An experimental and theoretical study of the structure of Lamotrigine in its neutral and protonated forms: Evidence of Lamotrigine enantiomers

Article abstract of DOI:10.1016/j.tet.2014.02.075

The energies, geometries, and NMR chemical shifts have been calculated at the B3LYP/6-311++G(d,p) level for 17 structures of the anticonvulsant drug Lamotrigine and 29 structures of protonated Lamotrigine, including tautomers and E/Z isomers of the imino groups. The calculations were compared with solid state (X-ray and CPMAS NMR) and solution experimental results both reported in the literature and determined in this work. The conclusion is that Lamotrigine exists as the diamino tautomer and that its protonation takes place on the N2 atom. Using ABTE and/or deuterated ABTE as chiral solvating agent, it has been demonstrated for the first time by NMR in solution that Lamotrigine is a racemate of rapidly interconverting enantiomers. The crystal structure of two new solvated salts of Lamotrigine, both saccharinates, has been determined. Both salts present the same arrangement in chains of Lamotrigine and saccharinate joined by hydrogen bonds and stacking interactions. No isostructurality is present because of the different arrangement of the chains in both crystal structures.

Full text of DOI:10.1016/j.tet.2014.02.075

Tetrahedron 70 (2014) 2784e2795
Contents lists available at ScienceDirect
Tetrahedron
journal homepage: www.elsevier.com/locate/tet
An experimental and theoretical study of the structure of
Lamotrigine in its neutral and protonated forms: evidence of
Lamotrigine enantiomersq
,
a
b
c
c
c
Ibon Alkorta a , Jose Elguero , Anna Font , Judit Galcera , Ignasi Mata , Elies Molins ,
*
ꢀ
Albert Virgili b
Instituto de Química Medica (IQM-CSIC), Juan de la Cierva, 3, E-28006 Madrid, Spain
Departament de Quìmica, Universitat Autonoma de Barcelona, Bellaterra, 08193 Cerdanyola, Barcelona, Spain
Institut de Ciencia de Materials de Barcelona (ICMAB-CSIC), Campus UAB, 08193 Bellaterra, Spain
a
ꢀ
b
ꢁ
c
ꢁ
a r t i c l e i n f o
a b s t r a c t
Article history:
The energies, geometries, and NMR chemical shifts have been calculated at the B3LYP/6-311þþG(d,p)
level for 17 structures of the anticonvulsant drug Lamotrigine and 29 structures of protonated Lamo-
trigine, including tautomers and E/Z isomers of the imino groups. The calculations were compared with
solid state (X-ray and CPMAS NMR) and solution experimental results both reported in the literature and
determined in this work. The conclusion is that Lamotrigine exists as the diamino tautomer and that its
protonation takes place on the N2 atom. Using ABTE and/or deuterated ABTE as chiral solvating agent, it
has been demonstrated for the first time by NMR in solution that Lamotrigine is a racemate of rapidly
interconverting enantiomers. The crystal structure of two new solvated salts of Lamotrigine, both sac-
charinates, has been determined. Both salts present the same arrangement in chains of Lamotrigine and
saccharinate joined by hydrogen bonds and stacking interactions. No isostructurality is present because
of the different arrangement of the chains in both crystal structures.
Received 10 January 2014
Received in revised form 18 February 2014
Accepted 26 February 2014
Available online 6 March 2014
Keywords:
Lamotrigine
X-ray crystallography
Dynamic NMR
Tautomerism
Axial chirality
GIAO calculations
Ó 2014 Elsevier Ltd. All rights reserved.
1. Introduction
From that review,9 and from a search in the literature, it ap-
pears that concerning Lamotrigine structural aspects, there is
There is no need for a long introduction justifying the impor-
tance of Lamotrigine 1 (Lamictal, GlaxoSmithKline) since it is the
treatment of choice for two serious illnesses: epilepsy1e4 and bi-
polar disorder.5e8 It is what it is known by the name of ‘stand alone’
drug. There is a 2012 review on Lamotrigine that covers all its
known properties at that time.9
much information about X-ray crystallography (including poly-
morphism), some NMR data used to characterize it but almost
none concerning computational studies. We have found only two
papers. In the first one a molecular modeling study of Lamo-
trigine/b-cyclodextrin inclusion complex was carried out using
the hybrid ONIOM2 method (the neutral diamino tautomer was
q
used for the calculations).10 In the second one, the IR and Raman
spectra were calculated at the B3LYP/6-31G(d,p) level for the same
tautomer.11
Dedicated to Professor Christian Roussel for his contributions to chirality
* Corresponding author. Tel.: þ34 91 562 29 00; fax: þ34 91 564 48 53; e-mail
address: ibon@iqm.csic.es (I. Alkorta).
