2

Tools Made of Light

The study of light plays a central role in understanding nature. Light from distant stars reveals the mysteries of the universe and served as a means for humans to circumnavigate the globe. Researchers knew about the existence of a strange form of light dating to the 18th century, but it took the 1895 discovery by Wilhelm Röntgen to identify and harness the light now called X rays. For his discovery’s impact on society, Röntgen won the first Nobel Prize in Physics in 1901. Scientists soon recognized that light comes in all different colors, the visible light from stars and Röntgen’s X rays were one and the same phenomenon, electromagnetic waves that travel with the same speed in a vacuum. In the 19th century, researchers discovered that gases in a plasma or a flame not only emit the characteristic light we see but also do so in discrete colors. These observations presented a paradox that eluded the mechanics of Newton and stimulated the birth of quantum mechanics in the early part of the 20th century. The expansion of spectroscopy—breaking up light into its component colors—across the electromagnetic spectrum, from X rays to radio frequencies, revealed the quantum structure of matter and unveiled the secrets of the universe.

In 1960, the first demonstration of the laser by Theodore Maiman defined a transformational moment not only in science and engineering but also for society. What was special about laser light? For one, the light, or photons, acted in a collective manner, the so-called coherence property. Consequently, light can have beam-like directional properties, where its composition approaches a single color, or alternatively it can be packed into a brief burst in time. Similar to the curiosity driven activity launched by Röntgen’s discovery, current-day scientists

are learning to manipulate and harness the light properties, while extending lasing into the X-ray regime. Over the past decade, these advances have produced exquisite tools for enabling unprecedented applications in conveying quantum information, precision tests of foundational physics, and producing movies of the motion of matter. Just as seeing different colors and intensities of starlight partly revealed the nature of the universe, using detectors based on interfering light has provided new eyes on the universe in the form of gravitational waves. This chapter will expand upon this query: What is so special about light?

GENERATING LIGHT WITH EXTREME PROPERTIES

Lasers have infiltrated everyday life to such a degree that their presence has become ubiquitous and unnoticed. Lasers are used for scanning barcodes at the grocery store checkout counters, for range finders for surveying, for displaying Blu-Ray discs for our viewing pleasure, and for providing precise tools for surgery. The invention and development of the laser has its origin grounded in fundamental atomic, molecular, and optical (AMO) research. In fact, the relationship between light and its applications has been synergistic: engineering new sources enables new science, while research demands drive the development of novel sources, and this remains a modus operandi. In this section, the committee will describe recent advances in the intensity (brightness), time duration, frequency (color), and coherence (oneness) of laser light.

Intensity

One measure of light is its intensity, a property defined as the power transferred per unit area. Power is the energy (total number of photons) per unit time. The peak power depends on the temporal bunching of the number of photons. Thus, a fixed number of photons arriving in 1 second will have their peak power increased by a billion-fold if bunched into 10⁻⁹ s (nanosecond). Currently, scientists have achieved peak powers of a few petawatts (1 PW ≡ 10¹⁵ W), equivalent to the combined solar power striking the states of California, Arizona, and Nevada on a sunny day at noon; but this power lasts for only 100 femtoseconds (10⁻¹³ s), a time quite short compared to 1 second. The intensity depends inversely on the beam area, which is proportional to the square of the radial size perpendicular to the propagation direction. The laser cavity or external optics define the area, which cannot be smaller than the wavelength squared. Focusing a PW-class laser beam to a diameter of one wavelength (1 micron ≡ 10⁻⁶ m) yields an intensity of 10²³ W/cm², which is approximately the current state of the art. Figure 2.1 shows the growth in intensity or equivalent electron energy as a function of year. In the early days of laser development, several technology advances (Q-switching, mode locking)

pushed intensity to a limit where the light caused catastrophic material damage thus clamping further gains as illustrated by the intensity plateau near 1970. In 1986, Mourou and Strickland circumvented this problem by the revolutionary technological concept of chirped pulse amplification (CPA; see Figure 2.2), for which they were awarded the 2018 Nobel Prize in Physics. Since that discovery, the intensity has rapidly risen in an analogous fashion to Moore’s law for computer chips. This advance resulted in scaling down the highest intensity lasers from facilities operation to table-top systems, thus opening numerous opportunities in AMO and plasma physics. CPA lasers producing intensities equivalent to the atomic field (regime of equivalent field strength as an electron bound to a proton in the hydrogen atom) enabled the rapid progress in ultrafast and extreme nonlinear science

discussed in Chapter 5. At the current state-of-the-art intensities (10²³ W/cm²), electrons become highly relativistic—that is, their motion approaches the speed of light, and must be described by physics beyond Maxwell’s equations. In addition, these high intensities are enabling the production of secondary particles (X rays, protons, neutrons) emerging from gas and solid plasmas that are finding applications in national security, astrophysics and cancer therapy.

Several ongoing frontier efforts in Europe, Korea, China, and Japan based on CPA are pushing the technology to new limits. Most notable is the Extreme Light Infrastructure (ELI) project of the European Union. The ELIs are composed of three pillars located in Hungary, Czech Republic, and Romania. These facilities will push the intensity by 100-fold, enabling the exploration of light-driven particle acceleration, nuclear physics, and zeptosecond (10⁻²¹ s) pulses. Clearly, Europe and Asia have taken leadership in a technology pioneered in the United States. The consequence of this has been a steady decay of technical competency and industrial innovation in the United States.

As shown in Figure 2.1, current CPA projects are all based on increasing the power by scaling the geometric growth of the amplifiers, which will ultimately plateau (see dashed line) similar to the earlier period before CPA. To push beyond this limit, novel optical architectures need to be developed. One approach under discussion is the “lambda-cubed shortcut.” The idea is to increase the intensity by decreasing the beam’s area and duration through shortening the wavelength.

Another transformative event occurred in fall 2009 with the operation of the world’s first X-ray free-electron laser (XFEL) at SLAC National Laboratory. The FEL concept is distinct from optical lasers for several reasons, but one major difference is that the lasing medium is not a physical material—for example, a crystal, or a diode—but a relativistic beam of electrons. In the 1970s, researchers developed FEL devices lasing in the mid-infrared/ultraviolet regime using optical resonators inserted into straight sections of a circular electron storage ring. However, operation was limited to the optical regime due to a combination of poor gain and low reflectivity of optics. Consequently, interest in FEL technology declined because these devices could not compete with traditional lasers. In the 1990s, a revolutionary approach emerged based on high-gain, single-pass lasing, which eliminated the need for resonator optics but at the cost of linear electron accelerators with high peak current. At first light, the Linac Coherent Light Source (LCLS) at SLAC performance at 1.5 Angstrom exceeded the peak spectral brilliance of all previous X-ray sources by more than a billion-fold. Truly revolutionary! The X rays had millijoule (mJ) energy, an attribute common to optical lasers but unprecedented from X-ray sources. Figure 2.3 presents a brief tutorial on the XFEL principle.

