Transitions in order and molecularity with temperature in gaseous metal oxidation reactions. The Sb-O2 system

Article abstract of DOI:10.1524/zpch.2000.214.11.1521

Examples of reactions which are termolecular near room temperature and gradually switch over to complex-forming bimolecular reactions with increasing temperature are discussed. The Sb/O₂ system, which is known from D. Husain's work to be termolecular at 300 K, has been studied here from 1165-1495 K at 12 to 52 mbar in a high-temperature fast-flow reactor (HTFFR). Sb was vaporized from a crucible and atomic resonance absorption spectrometry was used to measure the rates of Sb consumption. Under these conditions the reaction, identified as Sb + O₂ → SbO + O, is pressure-independent with k(T) = 6.3×10^-11 exp (-9107 K/T) cm³ molecule^-1 s^-1. The 2σ precision limits are ±8% and the corresponding estimated accuracy limits are ±24%. The results suggest that D_o(Sb-O) = 444±28 kJ mol^-1. The results are compared to those of other endothermic metal-atom O₂ abstraction reactions, all of which have pre-exponentials on the order of 10^-10 cm³ molecule^-1 s^-1 and activation energies for abstraction close to the endothermicity. This allows assessments of the relative importance of bi-, and ter- molecular kinetics, as illustrated for As + O₂. by Oldenbourg Wissenschaftsverlag, Muenchen.

Full text of DOI:10.1524/zpch.2000.214.11.1521

Zeitschrift für Physikalische Chemie, 214, 11, 1521Ϫ1531 (2000)
by Oldenbourg Wissenschaftsverlag, München

Transitions in Order and Molecularity
with Temperature in Gaseous Metal Oxidation
Reactions. The Sb-O System
2
By Biljana Cosic and Arthur Fontijn*
High-Temperature Reaction Kinetics Laboratory, The Isermann Department of Chemical
Engineering, Rensselaer Polytechnic Institute, Troy, NY 12180-3590, USA
Dedicated to Prof. Dr. Dr. h. c. mult. Jürgen Troe
on the occasion of his 60th birthday
(Received May 30, 2000; accepted July 20, 2000)
High Temperature Kinetics / Gas-phase / Molecularity / Metal Atoms /
Metal Oxidation / Rate Coefficients / Activation Energies / Sb / As / O
2
/
SbO Bond Energy
Examples of reactions which are termolecular near room temperature and gradually
switch over to complex-forming bimolecular reactions with increasing temperature are
discussed. The Sb/O system, which is known from D. Husain’s work to be termolecular
2
at 300 K, has been studied here from 1165Ϫ1495 K at 12 to 52 mbar in a high-tempera-
ture fast-flow reactor (HTFFR). Sb was vaporized from a crucible and atomic resonance
absorption spectrometry was used to measure the rates of Sb consumption. Under these
conditions the reaction, identified as Sb ϩ O
2
→ SbO ϩ O, is pressure-independent with
Ϫ11
3
Ϫ1 Ϫ1
k(T) ϭ 6.3ϫ10
exp (Ϫ9107 K/T) cm molecule s . The 2σ precision limits are
Ϯ8% and the corresponding estimated accuracy limits are Ϯ24%. The results suggest
Ϫ1
that D
thermic metal-atom O
0
(Sb-O) ϭ 444Ϯ28 kJ mol . The results are compared to those of other endo-
abstraction reactions, all of which have pre-exponentials on the
2
Ϫ10
3
Ϫ1 Ϫ1
order of 10 cm molecule
endothermicity. This allows assessments of the relative importance of bi-, and ter- molec-
ular kinetics, as illustrated for As ϩ O
s
and activation energies for abstraction close to the
2
.
1. Introduction
Luther and Troe have described and discussed the influence of temperature
on termolecular reactions [1]. Near or below room temperature many such
*
Corresponding author. E-mail: fontia@rpi.edu
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 6/7/15 5:45 PM

