Kinetics of ground-state Cd reactions with Cl2, O2, and HCl over wide temperature ranges

Article abstract of DOI:10.1021/jp963074z

Reactions of ground-state Cd(¹S) have been investigated in the 10-80 mbar pressure range using the high-temperature fast-flow reactor (HTFFR) technique. The Cd + Cl₂ → CdCl + Cl reaction yielded k(466-875 K) = 9.0 × 10^-10 exp (-5959K/T) cm³ molecule^-1 s^-1, with a 2σ precision of about ±15% and corresponding accuracy of ±27% For Cd + O₂ and Cd + HCl reactions the upper limits are k(300 K) < 1 × 10¹⁵ cm³ molecule^-1 s^-1 or < 5 × 10^-34 cm⁶ molecule^-2 s^-1. Orbital symmetry rules are used to deduce the reaction products and, along with electron configuration arguments, help explain the slowness of the latter two reactions. The importance of these results for the speciation of Cd, Hg, and Zn in waste incineration is discussed, and the consequences of the present findings for transition metal reactions are mentioned.

Full text of DOI:10.1021/jp963074z

J. Phys. Chem. A 1997, 101, 2323-2326
2323
Kinetics of Ground-State Cd Reactions with Cl2, O2, and HCl over Wide Temperature
Ranges
Jasmina Hranisavljevic and Arthur Fontijn*
High-Temperature Reaction Kinetics Laboratory, The Isermann Department of Chemical Engineering,
Rensselaer Polytechnic Institute, Troy, New York 12180-3590
ReceiVed: October 7, 1996; In Final Form: January 6, 1997X
Reactions of ground-state Cd(1S) have been investigated in the 10-80 mbar pressure range using the high-
temperature fast-flow reactor (HTFFR) technique. The Cd + Cl2 f CdCl + Cl reaction yielded k(466-875
K) ) 9.0 × 10-10 exp (-5959K/T) cm3 molecule-1 s-1, with a 2σ precision of about (15% and corresponding
accuracy of (27%. For Cd + O2 and Cd + HCl reactions the upper limits are k(300 K) < 1 × 10-15 cm3
molecule-1 s-1, or < 5 × 10-34 cm6 molecule-2 s-1. Orbital symmetry rules are used to deduce the reaction
products and, along with electron configuration arguments, help explain the slowness of the latter two reactions.
The importance of these results for the speciation of Cd, Hg, and Zn in waste incineration is discussed, and
the consequences of the present findings for transition metal reactions are mentioned.
1. Introduction
2. Experimental Section
Reactions of excited Cd(3P) and Cd(1P) atoms have been
extensively studied,1 but there is only one reported investigation
of Cd(1S) reactions, i.e., with O2, NO2, and N2O.2 Information
on Cd(1S) reactions is of inherent interest and is needed for
modeling of waste incineration because of the presence of this
highly toxic metal in many combustible wastes in environmen-
tally significant concentrations.3,4 Metals in incineration parti-
tion among residual ash, gases, and airborne submicron particles,
such as fly ash and aerosols. These metal-enriched combustion
emissions are particularly difficult to clean up and are considered
to be among the most hazardous forms of airborne pollution.3,4
Recent thermodynamic equilibrium calculations for simulated
incineration conditions have shown that Cd can start volatilizing
at ∼600 K in the form of CdCl2, but above 1300 K it is mostly
present as atomic Cd; the formation of CdO is not favored, even
under oxidizing conditions.5 However, equilibrium calculations
can be misleading, since on a given time scale the actual
concentrations of metallic species will be kinetically con-
trolled.4,6 Therefore, kinetic data on reactions leading to
cadmium chlorides and oxides are needed. Furthermore, the
Cd + Cl2 reaction is of interest, since because of its endother-
micity (34 kJ mol-1),7,8 it could show evidence for competition
between abstraction and addition channels. Such behavior has
been observed for the Cr + O2 and AlO + O2 reactions.9,10
This endothermicity is in contrast to most metal atom reactions
with Cl2, which tend to be strongly exothermic.
