The Chapman-type rearrangement in pseudosaccharins: The case of 3-(methoxy)-1,2-benzisothiazole 1,1-dioxide

Article abstract of DOI:10.1016/j.molstruc.2008.05.054

The thermal Chapman-type rearrangement of the pseudosaccharin 3-(methoxy)-1,2-benzisothiazole 1,1-dioxide (MBID) into 2-methyl-1,2-benzisothiazol-3(2H)-one 1,1-dioxide (MBIOD) was investigated on the basis of computational models and knowledge of the structure of the reactant and product in the isolated and solid phases. X-ray diffraction was used to obtain the structure of the substrate in the crystalline phase, providing fundamental structural data for the development of the theoretical models used to investigate the reaction mechanism in the condensed phase. The intra- and different intermolecular mechanisms were compared on energetic grounds, based on the various developed theoretical models of the rearrangement reactions. The energetic preference (ca. 3.2 kJ mol^-1, B3LYP/6-31+G(d,p)) of inter- over intramolecular transfer of the methyl group is predicted for the quasi-simultaneous transfer of the methyl groups model, explaining the potential of MBID towards [1,3′]-isomerization to MBIOD in the condensed phases. The predicted lower energy of MBIOD relative to MBID (ca. 60 kJ mol^-1), due to the lower steric hindrance in the MBIOD molecule, acts as a molecular motor for the observed thermal rearrangement.

Full text of DOI:10.1016/j.molstruc.2008.05.054

Journal of Molecular Structure 892 (2008) 343–352
Contents lists available at ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
The Chapman-type rearrangement in pseudosaccharins:
The case of 3-(methoxy)-1,2-benzisothiazole 1,1-dioxide
A. Kaczor a,b, , L.M. Proniewicz , R. Almeida ,a A. Gómez-Zavaglia a,d, M.L.S. Cristiano c,
b
a,c
*
e
e
A.M. Matos Beja , M. Ramos Silva , R. Fausto
Department of Chemistry, University of Coimbra, Coimbra, 3004-535, Portugal
Faculty of Chemistry, Jagiellonian University, Ingardena 3, 30-060 Krakow, Poland
Department of Chemistry and Biochemistry, F.C.T., and CCMAR, University of Algarve, Campus de Gambelas, 8005-039 Faro, Portugal
Faculty of Pharmacy and Biochemistry, University of Buenos Aires, C.P. 1113 Buenos Aires, Argentina
Department of Physics, University of Coimbra, Coimbra 3004-536, Portugal
a
b
c
d
e
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 2 April 2008
Received in revised form 27 May 2008
Accepted 30 May 2008
Available online 12 June 2008
The thermal Chapman-type rearrangement of the pseudosaccharin 3-(methoxy)-1,2-benzisothiazole 1,1-
dioxide (MBID) into 2-methyl-1,2-benzisothiazol-3(2H)-one 1,1-dioxide (MBIOD) was investigated on
the basis of computational models and knowledge of the structure of the reactant and product in the iso-
lated and solid phases. X-ray diffraction was used to obtain the structure of the substrate in the crystal-
line phase, providing fundamental structural data for the development of the theoretical models used to
investigate the reaction mechanism in the condensed phase. The intra- and different intermolecular
mechanisms were compared on energetic grounds, based on the various developed theoretical models
Keywords:
Pseudosaccharyl ether
Molecular structure
Reaction path
IR spectra
Matrix-isolation
DFT calculations
ꢀ1
of the rearrangement reactions. The energetic preference (ca. 3.2 kJ mol , B3LYP/6-31+G(d,p)) of inter-
over intramolecular transfer of the methyl group is predicted for the ‘‘quasi-simultaneous” transfer of
0
the methyl groups model, explaining the potential of MBID towards [1,3 ]-isomerization to MBIOD in
ꢀ1
the condensed phases. The predicted lower energy of MBIOD relative to MBID (ca. 60 kJ mol ), due
to the lower steric hindrance in the MBIOD molecule, acts as a molecular motor for the observed thermal
rearrangement.
Ó 2008 Elsevier B.V. All rights reserved.
1
. Introduction
Pseudosaccharins (substituted 1,2-benzisothiazole 1,1-dioxides)
Classically, the Chapman rearrangement is assumed to be intra-
molecular for aryl imidates [6–8] and alkoxypyrimidines [9]. On
the other hand, the Chapman-type thermal rearrangement of the
have attracted attention as significant intermediates in organic
synthesis, with particular emphasis on their O-ethers, known as
important intermediates for the hydrogenolysis of alcohols. The
key for the reactivity of this family of compounds is the strong
electron withdrawing ability of the pseudosaccharyl group which
simplest alkylpseudosaccharyl ether, 3-(methoxy)-1,2-benziso-
thiazole 1,1-dioxide (MBID), resulting in [1,3 ]-isomerization to
0
2-methyl-1,2-benzisothiazol-3(2H)-one
1,1-dioxide
(MBIOD)
(Fig. 1), was observed long ago to occur in the molten phase and
was suggested to take place via an intermolecular mechanism
[10]. Recently, it was demonstrated that this process can also hap-
pen in the crystalline state [5].
reflects in its exceptionally long and short C
A R
AO and C AO bonds
(
where R = heteroaromatic ring and A = aliphatic or aryl group),
respectively [1–3]. This characteristic structural feature of
pseudosaccharyl O-ethers is considered to be responsible for the
high ability of the pseudosaccharyl-oxygen system to act as nucle-
ofuge in reductive cleavage reactions catalyzed by transition met-
als [1,2,4]. The ability of pseudosaccharyl ethers towards the
cleavage of the weak C AO bond has also been observed in other
A
reactions, such as the Claisen- and Chapman-like thermal isomer-
izations resulting in formation of N-isomers [3,5].
H
18
C17 H20
H19
H10
H10
O15
C9
O15
H11
C6
H11
C6
H18
N8 C17
C9
C5
C4
C1
C2
Δ
C5
C4
C1
C2
N8
H20
S7
S7
H12
C3
H12
C3
H
19
O16
O16
H13
O14
O14
H13
*
Corresponding author. Address: Faculty of Chemistry, Jagiellonian University,
Fig. 1. The Chapman-type rearrangement of MBID (the atom numbering schemes
for MBID and MBIOD adopted throughout this paper are those presented in the
figure).
Ingardena 3, 30-060 Krakow, Poland.
E-mail address: kaczor@chemia.uj.edu.pl (A. Kaczor).