Another advance closely related to intensity is the repetition rate for delivering the light pulses. In an experiment, the real time to accumulate a signal is a

critical element in evaluating the feasibility for performing a measurement. Thus, the average power, defined as the total energy delivered over 1 second, is another critical experimental parameter. A standard commercial table-top CPA laser system operating in the near-infrared (NIR) produces 1-25 W of average power at repetition rates spanning 0.1 to 1 kHz. However, there has been a dramatic change in the average power capabilities of table-top lasers. The state-of-the-art table-top lasers are operating at kilowatt average power at 0.1 to 1 MHz repetition rates. These advances, which are mainly occurring in Europe, are rapidly moving onto the commercial market. The challenges associated with thermal management are being circumvented by new methods based on coherent combination of fiber amplifiers

outputs or optical parametric amplifiers. Likewise, in the X-ray regime, the next generation of XFEL coming online will also move performance from the 120 Hz of the initial LCLS project into the megahertz regime for the European XFEL and LCLS II in the United States. In these cases, the enabling technology is superconducting electron accelerators, which will push the average power to 200 W level.

Each of these advances enables more precise experiments and greater user access while transforming our understanding of light-matter interactions. AMO scientists play a central role in the technology and science. Some of the science enabled by these parameters are highlighted in Chapter 5.

Time

As illustrated in Figure 2.4, scientists are faced with the daunting task of clocking natural phenomena over a broad range of time scales. Over the past decade, the emergence of new transformational tools has expanded the time scales to include the clocking at the molecular and electronic levels.

The duration of a light pulse has far-reaching consequences beyond defining its power. The uncertainty principle dictates a strict inverse relationship between frequency and time. The shorter the time duration of a pulse, the larger the uncertainty in its frequency content, or conversely, a laser with a well-defined frequency

requires a long coherence time. The committee defers discussion of ultra-narrow laser operation for the next section. Instead, here it will examine time in the context of producing a “movie” of matter on specific physical or chemical time scales. A movie represents a series of snapshots that, when played back in sequence, reveals motion. The key is that the “camera shutter” is fast enough to freeze the motion in a single snapshot. For AMO scientists working with ultrafast technology, the “shutter” is the brevity of the light pulse. Figure 2.4 illustrates the time scale of natural and technological processes. The second is a good metric of human experience, the beat of a heart. The beating of a fly’s wing is a thousand times faster (10⁻³ s), while phenomena like shock propagation in glass or fast chemical kinetics is even faster (10⁻⁶ s). The lifetime of a fluorescing molecule or flipping a central processing unit (CPU) bit in a classical computer takes nanoseconds to picoseconds.

Biological movement and chemical kinetics typically happen at times longer than nanoseconds (1 ns ≡ 10⁻⁹ s). Moving to shorter light pulses, one enters the time scale defined by the rotational (picoseconds ≡ 10⁻¹² s) and vibrational (femtoseconds ≡ 10⁻¹⁵ s) motion of the nuclear constituents of a molecule. Laser pulse technology producing femtosecond durations are ideal probes of this molecular motion; pioneering work resulted in the 1999 Nobel Prize in Chemistry being awarded to the late Professor Ahmed Zewail.

Femtosecond light pulses from lasers remain the routine ultrashort tool to this day, but in 2001 a watershed moment occurred in laboratories in Paris and Vienna, when the first formation of attosecond (1 as ≡ 10⁻¹⁸ s) light pulses gave birth to the attosecond era, and enabled the study of electron motion in matter on its natural time scale. For example, the orbit time associated with the electron in the ground state of the Bohr atom is 150 as.

CPA technology was crucial for this discovery since attosecond pulse generation results only when an intense laser pulse possessing an atomic unit of field¹ interacts with an atomic sample (the bound electron regime shown earlier in Figure 2.2). This extreme nonlinear interaction leads to the process of high harmonic generation (HHG). In the frequency domain, HHG produces a coherent frequency comb or plateau of odd-order harmonics whose Fourier synthesis in time is a train of attosecond pulses or an isolated attosecond pulse. These tabletop sources can extend into the kilovolt spectral regime, although the majority of applications are performed in the extreme ultraviolet (XUV) to soft X-ray regime. Current state-of-the-art HHG sources have produced attosecond pulses with durations of two atomic units of time (1 au ≡ 24.2 as).

In the future, high-intensity relativistic interactions resulting from a laser interacting with a solid surface could break the attosecond barrier opening the

___________________

¹ The atomic unit of field is defined as the Coulomb field strength between the electron and proton in the ground state of the hydrogen atom and has a value of 50 V/Å.

possibility of zeptosecond bursts (10⁻²¹ s). In one model, the ionized (plasma) surface oscillates at relativistic speeds producing a Doppler shifted laser light at much higher frequencies: a harmonic comb extending deep into the X-ray regime. In principle, all the frequencies are phase-locked, thus capable of producing not only attosecond but also potentially zeptosecond (1 zs ≡ 10⁻²¹ s) bursts. At this time scale, scientists would be able to time resolve nuclear processes.

XFEL facilities are defining not only the intensity frontier in X-ray science but also a new vista in time. For decades, synchrotron facilities produced periodic burst of X rays of ~100 ps, but XFELs have demonstrated performance down to 1 fs, shortening pulse durations by five orders of magnitude. Furthermore, proof-of-principle experiments have demonstrated that the next-generation XFELs will have operational capabilities on the attosecond time scale.

Frequency, Bandwidth, and Coherence

The use of light across the electromagnetic spectrum is a foundational tool in modern-day AMO physics covering spectroscopy, cooling, clocks, ultrafast, metrology, and imaging. The color of light is often thought of as defined by its frequency, which is a measure of the number of oscillations of the electromagnetic wave in 1 second, expressed in Hertz (s⁻¹) in SI units. In vacuum, the frequency is proportional to the speed of light divided by the wavelength. In the context of science and technology, the electromagnetic frequency spectrum is continuous, extending from radio (kilohertz) to gamma-ray (yottahertz = 10²⁴ Hz) frequencies. A beam of light, whether a continuous wave or pulsed, is characterized by a carrier or central frequency and a distribution function—for example, bandwidth. Different parts of the frequency spectrum are utilized to probe and control specific aspects of matter. Visible light, detectable by the human eye, occupies a small portion of the electromagnetic spectrum from 0.4 to 0.8 petahertz (1 PHz = 10¹⁵ Hz). In spectroscopic applications, the frequency of the light is relevant for exciting specific quantum transitions in matter. Typical electronic transition frequencies from the ground state to other bound states occur in the visible to the vacuum ultraviolet (6 PHz), while excitation among Rydberg states, fine, and hyperfine structure occur at much lower frequencies. In molecules or solids, the additional degrees of freedom from nuclear movement have transition frequencies covering the microwave (gigahertz) to infrared (100 THz = 10¹² Hz) regions. At frequencies higher than the vacuum ultraviolet, light becomes ionizing, freeing electrons from valence and inner shell states. Determining the molecular structure occurs at hard X-ray frequencies (30 PHz to 20 exahertz), where the wavelength becomes smaller than the atomic distances in matter.

Two important properties that determine the light’s spectral content are bandwidth and coherence. The bandwidth is the range of frequencies contained in the

light, and the coherence defines the phase relationship between the different colors in space and time. For fully coherent Fourier transform limited light, the bandwidth is simply related the inverse duration. Thus, a two-cycle visible laser pulse has a bandwidth comparable to the carrier frequency, while a highly monochromatic laser requires continuous wave operation. However, the high temporal and spatial coherence of a laser makes it ideal for producing waveforms that span the visible and infrared regions of the spectrum. By harnessing nonlinear optical processes, it is possible to extend laser-like coherence properties over the entire ultraviolet and soft X-ray regions. These qualities are transforming the utility of light in ultraprecise measurement and ultrafast science.

