Observation of Light-Phase-Sensitive Photoemission from a Metal

We demonstrate that multiphoton-induced photoelectron emission from a gold surface caused by low-energy (unamplified) 4-fs, 750-nm laser pulses is sensitive to the timing of electric field oscillations with respect to the pulse peak. This observation confirms recent theoretical predictions and opens the door to measuring the absolute value of the carrier-envelope phase difference of few-cycle light pulses with a solid-state detector.

Figure 1

Schematic of the experiments. A 10-fs phase-controlled pulse train passes through a 1.5-mm-long single-mode fiber and a dispersive delay line consisting of ultrabroadband chirped mirrors to produce sub-5-fs pulses at a 24 MHz repetition rate. The carrier-envelope phase difference of the pulses can be shifted by known amounts by translation of one of a pair of thin fused-silica wedges. They are focused with an off-axis parabola onto a gold photocathode. The multiphoton-induced photocurrent is preamplified by an electron multiplier and selectively amplified by a lock-in amplifier triggered by the reference signal $R(t)$ at $f_{\mathrm{r}\mathrm{e}\mathrm{f}}=1\mathrm{M}\mathrm{H}\mathrm{z}$.

Figure 2

The reference signal $R(t)=R_0\cos (2\pi f_{\mathrm{r}\mathrm{e}\mathrm{f}}t)$ along with the modulated signal component $S(t)={\mathrm{S}}_0\cos (2\pi f_{\mathrm{r}\mathrm{e}\mathrm{f}}t+\theta )$ of the photocurrent. The phase shift $\theta$ can be directly measured by phase-sensitive lock-in detection. With the carrier-envelope phase difference maximizing the number of extracted photoelectrons inferred from a numerical study and intuitive considerations: ${\phi }_{\max }=-\pi /4$, the carrier-envelope phase difference can be determined in any pulse of the train as shown for a few representative cases assuming a somewhat idealized Gaussian pulse envelope. Knowledge of $\theta$ and ${\phi }_{\max }$ creates an unambiguous connection between $\phi$ and the instantaneous phase of the reference signal, allowing the absolute value of the carrier-envelope phase difference to be referenced directly to $R(t)$.

Figure 3

In-phase component, $S_0\cos \theta$, of the modulation of the photocurrent, $S(t)$, as a function of the change in path length through the fused silica glass wedges shown in Fig. 1. (a),(b) Photoemission signal recorded with pulses of a peak intensity of $I_p\approx 2\times {10}^{12}\mathrm{W}/{\mathrm{c}\mathrm{m}}^2$ and a duration (full width at half maximum) of ${\tau }_L=4.5\mathrm{f}\mathrm{s}$ and 4.0 fs, respectively. The experimental data (triangles) are corrected for a constant (nonoscillating) phase offset of electronic origin. The lines are obtained by modeling the decrease of the photocurrent using the power law ${\mathrm{S}}_0\sim I_p^x$ with $x=3.0$ and taking into account dispersive pulse broadening. Although the pulses broaden only by a few percent upon traveling a distance of a few tens of micrometers in fused silica, the resultant decrease in their peak intensity is sufficient to notably lower the photocurrent owing to the rapid $I_p^x$ scaling discussed above. $S_0$ decays faster in (b) simply because the shorter pulse broadens more rapidly upon propagation.