Laser fluorescence study of the S(1D) + CD4 reaction: Determination of the SD product internal state distribution

Article abstract of DOI:10.1016/j.cplett.2004.12.044

The dynamics of the S(¹D) + CD₄ → SD + CD ₃ reaction has been studied in a photolysis-probe experiment. S( ¹D) reagent was prepared by laser photolysis of CS₂, and the SD(X²Π) product was detected by laser fluorescence excitation. The nascent rotational/fine-structure state distribution of the SD(X ²Π) product was determined. The experimentally determined rotational state distribution within each fine-structure manifold agrees well with the predictions of a statistical 'prior' distribution, while the F ₁ to F₂ fine-structure branching ratio is higher than predicted statistically. The product state distribution is consistent with reaction through the formation and decay of an energized CD₃SD complex.

Full text of DOI:10.1016/j.cplett.2004.12.044

Chemical Physics Letters 402 (2005) 265–269
www.elsevier.com/locate/cplett
Laser fluorescence study of the S(1D) + CD4 reaction:
determination of the SD product internal state distribution
*
Ani Khachatrian, Paul J. Dagdigian
Department of Chemistry, The Johns Hopkins University, 3400 North Charles Street, Baltimore, MD 21218-2685, USA
Received 1 November 2004
Abstract
The dynamics of the S(1D) + CD4 ! SD + CD3 reaction has been studied in a photolysis-probe experiment. S(1D) reagent was
prepared by laser photolysis of CS2, and the SD(X2P) product was detected by laser fluorescence excitation. The nascent rotational/
fine-structure state distribution of the SD(X2P) product was determined. The experimentally determined rotational state distribu-
tion within each fine-structure manifold agrees well with the predictions of a statistical ÔpriorÕ distribution, while the F1 to F2 fine-
structure branching ratio is higher than predicted statistically. The product state distribution is consistent with reaction through the
formation and decay of an energized CD3SD complex.
ꢀ 2004 Elsevier B.V. All rights reserved.
1. Introduction
resolved dynamical studies of reactions involving SH
products is the rapid predissociation of SH(A2R+)
[8,9], which renders laser fluorescence detection of this
radical problematic. Wittig and co-workers [10,11]
investigated the dynamics of the reaction of D atoms
with OCS in bimolecular collisions and within a van
der Waals complex. In recent work, we have determined
the rotational/fine-structure state distribution of SD
products from the S(1D) + D2 reaction [12], to provide
information complementary to that from crossed beam
studies [13,14] and theoretical treatments of the dynam-
ics [15–17] of this reaction.
The dynamics of reactions of O(1D) atoms with
hydrocarbons have been extensively studied, in part be-
cause of their importance in atmospheric chemistry. The
most important of these reactions is O(1D) + CH4. This
reaction has been studied by both laser fluorescence and
infrared chemiluminescence detection of the products
and in a crossed beam study [1–6]. As with many
O(1D) reactions, this reaction proceeds through inser-
tion to form an energized intermediate, which decays
to products, with some memory of the initial prepara-
tion of the complex [5,6]. The dominant chemical path-
way of the O(1D) + CH4 reaction is formation of
OH + CH3 products.
In this Letter, we investigate the dynamics of the
S(1D) + CH4 reaction. As in the O(1D) + CH4 reaction,
several chemical exothermic pathways are possible:
By contrast, the corresponding reactions of the S(1D)
atom have been very little studied, and only in kinetic
measurements. A critical evaluation of kinetic data [7]
recommends a room-temperature rate constant of
P5 · 10ꢀ12 cm3 moleculeꢀ1 sꢀ1 for the S(1D) + CH4
reaction. A significant factor which has impeded state-
1
Sð DÞ þ CH4 ! SH þ CH4;
DHo0 ¼ ꢀ26:7 ꢁ 0:8 kJ molꢀ1
! H2CS þ H2;
;
ð1aÞ
ð1bÞ
DHo0 ¼ ꢀ200 ꢁ 8 kJ molꢀ1
:
The quoted reaction exothermicities were computed
from available heats of formation and the S atomic
excitation energy [18–21]. It should also be noted that
*
Corresponding author. Fax: +1 410 516 8420.
E-mail address: pjdagdigian@jhu.edu (P.J. Dagdigian).