1
522
B. Cosic and A. Fontijn
reactions show the typical fall-off behavior from purely third order at low
pressures to an approach to second order at high pressures. For most reac-
tions this can be ascribed to the energy-transfer mechanism
A ϩ B ↔ AB*
(1a)
(1b)
AB* ϩ M → AB ϩ M .
As the temperature rises the stability of AB* decreases and the approach to
the second-order regime requires increasing pressures; in most experiments
this regime then is no longer observable. With an additional increase in
temperature a further change can take place, when B is di-, or poly-atomic
species, i.e. the participation and eventual dominance of a bimolecular ab-
straction reaction. This has been found in endothermic metal atom and metal
monoxide reactions with O . The reactants and products of such reactions
2
are usually thermally stable over wide temperature ranges. The reaction
paths can then be described by:
M
A ϩ O
2
↔ AO*
2
–→– AO
2
[
[
↓
(2)
Ͼ
AO ϩ O .
Where abstraction is energetically favorable it dominates, for example
in the O reaction with Ba [2], Al [3], Sn [4], Zr, Hf, Os [5], and various
2
lanthanides [6]. Where abstraction is too endothermic, O forms association
2
or insertion complexes, e.g. with alkali metals, Mg, Ca, Sr [2, 7, 8], Fe [9],
Cu [7, 10], Cr and Ni [10]. However, in the limited temperature ranges of
these studies no transition from ter-, to bi-molecular reaction was seen.
Wider temperature ranges can be covered in the Rensselaer HTFFR (high-
temperature fast-flow reactor) and pseudo-static HTP (high-temperature
photochemistry) facilities [11, 12]. In the HTP study of Cr/O [13] and the
2
HTFFR measurements on the AlO/O system [14], both the transition from
2
termolecular-third to -second order and that from termolecular to bimolecu-
lar have been observed. Thus, at room temperature these studies show the
typical fall-off behavior of termolecular reactions. At temperatures several
hundred Kelvins higher only the third order regime is evident, while at still
higher temperatures, (around 800 K for Cr and around 1000 K for AlO)
plots of observed rate coefficients versus pressure show an intercept and
can be described by
kobs ϭ kbi ϩ kter[M].
(3)
With further increase in temperature the intercept grows and the slope de-
creases in magnitude. At temperatures of 1510 K and 1690 K both processes
still contribute to the observed reaction.
It is likely that the bimolecular reactions in these cases proceed through
the route indicated by the solid lines in Eq. (2), i.e. the reactions are part of
“
the colorful world of the complex-forming bimolecular reactions” [15].
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 6/7/15 5:45 PM

Transitions in Order and Molecularity with Temperature .. .
1523
Evidence for this comes from the AlO/O
2
reaction and its iso-electronic
equivalent BO/O . The BO reaction is exothermic for abstraction to BO ϩ
2
2
O, has a negative temperature dependence, and is in Ar bath gas pressure-
independent to at least 300 mbar, from room temperature to at least about
1
000 K [16, 17]. This is indicative of a type of reaction where an intermedi-
ate complex preferentially redissociates to reactants and the heat of associ-
ation exceeds the energy required for dissociation of the complex to prod-
ucts [18]:
BO ϩ O ↔ BO* → BO ϩ O .
(4)
2
3
2
No pressure-dependence channel could be observed under the investigated
conditions, contrary to the AlO/O reaction which is endothermic for ab-
2
straction. An ab initio calculation has suggested that the BO reaction goes
through the insertion complex OBO(O) [19]. Matrix-isolation study of the
AlO reaction has identified an equivalent OAlO(O) complex with hot
ϩ
Ϫ
3
atoms, though with cold reactants the Al O
complex is favored [20].
Similarly, for the Cr/O reaction such studies have found OCrO and CrOO
2
complexes, with the latter preferred for cold reactants [21], in agreement
with the suggestion from the room temperature study [10].
The present study was undertaken to obtain information on the kinetics
of antimony in combustion, especially municipal solid waste incineration.
The presence of Sb in the emissions from such environments presents a
toxicologic and carcinogenic hazard [22Ϫ24]. The properties of Sb and its
compounds limited this investigation to temperatures above 1150 K (see
below), where Sb ϩ O is indicated to proceed as a bimolecular reaction
2
Ϫ1
Sb ϩ O → SbO ϩ O ∆ H° ϭ 64Ϯ42 kJ mol [25, 26] .
2
f
298
(5)
Room temperature measurements on this reaction have been made by Hu-
sain et al. [27, 28]. There, a third-order reaction was observed,
Sb ϩ O
2
ϩ M → SbO
2
ϩ M
(6)
Ϫ31
6
Ϫ2 Ϫ1
s
for which k ϭ (3.2Ϯ0.3)ϫ10 cm molecule
for M ϭ N . No fall-
2
off behavior was evident in their work, within the pressure range 20Ϫ
1
50 mbar.
2. Technique
The experiments have been performed in a high-temperature fast-flow reac-
tor (HTFFR) using atomic resonance absorption spectrometry (ARAS). The
HTFFR technique [29] and current procedures have been elaborated on in
many earlier publications [11, 12, 30]. Briefly, a vertical mullite reaction
tube (2.2 cm i.d.) is heated radiatively by SiC resistant heating elements in
an insulated, water-cooled vacuum housing. Antimony atoms are generated
in a stream of bath gas N , by evaporation of Sb powder from an alumina
2
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 6/7/15 5:45 PM