0
022-2860/$ - see front matter Ó 2008 Elsevier B.V. All rights reserved.
doi:10.1016/j.molstruc.2008.05.054

3
44
A. Kaczor et al. / Journal of Molecular Structure 892 (2008) 343–352
Intermolecular Chapman-like rearrangements in the solid phase
129–130 (C; 1H NMR (300 MHz, CDCl
3
), 3.28 (3H, s), 7.80–7.90
were previously described for oxadiazoles [11,12]. Intermolecular
methyl migration in the solid state was also observed for methyl
p-dimethylaminobenzenesulfonate [13–15] and 2-(4 -nitroanili-
no)4,6-dimethoxy-1,3,5-triazine [16]. Interestingly, for both the
oxidiazoles and dimethylaminobenzenesulfonate the reaction in
the solid phase was found to be faster than in the melt, and this
observation was rationalized in terms of an optimal alignment of
the reacting atoms in the crystal lattice of the studied compounds
(2H, m), 7.92–7.96 (1H, d), 8.05–8.10 (1H, d); MS (EI): m/z 197
([M]+, 100%)). The synthesis of MBID was described elsewhere
[5,19].
0
The infrared spectra were recorded in the 500–4000 cm 1 range
ꢀ
using a Mattson Infinity 60AR series FT-IR spectrometer, with res-
ꢀ1
olution of 0.5 cm . The sample was co-deposited with xenon
(N45, Air Liquide) isolant gas onto a cryogenically cooled (20 K)
CsI window. The compound was sublimated at ca. 340 K from a
specially designed mini-furnace assembled inside the cryostat.
The selected temperature of the CsI optical substrate was obtained
using an APD Cryogenics closed-cycle helium refrigeration system
with a DE-202A expander. The temperature was measured directly
at the sample holder by a silicon diode temperature sensor, con-
nected to a digital controller (Scientific Instruments, Model 9650-
1) with the accuracy of 0.1 K. Detailed descriptions of experimental
conditions applied to obtain spectra of matrix-isolated MBID as
well as polycrystalline spectra of both compounds may be found
elsewhere [5,19].
[
11,13]. This type of rearrangements, which proceed faster in the
solid than in the melt, are recognized as ‘‘topochemically con-
trolled” and have also been previously observed for methyl group
transfer reactions in thiocyanurates [17] and 4,6-dimethoxy-3-
methyl-dihydro-triazine-2-one [18].
The observation that the rearrangement of MBID may take place
in the solid-state [5] motivated us to shed some light on the mech-
anism of this reaction. To account for the intermolecularity of the
observed reaction, the theoretically predicted intra- and various
intermolecular reaction pathways were investigated in this study.
In order to ensure applicability of the proposed models and to
characterize in deeper detail the relevant chemical species, ma-
trix-isolation infrared spectroscopy was used to obtain the vibra-
tional spectrum of the monomeric final product (MBIOD), which
was compared with previously obtained data for the monomeric
substrate (MBID) [19]. The crystallographic structure of the sub-
strate was determined through X-ray diffraction analysis.
X-ray data were collected on a Bruker APEX II-CCD diffractom-
eter, using a transparent plate shaped crystal with dimensions
0.26 ꢁ 0.16 ꢁ 0.06 mm. The crystallographic structure was solved
by direct methods using SHELXS-97 [20]. Refinements were carried
out with the SHELXL-97 package [20]. All refinements were made
2
by full-matrix least-squares on F with anisotropic displacement
parameters for all non-hydrogen atoms (Table 1 for details).
2.2. Computational
2
. Materials and methods
Geometry optimizations and calculations of vibrational fre-
2.1. Experimental
quencies of MBIOD were performed at the DFT/B3LYP level of the-
ory [21,22], using the 6-31+G(d,p), 6-31++G(3df,3pd) and 6-
MBIOD was obtained from MBID by heating a neat sample of
3
11++G(3df,3pd) basis sets.
this latter compound at ca. 185 °C and keeping the sample at this
temperature until all reactant has been consumed, as confirmed
by TLC analysis. MBIOD was isolated as colourless crystals: m.p.
The intrinsic reaction coordinate (IRC) reaction path for the
intramolecular rearrangement of MBID into MBIOD was calculated
at the B3LYP/6-31++G(3df,3pd) level of theory and the geometry
and other relevant properties of the transition state (TS) associated
to this process were obtained at the same level of theory with the
help of the synchronous transit-guided quasi-Newton method
Table 1
Summary of X-ray data collection and processing parameters for MBID structure
(
STQN-QST3) [23,24]. In order to compare these results with the
Chemical formula
Colour/shape
Formula weight
Space group
C
8
H
7
NO
3
S
investigated intermolecular models, the geometry and vibrational
frequencies of the substrate, product and TS state were recom-
puted at the B3LYP/6-31+G(d,p) level [giving energy values in close
agreement with those obtained at the higher B3LYP/6-
31++G(3df,3pd) level].
Three alternative reaction pathways modeling the intermolecu-
lar rearrangement were computed at the B3LYP/6-31+G(d,p) level
of theory. In model 1, a sequential process was considered: the
Colourless/plate
394.43/2
P1
ꢀ
Temperature, K
293(2)
430.08(10)
Triclinic
6.9412(9)
7.7281(10)
8.1558(11)
100.044(7)
92.511(7)
91.670(7)
2
Cell volume (Å 3
Crystal system
a (Å)
)
b (Å)
c (Å)
a
(deg)
b (deg)
(deg)
Formula units/unit cell
bond distance between the nitrogen atom of one molecule and
0
the methyl carbon atom of the second molecule (C17ꢂꢂꢂN ) was var-
8
c
ied stepwise and all other parameters optimized at each point, pro-
ducing an anion–cation pair; subsequently, a second methyl group
ꢀ
3
D
c
(Mg m
)
1.523
transfer for anion–cation and neutral–anion pathways was consid-
Diffractometer/scan
APEX II KAPPA CCD/
x,/
0
Radiation (Å) (graph. monochromated)
Max crystal dimensions (mm)
h Range (deg)
0.71073
0.26 ꢁ 0.16 ꢁ 0.06
2.54–28.34
ered by scanning the C ꢂꢂꢂN
8
distance, leading to the final products
1
7
0
0
(
‘‘sequential” model). In model 2, both C17ꢂꢂꢂN and C ꢂꢂꢂN
8
dis-
8
17
tances were varied simultaneously and the two methyl groups
Range of h, k, l
Reflections measured/independent
ꢀ9/9, ꢀ10/10, ꢀ10/10
transferred synchronously (‘‘simultaneous” transfer model). Final-
17231/2117 (Rint = 0.0221)
0
Reflections observed I > 2
Corrections applied
Computer programs
Structure solution
No. of parameters varied
GOF
r
1779
ly, in model 3, the C17ꢂꢂꢂN distance was varied up to ca. 1.9 Å, then
8
0
Lorentz and polarization effects
SHELXL, SHELXS, PLATON
Direct methods
119
1.062
fixed at this value while the C ꢂꢂꢂN
distance was varied up to full
1
7
8
0
first methyl group transfer and, finally, the C17ꢂꢂꢂN distance further
8
shortened to allow for the second methyl transfer (‘‘quasi-simulta-
neous” transfer model). In all considered models, the reactant and
all neutral products were fully optimized and their vibrational fre-
quencies subsequently computed at the B3LYP/6-31+G(d,p). Fur-
ther description on these models is given in the Models for the
MBID ? MBIOD rearrangement section.