The HTFFR technique and the present data analysis proce-
dures have been discussed in detail in earlier publications.11,12
The reactor used consists of an industrial grade quartz reaction
tube surrounded by SiC resistance heating elements, insulation,
and a water-cooled steel vacuum housing. Gaseous Cd atoms
were produced from Cd powder or granules, placed in a
resistively heated alumina crucible. The vapor was entrained
in a stream of Ar bath gas. Mixtures of oxidant, Cl2, O2, or
HCl, in Ar were introduced downstream from the evaporation
source through a movable quartz inlet. For a given rate
coefficient measurement, the total flow rate of these mixtures
was kept constant while the oxidant flow was varied.
A
retractable Pt-Pt/13% Rh thermocouple was used to measure
temperature in the reaction zone. Pressure was measured
downstream from the reaction zone using an MKS Baratron
pressure transducer. Gas flow rates were determined by
calibrated Hastings-Teledyne mass-flow controllers. Relative
Cd concentrations were monitored by atomic resonance absorp-
tion spectrometry (ARAS) using the Cd 5s2(1S0) to 6s(1P0)
resonance transition at 228.8 nm. The resonance radiation was
supplied by a Buck Scientific hollow cathode lamp, and the
signals were detected by a Thorn RFI/B-213F photomultiplier
tube equipped with a 228.5 nm (9.1 nm fwhm) filter and
transferred to a Keithley 614 electrometer.
Rate coefficient measurements were performed in the station-
ary inlet mode13 with reaction zone lengths of 10 or 20 cm.
The gas concentrations were chosen such that [Cd] , [oxidant]
, [Ar], providing for pseudo-first-order conditions in the
reactor. Individual rate coefficients, ki, were obtained from the
slope of linear plots of ln[Cd]relative vs [oxidant]. A weighted
In the present work the kinetics of the
Cd(1S) + Cl2 f products
Cd(1S) + O2 f products
(1)
(2)
linear regression was used to determine ki and σk .14 Typically,
i
five oxidant concentrations, providing variation by a factor of
and
5, were used.
The gases used were Ar from the liquid (99.998%) from
Praxair, Cl2 (99.9%), O2 (99.99%), and HCl (99.99%) all from
Matheson, 3.00% Cl2 (99.9%) in Ar (99.995%) from Linde, HCl
(99.995%) from Spectra. Cd powder (99.5%, 325 mesh) and
granules (99%, 5-20 mesh) were obtained from Aldrich. All
gases, except for Ar, flowed through Drierite (CaSO4) drying
towers.
Cd(1S) + HCl f products
(3)
reactions are investigated using the high-temperature fast-flow
reactor (HTFFR) technique.
X Abstract published in AdVance ACS Abstracts, February 1, 1997.
S1089-5639(96)03074-5 CCC: $14.00 © 1997 American Chemical Society

2324 J. Phys. Chem. A, Vol. 101, No. 12, 1997
Hranisavljevic and Fontijn
a
TABLE 1: Summary of Rate Coefficient Measurements of Cd + Cl2
P
[M]
[Cl2]max
initial Cd
absorption (%)
Vj
reaction zone
length (cm)
T
(K)
ki ( σk
i
(mbar)
(1017 cm-3
)
(1014 cm-3
)
(m s-1
)
(cm3 molecule-1 s-1
)
12.4
13.4
12.6
41.2
41.2
32.4
32.4
27.4
28.0
40.1
14.8
8.1
79.6
79.4
73.3
73.1
16.2
10.1
10.1
56.6
11.2
12.0
18.5
18.5
15.0
34.0
59.3
49.9
57.2
57.1
24.9
25.2
1.41
1.41
1.43
4.85
4.89
3.76
3.70
3.76
3.93
5.98
2.03
1.09
11.8
11.9
11.1
11.4
1.92
1.28
1.29
7.43
0.927
0.993
1.67
1.65
1.34
2.90
6.18
5.51
6.05
5.97
2.54
2.50
38.0
78.0
96.7
35.7
53.1
68.6
99.9
82.1
