Open Journal of Physical Chemistry, 2013, 3, 119-125
http://dx.doi.org/10.4236/ojpc.2013.33015 Published Online August 2013 (http://www.scirp.org/journal/ojpc)
An Adjusted Model for Simple 1,2-Dyotropic Reactions.
Ab Initio MO and VB Considerations
Henk M. Buck
Kasteel Twikkelerf 94, Tilburg, The Netherlands
Received April 1, 2013; revised May 1, 2013; accepted June 1, 2013
Copyright © 2013 Henk M. Buck. This is an open access article distributed under the Creative Commons Attribution License, which
permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ABSTRACT
With an adjusted model, we reconsider simple 1,2-dyotropic reactions with the introduction of a concept based on the
intramolecular dynamics of a tetrahedron (van ’t Hoff modeling). In fact the dyotropic reactions are strongly related to
conversions originated from neighbouring group participation or anchimeric assistance, defined as the interaction of a
center with a lone pair of electrons in an atom and the electrons present in a ϭ or π bond. The researchful 1,2-dyotropic
reactions, based on the 1,2-interchange of halogens, methyl and hydrogen taking place in a concerted fashion, are in
competition with the two-step reaction in which the neighbouring group participation or anchimeric assistance comes to
full expression by ionic dissociation of the other exchangeable (halogen) atom. As to be expected there is an essential
difference between halogen or methyl exchange regarding the number of electrons participating in the transition state.
This aspect becomes evident in the geometries of the corresponding transition state geometries. In this paper we refer to
ab initio MO calculations and VB considerations. We consider the 1,2-halogen exchange as a combination of two SN2
reactions each containing four electrons. The van ’t Hoff dynamics appears a useful model in order to illustrate the
computations in a straightforward manner.
Keywords: Type-I Dyotropic Reactions; MO Calculations; Van ’t Hoff Model Considerations; Halogen and Methyl
Exchange; Conflicting Models
1. Introduction
Recently a theoretical model has been given for type-I
1,2-dyotropic reactions of the type CH2X−CH2X focused
on the exchange of X, based on sophisticated ab initio
computations. Reactions of this type have been defined
by Reetz as isomerizations that involve an intramolecular
one-step migration of the two ϭ bonds [1,2]. In type-I the
shift is based on 1,2-interchange of atoms (halogens) or
groups (methyl) that may result in inversion of configu-
ration of the positions under consideration. We recon-
sider these identity reactions with the results based on a
linear three-center four electron bonding, known as SN2
reactions in combination with van ’t Hoff modeling [3,4].
We introduce an adjusted model for this type of ex-
change reactions. For the 1,2-interchange of dibromides,
we will also focus the attention on a more complex sys-
tem based on the mutarotation of 5α, 6β-dibromide cho-
lestane that rearranges in the more stable diequatorial 5β,
6α isomer. In this situation, we are dealing with a more
or less fixed geometry that disfavours the flexibility dur-
ing the reaction course. We also take into consideration
symmetry changes by substitution of CH2X−CH2X for
SiH2X−CH2X and the corresponding dynamics.
2. Results and Discussion
2.1. MO Calculations. Van ’t Hoff Model
Description for 1,2-Dyotropic Halogen
Exchange Reactions in CH2X*−CH2X with
X = F, Cl, Br, and I. A Dibromide
Isomerization in a More Complex and
Rigid System
The 1,2-dyotropic reactions we consider, are focused on
identity halogen exchange in CH2X*CH2X (X = F, Cl,
Br, and I). We also take notice of the results of hydrogen
and methyl migration. The transition state (TS) is given
in Figure 1.
The relevant ab initio data for these exchange reac-
tions are given in Table 1. These model calculations in
combination with activation energies (ΔE≠) have been
computed at ZORA-OLYP/TZ2P by Fernández et al.
[5]. They also extended their studies to corresponding
methyl and hydrogen shifts. We focus on the ratio
C
opyright © 2013 SciRes. OJPC

H. M. BUCK 120



TS,C XRP,C Xdd 

 
3


based on the distances
(d) indicated as R(d), given in Table 1. The results in this
table also demonstrate a linear relation between ΔE≠ and
R(d). Similar observations have been done before and
completely worked out [4].
We compare these results with the identity SN2 substi-
tution reactions as given by

