Professional Documents
Culture Documents
School of Chemical and Biomolecular Engineering, Georgia Institute of Technology, 311 Ferst
Drive N.W., Atlanta, Georgia 30332-0100
Received 8 April 2009; Revised 4 September 2009; Accepted 13 September 2009
DOI 10.1002/jcc.21440
Published online 11 November 2009 in Wiley InterScience (www.interscience.wiley.com).
Abstract: The Hammond-Lefer postulate asserts that transition states of exothermic reactions are reactant-like
(early), whereas transition states of endothermic reactions are product-like (late). Related postulates have been proposed to describe the sensitivity of activation barriers for reactions occurring on catalytic surfaces to the catalyst
structure. To evaluate the validity of these postulates for different chemical reactions, a general method for classifying transition states as either early or late is needed. One can envision a dimensionless reaction coordinate that
changes continuously and monotonically from 0 to 1 along a minimum energy reaction pathway. The value of the
dimensionless reaction coordinate for the transition state (WTS) classies transition states as (a) early when WTS
\ 0.5, (b) late when WTS [ 0.5, and (c) equidistant between reactants and products when WTS 5 0.5. In this
article, we derive such a dimensionless reaction coordinate and illustrate its usefulness for several different chemical reactions.
q 2009 Wiley Periodicals, Inc.
Key words: chemical reactions; reaction coordinate; transition state lateness; Hammond postulate; dimensional
analysis
Introduction
A transition state that resembles the reactant state more than the
product state is said to be early, whereas a transition state that
resembles the product state more than the reactant state is said
to be late. Whether a transition state is early or late can sometimes be inferred from a plot of energy versus reaction coordinate along the minimum energy reaction pathway. This direct
approach may not work, however, if atoms in the reactants or
products are separated by an arbitrarily large (innite) distance
or if changes in reaction coordinate are complicated. The development of a general method for classifying transition states as
either or late is challenging because of the different reaction
coordinate types that exist for different types of chemical reactions. For example, the reaction coordinate for NH3 inversion
is the dihedral angle, but the reaction coordinate for H2
dissociation is the interatomic distance. Here, we use the abbreviation MERP to denote the minimum energy reaction pathway
for an elementary reaction within the Born-Oppenheimer
approximation.1
Several heuristic principles relating properties of classes of
reactions to the lateness of transition states have been advanced.
The best known is the Hammond-Lefer postulate (HLP), which
asserts that for an elementary reaction step between two ground
1529
Theory
Basic Form of the Dimensionless Reaction Coordinate
Consider a dimensionless reaction coordinate that can be computed using only molecular geometries along a MERP. Let W(Y)
denote the value of this dimensionless reaction coordinate for
geometry Y on the MERP, where Y is the set of Cartesian coordinates specifying the atomic positions. W(Y) should be invariant
to rotations and translations of the whole system; therefore, the
dimensionless reaction coordinate should be a function of the set
of relative distances between atoms in the system. The relative
distance Dij between atom i and atom j is dened as
Dij
q
xi xj 2 yi yj 2 zi zj 2 ;
(1)
where (xi, yi, zi) is the Cartesian coordinate of atom i and (xj, yj,
zj) is the Cartesian coordinate of atom j. A reaction coordinate
measure is said to be dimensionless when it does not depend on
the scale used to measure distances. A homogeneous function is
a function f(x) for which f(gx) 5 gkf(x) for any constant g, and
the exponent k is called the homogenous degree.23 A function
that does not depend on the scale used to measure the independent variable x has homogenous degree zero. Because a dimensionless reaction coordinate is unitless, W(Y) cannot depend on
the distance measurement scale and thus must be a homogenous
function of degree zero in the distance measurement scale.
