4 Femtosecond Photodissociation Dynamics by Velocity Map Imaging
85
nificantly larger degree of rotational excitation is present, while no distinction can
be made as to whether it corresponds to the light outgoing CH 3 fragment or to the
[I/I ∗ · · · ICH 3 ] co-fragment, as both imply a reduction of the total available kinetic
energy.
In combination with ab initio calculations [15] that have produced optimized geometries for the dimer and energy values and oscillator strengths for the excited
states of the A band of (CH 3 I) 2 , these results have provided solid understanding
of the blue-shift of the A-band in the (CH 3 I) 2 dimer. This shift had been reported
previously and had been qualitatively understood in terms of the ground state dimer
being stabilized by a dipole–dipole interaction, while the molecular dipole is weakened by the valence state transition, which involves the promotion of an electron
located mainly on the I atom to an antibonding molecular orbital. With a lower
dipole–dipole interaction, the dimer in the valence state will not be as stabilized as
in the ground electronic state. Donaldson et al. [52] reported an experimental shift
of the peak of the absorption band by around 500–1000 cm −1 in conditions of dimer
formation with respect to monomer-only conditions. This value is compatible with
the value of about 0.2 eV (1600 cm −1 ) found here.
In addition, a significant difference in the anisotropy observed for the monomer
and dimer must be noted. Through the use of the multidimensional fit procedure
described in the methodology section, it has been possible to discriminate the angular character of the different contributions, with the finding that, for the I ∗ ( 2 P 1/2 )
channel, no loss of orientational preference happens upon dimerization, and this
is in contrast with the I( 2 P 3/2 ) channel, which shows a pronounced decrease in
the observed anisotropy. The implications of this observation will be discussed below.
One more feature marks a difference between monomer and dimer dissociation
results in the asymptotic situation: the dramatic increase in the I( 2 P 3/2 )/I ∗ ( 2 P 1/2 )
ratio observed through the CH 3 (ν = 0) fragment, from values of 0.14 ± 0.05 for
the monomer to 0.72 ± 0.05 for the dimer. Analogous measurements were performed by tuning the probe laser to 329.5 nm and 325.8 nm, which constitute a
resonant probe of CH 3 (ν 2 = 1) and CH 3 (ν 2 = 2), respectively (ν 2 being the umbrella mode of CH 3 ). A dramatic change of the I( 2 P 3/2 )/I ∗ ( 2 P 1/2 ) ratio, in the
same direction as that just described for CH 3 (ν = 0), was also observed in both
cases.
The decrease of the anisotropy of the I( 2 P 3/2 ) channel in the dimer and the
large change in the branching ratio need to be discussed conjointly. An increase
of the I( 2 P 3/2 )/I ∗ ( 2 P 1/2 ) ratio could be related to either a change in the main
absorbing states or in the efficiency of the coupling in the non-adiabatic crossing. Since the other possibly participating states, 1 Q 1 and 3 Q 1 , are of perpendicular nature, in contrast to the parallel 3 Q 0 state, an examination of the change
in anisotropy should permit the distinction between the two phenomena. If we
consider that the I( 2 P 3/2 ) fragment observed is originated partly through absorption to 3 Q 0 followed by crossing to 1 Q 1 , and partly through direct absorption to
one of the perpendicular states ( 3 Q 1 or 1 Q 1 ), then the resulting anisotropy is expected to reflect a mixture of parallel (β = 2) and perpendicular (β = −1) character. From the observation of experimental values for the β anisotropy parameter, a
85
nificantly larger degree of rotational excitation is present, while no distinction can
be made as to whether it corresponds to the light outgoing CH 3 fragment or to the
[I/I ∗ · · · ICH 3 ] co-fragment, as both imply a reduction of the total available kinetic
energy.
In combination with ab initio calculations [15] that have produced optimized geometries for the dimer and energy values and oscillator strengths for the excited
states of the A band of (CH 3 I) 2 , these results have provided solid understanding
of the blue-shift of the A-band in the (CH 3 I) 2 dimer. This shift had been reported
previously and had been qualitatively understood in terms of the ground state dimer
being stabilized by a dipole–dipole interaction, while the molecular dipole is weakened by the valence state transition, which involves the promotion of an electron
located mainly on the I atom to an antibonding molecular orbital. With a lower
dipole–dipole interaction, the dimer in the valence state will not be as stabilized as
in the ground electronic state. Donaldson et al. [52] reported an experimental shift
of the peak of the absorption band by around 500–1000 cm −1 in conditions of dimer
formation with respect to monomer-only conditions. This value is compatible with
the value of about 0.2 eV (1600 cm −1 ) found here.
In addition, a significant difference in the anisotropy observed for the monomer
and dimer must be noted. Through the use of the multidimensional fit procedure
described in the methodology section, it has been possible to discriminate the angular character of the different contributions, with the finding that, for the I ∗ ( 2 P 1/2 )
channel, no loss of orientational preference happens upon dimerization, and this
is in contrast with the I( 2 P 3/2 ) channel, which shows a pronounced decrease in
the observed anisotropy. The implications of this observation will be discussed below.
One more feature marks a difference between monomer and dimer dissociation
results in the asymptotic situation: the dramatic increase in the I( 2 P 3/2 )/I ∗ ( 2 P 1/2 )
ratio observed through the CH 3 (ν = 0) fragment, from values of 0.14 ± 0.05 for
the monomer to 0.72 ± 0.05 for the dimer. Analogous measurements were performed by tuning the probe laser to 329.5 nm and 325.8 nm, which constitute a
resonant probe of CH 3 (ν 2 = 1) and CH 3 (ν 2 = 2), respectively (ν 2 being the umbrella mode of CH 3 ). A dramatic change of the I( 2 P 3/2 )/I ∗ ( 2 P 1/2 ) ratio, in the
same direction as that just described for CH 3 (ν = 0), was also observed in both
cases.
The decrease of the anisotropy of the I( 2 P 3/2 ) channel in the dimer and the
large change in the branching ratio need to be discussed conjointly. An increase
of the I( 2 P 3/2 )/I ∗ ( 2 P 1/2 ) ratio could be related to either a change in the main
absorbing states or in the efficiency of the coupling in the non-adiabatic crossing. Since the other possibly participating states, 1 Q 1 and 3 Q 1 , are of perpendicular nature, in contrast to the parallel 3 Q 0 state, an examination of the change
in anisotropy should permit the distinction between the two phenomena. If we
consider that the I( 2 P 3/2 ) fragment observed is originated partly through absorption to 3 Q 0 followed by crossing to 1 Q 1 , and partly through direct absorption to
one of the perpendicular states ( 3 Q 1 or 1 Q 1 ), then the resulting anisotropy is expected to reflect a mixture of parallel (β = 2) and perpendicular (β = −1) character. From the observation of experimental values for the β anisotropy parameter, a