R1
0.0322
wR2
0.0865
2
2
Function minimized
R
w(
DF )
Diff. density final max/min (e Å 3)
ꢀ
0.278/ꢀ0.312

A. Kaczor et al. / Journal of Molecular Structure 892 (2008) 343–352
345
All the above-mentioned calculations were performed using the
GAUSSIAN 03 suite of programs [25]. Potential energy distributions
PEDs) of the normal modes were computed in terms of natural
internal coordinates [26] with the GAR PED program [27].
(
2
3
. Results and discussion
3.1. Geometry of MBID: comparison between X-ray (crystal) and
theoretical (isolated monomer) data
The structure of the isolated molecule of MBID was recently
studied in great detail using different theoretical models [19]. In
the present investigation its structure in the crystalline state has
been addressed using X-ray diffraction. The X-ray crystal structure
of MBID is presented in Figs. 2 and 3. Atomic coordinates, bond
lengths, valence angles, dihedral angles, and other crystallographic
data were deposited at the Cambridge Crystallographic Data Center
CCDC No. CCDC 674704.
Apart from O14 and O16, in the crystal the molecules of MBID are
mainly planar, with O15 and C17 sharing the plane of the two fused
rings (Fig. 2). The O14 and O16 atoms are not equally far-away from
Fig. 3. Crystal packing of MBID.
such plane. The distance between O16 and the least-squares plane
0
of the ring0 s is 1.1869(12) ÅA while for
.2613(12) ÅA . The torsion angle N AC AO15AC17 is 1.1(2)°. There
are no conventional hydrogen bonds between the molecules due
to the lack of donors. Even the weaker CAH. . . interactions are
O14 the distance is
the crystal it appears slightly distorted, as described above. The
similarity between the gas phase and crystal state MBID molecular
geometries supports the use of isolated molecule based theoretical
models to address the reactivity of this molecule in the condensed
phases.
1
8
9
p
not observed in this structure. The closest intermolecular contact
0
i
for O14 is with C17 (i: ꢀx, 1ꢀy, 1ꢀz) at a distance of 3.319(2) ÅA
ii
6
while for O16 the closest intermolecular contact is with C
(ii:
0
3.2. Geometry of MBIOD. Comparison of the substrate and the product
1
ꢀx, ꢀy, ꢀz) [3.443(2) ÅA ]. The main interaction governing the pack-
ing of the molecules is . . . interaction between the electron
clouds of neighboring saccharin derivatives. This interaction dis-
plays the usual slipped stacking geometry, with the interacting
p
p
p
There is only one minimum on the potential energy surface
for MBIOD. The stable conformer of MBIOD has C symmetry
s
p
and one of the hydrogen atoms of the methyl group in the
anti-periplanar arrangement relatively to the SAN bond. The
optimized B3LYP/6-311++G(3df,3pd) geometrical parameters of
MBIOD are given in Table 2.
systems antiparallel displaced. The centroid–centroid distance be-
tween consecutive centroids of the six-membered rings is
3
.9097(9) Å and the distance between consecutive centroid of
five-membered rings/centroid of six-membered rings is
0
MBIOD was found to be significantly more stable (ca. 58–
3
.9097(9) ÅA . The p. . .p interactions join the molecules in columns
ꢀ1
6
0 kJ mol depending on the level of theory applied) than MBID
that run along the a axis (Fig. 3).
[
19]. This difference in stability might be rationalized in terms of
The comparison between the geometrical parameters of the iso-
lated monomer of MBID, obtained at the B3LYP/6-311++G(3df,3pd)
level of theory [5], and those determined for the molecule in the
crystal (Table 2), reveals the excellent correlation between the
two sets of data, providing further evidence on the absence of spe-
cific strong intermolecular interactions in the crystal. Nevertheless,
differences between the geometries of the two compounds. Obvi-
ously, the tautomerization reaction must affect considerably the
9
length of the C AO15 bond.
9
Although an extremely short C AO15 bond is a characteristic
feature of the O-ethers of pseudosaccharins [1–3], as mentioned
in the Introduction, the conversion of the ether linkage (in MBID)
into a carbonyl bond (in MBIOD) further shortens this bond, from
s
in the gas phase the molecule is predicted to be strictly C , while in
1
.320 to 1.209 Å [B3LYP/6-311++G(3df,3pd) results]. This value is
even smaller than that normally associated with the CAO bond
length in a carbonyl of a ketone or aldehyde (1.23 Å). This change
constrains considerable redistribution of the electronic density
within the five-membered heterocyclic ring. The striking differ-
ence could be easily anticipated to take place in the length of the
C AN bond, which increases from 1.284 Å in MBID to 1.386 Å in
9
MBIOD. The changes in the lengths of the remaining bonds of the
isothiazole ring nearly cancel each other, with the CAS bond being
shortened by 0.023 Å and the NAS and C AC bonds being length-
2 9
ened by 0.012 and 0.010 Å, respectively. On the whole, the hetero-
cyclic ring is predicted to be more strained in MBID than in MBIOD,
with the sum of bond lengths being 0.100 Å smaller in the first
molecule, while the sum of the ring-internal angles is nearly equal
for both compounds. The higher steric hindrance of MBID com-
pared with that of MBIOD is then one of the main factors respon-
sible for the higher stability of the latter compound. It is also
worth noting that the changes in the geometry of the five-mem-
bered ring upon MBID ? MBIOD conversion practically do not
Fig. 2. ORTEP diagram of MBID, showing the displacement ellipsoids drawn at the
0% probability level.
5

3
46
A. Kaczor et al. / Journal of Molecular Structure 892 (2008) 343–352
Table 2
take place if the reaction is not topochemically controlled. An
interesting example of such methyl conversion is the rearrange-
ment of 2-(4 -nitroanilino) 4,6-dimethoxy-1,3,5-triazine into a
mixture of products, which occurs in the solid state although there
are no contacts between the reaction centers of the substrate in the
crystal [16].
Chosen experimental (X-ray; single crystal) and calculated geometrical parameters
for MBID and calculated geometrical parameters for MBIOD
0
Parameter
MBID
X-ray
MBIOD
DFT/B3LYP/
DFT/B3LYP/
6-311++G(3df,3pd)
6
-311++G(3df,3pd) [5]
Bond length (Å)
In order to understand the energetic factors controlling the ob-
served MBID ? MBIOD rearrangement, models for the intra- and
intermolecular processes shall be considered and compared.
The knowledge about the structures of the monomeric sub-
strate and product discussed above, also rationalized by the good
agreement between the theoretically predicted infrared spectra
of the two compounds and the corresponding matrix-isolation
spectra (see Fig. 4 and Table 3 for MBIOD, and Ref. [19] for equiv-
alent spectroscopic data for MBID), together with the X-ray data
obtained for the substrate, provides the basis for the development
of reliable theoretical models for the MBID ? MBIOD
rearrangement.