The eld of differential geometry provides some insights on
how to construct distance functions. Typically, differential distance ds on a differential manifold is described by an equation
of the form
ds
rX
X
gab dXa dXb ;
a
(2)
where gab is a metric tensor and {Xa} are the independent coordinates. A distance function of form (2) is not normalized. To
achieve normalization, the scaled distance from R to state Y on
DOI 10.1002/jcc
1530
(3)
(4)
(5)
R YpP
P
a
b gab dva dvb
R
WY R PpP
:
P
a
b gab dva dvb
R
XX
a
(7)
(8)
When |b| 1 the rst two terms in the series dominate and
inserting (9) into (8) yields
s
NN 1
1 :
(10)
k1b
2
s
2 9
:
b2 4
(11)
(6)
R Yq
P P
R Yq
P
2
2
i>j
j dvij
R
a dva
R
WY R q
R q
:
P
P P
2
P P
2
i>j
j dvij
R
a dva
R
(12)
DOI 10.1002/jcc
1531
Integral Approximation
PnY q
P P
2
j2
i>j
j vij j1 vij j
;
WY P qP
P
np
2
j2
i>j
j vij j1 vij j
bond
bond p
cov p
vsingle
CAB Dsingle
CAB Rcov
AB
AB
A RB
(18)
and
half p
vhalf
AB CAB DAB
p
vdw
cov
cov
CAB Rvdw
A RB RA RB =2 : 19
(13)
How can the constant offset distances {Gij} in eq. (3) be determined? We rst consider the case of diatomic molecules and
then show that the offset distances for these examples extend to
all chemical systems. Consider a diatomic molecule that contains
a covalent bond between atoms A and B. Let f be dened as the
ratio of the equilibrium bond length Dbond
AB and the sum of covalent radii:
vAB R vAB
vsingle bond v
ABsingle bond AB :
vAB R vAB P
vAB
0
(20)
The dimensionless reaction coordinate at which approximately half of the single bond is broken divides the early and
late reaction phases:
bond
vsingle
vhalf
AB
AB
W vhalf
:
0:5
AB
single bond
vAB
0
(21)
(14)
CAB
p
cov
p
vdw
cov
Rvdw
A RB 1 2 2RA RB
p
p
2 2 1
(22)
(15)
(16)
We propose that
Dhalf
AB
cov
vdw
vdw
Rcov
A RB RA RB
2
(17)
As the chemical state (charge, oxidation, spin, etc.) of an element can vary during the course of a chemical reaction, while
the {Gij} are constant by denition, a covalent radius independent of the chemical state of the element must be used for the
construction of a dimensionless reaction coordinate. For the rst
96 elements, the recommended covalent radii are those reported
by Cordero et al. based on statistical analysis of more than
228,000 experimental bond lengths from the Cambridge Structural Database.25 For four elements (C, Mn, Fe, and Co), they
reported different covalent radii for separate hybridization or
spin states; for these elements, we recommend averaging their
reported covalent radii over hybridization or spin states to give
.
Rcov equal to 0.73 (C), 1.50 (Mn), 1.42 (Fe), and 1.38 (Co) A
Unfortunately, a set of van der Waals radii has not been
reported that includes most of the elements in the periodic table.
To provide a comprehensive set of van der Waals radii, we
DOI 10.1002/jcc
1532
Rvdw
A RA 0:76 A:
(23)
0:76 A
Cij p
Rcov
Rcov
i
j :
p
2 1
(24)
(25)
DOI 10.1002/jcc
1533
0s1
p 2
q
R1 R1 R0
@
k x2 y2 z2 C
q d
dxdydzAdk
1 1 1
dk
0s1
:
W
p 2
q
R min
R1 R1 R0
2
d
2
2
@
A
k x y z C
dxdydz dk
1 1 1 q dk
1
(26)
1
0v
!
u
1
uR 1 R 1 R 0
p2 kx2 kx2 y2 z2
t
@
p
2p2 dxdydzAdk
1
1 1 1
2
R
W
kx y2 z2 C
1
:
0v
!