0009-2614/$ - see front matter ꢀ 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.cplett.2004.12.044

266
A. Khachatrian, P.J. Dagdigian / Chemical Physics Letters 402 (2005) 265–269
formation of CH3SH from S(1D) + CH4 is exothermic
by 330.9 0.4 kJ molꢀ1. In the O(1D) + CH4 reaction,
several other exothermic channels are available, involv-
ing formation of H and H2, but the OH + CH3 pathway
dominates (branching ratio determined to be 77% [6]).
Similarly, we expect pathway Eq. (1a) to be dominant
in the S(1D) + CH4 reaction.
Laser fluorescence excitation is employed here for the
determination of the rotational/fine-structure state dis-
tribution of the SD product from the S(1D) + CD4 isoto-
pic analog of reaction Eq. (1a) ½DHo0 ¼ ꢀ22:2 kJ molꢀ1ꢂ,
using the SD A2R+ ꢀ X2P electronic transition [22]. The
deuterated version of reaction Eq. (1a) has been studied
because the excited-state predissociation is sufficiently
slow in the deuterated isotopomer [8,9] that SD laser
fluorescence excitation spectra with good signal-to-noise
ratio could be recorded.
gies of the room-temperature CD4 reagent are 3.7 and
0.4 kJ molꢀ1, respectively, and the internal excitation
of the S(1D) reagent has already been taken into account
in the computation of the exothermicity for the
reaction. The average relative translational energy
(19.8 kJ molꢀ1) was obtained from the relative transla-
tional energy distribution calculated [12] by convolution
of the velocity of the S(1D) fragment from 193 nm pho-
tolysis of CS2 with an isotropic Maxwellian velocity dis-
tribution of the CD4 reagent. Because of the greater
mass of CD4 vs. D2, the velocity of S(1D) is more effi-
ciently converted into relative translational energy of
the reagents than in the S(1D) + D2 reaction. The total
energy available to the SD + CD3 products of the
S(1D) + CD4 reaction is thus 46.1 kJ molꢀ1. This energy
is sufficient to allow formation of SD products in the
v = 0 and 1 vibrational levels. Within v = 0, rotational
levels as high as J = 27.5 and 25.5 within the F1
(X = 3/2) and F2 (X = 1/2) manifolds, respectively, can
be populated.
2. Experimental
Laser fluorescence excitation spectra of the SD
A2R+ ꢀ X2P (0,0) band, with observation of emission
in the (0,1) band, were recorded in order to determine
the rotational/fine-structure state distribution of the
SD(X2P, v = 0) product of the S(1D) + CD4 reaction.
Excitation spectra were taken as a function of the delay
between the photolysis and probe lasers. The excitation
spectra at the shortest delays (200 and 350 ns) were quite
similar, and the spectra at 200 ns were employed to com-
pute the nascent SD internal state distribution. We also
searched for the presence of SD(X2P, v = 1) product by
scanning through the (0,1) band, with detection of emis-
sion in the (0,0) band. We were not able to detect vibra-
tionally excited SD products.
The relative intensities of the resolved SD lines were
converted to populations with the help of fluorescence
excitation line strength factors [25], assuming isotropic
MJ distributions. We also took into account the fact
that the spectrometer transmitted only a portion of the
detected SD A2R+ ꢀ X2P (0,1) emission band, namely
near the R1 head. Fluorescence excitation factors were
calculated with inclusion of the spectrometer transmis-
sion factor for all the emission lines associated with each
transition in the fluorescence excitation spectrum. For
most of the SD rotational/fine-structure levels, it was
possible to obtain redundant values of the populations
from the intensities of different lines originating from
the same level, while in some cases spectral congestion
prevented determination of the relative populations
from all the allowed lines. The populations were deter-
mined from averages of the populations from different
resolved excitation lines.
This study was carried out in the same apparatus em-
ployed in our recent study [12] of the S(1D) + D2 reac-
tion and has been described in detail previously
[12,23,24]. A brief description of the experiment is pre-
sented here. The S(1D) reagent was prepared by
193 nm photolysis of CS2 (1 mTorr) in a quasi-static
cell. The CD4 reagent (200 mTorr) and argon
(300 mTorr, to retard deposition of sulfur on the win-
dows) were introduced through the ends of the sidearms,
and the gas mixture was flowed slowly through the cell.
The excimer photolysis laser (Lambda Physik COMPex
102) and probe laser (frequency-doubled signal output
of a Continuum Sunlite EX optical parametric oscilla-
tor) radiation were counterpropagated through sidearms
on the cell.
