10570
Gericke, Kreher, and Reinsch: Determination of reaction geometries
TABLE I. Observed steric effect, (IЌϪIʈ)/(IЌϩIʈ), in the reaction of
␥
0
͐
maxP ␥,␦ϭ90°͒sin ␥͒d␥
͑ ͑
J
HϩHCN(3) with 3ϭ2,4 and ClϩHCN ͑3ϭ2, Jϭ9͒. The error of the
intensity ratio is below 0.01. The positive values indicate a linear reaction
geometry for both the Cl and the H atom reaction.
SE ␥ ͒ϭ
.
͑9͒
͑
max
␥
0
͐
maxP ␥,␦ϭ0°͒sin ␥͒d␥
͑ ͑
J
The integration of this equation can easily be performed.8
It should be mentioned that depolarization effects due to
the earth magnetic field and the nuclear spin of reactant B are
not considered.10–12 Precession of the nuclear spin leads to a
decrease of the initial molecular alignment. The depolariza-
tion is stronger for low J values. For high rotations the total
Reaction
Transition Delay time/ns (IЌϪIʈ)/(IЌϩIʈ)
50
50
0.065
0.022
0.00
HϩHCN ͑3ϭ2, Jϭ1͒
HϩHCN ͑3ϭ2, Jϭ6͒
HϩHCN ͑3ϭ2, Jϭ0͒
HϩHCN ͑3ϭ2, Jϭ5͒
HϩHCN ͑3ϭ4, Jϭ9͒
ClϩHCN ͑3ϭ2, Jϭ9͒
R(0)
R(5)
P(1)
P(6)
R(8)
R(8)
50
50
0.01
230
230
0.043
0.029
angular momentum F is essentially determined by J, be-
5
2
cause typical values of I, the nuclear spin, are below . In
the high J limit A(02) becomes independent of the excitation
branch and Eq. ͑7͒ reduces to (␦ϭ0),
The measured intensity ratios (IЌϪIʈ)/(IЌϩIʈ) for
different reactions are summarized in Table I, where the

P ␥,0͒ϰ 1ϩ
P2 cos ␥͒ .
͑10͒
͑
͑
ͩ
ͪ
10
ʈ
subscripts
or Ќ indicate a parallel or perpendicular
ជ
ជ
alignment between EJ and EV . The polarization of the lasers
were rotated by using /2 wave plates22 of zeroth order. The
beam walk has not been observed. Four hundred laser shots
for each polarization scheme were averaged to obtain the
data. The data for P(1) excitation were used to minimize
systematic errors due to beam walk or lengthy data
averaging. The observed deviation for P(1) excitation was
always below Ϯ0.01. The positive intensity ratio
demonstrates a preferred end-on attack of the hydrogen or
the chlorine atom on the hydrogen atom of the HCN
molecule. Thus, a linear reaction geometry is preferred for
both the HϩHCN→H2ϩCN and the ClϩHCN→HClϩCN
reaction. The observed intensity ratios of the different tran-
sitions at a delay time of 50 ns behave as one expects when
Eq. ͑2͒ is considered. However, HCN exhibits a nuclear spin
and a quadrupole moment that will strongly depolarize the
initial aligned HCN. Since this effect becomes negligible for
high rotations ͓Eq. ͑10͔͒ or for delay times р50 ns, the
The influence of the nuclear spin on the polarization mea-
surements can also be neglected, if the time of precession
p
is long compared to the observation ͑delay͒ time . On the
other hand, if the interaction between J and I is strong, i.e.
pӶ, then a time-averaged depolarization coefficient can
be calculated.12
The derivation of the above equations further assumes
stationary target and precursor molecules. However, Gilbert
et al.13 have shown that the degradation in alignment is not
likely to compromise seriously the viability of the experi-
ment. Even if the reduction in alignment cannot be neglected
completely in some cases ͑very heavy attacking atom and a
very light target molecule͒, it can be adequately treated
through an effective  parameter ͓Eq. ͑9͒ of Ref. 13͔.
III. COMPARISON WITH EXPERIMENTAL RESULTS
We have studied the reaction of aligned hydrogen atoms
generated in the 266 nm photodissociation of methylmercap-
tan, CH3SH, with HCN being excited ͑around 1.53 m͒ in a
range of the angle of attack, 0,␥
, can be extracted
͔
͓
max
from Eq. ͑9͒.22 We obtain ␥max HϩHCN( ϭ2) ϭ50°,
͓
͔
3
␥
HϩHCN( ϭ4) ϭ55°, and ␥
ClϩHCN( ϭ2)
͓ ͔
max 3
͓
max
͔
3
single rotational state of the first overtone of CH stretch-
3
ϭ65°. Since the experimental error is ca.Ϯ5°, we assume
that the cone of acceptance is slightly larger for the Cl reac-
tion in comparison to the HϩHCN reaction.
ing motion. Details of the experimental setup are given
elsewhere.14,15 The

parameter for the CH3SH
ϩh→HϩCH3S process is known to be ХϪ0.9616,17 and,
thus, a reaction plane is defined in the lab frame, as indicated
by Fig. 1. The reaction HϩHCN is endothermic and only
rovibrationally excited molecules can react with the pho-
tolytically generated hydrogen atoms.18,19 The collision en-
ergy and its distribution can be extracted from the experi-
mental work of Wilson et al.15 and is published elsewhere.20
Chlorine atoms are generated in the photodissociation of Cl2
at 355 nm and its spatial distribution is described by
ϭϪ1.21 The CN product molecules were analyzed by LIF.
The delay time between the CN analysis pulse and the HCN
excitation laser pulse was 50 and 250 ns, depending on the
system under investigation. Since the partial pressures were
slightly below 10 Pa single collision conditions are expected
and the product yield is directly proportional to the reactivity
In principle, there is the possibility of significant reagent
reorientation on an approach to the transition state. However,
all performed measurements with rovibrationally excited
HCN as well as all trajectory studies indicate a similar be-
havior for both the H and the Cl reaction,15,20 and a long
living transition state can be excluded. Since the approach of
the light and fast hydrogen atom is unlikely to reorientate the
HCN, we assume no substantial reagent reorientation.
IV. SUMMARY
The reaction geometry can be analyzed even in bulk ex-
periments when polarized light is used. Equations have been
developed that allow a relatively easy conversion of experi-
mental results to the angle of attack, ␥. At high reactant
rotations the dynamic range of the signal intensities at dif-
ferent polarization schemes is fairly low, but sufficient to
obtain a quantitative impression of the attack angle. For low
2
of the ͑aligned͒ reactants. The CN (2⌺←X ⌺) transition is
saturated and, thus, any influence of aligned CN products on
the signal can be neglected.
J. Chem. Phys., Vol. 107, No. 24, 22 December 1997
152.2.176.242 On: Mon, 01 Dec 2014 02:59:34
Gericke, Karl-Heinz
