Momentum-map-imaging photoelectron spectroscopy of fullerenes with femtosecond laser pulses

Photoelectron spectra of ${\mathrm{C}}_{60}$ and ${\mathrm{C}}_{70}$, ionized with ultrashort laser pulses, are measured with a momentum-map-imaging electron spectrometer. Above the photon energy, 1.6 eV, the spectra are essentially structureless and well described by Boltzmann distributions, with temperatures on the order of ${10}^4$ K. This result is similar to previous results for ${\mathrm{C}}_{60}$ using a time-of-flight electron spectrometer, which are confirmed in this study. Comparisons of electron energy distributions recorded for identical laser intensities but for different pulse durations demonstrate unambiguously that a significant fraction of the electrons are emitted in quasithermal processes and results argue strongly against a field ionization or direct multiphoton ionization mechanism. For electron energies above the photon energy, which account for about half the intensity, the whole signal is consistent with this quasithermal emission.

Figure 1

A schematic picture of the electron spectrometer. See main text for more details.

Figure 2

(a) Raw data, momentum map image of xenon after femtosecond laser ionization. (b) Inverted image, representing the $p_z=0$ slice of the original three-dimensional distribution. The whole three-dimensional distribution can be reconstructed using the cylindrical symmetry, caused by the laser polarization. The laser peak intensity is $6.4\times {10}^{13}$ W/cm${}^2$. The laser polarization is in the $y$ direction, parallel to the detector; $p_x$ and $p_y$ correspond to the momentum of the photoelectrons. Colors (online only) are saturated for display purposes.

Figure 3

Energy calibration of the electron spectrometer using the ATI peaks of ionized xenon atoms. The arrows have a length of 1.59 eV, which is the centroid photon energy of the fs laser and the distance between the ATI peaks.

Figure 4

$C_{60}$ momentum map, at laser fluence 1.84 J/cm${}^2$: (a) raw image, linear scale; (b) inverted image, logarithmic scale. The intense central rings are associated with Rydberg states [19]. The laser polarization is in the $y$ direction, parallel to the detector; $p_x$ and $p_y$ correspond the momentum of the photoelectrons. Colors (online only) are saturated for display purposes.

Figure 5

$C_{70}$ momentum map. All other data are the same as in Fig. 4.

Figure 6

$C_{60}$ electron and mass spectra. From top to bottom, the fs-pulse fluences are 2.19, 1.84, 1.70, 1.56, 1.42, 1.27, and 1.13 J/cm${}^2$ in the electron spectra. The spectra are cut at the noise level. The bottom curve corresponds to ns excitation. The mass spectrum corresponds to the fs-laser fluence 1.13 J/cm${}^2$.

Figure 7

$C_{70}$ electron and mass spectra. The laser fluences and durations are identical to those of Fig. 6.

Figure 8

The apparent electronic temperature as a function of laser fluence. Squares, ${\mathrm{C}}_{70}$; circles, ${\mathrm{C}}_{60}$. The dotted line is the fit to the temperatures vs laser fluence for ${\mathrm{C}}_{60}$ from [5]. The laser fluence has an overall systematic uncertainty of 10%, which is not included in the error bars.

Figure 9

Electronic temperature of ${\mathrm{C}}_{70}$ photoelectrons as a function of peak intensity. Crosses correspond to a fixed pulse duration, 150 fs, and varying pulse energy. Open squares correspond to a fixed laser fluence, 1.84 J/cm${}^2$, and varying pulse duration from 150 up to 1528 fs. The uncertainties in temperature are those of the fitting procedure, and those of the intensities are mainly from the autocorrelation function. The systematic uncertainty in the absolute fluence calibration is irrelevant for the comparison here and is not included. The arrows indicate the spectra used in Fig. 10.

Figure 10

Photoelectron spectra from ${\mathrm{C}}_{70}$ for stretched and short pulses with approximately the same electron temperatures. (a) Black, dotted, curve: 150 fs, 1.27 J/cm${}^2$, $T_a=1.24$ eV; red, full, curve: 680 fs, 1.84 J/cm${}^2$, $T_a=1.24$ eV. (b) Black, dotted, curve: 150 fs, 1.42 J/cm${}^2$, $T_a=1.32$ eV; blue, full, curve: 340 fs, 1.84 J/cm${}^2$, $T_a=1.30$ eV. The spectra are scaled to give identical values of the thermal parts above the photon energy, 1.59 eV.

Figure 11

The electronic temperatures for different pulse duration at fixed laser fluence (1.84 J/cm${}^2$) for ${\mathrm{C}}_{70}$. The experimental data are the stretched pulse data from Fig. 9. The simulated curves are the average electronic temperature of the ion at the time of emission, calculated for two different electron-vibrational coupling times. The dashed curve corresponds to a coupling time of 400 fs and a total of 190 photons (300 eV) absorbed in the pulse. The dotted line corresponds to 600 fs and 100 photons (160 eV). The lines are not fit curves and several sets of parameters give similar curves.