33
XCH XXCH XXCH+X


.
The results are given in Table 2. For a qualification of
the theoretical outcome we give the van ’t Hoff model
results. As we published before the van ’t Hoff model is
based on the transition from a regular tetrahedron into a
trigonal pyramid (TP) by moving the tetrahedral carbon
along the principal normal to the reaction center of the
triangle [8]. In the SN2 reaction mechanism as originally
proposed by Hughes and Ingold, the geometry of the TS
then corresponds with a trigonal bipyramid (TBP) via
backside attack of the incoming nucleophile, resulting in
inversion of carbon [9]. We then arrive to:

cos1cosR

 
in which θ is the van ’t Hoff tetrahedral angle. The value
for d [TS, C−X] can then be expressed by:


 
TS,C XcosRP,C XdRd    

The ideal R (cosθ) value is 1.333 with θ = 109.47˚.
By comparison the ratio values R(d) in Table 1 with
the corresponding values in Table 2, it is clear that going
from Cl to I there is a nearly constant difference. The
average values are 1.263 and 1.298, respectively. However,
it should be mentioned that the XCX angle in Figure 1
deviates from linearity for about 40˚. The ratio value of
1.333 for n = 4 can be intuitively obtained with:


11Rn n 2
in which n is the number of electrons in the TS [3].
This relation has been tested for identity methyl, pro-
ton, hydrogen atom, and hydride exchange reactions in
relation to three center four-(methyl cation and proton),
three-(hydrogen atom), and two electrons (hydride anion)
following the corresponding principal reaction coordinate
in the TS [3,8]. The R(n) values are then 1.333, 1.250,
and 1.167. These values are in good correspondence with
ab initio values and the van ’t Hoff model considerations.
In the case of four electrons and a three center carbon
configuration, hypervalency could be frozen as a TBP
configuration reflecting nicely the 1.333 value [3,4].The
significance for proton transfer focused on biochemical
networks could be clearly visualized with the van ’t Hoff
model making this and the other transfer reactions under-
standable in order to judge the computations [4].
From the description of the reaction type as proposed
by Fernández et al. [5] under investigation, four ϭ elec-
trons are involved. This results in a ratio value of 1.167
for each X−C−X bonding, a value that differs strongly
from the results in Table 1. Therefore the mechanism for
this type of reactions as presented in Figure 1 in combi-
nation with the computational results in Table 1 must be
considered from a model that includes an additional con-
tribution of four electrons in the TS. These extra elec-
trons can be delivered by one of the lone pairs of each
transferred X. Summarizing, each X delivers two ϭ elec-
trons (C−X bond) and one lone pair. This electron
Figure 1. Reaction pathway for with X = F, Cl, Br, and I.
*
22 22
CHXCHXCHX CHX
*
Table 1. Geometric values of distances (in Å), angles (in deg) and activation energies (ΔE≠ in kcal·mol−1) for the reaction
pathway as given in Figure 1a.
X TS, CX R(P), CX R(d)b ΔE≠ R(P), CC TS, CC TS, XCC TS, XX
F 1.874 1.398 1.340 65.1 1.523 1.401 68.0 3.476
Cl 2.283 1.803 1.266 42.6 1.517 1.412 72.0 4.343
Br 2.444 (2.510)c 1.982 (1.934)d 1.233 (1.298) 32.0 (28.0)c1.509 1.413 (1.417)c 73.2 (73.6)c 4.679
I 2.662 2.192 1.214 24.9 1.501 1.409 74.7 5.134
aThe distances, angles and activation energies are derived from computations of Fernández et al., [5]; bR(d) = TS,C−X/R( P),C−X; cFernández et al., [6]; dEx -
perimental distance for CH3Br [7].

H. M. BUCK 121
participation connects the results of Tables 1 and 2 in de-
monstrating a rather good correspondence between the
calculated distances and those obtained with the van ’t
Hoff model. In our opinion this electron participation in
the TS model may be an effective model for explaining
the 1,2-dyotropic halogen exchange reactions. Studying
the effects of electron-donating (D) and electron-ac-
cepting (A) substituents for the hydrogens of the eth-
ane moiety as (A)DXHC−CHXD(A), it appears that the
electron-donating substituents reduce the ΔE≠ in con-
trast with the acceptor substituents for the bromine
exchange [6]. This aspect may be understandable by
taken the Br−C−Br configuration. The displacement of
the electrons from C to Br will be facilitated by donor
substituents linked to carbon. This type of electron
transfer has been calculated for the SN2 reactions with
as TS. For X = Br calculations give
qCH3 = +0.188 and qBr = −0.594. This electron transfer
decreases from F to I [7].