C
C
1
AS
AC
7
1.767(1)
1.791
1.479
1.679
1.434
1.434
1.284
1.320
1.442
1.768
1.489
1.691
1.435
1.435
1.386
2
9
1.480(2)
1.653(1)
1.431(1)
1.430(1)
1.292(2)
1.309(2)
1.454(2)
S
7
S
7
S
7
AN
8
@O14
@O16
C
C
C
9
@N
AO15
17AO15
8
9
1.209 (C
–
9
@O15)
Angle (°)
C
C
C
C
C
C
C
2
6
1
3
1
1
1
AC
AC
AC
AC
1
1
2
2
AS
AS
AC
AC
AN
8
7
7
106.9(1)
107.5
130.4
109.4
130.0
94.9
110.2
127.3
113.4
126.5
92.0
111.8
111.8
109.6
109.6
118.7
115.4
108.9
126.5
130.3(1)
109.5(1)
129.7(1)
96.5(1)
111.3(1)
111.3(1)
109.1(1)
108.8(1)
117.8(1)
109.1(1)
118.0(1)
117.4(1)
124.6(1)
116.9(1)
109.5
9
9
AS
7
AS
7
AS
7
Fig.
5
presents the calculated IRC reaction path
@O14
@O16
110.4
110.4
109.3
109.3
119.7
109.9
118.4
117.6
124.0
117.0
105.3
110.0
110.0
[
B3LYP/6-31++G(3df,3pd)], with the optimized geometries of both
minima and the transition state for the intramolecular MBID ?
MBIOD rearrangement model.
In the transition state, the four-membered CAOACAN ring is
significantly distorted, with very long C17AO and C17AN distances
N
N
O
8
AS
AS
14@S
7
@O14
@O16
8
7
7
@O16
AS
@N
AO15
AO15
C
C
C
9
@N
8
7
2
AC
AC
9
8
(
2.162 and 2.222 Å, respectively) and short C
9 9
AO and C AN bonds
2
9
(
1.253 and 1.321 Å). The B3LYP/6-31++G(3df,3pd) computed en-
N
8
@C
9
8
N C
9
@O15: 124.7
ꢀ1
C
9
AO15AC17
15AC17AH18
15AC17AH19
15AC17AH20
C
9
AN AC17: 123.9
8
ergy for the TS is 287.2 and 230.6 kJ mol higher than those for
the product and reactant, respectively, showing that the predicted
activation energy for the (concerted) intramolecular rearrange-
ment is quite high (230.6 kJ mol ). The reconsideration of the
reaction at the B3LYP/6-31+G(d,p) level of theory, used in the com-
putationally more exigent intermolecular models discussed below,
brought similar values, namely an activation energy of
O
O
O
8
N AC17AH18: 106.9
8
N AC17AH19: 110.5
8
N AC17AH20: 110.5
109.5
109.5
ꢀ1
Dihedral angle (°)
C
3
AC
2
AC
1
AS
7
179.9(1)
1.3(1)
0.8
180.0
0.0
0.0
0.0
112.7
ꢀ112.7
180.0
ꢀ67.3
67.3
180.0
0.0
0.0
C
9
AC
2
AC
1
AS
7
S
7
AC
1
AC
6
7
7
7
7
7
7
AH13
AN
@O14
@O16
ꢀ1
C
C
C
C
C
C
C
C
C
C
C
C
C
C
2
2
2
6
6
6
1
1
3
3
1
9
9
2
AC
AC
AC
AC
AC
AC
AC
AC
AC
AC
AS
@N
@N
AC
AN
1
AS
AS
AS
AS
AS
AS
AC
AC
AC
AC
8
ꢀ2.0(1)
111.5(1)
ꢀ115.0(1)
177.1(1)
ꢀ69.5(2)
64.0(2)
0.2(2)
179.5(1)
ꢀ178.3(1)
1.0(2)
0.0
229.2 kJ mol and an energy difference between MBID and MBIOD
equal to 58.1 kJ mol
1
1
1
1
1
2
2
2
2
112.1
ꢀ112.1
180.0
ꢀ67.9
67.9
0.0
180.0
180.0
0.0
ꢀ1
.
The input geometry used to calculate the intermolecular rear-
rangement pathways was taken from the X-ray data (Fig. 6a).
Two MBID molecules from neighboring layers of the crystal were
chosen (Fig. 6b) to form the initial dimeric unit used in the
calculations.
In the first intermolecular model investigated, the rearrange-
ment can be considered as a sequential process in which the first
step is the transfer of the methyl group from one MBID molecule
to the nitrogen atom of the second molecule, giving rise to an an-
ion–cation pair. The initial formation of an anion–cation pair was
previously suggested by Dessolin et al. [11] as the initial step of
oxadiazoles rearrangement, leading to propagation of the reaction
via the methyl transfer between a neutral molecule and an anion or
a cation. A two-step mechanism involving a molecular pair inter-
mediate was also proposed for intermolecular thermal isomerisa-
AN
8
@O14
@O16
9
9
9
9
@N
AO15
@N
AO15
@C
8
0.0
180.0
180.0
0.0
8
7
AN
8
7
7
9
2.1(1)
0.0
0.0
8
AS
AS
@O14
@O16
ꢀ113.1
117.2(1)
ꢀ1.6(2)
179.1(1)
ꢀ178.3(1)
1.1(2)
ꢀ113.7
113.7
0.0
180.0
180.0
0.0
ꢀ114.0
114.0
0.0
8
9
@N
@C
AO15AC17
AO15AC17
8
AS
7
S
7
8
9
AO15
180.0
C
N
2
AC
@C
9
2 9 8
C C N C17: 180.0
8
9
–
influence the geometry of the six-membered ring. Indeed, the
HOMA aromaticity index [28] of the phenyl ring in MBID and
MBIOD (0.992 and 0.995, respectively) are practically identical
and also very close to that of benzene (1.000).
tion
of
methyl
p-dimethylaminobenzenosulfonate
into
zwitterionic p-trimethylammoniumbenzenosulfonate [15].
In our calculations we considered two pathways for the second
methyl transfer: anion–cation (starting from the point in the PES
corresponding to the minimum energy structure after the first
methyl group transfer) and anion–neutral molecule.
3.3. Models for the MBID ? MBIOD rearrangement
As it was previously suggested by Hettler [10], the Chapman-
In the crystal, there are two ‘‘types of pairs” of MBID molecules
0
like rearrangement of MBID takes place intermolecularly [5]. An
intermolecular methyl migration in the crystalline state is facili-
tated if a propitious arrangement of the substrate molecules in
the crystal occurs, as was previously demonstrated in the case of
oxadiazoles [11,12] and methyl p-dimethylaminobenzenosulfo-
nate [13–15]. Under these conditions, the intermolecular methyl
transfer does not require a considerable distortion of the molecules
in the crystal and may become energetically favored. Nevertheless,
the intermolecular methyl transfer in the solid state might also
(see Fig. 6), with different C17ꢂꢂꢂN distances (ca. 4 and 5 Å, respec-
8
tively). In the calculations, during the initial step of the sequential
0
intermolecular rearrangement, the C17ꢂꢂꢂN distance (
1
D ) was de-
8
creased in steps of 0.1 Å, starting from a value equal to the longest
distance of this structural parameter observed in the MBID crystal,
5.0 Å, while all other structural parameters were let to relax. Dur-
0
ing the initial steps (up to a value for the C17ꢂꢂꢂN distance equal ca.