167
170
169
76.0
282
284
18
9
49
50
49
14
14
15
15
18
17
17
17
32
9
9
6
6
37
55
55
9
73
70
61
62
76
24
11
22
17
17
40
41
20
10
20
20
10
20
10
20
10
20
20
20
20
20
20
10
20
20
10
10
20
10
20
10
20
20
20
20
20
10
20
10
637
687
634
615
610
625
635
527
516
486
530
534
488
484
479
466
610
571
568
552
875
875
804
810
806
849
695
656
685
692
712
729
(1.03 ( 0.14) × 10-13
(2.50 ( 0.32) × 10-13
(6.93 ( 0.85) × 10-14
(6.22 ( 0.52) × 10-14
(6.06 ( 0.49) × 10-14
(5.88 ( 0.69) × 10-14
(6.55 ( 0.56) × 10-14
(1.07 ( 0.09) × 10-14
(8.57 ( 0.80) × 10-15
(4.53 ( 0.34) × 10-15
(7.76 ( 0.87) × 10-15
(9.28 ( 1.64) × 10-15
(4.23 ( 0.29) × 10-15
(4.91 ( 0.39) × 10-15
(2.87 ( 0.18) × 10-15
(2.43 ( 0.25) × 10-15
(5.04 ( 0.58) × 10-14
(3.01 ( 0.43) × 10-14
(2.29 ( 0.33) × 10-14
(2.69 ( 0.18) × 10-14
(1.12 ( 0.15) × 10-12
(7.51 ( 1.06) × 10-13
(5.63 ( 0.59) × 10-13
(5.08 ( 0.49) × 10-13
(7.11 ( 0.78) × 10-13
(7.30 ( 0.57) × 10-13
(1.09 ( 0.09) × 10-13
(1.71 ( 0.11) × 10-13
(1.41 ( 0.09) × 10-13
(1.43 ( 0.10) × 10-13
(1.67 ( 0.17) × 10-13
(1.41 ( 0.16) × 10-13
35
19
24
14
35
18
17
20
40
41
19
21
13
13
44
33
42
18
48
49
38
30
25
49
48
34
29
49
50
48
396
406
130b
86.4b
111b
132b
20.2
54.3
16.4
38.9
31.9
5.02b
14.6b
5.50b
11.0b
10.9b
60.5
159
a The measurements are reported in the sequence in which they were obtained. b Used 3.00% Cl2 from Linde; otherwise, used pure Cl2 from
Matheson.
3. Results
K, and (13% at 835 K. Allowing (10% for the uncertainty
associated with the flow profile factor13 and (20% for possible
systematic errors, we obtain 2σ accuracy limits varying from
(28% at 466 K to (26% at 835 K.
3.1. Cd(1S) + Cl2. Table 1 summarizes the individual rate
coefficients measured and the conditions under which they were
obtained. Below 466 K rate coefficients for reaction 1 were
too low to be measured in our apparatus, and above 875 K
excessive evaporation of Cd occurred, even with the use of
granulated Cd and when the crucible was covered except for a
ca. 1 mm diameter hole. This excessive evaporation led to initial
Cd absorption higher than the recommended 50%13 and resulted
in nonlinear ln[Cd]relative vs [Cl2] plots, which were discarded.
The reaction conditions varied include pressure P from 8 to 80
mbar corresponding to total gas concentrations [M] from 9.3
× 1016 to 1.2 × 1018 molecule cm-3, maximum oxidant
3.2. Cd(1S) + O2. For reaction system 2 no measurable
reaction was observed over the temperature range 300-1000
K and an [O2]max range of 9 × 1015 to 1 × 1017 molecule cm-3
at similar conditions as employed for reaction 1, from which
we conclude k2 < 1 × 10-15 cm3 molecule-1 s-1 or k2 < 5 ×
10-34 cm6 molecule-2 s-1. The only previous study of this
reaction was made in a diffusion flame apparatus,2 which
suggested that a second-order reaction occurred with k(800-
960 K) ) 8.5 × 10-10 exp(-8250K/T) cm3 molecule-1 s-1
,
corresponding to an activation energy of 69 kJ mol-1. This
result is in conflict with the data presented here and is
inconsistent with the endothermicity for abstraction of 258 (
concentration [Cl2]max from 5 × 1014 to 4 × 1016 molecule cm-3
,
initial Cd absorption, which is a measure of initial [Cd] from 9
to 50%, average flow velocity Vj from 6 to 76 m s-1, and the
observed reaction zone length, which was either 10 or 20 cm.