URL: http://www.iqm.csic.es/are
0040-4020/$ e see front matter Ó 2014 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tet.2014.02.075

I. Alkorta et al. / Tetrahedron 70 (2014) 2784e2795
2785
The purpose of the present paper is to study theoretically the
structure (tautomerism and isomerism) of Lamotrigine and its
cation (relative energies) and calculate the chemical shifts of all
structures. To support the calculations some new structures of
Lamotrigine salts will be reported as well as the experimental de-
termination of the barrier to enantiomerization.12 In some papers,
the name lamotriginium is used for protonated Lamotrigine. As
with all international nonproprietary names (INNs) the WHO rec-
tautomers has been proposed.14,15 Note that the presence of two
tautomeric groups could favor the exo double-bonded tautomers,
for instance whilst 3(5)-pyrazolinones are mixtures of oxo and
hydroxy tautomers, 3,5-pyrazolidindiones are always dioxo ones.14
Therefore, amino/imino tautomers were expected, a priori, to be
relatively stable for Lamotrigine.
All possible isomers (excluding structural ones), both realistic
and highly improbable, are represented in Fig. 1 and the corre-
H
N
H
N
H
N
H
N
N
NH2
N
N
Cl
N
Cl
Cl
N
Cl
Cl
Cl
Cl
N
H
Cl
Cl
N
Cl
N
Cl
Cl
Cl
N
Cl
Cl
Cl
N
Cl
Cl
N
H
NH2
NH2
NH2
NH2
a
bZ
cE
bE
H
N
H
N
NH2
NH2
N
NH2
H
N
N
H
N
N
H
N
H
N
Cl
Cl
Cl
Cl
N
N
N
N
N
N
Cl
Cl
Cl
Cl
N
H
N
H
N
NH2
H
N
cZ
dZ
eE
Cl
dE
H
N
H
N
H
N
H
N
N
H
N
N
NH2
H
N
H
Cl
Cl
N
N
Cl
Cl
N
N
N
H
H
H
N
H
N
N
H
H
fEZ
eZ
fEE
fZE
H
N
H
N
H
N
H
N
H
H
N
N
H
N
N
N
N
N
Cl
Cl
N
H
NH2
N
H
NH2
gZ
gE
N
fZZ
NH2
H
N
NH2
Cl
N
N
N
H
N
N
N
H
hZ
hE
Fig. 1. The 17 isomers of Lamotrigine 1.
ommends to use always the INN without any modification: ‘Names
for different salts or esters of the same active substance should
differ only with regard to the inactive moiety of the molecule’.13
We also want to determine if Lamotrigine axial chirality could be
calculated and if enantiomerization barrier is high enough to be
measured. The fact that Lamotrigine is a racemate has never been
sponding dipole moments and relative energies are reported in
Table 1.
The differences in energy are very important to the point that
only tautomer 1a (diamino) should exist; the next in energy is the
1bE (3-imino/5-amino) that lies 38.0 kJ molꢀ1 higher. Note the
large influence of the E/Z isomerism on the energies.
reported. Lamotrigine complexes with S,S-ABTE [(S,S)-a,
a0–bis(tri-
fluoromethyl)-9,10-anthracenedimethanol] could be of different
energy and consequently the populations of both diastereoisomeric
complexes could be not exactly 50/50. Considering that this dif-
ference should be very low and the interconversion between both
complexes through free compounds very fast, we expected an equal
distribution of enantiomers.
Table 1
Dipole moments (
Fig. 1
m
, Debye) and relative energies (kJ molꢀ1) of the compounds of
Comp. Dipole Erel
Comp. Dipole Erel
Comp. Dipole Erel
1a
3.16
5.03
5.32
7.51
5.69
5.68
0.0 1dZ
38.0 1eE
62.9 1eZ
135.8 1fEE
102.6 1fEZ
70.3 1fZE
6.13
4.09
5.97
2.53
2.83
4.34
55.8 1fZZ
3.16
7.99
8.51
4.84
4.18
74.3
1bE
1bZ
1cE
1cZ
1dE
68.4 1gE
85.5 1gZ
63.6 1hE
76.3 1hZ
66.2
136.7
139.8
110.8
90.4
2. Results and discussion
2.1. Computational chemistry
2.1.1. Energies
2.1.1.2. Protonated Lamotrigine. All possible structures, tauto-
mers as well as E/Z isomers, are represented in Fig. 2 whereas the
corresponding relative energies are reported in Table 2.
