Temperature-Dependent Rate Constants for the Reactions of Gas-Phase Lanthanides with O2

Article abstract of DOI:10.1021/jp991647c

The reactivity of the gas-phase lanthanide atoms Ln (Ln = La-Yb with the exception of Pm) with O₂ is reported. Lanthanide atoms were produced by the photodissociation of [Ln(TMHD)₃] and detected by laser-induced fluorescence. For all the lanthanides studied with the exception of Yb, the reaction mechanism is bimolecular abstraction of an oxygen atom. The bimolecular rate constants (in molecule^-1 cm³ s^-1) are described in Arrhenius form by k[Ce(¹G₄)] = (3.0 ± 0.4) × 10^-10 exp(-3.4 ± 1.3 kJ mol^-1/RT); Pr(⁴I_9/2), (3.1 ± 0.7) × 10^-10 exp(-5.3 ± 1.5 kJ mol^-1/RT); Nd(⁵I₄), (3.6 ± 0.3) × 10^-10 exp(-6.2 ± 0.4 kJ mol^-1/RT); Sm(⁷F₀), (2.4 ± 0.4) × 10^-10 exp(-6.2 ± 1.5 kJ mol^-1/RT); Eu(⁸S_7/2), (1.7 ± 0.3) × 10^-10 exp(-9.6 ± 0.7 kJ mol^-1/RT); Gd(⁹D₂), (2.7 ± 0.3) × 10^-10 exp(-5.2 ± 0.8 kJ mol^-1/RT); Tb(⁶H_15/2), (3.5 ± 0.6) × 10^-10 exp(-7.2 ± 0.8 kJ mol^-1/RT); Dy(⁵I₈), (2.8 ± 0.6) × 10^-10 exp(-9.1 ± 0.9 kJ mol^-1/RT); Ho(⁴I_15/2), (2.4 ± 0.4) × 10^-10 exp(-9.4 ± 0.8 kJ mol^-1/RT); Er(³H₆), (3.0 ± 0.8) × 10^-10 exp(-10.6 ± 1.1 kJ mol^-1/RT); Tm(²F_7/2), (2.9 ± 0.2) × 10^-10 exp(-11.1 ± 0.4 kJ mol^-1/RT), where the uncertainties represent ±2σ. The reaction barriers are found to correlate to the energy required to promote an electron out of the 6s subshell. The reaction of Yb(¹S₀) with O₂ reacts through a termolecular mechanism. The limiting low-pressure third-order rate constants are described in Arrhenius form by k₀[Yb(¹S₀)] = (2.0 ± 1.3) × 10^-28 exp(-9.5 ± 2.8 kJ mol^-1/RT) molecule^-2 cm⁶ s^-1.

Full text of DOI:10.1021/jp991647c

7274
J. Phys. Chem. A 1999, 103, 7274-7279
Temperature-Dependent Rate Constants for the Reactions of Gas-Phase Lanthanides with
O2
Mark L. Campbell*,†
Chemistry Department, United States NaVal Academy, Annapolis, Maryland 21402
ReceiVed: May 19, 1999; In Final Form: July 16, 1999
The reactivity of the gas-phase lanthanide atoms Ln (Ln ) La-Yb with the exception of Pm) with O2 is
reported. Lanthanide atoms were produced by the photodissociation of [Ln(TMHD)3] and detected by laser-
induced fluorescence. For all the lanthanides studied with the exception of Yb, the reaction mechanism is
bimolecular abstraction of an oxygen atom. The bimolecular rate constants (in molecule-1 cm3 s-1) are described
in Arrhenius form by k[Ce(1G4)] ) (3.0 ( 0.4) × 10-10 exp(-3.4 ( 1.3 kJ mol-1/RT); Pr(4I9/2), (3.1 ( 0.7)
× 10-10 exp(-5.3 ( 1.5 kJ mol-1/RT); Nd(5I4), (3.6 ( 0.3) × 10-10 exp(-6.2 ( 0.4 kJ mol-1/RT); Sm(7F0),
(2.4 ( 0.4) × 10-10 exp(-6.2 ( 1.5 kJ mol-1/RT); Eu(8S7/2), (1.7 ( 0.3) × 10-10 exp(-9.6 ( 0.7 kJ mol-1/
RT); Gd(9D2), (2.7 ( 0.3) × 10-10 exp(-5.2 ( 0.8 kJ mol-1/RT); Tb(6H15/2), (3.5 ( 0.6) × 10-10 exp(-7.2
( 0.8 kJ mol-1/RT); Dy(5I8), (2.8 ( 0.6) × 10-10 exp(-9.1 ( 0.9 kJ mol-1/RT); Ho(4I15/2), (2.4 ( 0.4) ×
10-10 exp(-9.4 ( 0.8 kJ mol-1/RT); Er(3H6), (3.0 ( 0.8) × 10-10 exp(-10.6 ( 1.1 kJ mol-1/RT); Tm(2F7/2),
(2.9 ( 0.2) × 10-10 exp(-11.1 ( 0.4 kJ mol-1/RT), where the uncertainties represent (2σ. The reaction
barriers are found to correlate to the energy required to promote an electron out of the 6s subshell. The
reaction of Yb(1S0) with O2 reacts through a termolecular mechanism. The limiting low-pressure third-order
rate constants are described in Arrhenius form by k0[Yb(1S0)] ) (2.0 ( 1.3) × 10-28 exp(-9.5 ( 2.8 kJ
mol-1/RT) molecule-2 cm6 s-1.