Laser fluorescence excitation spectra were recorded in
order to determine the internal state distribution of the
SD reaction product. The SD fluorescence was detected
with a photomultiplier/gated-integrator combination
through a 1 m spectrometer operated at a resolution of
255 cmꢀ1. To minimize the effect of predissociation in
the SD(A2R+) state, the width of the integrator gate
was set to detect only a small portion (30 ns, delayed
30 ns after the probe laser pulse) of the fluorescence de-
cay profile [12]. Spectra were recorded with probe laser
pulse energies of 0.25 mJ in a 0.35 cm diam beam. Previ-
ous work [12] indicated that these conditions were with-
in the linear excitation regime.
3. Results
The detection sensitivity for the individual levels is
also affected by the increasing predissociation rate of
the SD(A2R+) excited levels as a function of N0 [8,9].
The use of a short integrator gate mitigates the effect
The total energy available to the products is the sum
of DHo0 and the internal and translational energies of the
reagents. The average rotational and vibrational ener-

A. Khachatrian, P.J. Dagdigian / Chemical Physics Letters 402 (2005) 265–269
267
of the varying excited-state predissociation rate on the
4. Discussion
relative detection sensitivity of different levels, and a cor-
rection for the effect of predissociation has been applied
to the measured intensities [12]. This correction becomes
increasingly large with increasing N0, as discussed previ-
ously [12].
The present study reports the internal state distribu-
tion of the SD(X2P) product from the reaction of
S(1D) with CD4, in particular the branching into the
spin–orbit and K doublet levels. Because of the presence
of the deeply bound CD3SD well, which can be accessed
from the reagents, this reaction should occur by inser-
tion dynamics, involving the formation and decay of a
collision complex. It is thus interesting to compare our
experimentally determined rotational state distribution
with a prediction from statistical theory. We compare
in Fig. 2 the measured state distribution with the so-
Fig. 1 presents the derived relative populations of the
rotational fine-structure levels of the SD(X2P) product
from the S(1D) + D2 reaction. The rotational state distri-
bution within each fine-structure manifold is fairly
broad, and the populations peak around J ꢃ 10.5. A sig-
nificant preference for formation of the lower, F1 fine-
structure manifold can be observed. Summing over all
observed rotational levels, the ratio of the population
in the F1 to F2 manifolds is found to be 3.23 0.13.
The populations of the K doublet levels of A0 and A00
symmetry [26], in the high-J limit, are separately plotted
in the two panels of Fig. 1. It can be seen that the pop-
ulations of the K doublet levels of A0 and A00 symmetry
for the same fine-structure manifold and rotational
angular momentum are essentially the same. The ratios
of the population of A0 to A00 levels, summed over all
observed rotational levels, are computed to equal
1.11 0.06 and 0.92 0.08 in the F1 and F2 manifolds,
respectively.
called ÔpriorÕ distribution [27]:
X
rðv; J; F iÞ ¼
rðv; J; F i; v1v2v3v4; JC; KCÞ;
ð2Þ
v1v2v3v4JCKC
where r(v,J,Fi;v1v2v3v4,JC,KC) is the correlated-product
state-resolved prior cross-section:
rðv; J; F i; v1v2v3v4; JC; KCÞ
1=2
/ ð2J þ 1Þð2JC þ 1Þ½E ꢀ eðv; J; F i; v1v2v3v4; JC; KCÞꢂ
:
ð3Þ
The sum over CD3 internal states in Eq. (2) reflects the
fact that we do not observe the CD3 product and hence
do not resolve its product states. In Eq. (3), E is the total
energy available to the SD + CD3 products, and
e(v,J,Fi;v1v2v3v4,JC,KC) is the internal energy of the cor-
related pair of SD and CD3 product states. Since their
energies are essentially identical, the SD(X2P) K doublet
levels associated with a given J and fine-structure
manifold are predicted to have equal populations.
We see in Fig. 2 that the shape of the experimentally
determined relative populations within a given fine-
structure manifold (F1 or F2) is well reproduced by a
0.05
F (A')
1
F (A")
1
0.04
0.03
0.02
0.01
0.05
0.04
0.00
0.05
F (A')
2
F (A")
2
0.04
0.03
0.02
0.01
0.00
F1 prior
0.03
0.02
F2 prior
0.01
0.00
0
2
4
6
8
10 12 14 16 18
0
5
10
15
20
25
rotational angular momentum J
rotational angular momentum J
Fig. 1. The derived rotation fine structure state distribution of
SD(X2P, v = 0) products from the reaction of S(1D) with CD4. Upper
panel: F1 (X = 3/2) fine-structure manifold; lower panel: F2 (X = 1/2)
fine-structure manifold. The distributions have been normalized so
that the sum of the populations in the v = 0 vibrational level is unity.