In laser science, the control of optical phases is paramount. In the spectral domain, continuous wave lasers are providing dramatically enhanced resolving power to see ever finer energy structures of matter. Ultrastable lasers that maintain optical phase coherence for tens of seconds make it possible to investigate optical transitions with resolution approaching one part in 10¹⁶. Many new scientific thrusts have emerged from that quest, such as testing for fundamental symmetries, developing sensors of increasing sensitivity, probing the quantum nature of many-body physics, and searching for new physics beyond the standard model. The best atomic clocks are now based on stable light interacting with degenerate quantum matter confined in laser fields. With significant increases in the quality factor of the atomic transition and the improved characterization and control of systematic effects, optical atomic clocks have progressed to an accuracy level of 10⁻¹⁸, which is two orders of magnitude improved from current standards.

The increased temporal resolution enabled by the combination of ultrafast lasers and extreme nonlinear optics opens the door to probe the fastest electron dynamics that occur on femtosecond to attosecond time scales. A femtosecond mode-locked (ML) laser generates a periodic train of short pulses in time that corresponds to a comb structure in the frequency domain. Thus, phase stabilization techniques can be straightforwardly applied to the pulse train to control both the repetition rate and the optical carrier frequency with respect to the pulse envelope. Equivalently, the broad spectral coverage of the frequency comb provides phase control of optical frequency markers across intervals of many hundreds of terahertz, enabling ultraprecise measurements and coherent spectroscopy that possess both ultrahigh spectral resolution and ultrawide spectral bandwidth.

MANIPULATING THE PROPERTIES OF LIGHT

The triumph of optical technology in the past decade is not only linked to the advances in generating unique properties of light but equally the ability to precisely manipulate these properties. In essence, the AMO scientist has the ability to sculpt

the light into a waveform that can guide the interaction with a quantum system. This section highlights some of these attributes.

Spectral Manipulation

As discussed above, ultrashort laser pulses are intrinsically broadband. In a method commonly termed pulse shaping, coherent bandwidth from an ML laser can be transformed into nearly arbitrarily shaped, user-defined waveforms. In one commonly used approach, the frequency components are spatially separated, manipulated in parallel in both amplitude and phase, and then put back together into a single beam. Thus, the output waveform is simply the inverse Fourier transform of the spatial pattern transferred onto the complex optical spectrum. Pulse shaping has found widespread applications both in technology and ultrafast optical science. User-defined ultrafast waveforms have been used to explore laser “optimal” control over photochemical reactions and quantum mechanical processes in matter, in implementation of multidimensional optical spectroscopies, in compressing pulses to durations approaching the oscillation period of visible light, and to deliver the shortest possible pulses at the foci of microscope objectives for nonlinear biomedical imaging and laser machining.

One important direction was inspired by the development of optical frequency comb (OFC) lasers. OFC sources highlight the discrete nature of the ML spectrum and provide a stable phase relationship between adjacent pulses in a ML train. This led to the quest for line-by-line shaping, also termed optical arbitrary waveform generation, in which spectral control is exercised independently on individual comb lines, with the consequence that unlike earlier pulse shaping experiments, the shaped field now expands to fill time. A challenge was that ML laser frequency combs generally have line spacing below ~1 GHz, too fine to be resolved by most pulse shapers. For this reason, line-by-line pulse shaping has been performed with new comb sources in which a continuous-wave laser is converted to a coherent broadband pulse field, either via strong electro-optic phase modulation or via nonlinear wave mixing in photonic microresonators (so-called Kerr combs). Such alternative comb sources provide much wider frequency spacings and bring new features such as flexible tuning and implementation in integrated photonics.

In an alternative direction, pulse shaping has impacted quantum optics. Although originally developed for manipulation of coherent ultrashort light pulses, it is equally applicable for spectral phase and amplitude filtering of any broadband optical signal. When applied to broadband time-energy signal and idler photons, generated for example via spontaneous parametric down conversion, it enables programmable reshaping of the time correlation function between the two photons. This operation is mathematically analogous to what occurs in classical ultrafast

optics but has a distinct physical interpretation. The most recent works involve manipulation and measurement of quantum states of light in which quantum information is encoded as superpositions of discrete frequency bins—a relatively new degree of freedom for quantum information with photons. Quantum information applications include generation of high-dimensional units such as qudits, which can carry multiple qubits per photon, robust transmission over fiber, frequency parallelism and routing, compatibility with on-chip microresonator sources, and potential for hyperentanglement with other photonic degrees of freedom such as different spatial or angular momentum modes.

Spatial Manipulation

Recently novel ways to spatially manipulate light have emerged, largely based on ideas drawn from solid-state phenomena. These manipulations open the door for structures that are reconfigurable and enable arbitrary localization and emission of light. Photonic topological insulators, for example, proposed a decade ago by Raghu and Haldane,² provide one-way spatial guiding of light that is robust against fabrication imperfections, even around sharp corners and bends. Photonic topological insulators have been experimentally realized in passive structures such as arrays of silicon rings or arrays of helically curved waveguides. Additionally, topological lasers with arbitrary geometries have been demonstrated, where the emission is spatially localized along the one-way edge mode. Photons are neutral particles, and hence there is no naturally existing gauge field that couples to photons. Nevertheless, effective gauge potentials for photons can be created in dynamically modulated photonic structures. In these recent demonstrations, light is spatially localized not by a boundary between a high-index region and a low-index region, but by a boundary between regions with different gauge potentials.³

Anderson localization leverages disorder and many-body interactions to strongly localize light by coherent interference from scatterers, in contrast to topological photonics that relies on order and periodicity for spatial guiding. Over the past decade, initial demonstrations of classical Anderson localization have been extended to the quantum regime with entangled photons, which show surprising behavior such as counterintuitive bunching or anti-bunching. Studies have also been carried out for localization and wave propagation in quasicrystals—a class

___________________

² S. Raghu and F.D.M. Haldane, Analogs of quantum-Hall-effect edge states in photonic crystals, Phys. Rev. A 78:033834, 2008.

³ For an extensive review of topological photonics and gauge potentials, see T. Ozawa, H.M. Price, A. Amo, N. Goldman, M. Hafezi, L. Lu, M.C. Rechtsman, D. Schuster, J. Simon, O. Zilberberg, and I. Carusotto, Topological photonics, Rev. Mod. Phys. 91:015006, 2019.

of structures made from building blocks that are arranged using well-designed patterns but lack translational symmetry.⁴

Metamaterials offer a unique approach for spatial light modulation with a subwavelength resolution. Metamaterials consist of a large number of spatially varying resonators arranged in a subwavelength-scale lattice. One can engineer each resonator almost independently via its geometry to manipulate the amplitude, phase, or polarization of light. This enables extremely large flexibility for full spatial control of the electromagnetic field. For example, metamaterials can realize an aberration-free lens with a high numerical aperture, which produces a strongly convergent wavefront. This complete spatial control of electromagnetic field enables the generation of holograms and structured light beams and the realization of cloaking. One can also exploit the high spatial resolution to multiplex several functionalities in one metamaterial. Metamaterials can be classified by the two types of subwavelength resonators—plasmonic and dielectric resonators. Plasmonic resonators allow deeply subwavelength sizes for extremely high spatial resolution, while suffering from inevitable material absorption from metals. Dielectric resonators, on the other hand, circumvent material absorption by using high-refractive-index dielectric materials such as silicon, silicon nitride, and titanium dioxide. In spite of the lower spatial resolution compared to their plasmonic counterparts, the high efficiency and the potential for Complementary metal–oxide–semiconductor (CMOS)-compatibility of all-dielectric metamaterials have attracted great attention lately. Currently, there is still a lack of effective approaches to tune the large number of resonators dynamically, which will require further research to incorporate tunable materials such as two-dimensional (2D) materials and phase-change materials.