8
3.6 Å, see Fig. 7) of the transfer, the energy of the system is pre-
dicted to fluctuate only slightly, the maximal change of energy

A. Kaczor et al. / Journal of Molecular Structure 892 (2008) 343–352
347
0
0
0
0
0
0
0
0
.6
.4
.2
.0
.6
.4
.2
.0
a
b
0.4
0.3
0.2
0.1
0.0
c
6
0
0
0
0
d
4
2
1800
1600
1400
1200
1000
800
600
Wavenumber/ cm-1
Fig. 4. Infrared spectra of MBIOD: (a) in the molten phase at 210 °C, (b) in the polycrystalline phase after cooling the melt to 20 °C, (c) isolated in the Xe matrix (20 K) and (d)
B3LYP/6-311++G(3df,3pd) calculated spectrum for the monomer.
being no more than 1.6 kJ mol 1. As this initial shortening of the
distance does not significantly affect the energy of the system, it
is suggested that up to some point the molecules in the pair behave
ꢀ
significant deviation from planarity, with the N
ꢂꢂꢂC0 ꢂꢂꢂO0 ꢂꢂꢂC0
8
1
7
15
9
0
0
dihedral and N
8
ꢂꢂꢂC ꢂꢂꢂO bond angles being 6.5° and 84.5°, respec-
1
7
15
tively. The product is a structure in which two MBIOD molecules
are arranged one on the top of the other (see Fig. 8). In turn, the
activation energy for the anion–neutral molecule methyl transfer
as practically non-interacting.
0
Along the methyl transfer, the C17ꢂꢂꢂN
8
(D
2
) distance shortens
reduces to
.9 Å and, from there, it is quite conserved (fluctuating in the range
.6–3.5 Å) until becomes ca. 2.1 Å. Then, a further shortening of
occurred (down to 3.245 Å) when the methyl group was trans-
ferred and becomes 1.437 Å. During the initial steps of methyl
transfer the model structure started to flatten but after the transi-
tion state was reached ( = 1.889 Å; see Fig. 6c) the two molecular
ꢀ1
down from 5.0 Å to approximately 3.6 Å while
D
1
process is even higher (ca. 314 kJ mol ).
3
3
As a whole, the sequential model for intermolecular transfer of
the methyl group, both for the anion–cation and anion–neutral
molecule second step, is predicted to have an activation energy
much higher than that corresponding to the intramolecular rear-
rangement. These results clearly demonstrate that the reaction
cannot occur this way.
D
1
D
2
D
1
D
1
units got significantly distorted out-of-plane relatively to each
other.
The second intermolecular model investigated assumes that
simultaneous one-step transfer of two methyl groups takes place.
This mechanism could be expected to be favored if proper align-
ment of molecules in the crystal occurred. This type of mechanisms
was suggested for the solid state methyl transfer of 6-methoxy-
3,5-dimethyl-tetrahydrotriazine-2,4-dione, based on the crystal
structure for this compound obtained by X-ray diffraction analysis
[18,29–32]. In the case of MBID, the activation energy for the
simultaneous methyl transfer process was predicted to be ca.
The energy profile corresponding to the above described single
methyl transfer process is given in Fig. 7. An activation energy of
ꢀ1
ca. 174 kJ mol with respect to the reactants’ minimum (the fully
optimized MBID dimer, where both CAN distances are equal to
3
.689 Å, see Fig. 7) was obtained. Comparison with the activation
ꢀ1
energy of the methyl intramolecular transfer (229.2 kJ mol
)
shows that this first step in the sequential intermolecular model
is favorable on energy grounds. However, as mentioned above, in
this sequential model, the first methyl group transfer corresponds
only to an introductory step for the reaction and requires the sub-
sequent transfer of a second methyl group in order to obtain the
final neutral MBIOD product.
ꢀ1
ꢀ1
286 kJ mol [B3LYP/6-31+G(d,p)] (see Fig. 7), i.e. ca. 55 kJ mol
higher than that estimated for the intramolecular process. In fact,
the MBID molecules in the crystal lattice are not favorably oriented
to allow for the simultaneous methyl transfer to take place effi-
ciently, since interaction between molecules from two different
layers separated by a large distance (ca. 5 Å) would be required.
In addition, steric hindrance due to repulsive interactions between
the two methyl groups being simultaneously transferred (both of
them placed in-between the two rearranging MBID molecules;
see Fig. 7) also contributes to make this process disfavored. In con-
sonance with this, the MBID ? MBIOD rearrangement does not
show the characteristics of a typical topochemically assisted reac-
tion. The reaction in the solid state is slower than in the liquid and
is considerably accelerated by melting [5].
Two different pathways for the second methyl group transfer
were investigated: anion–cation (starting from the point in the
PES corresponding to the minimum energy structure after the first
methyl group transfer) and anion–neutral molecule. The obtained
energy profiles for these two pathways are presented in Fig. 8.
Both considered reaction pathways for the transfer of the sec-
ond methyl group are predicted to require quite high energy. The
activation energy for the anion–cation methyl transfer is ca.
ꢀ
1
1
2
75 kJ mol , i.e. the energy of the transition state stays ca.
