Depletion kinetics of Mo(a7S3,a5S2,a5D J) by N2, SO2, CO2, N2O, and NO

Article abstract of DOI:10.1021/jp9532172

The gas phase collisional disappearance of Mo(a⁷S₃,a⁵S₂,a⁵D _J) in the presence of N₂, SO₂, CO₂, N₂O, and NO over the temperature range 294-621 K and in the total pressure range 10-600 Torr is reported. Mo atoms were produced by the 248 nm photodissociation of Mo(CO)₆ and MoCl₄ and detected by laser-induced fluorescence. The room temperature depletion rate constant of Mo(a⁵S₂) + N₂ is (2.3 ± 0.7) × 10^-11 cm³ s^-1; the depletion rate constants of the a⁵D_J are smaller and range from (16 ± 8) × 10^-12 for Mo(a⁵D₀) to (0.13 ± 0.04) × 10^-12 cm³ s^-1 for the other spin-orbit states. The depletion rate of all states of Mo by SO₂ are on the order of the collision rate. Mo(a⁷S₃) is found to be unreactive toward CO₂, and the rate constant for the reaction of Mo(a⁷S₃) with N₂O is expressed as k(T) = (2.0 ± 0.5) × 10^-10 exp[-(9.8 ± 0.3) kcal mol^-1/RT] cm³ s^-1. Termolecular kinetics are observed for Mo(a⁷S₃) + NO with k₀ = (2.6 ± 0.3) × 10^-29 cm⁶ s^-1, k_∞ = (5.8 ± 0.5) × 10^-11 cm³ s^-1, and F_c = 0.72 ± 0.07 at 296 K; the reaction rates for this reaction decrease with increasing temperature. Depletion rate constants of the reactions of the excited states of Mo with CO₂, N₂O, and NO are on the order of 10^-12-10^-10 cm³ s^-1; the depletion kinetics are complex and involve significant energy transfers. Results indicate that the electron configuration of Mo plays a role in its depletion kinetics. In all cases, Mo(4d⁵5s¹a⁵S₂) depletes faster than Mo(4d⁴5s²a⁵D_J). The inefficient reactions of Mo(a⁷S₃) with CO₂ and N₂O are attributed to the production of spin-forbidden states.

Full text of DOI:10.1021/jp9532172

7502
J. Phys. Chem. 1996, 100, 7502-7510
Depletion Kinetics of Mo(a7S3,a5S2,a5DJ) by N2, SO2, CO2, N2O, and NO
Roy E. McClean,* Mark L. Campbell, and Robert H. Goodwin
Chemistry Department, United States NaVal Academy, Annapolis, Maryland 21402
ReceiVed: October 31, 1995; In Final Form: January 30, 1996X
The gas phase collisional disappearance of Mo(a7S3,a5S2,a5DJ) in the presence of N2, SO2, CO2, N2O, and
NO over the temperature range 294-621 K and in the total pressure range 10-600 Torr is reported. Mo
atoms were produced by the 248 nm photodissociation of Mo(CO)6 and MoCl4 and detected by laser-induced
fluorescence. The room temperature depletion rate constant of Mo(a5S2) + N2 is (2.3 ( 0.7) × 10-11 cm3
s-1; the depletion rate constants of the a5DJ are smaller and range from (16 ( 8) × 10-12 for Mo(a5D0) to
(0.13 ( 0.04) × 10-12 cm3 s-1 for the other spin-orbit states. The depletion rate of all states of Mo by SO2
are on the order of the collision rate. Mo(a7S3) is found to be unreactive toward CO2, and the rate constant
for the reaction of Mo(a7S3) with N2O is expressed as k(T) ) (2.0 ( 0.5) × 10-10 exp[-(9.8 ( 0.3) kcal
mol-1/RT] cm3 s-1. Termolecular kinetics are observed for Mo(a7S3) + NO with k0 ) (2.6 ( 0.3) × 10-29
cm6 s-1, k∞ ) (5.8 ( 0.5) × 10-11 cm3 s-1, and Fc ) 0.72 ( 0.07 at 296 K; the reaction rates for this reaction
decrease with increasing temperature. Depletion rate constants of the reactions of the excited states of Mo
with CO2, N2O, and NO are on the order of 10-12-10-10 cm3 s-1; the depletion kinetics are complex and
involve significant energy transfers. Results indicate that the electron configuration of Mo plays a role in its
depletion kinetics. In all cases, Mo(4d55s1a5S2) depletes faster than Mo(4d45s2a5DJ). The inefficient reactions
of Mo(a7S3) with CO2 and N2O are attributed to the production of spin-forbidden states.
TABLE 1: Electronic Configurations and LIF Excitation
Informationa
Introduction
We recently reported an investigation of the reaction kinetics
of Mo(a7S3,a5S2,a5DJ,a5GJ) with O2.1 That study was an effort
to understand the dynamics of Mo oxidation reactions and to
determine possible oxidation mechanisms of transition metal
(TM) atoms. We found that the reaction kinetics of Mo with
O2 were independent of total pressure (Ar buffer), observations
consistent with bimolecular reaction kinetics. The a7S3, a5S2,
and a5GJ states have measured rate constants on the order of
the collision rate constant, and the a5DJ spin-orbit states have
smaller rate constants that are temperature dependent. Results
were interpreted in terms of a direct abstraction mechanism.1,2
In this mechanism, atomic states evolving from electron
configurations that correlate to the product metal oxide states
react faster than those atomic states that do not correlate with
the product states. Reaction barriers are expected to arise along
the surfaces for the states that do not correlate. In the Mo case,
atomic states evolving from the 4d55s1 configuration orbitally
correlate with 2δ212σ16π1 MoO(X5Πr) and several low-energy
molecular states.3 The a7S3 (ground state), a5S2, and a5GJ states
have 4d55s1 configurations, and the a5DJ term has a 4d45s2
configuration. The electron configurations and energies of the
Mo states studied here are summarized in Table 1.4,5 The a5GJ
term, which is not shown in Table 1, has as its lowest energy
electronic
config
4d55s1
energy
wavelength
(nm)
detection filter
(nm)b
state
a7S3
(cm-1
0
)
386.41
390.30
550.65
379.73
384.72
579.19
388.68
603.07
400
390c
600
600
600
550
600
550
a5S2
a5D0
a5D1
a5D2
a5D3
a5D4
4d55s1
4d45s2
4d45s2
4d45s2
4d45s2
4d45s2
10 768.33
10 965.97
11 142.83
11 454.42
11 858.54
12 346.31
a References 4 and 5. b Broad-band filter with fwhm ) 50 nm.
c Narrow-band filter with fwhm ) 10 nm.
range version of the harpoon mechanism used to explain the
large reaction cross sections of metal atoms with oxidants.2,6
The potential energy surfaces deriving from the reactant states
cross an ion-pair surface in this mechanism. Depletion ef-
ficiencies thus depend on the electron affinity of the oxidant,
OX, and on the ionization potential, IP, of the TM atom. The
electron-transfer mechanism has been applied to reactions
involving main group metal atoms as well as TM atoms. A
semiempirical, resonance interaction model has also been
developed to explain reactions of metal atoms with N2O.7 This
model predicts activation energies and correlates rate constants
and activation energies to the sum of IPs and s-p promotion
energies, PE. We compare the depletion kinetics of Mo states
with respect to electron configurations and test the models above
with our data.
The reactions of Mo(a7S3) with O2, CO2, N2O, and NO have
been observed in a shock tube apparatus at temperatures greater
than ≈1200 K,8 and a room temperature flow tube study of the
reactivity of Mo(a7S3) with O2, CO2, and N2O has recently been
reported.9 We are unaware of any other studies on the reactions
reported here.
16 641.08 cm-1
.