Introduction
gas-phase lanthanide metals with diatomic oxygen. In general,
the chemistry of condensed-phase lanthanides is very similar.
This similarity is due to the lanthanides differing only in the
number of inner shell f electrons. It might be expected that since
f electrons do not extend spatially as far as the outer 5d and 6s
valence electrons, that the lanthanides’ reactivity in the gas phase
will be unaffected as the f electrons fill across the series. Here
we report the first experiments determining the temperature-
dependent rate constants of the neutral, gas-phase lanthanides
reacting with O2.
Studies of temperature-dependent rate constants for both the
main group and transition metal reactions with O2 in the gas
phase have been extensive.1-13 Atmospheric scientists are
interested in these reactions owing to the presence of metallic
species in the atmosphere from meteor ablation14,15 and pollution
from industrial smokestacks.16 These reactions are also of
interest in chemical vapor deposition processes.17-19
The mechanism for the reaction of a gas-phase metal atom
with oxygen is dependent upon the thermodynamics of the
process. About half of the naturally occurring metals are able
to exothermically abstract an oxygen atom to form the metal
monoxide. In the cases where the abstraction process is
exothermic, the abstraction reaction is the observed mechanism
while the reaction barriers are observed to be 15 kJ/mol or
less.2,9-13
For many of the gas-phase metal atoms, the abstraction
reaction is endothermic. Thus, at moderate temperatures, the
termolecular reaction to produce the metal dioxide is the only
viable process. The main group metals (with the exception of
Ba, Ge, and Sn) and about half of the transition metals can only
react with O2 through the termolecular process. The termolecular
kinetics of these atoms is quite diverse. Brown et al. reported
on the 3d transition metal series and found that atoms with a
ground or low-lying electronic state with a single electron in
the valence s subshell react faster than those metals atoms with
ground states with a filled outer s subshell.3 This behavior has
also generally been observed for the 4d and 5d transition metal
series.4-8
Experimental Section
The experimental arrangement has been described in detail
elsewhere and will be described only briefly here.20 Pseudo-
first-order kinetic experiments ([Ln] , [O2]) were carried out
in an apparatus with slowly flowing gas using a laser photolysis/
laser-induced fluorescence (LIF) technique. Laser photolysis of
a lanthanide metal precursor [Ln(TMHD)3, where TMHD
represents the 2,2,6,6-tetramethyl-3,5-heptanedionato ion] pro-
duced the gas-phase lanthanide metal atoms, and laser-induced
fluorescence was used to probe their temporal behavior. The
reaction chamber is a stainless steel reducing four-way cross
with attached sidearms and a sapphire window for optical
viewing. The reaction chamber is enclosed within a convection
oven (Blue M, model 206F, Tmax ) 623 K) with holes drilled
to allow for the exiting sidearms and the telescoping of the LIF
signal to the PMT.
Lanthanide metal atoms were produced by the 248 nm
photodissociation of the appropriate precursor with the focused
output of an excimer laser (Lambda Physics Lextra 200). The
focusing lens (f ) 564 mm) was placed approximately one focal
length from the reaction zone. The ground states of the
Thus far, no experiments have been reported in the literature
regarding the determination of bulk rate constants for neutral,
† Henry Dreyfus Teacher-Scholar.