ꢀ1
53 kJ mol above that of the initial reactant (MBID dimer; see
The third model investigated (‘‘quasi-simultaneous” model) re-
sulted from a preliminary exploration of the potential energy sur-
also Fig. 7). In this transition state, the anion–cation pair shows a

3
48
A. Kaczor et al. / Journal of Molecular Structure 892 (2008) 343–352
Table 3
Assignment of experimental IR spectra of matrix-isolated (solid Xe, 20 K), polycrystalline and melted MBIOD in KBr pellet (500–4000 cm spectral range)
ꢀ
1
Calculateda B3LYP/6-311++G(3df,3pd)
Experimental
Approximate assignmentb
Xe, 20 K
Polycrystalline
Melt
(cm 1)
ꢀ
I (km mol
ꢀ1
)
m
(cm 1)
ꢀ
Ic
m
(cm 1)
ꢀ
Ic
m
(cm 1)
ꢀ
Ic
m
3
3
3
3
3
3
3
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
9
9
8
8
7
7
7
6
6
6
5
5
5
169
166
154
141
106
073
008
757
612
606
494
477
474
469
431
346
343
322
278
217
186
172
169
142
129
058
028
009
76
70
96
79
93
57
43
86
76
06
79
33
05
5.4
0.4
3.5
1.5
0.1
n.obs
n.obs
n.obs
n.obs
2977
2943
2905
–
–
–
–
w
w
w
S
3095
3084
3072
3035
3016
2952
2925
1739
1595
m
sh
w
w
w
w
w
S
3095
3084
3074
3030
w
w
w
w
m
m
m
m
m
m
m
m
m
m
d
CAH6R
CAH6R
CAH6R
CAH6R
CAH
CAH
CAH
C@O
M
M
M
7.1
2950
2925
1737
1593
w
sh
S
28.6
367.4
5.1
5.3
18.0
6.6
1742
n.obsd
–
–
w
w
CAC6R
CAC6R
n.obsd
1469/1467
1463
1458
1450
1411
1352/1351/1348
1343/1336
1319/1314/1312
1268
1225/1222/1212/1208/1206
1199/1197/1190/1192
1181
1176/1171
1161/1159
1127
1066/1061
1018
m/sh
sh
w
w
1469
1460
m
m
1464
1456
m
sh
M
dCACAH6R
d
M
6.7
8.3
dCACAH6R
21.2
208.0
28.4
257.2
41.5
41.5
118.7
5.4
131.8
0.6
7.3
15.7
0.8
0.0
0.6
65.7
0.1
31.4
9.5
36.3
18.6
4.3
20.8
31.5
62.7
14.8
25.6
m
1422
1328
m
S
1422/1417
1337
w/sh
S
d
m
m
m
M
S/S/m
w/m
m/S/sh
m
w/w/m/w/w
m/m/S/S
m
sh/S
w/w
w
w/w
w
–
w/sh
m/sh/m/sh
–
S@O
CAC6R
C AN, mC17AN
9
1315
1276
1221/1214
1187
1174/1160
S
m
m/sh
S
S/sh
1308
1264
1223/1213
1183
1176/1162
S
m
m/sh
S
sh/sh
dCACAH6R
d
m
M
,
m
CAC5R
S@O
dCACAH6R
m
d
S@O
1132
1127
1057
1022
999
978
968
n.obs
890
785
sh
w
m
w
w
m
sh
–
1137
1125
1059
1019
999
971
960
n.obs
882
787
sh
w
m
w
w
m
sh
–
M
dCACAH6R
dCACAC6R
,
mCAC6R
m
c
c
m
c
m
c
s
CAC6R
n.obs
6R
984/978
976/975/971/969
n.obs
6R
C17AN,
m
C
9
AN
6R
885
m
m
w
S
m
w
S
M
NAS, d , dXAC@O
CAC(@O)AN, c
6R
, c
6R 6R
786/784/782
751/750/749/748
747
688/685
676/673
611/610
588/586
531
w/w/w
m/m/m/m
sh
w/w
w/m
m/sh
m/m
m
751
749
dCACAC6R
dCACAC6R
s
6R
dXAS@O, dXAC@O
c
c
685
675
609
591
528
512
m
m
m
S
m
m
685
675
610
587
532
508
w
m
m
S
m
m
CAS(@O)AN
CAS(@O)AN, s
6R
510
m
dCAXAX5R
a
Frequencies scaled uniformly by the factor of 0.987 [15].
b
6
R and 5R refer to the six- and five-membered ring, respectively; M stands for methyl group;
m, d, c and s indicate stretching, bending, rocking and torsion modes,
respectively; X designates either a C, S or N atom.
c
Experimental intensities are provided as qualitative: S, strong; m, medium; w, weak; sh, shoulder.
Not identified probably due to overlap with the bands assigned to monomeric water.
d
face of the system in aiming to find the minimum reaction
pathway for the intermolecular transfer. This preliminary
inspection of the PES was carried out at a low-level of theory
The activation energy for this ‘‘quasi-simultaneous” methyl
ꢀ1
transfer process is ca. 226 kJ mol , in respect to the fully opti-
mized substrate. The predicted energy is then considerably lower
than the energy of all other considered intermolecular pathways
[
B3LYP/3-21G(d,p)], since it required the calculation of a two-
dimensional energy grid resulting from the PES scan along the
and
coordinates. This scan was performed by varying the CꢂꢂꢂN
distances in the 2.8–1.8 Å range, in steps of 0.1 Å; see Fig. 9.
ꢀ
1
D
1
of rearrangement and slightly lower (ca. 3.2 kJ mol ) than the
activation energy of the intramolecular transfer. The significant
energetic preference of the ‘‘quasi-simultaneous” mechanism of
rearrangement over the other intermolecular mechanisms results
from the fact that in its transition state a more favorable equilib-
rium is attained between the two most relevant factors determin-
D
2
The results of this preliminary PES study confirm that the most
energetically favorable methyl transfer pathway is indeed along
the CꢂꢂꢂN coordinates, with
D
1
first varying down to 1.8 Å and then
is kept at that value.
0
0
D
2
being shortened while
D
1
ing the activation barrier: the value of the N
8
ꢂꢂꢂC ꢂꢂꢂO angle (a
1
7
15
Following the indications provided by the preliminary examina-
more linear angle favors energetically the process) and the steric
repulsion between the methyl groups. As described before, the first
factor favors the simultaneous transfer of the methyl groups (for
tion of the PES along the CꢂꢂꢂN coordinates, calculations performed
at the B3LYP/6-31+G(d,p) level of theory were carried out: starting
0
0
with the geometry extracted from the X-day data, the
D
1
distance
this mechanism the N
8
ꢂꢂꢂC ꢂꢂꢂO angle at the transition state is
1
7
15
was initially changed down to 1.889 Å and from this point the
ca. 166°), but the second factor was highly unfavorable. In turn,
the repulsion between the methyl groups is minimized in the
sequential mechanism, which however is disfavored by the strong
shortening of the
frozen; when the
D
D
2
distance was conducted while was kept
D
1
2
distance reduces to 1.5 Å, the D distance
1
0
0
was then shortened down to 1.449 Å while
Fig. 10).
D
2
was allowed to relax
deviation from linearity of the N
8
ꢂꢂꢂC ꢂꢂꢂO angle (ca. 85°). On the
1
7
15
(
other hand, for the quasi-simultaneous mechanism, the

A. Kaczor et al. / Journal of Molecular Structure 892 (2008) 343–352
349
N
8
1
.321
2.222
C9
1
.253
C17
2
.162
O
1
5
250
200
150
100
TS [220.0]
50
0
-50
1.4
1.6
1.8
2.0
2.2
2.4
2.6
2.8
3.0
C -O (Å)
15
17
1
.458
N8
N8
1
.289
1
.388
C
17
C9
1
.323
O1
C
9
1
.213
5
O15
1
.444
C1
7
MBID [0.0]
MBIOD [-56.6]
Fig. 5. The predicted IRC reaction path [B3LYP/6-31++G(3df,3pd)] for the intramolecular rearrangement of MBID to MBIOD. Optimized geometries for the substrate, product
ꢀ
1
and the TS are shown, with the key bond lengths presented explicitly. The zero-point corrected energies relative to the substrate (in kJ mol ) are given in brackets. The
numbering of the relevant atoms is shown (bold).