Quantum Manipulation

The realization of strong interactions between individual light quanta (photons) is a long-standing goal in optical science and engineering that is of both fundamental and technological significance. While it has been known for more than half a century that light fields can interact with each other inside nonlinear optical media, at light powers corresponding to individual photons the nonlinearity of conventional materials is completely negligible. Remarkable advances in quantum optics over the past decade have recently culminated in experimental demonstrations of several methods to generate optical nonlinearities at the level of individual photons.

A long-standing goal in optical science has been the implementation of nonlinear effects at progressively lower light powers or pulse energies. The ultimate

___________________

⁴ Z. Vardeny, A. Nahata, and A. Agrawal, Optics of photonic quasicrystals, Nature Photon 7:177-187, 2013, https://doi.org/10.1038/nphoton.2012.343.

limit may be termed quantum nonlinear optics, where individual photons interact so strongly with one another that the propagation of light pulses containing one, two, or more photons varies substantially with photon number. While this regime is difficult to reach because of the small nonlinear coefficients of bulk optical materials, the potential payoff is significant. On one hand, the realization of quantum nonlinear optics could improve the performance of classical nonlinear devices, enabling—for example, fast energy-efficient optical transistors that avoid Ohmic heating. On the other hand, nonlinear switches activated by single photons could enable optical quantum information processing and communication, as well as other applications that rely on the generation and manipulation of nonclassical light fields.

Over the past decade, this long-standing goal of quantum nonlinear optics was achieved using the complementary approaches of cavity quantum electrodynamics (cQED) and Rydberg-blockade mediated photon-photon interactions. The cQED approach to enhance the atom-photon interaction probability, beyond what is possible with a tightly focused laser beam, is to make the photon pass through the atom repeatedly. This can be achieved by means of an optical cavity. An optical cavity comprises two (or more) highly reflective mirrors between which the light can bounce multiple times before escaping due to transmission or losses. In this case, the interaction probability is enhanced by the number of bounces the photon makes between the mirrors before leaving the cavity, which is conventionally quantified by the cavity “finesse.” Taking the multipass atom-photon interaction into account, one can define a quantity called cooperativity—that is, how close the interaction probability approaches unity—enabling strong nonlinear interactions at the single-photon level. Over the past decade, systems with cooperativity approaching 50 were realized. This approach led to the realization of a single-photon transistor and of a single-photon phase switch as well as quantum logic operations and entanglement of individual photons using the strong coupling of a single trapped atom or a single atom-like color center to the optical cavity. These techniques were already used to realize optically mediated interactions between two color centers in nanophotonic cavity and quantum logical gates between two trapped atoms.⁵ They pave the way toward integrated quantum networks involving multiple atomic nodes connected by guided light, as discussed in Chapter 4.

An alternative approach, which does not require optical resonators, makes use of strong atom-atom interactions associated with atomic Rydberg states. Atoms in such states have large orbits resulting in large polarizability and greatly

___________________

⁵ R.E. Evans, M.K. Bhaskar, D.D. Sukachev, C.T. Nguyen, A. Sipahigil, M.J. Burek, B. Machielse, et al., Photon-mediated interactions between quantum emitters in a diamond nanocavity, Science 362(6415:662-665, 2018, https://doi.org/10.1126/science.aau4691.

enhanced interaction strength. By strongly coupling incoming photons to such states, via, for example, electromagnetically induced transparency, strong long-range dipolar interaction between two Rydberg atoms can mediate a strong effective photon-photon interaction of propagating photons down to the level of individual light quanta. Realization of such systems has been a long-standing goal in nonlinear optics community. With this approach, researchers realized an optical medium that transmits one photon, but absorbs two; a bound state between two and three photons and a robust nonlinear phase shift between two individual photons.

Systems exhibiting strong photon-photon interactions are currently being explored to enable unique applications such as quantum-by-quantum control of photon states, single-photon switches and transistors, all-optical deterministic quantum logic, the realization of quantum networks for long-distance quantum communication, and the exploration of novel strongly correlated states of light and matter. Some of these exciting new developments are discussed in subsequent chapters of this report.

Squeezed States of Light

Quantum correlations of light extend beyond the single photon regime. Indeed, continuous wave light beams also have quantum descriptions and can be manipulated for applications in sensing and metrology. One such quantum state of light is a squeezed state, which can increase the precision of optical measurements, most notably in laser interferometers. Quantum uncertainty imposes a fundamental limit on the precision with which complementary quantities—for example, position and momentum of a particle—can be simultaneously measured. Squeezed states of light manipulate this limit by decreasing the noise in either the phase (or amplitude), while simultaneously increasing the noise in the orthogonal property—amplitude (or phase)—hence “squeezing” the minimum uncertainty. Thus, a measurement will have greater precision if the detected signal lies in its squeezed state, without contamination from the orthogonal property, which is necessarily much noisier.

The applications of squeezed light include radiometry, quantum sensing, and quantum key distribution. However, one of the great successes is its application to gravitational wave detectors, where squeezed light increases the sensitivity of the optical measurement at the output of the kilometers-long laser interferometers. This is a beautiful example of AMO techniques developed for a revolutionary astrophysics instrument, and is described in greater detail in Chapter 6.

In the near future, as squeezed light sources are engineered to be more portable and integrated, this technology can improve the precision of any quantum-limited optical measurement. Some significant challenges that must be overcome to realize

this are lower loss linear and nonlinear optical components, higher efficiency photodetectors, and optical interfaces.

EMERGING PLATFORMS

Light-based technologies have allowed the development of several “platforms” that may be used for many different applications. In this section, the committee describes a few of these versatile platforms and the advances they have enabled.

Optical Control of Color Centers in the Solid State

Crystallographic defects, dubbed color centers, particularly in wide-bandgap materials, have become fertile platforms for light-matter coupling, quantum information processing, and quantum sensing. Research on quantum control of atom-like impurities in the solid state aims to bridge the gap between the advantage of well-defined quantum properties present in isolated atoms and the possibility to create conditions for strong interactions and integration into nanoscale solid-state devices by combining the best of both worlds of atomic and condensed-matter qubits. Formed by introducing localized defects, such as low-concentration dopants, in solid crystals, these impurities feature tightly localized orbitals, with electronic states that resemble those of individual atoms. In other words, they behave essentially as single atoms frozen inside a solid lattice. Similar to atoms and ions, quantum mechanical superpositions of their states can be prepared and manipulated using coherent control techniques developed by the AMO physics community. Remarkably, this can sometimes be accomplished using optical techniques even under ambient, room-temperature conditions. Moreover, because of the tight localization of their wave functions, atom-like systems can be strongly coupled to other systems and degrees of freedom, such as nuclear spins in the lattice, photons, or phonons. Recent experiments show that atom-like systems may enable far-reaching control over quantum states, with applications ranging from information processing and quantum communication to biological sensing.