We extend the Mo + O2 study to an investigation of the
reactivity of Mo(a7S3,a5S2,a5DJ) with SO2, CO2, N2O, and NO.
SO2, CO2, and N2O are closed-shell molecules while NO is a
stable radical. There is a reasonable range of electron affinities
among these reactants and O2 such that the reactivity of different
electronic states of Mo toward different types of oxygen-
containing molecules can be assessed and the driving force for
reaction can be ascertained. The electron affinity, EA, is an
important quantity in the electron-transfer mechanism, a short-
X Abstract published in AdVance ACS Abstracts, April 1, 1996.
S0022-3654(95)03217-5
This article not subject to U.S. Copyright. Published 1996 by the American Chemical Society

Depletion Kinetics of Mo(a7S3,a5S2,a5DJ)
J. Phys. Chem., Vol. 100, No. 18, 1996 7503
Experimental Section
Experiments were carried out in a slowly flowing apparatus
using a laser photolysis/laser-induced fluorescence (LIF) tech-
nique. Reactions of Mo + oxidant were studied under pseudo-
first-order conditions where the pressure of the oxidant greatly
exceeded the pressure of the Mo precursor (and hence, Mo).
Details of the experiment have been described elsewhere1,2 and
are only summarized here. The reaction chamber was a stainless
steel cross that was contained in a convection oven. Molyb-
denum atoms were produced by the 248 nm photodissociation
of Mo(CO)6 or MoCl4 with the unfocused output of an excimer
laser operating at 15 Hz. Photolysis laser fluences inside the
chamber were ≈60 mJ/cm2, although fluences were varied over
a factor of ≈5 to study their effects on thermalization and
reaction rates. Molybdenum atoms were detected via LIF using
laser light from an excimer-pumped dye module. Neutral
density filters were used to ensure the kinetic results did not
depend on the dye laser fluence. The laser light from the
photolysis laser and dye module counterpropagated through the
chamber. Interference filters were used to isolate the LIF
signals. The Mo states4 and corresponding LIF transitions5 are
given in Table 1. A photomultiplier tube was used in collecting
the LIF which was subsequently sent to a gated boxcar sampling
module, and the digitized output was stored and analyzed by a
computer.
Figure 1. Production and decay of Mo(a5D0) following the 248 nm
photodissociation of Mo(CO)6 in Ar buffer. T ) 296 K, Ptot ) 20
Torr, [Mo(CO)6] ≈ 0.2 m Torr, and laser fluence ≈ 60 mJ/cm2. (a)
No added OX. (b) 32.2 mTorr of NO; solid line is an exponential fit
with τ ) 26.6 µs.
The molybdenum precursor was entrained in a flow of buffer
gas (Ar or N2). The diluted molybdenum precursor, buffer gas,
and oxidant flowed through mass flow meters and flow
controllers prior to admission to the reaction chamber. Total
flows were between 150 and 5100 sccm, depending on the total
pressure. Pressures were measured with MKS Baratron ma-
nometers, and chamber temperatures were measured with a
thermocouple.
were unable to obtain an adequate signal-to-noise ratio, S/N,
for the excited states of Mo at higher total pressures using N2
buffer. Additionally, the lifetimes of these states decreased with
increasing N2 pressures such that 10 Torr was the maximum
pressure that could be used to obtain accurate rate constants.
Following photodissociation of Mo(CO)6 in argon buffer, in
the absence of oxidant, significant growth was observed for all
states investigated. This growth is exhibited in Figure 1a for
Mo(a5D0). The growth is attributed to collisional quenching
of higher lying excited states and dissociation of photofragments.
We detected excited Mo states up to 16 800 cm-1 (a5GJ) (these
states decayed exponentially under all conditionsswith and
without added O2).1 Other workers have studied the 248 nm
photodissociation of Mo(CO)6 and have observed emission from
Mo states lying as high as 31 655 cm-1 for fluences >10 mJ
cm-2, indicating a multiphoton (>3 photons) process.10 Their
results and energetic considerations indicate that ground state
Mo is formed via a two-photon process, and the excited states
are produced by a three-photon process. Thus, the production
of Mo states up to ≈36 000 cm-1 is energetically possible.
Addition of oxidant (NO, in this case) eliminates the growth as
shown in Figure 1b; the solid line through the data is an
exponential fit to the equation
Reaction time was taken as the delay time between the
photolysis pulse and the dye-laser pulse. The delay time was
varied by a digital delay generator controlled by a computer.
Minimum delay times were typically 1-3 µs in order to avoid
any prompt emission from photofragments. The trigger source
for these experiments was scattered pump laser light incident
upon a fast photodiode. LIF decay traces consisted of 200-
500 points, each point averaged over 2-10 laser shots. LIF
intensities were proportional to Mo number densities.
Reagents. The following reagents were used as received:
Mo(CO)6 (Johnson Matthey, 99%), MoCl4 (Aldrich, 90%), Ar
(Potomac Airgas, Inc., 99.998%), N2 (Potomac Airgas, Inc.,
99.998%), SO2 (MG Industries, 99.98%), CO2 (MG Industries,
anaerobic grade, 99.9%), N2O (Union Carbide, 99.0%), NO
(Liquid Carbonic, 99.0%). MoCl4 was used exclusively at
temperatures above ≈540 K due to decomposition of Mo(CO)6
at temperatures above 540 K. Although we used MoCl4 in a
few experiments below 540 K, Mo(CO)6 was the precursor of
choice because it has a higher vapor pressure than MoCl4.
Below 540 K, measured rate constants were independent of the
Mo source.
I ) I0 exp(-t/τ)
(1)
where I and I0 are the LIF signals at time t and time 0,
respectively, and τ is the lifetime from which the pseudo-first-
order rate constant, 1/τ, is obtained. The fit is from a delay of
1.0 µs and extends for ≈4 lifetimes. The reproducibility of
our measured τ’s is ( 10% for single-exponential decays and
(25% for biexponential decays (discussed below). Conditions
were always chosen so that growth processes were unimportant
in obtaining rate constants, i.e., we ensured τ did not depend
on photolysis and dye-laser fluence, precursor partial pressure,
and total flow rate. We previously reported a kinetic study on
Mo + O2 using this identical experimental method.1 Our
measured rate constant for Mo(a7S3) + O2 agrees with that
obtained using a flow tube/laser vaporization technique,9
indicating growth processes were not important. The growth
Data Analysis and Results
Experiments were typically performed at 20 Torr total
pressure using argon bath gas at room temperature. Experiments
were performed in the temperature range 295-621 K. Precursor
partial pressures were ≈0.1-0.5 mTorr, and oxidant partial
pressures were usually less than a fraction of a percent of the
total pressure. Because energy-transfer processes and observ-
able consecutive-type depletion kinetics were found to be
important in these Mo/OX systems, rate constants were also
measured in N2 bath gas at a total pressure of 10 Torr. We

7504 J. Phys. Chem., Vol. 100, No. 18, 1996
McClean et al.
diminished with increasing total pressures. The use of MoCl4
without added oxidant also resulted in growth at short times;
however, the growth was not as pronounced. Growth at short
times was virtually nonexistent when using N2 buffer. The
absence of growth is attributed to the collisional relaxation of
photofragments that ultimately dissociate to produce Mo atoms
and to the quenching of higher energy states of Mo. 1/τ is given
by the expression
1/τ ) 1/τ0 + k[OX]
(2)
where k is the second-order rate constant, [OX] is the number
density of OX, and τ0 is the lifetime of Mo without added OX.
τ0 represents the lifetime of Mo in the presence of other species
in the reaction chamber and diffusion out of the detection zone;
τ0 is usually large compared to the lifetimes with added OX.
Second-order rate constants are determined from plots of 1/τ
versus oxidant partial pressure such as that shown in Figure 2.