Fig. 6. (a) MBID pair arrangements in the crystal, (b) the input structure used in the theoretical intermolecular rearrangement models and (c) the geometry of the model at
the transition state after flattening of the structure.

3
50
A. Kaczor et al. / Journal of Molecular Structure 892 (2008) 343–352
single transfer
simultaneous transfer
[
286.1]
300
200
100
0
[
174.0]
[74.9]
[
0.0]
-100
[
-104.2]
5.0
4.5
4.0
3.5
17 N’
8
3.0
Å)
2.5
2.0
1.5
…
C
(
Fig. 7. Calculated reaction paths [B3LYP/6-31+G(d,p)] for the intermolecular rearrangement of MBID to MBIOD: (h) the first methyl group transfer in the sequential model;
ꢀ
1
(
j) the simultaneous transfer of two methyl groups. Relative energies (in kJ mol ) of reactant, product and transition state species are given in brackets.
anion-neutral transfer
anion-cation transfer
[313.7]
300
200
100
[
252.8]
[
77.5]
0
[0.0]
[
[
-22.4]
-100
-107.0]
3.5
3.0
2.5
2.0
1.5
C’17…N
8
(Å)
Fig. 8. Calculated reaction paths [B3LYP/6-31+G(d,p)] for the second step of the sequential intermolecular rearrangement of MBID into MBIOD: (h) via anion–cation; (d) via
ꢀ
1
anion–neutral molecule (d). The energies (kJ mol ) relative to the fully optimized substrate (for anion–cation reaction) and to the starting structure (for the anion–neutral
reaction) are given in brackets.
ꢂꢂꢂC ꢂꢂꢂO0 angle is ca. 130° in the transition state and the repul-
0
weak, since in this case the hydrogen atoms are directed in oppo-
sition to each other, conversely to what happens in the case of the
N
8
17
15
sive interactions between the methyl groups are also relatively

A. Kaczor et al. / Journal of Molecular Structure 892 (2008) 343–352
351
22 ...888
-
50
0
5
0
2
2...666
1
1
2
2
3
3
00
50
00
50
00
50
2
2...444
2
2...222
2
.0
2
.. 00
1
.8
1
.. 88
1
.8
2.0
2 .. 00
2.2
2 .. 22
2.4
2 .. 44
2.6
2 .. 66
2.8
2 .. 88
1
.. 88
Δ1 (C17…N’ )/ Å
8
Fig. 9. Two-dimensional potential energy profile for MBID ? MBIOD rearrangement (energy in kJ molꢀ1).
[226.0]
2
50
00
50
00
2
1
1
50
[
0.0]
0
-50
-
100
4.0
1.0
3.5
[
2
-104.2]
1.5
3.0
.0
2.5
2.5
3.0
2.0
3.5
1
.5
4.0
Fig. 10. The predicted reaction path [B3LYP/6-31+G(d,p)] for the ‘‘quasi-simultaneous” intermolecular methyl transfer (MBID ? MBIOD). The energies relative to the fully
optimized substrate are given in brackets.
simultaneous transfer, where the methyl groups are in close prox-
imity and interact repulsively via four hydrogen atoms; compare
Figs. 7 and 10).
these two factors (unfavorable deviation from linearity of the
O
15ꢂꢂꢂC ꢂꢂꢂN17 angle and favorable lack of hindrance in the TS of
9
the intramolecular process) result in the comparable activation en-
ergy of the intra- and ‘‘quasi-simultaneous” intermolecular
rearrangements.
Compared to the activation energy associated to intramolecular
transfer, the activation energy of the ‘‘quasi-simultaneous” inter-
ꢀ1
molecular process is only slightly lower (ca. 3.2 kJ mol ), although
We believe that MBID is a boundary case, for which the prefer-
ence towards an inter- over intramolecular process exists. It is ex-
pected that the steric hindrance due to a substituent bigger than
the methyl group would significantly increase the energy of the
transition state (see Fig. 10), thus making the intermolecular rear-
an upmost bending of the O15ꢂꢂꢂC ꢂꢂꢂN17 angle (60.8°) is predicted
9
for the transition state of the intramolecular rearrangement. How-
ever, the intramolecular process is the only one in which the tran-
sition state is not sterically crowded. The conjugated effects of

3
52
A. Kaczor et al. / Journal of Molecular Structure 892 (2008) 343–352
rangement improbable on energetic grounds (as was noticed be-
fore, the classical Chapman rearrangement is assumed to be intra-
molecular for aryl imidates) [6–8]. The reaction is not
topochemically controlled, therefore, it is expected that a prerequi-
site for the transfer is a slight reorganization of the molecules in
the lattice with the shortening of both CAN distances (see Fig. 7),
allowing the subsequent ‘‘quasi-simultaneous” transfer of the
methyl groups. The computations suggest that the initial process
of molecular reorganization does not require significant energy.
However, the rigidity of the lattice, not taken into account in our
dimeric-type models, shall definitely hinder in some amount this
mobility strongly increases and a proper alignment of the substrate
molecules may be more easily attained.
Acknowledgements
Calculations were done at the Academic Computer Center
‘‘Cyfronet”, Krakow, Poland (Grant KBN/SGI_ORIGIN_2000/UJ/044/
1999), which is acknowledged for computing time. The research
was supported by the Portuguese Fundação para a Ciência e a Tecn-
ologia (Grant FCT #SFRH/BPD/26590/2006 and Projects POCI/QUI/
59019/2004, POCI/QUI/58937/2004 and PTDC/QUI/67674/2006).
AGZ is member of the research career Conicet (National Research
Council, Argentina). Authors wish to acknowledge Dr. Marcin And-
rzejak for the helpful ideas contributing to this work.
‘
‘pre-orientation” required to start the process, thus being a deci-
sive factor in explaining the observed greater facility of MBID to
undergo the rearrangement after melting [5], when the molecular
mobility strongly increases and a proper alignment of the substrate
molecules may be more easily attained.
References
4
. Conclusions
[1] N.C.P. Araújo, A.F. Brigas, M.L.S. Cristiano, L.M.T. Frija, E.M.O. Guimarães, R.M.S.
Loureiro, J. Mol. Catal. A Chem. 215 (2004) 113.
[
2] L.M.T. Frija, M.L.S. Cristiano, E.M.O. Guimarães, N.C. Martins, R.M.S. Loureiro,
J.F. Bickley, J. Mol. Catal. A Chem. 242 (2005) 241.
The mechanism for the thermal Chapman-type rearrangement
ꢀ1
of MBID towards its energetically lower (ca. 60 kJ mol ) tauto-
mer, MBIOD, was investigated in detail. Based on the knowledge
of the structure of the substrate and product under different
experimental conditions, various models for the rearrangement
reaction were evaluated through quantum chemical theoretical
methods.