10.1021/jp991647c This article not subject to U.S. Copyright. Published 1999 by the American Chemical Society
Published on Web 08/24/1999

Reactions of Gas-Phase Lanthanides with O2
J. Phys. Chem. A, Vol. 103, No. 36, 1999 7275
TABLE 1: Laser-Induced Fluorescence Transitions Used to
Probe the Ground State Lanthanide Atoms
lanthanide wavelength (nm)a
lanthanide wavelength (nm)a
Ce(1G4)
491.532
513.344
492.453
380.394
459.403
398.784
Tb(6H15/2
Dy(5I8)
)
432.643
421.172
405.393
400.796
418.762
398.799
Pr(4I9/2
)
Nd(5I4)
Ho(4I15/2
Er(3H6)
)
Sm(7F0)
Eu(8S7/2
Gd(9D2)
)
Tm(2F7/2
Yb(1S0)
)
a Wavelengths taken from ref 21.
lanthanide metal atoms were detected via LIF using an excimer-
pumped dye laser (Lambda Physics Lextra 50/ScanMate 2E).
The wavelengths utilized to probe the ground state lanthanide
atoms are listed in Table 1. The fluorescence was detected at
90° to the counterpropagated laser beams with a three-lens
telescope imaged through an iris. Interference filters were used
to isolate the LIF. A photomultiplier tube (Hamamatsu R375)
was used in collecting the LIF which was subsequently sent to
a gated boxcar sampling module (Stanford Research Systems
SR250), and the digitized output was stored and analyzed by a
computer. Real time viewing of the photolysis prompt emission
and LIF signal were accomplished using a LeCroy model 9360
digital oscilloscope.
Nitrogen gas was used as the primary buffer gas; however,
for the majority of lanthanides, a mixture of both N2 and
methane was used. As a result of their low vapor pressures at
room temperature, the lanthanum precursors required heating
in order to get enough precursor into the gas phase. The
lanthanide metal precursor vapor was entrained in a flow of N2
buffer gas. The diluted precursor, N2 and CH4 buffer gases and
O2 flowed through calibrated mass flow meters and flow
controllers (MKS Types 1459C and 0258C, and Matheson
models 8102 and 8202-1423) prior to admission to the reaction
chamber. Each sidearm window was purged with a slow flow
of buffer gas to prevent deposition of metal atoms and other
photoproducts. Total flows varied between 250 and 900 sccm
(sccm ) standard cubic centimeters per minute); however, most
experiments used a flow of approximately 700 sccm. Pressures
were measured with MKS Baratron manometers, and chamber
temperatures were measured with a thermocouple.
Figure 1. Typical decay curves with added O2 at 448 K: (a) Tb(6H15/2),
Ptotal ) 5.0 Torr, P(O2) ) 12.4 mTorr, τ ) 66.1 µs; (b) Nd(5I4), Ptotal
) 5.0 Torr, P(O2) ) 4.9 mTorr, τ ) 99.3 µs; (c) Er(3H6), Ptotal ) 20.0
Torr, P(O2) ) 14.2 mTorr; τ ) 143 µs. The solid lines through the
data are exponential fits. The inset is a ln plot of the data.
buffer gas pressure, kd represents the depletion rate constant
due to the reaction of the lanthanide metal with precursor
molecules and fragments and diffusion out of the detection zone,
and τ is the observed time constant for lanthanide metal
depletion in the presence of a given [O2], precursor, and buffer
gas pressure. The LIF signal is proportional to the lanthanide
metal number density, [Ln]. Thus, for a simple pseudo-first-
order system in which production of excited states is insignifi-
cant, the LIF signal of the ground state at a time t after photolysis
is
[Ln](t) LIF(t) ) LIF0 exp(-t/τ)
(2)
where LIF0 is the initial LIF signal immediately following the
photolysis laser pulse. LIF decays which exhibit single-
exponential behavior were fitted using a least-squares procedure
to determine τ.
In some cases, as a result of the photolysis of the lanthanide
metal precursor, atoms in excited states are produced in
significant quantities. Where possible, it is best to find a
quencher gas to help relax excited states as rapidly as possible.
Methane was used as a quencher gas for all of the lanthanides
studied with the exception of Ce, Eu, Ho, and Yb. For a ground
state in which quenching of excited states after the initial
photolysis laser pulse is important, the time dependence of the
ground state’s concentration will exhibit biexponential behavior.
For cases where the decay rate of an excited state down to the
ground state is fast compared to the removal rate of the ground
state, cascade processes are unimportant at long delay times
and the ground state removal rate is determined by analysis of
the decay at long times. Biexponential behavior of this type
was observed for a few of the lanthanide metals. Typical plots
of exponential decays are shown for terbium, neodymium, and
erbium in Figure 1. For those lanthanides where growth was
observed, time constants were determined by adjusting the range
of the linear regression analysis; that is, the range did not include
the beginning of the decay when growth was present. Decays
were typically analyzed after a delay of up to one reaction
lifetime and included data for a length of two to three reaction
lifetimes
The delay time between the photolysis pulse and the dye laser
pulse was varied by a digital delay generator (Stanford Research
Systems DG535) controlled by a computer interfaced through
a Stanford Research Systems SR245 computer interface. The
trigger source for these experiments was scattered pump laser
light incident upon a fast photodiode. LIF decay traces typically
consisted of 200 data points, each point averaged from two to
four laser shots.