To ascertain that individual rate coefficient measurements, ki,
were independent of the conditions used, plots of [k(T) - ki]/
k(T) vs P, [M], [Cl2]max, initial Cd absorption, Vj, and reaction
zone length were examined. Here k(T) represents the rate
coefficient at each individual temperature calculated from eq 4
below. The data are well fitted by a linear regression fit15 based
84 kJ mol-1 7,17
The use of diffusion flames for rate coefficient
.
measurements has previously been shown to be often inac-
curate.18
3.3. Cd(1S) + HCl. For this system no measurable reaction
was observed over the temperature range 300-1000 K and an
[HCl]max range 6 × 1015 to 9 × 1016 molecule cm-3 at operating
conditions similar to those for the other two reactions. Based
on this, the upper limit for reaction 3 is k3 < 1 × 10-15 cm3
on ki and σk to the form k(T) ) A exp(-E/T):
i
molecule-1 s-1 or k3 < 5 × 10-34 cm6 molecule-2 s-1
.
k1(466-875 K) )
4. Discussion
9.08 × 10-10 exp(-5959 K/T) cm3 molecule-1 s-1 (4)
These studies have revealed that several of the potential
reactions occur, if at all, at immeasurably slow rates. To explain
these observations, we invoke thermochemical, molecular
orbital, and symmetry arguments. This discussion addresses
2
with associated variances and covariances16 σΑ ) 4.64 ×
2
2
10-2A2, σΑΕ ) 2.83 × 101A, and σΕ ) 1.79 × 104. The
resulting 2σk precision limits are (18% at 466 K, (9% at 670

Cd Reactions
J. Phys. Chem. A, Vol. 101, No. 12, 1997 2325
Figure 1. Summary of the Cd(1S) + Cl2 f CdCl + Cl rate coefficent
measurements: (O) individual rate coefficient measurements; (-) fit
of the measurements (eq 4).
Figure 2. Analysis of the Cd(1S) + Cl2 reaction system: (a) collision
complex with C2V symmetry; (b) molecular orbital representation of
the insertion of Cd(1S) into the Cl2 bond; (c) molecular orbital
representation of the abstraction reaction.
reaction systems 1, 2, and 3 separately. Additionally, predictions
are made for some Cd(3P) reactions based on symmetry
arguments.
Cd + O2 f OCdO
Cd + O2 f CdOO
(9)
4.1. Cd(1S) + Cl2. The preexponential of reaction 1 is
comparable to those of other metal atom Cl2 reactions.19-25 The
following three paths can be envisaged:
(10)
Abstraction (reaction 8) would be 258 kJ mol-1 endothermic,
which makes it too slow for measurement at the temperatures
of interest. The absence of insertion (reaction 9) follows again
from symmetry considerations. In O2 the interacting orbitals
are partially occupied degenerate πg orbitals. These transform
under C2V symmetry as a2 + b2. Hence, their interaction with
the a1 ground-state Cd 5s orbital under C2V symmetry would
again be forbidden.