The nitrogen-vacancy (NV) center in diamond is a prototypical example of an atom-like system. This quantum defect is formed by a nitrogen impurity next to a missing carbon, or vacancy, and can be created naturally or by nitrogen ion implantation and annealing. The electrons occupying the dangling bonds around the vacancy play the role of the electrons bound to the nucleus of an atom or ion, exhibiting long-lived spin states and well-defined optical transitions. The electronic states lie deep within the wide indirect bandgap of diamond, and are thereby energetically isolated from the Bloch states of the crystal. Furthermore, the low spin-orbit coupling in diamond, as well as its mostly spinless carbon-12 lattice, create an ideal solid-state environment for spins. Thus, despite the fact that the NV defect is

surrounded by nearest-neighbor atoms only angstroms away, its states are so well isolated from environmental perturbations that their coherence properties can be comparable to those of an atom trapped in ultrahigh vacuum. Over the past decade, NV centers have been used to realize long-lived quantum memory and multiqubit quantum registers, realize and probe novel phases of matter away from equilibrium such as discrete time crystals (see Chapter 4), create long-distance entanglement that has been used for first loophole-free violation of Bell inequalities, and realize unique applications for nanoscale sensing. Specifically, magnetometers based on NV centers were able to record nuclear magnetic resonance (NMR) signals from individual molecules and picoliter-scale samples with sufficient sensitivity to perform analytical NMR spectroscopy of single cells (see Chapter 4). These techniques are already being employed for practical applications in biomedical diagnostics.

Moreover, over the past few years many other atom-like defects, both in diamond and other wide-bandgap materials, have been actively explored. Some of them may possess properties that are superior to those of NV centers for certain applications. For example, a silicon-vacancy (SiV) centers in nanophotonic diamond system was used to efficiently generate a coherent optical interface for the generation, storage, entanglement, and manipulation of individual photons. These features, which arise from an inversion symmetry associated with SiV centers, make them an intriguing alternative to NV centers, with potentially superior properties for quantum optics and quantum networking applications For instance, very recently, SiV centers were employed for the first demonstration of memory-enhanced quantum key distribution (see Chapter 4). At the same time, defects in silicon carbide and 2D materials are being actively explored. Identifying and custom-designing atom-like systems with desired properties is an active topic in materials science research.

Integrated Optics

In the past decade, the photonic community witnessed a complete transformation of optics. Scientists went from being able to miniaturize a handful of devices to being able to define and control the flow of light using thousands of monolithically integrated optical components—all on a silicon chip. The key to enable this massive integration is the decrease of optical loss. In the past decade, propagation losses in waveguides have decreased by orders of magnitude thanks to advances in semiconductor processing. Chip-scale long optical delays (see, for example, Figure 2.5) can now be realized, due to the recent development of photonic platforms that can transmit light confined to submicron-size waveguides, over very long distances with low optical losses.

The field of integrated optics and in particular silicon photonics is rapidly evolving and is now enabling completely new applications, ranging from Lidar to quantum platforms. While the initial drive for silicon photonics in the early 2000s

was the ability to transmit and manipulate ultrahigh bandwidth with low power dissipation, new applications are continually emerging. This is partly due to the development of novel chip-scale devices and materials compatible with silicon photonics. Many of these technologies and devices can manipulate light across the whole visible, infrared, and mid-infrared spectrum.

For nonlinear optical phenomena that enable the generation and manipulation of light across different spectral domains, chip-based platforms offer several key advantages over bulk and fiber geometries. The first is that most of the photonic-chip materials (e.g., silicon nitride, lithium niobate) have relatively high effective nonlinearities as compared to silica glass. More importantly, a critical property of these materials is the high refractive index contrast between the core waveguide and the cladding layer. For example, materials such as silicon and silicon nitride have refractive indices of 3.4 and 2.0, respectively, at 1.55 μm wavelength, and are typically covered by a silicon oxide cladding layer with an index of 1.46. This large index contrast allows for light to be strongly confined to an area smaller than the wavelength squared, which greatly enhances the effective nonlinearities due to the increase in power, over distances that can be as long as 1 m. This tight confinement also allows for dispersion engineering by tuning slightly the size and shape of the waveguide to phase match optical parametric processes over extremely broad bandwidths.

As an example of the power of integrated optics for nonlinear optics, microresonator Kerr combs have been demonstrated in a variety of platforms, including in a silicon platform that is fully compatible with standard microelectronic processing platforms. Their spectral coverage extends to the visible and mid-infrared, and repetition rates in the microwave regimes have been achieved. More importantly, such chip-scale frequency combs can generate fully coherent optical frequency combs with a bandwidth limited only by dispersion and transparency. Such soliton microcombs have in the past years been applied successfully to ultrafast ranging,

astrophysical spectrometer calibration, terabit per second data communication, as well as dual comb spectroscopy.

The area of nonlinear optics in these integrated platforms could enable a technology that is robust enough to be deployed at a large scale, and operate with power levels that are relevant for use in satellites or mobile devices. In a similar vein, optical losses in fibers, on the order of decibels per kilometer (dB/km; 1 dB ~ 25 percent), are more than four orders of magnitude smaller than those in integrated photonics, but little is known about where the latter losses stem from, and how they can be mitigated. This will require innovative approaches beyond the traditional ones employed to date. Likewise, integrated nonlinear photonics can be leveraged with new materials, such as gallium phosphide (GaP), which are promising, but have not been widely explored. Innovative inverse design approaches may herald unprecedentedly flat and complex dispersion profiles, enabling synthesis of light of virtually any color on a chip, with any desired spectral envelope, in a fully coherent manner, at low powers—both pulsed and continuous wave.

There is an urgent need for optical materials that are widely tunable to enable light generation, manipulation, processing, and detection on a single platform. Current integrated optics platforms fall short. Silicon offers ideal compatibility with electronics, but it is a poor light emitter and a poor detector at telecom wavelengths, and it suffers from losses when it is modulated. III-V materials are excellent light emitters and detectors but lack compatibility with CMOS electronics for massive integration. Currently, heterogeneous integration of silicon photonics with bonded III-V materials represent the state of the art.

Two-dimensional materials can enable true monolithic integration of light generation, manipulation, and detection on a massively integrated platform with CMOS electronics (see Figure 2.6). These 2D materials are crystalline, comprise a single layer of atoms, and already have a breathtaking number of applications since their discovery in the early 2000s. In the past decade, room-temperature lasing, high-speed modulation, and high-speed detectors have been demonstrated using 2D materials integrated on photonic devices. The quantum confinement in the direction perpendicular to the 2D plane leads to novel electronic and optical properties that are not present in the bulk counterpart of these materials and other three-dimensional (3D) photonic materials such as gallium arsenide (GaAs) and silicon.