2.1.1.1. Neutral Lamotrigine. Little is known about the tautom-
erism of amino-1,2,4-triazines; although the amino tautomers ap-
pear to be largely predominant, the participation of imino

2786
I. Alkorta et al. / Tetrahedron 70 (2014) 2784e2795
H
N
NH2
H
H
N
NH2
Cl
N
NH2
N
NH3
N
NH2
Cl
N
Cl
Cl
Cl
Cl
N
Cl
Cl
Cl
Cl
N
Cl
Cl
Cl
N
Cl
Cl
N
N
Cl
Cl
Cl
Cl
N
Cl
N
Cl
N
Cl
N
H
NH2
NH2
NH2
NH2
NH3
ae
ac
ad
aa
ab
H
N
H
N
H
N
H
N
H
N
H
N
H
N
N
H
Cl
N
H
N
H
H
N
N
H
Cl
N
N
Cl
N
N
Cl
N
Cl
N
N
N
H
NH2
baE
NH2
bcE
NH3
beE
NH2
NH2
baZ
bcZ
H
N
H
N
H
N
H
N
H
N
H
N
H
H
N
N
H
H
N
N
N
N
N
H
Cl
N
Cl
Cl
N
Cl
Cl
H
Cl
Cl
N
Cl
N
N
N
Cl
N
H
H
H
NH2
caZ
NH2
NH3
beZ
NH2
NH2
caE
cbE
cbZ
H
N
H
N
H
N
H
N
NH2
Cl
H
NH2
N
N
N
H
NH2
H
N
N
H
Cl
N
H
Cl
N
N
N
N
N
Cl
N
N
H
N
daE
N
H
NH3
ceE
N
dcE
NH3
H
daZ
ceZ
H
N
H
N
H
N
NH2
H
N
NH2
NH3
Cl
NH3
Cl
H
N
NH2
Cl
Cl
N
Cl
N
H
Cl
Cl
N
Cl
N
Cl
N
Cl
Cl
N
H
N
Cl
H
N
N
H
Cl
H
N
H
H
N
N
N
N
ddE
N
N
N
H
eaZ
ddZ
dcZ
eaE
H
N
NH2
N
N
NH3
H
N
N
NH3
Cl
NH2
N
Cl
N
Cl
N
N
N
N
H
Cl
H
N
Cl
H
N
H
H
N
H
H
ebZ
edZ
edE
ebE
Fig. 2. The 29 isomers of protonated Lamotrigine.
Table 2
Relative energies (kJ molꢀ1) of the isomers of Fig. 2
Comp.
Erel
21.2
0.0
88.7
120.3
136.7
174.6
203.4
99.5
Comp.
Erel
Comp.
1HþdcZ
Erel
1Hþaa
1HþbeZ
1HþcaE
1HþcaZ
1HþcbE
1HþcbZ
1HþceE
1HþceZ
1HþdaE
1HþdaZ
1HþdcE
208.7
169.2
141.0
99.5
115.4
252.0
249.8
97.1
107.1
94.2
115.4
256.7
279.7
1Hþab
1Hþac
1HþddE
1HþddZ
1HþeaE
1HþeaZ
1HþebE
1HþebZ
1HþedE
1HþedZ
1Hþad
1Hþae
90.6
1HþbaE
1HþbaZ
1HþbcE
1HþbcZ
1HþbeE
334.0
304.1
197.6
172.5
94.2
90.6
194.1
Fig. 3. The molecular electrostatic potential (MEP) of the isolated Lamotrigine. The red
and blue isosurfaces correspond to þ0.06 and ꢀ0.06 au, respectively.
The most stable structure is that protonated on the N2 ring ni-
trogen,1Hþab, the only one found in the solid state (see below). The
following one is 1Hþaa (21.2 kJ molꢀ1) protonated on N1; all the
remaining cations are much more unstable; it was known that
aminoazines protonated on ring nitrogen atoms.14 To this pro-
pensity of N1 and N2 to be protonated corresponds the following
MEP of Lamotrigine, see Fig. 3 where the electron density follows
the order N2zN1>>N4.