Cd + Cl2 f CdCl + Cl
Cd + Cl2 f CdClCl
Cd + Cl2 f ClCdCl
(5)
(6)
(7)
End-on collision would be free of any symmetry restrictions
and could result in diradical O2 combining with Cd(1S) to form
an association complex (reaction 10). As discussed by Brown
et al.,27 metals with the ns1 electronic configuration react rapidly
with O2 via an association mechanism, and those with the ns2
configuration react slowly. They pointed out that association
of transition metal atoms with O2 proceeds by the end-on
interaction of the metal s orbital with one of the half-filled O2
πg orbitals to form two new orbitals, one bonding and one
antibonding. For reactions of metals with the ns1 electronic
configuration, one electron from the O2 πg orbital and a single
s electron from the metal fill the bonding orbital, resulting in a
stable configuration. For reactions involving metals with the
ns2 configuration, such as Cd, there is, in addition to the
electrons in the bonding orbital, one electron in the antibonding
orbital, which destabilizes the interaction. Hence, association
also should not be expected to proceed at measurable rates, in
agreement with the experimental results that show an absence
of any reaction in the Cd(1S)-O2 reaction system.
The reaction was found to be pressure independent. Thus, there
is no experimental evidence for association or insertion reactions
6 and 7. The unlikelihood of any such form of an addition
process is in accord with the following theoretical consider-
ations.
Cl2, as the more electronegative reactant, would attract an
electron pair from Cd(1S) into its unoccupied σ* orbital. This
orbital is strongly antibonding, and its interaction with the Cd
atom would invariably result in bond breaking,26 i.e., in
abstraction (reaction 5) or insertion (reaction 7) but not in
association.
In order for Cd(1S) to insert into Cl2, collision has to happen
in a direction nearly perpendicular to the Cl-Cl bond. In the
limiting case the collision will occur under C2V symmetry, as
shown in Figure 2a. For Cd and Cl2, the interacting orbitals
are the doubly occupied 5s and the unoccupied σ* orbital.
Those transform under C2V symmetry as a1 and b2, respectively.
Symmetry rules predict large activation barriers for the interac-
tion of species that have different symmetries, i.e., they are
forbidden.26 In the language of molecular orbital theory, the
net overlap between the Cd 5s orbital and the σ* orbital of Cl2
would be zero, as shown in Figure 2b, where the “+” and “-”
signs refer to the parity of the wave function. The orbital
overlap yields two identical lobes of opposite sign that cancel
each other out. Hence insertion also should not occur.
The end-on collision of Cd with Cl2, resulting in abstraction,
would produce nonzero overlap (Figure 2c), and it is therefore
allowed. Thus, reaction 5 should be the operative reaction.
4.2. Cd(1S) + O2. The possible channels here are
4.3. Cd(1S) + HCl. The paths here could be
Cd + HCl f CdCl + H
Cd + HCl f CdClH
(11)
(12)
Cd + HCl f HCdCl
(13)
Abstraction (reaction 11) is endothermic by 224 kJ mol-1 7,8
and therefore does not need to be taken into consideration in
this work. Association (reaction 12) can be ruled out on the
basis of similar reasoning as in the Cl2 case, i.e., HCl would
react with its strongly antibonding σ* orbital, and this would
Cd + O2 f CdO + O
(8)

2326 J. Phys. Chem. A, Vol. 101, No. 12, 1997
Hranisavljevic and Fontijn
Hardwidge, E.; Tsunashima, S; Greig, G.; Strausz, O. P. J. Am. Chem. Soc.
1974, 96, 1946. McAlduff, E. J.; Yuan, Y. H. J. Photochem. 1976, 5, 297.
Breckenridge, W. H.; Konar, R. S.; Darwent, B. deB. J. Chem. Soc. Faraday
Trans. 1978, 74, 1545. Renlund, A. M. J. Phys. Chem. 1979, 83, 303.
Takahashi, O.; Sotowa, C.; Saito, K.; Ahmed, O.; Yamamoto, S. J. Chem.
Soc., Faraday Trans. 1995, 91, 3795. Greene, T. M.; Andrews, L.; Downs,
A. J. J. Am. Chem. Soc. 1995, 117, 8180.
(2) Kashireninov, O. E.; Manelis, G. B.; Repka, L. F. Russ. J. Phys.
Chem. 1982, 56, 630.