1
524
B. Cosic and A. Fontijn
Fig. 1. An example rate coefficient determination plot. Observed reaction zone length
Ϫ1
2
0 cm; P ϭ 26.9 mbar; υ ϭ 36 m s ; T ϭ 1260 K. The two measurements at the lowest
2
[O
] are taken first and last, respectively, to evaluate and correct for any drift.
crucible, heated resistively by tungsten wire. Downstream from the Sb
source, a mixture of oxidant and diluent was introduced through a movable
multi-perforated ring inlet, situated 10 or 20 cm upstream from the obser-
vation plane. Relative Sb concentrations, [Sb] were measured by monitor-
rel
4
4
ing one of three transitions: (Sb(6 P5/2) ← Sb(5 S3/2)) at λ ϭ 206.8 nm,
4
4
4
4
(
Sb(6 P ) ← Sb(5 S )) at λ ϭ 217.6 nm, and (Sb(6 P ) ← Sb(5 S )) at
3
/2
3/2
1/2
3/2
λ ϭ 231.1 nm. Resonance radiation was produced by a hollow cathode
lamp. The signal was filtered through an F/3.7 monochromator and detected
by a Thorn EMI 9831QA photomultiplier tube, the output of which was
measured with a Keithley 614 electrometer. The temperature of the reaction
zone was measured by a retractable Pt-Pt/13% Rh thermocouple, and cor-
rected for radiative heat effects [29, 31]. Pressure was determined using
an MKS Baratron transducer; gas flow rates were measured by calibrated
Hastings-Teledyne mass-flow controllers.
An estimate of the upper limit of absolute [Sb] was obtained from the
21
antimony rate of evaporation. The Sb loss rate, on the order of 10 mole-
Ϫ1
cule h , provided [Sb] 10Ϫ30 times smaller than [O ], assuming all Sb
2
was present in the monoatomic form, which is not the case (see Results
section). Thus, pseudo first-order conditions pertained.
Rate coefficient measurements were made in the stationary inlet mode
with reaction zone lengths of 10 or 20 cm. For a single rate coefficient
measurement, k
lengths. Sb concentrations, proportional to Ϫln(I/I
to that obtained at the lowest [O ] used [29, 31]. (Here I and I represent,
i
, five different [O
2
] were used at the fixed reaction zone
0
), were measured relative
2
0
respectively, the light intensity with and without metal atoms present.) Plots
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 6/7/15 5:45 PM

Transitions in Order and Molecularity with Temperature .. .
1525
2
of such ln[Sb]rel versus [O ], as shown in Fig. 1, yield straight lines with
slopes Ϫkt, where t is reaction time. A weighted linear regression was used
to determine k and σ [32] for each individual measurement.
i
i
The chemicals used were Sb powder (99.5%) Alfa Aesar, N
from the liquid, Praxair, and O (99.99%) from Matheson, which was flow-
ed through Drierite (CaSO ) drying towers.
2
(99.995%)
2
4
3. Results
The individual rate coefficients and the conditions under which they were
obtained are summarized in Table 1. The range of temperatures and pres-
sures covered in this study was somewhat limited, considering that the ap-
proximate operating ranges of HTFFRs span 300 to 1800 K and 10 to
1
50 mbar. In equilibrium the vapor phase of antimony is known to consist
to a major degree of oligomers [33, 34], while the concentration of free
atoms increases with temperature towards the boiling point [34]. As a result,
even though Sb is a rather volatile element, sufficient concentrations of
Sb atoms to allow absorption measurements could only be produced and
maintained at the temperatures of the reactor above about 1100 K and
source temperatures of around 1800 K. For the same reason the work was
additionally constrained to low and moderate pressures. The highest tem-
perature used was 1495 K, since at higher temperatures the product started
to dissociate, as manifested initially through the rate coefficients becoming
dependent on residence time (between 1500 and 1550 K), and above 1550 K
by a drop of k values at increasing temperatures. The other reaction con-
17
ditions varied were total concentration [M] from (0.64 to 2.62)ϫ10 mole-
Ϫ3
cule cm , corresponding to pressure P from 12 to 52 mbar, maximum oxi-
15
Ϫ3
dant concentration [O
average flow velocity υ from 20 to 63 m s , percent of Sb absorption,
-I)/I , from 18 to 55%, and observed reaction zone length z, 10 or 20 cm.
Over the limited temperature range used the data show no deviation
2
]max from (2.64 to 13.2)ϫ10 molecule cm ,
Ϫ1
(I
0
0
from Arrhenius behavior, Fig. 2. A fit to the k(T) ϭ A exp(ϪE K/T) form
using weighted linear regression [35, 36] yields
Ϫ11
3
Ϫ1 Ϫ1
s
k(1165Ϫ1495 K) ϭ 6.25ϫ10 exp(Ϫ9107 K/T) cm molecule
.
(7)
Examination of residual plots [k(T)Ϫk ]/k(T) vs. P, [M], [O
i
2
]
max, υ, and Rey-
nolds number Re ϭ (2υaP)/(νRT) [37] shows the individual rate coefficients
k to be independent of these parameters. The calculated variances and co-
i
2
A
Ϫ2
2
2
E
5
1
variance are σ
ϭ 7.48ϫ10 A , σ ϭ 1.25ϫ10 and σAE ϭ 9.64ϫ10 A.
The resulting Ϯ2σ precision limits vary from 7% at 1165 K to 8% at
k
1
495 K and the corresponding confidence limits from 23% to 24%, when
allowing Ϯ10% for uncertainty in the velocity profile factor η [29, 38] and
Ϯ20% for possible systematic errors.
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 6/7/15 5:45 PM