u
1
2 kx2 kx2 y2 z2
R min u
R
R
R
p
@t 1 1 0
p
2p2 dxdydzAdk
1 1 1
1
2
(27)
kx y2 z2 C
(28)
0v
1
0
1
u
1
u
2
2
C
Z Z Z
Z Bu
2
2
2
C
Bu 1 1 0 B
C
Bp k x k x y z
C
Bu
Cdk
dxdydz
B
C
q
Bu
C
2p2 A
@
1 @t 1 1 1
A
2
k x y2 z2 C
:
1
0v
0
1
u
1
u
2
2
C
Z Z Z
Z =s Bu
2
2
2
C
C
Bu 1 1 0 B
3
Bp K X K X Y Z
C
CdK:
Bu
s2p
dXdYdZ
B
C
q
2p2
C
Bu
A
1 @t 1 1 1 @
A
2
C
2
2
K X Y Z s
Thus, in the case G 5 0 this reaction and its dimensionless reaction coordinate exhibit self-similar scaling; that
is, the approach of an ion to a distance L from the surface
(29)
(30)
is a magnied copy of the approach of an ion to a distance 5 sL from the surface. Comparing eqs. (27) and
(30) gives
DOI 10.1002/jcc
1534
W
32p
min
(31)
could be as low as 2. Insomuch as the value of the dimensionless reaction coordinate is not unduly sensitive to the value of p,
an intermediate value of p 5 4 should give good results in
nearly all cases.
!1
!1
min
32p
0;
(32)
(33)
q2
for > min :
4pe0
(34)
Comparing eqs. (31) and (34), the slope of E(Y) versus W(Y) is
constant iff. p 5 5/2.
The lower bound p [ 1.5 also applies to other systems containing an innite number of atoms. Consider a system with q
atoms per unit volume. Let r be an independent variable equal
to the distance from some reactive atom A in the system. For
image Y on the MERP, we have
Z
Y
R
s
X X 2 Z
dvij
i>j
s
Z Z Z
2
q d r Cij p dxdydz:
Summary
(35)
Results
Dissociation of Covalent Diatomic Molecules
DOI 10.1002/jcc
1
1
1
1
3
3
3
5
1
3
3
3
DOI 10.1002/jcc
Spin multiplicity of the low energy state. The isolated atoms had the following spin multiplicities: H (2), He(1), Li (2), Na (2), Cu (2), Ag (2), Au (2),
C (3), and O (3).
99
94
84
83
99
98
99
100
83
99
99
99
0.84
0.75
0.84
0.85
0.86
0.86
0.86
0.86
0.85
0.75
0.75
0.75
0.80
0.62
0.80
0.83
0.86
0.85
0.84
0.84
0.81
0.76
0.75
0.75
0.76
0.51
0.76
0.81
0.86
0.85
0.82
0.82
0.78
0.78
0.76
0.76
1
1
1
1
1
1
1
5
1
1
1
1
94
84
61
60
86
89
96
99
61
88
90
90
0.73
0.62
0.72
0.74
0.75
0.75
0.74
0.74
0.73
0.75
0.75
0.75
0.69
0.50
0.69
0.73
0.76
0.76
0.74
0.74
0.71
0.76
0.75
0.75
0.66
0.40
0.66
0.72
0.78
0.77
0.74
0.74
0.69
0.78
0.76
0.76
1
1
1
1
1
1
1
1
1
1
1
1
61
52
25
24
46
49
57
64
25
47
49
48
0.49
0.40
0.49
0.51
0.52
0.52
0.51
0.51
0.50
0.52
0.52
0.52
0.48
0.31
0.48
0.52
0.56
0.55
0.53
0.53
0.50
0.56
0.55
0.55
0.47
0.24
0.47
0.53
0.61
0.59
0.56
0.56
0.50
0.61
0.59
0.58
1
1
1
1
1
1
1
1
1
1
1
1
4.71
9e-4
1.03
0.72
2.03
1.72
2.29
10.92
0.86
1.87
2.47
2.20
1.20
5.38
1.05
0.96
0.84
0.87
0.91
0.82
1.00
0.85
0.87
0.88
H2
He2
Li2
Na2
Cu2
Ag2
Au2
CO
LiNa
CuAg
CuAu
AgAu
0.74
3.01
2.70
3.18
2.21
2.53
2.48
1.14
2.95
2.36
2.33
2.48
DE/Ed
(%)
W
p56
W
p 5 2.5
DE/Ed
(%)
W
p56
W
p 5 2.5
D0
)
(A
Ed
(eV)
(2s11)a
W
p 5 1.5
D0 1 0.76 A
(2s11)a
W
p 5 1.5
W
p 5 2.5
W
p56
DE/Ed
(%)
(2s11)a
W
p 5 1.5
D0 1 2.28 A
D0 1 1.52 A
D0
2s11a
beyond
value when the bond length has been extended 0.76 A
its equilibrium value [see eqs. (19), (21), and (23)].