3
XCH X



The reactions described are relatively simple in their
geometry. Therefore we will consider a 1,2-exchange
reaction for a more complex system as the diaxial 5α,
6β-dibromide isomerization of cholestane into the stable
diequatorial 5β, 6α-dibromide. This is illustrated in a
simplified way in Figure 2. Both dibromides when
treated with NaI in acetone undergo trans elimination
with regeneration of cholestane in which the 5α, 6β-di-
bromide reacts much faster because the bromine and
carbons all concerned lie in one plane and are in a fa-
voured position for a four-center TS [9].
Calculations on the bromine exchange have been car-
ried out by Fernández et al. for a simplified system (see
Table 1) and the more complex system 5α, 6β dibromo-
cholestane. Because of the rigid structure of cholestane,
there is a significant difference in the various distances in
the TS [6]. The calculations give for the top C(5)−Br =
2.766 Å and C(6)−Br = 2.673 Å, and for the bottom
C(5)−Br = 2.828 Å and C(6)−Br = 2.540 Å. These data
differ from the simple configuration as illustrated in
Figure 1 with corresponding values for the bond length
of C−Br in the TS as given in Table 1. The average value
is 2.693 Å corresponding with an R(d) value of 1.392 and
1.359 for the experimental C−Br distance of 1.934 Å and
the calculated value of 1.982 Å, respectively, as given in
Table 1. These values are in good correspondence with
the proposed van ’t Hoff model as a realistic approach
for the correctness of the ab initio calculations.
For the isomerization of the 5α, 6β dibromocholestane
also the activation parameters, as a first order in dibro-
mide, were determined in chloroform [11]. These values
are ΔH≠ = 19.9 kcal·mol−1, ΔE≠ = 20.6 kcal·mol−1, ΔS≠ =
−14.3 cal·mol−1K−1, and ΔG≠ = 24.2 kcal·mol−1. The
value of ΔE≠ can be compared with the values in Table 1
for the D2h symmetry. The negative value for ΔS≠ has
been interpreted that fewer degrees of freedom are
available than in the ground state which seems consistent
with a dyotropic reaction. An ionization with internal
return would show a positive value for ΔS≠. In these con-
siderations the nature of the medium may play an essen-
tial role. As has been proposed that a decrease in polarity
Table 2. Geometric values of distances (in Å) for the reaction pathway of the exchange reaction X + CH3  X via a trigonal
bipyramidal [XCH3X] transition statea. A comparison with the van ’t Hoff modelb.
ab initio Van ’t Hoff
X R(P), CX TS, C_X R(d)c R(P), CXd TS, CX R(cosθ)e
F 1.396 1.860 1.332 1.383 1.828 1.322
Cl 1.791 2.360 1.318 1.776 2.343 1.319
Br 1.959 2.510 1.281 1.934 2.522 1.304
I 2.157 2.720 1.261 2.132 2.812 1.319
aThe ab initio results are derived from computations of Bento et al. [10]; bThe model results are obtained from the dynamics of a regular tetrahedron as originated
by van ’t Hoff into a trigonal bipyramid. See text; cR(d) = TS, CX/R(P),CX; dThe experimental distances are derived from CH3X [7]; eR (cosθ) = 1 cosθ in
which θ is the experimental tetrahedral angle HCX in CH3X [7].
Figure 2. A simplified model for the isomerization of 5α, 6β dibromocholestane in the corresponding 5β, 6α isomer.

