Detailed model for hot-dense aluminum plasmas generated by an x-ray free electron laser

Abstract

The possibility of creating hot-dense plasma samples by isochoric heating of solid targets with high-intensity femtosecond X-ray lasers has opened up new opportunities in the experimental study of such systems. A study of the X-ray spectra emitted from solid density plasmas has provided significant insight into the X-ray absorption mechanisms, subsequent target heating, and the conditions of temperature, electron density, and ionization stages produced (Vinko et al. Nature 482, 5962). Furthermore, detailed analysis of the spectra has provided new information on the degree of ionization potential depression in these strongly coupled plasmas (Ciricosta et al. PRL 109, 065002). Excellent agreement between experimental and simulated spectra has been obtained, but a full outline of the procedure by which this has been achieved has yet to be documented. We present here the details and approximations concerning the modelling of the experiment described in the above referenced work. We show that it is crucial to take into account the spatial and temporal gradients in simulating the overall emission spectra, and discuss how aspects of the model used affect the interpretation of the data in terms of charge-resolved measurements of the ionization potential depression.

Measurements of the K -Shell Opacity of a Solid-Density Magnesium Plasma Heated by an X-Ray Free-Electron Laser

“…For each thickness, the figure shows the series of K-α emission lines corresponding to transitions of L electrons filling K-shell holes, created by the LCLS photons, in ions with a different number of L-shell holes, labelled by their charge state (noting cold solid density Mg already has 2 free electrons, and the cold K-α comes from charge state III). As in previous experiments [12,13,16] the temperatures attained are too low for the targets to emit thermally on such transitions, and the line spectra recorded thus correspond to radiation that is only emitted during the short duration of the FEL pulse itself convolved with the lifetime of the K-shell core-hole state (which is dominated by Auger decay, and is typically of order 3 fs for most states, save the helium-like state which has no Auger channel), and are thus a probe of the target under truly isochoric conditions, which further allows the electron density during the emission from a particular charge state to be deduced from the ion stage.…”

“…Targets comprising thin (24 nm -487 nm ± 2 nm) foils of Mg were irradiated by ∼ 50 fs pulses of X-rays (shorter than the nominal electron bunch duration, consistent with previous work [16,[21][22][23]), with a photon energy of 1540 eV, and nominal pulse energy of 1.8 mJ prior to transmission through the beamline focussing optics, which was estimated from the mirror reflectivities to be ∼50% [16]. The target was oriented at an angle of 45…”

“…The slight discrepancies observed for the helium-like lines -where we find the deduced opacity is lower than expected -may be attributed to a lower final temperature than expected from SCFLY simulations for our spot parameters. We find that just by simply decreasing the laser intensity in our simulations by a factor of two gives the same evolution for the earlier charge states -which are still swept through in time, and still in LTE under the same temperature and density conditions [16], yielding the same opacity -but now reduces the opacity of the final charge states as, being last to be produced, their populations are the most sensitive to the total energy in the FEL pulse.…”

“…Previous studies of such plasmas have shown that, owing to the high electron densities and thus large collisional rates [15], the systems are extremely close to LTE, with the populations of the majority of the atomic states being determined by the instantaneous electron temperature [16]. More precisely, for low-to-medium Z materials such as magnesium studied here, whilst the ground state populations are mainly determined by the collisional physics, and are thus in LTE, during the FEL pulse at any instant a few percent of the ions have a K-shell core-hole induced by the intense FEL photoionization, and in this restricted sense this small number of atoms is driven out of LTE.…”

“…To illustrate this, the experimental data were simulated using the collisional-radiative code SCFLY [26] which has previously been shown to produce excellent agreement with spectra produced by LCLS interacting with solid targets [16]. Simulations over a range of FEL intensities calculate the local time-dependent pop- ulations of super-configurations within the target, with the only user input being the time-dependent incident X-ray pulse.…”