3. A signicant fraction of the bond energy should be lost and a
signicant fraction of the bond energy should be retained
beyond its equilibwhen the bond length is extended 0.76 A
rium value. (Both the early and late reaction phases should
be accompanied by substantial bond energy changes.).
4. The bond energy is mostly lost when the bond length has
beyond its equilibrium value.
been extended 1.52 A
(According to #2 above this amount of bond extension should
occur late in the reaction. Thus, most of the bond energy
should be lost at this point in the reaction.).
1535
1536
)
[H1
H2
X]{ distances (A
X5
H
F
Cl
Br
I
)
H
X D0 (A
0.74
0.92
1.28
1.41
1.61
[0.77]b
[0.94]
[1.30]
[1.45]
[1.65]
Erxn (ev)
Ea (eV)
H1
H2
H2
X
H1
X
WTSa
0
21.33 [21.04]
0.16 [0.28]
0.62 [0.82]
1.22 [1.33]
0.42
0.08 [0.32]
0.37 [0.59]
0.65 [0.99]
1.21 [1.43]
0.93
0.77 [0.79]
0.99 [1.01]
1.27 [1.26]
1.53 [1.46]
0.93
1.52 [1.41]
1.43 [1.43]
1.46 [1.49]
1.63 [1.67]
1.86
2.06 [2.20]
2.42 [2.44]
2.73 [2.75]
3.16 [3.13]
1
0.05 [0.04]
1.7 [1.8]
11 [12]
40 [35]
and transition state to determine the relative lateness of transition states for the reactions H2 1 X ? H 1 HX, where X 5 F,
Cl, Br, and I.14 In agreement with HLP, they found that the transition state for X 5 F (exothermic) was early, whereas the transition states for X 5 Cl, Br, and I (endothermic) were late. To
perform their analysis, Nalewajski and Broniatowska used the
GVB 11 1 2 geometries reported by Dunning,38 which are
displayed in [] in Table 2. Also displayed in Table 2 are the
energies and geometries we computed at the CCSD(T) level of
theory using all-electron aug-cc-pVTZ basis sets2932 on H, F,
and Cl and aug-cc-pVTZ-PP basis sets33,34 including relativistic
small core pseudopotentials on Br and I. Coefcients for these
basis sets and their diffuse functions were obtained from the
EMSL basis set exchange.35,36 The CCSD(T) geometries were
similar to those previously reported by Dunning using the generalized valence bond method. We computed [H
H
F]{ to be
slightly nonlinear, whereas all of the remaining [H
H
X]{
{
were linear. The bond distances in [H
H
X] allow one to
unambiguously classify the transition states as early or late for
these four reactions. Specically, we can dene a parameter
c
DHH TS DHH R
DH1 X TS DHX P
(36)
DOI 10.1002/jcc
X5
10% smaller
covalent
radii
10% larger
covalent
radii
Rvdw 2 Rcov
5 0.86 A
Rvdw 2 Rcov
5 0.66 A
H
F
Cl
Br
I
0.50
0.39
0.54
0.62
0.68
0.50
0.39
0.57
0.66
0.71
0.50
0.41
0.55
0.62
0.67
0.50
0.37
0.56
0.66
0.72
1537
X
C
C (8)39
Reactant
Product
C
H2C
HFC
C
HC B CH
HC B CF
F2C
C
FC B CF
Migrating
atom (X)
WTSa
rxn type
Agrees
HLP?
TS
react.
H
H
F
F
0.530.55
0.190.21
0.570.59
0.560.57
Exothermic
Exothermic
Exothermic
Exothermic
No
Yes
No
No
54.9
81.5
57.5
58.5
120.4
109.8
133.0
123.0
DOI 10.1002/jcc
1538
geometries given by Li et al.21,22,41 In computing WTS for surface reactions, periodic copies of the surface atoms are generated along the surfaces two lattice directions, whereas periodic
copies of the adsorbate atoms are not used. This corresponds to
the actual periodicity in the real chemical system, but differs
from that used in the DFT slab model where all atoms are periodically replicated along three lattice directions. In general, WTS
should be computed using the periodicity of the real system.