The slope yields the observed bimolecular rate constant, k. Such
plots typically extend over a 10-fold partial pressure (of OX)
range. Experiments described here were repeated at different
total pressures and temperatures, depending on the system.
Since the decay kinetics of Mo with OX were also investi-
gated in N2, we first present quenching (physical) results due
to N2 as these results are instrumental in interpreting the decay
kinetics of Mo + OX in N2 buffer.
Figure 2. Typical plot for determining second-order rate constants.
Conditions are the same as for Figure 1. The slope yields k ) (2.9 (
0.8) × 10-11 cm3 s-1
.
Quenching by N2. Single-exponential temporal behavior was
not observed for Mo(a5S2) and Mo(a5D0) in the presence of
added N2, as observed in Figure 3a for Mo(a5D0). Thus, a
similar decay scheme is indicated for both states. Under these
conditions, a relatively fast decay rate of Mo(a5D0) is observed
at times less than ≈50 µs, and a slower decay rate is observed
at longer times. The decay rates of both portions of the decay
trace depend on added N2. Thus, we attribute this behavior to
the collisional deactivation of a higher energy state to Mo(a5D0).
The solid line through the experimental points in curve a is a
biexponential fit. The fit assumes a kinetic model of the form
Figure 3. Collisional disappearance of Mo(a5D0) by N2. T ) 295 K,
Ptot ) 20 Torr (Ar buffer), precursor ) Mo(CO)6. (a) Biexponential
decay plot for 163 mTorr of added N2; solid line is a biexponential fit
with τ1 ) 9.27 µs, τ2 ) 249 µs. (b) Determination of the second-
order rate constants. See text for results of fit.
Mo(a5D0) + N2 f Mo′ + N2 kfast
Mo′′ + N2 f Mo(a5D0) + N2 kslow
(3)
(4)
TABLE 2: Measured Second-Order Rate Constants for the
Collisional Disappearance of Mo(a5S2,a5DJ) by N2 in Ar
Buffera
where Mo′ and Mo′′ are states lower and higher in energy,
respectively, than the a5D0 state. The rate equations for reactions
3 and 4 can be solved to yield the biexponential expression
k’s (10-12 cm3 s-1
)
b
Mo state
a5S2
T (K)
kfast
kslow
ktrue
[a5D0] ) A1 exp(-t/τ1) + A2 exp(-t/τ2)
A1 ) [a5D0]0 - [Mo′′]0kslow/(kfast - kslow
A2 ) [Mo′′]0kslow/(kfast - kslow
(5)
296
455
296
296
455
23 ( 7
c
16 ( 8
0.18 ( 0.09
0.46 ( 0.23
0.21 ( 0.11
23 ( 7
c
16 ( 8
)
(5a)
(5b)
a5D0
a5D1-4
0.13 ( 0.04
0.40 ( 0.12
)
b
a Total pressure ) 20-50 Torr. ktrue ) rate constant for the
depletion of the state in question. c Did not attempt to measure.
where [ ]0 indicates initial number densities. τ1 and τ2 are
determined by a fit of the decay profile using a nonlinear fitting
routine. kfast and kslow are determined from plots of 1/τ1 vs PN
mentioning that the long time temporal behavior observed in
Figure 3a is independently exponential over 3-4 lifetimes (a
log I versus t plot yields the same τ2, to within experimental
error). The J ) 1-4 spin-orbit states of the a5DJ term decayed
exponentially with added N2, although some of them exhibited
a minor growth for 1-3 µs following the photolysis pulse.
Rate constants for the Mo/N2 system are summarized in Table
2. The listed ktrue values are the rate constants for the depletion
of the states in question. Note the relatively large value of
k(a5S2). In 10 Torr of neat N2 buffer, Mo(a5S2) has virtually
decayed away in less than 1 µs; therefore, the temporal behavior
of Mo(a5S2) in neat N2 buffer represents the decay rates of higher
2
and 1/τ2 vs PN , respectively, as shown in Figure 3b. For N2
2
pressures greater than ≈0.8 Torr, τ1 was too short to obtain
meaningful kfast values. In fact, the rapid decay observed in
Figure 3a is virtually nonexistent at N2 pressures above ≈1 Torr,
and only the long time behavior is observed. The model
represented by reactions 3 and 4 is oversimplified. Although
written as a two-step mechanism, it could involve three or more
steps with kslow being a weighted average of the respective rate
constants. Biexponential behavior is still predicted for Mo′′ +
N2 having additional exit channels.11 We estimate the overall
uncertainty1 in kfast and kslow to within (50%. It is worth

Depletion Kinetics of Mo(a7S3,a5S2,a5DJ)
J. Phys. Chem., Vol. 100, No. 18, 1996 7505
TABLE 3: Measured Second-Order Rate Constants for the
Collisional Disappearance of Mo(a7S3,a5S2,a5DJ) by SO2 at
297 Ka
k’s (10-10 cm3 s-1
)
Mo state
kfast
kslow
k(in N2)
ktrue
a7S3
1.3 ( 0.3
3.8 ( 1.9
d
1.3 ( 0.3b
3.8 ( 1.9
d
a5S2
c
1.7 ( 0.5
1.8 ( 0.6
a5D0
a5D1-4
2.2 ( 0.8
1.8 ( 0.9
1.8 ( 0.6
a Ptot ) 10-40 Torr in Ar and 10 Torr in N2. kfast and kslow are
Figure 4. Mo(a7S3) + SO2. T ) 295 K, Ptot ) 40 Torr (Ar buffer),
precursor ) MoCl4, 15.7 mTorr SO2; solid line is an exponential fit
with τ ) 10.4 µs.
measured biexponential rate constants in Ar buffer. States without a
k
b
slow decayed exponentially in Ar buffer. Ptot(in Ar) ) 10-200 Torr.
c A small tail whose temporal behavior appeared to depend on the SO2
partial pressure. We were unable to obtain consistent lifetimes due to
low S/N. d Unable to measure and therefore could not determine ktrue
.
See text.
+ OX interaction in question.) Identical rate constants using
both Ar and N2 buffers are observed for J ) 1-4 of the a5DJ
term. On the basis of our measured rate constants of the other
states of Mo interacting with SO2, we estimate k(a5D0) ≈ 3 ×
10-10 cm3 s-1
.
Mo + CO2. Ground state Mo is unreactive toward CO2. We
report an upper limit of k ) 10-15 cm3 s-1. We attempted to
measure rate constants for this reaction over the temperature
range 296-621 K and over the total pressure range 100-600
Torr using argon buffer. The room temperature rate constant
is probably much less than 10-15 cm3 s-1 since the activation
energy for this reaction over our temperature range is probably
large. Akhmadov et al. measured an activation energy of 19.5
kcal mol-1 using a shock wave technique.8 With such a large
activation energy, we would not be able to measure a depletion
rate, even at 621 K, the maximum attainable temperature of
our oven. Lian et al.9 reported the value k ) 5.3 × 10-14
((30%), which is 50 times larger than our upper limit. Since
k for the reaction of Mo(a7S3) with O2 is near the gas kinetic
rate,1,9 an impurity level of 500 ppm O2 would result in a rate
constant near the value reported by Lian et al. When we used
anaerobic grade CO2 (<10 ppm O2), we did not observe any
decay due to CO2. However, we did observe some decay using
standard grade CO2. It is possible that their larger value is due
to impurities. Nevertheless, both these studies show that
Mo(a7S3) is unreactive toward CO2.
Figure 5. Mo(a5D0) + SO2. T ) 296 K, Ptot ) 40 Torr (Ar buffer),
precursor ) Mo(CO)6, 4.64 mTorr SO2; solid line is a biexponential
fit with τ1 ) 2.89 µs, τ2 ) 28.7 µs.
energy state(s). The rate constants for the a5D1-4 spin-orbit
states are identical, to within experimental uncertainty, and have
a value of (1.3 ( 0.4) × 10-13 cm3 s-1 at room temperature.