1
526
B. Cosic and A. Fontijn
Table 1. Summary of rate coefficient measurements on Sb ϩ O .
2
T
K
P
[M]
[O
2
15
]
υ
z
cm
% Sb
k
i
Ϯσki
3
1
7
Ϫ3
Ϫ3
Ϫ1
Ϫ1 Ϫ1
s
mbar 10 cm 10 cm m s
absorption cm molecule
Ϫ14
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
165 35.3
2.20
2.08
2.16
1.25
1.26
1.08
1.08
1.26
1.25
1.25
1.24
1.53
1.18
1.18
1.27
1.26
1.51
1.17
1.17
1.18
1.22
1.23
1.49
1.18
1.42
1.11
1.12
1.25
1.44
2.02
1.25
1.42
1.39
1.01
1.23
1.09
2.00
1.16
1.53
1.54
0.95
2.43
2.45
1.45
1.21
1.93
1.92
1.92
12.0
8.87
22
23
22
31
31
36
36
31
45
45
45
26
40
40
30
31
26
33
33
33
39
39
26
33
27
35
35
31
27
24
31
27
28
38
31
36
21
36
25
36
46
23
23
33
47
27
27
27
10
20
20
20
20
20
20
20
20
20
20
10
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
20
10
20
20
20
20
20
20
20
10
20
20
20
20
10
10
10
20
10
20
20
31
51
39
39
37
37
33
33a
26
(1.89Ϯ0.15)ϫ10
(2.61Ϯ0.21)ϫ10
(2.49Ϯ0.18)ϫ10
(2.69Ϯ0.25)ϫ10
(2.99Ϯ0.27)ϫ10
(3.72Ϯ0.38)ϫ10
(3.45Ϯ0.35)ϫ10
(3.64Ϯ0.33)ϫ10
(4.57Ϯ0.49)ϫ10
(4.59Ϯ0.52)ϫ10
(4.26Ϯ0.39)ϫ10
(1.85Ϯ0.15)ϫ10
(2.13Ϯ0.21)ϫ10
(2.61Ϯ0.26)ϫ10
(3.67Ϯ0.32)ϫ10
(3.02Ϯ0.28)ϫ10
(2.63Ϯ0.23)ϫ10
(2.28Ϯ0.20)ϫ10
(3.53Ϯ0.33)ϫ10
(3.16Ϯ0.33)ϫ10
(3.87Ϯ0.46)ϫ10
(5.63Ϯ0.81)ϫ10
(2.70Ϯ0.24)ϫ10
(3.11Ϯ0.28)ϫ10
(3.87Ϯ0.31)ϫ10
(4.24Ϯ0.46)ϫ10
(4.09Ϯ0.40)ϫ10
(4.25Ϯ0.39)ϫ10
(3.09Ϯ0.30)ϫ10
(2.50Ϯ0.21)ϫ10
(4.71Ϯ0.43)ϫ10
(2.92Ϯ0.23)ϫ10
(3.04Ϯ0.25)ϫ10
(3.98Ϯ0.39)ϫ10
(4.80Ϯ0.43)ϫ10
(3.11Ϯ0.30)ϫ10
(4.93Ϯ0.36)ϫ10
(3.90Ϯ0.35)ϫ10
(5.29Ϯ0.40)ϫ10
(4.46Ϯ0.33)ϫ10
(4.44Ϯ0.46)ϫ10
(3.54Ϯ0.26)ϫ10
(3.79Ϯ0.28)ϫ10
(5.11Ϯ0.54)ϫ10
(5.67Ϯ0.60)ϫ10
(6.54Ϯ0.50)ϫ10
(6.40Ϯ0.47)ϫ10
(5.71Ϯ0.47)ϫ10
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
174 33.8
174 35.0
181 20.3
182 20.5
182 17.7
187 17.7
187 20.6
194 20.7
194 20.7
195 20.5
195 25.2
202 19.6
202 19.6
203 21.1
204 21.0
205 25.1
205 19.5
205 19.5
205 19.6
206 20.3
206 20.4
206 24.7
207 19.7
214 23.7
214 18.6
214 18.7
216 21.0
222 24.3
229 34.3
230 21.1
232 24.1
235 23.6
240 17.3
243 21.2
246 18.8
247 34.6
250 20.1
252 26.5
260 26.9
264 16.6
268 42.6
274 43.1
278 25.6
278 21.4
285 34.2
292 34.2
292 34.2
5.64
6.49
6.52
5.61
7.18
6.50
6.65
6.65
4.49
b
24a
28
11.8
35a
5.02
5.02
5.40
6.60
7.89
7.89
6.14
7.32
5.20
3.20
7.77
7.34
7.41
5.80
6.93
6.53
7.54
5.28
6.51
7.41
7.25
5.30
6.45
5.70
9.50
5.59
8.01
4.57
6.59
8.91
37
37a
a
19
32
31a
30
29
27
34
32
27
27
27
27
25
25
37
41
26
35
36a
30
29a
33
a
24a
32
30
34
37
51
42
55
13.2
6.17
6.43
4.67
4.63
4.63
b
52
47
37a
29
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 6/7/15 5:45 PM