Graphene is a crystalline form of carbon in a sheet that is one atom thick. Graphene has unique optical properties such as a widely tunable refractive index and a zero bandgap, and it has been shown to be compatible with CMOS electronic manufacturing. Over the past decade, tremendous advances have been made in integrated detectors and modulators capable of more than 100 Gigabits per second bandwidths. Developing techniques to dope graphene are key to extracting its full potential. Graphene integration with deposited or amorphous materials enables a true electro-optic effect on otherwise passive platforms (e.g., silicon nitride,

silicon dioxide). It also enables low-power, large modulation in silicon without the carrier effects. Transition metal dichalcogenides (TMDCs), another 2D material, can be integrated with optical chips to enable generation of classical and quantum light. A recent demonstration of the presence of isolated defect single-photon emitters in 2D tungsten diselenide shows the potential of TMDCs to transform quantum information science. These single-photon emitters promise to be easier to integrate to photonic devices than their solid-state counterparts, such as NV centers in diamond. They are also expected to have higher tunability and sensitivity to ambient conditions. Yet another 2D material, black phosphorus, has a tunable bandgap that bridges those of graphene and TMDCs.

Integration of 2D materials with integrated optic platforms will transform the field of integrated optics and revolutionize communications, sensing, signal processing, and quantum information science, all important topics in AMO. It will enable integrated optics in three dimensions, where multiple photonic layers, each with its own sources, detectors, modulators, and sensors, can be monolithically integrated into a single electronic-photonic chip.

Optomechanics

Optomechanics (OM) refers to systems in which there is transduction between light (as photons) and mechanical motion (generally as phonons). When laser light reflects off mirrors that are free to move, their mutual OM interaction is mediated

by radiation pressure. Resonators, such as cavities, enable coupling of the light to and even cooling of the mechanical modes, as much as the atomic degrees of freedom. Stimulated by advances in nanofabrication and the quest for more precise measurements of force and displacement, OM interactions have been applied to probe fundamental quantum phenomena in objects over nano- to macro-size scales, and have opened up applications ranging from quantum information to precision quantum sensing.

OM interactions have been used to probe fundamental scientific questions such as, What is the ultimate limit to the precision of measurements? Where is the boundary between classical and quantum behavior? and How does decoherence manifest? On the more applied side, OM coupling on mass scales ranging from nanogram to gram have been deployed to create the tools of quantum information such as quantum memories for information storage and so-called hybrid systems for information transport. Other applications include quantum sensing of forces, displacements, spins, magnetic fields—for example, OM has also been used to create exotic quantum states of light—for example, squeezed light—and of mechanical motion—for example, phonon Fock states. Many phenomena that have been observed with light-atom interactions and with nonlinear optical systems have been replicated with light-mechanics interactions. OM research has both enabled and been enabled by technical advances in nanofabrication, cryogenic systems, and novel materials science and engineering.

The greatest impediment to realizing the full potential of OM interactions in the quantum regime is thermal noise of the mechanical oscillators. Most OM systems are cryogenically or laser cooled to mitigate the effect of thermal noise to reach the quantum regime. As nanofabrication technologies and the availability of low thermal dissipation materials improve, design and control of the mechanical oscillators will get more sophisticated. The lower thermal noise of future oscillators will allow quantum fluctuations to fully dominate the motion of the mechanical oscillators, perhaps even at room temperature. OM is a rapidly advancing field with a bright future.

Attosecond Light Sources

In 2001, attosecond light bursts were first demonstrated in the laboratory using HHG in gases. This announcement meant that scientists now had a tool for following in time the electron motion in matter. Rapid progress in both attosecond source development and applications ensued. Table-top attosecond extreme ultraviolet sources (10 to 150 eV) driven by commercial Ti:Sapphire CPA lasers centered at 800 nm are now operating in many laboratories around the world. The typical repetition rate is 1 to 3 kHz. Over the past decade, the spectral range of high-order harmonics has been extended from the XUV into water window soft X rays

(282 to 533 eV) by exploiting the fundamental wavelength scaling toward higher HHG photon energy enabled by long-wavelength infrared laser drivers (>1.5 μm). Similarly, pulse durations have steadily decreased, with a current record of ~50 as. Figure 2.7 is a typical table-top setup for performing attosecond photoionization experiments. The principles of generating trains or isolated attosecond pulses are similar, and require only different shaping of the fundamental field. The concept of the apparatus is very similar to a synchrotron or XFEL platform: the attosecond source is the beamline, which can support different end-stations—that is, electron or photon detection—except in this case, it is all done in a laboratory-scale, ultrahigh vacuum setup.

In the next decade, attosecond X-ray open-access user facilities based on gas-phase HHG and XFELs will provide unique opportunities to scientists in many fields to study electron dynamics. The advances in high average power ultrafast laser sources offers a path for attosecond generation with significantly enhanced repetition rates spanning 0.1-1.0 MHz, allowing application of more sensitive detection schemes, such as multiparticle coincidence. Likewise, new high-power femtosecond mid-infrared lasers will enable the generation of kiloelectronvolt X rays with pulse duration approaching one atomic unit of time (24.2 as).

HARNESSING THE PROPERTIES OF LIGHT

One of the strengths of AMO science is that the tools developed have broad impact on science and technology, which enables the development of new frontiers. One excellent example is the development of the optical tweezer by Arthur Ashkin (2018 Nobel Prize in Physics), a technique that has now become a standard tool in biology and atomic/quantum physics, but whose intellectual inception was understanding the basic principles of the light force. As elaborated throughout this study, this tradition continues to this day. This section illustrates how AMO scientists are harnessing the fundamental properties of light to generate new tools for metrology and sensing.

Ultrafast X-ray Metrology

Since the beginning of the 21st century, our ability to measure light pulses has achieved a high degree of sophistication. Using the “zoo” of existing optical pulse characterization techniques, such as FROG, TADPOLE, and SPIDER,⁶ researchers can precisely measure the spectral amplitude and phase of complex ultrafast light pulses. In addition, interferometric techniques based on nonlinear optics or photoelectron spectroscopy can determine the relative carrier envelope phase (CEP) of an ultrashort pulse with roughly 0.1 radian precision. In addition, the stereo-above threshold ionization phase meter method⁷ can determine the CEP for a single shot, thus allowing the tagging of each individual laser pulse.

The challenge of moving ultrafast science toward shorter wavelengths is captured by a simple question: how do you measure the pulses’ temporal properties? The AMO community has developed precise metrology for the complete

___________________

⁶ For a description of the optical methods see R. Trebino, K.W. DeLong, D.N. Fittinghoff, J.N. Sweetser, M.A. Krumbügel, and B.A. Richman, Measuring ultrashort laser pulses in the time-frequency domain using frequency-resolved optical gating, Rev. Sci. Instrum. 68:3277, 1997.