(3) Linak, W. P.; Wendt, J. O. L. Prog. Energy Combust. Sci. 1993,
19, 145.
(4) Seeker, W. M. 23rd Symposium (International) on Combustion;
The Combustion Institute: Pittsburgh, PA, 1990; p 867.
(5) Verhulst, O.; Buekens, A.; Spencer, P. J.; Eriksson, G. EnViron.
Sci. Technol. 1996, 30, 50.
(6) Fontijn, A.; Blue, A. S.; Narayan, A. S.; Bajaj, P. N. Combust.
Sci. Technol. 1994, 101, 59.
(7) Chase, M. W., Jr.; Davies, J. R.; Downey, J. R., Jr.; Frurip, D. J.;
McDonald, R. A.; Syverud, A. N. Janaf Thermochemical Tables, 3rd ed. J.
Phys. Chem. Ref. Data, Suppl. 1 1985, 14.
(8) Huber, K. P.; Herzberg, G. Molecular Spectra and Molecular
Structure. IV. Constants of Diatomic Molecules; Van Nostrand Reinhold:
New York, 1979.
invariably lead to bond breaking. However, Cd insertion into
the H-Cl bond (reaction 13) should be free of restrictions, since
the collision complex formed would not possess any symmetry
elements.28 Such a process has been observed in the reaction
of ground-state Cu with HCl, where in addition to CuCl, an
HCuCl insertion product was identified.29 However, Cu is an
open shell atom. The closed shell structure of Cd apparently
creates a sufficiently large barrier to make reaction 13 too slow
to be observable under the conditions investigated.
4.4. Cd(3P) Reactions. Although insertion can thus be ruled
out for Cd(1S) + Cl2 and Cd(1S) + O2, it has been observed
for Cd(3P) + H2.30 This reaction is allowed, since Cd(3P) has
a 5py orbital of b2 symmetry available for bonding with the H2
σ* orbital of b2 symmetry. Similar considerations would suggest
that further Cd(3P) insertion reactions can occur. Hence, for
example, the symmetry of Cd(3P) + Cl2 insertion is the same
as that of H2 insertion. The interacting orbitals of these reactants
have a finite net orbital overlap. Similarly, Cd(3P) could insert
into the O2 bond, since interaction with the πyz orbital is allowed.
Note, however, that interaction with the πxy orbital is forbidden.
The use of symmetry rules can also explain other observations
in the literature. Thus, it has been reported that ground-state
Cu(2S) does not insert into O2, but Cu(2P) does.31 This situation
is analogous to the Cd case, since the interacting orbitals of
Cu(2S) and Cu(2P) are again of, respectively, s and p character.
(9) Narayan, A. S.; Slavejkov, A. G.; Fontijn, A. 24th Symposium
(International) on Combustion; The Combustion Institute: Pittsburgh, PA,
1992; p 727.
(10) Belyung, D. P.; Fontijn, A. J. Phys. Chem. 1995, 99, 12225.
(11) Fontijn, A.; Futerko, P. M. In Gas-Phase Metal Reactions; Fontijn,
A, Ed.; North Holland: Amsterdam, 1992; Chapter 6.
(12) Slavejkov, A. G.; Futerko, P. M.; Fontijn, A. 23rd Symposium
(International) on Combustion; The Combustion Institute: Pittsburgh, PA,
1990; p 155.
(13) Fontijn, A.; Felder, W. In ReactiVe Intermediates in the Gas Phase;
Setser, D. W., Ed.; Academic Press: New York, 1979; Chapter 2.
(14) Irvin, J. A.; Quickenden, T. I. J. Chem. Educ. 1983, 60, 711.
(15) Press, W. H.; Flannery, B. P.; Teukolsky, S. A.; Vetterling, W. T.
Numerical Recipes; Cambridge University: Cambridge, 1986; Chapter 14.
(16) Wentworth, W. E. J. Chem. Educ. 1965, 42, 96, 162.
(17) Lide, D. R., Ed. CRC Handbook of Chemistry and Physics, 73rd
ed.; CRC Press: Boca Raton, FL, 1992.