Table S1 (Supplementary data) contains all the available in-
formation from the last update of the CSD, together with Fig. S1
reporting the formulas of some compounds of Table S1 (struc-
tures 6e11).
The calculated geometry of Lamotrigine 1a (Fig. 1) could be
compared with the experimental X-ray structure (EFEMUX01)
(Fig. 4).
2.1.1.3. Geometry of Lamotrigine. In this section we will discuss
the X-ray structures of neutral Lamotrigine as reported in the CSD.16

I. Alkorta et al. / Tetrahedron 70 (2014) 2784e2795
2787
1a
EFEMUX01
1.348
Ha
1.340
1.323
1.325
1.350
1.312
ΣN = 355.4º
Σ
N = 359.4º
Ha
N
N
N
N
N
N
1.360 (C-NH2)
1.335 (C-NH2)
Cl
N
Hb
1.340
1.328
Cl
N
Hb
1.351
1.325
Cl
N
Cl
N
Cl
C2'
N(5)H2
Hb
Hb
Ha
Ha
1.358 (C-NH2)
1.338 (C-NH2)
1.428
1.433
N1
H
Σ
ΣN = 355.1º
N = 360.0º
C6'
C6-C1': 1.489
N1-C6-C1'-C2'(Cl): 116.6º
C6-C1': 1.498
N1-C6-C1'-C2'(Cl): 111.6º
View along the C6-C1' bond
2
ꢁ
Fig. 4. Calculated and experimental geometries of Lamotrigine (distances in A). SN corresponds to the sum of the three angles around the N atom (360 for an sp atom and 327.8ꢁ
ꢂ
for an sp3 atom).
The experimental geometry is acceptably well reproduced by
the calculations although the amino groups are more planar in the
crystal than in the gas phase, this being probably due to the exis-
tence of intermolecular hydrogen bonds (see further on). The
biaryl-type conformations12 of 1a and EFEMUX01 are almost
identical with the chlorine atom at position 2 pointing towards the
amino group at position 5.
in the range that can be measured by dynamic NMR (DNMR) in the
presence of a chiral additive or by dynamic chiral HPLC at low
temperatures (cryo-chromatography).
The N1eC6eC10eC20(Cl) torsion angles are ꢀ14.3ꢁ for a and
þ174.5ꢁ for b. The amino groups are no longer sp2 but intermediate
between sp2 and sp3, especially in a.
Concerning the intermolecular hydrogen bonds (IMHB), ex-
cluding the frequent cases where the HB involves the ‘solvent’
molecule, all possibilities of homodimers are found (Fig. 6). We
have named them from the amino group (positions 3 and 5) to the
ring nitrogen atoms [N(2) and N(4)], for instance, 3,2/3,2 corre-
sponds to two bonds from the amino group at position 3 to the
nitrogen atom at position 2. Only a heterodimer is found in the case
We have calculated the barrier to the rotation about the C6eC10
bond through a near planar structure that corresponds to a race-
mization process. There are two possibilities that correspond to
diastereomers: that the Cl atom at position 20 is on the side of N1
(Fig. 5a) and that it is on the side of the 5-NH2 (Fig. 5b). The barriers
are almost the same, 61.9 and 62.1 kJ molꢀ1, respectively, therefore
Fig. 5. The two transition states for racemization.
Cl
H
Cl
Cl
Cl
H
H
N
H
N
N
ΣN = 354.7º
N
N
N
N
ΣN = 359.5º
Cl
4
N
4
N
5
3
N
2
N
H
H
H
N = 352.9º
Σ
Cl
H
N
N
H
N
H
N
H
3
ΣN = 359.4º
H
H
H
N
H
N
5
H
N
H
N
H
N = 360.0º
Σ
ΣN = 352.1º
3
N
4
N
4
N
N
2
H
H
3
H
H
N
H
Cl
Cl
N
N
N
H
N
N
N
Cl
Cl
Cl
Cl
3,4/3,4
5,4/5,4
3,2/3,2
Fig. 6. Hydrogen-bonded homodimers of tautomer 1a.

2788
I. Alkorta et al. / Tetrahedron 70 (2014) 2784e2795
of EFEMUX01 (5,2/3,4). There is also an example of a crystal con-
taining two homodimers, BEGZUI, thus, it will appear two times in
the list given below.