⁷ The stereo-above threshold ionization method is a strong field photoionization technique for few-cycle pulses that is sensitive to small changes in the pulse’s CEP.

characterization of optical laser pulses; these techniques are now being translated with the same level of sophistication into the XUV/X-ray regime. The cumulative research in strong field atomic physics has provided a means for retrieving the temporal information. In one variant known as RABBITT (Reconstruction of Attosecond Beating By Interference of Two-photon Transitions), the quantum interference of degenerate photoionization pathways results in modulation sidebands produced by XUV+IR fields. The resulting photoelectron spectrogram conveys information on the temporal structure of a train of attosecond pulses. Temporal measurement of isolated attosecond pulses uses a streaking method wherein the XUV attosecond pulse ionizes a target atom in the presence of a relatively intense, reference optical field. The freed photoelectron will receive an additional momentum kick from the reference field’s vector potential, and this depends on the relative phase between the XUV and optical fields, controlled with an interferometer. Thus, the time-dependent vector potential of the reference field produces a trace similar to a conventional laboratory oscilloscope, except with a petahertz bandwidth. These techniques have gained wide acceptance for characterizing attosecond pulses from the XUV to soft X rays, and interrogating attosecond dynamics in nature. The clocking precision achieved in these measurements has achieved subattosecond performance.

The emergence in 2009 of the LCLS XFEL facility posed a new question for temporal metrology in the hard X-ray regime. How do you measure these short burst of X rays? The AMO community had the answer by translating the table-top techniques developed a decade earlier for measuring attosecond pulses. Although facility operation posed some interesting challenges for the application to the XFEL platforms, AMO scientists contributed to their solution, and in doing so had the opportunity for exploring new physics. These AMO techniques have become a staple on all XFEL facilities.

Generation of Optical Clocks

Improvement of spectral resolution has been a key driver behind many scientific and technological breakthroughs over the past century, including the invention of the laser and the realization of ultracold matter. The development of optical atomic clocks in recent years has benefited from the same pursuit. A number of key ideas and enabling technologies have accelerated the progress of clock performance. These include the search for and use of exceedingly high-quality atomic transitions in the visible regime, quantum state engineering of single trapped ions and of many-atom systems in specially designed optical lattices, the development of ultrastable lasers, and the invention of optical frequency combs.

Precise control of optical phases is a dominant theme in laser science. In the spectral domain, continuous wave lasers are providing dramatically enhanced

resolving power to see even finer energy structures of matter. Ultrastable lasers that maintain optical phase coherence for tens of seconds make it possible to investigate optical transitions with spectral resolution approaching 1 part in 10¹⁶. Through the powerful tool of optical frequency combs that unite stable optical phase control and ultrafast science, unprecedented spectral resolving power has been established across the entire visible spectrum and beyond.

Precise quantum state engineering of individual atoms has led to extraordinary measurement performance. The pioneering work of controlling single trapped ions for frequency standards has provided a rich set of scientific insights and technology advances. More recently, the use of many atoms, all confined in optical lattices with precisely engineered properties of trapping potentials that are independent of the two clock states, has greatly enhanced signal strengths for clock spectroscopy, leading to new levels of clock precision. The best atomic clocks have progressed to measurement uncertainty at the level of 10⁻¹⁸. Together with the progress of quantum matter and simulations, the prospect for further progress seems very promising. Furthermore, AMO scientists have created new opportunities for understanding and constructing quantum matter where many-body physics is no longer feared as a hurdle for precision measurement, but rather a new frontier to advance precision and accuracy. Shown in Figure 2.8 is the novel JILA 3D optical lattice clock, which contains a quantum gas of fermionic atoms that are spatially correlated to guard against motional and collisional effects. This type of system will provide a powerful platform to integrate many-body quantum state engineering with quantum metrology, creating exciting opportunities to advance precision beyond the standard quantum limit.

The convolution of precision control of light and matter is helping bridge different disciplines in physics and fostering new capabilities to probe fundamental and emerging phenomena. Many new scientific frontiers (see Chapter 6) have emerged and are being elucidated through the quest of building a better atomic clock, for example, tests of fundamental physics, the development of sensors of increasing sensitivity, the probe of quantum many-body physics, and the search for new physics beyond the standard model.

Light Propagation: Sensing and Control

The highly directional character of laser beams makes them ideal for remote (long distance) detection of chemical pollutants on the one hand, and for distance surveying on the other hand—for example, for mapping or use by autonomous vehicles. There are many examples of AMO-derived sensing technologies that have given rise to profound impact on environmental and other monitoring activities. An important application is the use of optical fibers to measure physical, chemical, and bio/medical phenomena. Fundamental progress in the understanding of light propagation, relativistic interferometry, the band structure of ions in glassy media, and absorption/scattering processes in “exotic” optical media have all contributed to the success of fiber optic sensors in civilian and defense applications.

A traditional approach to remote detection involves Laser-Induced Breakdown Spectroscopy (LIBS). In LIBS, a laser beam creates a small plasma plume on some remote target. As the plasma recombines, the constituents emit characteristic frequencies that can identify the materials spectroscopically. Two problems exist: (1) during propagation, laser light spreads, producing a lower intensity on target; and (2) the emitted recombination light suffers a large radial loss as it propagates back to the remote detector.

Recently, an exciting alternative to LIBS was developed. An ultrashort laser pulse propagating through the atmosphere will have sufficient peak power to drive the nonlinear effect of filamentation. In light filamentation (LF), the air acts like a lens, and focuses the laser down to the point where air molecules are ionized and a plasma forms; the plasma arrests any further spatial collapse, and the competing interactions result in stable propagation of a coherent light/plasma beam. LFs have extraordinary properties: (1) they do not diffract, thus producing high-intensity beams on distant targets; and (2) their coherence spans the entire visible range making them ideal for spectroscopic analysis.

Recent research has shown LFs can create terahertz (THz = 10¹² cycles/s) radiation at the remote target. Terahertz radiation is important for spectroscopic analysis but does not propagate well through the atmosphere. LFs create the terahertz energy near the target, thus minimizing atmospheric loss. In addition, LFs

can produce a local disruptive electromagnetic pulse. Recently, the Army Research Office demonstrated remotely disabling a drone from one shot of an LF.

In the future, new schemes of light/plasma interactions on distant targets are expected. Researchers are exploring air lasing to create backward-propagating beams, which would alleviate the propagation loss. Additionally, a unique solution to Maxwell’s equations supports a “flying torus” structure, which is quite unlike a conventional electromagnetic wave. In such structures, the magnetic field consists of loops, the plane of which are perpendicular to the direction of propagation, and the electric field wraps around the magnetic field. Although theoretically known for some time, flying torii have eluded generation and detection.

Another recently emerged powerful technique for broadband spectroscopy and sensing is laser frequency combs. While initially developed for frequency metrology, frequency combs have become a powerful tool for new approaches to broadband molecular spectroscopy. The broad coherent spectrum of sharp comb lines permits novel approaches to linear and nonlinear spectroscopy that often outperforms other state-of-the-art techniques with respect to resolution, accuracy, recording time, and sensitivity.

Comb laser beams enable both local sensing at very high sensitivity and integration over columns of kilometers of open path. The broad spectral bandwidth enables simultaneous interrogation of multiple species and transitions, improving the consistency and reliability of measurements. When comb lines are resolved, the contribution of the instrument to the experimental line profiles becomes negligible. A number of proof-of-concept demonstrations of broadband frequency comb spectroscopy for sensing applications already highlight its great potential. A dual-comb spectrometer operated in the mid-infrared spectral region can achieve open-path monitoring of ammonia from tens of meters away on millisecond time scales. Similarly, open-path comb-based mid-infrared measurements achieved detection of atmospheric CH₄ and H₂O on millisecond time scales. Comb spectroscopy has also been employed to measure the concentration of radicals involved in the atmospheric chemistry of the marine boundary layer such as BrO, IO, and NO₂ on minute time scales and with detection limits better than 10 parts per trillion volume in East Antarctica.