H. M. BUCK 122
will favour an intermediate or TS in which negligible
charge separation is involved favouring a dyotropic reac-
tion [11,12].
2.2. VB Considerations. Van ’t Hoff Model
Description for 1,2-Dyotropic Halogen
Exchange Reactions. The Influence of the
C−C Bonding on the Electron Distribution
in the X−C−X Transition State
An instructive visualization of the TS with four electrons
has been given in Figure 3.
This configuration shows an explicit contribution of
two electrons for the formation of a double bond charac-
ter of the C−C bonding. The MO calculations show an
average value of 1.409 Å for X = F (1.401 Å), Cl (1.412
Å), Br (1.413 Å), and I (1.409 Å), resulting in a partial
double bond character for the C−C bonding. Expressed in
bond orders with the corresponding number of electrons
in parenthesis, we then calculate 0.626 (1.252 e), 0.560
(1.120 e), 0.554 (1.108 e), and 0.578 (1.156 e), respec-
tively. With the expression of R(n), vide supra, we then
obtain 1.281, 1.287, 1.287, and 1.285 respectively, based
on eight electrons with 4 − q/2 electrons per X−C−X
configuration in which q is the electron density of the
partial C−C double bond. The R(n) values are in corre-
spondence with the average value of R(d) i.e. 1.263 in
Table 1, taking into account the deviation from linearity.
Using the van ’t Hoff model for the equatorial bonding,
then:

sin sinR

 
The value for d[TS, C−C] can then be expressed by:


 
TS,CCsinRP,CCdRd   

With the tetrahedral XCC angles 107.68˚, 109.79˚,
109.61˚, 109.93˚, respectively, we obtain for d [TS, C−C]
the corresponding values 1.451 Å, 1.427 Å, 1.421 Å, and
1.411 Å. There is significant deviation from the C−F
distance of 1.401 Å as given in Table 1. From these con-
siderations it is clear that the TS geometry of these type
of reactions must described by eight electrons and not by
four electrons as was supposed. Therefore it is of interest
to take notice of the methyl exchange reactions instead of
Figure 3. A characteristic VB configuration for a dyotropic
reaction of 1,2-X exchange.
halogen. In that case we are not dealing with extra elec-
trons as in the case of the additional lone pairs of the
halogens. The results will be discussed in the next sec-
tion.
2.3. MO Calculation. Van ’t Hoff Model for
1,2-Dyotropic Methyl Exchange Reactions
From the ab initio calculations it is clear that the C−C
bond distance in the TS (1.350 Å) is very close to the
ethylenic bond, corresponding with 1.859 e. Since no
extra electrons are available, only via hyperconjugation
of the methyl group, we calculated that only 1.071 e re-
mains for each H3C−C−CH3 configuration. This result
has no physical meaning in a three-center bonding. Ap-
parently, we are dealing with a different TS complex
than in the case of halogen exchange. In our opinion,
Figure 3 is a good representation. It is clear that in this
case we are approaching a dissociative TS. A similar
conclusion can be drawn for a corresponding 1,2-hy-
drogen exchange reaction. For a better understanding of
the different TS complexes of the halogen and methyl
exchange reactions it is obvious to consider halogen and
methyl migration at one side of the C−C linkage. At first
we will discuss the 1,2-methyl migration with corner-
protonated cyclopropane [13]. As to be expected the
corner-protonated cyclopropane, which can be consid-
ered as an intermediate in this methyl migration, is
closely related to the stable nonclassical 2-norbornyl
cation. The distances based on the triangle CXC, in
which X = CH3, are C−C 1.399 Å (1.394 Å) and C−CH3
1.803 Å (1.829 Å), corresponding values for the non-
classical 2-norbornyl cation are given in parenthesis [14].
Comparison of the geometry of the corner-protonated
cyclopropane with the structure in Figure 3 for X = CH3,
then there is with respect to the former one a decrease
in CXC angle of 13.94˚, a decrease in C−C bond distance
of 0.049 Å (3.50%) and an increase in C−X distance of
0.667 Å (36.99%). This dramatic increase in C−X
bond length in the 1,2-methyl migration asks for a simi-
lar analysis of the corresponding halonium geometries.
These halonium ions are known in the triangle geometry
as has been established from the NMR work of Olah et al.
[15]. We also mention a stable bromonium ion in the
reaction of adamantylideneadamantane with bromine by
Strating et al. [16]. The MO results are given in the next
section.
2.4. A Comparison between the MO Results of
the 1,2-Cyclic Halonium Ion and the
Transition State of 1,2-Dyotropic Halogen
Exchange
A coordinate halonium structure of the halogen exchange
is given in Figure 4.