Fig. 2. Comparison of the experimentally determined product
SD(X2P, v = 0) rotational fine-structure state distribution (symbols)
with a prior statistical distribution (solid lines). The latter has been
scaled so that the distribution in the F1 fine-structure manifold matches
the experimental distribution.

268
A. Khachatrian, P.J. Dagdigian / Chemical Physics Letters 402 (2005) 265–269
statistical distribution. This good agreement of the
experimental and prior state distributions is consistent
with formation and decay of an energized CD3SD com-
plex in the dynamics of the S(1D) + CD4 reaction.
While there is sufficient energy to populate SD rota-
tional levels up to J = 25.5–27.5, the experimental and
statistical rotational state distributions peak at much
lower values of J. It is also interesting to note that the
degree of rotational excitation of the SD product from
the S(1D) + CD4 reaction is much less than that
for the SD product of the S(1D) + D2 reaction [12], de-
spite the fact that both reactions proceed by formation
and decay of an energized complex and the energy avail-
able to the former reaction is significantly greater than
for the latter. This reflects the approximately statistical
partitioning of the available energy into the SD and
CD3 internal degrees of freedom.
spin–orbit constant A and the rotational constant B
(A/B = ꢀ7.5 and ꢀ76.9, respectively [22,28]). The small
value of this ratio for OH(X2P) implies that this state
approaches the HundÕs case (b) limit at fairly low values
of J. Within the case (b) limit, the K doublet levels can
display different symmetries of the electronic wave func-
tion with respect to the plane of molecular rotation, and
collisional K doublet propensities can be anticipated
[29]. By contrast, SD(X2P) remains near the HundÕs
case (a) limit for all rotational levels energetically acces-
sible in the S(1D) + CD4 reaction.
Acknowledgement
This research has been supported by the National Sci-
ence Foundation under Grant No. CHE-0413743.
We observe in Fig. 2 that a slight preference for for-
mation of the lower-energy F1 fine-structure manifold
over that for the F2 manifold is predicted in the prior
distribution. The ratio of the total populations in the
F1 to F2 manifolds is computed for the prior distribution
to equal 1.64. The slightly unequal populations result
from the slightly greater energy available to relative
translational energy and CD3 internal energy in the case
of SD products formed in the F1 manifold, and resulting
larger density of available states, than for products
formed in the F2 manifold. Our experimentally deter-
mined SD(X2P) internal state distribution displays an
even greater propensity for formation of products in
the F1 manifold. As noted above, we determine the
fine-structure branching ratio to equal 3.23 0.13.
This fine-structure propensity contrasts with the lack
of such a propensity in the isoelectronic O(1D) + CH4
reaction [1,2]. The exothermicity of this reaction
ðDHo0 ¼ ꢀ191:3 ꢁ 1:2 kJ molꢀ1Þ is much greater than
that of the S(1D) + CH4 reaction. Moreover, the spin–
orbit constant A for OH(X2P) is much less than for
SH(X2P) (A = ꢀ139 and ꢀ377 cmꢀ1, respectively [28]).
Thus, the fine-structure splitting is a much smaller frac-
tion of the total energy available to the products of
O(1D) + CH4 than for S(1D) + CH4. The observed
enhancement of the production of SD in the lower, F1
fine-structure manifold over that predicted by the prior
distribution could be the result of a tendency of the ener-
gized CD3SD complex to dissociate adiabatically to the
lowest fine-structure state.
References
[1] A.C. Luntz, J. Chem. Phys. 73 (1980) 1143.
[2] C.R. Park, J.R. Wiesenfeld, J. Chem. Phys. 95 (1991) 8166.
[3] P.M. Aker, J.J.A. OÕBrien, J.J. Sloan, J. Chem. Phys. 84 (1986)
745.
[4] S. Satayapal, J. Park, R. Bersohn, B. Katz, J. Chem. Phys. 91
(1989) 6873.
[5] M. Brouard, H.M. Lambert, C.L. Russell, J. Short, J.P. Simons,
Faraday Discuss. 102 (1995) 179.