Materials. The Ln(TMHD)3 precursors were obtained from
Strem Chemical. O2 (MG Industries, 99.8%), CH4 (Linde,
ultrahigh purity grade, 99.99%), and N2 (Potomac Airgas, Inc.,
99.998%) were used as received.
Data Analysis and Results
The decay rates of the gas-phase lanthanide atoms as a
function of oxygen pressure were investigated as a function of
temperature. Under pseudo-first-order conditions in which no
production processes occur after the initial photolysis event, the
decrease in the [Ln] with time following the photolysis laser
pulse is given by:
-d[Ln]/dt ) (kobs[O2] + kd)[Ln] ) (1/τ)[Ln]
(1)
Once the reaction time constant has been determined for a
series of oxygen number densities, the slope of a plot of 1/τ vs
where kobs is the observed rate constant due to O2 at a fixed

7276 J. Phys. Chem. A, Vol. 103, No. 36, 1999
Campbell
TABLE 2: Temperature Dependence of the Bimolecular
Rate Constants (10-11 molecule-1 cm3 s-1) for the Reaction
of Ln(g) + O2(g) f LnO(g) + O(g)
T (K) Ce Pr Nd Sm Eu Gd Tb Dy Ho Er Tm
298a
348
373
3.0
0.39
1.7 0.8 0.50 0.43 0.32
9.8 5.4 4.3
0.70 4.8 2.6 1.2 0.92 0.72 0.59
9.8 5.5 4.7 3.0 0.68 5.0 3.3 1.4 1.0 0.84 0.78
398 11.0 6.0 5.7 3.9 0.83 5.4 4.7 1.9 1.4 1.3 0.99
423 11.0 6.2 6.6 3.9 1.0 6.2 4.1 2.0 1.6 1.8 1.3
448 13.0 8.2 7.4 4.7 1.4 6.5 4.7 2.3 2.1 1.9 1.4
473 12.0 9.5 7.5 5.6 1.4 7.4 5.4 2.5 2.3 1.8 1.6
498 14.0 9.8 7.7 5.8 1.6 8.1 5.8 3.0 2.2 2.6 2.0
523
548
573
598
623
9.5 9.0 6.0 1.8 8.3 7.4 3.6 3.1 2.3 2.1
9.3 9.5 5.6 2.1
9.6 10.0 6.3 2.4
7.6 4.0 2.9 3.1 2.7
8.0 4.5 3.2 3.2 2.8
8.2 4.8 3.3 3.5 3.0
10.0
11.0
2.5
Figure 2. Typical plots for determining ksecond for Ln + O2 at 448 K.
The solid line for each set of data is a linear regression fit from which
ksecond is obtained.
8.7
3.6 3.8 3.3
a Room-temperature varied from 296 to 300 K.
Figure 4. Arrhenius plots for the collisional disappearance of Ln with
O2. Each solid line is a linear regression fit of the rate constants to the
equation k(T) ) A e-E /RT
.
a
Figure 3. Typical plots for determining ksecond for Yb(1S0) + O2
illustrating the third-order behavior of this reaction. The solid line for
each set of data is a linear regression fit from which ksecond is obtained.
TABLE 3: Arrhenius Parameters and Promotion Energies
for the Lanthanides
A (10-10 molecule-1
cm3 s-1
Ea
(kJ/mol)
6s2-6s1 promotion
[O2] yields the observed second-order rate constant, kobs. Figure
2 illustrates examples of these types of plots for cerium,
gadolinium, terbium, dysprosium, and erbium. The rate constants
for the reaction of Yb with O2 were found to be pressure
dependent. The slopes of the linear fits in Figure 3 illustrate
the pressure and temperature dependence of the second-order
rate constants for this reaction. The relative uncertainty (i.e.,
the reproducibility) of the second-order rate constants is
estimated at (20% based on repeated measurements of rate
constants under identical temperature and total pressure condi-
tions. The absolute uncertainties are conservatively estimated
to be (30% and are based on the sum of the statistical scatter
in the data, uncertainty in the flowmeter and flow controller
readings (5%) and the total pressure reading (1%), uncertainties
due to incomplete gas mixing, and uncertainties due to
incomplete relaxation of the excited electronic states to the
ground state.