The results are shown in Table 5. The WTS values indicate that
b-H elimination of ethyl over Cu(100), Cu(110), Cu(111), and
Cu(221) has an early transition state. Because the reaction is
endothermic, this disagrees with HLP. Because the span in Ea is
slightly greater for the reverse reaction (0.15 eV) than the forward reaction (0.11 eV), an early transition state is in agreement
with Gellmans structure sensitivity postulate (SSP). As the
structure of the reactant was not varied in this example, Gellmans RSP does not apply.
The structure of the reactant was varied for b-H elimination
of OCH2CH3-xFx over Cu(111), thus providing a test of the RSP.
Because these reactions were endothermic, a late transition state
is predicted by HLP. The WTS values indicate an equidistant
transition state for x 5 0 and a late transition state for x 5 2 or
3. (The geometry for x 5 1 was not available for analysis.)
Because the transition state was not early, the RSP predicts that
Ea for the forward reaction will vary more than Ea for the
reverse reaction as x is varied. This was indeed the case as the
span in computed Ea for the forward reactions was 0.92 eV,
whereas that for the reverse reaction was only 0.07 eV.
A similar set of reactions was studied over Pt(111). In this
case, the reaction was exothermic so HLP predicted an early
transition state. The WTS values also indicated an early transition
state. According to the RSP, the reverse reaction should have a
larger variation in Ea than the forward reaction as x is varied.
The span in computed Ea values for the forward reaction was
0.32 eV, whereas that for the reverse reaction was 0.14 eV.
However, the data point x 5 3 has substantially higher Ea,f and
substantially lower Erxn than x 5 0, 1, and 2. If x 5 3 is
Table 5. Computed WTS Values for b-H Elimination of Ethyl and Fluorinated Ethoxy Species.
Surface
Reactant
Ea,fa (eV)
Erxna (ev)
WTSb
TS
type
Agrees
SSP?
Agrees
HLP?
Agrees
RSP?
Cu(100)
Cu(110)
Cu(111)
Cu(221)
Cu(111)
Cu(111)
Cu(111)
Pt(111)
Pt(111)
Pt(111)
Pt(111)
CH2CH3
CH2CH3
CH2CH3
CH2CH3
OCH2CH3
OCH2CF2H
OCH2CF3
OCH2CH3
OCH2CFH2
OCH2CF2H
OCH2CF3
0.76
0.65
0.69
0.66
1.29 [1.26]c
1.58 [1.52]c
1.75 [1.83]c
0.84
0.80
0.83
1.12
0.12
0.16
0.10
0.13
0.69
1.01
1.10
20.33
20.27
20.20
20.04
0.410.43
0.440.46
0.420.44
0.410.42
0.490.51
0.550.57
0.540.56
0.320.36
0.340.38
0.310.35
0.360.42
Early
Early
Early
Early
Equidistant
Late
Late
Early
Early
Early
Early
Yes
Yes
Yes
Yes
Yes
n.a.
n.a.
Yes
n.a.
n.a.
n.a.
No
No
No
No
No
Yes
Yes
Yes
Yes
Yes
Yes
n.a.
n.a.
n.a.
n.a.
Yes
Yes
Yes
n.a.
n.a.
n.a.
n.a.
DOI 10.1002/jcc
1539
the pathway with late transition state as the structures of the catalyst and nonreactive substituents, respectively, are varied.
Although the SSP has been developed for and applied to catalytic reactions occurring on solid metal surfaces, there does not
appear to be any reason why it could not also be applied to reactions occurring on other kinds of catalysts. Here, we consider
whether SSP correctly describes structure sensitivity for homogeneous catalysis for b-boryl elimination catalyzed by Ru and
Os complexes. DFT-optimized geometries for the reactions
shown in Figure 3 were reported by Lam et al.48 Table 6 shows
the computed WTS values for reactions (a) ? (c) and (d) ? (f).