The large rate constant at 455 K gives an activation energy of
≈1.9 kcal/mol and an A factor of ≈3.2 × 10-12 cm3 s-1
.
Mo + SO2. Figure 4 shows a decay profile for the reaction
of Mo(a7S3) with SO2 using the precursor MoCl4. Note the
growth at t ≈ 1-7 µs. The exponential fit is from 8 to 60 µs.
This growth was observed under all conditions, using both
MoCl4 and Mo(CO)6. We attribute the growth to thermalization
of photofragments and the decay of excited states of molybde-
num, Mo*, to ground state Mo via collisions with SO2. (The
growth exhibited at short times in Figure 4 was also observed
in some of the other Mo + OX systems studied here and will
be noted where applicable.)
The decays of Mo(a5S2) in the presence of SO2 was, for the
most part, exponential, with a small tail presenting itself after
≈90% decay in Ar buffer. The small tail might be due to the
collisional deactivation of a higher energy state to Mo(a5S2).
k(a5S2) is (3.8 ( 1.9) × 10-10 cm3 s-1 in Ar, but in N2 it is (1.7
( 0.5) × 10-10 cm3 s-1. As noted above in the N2 section, the
depletion kinetics of Mo(a5S2) in N2 buffer is due to another
The decay kinetics of the excited states of Mo in the presence
of CO2 are quite complex. In Ar buffer, biexponential decay
kinetics were observed for Mo(a5S2) and Mo(a5D0). Single-
exponential decay kinetics in Ar were observed only for
Mo(a5D4). In Ar, the a5D1 and a5D3 states were, for the most
part, exponential, with a small tail that became evident after
≈80% decay. The temporal dependence of the tails appeared
to depend on the CO2 partial pressure, but we were unable to
obtain consistent lifetimes due to low signal-to-noise ratios.
Mo(a5D2) had a rapid decay at relatively short times followed
by a longer, exponential decay; we were unable to obtain an
accurate measure of kfast for this reaction. The depletion kinetics
were also measured in N2 buffer. We were able to obtain rate
constants for only Mo(a5S2) and two states in the a5D term.
The depletion rate constants are listed in Table 4. There appears
to be no consistency between the rate constants measured in
Ar and N2 buffer. The kinetics in the Mo/CO2 system are
obviously complex and involve energy-transfer processes and
possibly mixing between the states.
Mo state. Therefore, the actual rate constant, ktrue, is kfast
.
We are unable to report a “true” depletion rate constant for
Mo(a5D0) + SO2 because the decay rates in Ar buffer were too
rapid at relatively short times, and the intensity of the slower
decay portion of the decay trace accounted for as much as 70-
90% of the total LIF intensity, depending on conditions. This
short-time nonexponential behavior is shown in Figure 5. The
long-time temporal behavior is exponential. We were unable
to measure depletion rates of Mo(a5D0) + SO2 in N2 buffer
because of a low S/N. Mo(a5D1-4) decay profiles exhibited
exponential behavior and thus were fitted using eq 1 in the form
log I Vs time. The measured second-order rate constants for
the reactions of Mo with SO2 are all on the order of the gas
kinetic rate constant (≈2 × 10-10 cm3 s-1) as seen in Table 3.
(In this table and all subsequent tables, kfast and kslow represent
measured biexponential rate constants in Ar buffer for the Mo
Mo + N2O. Ground state Mo was found to be unreactive
toward N2O at room temperature. Unlike the Mo(a7S3)/CO2
system, observable reaction rates were measured for Mo(a7S3)
+ N2O as the temperature was increased. The measured rate

7506 J. Phys. Chem., Vol. 100, No. 18, 1996
McClean et al.
TABLE 4: Measured Second-Order Rate Constants for the
Collisional Disappearance of Mo(a7S3,a5S2,a5DJ) by CO2 at
297 Ka
k’s (10-12 cm3 s-1
)
Mo state
kfast
kslow
k(in N2)
ktrue
a7S3
a5S2
a5D0
a5D1
a5D2
a5D3
a5D4
<0.001
38 ( 19
15
4.2 ( 1.3
14 ( 4
38 ( 19
3.0 ( 0.9
15
c
c
c
c
5.0 ( 1.5
b
18 ( 5
>5
3.0 ( 0.9
b
18 ( 5
4.0 ( 2.0
2.9 ( 0.8
d
d
Figure 6. Arrhenius plot for Mo(a7S3) + N2O. Ptot ) 100-600 Torr
(Ar buffer). Solid line is a weighted fit to the expression k(T) ) A
exp(-Ea/RT). Error bars represent (30% uncertainties. See text for
results of fit.
a Ptot(a7S3) ) 100-600 Torr in Ar. Ptot(Mo*) ) 10-100 Torr in
Ar and 10 Torr in N2. b A small tail whose temporal behavior appeared
to depend on the CO2 partial pressure. We were unable to obtain
consistent lifetimes due to low S/N. c Unable to ascertain actual value
of k. See text for discussion. d Too complex to adequately fit.
TABLE 5: Measured Second-Order Rate Constants for the
Collisional Disappearance of Mo(a7S3,a5S2,a5DJ) by N2Oa
b
Mo state T (K)
ktrue
kslow
k(in N2)
a7S3
297 ,0.1 × 10-14
400
426
448
480
523
561
585
606
621
296
296
297
398
473
621
297
398
472
621
296
397
470
621
300
398
473
621
0.10 × 10-14
0.21 × 10-14
0.34 × 10-14
0.65 × 10-14
1.4 × 10-14
3.0 × 10-14
4.0 × 10-14
5.9 × 10-14
6.8 × 10-14
Figure 7. Arrhenius plots at Ptot ) 20 Torr. (a) Mo(a5D4) + NO; (b)
Mo(a5D3,4) + N2O; (c) Mo(a5D1,2) + N2O. Error bars represent (30%
uncertainties, except for Mo(a5D4) + NO where they are (40%. Solid
lines are weighted fits to the expression k(T) ) A exp(-Ea/RT). See
text for results of fits.
a5S2
a5D0
a5D1
25 × 10-11 (fast) 1.3 × 10-11
1 × 10-11 (fast) 0.60 × 10-11
0.70 × 10-11
0.85 × 10-11
1.1 × 10-11
0.77 × 10-11
activation energy of 8.63 ( 6.33 kcal/mol; this value was
measured in a shock tube apparatus and is in agreement with
our value.
0.96 × 10-11
1.3 × 10-11
a5D2
a5D3
a5D4
0.84 × 10-11
0.87 × 10-11
1.3 × 10-11
1.4 × 10-11
The decay kinetics of Mo(a5S2) and Mo(a5D0) with N2O were
biexponential in Ar buffer while the decay kinetics of the a5D1-4
spin-orbit states were exponential with a little growth at short
times (similar to the decay plots involving SO2 in Figure 4).