Transitions in Order and Molecularity with Temperature .. .
1527
Table 1. (Continued).
T
K
P
[M]
[O
2
15
]
υ
z
cm
% Sb
k
i
Ϯσki
3
1
7
Ϫ3
Ϫ3
Ϫ1
Ϫ1 Ϫ1
s
mbar 10 cm 10 cm m s
absorption cm molecule
Ϫ14
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
295 26.0
1.46
1.47
1.47
0.95
0.96
1.75
1.70
1.72
1.30
1.32
1.49
2.01
1.65
1.82
0.92
1.48
1.59
1.26
0.63
0.62
1.36
1.37
1.45
1.45
1.64
1.45
1.47
1.46
0.97
0.97
0.98
0.64
0.64
2.62
1.93
0.79
1.02
1.03
1.13
1.42
1.41
6.19
6.24
7.45
4.44
5.30
8.83
5.40
5.47
4.64
4.82
5.15
5.25
4.83
5.25
6.33
7.48
3.71
3.65
3.51
3.53
3.23
4.79
6.14
6.14
3.84
6.14
8.00
4.39
2.92
4.70
4.14
5.70
5.69
5.34
4.68
4.04
5.23
5.27
3.20
2.64
4.13
33
33
32
46
46
30
30
30
40
34
32
24
34
24
47
32
33
34
85
85
38
38
33
33
32
33
33
33
49
49
49
63
63
20
27
50
39
38
39
39
40
20
20
20
20
20
10
20
20
20
20
20
10
20
10
10
20
20
20
20
20
20
20
20
20
20
10
10
20
20
20
20
20
20
10
10
20
20
20
20
20
20
52b
40
38
(5.37Ϯ0.41)ϫ10
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ13
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ14
Ϫ13
Ϫ14
Ϫ13
Ϫ13
Ϫ13
Ϫ14
Ϫ13
Ϫ13
Ϫ14
Ϫ13
Ϫ13
Ϫ13
Ϫ13
300 26.3
301 26.4
303 17.1
303 17.2
315 31.7
323 31.0
324 31.4
324 23.7
328 24.1
336 27.5
342 37.2
342 30.6
344 33.7
348 17.0
348 27.5
351 29.6
358 23.9
360 11.7
363 11.7
364 25.7
364 25.8
365 27.3
365 27.3
375 31.1
379 27.5
381 28.0
392 28.0
392 18.6
409 18.9
412 19.0
421 12.5
423 12.5
425 51.5
426 38.0
428 15.5
430 20.1
430 20.3
446 22.6
482 29.0
495 29.2
(4.02Ϯ0.31)ϫ10
(4.34Ϯ0.33)ϫ10
(6.41Ϯ0.62)ϫ10
(5.34Ϯ0.53)ϫ10
(8.00Ϯ0.77)ϫ10
(8.16Ϯ0.64)ϫ10
(8.43Ϯ0.64)ϫ10
(9.09Ϯ0.76)ϫ10
(8.96Ϯ0.77)ϫ10
(6.92Ϯ0.59)ϫ10
(7.88Ϯ0.58)ϫ10
(7.90Ϯ0.60)ϫ10
(7.42Ϯ0.54)ϫ10
(7.91Ϯ0.82)ϫ10
(5.08Ϯ0.38)ϫ10
(7.28Ϯ0.53)ϫ10
(7.47Ϯ0.59)ϫ10
(1.00Ϯ0.13)ϫ10
(9.81Ϯ1.27)ϫ10
(7.86Ϯ0.61)ϫ10
(7.71Ϯ0.59)ϫ10
(6.45Ϯ0.53)ϫ10
(6.38Ϯ0.49)ϫ10
(7.64Ϯ0.64)ϫ10
(8.72Ϯ0.89)ϫ10
(8.23Ϯ0.63)ϫ10
(9.47Ϯ0.71)ϫ10
(1.15Ϯ0.12)ϫ10
(7.76Ϯ0.81)ϫ10
(1.09Ϯ0.11)ϫ10
(1.10Ϯ0.14)ϫ10
(1.09Ϯ0.16)ϫ10
(9.45Ϯ0.75)ϫ10
(1.10Ϯ0.08)ϫ10
(1.10Ϯ0.13)ϫ10
(9.82Ϯ0.95)ϫ10
(1.02Ϯ0.09)ϫ10
(1.17Ϯ0.11)ϫ10
(1.43Ϯ0.11)ϫ10
(1.13Ϯ0.09)ϫ10
b
45
47a
27
31a
a
30a
30
52b
40
26
34
43
37a
39
43
40
42
44
42
35
51
53a
21
50
49
49
45a
34
b
38
18
20
47
45
25
25
25
31
27
41
a
The 217.6 nm absorption line was used.
The 231.1 nm line was used. In all other experiments the 206.8 nm line was used.
b
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 6/7/15 5:45 PM