H. M. BUCK 123
However, this intermediate follows a classical two-
step mechanism resulting in the same stereochemistry as
the one-step 1,2-dyotropic halogen exchange reaction. As
starting point for the reaction profile a C2h symmetry (R,
P) is selected. Via a conrotatory process, both halogens
reach a geometry that demonstrates a TP configuration in
which the central carbon is in-plane with 2H’s and the
other CH2 group, and the halogen is located in the axial
position of the TP. From UV-vis and NMR spectroscopic
measurements in combination with MO calculations
based on model systems as the proton complexes of 1,
1-diphenyl-2-halogeno (Cl, Br, and I) ethylenes with
para electron-donating substituents:
 
22
ArC1C2HXXCl, Br, andI

in which Ar is the aryl group with para substituents as
OCH3 and N(CH3)2, a TP geometry has been proposed
for C(2) as center in plane of the triangle formed by C (1)
and the 2H’s with X in a C(2)−X axial position [17,18].
With MO symmetry arguments related to the twofold
axis of symmetry of the carbenium ions, the A HOMO-
S LUMO transition appears a good criterion for deter-
mining the electron shift from C(2) to X. Generally for a
good fit between the UV-vis and NMR spectroscopy
with the MO calculations, a shift of about 0.6 e has been
taken place in the direction of X for the C(2)−X bonding.
A similar exclusive shift is absent for C(2)−F. In that
case the dominant electronegativity of F has already de-
pleted the C(2) electron density in the tetrahedral con-
figuration that cancels change in hybridization from sp3
into sp2 [18]. Summarizing, the TP model is valid for Cl,
Br and I whereas F preserves its tetrahedral configuration
in the bonding with C(2).
Furthermore from other data it is well known that F is
not able to form a triangle halonium ion [15]. After
reaching this location on the reaction coordinate, there is
an inversion of charge on the halogen by coordination
with the other carbon of the ethane linkage. The reaction
then proceeds via D2h symmetry as shown in Figure 3.
The differences in geometry between the corner-proto-
nated cyclopropane and the 1,2-dyotropic methyl ex-
change TS are much more pronounced than the corre-
sponding geometric differences between the halonium
ions [19] and the 1,2-dyotropic halogen TS. For the CXC
Figure 4. A coordinate halonium structure of the 1,2-dyo-
tropic reaction.
angle, the decrease is 9.84˚ (Cl) and 6.34˚ (Br). The in-
crease for the C−X distance is 0.407 Å (21.70%, Cl) and
0.335 Å (15.88%, Br). The differences for the C−C bond
length are of minor significance. This explains the fun-
damental distinction between the methyl and halogen
exchange reactions.
The position of F on the energy profile is now also
clear. In the 1,2-F migration ΔE≠ (65.1 kcal·mol−1) is
much higher than for the corresponding halogens (Cl
42.6 kcal·mol−1, Br 32.0 kcal·mol−1, and I 24.9 kcal·
mol−1). Accommodation of positive charge on F in com-
parison with the other halogens is in fact an unfavourable
model for double migration. Although the locations on
the energy profile are focused on the interaction of one of
the lone pairs halogens with the other carbon of the C−C
linkage, there may be another profile for the 1,2-F migra-
tion.
The contrast between F and the other halogens is clear.
For the geometries in Figure 1, the following CXC bond
angles were calculated 43.90˚ (X = F), 36.02˚ (X = Cl),
33.61˚ (X = Br), and 30.69˚ (X = I). A similar behaviour
is found for the open structures of the dialkylhalonium
ions 3

3
HC X CH


 .The calculations show for the
CXC bond angles 120.2˚ (X = F), 105.0˚ (X = Cl),
101.4˚ (X = Br), and 97.7˚ (X = I). The cations for X =
Cl, Br, and I have been prepared as long-lived cations.
However, no stable dialkylfluoronium ion has been ob-
tained [20]. The exclusive electronegativity of F with
respect to the other halogens determines the expansion of
the XCX angle. For simplicity it means that F aims at
an increase of its s character. In that respect it is of inter-
est to mention the results of the calculations of methy-
lated dimethylhalonium ions. Theoretically it has been
found that methylated dimethylhalonium ions accommo-
date a tetrahedral configuration whereas

2
33
CH F



has a D3h symmetry [20]. Thus by going from C3v to D3h
symmetry, F increases its s character.
Recently, there was mechanistic evidence for a sym-
metrical intermediate in solution [21]. The CFC bond
angle and the C−F distance calculated from this fluoronium
ion derived from a fixed configured precursor, correspond
with the values as given for the