a
lanthanide
)
energy (cm-1
)
b
La(2D3/2
)
4.0 ( 0.6
3.0 ( 0.4
3.1 ( 0.7
3.6 ( 0.3
2.4 ( 0.4
1.7 ( 0.3
2.7 ( 0.3
3.5 ( 0.6
2.8 ( 0.6
2.4 ( 0.4
3.0 ( 0.8
2.9 ( 0.2
3.1 ( 1.7
3.4 ( 1.3
5.3 ( 1.5
6.2 ( 0.4
6.2 ( 1.5
9.6 ( 0.7
5.2 ( 0.8
7.2 ( 0.8
9.1 ( 0.9
9.4 ( 0.8
10.6 ( 1.1
11.1 ( 0.4
2668
2369
6714
Ce(1G4)
Pr(4I9/2
)
Nd(5I4)
8475
Sm(7F0)
10801
12924
6378
Eu(8S7/2
Gd(9D2)
)
Tb(6H15/2
Dy(5I8)
)
8190
15567
15855
16321
16742
Ho(4I15/2
Er(3H6)
)
Tm(2F7/2
)
a Reference 22. b Arrhenius parameters calculated from the rate
constants in ref 2.
lanthanide to the equation k(T) ) A e-E /RT was performed; the
frequency factors, A, and activation energies, Ea, are listed in
Table 3.
a
Measured rate constants for the ground states of the lan-
thanides (with the exception of Pm and Yb) reacting with O2
are listed in Table 2. The total pressure for the experiments
involving Ce, Pr, Nd, Sm, Eu, Gd, and Tb was 5.0 Torr. The
experiments involving Dy, Ho, Er, and Tm were performed at
a total pressure of 20.0 Torr. Extensive pressure dependence
studies were not performed; however, rate constants for Ce at
5 and 10 Torr at 373 K and Gd at 5 and 10 Torr at 348 K were
found to be independent of total pressure within experimental
uncertainty. An Arrhenius plot (ln k vs 1/T) of the rate constants
is shown in Figure 4. A fit of the rate constants for each
Measured rate constants for Yb(1S0) with O2 as a function of
temperature and pressure are listed in Table 4. The low-pressure,
third-order rate constants, k0, for each temperature were ap-
proximated by fitting the second-order rate constants to a
quadratic equation in buffer gas number density. The third-order
rate constant was determined from the slope of the fit at [M] )
0. Representative fits of the data are shown in Figure 5, and
the calculated k0 values are listed in Table 4. A fit of the third-
order rate constants to the Arrhenius equation yields k0 ) (2.0
( 1.3) × 10-28 exp(-9.5 ( 2.8 kJ mol-1/RT) molecule-2 cm6
s-1 (Figure 6).

Reactions of Gas-Phase Lanthanides with O2
J. Phys. Chem. A, Vol. 103, No. 36, 1999 7277
TABLE 4: Second-Order Rate Constants (10-12 molecule-1
cm3 s-1) and Low-Pressure Limiting Third-Order Rate
Constants, k0, (10-29 molecule-2 cm6 s-1) for the Reaction:
Yb + O2 + N2 f YbO2
T (K) 2.5 Torr 5.0 Torr 7.5 Torr 10.0 Torr
k0
297
373
423
473
523
573
623
0.38
0.58
0.97
0.86
0.94
1.3
0.46
0.84
1.4
1.2
1.6
2.2
1.8
0.55
1.1
1.8
1.5
2.0
2.6
2.5
0.64
1.3
2.1
1.8
2.5
2.9
3.1
0.42 ( 0.12
0.86 ( 0.13
1.7 ( 0.2
1.6 ( 0.3
2.1 ( 0.3
3.4 ( 0.2
2.9 ( 0.5
1.2
Figure 7. Plot of the experimental activation energies vs the electronic
promotion energy from the ground state to the lowest energy state with
a 6s1 electron configuration.
the vaporized metal reacting with O2.27-31 Thus, an abstraction
of an oxygen atom to produce the metal oxide is the indicated
mechanism for all the lanthanides with the exception of
ytterbium.
Energy barriers of only one of the lanthanides reported here
have been previously reported for the abstraction reaction with
oxygen. Using a beam-beam experiment, a threshold energy
of 14.5 ( 1.9 kJ/mol was determined for the Eu + O2 reaction.32
This threshold value is slightly higher than the activation energy
reported here; however, this threshold energy was determined
from the translational energy component only so that a direct
comparison with the value reported here cannot be made.
The experiments in this study indicate the gas-phase lan-
thanides show differences of up to almost 2 orders of magnitude
in their rate constants for the reactions with oxygen. Similar to
the results observed here, studies of gas-phase lanthanides with
N2O showed diverse kinetic behavior. In the study of the N2O
reactions, it was shown that the activation energies correlated
with the excitation energy required to promote an s electron
from an s2 to an s1 configuration.33 Listed in Table 3 are the
energies required to promote an electron from the ground state
to a configuration with a single electron in the 6s subshell.