1
528
B. Cosic and A. Fontijn
Fig. 2. Arrhenius plot of the Sb ϩ O
2
rate coefficients: (᭺) measurements taken at
2
06.8 nm; (᭹) measurements taken at 217.6 nm; (᭝) measurements taken at 231.1 nm;
(––––) best fit to the measurements (Eq. (7)).
These results are in agreement with the bimolecular reaction (5) as the
operating reaction above 1150 K, as compared to the room temperature ob-
servations of Husain [27, 28] where only a termolecular reaction could be
observed. The regime where the termolecular and bimolecular reactions
compete, as observed for the Cr and AlO reactions with O , thus falls be-
2
tween these two temperature limits for Sb ϩ O
2
.
Ϫ1
The above activation energy, E corresponds to 76 Ϯ3 kJ mol . It al-
a
lows re-examination of the heat of reaction (5) and the literature value for
Ϫ1
the Sb-O bond energy, D (Sb-O) ϭ 430Ϯ42 kJ mol [25]. Taking E ϩ
0
a
2
σ as the upper limit value for the endothermicity of reaction (5) at 1330 K
Ϫ1
yields a bond enthalpy at that temperature of Ն425 kJ mol . By using the
following values of [H1330ϪH
Ϫ1
0
]: for Sb 27.7 kJ mol
[34], SbO
0
Ϫ1
Ϫ1
4
4
8.0 kJ mol [25], and O 28.5 kJ mol [26], we obtain D (Sb-O) Ն
Ϫ1
17 kJ mol . Taking this as the lower limit and the above literature value
Ϫ1
as the upper limit, yields D (Sb-O) ϭ 444Ϯ28 kJ mol . Similarly, this
value can be converted to a bond enthalpy at 298 K of 448Ϯ28 kJ mol ,
which gives ∆ H° (SbϩO →SbOϩO) ϭ 50Ϯ28 kJ mol .
0
Ϫ1
Ϫ1
r
298
2
4. Discussion
The mechanism of antimony oxidation by O may be compared to those of
2
As and Bi, the neighboring group 15 elements of the periodic table. These
were studied in He bath gas, at room temperature only, by Husain et al.
[39Ϫ41]. For the As reaction, they obtained by absorption and fluorescence
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 6/7/15 5:45 PM