33
HC F CH

 ion
[20].
2.5. A Comparison of 1,2-SiH3 and CH3 Shifts as
Substituents in Ethane. MO Calculations
and Van ’t Hoff Model Consideration
We consider reactions as illustrated in Figure 1 in which
X = SiH3. Like CH3, Si has not the capacity to deliver
extra electrons. So it is to be expected that the geometry
for the TS of the SiH3 shift is in correspondence with the
methyl shift. In fact it means that two electrons are
available for each H3Si−C−SiH3 configuration. This elec-

H. M. BUCK 124
tron participation results in conflicting values for R(n)
and R(d). In order to escape this split, the TS can be de-
scribed as given in Figure 3. This is in excellent agree-
ment with the ethylenic bond distance of 1.354 Å corre-
sponding with 1.81 e. This geometry involves a dissocia-
tive state. The calculations show that the ΔE≠ for the
SiH3 shift (101.7 kcal·mol−1) is smaller than for the CH3
shift (131.0 kcal·mol−1). This aspect can be qualitatively
explained by orbital expansion of Si compared with car-
bon.
2.6. Symmetry Change in the 1,2-Dyotropic
Halogen Exchange Reactions
Changing the C−C linkage through Si−C is of interest in
consequence of loss of its symmetry and an increasing
coordination ability of Si compared with carbon [5]. This
aspect is observable for the 1,2-F migration between Si
and carbon. In that specific case there is a pronounced
asymmetry in F shift as follows from the different angles
in the triangle CSiF 76.31˚, SiCF 50.39˚, and CFSi
53.30˚. For comparison the corresponding results are
given for the symmetric TS of the 1,2-exchange reaction
of the 1,2-disubstituted fluoroethane i.e., FCC 68.04˚ and
CFC 55.98˚. Thus in the asymmetric TS the deviation
from linearity is substantially decreased compared with
the symmetric one. This aspect is recognized in ΔE≠. For
the asymmetric TS 48.9 kcal·mol−1 has been calculated
whereas for the symmetric TS 65.1 kcal·mol−1 is found. It
involves that Si accommodates a fifth ligand easier than
carbon with a geometry closely related to a TBP con-
figuration that results in lowering of the ΔE≠ of the TS
[22]. This aspect is also reflected in the value of R(d) for
the Si−F bond in the transition intermediate that ap-
proaches its normal bond length. However, for coordina-
tion of the other F with Si the displacement of F is con-
siderable, resulting in a high R(d) value of 1.582 for the
C−F bond that influences the transition for the 1,2-F mi-
gration in a negative manner. In the corresponding 1,2-
disubstituted chloroethane, the difference between the
angles CSiCl and SiCCl is 7.21˚ whereas in the former
one a value of 25.92˚ is found. The R(d) values of Si−Cl
and C−Cl are 1.212 and 1.353, respectively with an average
value of 1.283. This value is in correspondence with the
average value of 1.309 of the fluorine exchange. This dif-
ference is also reflected in the ΔE≠ values. The 1,2-Cl mi-
gration is 9.8 kcal·mol−1 in favour over the 1,2-F exchange.
3. Conclusion
It has been suggested that type-I 1,2-dyotropic reactions
as presented in this paper are considered as four-mem-
bered transition states, involving a concerted exchange
migration of the X atoms or groups in CH2X−CH2X. The
discussion is based on X = halogen, methyl and hydrogen.
According to our results there is a fundamental differ-
ence in the description of this dyotropic reaction with
others concerning the participation of the number of
electrons in the TS based on a clear distinction between
the halogen and the methyl and hydrogen exchange. The
difference is clear. The halogen exchange takes profit
from the presence of its lone pair electrons. This “cata-
lyzing” effect is absent for methyl and hydrogen ex-
change, explaining the relatively high ΔE≠ values of
131.0 and 145.2 kcal·mol-1, respectively, in comparison
with the halogens as shown in Table 1. A similar effect
is found for the R(d) values of 1.613 and 1.697, respec-
tively, compared with the R(d) values of the halogens in
Table 1. The differences between methyl and hydrogen
migration as expressed in ΔE≠ and R(d) are in fact a
measure for the effect of hyperconjugation of the methyl
group in the exchange reaction. The high values for both
R(d)’s of the methyl and hydrogen binding in the TS
(much higher than the van ’t Hoff value of 1.333) are an
indication for a loose complex binding or a dissociative
state as illustrated in Figure 3 and is confirmed by the
distance of the C−C bond that approaches the double
bond character. The effect of the SiH3 transfer is still
more explicit compared with the CH3 migration, result-
ing in a decrease in ΔE≠ value of 29.3 kcal·mol−1 as a
result of Si-orbital expansion. Generally, there is a strict
linear relation between ΔE≠ and R(d) for the halogen ex-
change. This relation is supported by definition that for