All states decayed exponentially in N2 buffer. The rate constants
measured in both Ar and N2 buffer are consistent, as seen in
Table 5. k(a5D1) in Ar and N2 appear to be different; however,
they are within their combined uncertainties. Mo(a5D1) and
Mo(a5D2) appear to decay at identical rates, to within experi-
mental uncertainty, and so do Mo(a5D3) and Mo(a5D4). Ar-
rhenius plots for Mo(a5D1,2) and Mo(a5D3,4) are shown in Figure
7. The rate constants can be expressed as
0.95 × 10-11
1.0 × 10-11
1.3 × 10-11
1.3 × 10-11
1.5 × 10-11
1.7 × 10-11
2.2 × 10-11
0.92 × 10-11
1.3 × 10-11
1.7 × 10-11
2.5 × 10-11
a Units are cm3 s-1. Uncertainties are (30% except for a5S2(in Ar)
and a5D0, which are (50%. Ptot(a7S3) ) 100-600 Torr in Ar.
b
Ptot(Mo*) ) 10-100 Torr in Ar and 10 Torr in N2. ktrue ) rate constant
for the depletion of the state in question in Ar buffer. States with a
k(a5D1,2) ) (2.0 ( 0.4) × 10-11
×
kslow decayed biexponentially in Ar buffer with kfast ) ktrue
.
exp[-(0.6 ( 0.2 kcal mol-1)/RT) cm3 s-1 (7)
constants are listed in Table 5 and are plotted in Arrhenius form
in Figure 6. Relatively high total pressures were required for
the reaction Mo(a7S3) + N2O because relatively high N2O
pressures were needed for any observable decays of ground state
Mo atoms. The Arrhenius equation obtained from the weighted
fit in Figure 6 is
k(a5D3,4) ) (4.7 ( 0.8) × 10-11
×
exp[-(0.9 ( 0.2 kcal mol-1)/RT) cm3 s-1 (8)
Mo + NO. Variations with total pressure of the second-
order rate constants for the reaction of Mo(a7S3) with NO are
shown in Figure 8, and the rate constants are listed in Table 6.
Note that the rate constants increase with increasing total
pressure and decreasing temperature. These observations are
consistent with a termolecular mechanism. The solid lines
through the data in Figure 8 are fits to the simplified falloff
expression of Troe12
k(a7S3) ) (2.0 ( 0.5) × 10-10
×
exp[-(9.8 ( 0.3 kcal mol-1)/RT] cm3 s-1 (6)
where the uncertainties represent one standard deviation (preci-
sion). We did not observe a significant decay rate at room
temperature. Lian et al.9 also reports no reaction at room
temperature. Extrapolation of our results to 298 K gives k(298
K) ) 1.3 × 10-17 cm3 s-1. Akhmadov et al.8 reported an
k0[M]
log Fc
log k ) log
+
(9)
1 + [log(k0[M]/k∞)]2
(
)
1 + (k0[M]/k∞)

Depletion Kinetics of Mo(a7S3,a5S2,a5DJ)
J. Phys. Chem., Vol. 100, No. 18, 1996 7507
TABLE 8: Measured Second-Order Rate Constants for the
Collisional Disappearance of Mo(a5S2,a5DJ) by NOa
k’s (10-11 cm3 s-1
)
Mo state
a5S2
T (K)
ktrue
kslow
k(in N2)
296
373
450
296
296
374
450
295
373
413
446
621
7.2 (fast)
b
3.3
4.2
5.3
3.3
b
b
b
a5D0
3.0
3.1
4.4
4.9
3.8
4.2
4.7
4.8
7.6
a5D1-3
a5D2
3.9
b
b
3.8
b
b
a5D4
Figure 8. Pressure dependence of the reaction Mo(a7S3) + NO at 296
and 450 K. Error bars represent (30% uncertainties. Solid lines are
fits to Troe’s falloff expression (see text).
b
b
TABLE 6: Measured Second-Order Rate Constants for the
Collisional Disappearance of Mo(a7S3) by NO in Ar Buffera
a Ptot ) 10-100 Torr in Ar and 10 Torr in N2. Uncertainties are
(40% except for a5S2 (fast and slow), which are (50%. States with
a kslow decayed biexponentially in Ar buffer with kfast ) ktrue.
b Did not
T
(K)
Ptot
k
T
(K)
Ptot
k
(Torr) (10-11 cm3 s-1
)
(Torr) (10-11 cm3 s-1
)
attempt to measure.
296
10
20
35
50
75
100
125
150
175
195
200
210
250
400
500
600
20
0.59
1.0
1.5
1.9
2.2
2.7
2.7
2.8
2.8
3.1
3.3
3.2
3.6
4.2
4.3
4.4
450
10
20
35
50
75
100
150
200
250
300
350
400
0.34
0.60
0.71
1.0
1.4
1.6
2.2
2.2
2.6
3.5
The decay kinetics of Mo(a5S2) in the presence of added NO
were biexponential in Ar buffer and exponential in N2 buffer.
The decay kinetics of the a5DJ states with NO were all
exponential in Ar and N2. We see in Table 8 that results in Ar
and N2 buffer are consistent. The temperature dependence of
k(a5D4) is given in Figure 7, and the Arrhenius expression
obtained is
2.5
3.5
k(a5D4) ) (1.3 ( 0.4) × 10-10
×
exp[-(0.8 ( 0.3 kcal mol-1)/RT) cm3 s-1 (10)
The other a5DJ states are expected to have similar Arrhenius
337
373
373
376
413
447
516
621
0.86
0.72
3.0
0.77
0.71
2.1
parameters, as seen from k(T) for Mo(a5D2) in Table 8.
20
200
20
20
200
20
Discussion
The Mo + N2 and Mo + OX systems are first discussed
separately. We then offer general comments about TM deple-
tion kinetics. Because we investigated several atomic states
interacting with the oxidants and a proposed model may apply
to only one system, we find this order to be a favorable approach
in our discussion.
0.41
0.24
20
a Uncertainties are (30%.
TABLE 7: Falloff Modeling Parameters for the Reactions
of Ti(a3F2), Cr(a7S3), Fe(5D4), and Mo(a7S3) with NO in Ar
Buffer
Mo + N2. Physical quenching is the only exit channel
available for excited states of molybdenum interacting with N2.
TM T (K)
Tia
k0 (cm3 s-1
)
k∞ (cm3 s-1
)
Fc
The vibrational frequency of N2, ωe(N2) ) 2360 cm-1 16
, is not
300 (5.8 ( 2.6) × 10-31
e
e
in resonance with any energy differences, ∆E, between the Mo
states,4 thus making direct E-V,R energy transfer improbable.
This implies a collision complex is formed that facilitates the
quenching. We omit a detailed discussion on the fate of the
quenched Mo species, but several observations stand out that
are worth mentioning. The rate constant of Mo(a5S2) is
significantly larger than the rate constant of the J ) 1-4 a5DJ
spin-orbit states, but it is only slightly larger than the rate
constant of Mo(a5D0). The a5S2 and a5D0 states are 197.64 cm-1
apart, and mixing between these states might be induced by
collisions with N2. Identical rate constants for the J ) 1-4
a5DJ spin-orbit states indicate either identical rate constants
or rapid interconversion among said states under our conditions.
Note that k(a5D1-4) ≈ kslow(a5D0) ≈ kslow(a5S2) and that the
depletion rate of Mo(a5S2) + N2 in Ar buffer is relatively fast
(see Table 2) such that, in neat N2 buffer, the observed depletion
rate of Mo(a5S2) in the presence of OX actually represents decay
of another state. Also note that in our Mo/OX systems using
N2 buffer k(a5S2) ≈ k(a5DJ). These observations indicate that
either one or more of the Mo(a5DJ) spin-orbit states and/or
Crb 298 (1.2 ( 0.2) × 10-30 (3.2 ( 0.8) × 10-11 0.7 (fixed)
Fec 296 2.3 × 10-32
≈1 × 10-12
≈0.7
Mod 296 (2.6 ( 0.3) × 10-29 (5.8 ( 0.5) × 10-11 0.72 ( 0.07
Mod 450 (1.9 ( 0.3) × 10-29 (7.5 ( 1.2) × 10-11 0.6 (fixed)
(1.7 ( 0.2) × 10-29 (6.0 ( 0.9) × 10-11 0.7 (fixed)
(1.7 ( 0.2) × 10-29 (5.1 ( 0.7) × 10-11 0.8 (fixed)
a Reference 14. b Reference 13. c Reference 15; error estimates not
given. d This work. e Not measured.
where k0 is the limiting low-pressure third-order rate constant,
k∞ is the limiting high-pressure second-order rate constant, [M]
is the buffer gas number density, and Fc is the broadening factor.