Transitions in Order and Molecularity with Temperature .. .
1529
Ϫ13
Ϫ13
3
Ϫ1 Ϫ1
measurements 5.0ϫ10 and 3.8ϫ10 cm molecule s , respectively.
No definite mechanism could be deduced, because the measurements were
constrained to a pressure of 26 mbar. The Bi reaction which was studied
over a wide pressure range was shown to be termolecular. Abstraction in
Ϫ1
this case would be too endothermic, 161Ϯ13 kJ mol [25, 26], to contrib-
ute at ambient temperature. On the other hand, abstraction for As is only
Ϫ1
1
9Ϯ8 kJ mol endothermic [26, 42]. The present study has shown for Sb
that when it is energetically accessible a bimolecular channel is indeed fa-
vored. Studies of other endothermic, bimolecular metal-atom reactions, such
as Cr [13], Fe, Mn [43], Eu [6], and Sb with O have shown that the value
2
of the activation energies appears to arise predominantly from the thermo-
chemistry. This suggests that even at room temperature the As ϩ O reac-
2
tion would proceed predominantly by abstraction with a pre-exponential on
Ϫ10
3
Ϫ1 Ϫ1
the order of 10 cm molecule s , which is of the same order as for
these other endothermic O abstraction reactions.
2
Transition from predominantly termolecular to predominantly bimolecu-
lar reactions with increasing temperature is not limited to O reactions.
2
However, the possibilities are somewhat limited. Such observations require
thermal stability at high temperatures, both of the molecule combining with
the metallic species (be it by association or insertion) [44] and of the ab-
straction product formed. Moreover, few reactions have been studied over
sufficiently wide temperature ranges. In the studies of the Cr/NO at 298 K
[
10] and Mo/NO from 296Ϫ620 K [45, 46] systems termolecular fall-off
behavior has been established. Shock tube studies in the 1200Ϫ3000 K have
shown abstraction to occur for these reactions [47]. While pressure was not
varied in the high-temperature experiments the positive activation energies
Ϫ1
obtained, 160Ϯ41 and 36Ϯ12 kJ mol , respectively, strongly indicate at
least a major bimolecular component. The Al/CO system has been found
2
to display mainly fall-off behavior at room temperature, with an abstraction
component, while in the 750Ϫ1215 K range abstraction dominates [3, 48,
4
9]. Competition between termolecular and bimolecular reactions has also
been noted in room temperature measurements on the Pt/N O system [50].
2
Acknowledgements
This work was supported under NSF grant CTS-9905265, Dr. Farley Fisher,
Program Director. We thank William Flaherty, Charles Fung, and Christine
Brown for their assistance.
References
1
. K. Luther and J. Troe, in Reactions of Small Transient Species. Kinetics and Ener-
getics (Eds. A. Fontijn and M. A. A. Clyne, Academic Press, London (1983), Chap. 2.
. J. M. C. Plane, in Gas-Phase Metal Reactions, A. Fontijn, Ed., North-Holland,
Amsterdam (1992), Chap. 3.
2
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 6/7/15 5:45 PM

