Control of photoionization by resonant phase-locked pulse pairs

TL;DR

This paper addresses coherent control of photoionization in He and H using resonant phase-locked pump–probe pulses to steer dressed-state populations in a two-photon resonance $1+1$ scheme. It develops a Floquet-based, non-Hermitian two-level model to predict when a dressed state becomes stabilized, with ionization rates given by $oxed{oldsymbol{ extlambda}=E-i oldsymbol{ extGamma}/2}$ and the probe CEP selecting the dressed state during the second pulse. Ab initio RTDCIS/TDCIS calculations validate the predictions, showing that circular polarization at high intensity suppresses probe ionization by steering population into the stabilized state, yielding a single Autler–Townes-like peak and Ramsey-like fringes in the photoelectron signal; linear polarization reduces control due to multiple continua. The results demonstrate a path to dynamic control of bound states in the continuum in the XUV regime and have implications for short-wavelength FEL experiments where phase-locked pulses can manipulate dressed-state dynamics.

Abstract

We study the nonlinear and resonant process of two-photon ionization of atoms (He and H) in a pump-probe scheme. The pump pulse prepares the quantum system in a superposition of the ground state and an excited bound state. By varying the phase difference between the pulses, we show how it is possible to coherently control the dressed-state population during the probe pulse. Our main result is that for certain laser parameters, the control over the dressed state population leads to strong control of the ionization probability during the probe pulse. The effect arises due to one of the dressed states becoming stabilized against ionization. Contrasting effects from circular and linear polarized pulses demonstrate how such ``bound states in the continuum'' are sensitive to the degeneracy of the coupled continuum.

Figures (7)

• Figure 1: Conceptual picture of the proposed pump--probe scheme with a delay $\tau$. A pump pulse prepares the atom in a superposition of dressed states, $\left| + \right\rangle$ and $\left| - \right\rangle$ (left side). The atom can photoionize from the $\left| - \right\rangle$ state, but not from the $\left| + \right\rangle$ state (marked by STOP signs), as shown in the middle cartoon. By controlling the CEP phase of the probe pulse $\phi$, the final population can be targeted to either dressed state (right side). If mostly the $\left| + \right\rangle$ state is populated during the second pulse then the total amount of ionization will be suppressed, as shown in the lowest part of the cartoon.
• Figure 2: Dressed-state ionization rates extracted from the imaginary part of the corresponding quasienergies. In all cases, the blue curves correspond to the $\left| - \right\rangle$ dressed state, and the orange ones to the $\left| + \right\rangle$ dressed state. Linear polarization is used for (a) and (c), and circular polarization for (b) and (d). The top row shows results for hydrogen (with $\omega = 0.375$ a.u.), and the bottom row for helium. The calculations for He were performed with an $\omega$ corresponding to the field-free resonance for each model of He, CIS ($\omega = 0.79721$ a.u., solid lines), $V^{\textrm{He}}_1$ ($\omega = 0.78118$ a.u., dotted lines), and $V^{\textrm{He}}_2$ ($\omega = 0.77679$ a.u., dashed lines).
• Figure 3: (a) Population of $^1P$ state in helium interacting with a pump-probe scheme of linearly polarized pulses tuned at frequency $\omega=21.7$ eV with the intensity of $I = 5\times 10^{13}$ W/cm$^2$, pulse durations $\tau_1 = \tau_2 = 24.8$ fs, and delay of $t_0 = 26.6$ fs, shown as a function of time and probe CEP. (b) Time evolution of populations of the ground ($^1S$) and excited ($^1P$, $^3P$) states for the fixed probe CEP of $1.26$ rad.
• Figure 4: Bloch sphere representation of the state evolution during the pump (a) and probe (b) pulses for two CEP values of the probe pulse: $\varphi = 1.26$ rad (solid) and $\varphi = 4.40$ rad (dotted). The torque vectors of pump and probe pulses are along the $x$ and $y$ directions, respectively (see main text for details).
• Figure 5: (a) Photoelectron spectra as a function of probe CEP. The pulse parameters are identical to those of Fig. \ref{['fig:rtdcis']}, except for the intensity, which in this case is $I=1 \times 10^{14}$ W/cm$^2$. (b) Lineouts from (a) for a few values of the probe CEP. The vertical dotted line indicates the expected location of the upper component ($18.74$ eV) of the AT-doublet based on the location of the lower component ($18.15$ eV) and the Rabi frequency at the peak of the pulse ($0.587$ eV). (c) The ratio of the ionization probability of the probe, $P_2$, and pump, $P_1$, pulses as a function of probe CEP. (d) Cumulative photoelectron flux during the pump-probe sequence, for the same values of the probe CEP as in (b).