(18) Nien, C.-F.; Rajasekhar, B.; Plane, J. M. C. J. Phys. Chem. 1993,
97, 6449.
(19) Talcott, C. L.; Ager, J. W., III; Howard, C. J. J. Chem. Phys. 1986,
84, 6161.
(20) Maya, J.; Davidovits, P. J. Chem. Phys. 1974, 61, 1082.
(21) Silver, J. A. J. Chem. Phys. 1986, 84, 4718.
(22) Vinckier, C.; Christiaensen, P. J. Phys. Chem. 1992, 96, 8423.
(23) Rogowski, D. F.; Marshall, P.; Fontijn, A. J. Phys. Chem. 1989,
93, 1118.
(24) Vinckier, C.; Verhaeghe, T.; Vanhees, I. J. Chem. Soc., Faraday
Trans. 1996, 92, 1455.
(25) Belyung, D. P.; Zhang, Q.; Hranisavljevic, J.; Fontijn, A. J. Phys.
Chem., to be submitted.
5. Conclusions
Among the reactions investigated here, only Cd + Cl2 f
CdCl + Cl proceeds at experimentally observable rates. Similar
considerations, as advanced above, can be used to predict the
behavior of the reactions of the ground states of the other two
group 12 metals with the same reactants. Thus, Zn + Cl2 f
ZnCl + Cl should likewise occur. Hg + Cl2 f HgCl + Cl
would be highly endothermic32 and would under normal
combustion conditions not proceed at measurable rates. Neither
do we expect a reaction of Zn and Hg with O2 and HCl. These
considerations are important, since both metals are significant
toxics present in wastes. Therefore, we suggest that in incinera-
tors Cd and Zn could form chlorides homogeneously but that
for Hg only heterogeneous routes are open for the present
reactants.
(26) Pearson, R. G. Symmetry Rules for Chemical Reactions; Orbital
Topology and Elementary Processes; John Wiley: New York, 1976; Chapter
1.
(27) Brown, C. A.; Mitchell, S. A.; Hackett, P. A. J. Phys. Chem. 1991,
95, 1062.
The symmetry arguments, advanced above, can also be
applied to transition metal reactions. These are known to
proceed by the interaction of the metal s orbitals, which have a
larger spatial extent than the partially filled d orbitals.33
(28) Shuler, K. E. J. Chem. Phys. 1953, 21, 624.
(29) Belyung, D. P.; Hranisavljevic, J.; Kashireninov, O. E.; Santana,
G. M.; Fontijn, A.; Marshall, P. J. Phys. Chem. 1996, 100, 17835.
(30) Breckenridge, W. H. J. Phys. Chem. 1996, 100, 14840.
(31) Ozin, G. A.; Mitchell, S. A.; Garcia-Prieto, J. J. Am. Chem. Soc.
1983, 105, 6399.
Acknowledgment. This work was supported by National
Science Foundation Grants CTS-9301655 and CTS-9632492.
We thank Dr. H. B. Hollinger and D. P. Belyung for helpful
discussions and B. Banovic for assistance with some of the
experiments.
(32) The heats of these processes are in kJ mol-1, listed after each
reaction:6 Zn + Cl2 f ZnCl + Cl, 14; Zn + O2 f ZnO + O, 227; Zn +
HCl f ZnCl + H, 203; Hg + Cl2 f HgCl + Cl, 139; Hg + O2 f HgO
+ O, 230; Hg + HCl f HgCl + H, 328.
References and Notes
(33) Walch, S. P.; Bauschlicher, C. W., Jr. In Comparison of Ab Initio
Quantum Chemistry with Experiment for Small Molecules; Bartlett, R. J.,
Ed.; Reidel: Dordrecht, 1985; p 17.
(1) Steacie, E. W. R; Leroy, D. J. J. Chem. Phys. 1943, 11, 164.
Tsunashima, S.; Sato, S. Bull. Chem. Soc. Jpn. 1967, 40, 2987. Young, P.;