Unified Time and Frequency Picture of Ultrafast Atomic Excitation in Strong Laser Fields

Excitation and ionization in strong laser fields lies at the heart of such diverse research directions as high-harmonic generation and spectroscopy, laser-induced diffraction imaging, emission of femtosecond electron bunches from nanotips, self-guiding, filamentation and mirrorless lasing during propagation of light in atmospheres. While extensive quantum mechanical and semiclassical calculations on strong-field ionization are well backed by sophisticated experiments, the existing scattered theoretical work aiming at a full quantitative understanding of strong-field excitation lacks experimental confirmation. Here we present experiments on strong-field excitation in both the tunneling and multiphoton regimes and their rigorous interpretation by time dependent Schrödinger equation calculations, which finally consolidates the seemingly opposing strong-field regimes with their complementary pictures. Most strikingly, we observe an unprecedented enhancement of excitation yields, which opens new possibilities in ultrafast strong-field control of Rydberg wave packet excitation and laser intensity characterization.

Figure 1

Sketch of the experimental setup. A focused linearly polarized laser beam with a pulse duration of 45 fs full width at half maximum and a center wavelength of either 800 or 400 nm intersects a collimated effusive beam of rare gas atoms midway between two field plates, where high voltages (HV) can be applied for pulsed field ionization (see Supplemental Material [26]). The atomic beam A is directed towards a position-sensitive multichannel plate (MCP) detector, which detects excited atoms ${\mathrm{A}}^{*}$ due to their high internal energy ($\approx 10\mathrm{eV}$ or more). Ground state atoms are not detected. A typical distribution of excited atoms is shown color coded. For the analysis we take only signals stemming from the immediate vicinity of the focal plane (red rectangle) to reduce volume-averaging effects.

Figure 2

Yield of strong-field excited Ar (black disks) and Ne (green squares) atoms, as a function the laser intensity (lower $x$ axis). The Keldysh $\gamma$ parameter for argon is shown on the upper $x$ axis. (a) 400 nm laser wavelength. Volume-averaged TDSE results for the all-electron (AE; red dashed line) and core-eliminated (CE; blue dotted line) model potentials are shown for comparison. Selected channel-closure intensities [$I_m=3509{\omega }^2(m\omega -I_p)$, where $I_m$ is in $\mathrm{TW}{\mathrm{cm}}^{-2}$] are indicated by black (Ar, top) and green (Ne, bottom) vertical arrows. (b) 800 nm laser wavelength. Channel closings are found at $\approx 26\mathrm{TW}{\mathrm{cm}}^{-2}$ intervals, and are not shown. Experimental intensities are calibrated on the calculated position of the six-photon channel closing in argon.

Figure 3

Probability of ionization and neutral excitation in Ar, as a function of peak intensity (in $\mathrm{TW}{\mathrm{cm}}^{-2}$). Probabilities above (below) brown horizontal line at $P\approx 0.01$ are shown on a linear (logarithmic) scale. (a) Excitation by a 400 nm field. Red dashed line: AE potential; Blue dotted line: CE potential. Intensities of $m$-photon channel closings are indicated on the top $x$ axis. (b) Probability of neutral excitation in argon for 800-nm driving field. (c) Combined probability of excitation and ionization.

Figure 4

Distribution of the principal quantum number $n$ of the Rydberg states in Ar after strong-field excitation. Unless noted otherwise, data are for the 400 nm driving field. In (a) the principal quantum number in the abscissa corresponds to $n$, in (b) and (c) to $n_{\mathrm{MAX}}$ (a). Calculated (AE) distributions at different peak intensities close to the six-photon peak in Fig. 3. The distributions are normalized to 1 over Rydberg state population with $4\le n\le 23$. The CE distributions are qualitatively similar. (b) Experimental cumulative distributions of $n$ at 180, 260, and $1100\mathrm{TW}{\mathrm{cm}}^{-2}$ (400 nm) and $770\mathrm{TW}{\mathrm{cm}}^{-2}$ (800 nm). The error bars include the statistical count error only. (c) Theoretical cumulative distributions after volume averaging.