Figure 7 shows there is a strong correlation between the
measured activation energies and the 6s2-6s1 promotion ener-
gies. Since the 6s orbital is the outermost valence orbital and
the primary bonding orbital of the lanthanide, the interaction
of the 6s orbital with O2 is expected to be the dominant
interaction at long range. The character of the potential upon
the onset of orbital overlap will be determined by the interaction
between the s orbital of the lanthanide and the singly occupied
π* antibonding orbitals of O2. The ground states of the
lanthanides are expected to be repulsive with O2 because the
6s orbital in all the lanthanides is doubly occupied. Thus, the
promotion of an electron out of the 6s subshell would be
expected to reduce this repulsion. Our experimental results
indicate the reduction of this repulsive interaction is important
for these reactions since the activation energies are strongly
correlated to the s2-s1 promotion energies.
Figure 5. Pressure/temperature dependence of the reaction of Yb(1S0)
+ O2 in N2 buffer. The solid lines are quadratic fits in [M]. The value
of kthird (Table 3) for each temperature is determined from the slope of
the fit at [M] ) 0.
Figure 6. Arrhenius plot of the third-order rate constants for Yb(1S0)
with O2. The solid line is a linear regression fit of the rate constants to
the equation k0(T) ) A e-E /RT
.
a
Discussion
With the exception of europium and ytterbium, the abstraction
reactions of the gas-phase lanthanides with oxygen are exo-
thermic.23,24 The Eu + O2 reaction is approximately thermo-
neutral (∆H ) 12 ( 17 kJ/mol) so that the measured barrier
might be partially due to a thermodynamic constraint. The Yb
+ O2 reaction is 80 kJ/mol endothermic so that the abstraction
reaction is precluded at moderate temperatures due to the
thermodynamics of the process. Although extensive pressure-
dependent studies were not performed for these reactions (with
the exception of ytterbium), the instances in which pressure was
varied indicated termolecular processes were unimportant. The
Arrhenius behavior (i.e., activation energies of less than 15 kJ/
mol and frequency factors near the gas kinetic collision rate)
observed for these reactions is typical of the bimolecular
abstraction mechanism seen with other metal atoms with
O2.2,25,26 For the cases of Ce, Sm, Gd, Dy, and Ho, the metal
oxide product has been directly observed from the reaction of
The electronic structure of the lanthanides has been shown
to be important in other bonding and kinetic circumstances. The
trends in bond dissociation energies of the lanthanide monoxides
can be accounted for nearly quantitatively by variations in
electronic transition energies.34-36 This model assumes the
binding energies of the monoxides are nearly equal when the
metal atom is in the 4fn-15d1 configuration so that the bond
dissociation energy values for those metals requiring 4f-5d
promotion are reduced below the baseline energy by the amount

7278 J. Phys. Chem. A, Vol. 103, No. 36, 1999
Campbell
of this promotion energy. This model actually assumes the metal
Summary
oxide is described predominantly by the ionic structure M2+O2-
,
We have measured the second-order rate constants as a
function of temperature for the reactions of ground state
lanthanide atoms with oxygen. With the exception of Yb, the
gas-phase lanthanides abstract an oxygen atom to produce the
metal oxide. Energy barriers vary from 3.4 for Ce(1G4) to 11.1
kJ/mol for Tm(2D3/2). The reaction barriers are found to correlate
to the energy required to promote an electron out of the filled
6s subshell. The reaction of Yb(1S0) with O2 proceeds through
a termolecular process with a barrier of 9.5 kJ/mol.
although the correlation between experimental and predicted
values of the bond dissociation energies is quite good if the
free atom values of the promotion energies are used.36 In fact,
Murad indicated that for praseodymium, the Pr+O- structure is
the more important configuration.37 This structure results from
the combination of the 4f36s1 (Pr+) and 2p5 (O-) states. Thus,
the requirement for reaction involves an ionic/covalent curve-
crossing in which the covalent surface asymptotically correlates
to the s1 configuration. Therefore, the smaller the difference
between the asymptotic energies of the lowest s1 and s2 energy
states, the greater the probability of reaction. Murad’s analysis
was only for Pr because the spectroscopic data available was
limited to Pr. However, the observed promotion energy relation
might indicate the M+O- structure is also important for the other
lanthanide monoxides.
Acknowledgment. This research was supported by a Cottrell
College Science Award of Research Corporation.