[6] J.J. Lin, J. Shu, Y.T. Lee, X. Yang, J. Chem. Phys. 113 (2000)
5287.
[7] K. Schofield, J. Phys. Chem. Ref. Data 8 (1979) 723.
[8] M. Kawasaki, H. Sato, G. Inoue, M. Suzuki, J. Chem. Phys. 91
(1989) 6758.
[9] M.D. Wheeler, A.J. Orr-Ewing, M.N.R. Ashfold, J. Chem. Phys.
107 (1997) 7591.
[10] D. Hausler, J. Rice, C. Wittig, J. Phys. Chem. 91 (1987) 5413.
[11] E. Bohmer, K. Mikhaylichenko, C. Wittig, J. Chem. Phys. 99
(1993) 6545.
[12] A. Khachatrian, P.J. Dagdigian, J. Chem. Phys. 122 (2005) in
press.
[13] S.-H. Lee, K. Liu, Chem. Phys. Lett. 290 (1998) 323.
[14] S.-H. Lee, K. Liu, in: R. Campargue (Ed.), Atomic and
Molecular Beams: The State of the Art 2000, Springer, Berlin,
2000, p. 543.
[15] S.D. Chao, R.T. Skodje, J. Phys. Chem. A 105 (2001) 2474.
[16] L. Banares, F.J. Aoiz, P. Honvault, J.-M. Launay, J. Phys. Chem.
A 108 (2004) 1616.
[17] B. Maiti, G.C. Schatz, G. Lendvay, J. Phys. Chem. A 108 (2004)
8772.
[18] M.W. Chase Jr., J. Phys. Chem. Ref. Data, Monograph No. 9. J.
Phys. Chem. Ref. Data, Monograph No.9, 1998.
[19] R.C. Shiell, X.K. Hu, Q.J. Hu, J.W. Hepburn, J. Phys. Chem. A
104 (2000) 4339.
The populations of the K doublet levels associated
with a particular rotational angular momentum J and
fine-structure manifold of the SD(X2P) product from
the S(1D) + CD4 reaction have been found to have
essentially equal populations. This observation contrasts
with the K doublet propensities observed [1,2] in the
OH(X2P) product of the O(1D) + CH4 reaction. One
very significant difference between the SD(X2P) and
OH(X2P) states is the magnitude of the ratio of the
[20] NIST Atomic Spectra Database v2.0, 1999.
[21] B. Ruscic, J. Berkowitz, J. Chem. Phys. 98 (1993) 2568.
[22] D.A. Ramsay, J. Chem. Phys. 20 (1952) 1920.
[23] B. Nizamov, P.J. Dagdigian, Y.-R. Tzeng, M.H. Alexander, J.
Chem. Phys. 115 (2001) 800.
[24] X. Tan, P.J. Dagdigian, Chem. Phys. Lett. 375 (2003) 532.
´
[25] P.J. Dagdigian, in: G. Scoles, U. Buck, D. Laine (Eds.), Atomic
and Molecular Beam Methods, Oxford U.P., New York, 1988, pp.
596–630.

A. Khachatrian, P.J. Dagdigian / Chemical Physics Letters 402 (2005) 265–269
269
[26] M.H. Alexander, P. Andresen, R. Bacis, R. Bersohn, F.J. Comes,
P.J. Dagdigian, R.N. Dixon, R.W. Field, G.W. Flynn, K.-H.
Gericke, E.R. Grant, B.J. Howard, J.R. Huber, D.S. King, J.L.
Kinsey, K. Kleinermanns, K. Kuchitsu, A.C. Luntz, A.J. Mac-
Caffery, B. Pouilly, H. Reisler, S. Rosenwaks, E. Rothe, M.
Shapiro, J.P. Simons, R. Vasudev, J.R. Wiesenfeld, C. Wittig,
R.N. Zare, J. Chem. Phys. 89 (1988) 1749.
[27] R.D. Levine, B.B. Bernstein, Molecular Reaction Dynamics
and Chemical Reactivity, Oxford University Press, New York,
1987.
[28] K.P. Huber, G. Herzberg, Molecular Spectra and Molecular
Structure. IV. Constants of Diatomic Molecules, Van Nostrand
Reinhold, New York, 1979.
[29] M.H. Alexander, P.J. Dagdigian, J. Chem. Phys. 80 (1984) 4325.