Seeing Light’s Effect on Atoms — Within Picoseconds

For the first time in a laboratory setting, NIST scientists have made stop-action x-ray measurements of the way visible light interacts with atoms and molecules on near-instantaneous time scales.

Using an innovative tabletop system — described in a recent issue of Physical Review X (first-authored by NIST scientist Luis Miaja-Avila) — the researchers obtained results with 10 times better time resolution than is available at large x-ray synchrotron facilities costing hundreds of millions of dollars, and collected x-rays with 100 to 1000 times better efficiency than current techniques.

“Chemical reactions driven by light are very important in the natural world and in new technologies for solar energy harvesting such as photovoltaics and the production of biofuels,” says Joel Ullom, leader of the Quantum Sensors Group in NIST’s Physical Measurement Laboratory that developed the new system with colleagues at the University of Colorado.

A Table-Top Setup for X-Ray Emission Spectroscopy

“The beginning and end states of reactions are often known, but not the pathways from start to finish,” Ullom says. “Understanding the ultrafast electronic and atomic movements that comprise photoactivity is a frontier of modern science and the development of techniques to directly observe these movements is an important first step.”

The system is used to investigate complex processes triggered by light in both inorganic and biological material. Such processes – from the generation of oxygen in photosynthesis to the generation of electricity in a photovoltaic cell – typically occur in ultra-fast cascades of electron transfers and atomic movements lasting only a tiny fraction of a nanosecond (ns, 10^-9 s).

Measuring reaction dynamics within such narrow time intervals has been difficult because time resolution better than about 60 picoseconds (ps, 10^-12 s) could only be achieved at one of the world’s two most powerful x-ray free-electron lasers. But the NIST researchers, with colleagues at the University of Colorado, have shown that their tabletop system can resolve time intervals as small as 6 ps.

That can have significant practical impact for probing the quantum states of complex molecular systems. “Various research groups are focusing on the problem of improving the efficiency of solar cells by finding materials that promote power conversion efficiency,” Ullom says. “Often these groups perform experiments in the optical or near optical part of the spectrum and they have to make assumptions about how optical properties are related to the underlying electronic states and atomic configurations.

“X-rays, on the other hand, are element-specific and highly sensitive to both atomic and electronic structure. Until now, most time-resolved experiments have been performed in large facilities such as synchrotrons or free-electron lasers. Our apparatus is the first to perform a laboratory-based time-resolved x-ray emission spectroscopy experiment. We believe we can study the ultrafast dynamics of photoactive materials with an apparatus that can be operated in a conventional laboratory at much lower cost.”

“The characteristic size and power consumption of our apparatus are meters and kilowatts, whereas the same quantities for large x-ray facilities are hundreds of meters and megawatts” says Galen O’Neil, one of the NIST scientists who performed the experiment. “Our apparatus thus has the potential to allow more routine study of actuation pathways and time scales in a wide range of materials.”

The method — time-resolved x-ray emission spectroscopy (TR-XES) — requires high-energy (“hard”) x-rays delivered, with exquisitely accurate timing, in a two-stage, “pump-probe” process. First, a beam of visible light (the pump beam) is directed onto the sample material to start a photoreaction. Then a beam of hard x-rays (the probe beam) is directed onto the same location.

The pump beam initiates a cascade of changes in the sample. Atoms in the sample then absorb the incident probe-beam x-rays and in turn fluoresce, emitting x-rays of different energies. The properties of those emitted x-rays reveal exactly how the visible light altered the internal states of atoms and electrons in the sample at that instant. Changing the timing delay between pump and probe shows reactions at different stages of the sequence, much as a stroboscope produces a series of “snapshots” of objects in motion.

By measuring the spectrum of emitted x-rays, investigators can identify and distinguish between very specific electronic and chemical transitions, and determine which molecules perform best in applications such as optoelectronics, photovoltaics, photocatalysis, energy and data storage, and optical display technologies.

How It Works

The NIST apparatus employs a highly unusual system for generating its visible-light and x-ray beams. It begins with the pulsed output (35 fs duration) from a near-infrared laser (800 nm) that is split into two separate beams.

The first beam is sent through a device that doubles its frequency. The resulting 400 nm violet pump light is focused onto the sample to start the reaction of interest. (See animation above.)

The second beam is focused on a 0.1 mm cylindrical water jet in a vacuum chamber. When the IR photons hit the water, a plasma is created. The electric field of the laser beam propels electrons in the plasma, which then slam into the adjacent water jet and come to an abrupt stop. That releases energy in the form of hard x-rays, which are gathered and channeled onto a spot on the sample smaller than the diameter of a human hair. The pump and probe beams strike the sample at the same location, so very precise alignment is required.

There are two potential problems with TR-XES implemented in a pump-probe scheme. The first is the relatively small number of emitted x-ray photons that can be measured. The signal consists of the emitted x-rays, which diverge equally in all directions from the sample. Because only a fraction of those photons can be captured and measured, the signal is typically several orders of magnitude weaker than the probe beam.

The second concern is that the technique requires quantifying the emitted x-ray energies to within a few electron volts (eV). For reference, a single photon of visible red light has an energy content of about 1.8 eV. The x-ray photons measured in the NIST apparatus have energies from 3,000 eV to 15,000 eV. That means that they must be measured to an accuracy around 1 part in 1,000 or better.

This is a formidable challenge. Traditional methods employ grating or crystal spectrometers which are difficult to arrange into an adequate collection area. The tabletop device uses a completely different kind of detector, a superconducting transition edge sensor (TES), which NIST scientists have been improving for more than a decade. When a photon strikes the TES, it temporarily heats up the circuit enough to cause the resistance to rise and the current flow to drop; the magnitude of those effects is dependent on the energy of the incoming photon.

The tabletop system uses an array of 240 superconducting TESs (for a total area of about 23 mm²) cooled to 0.115 K to suppress thermal and electrical noise, and configured to capture a maximum amount of a limited number of emitted x-ray photons.  X-ray spectrometers constructed from arrays of TES detectors are a NIST invention.

The collection efficiency of the NIST detector system allows scientists to make measurements using two or three orders of magnitude fewer photons than equivalent measurements from a synchrotron. This reduction is particularly important when examining biological samples, such as reactions that play a key role in photosynthesis. “Photosynthesis has proven very difficult to study because x-ray analysis techniques damage the material even as they peer inside it,” says Miaja-Avila, who helped conduct the experiments and led the preparation of the journal article. “Our system could be very beneficial for studies of such damage-prone materials.”

“This measurement system breakthrough on a tabletop represents a major advance in x-ray material science instrumentation, combining the technological expertise of laser-generated x-ray production with high-resolution, high-efficiency x-ray collection,” says Bob Hickernell, a division chief of the NIST Physical Measurement Laboratory, where the research was performed. “It offers unique access for scientists and engineers to an important range of time-resolved x-ray measurements and practical applications in a small lab setting that otherwise requires scarce and expensive x-ray beam time in a large-scale facility.”