238
M.E. Vaida and T.M. Bernhardt
IND(1) pathway exhibits a maximum at a translational energy 1.2 eV below that
of the DIR channel. This shift was attributed to a strong collision suffered by the
CH 3 on leaving the molecular layer. In addition, the IND(1) fragments exhibit a
similar angular dependence as those of the DIR channel (peaking around 40° from
the surface normal for 0.75 ML coverage). This means that the IND(1) CH 3 maintain the memory of the prior Br–C bond direction, similar to the DIR CH 3 , despite
having collided strongly enough to lose 45 % of their initial translational energy. A
molecular mechanism that would be consistent with these observations is a methyl
exchange reaction as shown schematically in the middle graphics of Fig. 10.4b.
Finally, the methyl fragments produced via the third pathway IND(2) exhibit the
lowest translational energy. More importantly, however, the angular distribution of
these fragments has been determined to be broad with approximately a cos 2 distribution, which indicates that these fragments did not retain memory of the prior
Br–C bond direction. These photofragments were attributed to CH 3 escaping from
adsorbed CH 3 Br molecules with their C–Br bond axes nearly parallel to the surface plane, which inevitably leads to strong collisions with neighboring adsorbate
molecules prior to desorption. Potential exchange reactions might also be feasible
as illustrated in the bottom graphics of Fig. 10.4b.
This example demonstrates that photoreactions induced in surface-aligned adsorbate layers might indeed exhibit fairly complex, yet nevertheless distinguishable
reaction pathways. The question whether the corresponding molecular reaction dynamics in these systems can be revealed by coherent excitation and time-resolved
detection on the timescale of the molecular motion, namely with fs resolution, will
be addressed in Sect. 10.3 of this chapter. For this purpose, in the subsequent section
an experimental technique will be put forward which permits to monitor surfacealigned reactions with fs-time-, mass-, and fragment velocity-resolution.
10.2 Surface Pump-Probe Fs-Laser Mass Spectrometry
The major novelty of the experimental time-resolved laser-spectroscopy approach
presented in this chapter consists in the detection of neutral photoreaction products
via fs time-delayed multi-photon ionization directly at the surface. A schematic illustration of the experimental arrangement used for the surface pump-probe fs-laser
mass spectrometry investigations is presented in Fig. 10.5.
The experiments are carried out in an ultra high vacuum (UHV) surface science
apparatus (base pressure < 1 × 10 −10 mbar) equipped with standard tools for surface preparation and analysis [92]. The fs-laser light is produced with a commercial
Ti:Sapphire oscillator continuously pumped by a 6 W Spectra Physics Millennia
Nd:YVO4 laser. The 800 nm laser pulses are amplified with a Nd:YLF laser pumped
Ti:Sapphire amplifier (Spectra Physics Spitfire XP) to yield 35 fs pulses with a typical power of 3 W at a repetition rate of 1 kHz. The pump laser pulses at 266 nm
are generated by frequency tripling of the fundamental wavelength in a homebuilt
third harmonic generator. The probe laser beam is tuned to a central wavelength
M.E. Vaida and T.M. Bernhardt
IND(1) pathway exhibits a maximum at a translational energy 1.2 eV below that
of the DIR channel. This shift was attributed to a strong collision suffered by the
CH 3 on leaving the molecular layer. In addition, the IND(1) fragments exhibit a
similar angular dependence as those of the DIR channel (peaking around 40° from
the surface normal for 0.75 ML coverage). This means that the IND(1) CH 3 maintain the memory of the prior Br–C bond direction, similar to the DIR CH 3 , despite
having collided strongly enough to lose 45 % of their initial translational energy. A
molecular mechanism that would be consistent with these observations is a methyl
exchange reaction as shown schematically in the middle graphics of Fig. 10.4b.
Finally, the methyl fragments produced via the third pathway IND(2) exhibit the
lowest translational energy. More importantly, however, the angular distribution of
these fragments has been determined to be broad with approximately a cos 2 distribution, which indicates that these fragments did not retain memory of the prior
Br–C bond direction. These photofragments were attributed to CH 3 escaping from
adsorbed CH 3 Br molecules with their C–Br bond axes nearly parallel to the surface plane, which inevitably leads to strong collisions with neighboring adsorbate
molecules prior to desorption. Potential exchange reactions might also be feasible
as illustrated in the bottom graphics of Fig. 10.4b.
This example demonstrates that photoreactions induced in surface-aligned adsorbate layers might indeed exhibit fairly complex, yet nevertheless distinguishable
reaction pathways. The question whether the corresponding molecular reaction dynamics in these systems can be revealed by coherent excitation and time-resolved
detection on the timescale of the molecular motion, namely with fs resolution, will
be addressed in Sect. 10.3 of this chapter. For this purpose, in the subsequent section
an experimental technique will be put forward which permits to monitor surfacealigned reactions with fs-time-, mass-, and fragment velocity-resolution.
10.2 Surface Pump-Probe Fs-Laser Mass Spectrometry
The major novelty of the experimental time-resolved laser-spectroscopy approach
presented in this chapter consists in the detection of neutral photoreaction products
via fs time-delayed multi-photon ionization directly at the surface. A schematic illustration of the experimental arrangement used for the surface pump-probe fs-laser
mass spectrometry investigations is presented in Fig. 10.5.
The experiments are carried out in an ultra high vacuum (UHV) surface science
apparatus (base pressure < 1 × 10 −10 mbar) equipped with standard tools for surface preparation and analysis [92]. The fs-laser light is produced with a commercial
Ti:Sapphire oscillator continuously pumped by a 6 W Spectra Physics Millennia
Nd:YVO4 laser. The 800 nm laser pulses are amplified with a Nd:YLF laser pumped
Ti:Sapphire amplifier (Spectra Physics Spitfire XP) to yield 35 fs pulses with a typical power of 3 W at a repetition rate of 1 kHz. The pump laser pulses at 266 nm
are generated by frequency tripling of the fundamental wavelength in a homebuilt
third harmonic generator. The probe laser beam is tuned to a central wavelength