The distribution of homodimers in the compounds of Table S1
is:
E configuration). Assuming that the relative energies should be
similar to those calculated for the 2H-tautomers and since the most
stable is the bE isomer, the relative energies are 9bE (0.0), 9bZ
(24.9), 9dE (32.3), 9dZ (17.8), 9fEE (25.6), 9fEZ (38.3), 9fZE (28.2)
and 9fZZ (36.3 kJ molꢀ1). It appears that 9dE is one of the less stable
isomers. Taking into account that when replacing an N2eH by an
N2eMe some stabilizing interactions may be lost, we have calcu-
lated the relative energies of the eight compounds of Fig. 7: 9bE
(0.0), 9bZ (27.1), 9dE (35.8), 9dZ (21.4), 9fEE (28.5), 9fEZ (41.0), 9fZE
(31.6), and 9fZZ (39.4 kJ molꢀ1) that are very similar excluding the
above hypothesis.
3,2/3,2: BEGZUI, SAXHUW; total 2.
3,4/3,4: SAXJAE, WUVKII, WUVLIJ, WUVLOR; total 4.
5,4/5,4: BEGZUI, KADPAG, OVUMEY, PEZKEM, SAXHOQ, YERTAR;
total 6.
The relative energy calculations of the three homodimers (the
interaction energies correspond to those of the dimer minus twice
those of the monomer, always 1a) are given below; all values in
2.1.1.4. Geometry of protonated Lamotrigine. Since there are
many available structures (Table S1), all of them corresponding to
the 1Hþab structure, we have selected the simplest anions, chloride
(YUCRAQ) and nitrate (YUCREU, two independent molecules). The
comparison between calculated and experimental results (Fig. 8) is
acceptable. The amino groups, both in the calculated and the
crystallographic structures, are almost perfectly sp2 and more
planar than in the neutral molecules of Fig. 5. This fact and the very
long C5eC6 bond suggest the major participation of the resonance
structure represented on the right side of Fig. 8.
kJ molꢀ1
.
3,2/3,2¼0.0 (interaction energy: ꢀ58.9)
3,4/3,4¼29.4 (interaction energy: ꢀ29.5)
5,4/5,4¼23.7 (interaction energy: ꢀ35.2)
As it can be seen the calculated order of stability is 3,2/
3,2>>5,4/5,4>3,4/3,4 while the frequency in crystals is 5,4/
5,4>3,4/3,4>3,2/3,2, but the experimental set is too small to draw
any firm conclusion.
It is worth noticing that the planarity of the amino groups in the
dimers is larger in those involved in HBs (about 359.6ꢁ) than in
those not involved (352.7ꢁ) in agreement with Fig. 4 data, where
the calculated monomer corresponds to the free NH2 while the
crystal structure (EFEMUX01) corresponds to hydrogen-bonded
NH2 groups.
We have calculated the barrier to the rotation about the C6eC10
bond of 1Hþab. We have obtained 66.9 kJ molꢀ1 (Cl atom at posi-
tion 20 is on the side of N1) and 69.6 kJ molꢀ1 (Cl atom on side of the
5-NH2). Compared to the 61.9 and 62.1 kJ molꢀ1 barriers of 1a, both
show an increase; the fact that the increase of the second barrier is
larger (7.5 instead of 5.5 kJ molꢀ1) is due to an increase in the
planarity of the 5-amino group (from 355.1ꢁ to 359.9ꢁ, Figs. 4 and 8,
respectively).
Compound 9 (see Supplementary data) owing to the 2-methyl
substituent has a reduced number of possible isomers (Fig. 7). Its
structure (VIYQAW) corresponds to 9dE (aminoeimino tautomer of
The conformation about the C6eC10 bond is like in the
neutral molecule (chlorine towards the NH2) for the two in-
dependent molecules of the nitrate of Lamotrigine, YUCREU
CH3
N
CH3
N
CH3
N
H
N
CH3
N
NH2
N
NH2
Cl
Cl
N
Cl
Cl
N
H
H
Cl
Cl
N
Cl
N
H
Cl
Cl
N
Cl
Cl
N
Cl
N
N
Cl
Cl
N
N
H
NH2
NH2
9dZ
9bE
9bZ
9dE
CH3
N
CH3
N
H
N
CH3
N
CH3
N
H
N
N
H
N
H
Cl
N
N
H
N
N
Cl
N
N
N
N
N
N
H
H
N
H
N
H
H
H
9fZE
9fEE
9fZZ
9fEZ
Fig. 7. The eight possible tautomers and isomers of 2-methyl Lamotrigine.
