In this work we demonstrate spectrally resolved x-ray scattering from electron-plasma waves in shock-compressed deuterium and proton-heated matter. Because the spectral signature of inelastic x-ray scattering is strongly dependent on the free electron density of the system, it is used to infer ionization in dynamically heated samples. Using 2-6 ns, 500 J laser pulses from LLNL's Janus laser, we shocked liquid deuterium to pressures approaching 50 GPa, reaching compressions of 4 times liquid density. A second laser produced intense 2 keV x-rays. By collecting and spectrally dispersing forward scattered photons at 45 degrees, the onset of ionization was detected at compressions of about 3 times in the form of plasmon oscillations. Backscattered x-rays bolstered this observation by measuring the free electron distribution through Compton scattering. Comparison with simulations shows very close agreement between the pressure dependence of ionization and molecular dissociation in dynamically compressed deuterium.

In a second set of experiments, a 10 ps, 200 J Titan laser pulse was split into two beams. One created a stream of MeV protons to heat samples of boron and boron-nitride and the other pumped 4.5 keV K-alpha radiation in a titanium foil to probe the hot target. We observed scattered x-rays 300 ps after heating, noting a strong difference in average ionization between the two target materials at temperatures of 16 eV and very similar mass densities. Comparison with electron structure calculations suggests that this difference is due to a persistence of long-range ion structure in BN resulting in high-temperature band structure.

These results underscore the importance of understanding the complex electron structure of materials even at electron-volt temperatures and gigapascal pressures. Our results provide new data to guide the theoretical modeling of warm, dense matter important to understanding giant planets and inertial fusion targets.

Next generation ultrabright light sources will operate at megahertz repetition rates with temporal resolution in the attosecond regime. For an X-Ray Free Electron Laser (FEL) to operate at such repetition rate requires a high quantum efficiency (QE) cathode to produce electron bunches of 300 pC per 1.5e-6 J incident laser pulse. Semiconductor photocathodes have sufficient QE in the ultraviolet (UV) and the visible spectrum, however, they produce picosecond electron pulses due to the electron-phonon scattering. On the other hand, metals have two orders of magnitude less QE, but can produce femtosecond pulses, that are required to form the optimum electron distribution for high efficiency FEL operation. In this work, a novel metallic photocathode design is presented, where a set of nano-cavities is introduced on the metal surface to increase its QE to meet the FEL requirements, while maintaining the fast time response.

Photoemission can be broken up into three steps: (1) photon absorption, (2) electron transport to the surface, and (3) crossing the metal-vacuum barrier. The first two steps can be improved by making the metal completely absorbing and by localizing the fields closer to the metal surface, thereby reducing the electron travel distance. Both of these effects can be achieved by coupling the incident light to an electron density wave on the metal surface, represented by a quasi-particle, the Surface Plasmon Polariton (SPP).

The photoemission then becomes a process where the photon energy is transferred to an SPP and then to an electron. The dispersion relation for the SPP defines the region of energies where such process can occur. For example, for gold, the maximum SPP energy is 2.4 eV, however, the work function is 5.6 eV, therefore, only a fourth order photoemission process is possible. In such process, four photons excite four plasmons that together excite only one electron. The yield of such non-linear process depends strongly on the light intensity.

In this work, the structure consisted of rectangular nano-grooves (NGs) arranged in a subwavelength grating on a metal surface is presented that provides a dramatic increase in the metal's absorption, field localization, and field enhancement. When light is polarized perpendicular to the orientation of the grooves a standing SPP wave is excited along the vertical walls in the NGs, that act as Fabry-Perot resonators. By adjusting the geometry of the NGs and the period of the subwavelength grating the resonance can be fine tuned to a desired position, for example, the laser fundamental wavelength, anywhere from the UV to the near infrared (NIR).

Two types of gratings are presented: (a) a gold grating with period of 600 nm, and (b) an aluminum-gold grating with a period of 100 nm; both with resonance at 720 nm. In each case, strong on-resonance absorption was observed, with over 98% for grating (b). Unlike the grating-coupled SPP waves, where the angle is well defined by the momentum matching condition, the resonant NGs allow coupling to the standing modes at a range of angles of incidence, referred to as the angular bandwidth. A new model for the on-resonance absorption based on the ensamble action of the NGs is presented that serves as the basis for a design of an NG grating with an ultrawide spectral as well as angular bandwidth. For sample (b), the angular bandwidth is 80 degrees, corresponding to an opening angle of 160 degrees.

The photoemission enhancement for such a grating was measured to be seven orders of magnitude for a four-photon photoemission. This is an incredible result demonstrating the power of the plasmonic grating presented, which is an efficient light trapper and field enhancer for a non-linear processes. These results demonstrate that the metal photocathode prepared with a NG grating on the metal surface will provide sufficient pulse charge driven by a 1e-6 J 15 fs pulsed laser at 800 nm for the optimum FEL operation.

