part of a year working with Ralph Storz, building a fairly complex chain of lasers, which started with a lamp-pumped infrared solid state laser, and added lots of nonlinear optics and amplifiers to do this task. It worked. We produced 35 nm light, the seventh harmonic of 248 nm amplified laser light, as well as lots of other kinds of vacuum ultraviolet light that we could use. 

And how did you utilize the radiation in experiments?

We set about using ultrafast VUV pulses to develop time-resolved ARPES, that is, time and angle-resolved photoemission spectroscopy. We weren’t surface physicists, but we hired a postdoc, Richard Haight, who knew all about traditional ARPES (not time-resolved) and we got lots of other advice from colleagues at Bell Labs, and within a year or two we were making measurements of electronic dispersion on well-characterized photoexcited surfaces of semiconductors – measurements that were not possible before. It was great fun, learning about the problems of surface physics.

Meanwhile, however, back at the source, we were about to be upstaged big-time. A discovery was made a few years later in strong-field ultrafast laser-atom physics, and not by us.  It turns out, if we had only focused our front-end infrared laser directly into the helium gas without all those other amplifiers and stuff, we would have discovered the phenomenon of high harmonic generation, and at the same time we would have made all the VUV light we needed for our new surface physics applications. This is a great example of how science doesn’t advance on a linear or uniform path. 

You developed broadband coherent THz radiation, the so-called "half-cycle pulses". What does this radiation look like and what is the main advantage of this radiation for THz spectroscopy?

I was introduced to THz radiation by a colleague at Bell Labs in Murray Hill, Ben Greene. Ben was doing something very simple: hitting the surface of a polished semiconductor wafer with a laser beam, and looking at the THz radiation that comes off. When I left Bell Labs for the University of Michigan, I continued this work with my postdoc Bob Jones. The wafer had two electrodes attached to the surface with a gap, so it was acting as the dielectric material in a capacitor. When a laser pulse excites photocurrent in the gap, it suddenly shorts out the voltage between the electrodes. The step-function change in the electric field propagates away from the gap in the form of a half-cycle pulse, that-is, a traveling wave of electric field that turns on and then turns off, but doesn’t oscillate the way ordinary radiation does. 

What did you use this radiation for?

We started to use this radiation to interrogate Rydberg atoms, which are atoms with one highly excited electron. We found that we could ionize Rydberg atoms with half-cycle pulses, but the ionization law worked in a completely new way. It wasn’t like ordinary photoemission and it also wasn’t like field ionization. Instead, the electron gets kicked by the half -cycle pulse, very much like a kicking a football. The ionization probability depends on the strength of the impulse delivered by the half cycle pulse and on the shape of the electron orbit, not on other properties like its central frequency compared to the electron binding energy, as in photoemission, or its peak field, as in field ionization. 

We then used half-cycle pulses to study lots of effects in Rydberg atoms, and I could go on for a long time about that; but my favorite use for these pulses was a quantum register decoder. My student Haidan Wen and a postdoc Chitra Rangan found we could encode a number in the quantum phase of a Rydberg electron wave packet, and then use the half-cycle pulse to decode it by turning the phase-encoded number into amplitude information. As far as I know this method of storing information in quantum phase is not on anyone’s critical path to building quantum computer, nor should it be; but some people are still interested in using Rydberg atoms for quantum technologies, so you never know what might be useful in the future.

You are interested in how ultrafast lasers can control molecular dynamics, on ultrafast time scales. Can you give us a short overview why such short times are interesting?

The atomic geometry in a molecule, even a very simple molecule, can move in complex ways during a chemical transformation. This motion is governed by the forces between the atoms, which in turn is determined by the charge distribution of the electrons in the molecular bonds. Intense short pulses of light can move the electrons, as I’ve just described. Therefore it stands to reason that coherent light should be able to control the internal motion of the molecule. 

This simple notion, which goes back to the earliest days of ultrafast science, has been very difficult to use in practice; but it still motivates a lot of research today. Because the problem is so complex, we developed lasers with computer programmable pulse shapes, and then built a search program into the experimental protocol so that the experiment could find its own optimal pulse by systematically improving the pulse shape guided by the success of the previous shapes. These days we would call this machine learning.

You also use attosecond pulses in your labs. Can you explain how this technology helps you doing research?

Electrons move across molecular bonds in a few hundred attoseconds. My colleagues James Cryan and Ago Marinelli in our PULSE Institute at Stanford have developed ways to shorten the X-ray laser pulses to less than a femtosecond and use them for ultrafast measurements. We also know how to make attosecond pulses with high harmonics, which are far weaker than X-ray laser pulses, but can be even shorter. These days we use both kinds of attosecond pulses to clock the motion of electrons. For example, during photoemission the time it takes for an electron to leave the molecule depends on its path through the nuclei and other electrons that are staying behind, and we can see that.

And last but not least, can you give us a short impression how the always growing knowledge of light-matter interactions change the world in, let’s say, 20 years?

Changing the world is a tall order, but it’s a pretty safe bet that high-powered lasers will continue to improve, and that new scientific discoveries will follow. We can already make focused laser fields that are many thousands of times stronger than needed for field ionization of atoms. And we can make use of even stronger fields by backscattering this laser from ultra-relativistic electrons made in the SLAC linac. I’ve been working with colleagues David Reis and Sebastian Meuren at SLAC to set up an experiment to do this now, as are other groups around the world working at other accelerators. Our goal is to collide the backscattered laser light with forward traveling light to create combined superstrong laser fields that can probe the vacuum itself. Let me explain: Physicists know that the vacuum isn’t really empty. It’s filled with so-called virtual charged particles. With enough laser intensity we can field ionize these charges, so that they come pouring out of the vacuum.

Author: Thorsten Naeser

Former Stanford University graduate student Jahee Kim and Phil Bucksbaum
(Photo by Peter Ginter courtesy of SLAC National Accelerator Laboratory)

Ramsey Prize: https://www.aps.org/programs/honors/prizes/ramsey.cfm