We obtained large uncertainties in k∞ and Fc (>100% for Fc)
when we fitted eq 9 to the 450 K data. (Not many points are
in the falloff region.) Since Fc ) 0.72 for the 296 K data, as
seen in Table 6, we fixed Fc at 0.6, 0.7, and 0.8 to obtain k0
and k∞ at 450 K. Values of the parameters in eq 9 are listed in
Table 7 and are compared to parameters from other TM/NO
complexation systems.13-15 Fc was fixed at 0.6, 0.7, and 0.8
for Cr + NO because only a small portion of the falloff range
was observed;13 we list only the median value.

7508 J. Phys. Chem., Vol. 100, No. 18, 1996
McClean et al.
another Mo state are connected to Mo(a5D0) and/or Mo(a5S2)
via collisions with N2.
Mo + SO2. The most likely reaction for ground state Mo is
Our results indicate that significant energy-transfer processes
and mixing are important for the Mo* + CO2 interactions. The
rate constant at 297 K for Mo(4d55s1 a5S2) + CO2 is relatively
large and is expected to be larger than the depletion rate constant
of the 4d45s2 a5DJ term on account of the closed s shell in the
a5DJ term. The depletion rates of the excited states are not
(solely, if at all) driven by an electron-transfer mechanism since
IP(a5S2) > IP(a5DJ) and k(a5S2) . k(a5DJ). As seen from the
measured rate constants in Table 4, the kinetics in this system
are too complex to draw any definitive conclusions.
Mo(a7S3) + SO2(X1A1) f MoO(X5Πr) + SO(X3Σ-)
∆H298 ) -11.0 kcal/mol17 (11)
The large rate constant at 297 K indicates a negligible barrier
to reaction. For the Mo* + SO2 interactions, we cannot
unequivocally state that physical quenching is not an important
exit channel; two or more exit channels may be operative to
varying degrees of importance. It is difficult to rationalize the
rate constants for the Mo + SO2 system in terms of the direct
abstraction mechanism since all of the rate constants are large;
only the ground state and the a5S2 state orbitally correlate with
the low-energy MoO states that are energetically accessible.3
Mo(a5S2) + SO2 has the largest rate constant of the Mo states
investigated in the Mo/SO2 system. Direct physical quenching
at such a rapid rate via E-V,R transfer to the ground state is
improbable since the ∆E of the Mo states are not in resonance
with any vibrational frequencies of SO2.18 It is conceivable
that its relatively large rate constant is due to a long-range
attraction that leads to an entrance channel controlled reaction;
i.e., efficient quenching might take place via a charge-transfer
intermediate complex. It is also conceivable that the interaction
of Mo(a5DJ) with SO2 might take place via a charge-transfer
complex. Small differences on the potential energy surfaces
of Mo(4d55s1 a5S2) and Mo(4d45s2 a5DJ) might account for the
smaller rate constant of the a5DJ term; the closed s shell is
expected to introduce more electronic repulsive effects. The
nonexponential decay plots of Mo(a5D0) might be due to
physical quenching from higher energy states.
Mo + N2O. Results for the reaction Mo(a7S3) + N2O are
consistent with a bimolecular reaction. Thus, the most likely
reaction channel is
+
Mo(a7S3) + N2O(X1Σ+) f MoO(X5Πr) + N2(1Σg )
∆H298 ) -103 kcal/mol17 (13)
The relatively large activation energy of 9.8 kcal/mol may be
due to the reaction being spin forbidden, as explained above
for CO2. The excited electronic states of MoO that are
energetically feasible for this reaction are also quintets,3 thus
making their production spin forbidden. The simple direct
abstraction mechanism, which only considers the conservation
of orbital occupancy, would predict a small or zero activation
energy.
The semiempirical, resonance interaction model uses the
modified Arrhenius equation k(T) ) AT1/2 exp(-E/RT) in
determining the Arrhenius parameters.7 Using this equation,
we obtain the result
k(T) ) (5.2 ( 1.1) × 10-12T1/2
×
The depletion rates of Mo + SO2 can be compared to the
rates of other SO2 reactions. The reaction kinetics of ground
state Ti with SO2 is the only other TM/SO2 system reported.14
This reaction has an appreciable rate at room temperature: k )
exp[-(9.3 ( 0.2 kcal mol-1)/RT] cm3 s-1 (14)
for Mo(a7S3) + N2O. Our activation energy does not agree
with the model’s predicted value of 6.72 kcal mol-1. The model
assumes that one of the resonance structures has a metal bonding
orbital of s character. Although the activated complex structure
for the reaction is not known, chemical bonding in MoO
involves the d orbitals, and the low-energy states of MoO have
a single electron in the 12σ molecular orbital which is
nonbonding and primarily centered on Mo.3 The 12σ molecular
orbital is of Mo 5s character and has significant mixing with
the Mo 5p orbital. Assuming the conservation of orbital
occupancy along the reaction coordinate, the poor agreement
between our measured value and the predicted value may be in
the model’s neglect of the TM’s d electrons and of the electronic
states of the products formed.
(6.09 ( 0.95) × 10-11 cm3 s-1. The reaction of B with SO2
19
proceeds at approximately the collision rate, and the reaction
of Ba with SO220-22 has a relatively large reaction cross section.
The fast rates of these reactions were attributed to an electron-
transfer mechanism (harpooning in the case of Ba). A high-
temperature kinetic study of the endothermic reaction of Al with
SO2 has been reported,23 and reactions of Na24 and K25 with
SO2 have been observed to produce stable adducts with k∞ on
the order of the collision rate. Results of these investigations
were also explained in terms of an electron-transfer mechanism.
On the basis of the limited amount of data on metal atom +
SO2 reactions, it appears that SO2 is very reactive toward metals.
SO2 has the largest EA of N2, O2, and the oxidants in this
paper.26-31
The 298 K rate constants for the ground state reactions of
Cr13,32 and W,33 which also lie in group 6, with N2O have also
been measured: kCr ) (1.0 ( 0.1) × 10-14 and kW ) 2.9 ×
Mo + CO2. Ground state Mo is unreactive toward CO2
under the conditions of our experiments despite the reaction’s
large exothermicity of 15.5 kcal/mol.17 The presence of a dn-1
s1 electron configuration is clearly not a sufficient condition
for chemical reactivity. Thus, the direct abstraction mechanism,
as defined, cannot account for this observation. This lack of
reactivity might be due to the reaction
10-15 ((30%) cm3 s-1
. The resonance interaction model
predicts kW < kMo < kCr. However, with the extrapolated kMo
) 1.3 × 10-17 cm3 s-1, we see that the resonance model does
not predict the correct ordering of the rate constants (kMo < kW
< kCr) based on the sum of IP and s-p PE.7 The electron-
transfer mechanism has also been applied to reactions involving
TM atoms and OX. For a given OX, a TM atom having a
smaller IP than other TMs is expected to exhibit faster reaction
rates. The IPs of Cr, Mo, and W are 6.77, 7.10, and 7.98 eV,
respectively.4 Clearly, the electron-transfer mechanism does not
explain the rate constant ordering of group 6 TM + N2O
reactions. Thus, the production of spin-forbidden states seems
to be the major constraint to the Mo + N2O reaction.
+
Mo(a7S3) + CO2(X1Σg ) f MoO(X5Πr) + CO(X1Σ+)
∆H298 ) -15.5 kcal/mol17 (12)
being spin forbidden. Without knowledge of the potential
energy surface and another viable approach, we can only
speculate about the production of spin-forbidden states being
cause for the lack of reactivity.