[3] N.C.P. Araújo, P.M.M. Barroca, J.F. Bickley, A.F. Brigas, M.L.S. Cristiano, R.A.W.
Johnstone, R.M.S. Loureiro, P.C.A. Pena, J. Chem. Soc., Perkin Trans. 1 (2002)
1213.
[
4] A.F. Brigas, R.A.W. Johnstone, Tetrahedron Lett. 31 (1990) 5789.
[5] R. Almeida, A. Gómez-Zavaglia, A. Kaczor, M.L.S. Cristiano, M.E.S. Eusébio,
T.M.R. Maria, R. Fausto, Tetrahedron 64 (2008) 3296.
[
[
6] H.M. Relles, J. Org. Chem. 33 (1968) 2245.
7] O.H. Wheeler, F. Roman, O. Rosado, J. Org. Chem. 34 (1969) 966.
Comparison of the predicted energetics for the various studied
models indicates that among the intermolecular rearrangements
investigated only that involving the ‘‘quasi-simultaneous” transfer
of two methyl groups is energetically favored relatively to the
intramolecular migration model. Both the sequential and simulta-
neous mechanisms for methyl transfer were found to require con-
[8] K.B. Wiberg, B.I. Rowland, J. Am. Chem. Soc. 77 (1955) 2205.
[9] D.J. Brown, R.V. Foster, J. Chem. Soc. (1965) 4911.
10] H. Hettler, Tetrahedron Lett. 15 (1968) 1793.
11] M. Dessolin, O. Eisenstein, M. Golfier, T. Prangé, P. Sautet, J. Chem. Soc., Chem.
Commun. (1992) 132.
[12] M. Dessolin, M. Golfier, J. Chem. Soc., Chem. Commun. (1986) 38.
13] C.N. Sukenik, J.A.P. Bonapace, N.S. Mandel, P.-Y. Lau, G. Wodo, R.G. Bergman, J.
Am. Chem. Soc. 99 (1977) 851.
[
[
[
ꢀ1
siderably more energy to take place (27 and 80 kJ mol for anion–
cation and anion–neutral molecule sequential mechanism, respec-
[
14] J.A.R.P. Sarma, J.D. Dunitz, Acta Crystallogr. B 46 (1990) 780.
[15] A. Gavezzotti, M. Simonetta, Nouv. J. Chim. 2 (1977) 69.
[16] H. Taycher, M. Botoshansky, V. Shteiman, M. Kaftory, Supramol. Chem. 13
ꢀ
1
tively, and 60 kJ mol for the simultaneous process).
(
2001) 181.
According to the ‘‘quasi-simultaneous” mechanism, the transfer
of the second methyl group starts when the first methyl group is
not yet fully transferred (when the CAN distance equals ca.
[
[
[
17] M. Greenberg, V. Shteiman, M. Kaftory, Acta Crystallogr. B Struct. Sci. 57 (2001)
428.
18] E. Handelsman-Benory, M. Botoshansky, M. Greenberg, V. Shteiman, M.
Kaftory, Tetrahedron 56 (2000) 6887.
19] A. Kaczor, R. Almeida, A. Gómez-Zavaglia, M.L.S. Cristiano, R. Fausto, J. Mol.
Struct. 876 (2008) 77.
20] G.M. Sheldrick, Acta Crystallogr. A 64 (2008) 112.
21] A.D. Becke, Phys. Rev. A 38 (1988) 3098.
ꢀ1
1
.9 Å), the activation energy of the process being ca. 226 kJ mol ,
slightly below to that corresponding to the intramolecular rear-
rangement. Two factors were found to be fundamental in control-
ling the energy of the methyl group transfer process: the deviation
[
[
[
22] C.T. Lee, W.T. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
[23] C.Y. Peng, P.Y. Ayala, H.B. Schlegel, M.J. Frisch, J. Comput. Chem. 17 (1996) 49.
[24] C.Y. Peng, H.B. Schlegel, Isr. J. Chem. 33 (1993) 449.
0
0
from linearity of the N
8
ꢂꢂꢂC ꢂꢂꢂO angle and the steric hindrance in
1
7
15
the transition state structure. The ‘‘quasi-simultaneous” intermo-
lecular methyl transfer was found to be the unique mechanism
in which these two factors are well balanced (e.g., the simultaneous
methyl transfer mechanism and the intramolecular mechanism are
very much disfavored by the first factor, while the sequential
mechanism is strongly destabilized by the second one). The ob-
served reaction is not topochemically controlled due to long dis-
tance in the crystal (ca. 5 Å) between the nitrogen and carbon
atoms directly involved in the rearrangement. Therefore, the pre-
orientation of the molecules in the lattice is necessary in order to
allow for the methyl group transfer to take place. Although the
computations predict that this process is not energetically expen-
sive per se, the rigidity of the crystalline lattice, not taken into ac-
count in our dimeric-type models, shall hinder to some extent this
[
25] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman,
J.A. Montgomery Jr., T. Vreven, K.N. Kudin, J.C. Burant, J.M. Millam, S.S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G.A.
Petersson, H. Nakatsuji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa,
M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J.E. Knox,
H.P. Hratchian, J.B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R.E.
Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C. Pomelli, J.W. Ochterski, P.Y.
Ayala, K. Morokuma, G.A. Voth, P. Salvador, J.J. Dannenberg, V.G. Zakrzewski, S.
Dapprich, A.D. Daniels, M.C. Strain, O. Farkas, D.K. Malick, A.D. Rabuck, K.
Raghavachari, J.B. Foresman, J.V. Ortiz, Q. Cui, A.G. Baboul, S. Clifford, J.
Cioslowski, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaromi, R.L.
Martin, D.J. Fox, T. Keith, M.A. Al-Laham, C.Y. Peng, A. Nanayakkara, M.
Challacombe, P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, C. Gonzalez, J.A.
Pople, GAUSSIAN 03, Revision C.02, Gaussian, Inc., Wallingford, CT, 2004.
26] P. Pulay, G. Fogarasi, F. Pang, J.E. Boggs, J. Am. Chem. Soc. 101 (1979) 2550.
[
[27] J.M.L. Martin, C. Van Alsenoy, GAR2PED, University of Antwerp, 1995.
[
[
[
28] J. Kruszewski, T.M. Krygowski, Tetrahedron Lett. 36 (1972) 3839.
29] M.L. Tosato, J. Chem. Soc., Perkin Trans. 2 (1979) 1371.
30] M.L. Tosato, J. Chem. Soc., Perkin Trans. 2 (1984) 1593.
‘
‘pre-orientation” required to start the process, thus being a deci-
sive factor in explaining the observed greater facility of MBID to
undergo the rearrangement after melting [5], when the molecular
[31] M.L. Tosato, L. Soccorsi, J. Chem. Soc., Perkin Trans. 2 (1982) 1321.
[32] R.J. Boyd, C. Kim, Z. Shi, N.W.S. Weinberg, J. Am. Chem. Soc. 115 (1993) 10147.