Atoms and molecules have most of their oscillator strength in the vacuum ultraviolet (VUV)

and extreme ultraviolet (XUV), between the wavelengths of 200 nm and 30 nm. However,

most femtosecond spectroscopy has been restricted to the visible and infrared due to a

lack of sufficiently intense VUV and XUV femtosecond light sources. This thesis discusses

extensions of pump/probe spectroscopy to the VUV and XUV, and its application to the

dynamics of ethylene and oxygen molecules excited at 161 nm.

I begin with a detailed discussion of the short wavelength light source used in

this work. The source is based on the high order harmonics of a near infrared laser and

can deliver > 10¹⁰ photons per shot in femtosecond pulses, corresponding to nearly 10

MW peak power in the XUV. Measurements of the harmonic yields as a function of the

generation conditions reveal the roles of phase matching and ionization gating in the high

order harmonic generation process.

Pump/probe measurements are conducted using a unique VUV interferometer, capable

of combining two different harmonics at a focus with variable delay. Measurements of

VUV multiphoton ionization allows for characterization of the source and the interferometer.

In molecules, time resolved measurements of fragment ion yields reveal the femtosecond

dynamics of the system. The range of wavelengths available for pump and probe allows the

dynamics to be followed from photo-excitation all the way to dissociation without detection

window effects.

The dynamics in ethylene upon π → π* excitation are protypical of larger molecules

and have thus served as an important test case for advanced ab initio molecular dynamics

theories. Femtosecond measurements to date, however, have been extremely lacking. In

the present work, through a series of pump probe experiments using VUV and XUV pulses,

time scales for the non-adiabatic relaxation of the electronic excitation, hydrogen migration

across the double bond, and H2 molecule elimination are measured and compared to theory.

In the simpler oxygen molecule, excitation in the Schuman-Runge continuum leads

to direct dissociation along the B ³Σ_u^- potential energy curve. The time resolved photoion

spectra show that the total photoionization cross section of the molecule resembles two

oxygen atoms within 50 fs after excitation.

This dissertation will present a series of new tools, together with new techniques, focused on the understanding of warm and dense matter. We report on the development of a high time resolution and high detection efficiency x-ray camera. The camera is integrated with a short pulse laser and an x-ray beamline at the Advanced Light Source synchrotron. This provides an instrument for single shot, broadband x-ray absorption spectroscopy of warm and dense matter with 2 picosecond time resolution. Warm and dense matter is created by isochorically heating samples of known density with an ultrafast optical laser pulse, and X-ray absorption spectroscopy probes the unoccupied electronic density of states before the onset of hydrodynamic expansion and electron-ion equilibrium is reached. Measured spectra from a variety of materials are compared with first principle molecular dynamics and density functional theory calculations.

In heated silicon dioxide spectra, two novel pre-edge features are observed, a peak below the band gap and absorption within the band gap, while a reduction was observed in the features above the edge. From consideration of the calculated spectra, the peak below the gap is attributed to valence electrons that have been promoted to the conduction band, the absorption within the gap is attributed to broken Si-O bonds, and the reduction above the edge is attributed to an elevated ionic temperature.

In heated copper spectra, a time-dependent shift and broadening of the absorption edge are observed, consistent with and elevated electron temperature. The temporal evolution of the electronic temperature is accurately determined by fitting the measured spectra with calculated spectra. The electron-ion equilibration is studied with a two-temperature model.

In heated nickel spectra, a shift of the absorption edge is observed. This shift is found to be inconsistent with calculated spectra and independent of incident laser fluence. A shift of the chemical potential is applied to the calculated spectra to obtain satisfactory agreement with measured spectra.

High energy density physics (HEDP) is an emerging field that seeks to investigate the properties of matter at extreme conditions. High energy density conditions occur in materials with pressures exceeding 1 Mbar, or pressures that exceed Earth’s atmospheric pressure by a factor of more than a million. A regime of HEDP of particular interest is warm dense matter (WDM) physics, which describes the behavior of materials at near solid densities and 10’s of eV temperatures. WDM occurs in astrophysical objects, such as giant planets and brown dwarfs, and is also generated in inertial confinement fusion (ICF) experiments. X-ray Thomson scattering (XRTS) offers a powerful tool to probe the equation of state of WDM. XRTS spectra consist of two components: elastically scattered photons with the frequency of the original x-ray source and inelastically scattered photons that are down- shifted in frequency. The Compton-shifted profile of inelastically scattered x-rays can be analyzed to return the sample’s electron density and electron temperature. The ratio of elastically to inelastically scattered x-rays relates to the number of tightly bound versus free electrons, and thus reflects the ionization state.

This thesis discusses the results of XRTS experiments on WDM performed at the OMEGA Laser facility. The first experiment presents and discusses XRTS results from 1 mm diamond spheres. The scattering spectra show evidence of higher ionization than predicted by several commonly-applied ionization models. A second experiment analyzed the contributions to elastic scattering from a small argon impurity in imploding beryllium capsules. The exper- iment found that less than 1 at.% of argon significantly affects the elastic scattering signal strength, and concluded that impurities in a sample should be considered before drawing conclusions from elastic scattering signals. The final experiment uses XRTS to measure the electron temperature and ionization state in isochorically heated materials used in ion stopping power experiments. The results from these experiments demonstrate the power of XRTS to measure ionization in WDM to benchmark theoretical modeling.