Depletion Kinetics of Mo(a7S3,a5S2,a5DJ)
J. Phys. Chem., Vol. 100, No. 18, 1996 7509
Mo(a5S2) interacts with N2O at the collision rate. The
chemical reaction
times (similar to the decay plots involving SO2 in Figure 4) for
Mo(a7S3) + NO.
General Comments. The oxidants CO2 and N2O appear
unique among the oxidants studied here. Both are linear, closed-
shell molecules; their reactions with ground state Mo are spin-
forbidden but exothermic, and they have very small electron
affinities (negative in the case of CO2),27,29 yet N2O has an
observable reaction rate with ground state Mo while CO2 does
not. Reaction 13 has a significantly larger exothermicity than
reaction 12; however, previous studies on other TM/OX systems
show no correlation between reaction efficiency and enthalpy
of reaction.14,36,37 For the V + CO2, N2O36,37 and Ti + CO2,
N2O14,37 reactions, rate constants for the N2O reactions are only
slightly larger than those for the CO2 reactions. Although CO2
and N2O do not correlate with their ground state fragments,
their reactions with V and Ti are spin-allowed. This may
explain why the rate constants for the V and Ti reactions are
significantly larger than those for the Mo + CO2, N2O reactions.
SO2 is also a closed-shell molecule. However, it has a
+
Mo(a5S2) + N2O(X1Σ+) f MoO(X5Πr) + N2(1Σg ) (15)
in which Mo(a5S2) orbitally correlates with MoO(X5Πr), may
be operative. Direct E-V,R energy transfer is improbable since
no vibrational frequencies of N2O18 and ∆E(a5S2-a7S3) are in
resonance. The depletion of Mo(a5S2) might be a direct O atom
abstraction taking place on an attractive, excited state surface,
unlike the ground state surface where a barrier is encountered.
The depletion rate of Mo(a5S2) is at least 19 times the depletion
rates of the a5DJ states. The reaction of Mo(a5DJ) with N2O to
produce ground state MoO would be slower than reaction 15
(as observed) according to the proposed direct abstraction
mechanism2 since the a5DJ term has a 4d45s2 electron config-
uration. Although we cannot identify the exit channels for
Mo(a5DJ), our results indicate some energy transfer is occurring.
The small activation energies (≈1 kcal/mol or less) measured
for Mo(a5DJ) + N2O indicate a loose transition state and
negligible barriers along the surfaces connecting reactants and
products.
significantly larger EA31 than CO2 and N2O,29 which seems
27
to have a favorable effect on its reactivity with metal atoms.
We now turn our attention to the electron-transfer mechanism.
The electron affinities of colliders studied in this work fall in
the order EA(N2) < EA(CO2) < EA(NO) < EA(N2O) < EA-
(O2) < EA(SO2).26-31 The electron-transfer mechanism cor-
rectly predicts for the bimolecular reactions of Mo(a7S3) with
CO2, N2O, O2, and SO2 that k(CO2) < k(N2O) < k(O2) < k(SO2)
at 298 K, but it is inconsistent with the depletion rates of Mo-
(a5S2) with all of the colliders mentioned above. (Valence
interactions are expected to be more important for reactions
involving N2, CO2, and N2O on account of their small EAs
which cause relatively short curve crossings (within the hard-
sphere collision diameter); therefore, the concept of electron
transfer in N2, CO2, N2O + TM reactions, especially when the
TM has a high IP, may not be valid). The electron-transfer
mechanism is also inconsistent with the reaction rates of several
oxidants with a given TM atom (Sc, Ti, V).14,36,37 It accounts
for the ordering of the rate constants of the 3d TM atoms Sc,
Ti, and V with a given oxidant37 but fails when the 4d and 5d
TM atoms Mo1 and W2 are included. This mechanism is also
inconsistent with the depletion rates of Mo (including Mo*)
Mo + NO. Results of the reaction Mo(a7S3) + NO indicate
a termolecular reaction mechanism. Production of MoO from
this reaction is not thermodynamically feasible under our
experimental conditions at 298 K as it is endothermic by 8.2
kcal/mol at 298 K.17 As seen in Table 7, k0 is an order of
magnitude larger than k0 for Cr (+NO), which lies above Mo
in the periodic table (CrO formation is also endothermic at 298
K).17 The value of k∞ for Mo indicates a small barrier to adduct
formation. Accurate RRKM34 calculations could not be per-
formed because interaction potentials, well depths, vibrational
frequencies, etc., of MoNO are not known. However, the rather
large value of k0 indicates a higher Mo-NO binding energy
(>20 kcal/mol) than Cr-NO. For the Ti/NO system, TiO
formation is the dominant channel.14 However, adduct forma-
tion was indicated at room temperature. The reaction was in
its third-order kinetic regime up to 200 Torr, the maximum total
pressure at which the reactions were studied.
Interestingly, k0 for Cr and Mo, which have dn-1s1 configura-
tions, are larger than k0 for Ti and Fe, which have dn-2s2
configurations. In the s1 configuration, the singly occupied
orbital of the TM might overlap favorably with the single
electron of NO, forming a bond. This argument has been used
to explain the dioxygen complexes of the 3d TM atoms.35 The
3d work has shown that TMs having a dn-1s1 electron config-
uration react faster than those with a dn-2s2 configuration. Even
if the d electrons of the TM atom were involved in the bonding,
the reactivity of the s2 configuration would still be expected to
be less than that of the s1 configuration because of its diffuse
nature. It would be interesting to perform a similar study on
TM + NO systems to test the effect of the atomic electron
configuration.
1
with O2 and the oxidants studied here. Higher energy states
should deplete at a faster rate since they effectively have lower
IPs. We find that the 4d55s1 a5S2 state, which is the lowest
energy excited state of Mo, depletes faster than the 4d45s2 a5DJ
term in the presence of the oxidants, lending support for the
importance of electron configurations in the depletion kinetics
of TMs. In the case of Mo + O2, the 4d55s1 a7S3 ground state
also depletes faster than the a5DJ term, which has over 10 000
cm-1 more electronic energy. Thus, even though some charge
transfer might be operative to some extent, as discussed above
for the SO2 reactions, there are shortcomings to the electron-
transfer mechanism in explaining the reactivity of TM atoms.
Fast reactions such as alkali metal atoms + halogen molecules
can be satisfactorily explained by the harpoon mechanism; the
alkali metal atoms have relatively small IPs, and the halogen
molecules have relatively large EAs. Thus, such reactions are
entrance-controlled. The reactions of TM atoms with oxidants
(those studied here) are expected to experience valence interac-
tions since the IPs of the TM atoms are relatively large, and
the EAs of the oxidants are generally smaller than those of the
halogens.
Though an extensive investigation into the relative efficiencies
of different quenchers was not undertaken, removal rate
constants of Mo(a7S3) + NO using He, N2, and SF6 buffers
were measured at a total pressure of 20 Torr. The rate constants
kHe < kAr < k < k follow the trend of larger and/or more
N2
SF6
complex molecules being more efficient energy stabilizers.
Energetically, the excited states of Mo can react with NO to
produce ground state MoO. However, as with the other Mo +
OX systems, we cannot distinguish between chemical and
physical quenching for any of the excited states. There is some
energy transfer occurring by virtue of the biexponential decay
kinetics of Mo(a5S2). Another indication of physical quenching
is the slight nonexponential temporal behavior at relatively short
It has been shown here and elsewhere2,35 that the electronic
structure of TM atoms plays a role in their depletion kinetics.
States (ground and excited) evolving from a dn-1s1 electron
configuration deplete faster than those evolving from a dn-2s2

7510 J. Phys. Chem., Vol. 100, No. 18, 1996
McClean et al.