For both Ru and Os catalysts, the transition state from (a) ? (c)
was early. This reaction violates HLP because it is endothermic
with an early transition state. Because the transition state for this
reaction is early, SSP predicts a larger structure sensitivity for
the reverse reaction. For this reaction, Ea values for the forward
reaction differed by 3.4 (3.4) kcal/mol for the two metals and
differed by 4.7 (4.2) kcal/mol for the reverse reaction. (Free
energies are in parentheses, whereas the electronic energies are
not.) Thus, the computed Ea values agree with the SSP.
DOI 10.1002/jcc
1540
(a)
(b)
(c)
(d)
(e)
(f)
0.0 (0.0)
0.0 (0.0)
15.7 (14.5)
19.1 (17.9)
3.6 (4.0)
11.7 (11.6)
27.4 (24.2)
21.8 (1.4)
12.8 (14.9)
17.8 (20.3)
21.8 (0.2)
22.6 (20.2)
0.400.42
0.440.48
0.600.62
0.440.46
Relative electronic energies and free energies (in parentheses) from Lam et al.48
WTS values over the range 1.5 p 6.
For the reaction of (d) ? (f), the WTS values indicate a late
transition state for Ru and an early transition state for Os. This
result is in agreement with the HLP because the reaction over
Ru is endothermic, whereas that over Os is exothermic. Because
a change in catalyst resulted in a change between early and late
transition state, SSP does not make any prediction in this case
as to whether the forward or reverse reaction has higher structure sensitivity.
Conclusions
The main conclusion of this work is that it is possible to develop
a generally applicable measure of transition state lateness based
on geometry changes along the MERP. Descriptors of transition
state lateness based on the wavefunction, electron density,
atomic charges, or bond orders may not change continuously
along the reaction coordinate, especially if the MERP involves a
change in spin multiplicity. Because molecular geometry
changes continuously along the reaction coordinate for all chemical reactions, a descriptor of transition state lateness that can be
computed directly from molecular geometry changes is desirable. We developed such a descriptor, WTS, that can be computed
directly from a series of images along the MERP. A minimum
of three images is required: the reactant, transition state, and
product geometries. This quantitative measure of transition state
lateness is based on a dimensionless reaction coordinate that
changes continuously and monotonically from 0 for the reactant
to 1 for the product. For early transition states WTS \ 0.5, for
late transition states WTS [ 0.5, and for WTS 5 0.5 the transition state is equidistant between reactants and products. The usefulness of WTS was demonstrated for a series of reactions.
For dissociation of covalent diatomic molecules, the dimensionless reaction coordinate was found to reasonably describe
the transition between early (reactant-like) and late (productlike) geometry. Bond elongation by signicantly less than 0.76
resulted in more than 50% retention of the bond energy on
A
average, and for this displacement range the dimensionless reaction coordinate was less than 0.50 indicating the early reaction
phase. On the other hand, bond elongation by signicantly more
resulted in less than 50% retention of the bond
than 0.76 A
energy on average, and for this displacement range the dimensionless reaction coordinate was more than 0.50 indicating the
resulted in
late reaction phase. Bond elongation by 0.76 A
50% retention of the bond energy on average, and for this dis-
placement the dimensionless reaction coordinate was 0.50 indicating the transition between early and late reaction phases. The
dimensionless reaction coordinate values were found to be relatively insensitive to the value of the exponent p in eq. (3) over
the range 1.5 \ p 6.
Some postulates in chemistry and catalysis are based on the
idea that reactions with early transition states behave differently
than reactions with late transition states. The HLP asserts that
endothermic reactions have a late transition state and exothermic
reactions have an early transition state. The SSP asserts that
activation barriers for surface-catalyzed reactions with late transition states are more sensitive to changes in the surface structure than reactions with early transition states. The RSP asserts
that activation barriers for catalytic reactions are more sensitive
to the structure of nonreactive groups on the reactant if the transition state is late compared to early. A reliable method for classifying transition states as either early or late is needed to use or
evaluate the validity of these postulates.
The reaction H2 1 X ? H 1 HX provided an interesting
test case because with X 5 H the reaction is thermoneutral,
with X 5 F the reaction is exothermic, and with X 5 Cl, Br, or
I the reaction is endothermic. As this reaction is known to follow the HLP,14 a switch from early to equidistant to late transition state occurs when X is varied, and WTS captured this trend.