1H+ab
Average of three structures
1.360
Ha
N
1.342
Ha
N
1.347
1.349
H
N
H
N
1.295
Cl
1.292
Cl
ΣN = 359.3º
ΣN = 360.0º
Hb 1.336 (C3-NH2)
1.327
Cl
N
Cl
N
Hb 1.314 (C3-NH2)
1.339
N
N
H
N
H
N
1.331
1.328
N
N
Hb
Hb
Ha
Ha
Cl
N
H
1.325 (C5-NH2)
1.317 (C5-NH2)
Cl
N
1.468
1.457
ΣN = 359.6º
Σ
N = 359.9º
N
C6-C1': 1.487
C6-C1': 1.493
H
H
N1-C6-C1'-C2'(Cl): 111.2º
N1-C6-C1'-C2'(Cl): 97.6º
Fig. 8. The geometry of protonated Lamotrigine.

I. Alkorta et al. / Tetrahedron 70 (2014) 2784e2795
2789
[N1eC6eC10eC20(Cl)¼111.6ꢁ] but in the chloride, YUCRAQ, the Cl
To get additional data about the structure of protonated Lamo-
trigine, the structures of two new solvated salts with saccharine as
counter-anion have been determined by X-ray diffraction. While
a crystal structure of a nonsolvated Lamotrigine saccharinate is
already known (WUVKOO in S1),17 the asymmetric unit of the new
salts contains, together with a Lamotrigine saccharinate unit, two
half 1,4-dioxane molecules (LMSC-dio) or one tetrahydrofuran
(LMSC-THF).
Both crystal structures form chains of Lamotrigine dimers al-
ternated with saccharinate dimers (Fig. 10). Molecules in the same
dimer interact by stacking while two hydrogen bonds join contig-
uous dimers. The stacking interactions are formed by the di-
chlorobenzene and the benzene rings in the Lamotrigine and the
saccharinate, respectively. The hydrogen bonds involve N2 and N3
as acceptors in a NH/N and a NH/O interaction, respectively, with
the saccharinate, giving rise to the same R22(8) motif already ob-
served in WUVKOO01. This supramolecular motif is very common
in Lamotrigine salts, being systematically present in the large
family of dicarboxylates.18 Stacking of chlorobenzene rings in
Lamotrigine is also observed in other crystal structures, including
that of the Lamotrigine itself (EFEMUX01).19
at
position
20
point
towards
the
N(2)Hþ
group
[N1eC6eC10eC20(Cl)¼ꢀ72.4ꢁ].
For the cations (tautomer 1Hþab), besides interaction with the
‘solvents’, interactions with the anions are predominant, being the
one involving the NHþ(2) the only one observed. In the case of
cations, only the 5,4/5,4-homodimer has been found (Fig. 6).
Compounds 10 and 11 (see Table S1) have a 1Hþab type struc-
ture with the proton of the NþeH at position 2 replaced by a methyl
group (10) or by an iso-propyl group (11).
To compare 1a and 1Hþab we have calculated the Wiberg in-
dices within the NBO analysis (Fig. 9).
ꢂ
Note that to the very long C5eC6 bond (neutral 1.428 A, cation
ꢂ
1.468 A) correspond bond orders of 1.19 and 1.08, i.e., close to single
bond. The only distances that can be compared are the CeN bonds
because there are six of them, that is, 12 points for both molecules.
There is a rough relationship between bond distance and bond
order: bond distance¼1.51e0.13 bond order, R2¼0.88.
Bond orders:
1.50
1.17
H
N
In both LMSC-dio and LMSC-THF these chains are arranged in
layers, appearing CH/Cl interactions between neighbor chains in
the same layer. CH/O and stacking interactions between saccha-
rinates are observed in LMSC-dio and LMSC-THF, respectively,
giving rise to a different conformation of the chains inside the layer
in both cases. However, the most important difference between
both crystal structures is in the arrangement of the layers in the
crystal packing.