(8) (a) Akhmadov, U. S.; Zaslonko, I. S.; Smirnov, V. N. Kinet. Catal.
1988, 29, 251. (b) Akhmadov, U. S.; Zaslonko, I. S.; Smirnov, V. N. Kinet.
Catal. 1988, 29, 808.
(9) Lian, L.; Mitchell, S. A.; Rayner, D. M. J. Phys. Chem. 1994, 98,
11637.
(10) Tyndall, G. W.; Jackson, R. L. J. Phys. Chem. 1991, 95, 687 and
references therein.
(11) The rate equations can be solved to yield biexponential expressions
for Mo(a5D0) or any other Mo state that undergoes a similar reaction scheme;
see ref 1. Additionally, modeling computations were performed on the
decay schemes presented in this paper using a “Kinetics Integration Package
from ARSoftware”. The qualitative features of the nonexponential decay
plots were reproduced. The computations are, at best, semiquantitative,
because the initial number densities of the formed Mo atoms were not
determined.
electron configuration in many cases. The diffuse, closed s
subshell in the s2 configuration causes electronic repulsive
effects, thus reducing the reactivity. Additionally, the dn-1s1
electron configuration in some of the TM atoms orbitally
correlate with their low-energy TM monoxides and (probably)
causes the enhanced reactivity relative to the dn-2s2 states. Such
kinetic results have been observed for bimolecular as well as
association reactions. Our results generally follow along this
same pattern. However, other factors such as the electronic
structure of the oxidant and product states appear important, as
indicated above for the Mo(a7S3) + CO2 and N2O reactions.
(12) Troe, J. J. Phys. Chem. 1979, 83, 114.
(13) Parnis, J. M.; Mitchell, S. A.; Hackett, P. A. J. Phys. Chem. 1990,
Summary and Conclusions
94, 8152.
Our results indicate reaction and energy-transfer processes
for the collisional disappearance of Mo(a7S3), Mo(a5S2), and
Mo(a5DJ) by N2, SO2, CO2, N2O, and NO. Bimolecular reaction
kinetics are observed for the reactions of Mo(a7S3) with SO2
and N2O, termolecular reaction kinetics with NO, and no
reactions with CO2. The 4d55s1 a5S2 excited state depletes faster
than the higher energy 4d45s2 a5DJ term in all cases. Rate
efficiencies do not appear to depend entirely on the electron
configuration of Mo, nor entirely on the propensity for electron
transfer, as observed in other works.1,2,14 The resonance
interaction model is inconsistent in explaining the rate constants
(14) Campbell, M. L.; McClean, R. E. J. Phys. Chem. 1993, 97, 7942.
(15) Mitchell, S. A.; Hackett, P. A. J. Chem. Phys. 1990, 93, 7822.
(16) Herzberg, G. Molecular Spectra and Molecular Structure I, Spectra
of Diatomic Molecules, 2nd ed.; Van Nostrand: Princeton, 1950.
(17) Chase, M. W., Jr., Davies, C. A., Downey, J. R., Jr., Frurip, D. J.,
McDonald, R. A., Syverud, A. N., Eds. JANAF Thermochemical Tables,
3rd ed.; J. Phys. Chem. Ref. Data 1985, 14 (Suppl. 1).
(18) Herzberg, G. Molecular Spectra and Molecular Structure III,
Electronic Spectra and Electronic Structure of Polyatomic Molecules; Van
Nostrand Reinhold: New York, 1966.
(19) DiGiuseppe, T. G.; Davidovits, P. J. Chem. Phys. 1981, 74, 3287.
(20) Smith, G. P.; Zare, R. N. J. Am. Chem. Soc. 1975, 97, 1985.
(21) Behrens, R., Jr.; Freedman, A.; Herm, R. R.; Parr, T. P. J. Am.
Chem. Soc. 1976, 98, 294.
for the reactions of Mo (and Cr and W) with N2O.
A
(22) Freedman, A.; Parr, T. P.; Behrens, R., Jr.; Herm, R. R. J. Chem.
Phys. 1979, 70, 5251.
comparison of our results with other TM systems indicates that
atomic structure (in particular, the dn-1s1 configuration) is a
driving force in the depletion kinetics and that the production
of spin-forbidden states might be responsible for reduced
reactivities. State-to-state dynamic studies and theoretical
investigations on TM + OX systems would cast further light
on possible TM oxidation mechanisms.
(23) Fontijn, A.; Felder, W. J. Chem. Phys. 1979, 71, 4854.
(24) Shi, Y.; Marshall, P. J. Phys. Chem. 1991, 95, 1654.
(25) Goumri, A.; Laakso, D.; Rocha, J.-D. R.; Francis, E.; Marshall, P.
J. Phys. Chem. 1993, 97, 5295.
(26) EA(N2) ) -1.6 eV: Rosenstock, H. M.; Draxl, K.; Steiner, B.
W.; Herron, J. T. J. Phys. Chem. Ref. Data 1977, 6 (Suppl. 1).
(27) EA(CO2) ) -0.6 eV: Compton, R. N.; Reinhardt, P. W.; Cooper,
C. D. J. Chem. Phys. 1975, 63, 3821.
(28) EA(NO) ) 0.026 eV: Travers, M. J.; Cowless, D. C.; Ellison, G.
Acknowledgment. This research was supported by the Naval
Academy Research Council, the Office of Naval Research, and
a Cottrell College Science Award of Research Corporation.
B. Chem. Phys. Lett. 1989, 164, 449.
(29) EA(N2O) ) 0.22 eV: Hopper, D. G.; Wahl, A. C.; Wu, R. L. C.;
Tiernan, T. O. J. Chem. Phys. 1976, 65, 5474.
(30) EA(O2) ) 0.451 eV: Travers, M. J.; Cowless, D. C.; Ellison, G.
B. Chem. Phys. Lett. 1989, 164, 449.
(31) EA(SO)2 ) 1.107 eV: Nimlos, M. R.; Ellison, G. B. J. Phys. Chem.
1986, 90, 2574.
References and Notes
(1) Campbell, M. L.; McClean, R. E.; Harter, J. S. S. Chem. Phys.
Lett. 1995, 235, 497.
(32) Fontijn, A.; Blue, A. S.; Narayan, A. S.; Bajaj, P. N. Combust.
(2) Campbell, M. L.; McClean, R. E. J. Chem. Soc., Faraday Trans.
1995, 91, 3787 and references therein.
(3) Hamrick, Y. M.; Taylor, S.; Morse, M. D. J. Mol. Spectrosc. 1991,
146, 274.
(4) Moore, C. E. Atomic Energy LeVel as DeriVed from the Analysis
of Optical Spectra; Natl. Stand. Ref. Data Ser. (U.S. Natl. Bur. Stand.) 1971,
NSRDS-NBS 35.
(5) Whaling, W.; Hanaford, P.; Lowe, R. M.; Biemont, E.; Grevesse,
N. J. Quant. Spectrosc. Radiat. Transfer 1984, 32, 69.
(6) Herschbach, D. R. AdV. Chem. Phys. 1966, 10, 319.
(7) Futerko, P. M.; Fontijn, A. J. Chem. Phys. 1991, 95, 8065.
Sci. Technol. 1994, 101, 59.
(33) Harter, J. S. S.; Campbell, M. L.; McClean, R. E. Manuscript in
preparation.
(34) Robinson, P. J.; Holbrook, K. A. Unimolecular Reactions; Wiley-
Interscience: New York, 1972.
(35) Brown, C. E.; Mitchell, S. A.; Hackett, P. A. J. Phys. Chem. 1991,
95, 1062.
(36) McClean, R. E.; Pasternack, L. J. Phys. Chem. 1992, 96, 9828.
(37) Ritter, D.; Weisshaar, J. C. J. Phys. Chem. 1990, 94, 4907.
JP9532172