EabRd

 then must apply a + b = 0 which
has been established. The methyl and hydrogen shift de-
viates from the halogen linearity. Finally, it is our con-
clusion that the mechanistic view of Fernández et al. [5]
based on a qualitative VB analysis as given in Figure 3
is far from complete. However, this approach is usable
for the methyl and hydrogen exchange because of its
reduced tendency for bonding in consequence of the
available electrons in the transition state. In fact the dif-
ferences in the 1,2-X migration are based on the overall
number of electrons participating in the transition state.
This situates the halogens with their additional lone pairs
in a complete different position as methyl and hydrogen.
REFERENCES
[1] M. T. Reetz, “Dyotropic Rearrangements, a New Class of
Orbital-Symmetry Controlled Reactions. Type I,” Ange-
wandte Chemie International Edition in English, Vol. 11,
No. 2, 1972, pp. 129-130. doi:10.1002/anie.197201291
[2] I. Fernández, F. P Cossío and M. A. Sierra, “Dyotropic
Reactions: Mechanisms and Synthetic Applications,” Che-
mical Reviews, Vol. 109, No. 12, 2009, pp. 6687-6711.
doi:10.1021/cr900209c
[3] H. M. Buck, “A Combined Experimental, Theoretical,
and Van ’t Hoff Model Study for Identity Methyl, Proton,
Hydrogen Atom, and Hydride Exchange Reactions. Cor-