1.67
1.18
N
1.18
N
1.33
1.30
1.34
1.20
1.31
N
NH2
NH2
Cl
Cl
Cl
N
Cl
N
1.38
1.22
1.36
1.36
NH2
NH2
1.08
1.19
1H+ab
1a
In LMSC-dio, chains in neighbor layers are parallel (Fig. 11). The
layers are connected by two NH/O bonds, one of them part of
a R24(8) motif involving two Lamotrigine and two saccharinate
moieties. The two symmetry independent solvent molecules are in
interstices between the layers, bridging them through two NH/O
bonds in one case, and through CH/p interactions with two tri-
azine rings in the other.
In LMSC-THF, chains in neighbor layers are crisscrossed (Fig. 12).
Solvent molecules are in interstices between layers, bridging them
by NH/O contacts. While there is no hydrogen bonding between
layers, the proximity of an oxo-group in the saccharine to the
C-N bond distances (Å):
1.360
1.325
Cl
1.295
Cl
1.348
NH2
1.336
1.327
1.360
H
N
N
NH2
N
N
1.340
1.328
1.358
Cl
N
Cl
N
1.331
1.325
NH2
NH2
1H+ab
1a
Fig. 9. NBO analysis. Double bond, bond orderz2; single bond, bond orderz1.
Fig. 10. Chains of Lamotrigine and saccharinate in the crystal structures of LMSC-dio (top) and LMSC-THF (bottom). Hydrogen bonds are marked with blue lines.

2790
I. Alkorta et al. / Tetrahedron 70 (2014) 2784e2795
The assignment of the signals of the spectrum we determined
was carried out using 2D-experiments (HSQC and HMBC). The lit-
erature data were assigned by analogy.
The GIAO calculated absolute shieldings [transformed into
chemical shifts by means of equation
d
1H¼31.0e0.97
s
1H (see
computational details)] of the three protons of the o-dichlor-
ophenyl group (7.20H50, 7.33H60, and 7.43 ppm H40) are not in good
agreement with the experimental results (there is even a crossing)
but since the experimental data are from dichloromethane and
dimethylsulfoxide solutions, we calculated the GIAO/PCM(CD2Cl2)
and the GIAO/PCM(DMSO) absolute shieldings and transformed
them into chemical shifts.
In DMSO, the calculated signals of the amino groups are at
higher fields but this is expected for these protons,24 very sensitive
to concentration and solvent. What is more interesting is the dif-
ference in broadening between them. This may be due to a differ-
ence in the rotation rate about the CeN bond, the broader
corresponding to the slower rate, but we carry out B3LYP/6-
311þþG(d,p) calculations and the barriers are almost the same,
55.0 (H towards N2) and 59.2 (H towards N4) kJ molꢀ1 for the 3-
amino and 58.5 (H towards C6) and 51.1 (H towards N4) kJ molꢀ1
for the 5-amino group. Another possibility was that the aniso-
chrony of the protons would be very different (the broader should
correspond to the larger Dd); according to the GIAO calculations in
the 3-amino group the protons are separated by 0.59 and 0.60 ppm,
while in the 5-amino group, the separation is only 0.38 and
0.34 ppm. Therefore, the broad signals were assigned to the 5-
amino group.
Fig. 11. Detail of the interactions between layers in LMSC-dio. Two chains fragments in
different layers are plotted. The inter-layer contacts are (a) the R24(8) motif, (b) NH/O
bond, (c) 1,4-dioxane bridging both chains. The 1,4-dioxane molecules forming CH/
p
bonds with one chain are sketched (d).
13C results in DMSO-d6 have been reported twice: in Ref. 9 the
following values are given: 128.5, 130.6, 130.7 (CH of C4, C5, and
C6),131.6,132.0,136.8,138.3 (C of C3, C6 of the triazine, C2, C3 of the
phenyl ring), 154.1 and 162.1 ppm (C of C3, C5). In another paper
only unassigned eight signals were reported with almost identical
chemical shifts: 128.4, 130.6, 131.6, 132.0, 136.8, 138.3, 154.1, and
162.1 ppm (probably the 130.6 ppm signal corresponds both to the
130.6 and 130.7 of the preceding paper).22 In the solid state
(CPMAS) the resolution was less good and only signals at 129.3,
132.6, 140.2, 154.4, and 161.7 ppm were observed22 that roughly
correspond to the average of DMSO-d6 pairs save for the carbon
atoms bearing amino groups. In Table 3 are also reported our ex-
perimental assigned values. Since the X-ray structure of Lamo-
trigine corresponds to 1a, the CPMAS data must correspond to 1a
and since the values in the solid state are very similar to those in
DMSO-d6 we must conclude that they also correspond to 1a.