More recently, dual-comb spectroscopy with resolved comb lines interrogated atmospheric trace gases with high sensitivity over a 2-km open-air path, retrieving path-integrated concentrations of several greenhouse gases with remarkable consistency across repeated measurements. With continued progress in fiber laser technology, comb-based spectrometers are now transportable. Field-deployed instruments have been successfully used to make measurements in an industrial gas turbine, and identify and quantify leaks from methane sources over multiple-square-kilometer regions at emission rates three orders of magnitude lower than conventional approaches. Furthermore, other sampling techniques may diversify

the tools for spectroscopic diagnostics. For instance, the combination of laser-induced breakdown spectroscopy and dual-comb spectroscopy lays the groundwork for in situ analysis of soils, rocks, and minerals.

THE FUTURE FOR NEW TOOLS MADE OF LIGHT

History has shown the transformational impact on science, technology, and society enabled by advances in light engineering. Whether it was the discovery of X rays by Röntgen, the first demonstration of the laser by Maiman, or the development of an efficient blue diode by Akasaki, Amano, and Nakamura, the global impact on society is immeasurable. Based on the emergence of novel light capabilities inspired by grand challenge needs in science, it is easy to predict a continuation of transformational impact.

Given the amazing capabilities in the control of optical coherence, manifested in either the spectral or temporal domains, light-matter interactions are entering a new phase to drive new scientific discoveries and novel technology development. The light’s high temporal and spatial coherence makes it possible to produce waveforms that span the infrared, visible, ultraviolet, and even soft X-ray spectral regions. In the spectral domain, continuous wave lasers are providing dramatically enhanced resolving power to see ever finer energy structures of matter. Ultrastable lasers that maintain optical phase coherence for many seconds make it possible to investigate optical transitions with resolution better than 1 part in 10¹⁵. Many new scientific thrusts have emerged from that quest, such as testing for fundamental symmetries, developing sensors of increasing sensitivity, probing the quantum nature of many-body physics, and searching for new physics beyond the standard model. The best atomic clocks are now based on stable light interacting with precisely controlled atoms and ions. These systems have strong potential for continued progress in the coming decade.

The production and control of quantum states of light will play critical roles in the emerging networks of quantum communication, as discussed in Chapter 4. At the same time, the squeezed state of light is already employed to reduce quantum noise in the current gravitational wave detectors at the Laser Interferometer Gravitational-Wave Observatory (LIGO), significantly improving the detection sensitivity of compact binary coalescences. The development of platforms that combine quantum information science and integrated photonics will greatly advance the applications, sensing, communications, and computing.

Watching, clocking, and controlling all the constituents of matter remains a dream of science, having been an active area of research for several decades. The progress has been steady and insightful, but the dream remained just that. However, several recent developments in ultrafast X-ray science could provide a turning point in realizing the dream (see Chapter 5). First, the continued advances in attosecond

science and technology are providing visualization of the fastest constituent of matter, the electron. The emergence of high-average-power ultrafast lasers will now enable high-fidelity measurement tools for watching this movement. Second, the development of XFEL facilities are a revolutionary concept that brings “ultrafast” into the vernacular of X-ray science. The well-known utility of X rays for determining molecular structures will now record molecular movies. It is expected that as both technologies progress, the next decade could see the realization of the dream.

The frontier of intense laser-matter interactions has seen a major increase in activity. The holy grail is to reach sufficiently high intensity to break the vacuum: to create something from nothing. The current state-of-the-art laser is seven orders of magnitude too low to achieve that now except with the Lorentz transformation using ultra-relativistic electron accelerators. However, efforts in Europe and Asia are pushing the technology to raise the highest power obtainable from a laser into the tens of petawatt range and beyond. These laser developments need a facility-type environment for operation and thus significant investment. The U.S. funding for this frontier has lagged behind, but recent federal activity in response to the 2018 National Academies of Sciences, Engineering, and Medicine report Opportunities in Intense Ultrafast Lasers⁸ could improve the landscape in the United States. This investment will be central for the United States to regain some leadership in an area that has important national security consequences. The XFELs have an additional transformative and unique quality at X-ray frequencies—extraordinary intensity. This raises another frontier question: how does an intense X-ray pulse interact with matter? The only thing that is known is that it will be dramatically different from known interactions using visible lasers. In fact, this regime may not be compatible with current theoretical approaches, and will largely be pioneered by experiments. As the XFEL engineering improves performance, and as new facilities in Europe and Asia increase accessibility, it may be possible to begin the exploration of this physics, some of which is highlighted in Chapter 5.

FINDINGS AND RECOMMENDATIONS

Finding: The past decade has seen revolutionary advancements in ultrafast light source development spanning the XUV and X-ray spectral regime. The ability to control and manipulate these tools made of light is enabling new applications that extend beyond AMO physics. New platforms are emerging in quantum information science, remote sensing, and clocking ultrafast electron dynamics in all phases of matter. Thus, the frontiers lie at the interdisciplinary intersection of physics, engineering, chemistry, materials science, and biology.

___________________

⁸ National Academies of Sciences, Engineering, and Medicine, 2018, Opportunities in Intense Ultrafast Lasers: Reaching for the Brightest Light, The National Academies Press, Washington, DC.

Finding: Advances in ultrafast X-ray science has become increasingly demanding of resources that are beyond the ability of single principal investigator (PI) funding models. XFELs are large-scale facilities that need the management infrastructure of a National Laboratory. However, table-top systems, such as attosecond- and petawatt-class lasers, have evolved to the level of a mid-scale facility requiring operational management and safety infrastructure. The United States has lagged behind the rest of the world in capitalizing on the opportunities enabled by these mid-scale facilities, training of a workforce at the cutting edge of technology, and the economic benefits of industrial growth.

Key Recommendation: U.S. federal agencies should invest in a broad range of science that takes advantage of ultrafast X-ray light source facilities, while maintaining a strong single principal investigator funding model. This includes the establishment of user facilities in mid-scale university-hosted settings.

Finding: Despite the enormous advances of integrated linear and nonlinear photonics based on silicon, there is a need for ultra-low-loss platforms where light can be generated with ultrahigh efficiency, switched and detected, especially for quantum-related applications. Such a platform would probably be formed via the integration of multiple materials on silicon.

Finding: Systems exhibiting strong photon-photon interactions are currently being explored, to enable unique applications such as quantum-by-quantum control of light fields, single-photon switches and transistors, all-optical deterministic quantum logic, the realization of quantum networks for long-distance quantum communication, and the exploration of novel strongly correlated states of light and matter.

Finding: As nanofabrication technologies and the availability of high optical quality, low thermal dissipation materials improve, design and control of the mechanical oscillators will get more sophisticated. The lower thermal noise of future oscillators will allow quantum fluctuations to fully dominate the motion of the mechanical oscillators, perhaps even at room temperature, creating a versatile quantum resource for a variety of applications.

Key Recommendation: The federal government should provide funding opportunities for both basic and applied research that enable the development of industrial platforms, such as foundry offerings, and interdisciplinary academic laboratories to support the integration of photonics and engineered quantum matter.