H. M. BUCK 125
relation with Three-Center Four-, Three-, and Two-Elec-
tron Systems,” International Journal of Quantum Chem-
istry, Vol. 108, No. 9, 2008, pp. 1601-1614.
doi:10.1002/qua.21683
[4] H. M. Buck, “A Linear Three-Center Four Electron Bond-
ing Identity Nucleophilic Substitution at Carbon, Boron,
and Phosphorus. A Theoretical Study in Combination
with Van ’t Hoff Modeling,” International Journal of
Quantum Chemistry , Vol. 110, No. 7, 2010, pp. 1412-
1424. doi:10.1002/qua
[5] I. Fernández, F. M. Bickelhaupt and F. P. Cossío, “Type-I
Dyotropic Reactions: Understanding Trends in Barriers,”
Chemistry—A European Journal, Vol. 18, No. 39, 2012,
pp. 12395-12403. doi:10.1002/chem.201200897
[6] I. Fernández, M. A. Sierra and F. P. Cossío, “Stereoelec-
tronic Effects on Type I 1,2-Dyotropic Rearrangements in
Vicinal Dibromides,” Chemistry—A European Journal,
Vol. 12, No. 24, 2006, pp. 6323-6330.
doi:10.1002/chem.200501517
[7] M. N. Glukhovtsev, A. Pross and L. Radom, “Gas-Phase
Identity SN2 Reactions of Halide Anions with Methyl
Halides. A High-Level Computational Study,” Journal of
the American Chemical Society, Vol. 117, No. 7, 1995, pp.
2024-2032. doi:10.1021/ja00112a016
[8] H. M. Buck, “A Model Investigation of Ab Initio Geome-
tries for Identity and Nonidentity Substitutions with Three-
Center Four- and Three-Electron Transition States,” In-
ternational Journal of Quantum Chemistry, Vol. 111, No.
10, 2011, pp. 2242-2250. doi:10.1002/qua.22529
[9] T. H. Lowrey and K. S. Richardson, “Mechanism and
Theory in Organic Chemistry,” Harper & Row, New
York, 1976.
[10] A. P. Bento and F. M. Bickelhaupt, “Nucleophilicity and
Leaving-Group Ability in Frontside and Backside SN2
Reactions,” The Journal of Organic Chemistry, Vol. 73,
No. 18, 2008, pp. 7290-7299. doi:10.1021/jo801215z
[11] D. H. R. Barton and A. J. Head, “Long-Range Effects in
Alicyclic Systems. Part I. The Rates of Rearrangement
of Some Steroidal Dibromides,” Journal of the Chemical
Society, 1956, pp. 932-937. doi:10.1039/jr9560000932
[12] C. A. Grob and S. Winstein, “Mechanismus der Mutaro-
tation von 5, 6-Dibromcholestan,” Helvetica Chimica Acta,
Vol. 35, No. 3, 1952, pp. 782-802.
doi:10.1002/hlca.19520350315
[13] L. Radom, J. A. Pople, V. Buss and P. V. R. Schleyer,
“Molecular Orbital Theory of the Electronic Structure of
Organic Compounds. XI. Geometries and Energies of
C3H7 Cations,” Journal of the American Chemical Society,
Vol. 94, No. 2, 1972, pp. 311-321.
doi:10.1021/ja00757a001
[14] P. R. Schreiner, D. L. Severance, W. L. Jorgensen, P. von
Schleyer and H. F. Schaefer III, “Energy Difference be-
tween the Classical and the Nonclassical 2-Norbornyl
Cation in Solution. A Combined Ab Initio—Monte Carlo
Aqueous Solution Study,” Journal of the American
Chemical Society, Vol. 117, No. 9, 1995, pp. 2663-2664.
doi:10.1021/ja00114a037
[15] G. A. Olah, G. K. S. Prakash, Á. Molnár and J. Sommer,
“Superacid Chemistry,” 2nd Edition, John Wiley & Sons,
Inc., Hoboken, 2009. doi:10.1021/ja00114a037
[16] J. Strating, J. H. Wieringa and H. Wynberg, “The Isola-
tion of a Stabilized Bromonium Ion,” Journal of the Che-
mical Society D Chemical Communications, No. 16, 1969,
pp. 907-908.
[17] H. M. Buck, “Mechanistic Models for the Intramolecular
Hydroxycarbene-Formaldehyde Conversion and Their In-
termolecular Interactions: Theory and Chemistry of Ra-
dicals, Mono-, and Dications of Hydroxycarbene and Re-
lated Configurations,” International Journal of Quantum
Chemistry, Vol. 112, No. 23, 2012, pp. 3711-3719.
doi:10.1002/qua.24127
[18] H. M. Buck, “Trigonal Pyramidal Carbon Geometry as
Model for Electrophilic Addition-Substitution and Elimi-
nation Reactions and Its Sinificance in Enzymatic Proc-
esses,” International Journal of Quantum Chemistry, Vol.
107, No. 1, 2007, pp. 200-211. doi:10.1002/qua.21061
[19] D. F. Shellhamer, D. C. Gleason, S. J. Rodriguez, V. L.
Heasley, J. A. Boatz and J. J. Lehman, “ Correlation of
Calculated Halonium Ion Structures with Product Distri-
butions from Fluorine substituted Terminal Alkenes,”
Tetrahedron, Vol. 62, No. 50, 2006, pp. 11608-11617.
doi:10.1016/j.tet.2006.09.071
[20] G. A. Olah, G. Rasul, M. Hachoumy, A. Burrichter and G.
K. S. Prakash, “ Diprotonated Hydrogen Halides (H3X2+)
and Gitonic Protio Methyl- and Dimethylhalonium Dica-
tions (CH3XH22+ and (CH3)2XH2+). Theoretical and Hy-
drogen-Deuterium Exchange Studies,” Journal of the
American Chemical Society, Vol. 122, No. 12, 2000, pp.
2737-2741. doi:10.1021/ja994044n
[21] M. D. Struble, M. T. Scerba, M. Siegler and T. Lectka,
“Evidence for a Symmetrical Fluoronium Ion in Solu-
tion,” Science, Vol. 340, No. 6128, 2013, pp. 57-60.
doi:10.1126/science.1231247
[22] A. Streitwieser Jr. and C. H. Heathcock, “Introduction in
Organic Chemistry,” 3rd Edition, Macmillan Publishing
Company, New York, 1985.
doi:10.1126/science.1231247