The rearrangement of vinylidene to acetylene has been
reported to violate HLP,39,40 and WTS conrmed this. We also
showed that F migration in HFC
C and F2C
C violates HLP,
whereas H migration in HFC
C follows HLP. Our ndings are
conrmed by a comparison of angles in the reactant, transition
state, and product geometries.
In some cases, SSP and HLP are in conict. Specically, b-H
elimination of ethyl on different Cu surfaces has been reported as
slightly endothermic and with slightly less structure sensitivity for
the forward reaction than for the reverse reaction.22 Accordingly,
HLP predicts a late transition state, but SSP is valid if the transition state is early but not late. The computed WTS values for this
reaction indicated an early TS, which supports the SSP. A similar
situation was encountered with b-boryl elimination catalyzed by
Ru and Os organometallic complexes. The reaction of (a) ? (c)
in Figure 3 was slightly endothermic and had slightly less structure sensitivity for the forward reaction than for the reverse reaction. The computed WTS values for this reaction indicated an early
TS, which supports the SSP but not the HLP. This suggests that
the SSP may be applied to other types of catalytic reactions
besides those occurring on solid metal surfaces.
DOI 10.1002/jcc
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
1541
25. Cordero, B.; Gomez, V.; Platero-Prats, A. E.; Reves, M.; Echeverria,
J.; Cremades, E.; Barragan, F.; Alvarez, S. Dalton Trans 2008, 2832.
26. Bondi, A. J Phys Chem 1964, 68, 441.
27. Smit, B.; Frenkel, D. Understanding Molecular Simulation; Academic Press: London, 2001.
28. Gaussian 03, Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria,
G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, Jr., J. A.;
Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S.
S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.;
Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.;
Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.;
Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.;
Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.;
Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi,
R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.;
Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D.
K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.;
Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.;
Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.;
Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara,
A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong,
M. W.; Gonzalez, C.; Pople, J. A.; Gaussian, Inc., Wallingford CT,
2004.
29. Dunning, T. H. J Chem Phys 1989, 90, 1007.
30. Kendall, R. A.; Dunning, T. H.; Harrison, R. J. J Chem Phys 1992,
96, 6796.
31. Woon, D. E.; Dunning, T. H. J Chem Phys 1993, 98, 1358.
32. Woon, D. E.; Dunning, T. H. J Chem Phys 1994, 100, 2975.
33. Peterson, K. A.; Figgen, D.; Goll, E.; Stoll, H.; Dolg, M. J Chem
Phys 2003, 119, 11113.
34. Peterson, K. A.; Puzzarini, C. Theor Chem Acc 2005, 114, 283.
35. Feller, D. J Comput Chem 1996, 17, 1571.
36. Schuchardt, K. L.; Didier, B. T.; Elsethagen, T.; Sun, L. S.; Gurumoorthi, V.; Chase, J.; Li, J.; Windus, T. L. J Chem Inf Model
2007, 47, 1045.
37. Feller, D.; Sordo, J. A. J Chem Phys 2000, 113, 485.
38. Dunning, T. H. J Phys Chem 1984, 88, 2469.
39. Loh, Z. H.; Field, R. W. J Chem Phys 2003, 118, 4037.
40. Petersson, G. A.; Tensfeldt, T. G.; Montgomery, J. A. J Am Chem
Soc 1992, 114, 6133.
41. Li, X. PhD Dissertation, Chemistry Department, Carnegie Mellon
University: Pittsburgh, PA, 2007; p. 1.
42. Dai, Q.; Gellman, A. J. J Phys Chem 1993, 97, 10783.
43. Forbes, J. G.; Gellman, A. J. J Am Chem Soc 1993, 115, 6277.
44. Ye, P. P.; Gellman, A. J. J Phys Chem B 2006, 110, 9660.
45. Ye, P. P.; Gellman, A. J. J Am Chem Soc 2008, 130, 8518.
46. Greeley, J.; Mavrikakis, M. J Am Chem Soc 2002, 124, 7193.
47. Greeley, J.; Mavrikakis, M. J Catal 2002, 208, 291.
48. Lam, K. C.; Lin, Z. Y.; Marder, T. B. Organometallics 2007, 26,
3149.
DOI 10.1002/jcc