References and Notes
(1) Plane, J. M. C. In Gas-Phase Metal Reactions; Fontijn, A., Ed.;
Elsevier: Amsterdam, 1992; p. 29.
The reaction of the lanthanide ions with hydrocarbons also
indicates a relationship between the reactivity and electron
configuration. For the reactions of lanthanide ions with
hydrocarbons,38-42 it has been established that large variations
in the activation of C-H and C-C bonds correlate with the
energies required to promote a nonbonding 4f electron to a
reactive 5d or 6s valence orbital. Insertion into a C-H or C-C
bond apparently requires an electronic configuration on the metal
center which is capable of forming two σ bonds. The f orbitals
do not extend far enough spatially to be involved in the bonding;
thus, the promotion of an f electron to a valence orbital is
required to facilitate reaction. Thus, as the energy required to
promote a metal ion electron from an fns1configuration to an
fn-1d1s1 configuration increases, the activation barrier for the
reaction increases. Thus, the rich chemistry of the gas-phase
lanthanide atoms and ions appears to result from the variation
in the electronic energy states in these systems.
(2) Campbell, M. L. Chem. Phys. Lett. 1998, 294, 339 and references
therein.
(3) Brown, C. E.; Mitchell, S. A.; Hackett, P. A. J. Phys. Chem. 1991,
95, 1062.
(4) Campbell, M. L. J. Chem. Soc., Faraday Trans. 1996, 92, 4377
and references therein.
(5) Vinckier, C.; Christiaens, P.; Hendrickx, M. In Gas-Phase Metal
Reactions; Fontijn, A., Ed.; Elsevier: Amsterdam, 1992; p 57.
(6) Campbell, M. L. Laser Chem. 1998, 17, 219.
(7) Helmer, M.; Plane, J. M. C. J. Chem. Soc., Faraday Trans. 1994,
90, 395.
(8) Campbell, M. L. J. Chem. Soc., Faraday Trans. 1998, 94, 353.
(9) Garland, N. L.; Nelson, H. H. Chem. Phys. Lett. 1992, 191, 269.
(10) LePicard, S. D.; Canosa, A.; Travers, D.; Chastaing, D.; Rowe, B.
R.; Stoecklin, T. J. Phys. Chem. 1997, 101, 9988.
(11) Brown, A.; Husain, D. Can. J. Chem. 1976, 54, 4.
(12) Wiesenfeld, J. R.; Yuen, M. J. J. Phys. Chem. 1978, 82, 1225.
(13) Fontijn, A.; Bajaj, P. N. J. Phys. Chem. 1996, 100, 7085.
(14) Granier, C.; Jegou, J. P.; Megie, G. Geophys. Res. Lett. 1989, 16,
243.
(15) Plane, J. M. C. Int. ReV. Phys. Chem. 1991, 10, 55.
(16) Fontijn, A.; Blue, A. S.; Narayan, A. S.; Bajaj, P. N. Combust.
Sci. Technol. 1994, 101, 59.
The pressure dependence of the Yb + O2 reaction indicates
a termolecular addition reaction to produce YbO2 with a barrier
of 9.5 ( 2.8 kJ/mol. It is well-known that third-order reactions
of closed-shell metal atoms with O2 are characterized by an
energy barrier. This barrier is located in the entrance channel
at the crossing of the covalent and ionic potential energy
surfaces.43 With its [Xe]4f146s2 configuration, ytterbium has the
same filled outer s subshell as the alkaline earth metals. In its
reaction with O2, ytterbium is actually more similar to the other
alkaline earths than barium. The abstraction reaction of barium
with O2 is exothermic and proceeds as a bimolecular process
with a barrier of 7.06 ( 0.23 kJ/mol.25 The abstraction reactions
of magnesium, calcium, and strontium with O2 are endothermic
and not thermodynamically feasible at moderate temperatures.
These alkaline earths are observed to undergo termolecular
processes with O2.43-46 In each of these cases, a barrier has
been observed from the temperature dependence of the limiting
low-pressure, third-order rate constant. The barriers for mag-
nesium, calcium, and strontium have been reported by Vinckier
and co-workers as 16.8 ( 0.8, 10.2 ( 0.4, and 8.7 ( 0.6 kJ/
mol, respectively.43-45 Vinckier rationalized the trend in these
barriers based on a model which assumes a partial charge
transfer between the metal atom and oxygen molecule; that is,
there should be a qualitative correlation between the activation
energy and ionization energy. The ionization energies of Mg,
Ca, Sr, and Yb are 7.64, 6.11, 5.69, and 6.25 eV, respec-
tively.22,47 The barrier observed here for ytterbium is similar to
the barriers observed for calcium and strontium. The uncertain-
ties in the activation energies precludes a more definite
conclusion regarding the charge transfer mechanism.