1
530
B. Cosic and A. Fontijn
3. N. L. Garland, in Gas-Phase Metal Reactions, ibid, Chap. 5; N. L. Garland and H.
H. Nelson, Chem. Phys. Lett. 191 (1992) 269.
4
5
. A. Fontijn and P. N. Bajaj, J. Phys. Chem. 100 (1996) 7085.
. M. L. Campbell, J. Chem. Soc. Faraday Trans. 95 (1998) 1687; J. Phys. Chem. 100
(1996) 19430.
6
7
. M. L. Campbell, J. Phys. Chem. A 103 (1999) 7274.
. C. Vinckier, P. Christiaens and M. Hendrickx, in Gas-Phase Metal Reactions, ibid,
Chap. 4.
8
9
. C. Vinckier and J. Helaers, J. Phys. Chem. A 102 (1998) 8333.
. M. Helmer and J. M. C. Plane, J. Chem. Soc. 90 (1994) 395.
1
0. S. A. Mitchell, in Gas-Phase Metal Reactions, ibid, Chap. 12; J. M. Parnis, S. A.
Mitchell and P. A. Hackett, J. Phys. Chem. 94 (1990) 8152.
1
1
1
1. A. Fontijn and P. M. Futerko, in Gas-Phase Metal Reactions, ibid, Chap. 6.
2. A. Fontijn, Pure and Appl. Chem. 70 (1998) 469.
3. A. S. Narayan, A. G. Slavejkov and A. Fontijn, 24th Symp. (Intl.) on Combustion,
The Combustion Institute, Pittsburgh (1992), p. 727; A. S. Narayan, Ph.D. Thesis,
Rensselaer Polytechnic Institute (1992).
1
1
1
4. D. P. Belyung and A. Fontijn, J. Phys. Chem. 99 (1995), 12225.
5. J. Troe, J. C. S. Faraday Trans. 90 (1994) 2303.
6. D. P. Belyung, G. T. Dalakos, J.-D. R. Rocha and A. Fontijn, 27th Symp (Intl.) on
Combustion, The Combustion Institute, Pittsburgh, 1998, p. 227.
7. C. T. Stanton, N. L. Garland and H. H. Nelson, J. Phys. Chem. 95 (1991) 8741.
8. D. M. Golden, J. Phys. Chem. 83 (1979) 108.
1
1
1
2
2
2
9. P. Saxe, as quoted in Ref. 17.
0. L. Andrews, T. R. Burkholder and J. T. Yustein, J. Phys. Chem. 96 (1992) 10182.
1. G. V. Chertihin, W. D. Bare and L. Andrews, J. Chem. Phys. 107 (1997) 2798.
2. D. A. Tillman, Trace Metals in Combustion Systems Academic Press, San Diego
(
1994), Chap. 1.
2
3. F. Paoletti, P. Sirini, H. Seifert and J. Vehlow, Abstracts of The Sixth International
Congress on Toxic Combustion Byproducts, University of Karlsruhe (1999), p. 50.
4. R. J. Lewis, Sr., Hazardous Chemicals Desk Reference, Van Nostrand Reinhold, New
York (1991).
2
2
2
5. J. B. Pedley and E. M. Marshall, J. Phys. Chem. Ref. Data 12 (1983) 967.
6. M. W. Chase, Jr., NIST-JANAF Thermochemical Tables, J. Phys. Chem. Ref. Data
(1998) Monograph 9.
2
2
2
7. D. Husain and N. K. H. Slater, J. Photochem. 7 (1977) 59.
8. D. Husain, L Krause and N. K. H. Slater, J. Chem. Soc. Faraday II 73 (1977) 1706.
9. A. Fontijn and W. Felder, in Reactive Intermediates in the Gas Phase. Generation
and Monitoring, D. E. Setser, Ed., Academic Press, New York (1979), Chap. 2.
0. A. G. Slavejkov, P. M. Futerko and A. Fontijn, 23rd Symp. (Intl.) on Combustion,
The Combustion Institute, Pittsburgh (1990), p. 155.
3
3
3
3
3
1. A. Fontijn and W. Felder, J. Phys. Chem. 83 (1979) 24.
2. J. A. Irvin and T. I. Quickenden, J. Chem. Educ. 60 (1983) 711.
3. R. E. Honig and D. A. Kramer, RCA Review 30 (1969) 285.
4. R. Hultgren, P. D. Desai, D. T. Hawkins, M. Gleiser, K. K. Kelley and D. D. Wagman,
Selected Values of Thermodynamic Properties of the Elements, American Society for
Metals, Metals Park, Ohio (1973).
3
5. W. H. Press, B. P. Flannery, S. A. Teokolsky and W. T. Vetterling, Numerical Recipes,
Cambridge University, Cambridge (1986) Chap. 14.
3
3
6. W. E. Wentworth, J. Chem. Educ. 42 (1965) 96, 162.
7. This is the Reynolds number for flow to fully develop in a tube, where a is tube
radius and ν bath gas viscosity, see M. C. Potter and J. F. Foss, Fluid Mechanics,
Great Lakes Press, Okemos, MI (1982), p. 274.
3
8. A. Fontijn and W. Felder, J. Phys. Chem. 83 (1979) 24.
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 6/7/15 5:45 PM

Transitions in Order and Molecularity with Temperature .. .
1531
3
4
4
4
4
9. D. Husain and P. E. Norris, J. Chem. Soc. Faraday Trans. II, 73 (1977) 1815.
0. D. Husain and N. K. H. Slater, J. Chem. Soc. Faraday Trans. II, 74 (1978) 1222.
1. C. F. Bell and D. Husain, J. Photochem. 24 (1984) 223.
2. K. H. Lau, R. D. Brittain and D. L. Hildenbrand, Chem. Phys. Lett. 81 (1981) 227.
3. U. S. Akhmadov, I. S. Zaslonko and V. N. Smirnov, Kinetics and Catalysis 29 (1988)
251.
4
4. J. Hranisavljevic and A. Fontijn, J. Phys. Chem. A 101 (1997) 2323; J. Hranisavl-
jevic, Ph.D. Thesis, Rensselaer Polytechnic Institute (1997) and references quoted
therein.
4
4
5. T. Wakabayashi, Y. Nakai and Y. Ishikawa, Chem. Lett. (1997) 331.
6. R. E. McClean, M. L. Campbell and R. H. Goodwin, J. Phys. Chem. 100 (1996)
7
502.
7. U. S. Akhmadov, I. S. Zaslonko and V. N. Smirnov, Kinetics and Catalysis 29 (1988)
08.
4
8
48. J. M. Parnis, S. A. Mitchell and P. A. Hackett, Chem. Phys. Lett. 151 (1988) 485.
49. N. L. Garland, C. H. Douglass and H. H. Nelson, J. Phys. Chem. 96 (1992) 8390.
50. M. L. Campbell, J. Chem. Soc. Faraday Trans. 94 (1998) 353.
Brought to you by | New York University Bobst Library Technical Services
Authenticated
Download Date | 6/7/15 5:45 PM