(17) Green, M. L.; Gross, M. E.; Papa, L. E.; Schnoes, K. J.; Brasen,
D. J. Electrochem. Soc. 1985, 132, 2677.
(18) Ouyang, M.; Hiraoka, H. Mater. Res. Bull. 1997, 32, 1099.
(19) Jones, S. L.; Kumar, D.; Singh, R. K.; Holloway, P. H. Appl. Phys.
Lett. 1997, 71, 404.
(20) Campbell, M. L.; McClean, R. E. J. Chem. Soc., Faraday Trans.
1995, 91, 3787.
(21) Meggers, W. F.; Corliss, C. H.; Scribner, B. F. Tables of Spectral-
Line Intensities, Part I Arranged by Elements; NBS Monograph 145; U.S.
Government Printing Office: Washington, DC, 1975.
(22) Martin, W. C.; Zalubas, R.; Hagan, L. Atomic Energy Levelss
The Rare-Earth Elements. Natl. Stand. Ref. Data Ser. (U.S. Natl. Bur. Stand.)
1978, NSRDS-NBS 60.
(23) Wagman, D. D.; Evans, W. H.; Parker, V. B.; Schumm, R. H.;
Halow, I.; Bailey, S. M.; Churney, K. L.; Nuttall, R. L. J. Phys. Chem.
Ref. Data 1982, 11 (Suppl. 2).
(24) Pedley, J. B.; Marshall, E. M. J. Phys. Chem. Ref. Data 1983, 12,
967.
(25) Nien, C.-F.; Plane, J. M. C. J. Phys. Chem. 1991, 94, 7193.
(26) Campbell, M. L. J. Phys. Chem. A 1998, 102, 892.
(27) Linton, C.; Dulick, M.; Field, R. W.; Leyland, P. C.; Barrow, R.
F. J. Mol. Spectrosc. 1983, 102, 441.
(28) Linton, C.; Bujin, G.; Rana, R. S.; Gray, J. A. J. Mol. Spectrosc.
1987, 126, 370.
(29) Carette, P.; Hocquet, A.; Douay, M.; Pinchemel, B. J. Mol.
Spectrosc. 1987, 124, 243.
(30) Linton, C.; Gaudet, D. M.; Schall, H. J. Mol. Spectrosc. 1983, 102,
441.
(31) Liu, Y. C.; Linton, C.; Schall, H.; Field, R. W. J. Mol. Spectrosc.
1984, 104, 72.
(32) Dirscherl, R.; Michel, K. W. Chem. Phys. Lett. 1976, 43, 547.
(33) Campbell, M. L. J. Chem. Phys. 1999, 111, 562.
(34) Ames, L. L.; Walsh, P. N.; White, D. J. Phys. Chem. 1967, 71,
2707.
(35) Hildenbrand, D. L. Chem. Phys. Lett. 1977, 48, 340.
(36) Murad, E.; Hildenbrand, D. L. J. Chem. Phys. 1980, 73, 4005.
(37) Murad, E. Chem. Phys. Lett. 1978, 59, 359.

Reactions of Gas-Phase Lanthanides with O2
J. Phys. Chem. A, Vol. 103, No. 36, 1999 7279
(38) Schilling, J. B.; Beauchamp, J. L. J. Am. Chem. Soc. 1988, 110,
15.
(39) Yin, W. W.; Marshall, A. G.; Marcalo, J.; de Matos, A. P. J. Am.
Chem. Soc. 1994, 116, 8666.
(43) Vinckier, C.; Helaers, J. J. Phys. Chem. A 1998, 102, 8333.
(44) Vinckier, C.; Christiaens, P. Bull. Soc. Chim. Belg. 1992, 101, 10.
(45) Vinckier, C.; Remeysen, J. J. Phys. Chem. 1994, 98, 10535.
(46) Nien, C.-F.; Rajasekhar, B.; Plane, J. M. C. J. Phys. Chem. 1993,
(40) Cornehl, H. H.; Heinemann, C.; Schroder, D.; Schwarz, H.
Organometallics 1995, 14, 992.
(41) Gibson, J. K. J. Phys. Chem. 1996, 100, 15688.
(42) Gibson, J. K.; Haire, R. G. J. Phys. Chem. A 1998, 102, 10746.
97, 6449.
(47) Moore, C. E. Atomic Energy Levels as Derived from the Analysis
of Optical Spectra. Natl. Stand. Ref. Data Ser. (U.S. Natl. Bur. Stand.) 1971,
NSRDS-NBS 35; Vols. I and II.