Often the carbon atoms bearing chlorine atoms are not well
reproduced by the calculations,25,26 so assuming a ‘chlorine effect’
the following equation is found that includes both 1H and 13C NMR
data determined in CD2Cl2: Exp. CD2Cl2¼(0.999ꢂ0.001) Calcd
CD2Cl2 ꢀ(7.5ꢂ1.1) Cl, n¼14, R2¼0.9999. Similar negative values for
the ‘chlorine effect’ were found in other publications.26,27
Fig. 12. Detail of the interaction between layers in LMSC-THF. One chain fragments and
its contacts with one of the contiguous layers are plotted. The inter-layer contacts are
(a) lone pair/p, identified by a short O/N distance and (b) THF molecule bridging
two chains. Each of the molecules in the contiguous layer belongs to a different chain.
In Table 3 are reported the calculated 15N NMR chemical shifts.
Note the considerable solvent effects on N1 and N2 and the good
agreement of the three experimental chemical shifts with the cal-
culated values for structure 1a in DMSO.
triazine ring suggests the formation of an inter-layer lone pair/
p
interaction.20
2.1.2. NMR chemical shifts
2.1.2.1. Neutral Lamotrigine. All the data are gathered in
Table 3.
2.1.2.2. Protonated Lamotrigine. The 1H NMR spectra of pro-
tonated Lamotrigine have been reported21 for different anions
(Table 4): in average the signals appear at 7.79* (H4 or H6), 7.44
(H5), 7.68* (H6 or H4), and 6.06 ppm (a NH2) with coupling con-
stants of JHH¼7.8 (7.79), JHH¼7.6 (7.68), and JHH¼1.65 Hz. The
authors named these cations ammonium salts and represented
them with structure 1Hþad, but according to our calculations this
structure is 120.3 kJ molꢀ1 above 1Hþab (Table 2). The calculated
chemical shifts show the same problems as in the neutral molecule
concerning H6 and the amino groups. But here the situation is more
complex because the authors, without proof, assign to the cations
The 1H NMR spectrum of Lamotrigine has been published four
times (always in DMSO-d6).9,21e23 Only the signal of H5 was
assigned because it is a triplet, those of H4 and H6 as well as both
NH2 groups were unassigned. The averaged values are 7.35ꢂ0.02
(H40 or H60), 7.43ꢂ0.02 (H50), 7.69 (H60 or H40), 6.42ꢂ0.03, and
6.68ꢂ0.03 ppm (NH2 protons). The coupling constants of the ABC
3
3
4
3
3
system are JHH¼7.9ꢂ0.1 (7.69 and 7.43), JHH¼7.6ꢂ0.1 (7.35 and
4
4
7.43), and JHH¼1.65 Hz (7.43 ppm). The JHH coupling constant of
6.3 Hz reported in Ref. 9 is obviously wrong. From the spectrum
4
they reported, a value of JHH about 2 Hz can be estimated.

I. Alkorta et al. / Tetrahedron 70 (2014) 2784e2795
2791
Table 3
Chemical shifts (d
, ppm) and 1He1H coupling constants (Hz) of Lamotrigine
Atom
Calcd gas phase
Calcd CD2Cl2
Calcd DMSO-d6
Exp. CD2Cl2
Exp. DMSO-d6
Exp. DMSO-d6
Exp. DMSO-d6
Exp. CPMAS
Ref.
H40
This work
7.43
This work
7.64
This work
7.69
This work
7.66
This work
7.71
9
7.70
22
7.70
22
d
J¼7.6
J¼2.0
7.44
J¼7.6
J¼7.6
7.42
J¼7.6
J¼2.0
5.15 (b)
J¼8.0
J¼1.6
7.45
J¼8.0
J¼7.6
7.37
J¼7.6
J¼1.6
6.45 (b)
J¼7.8
J¼6.5
7.45
J¼7.8
J¼7.8
7.37
J¼7.6
J¼6.3
6.46 (b)
H50
7.20
7.33
7.45
7.32
7.48
7.33
7.42
7.35
6.69
6.43
d
d
d
d
H60
3-NH2
5-NH2
4.67, 4.07
aver. 4.37
Dd¼0.60
4.03, 3.69
aver. 3.86
Dd¼0.34
159.4
150.6
139.0
140.0
139.5
4.77, 4.39
aver. 4.58
Dd¼0.38
4.64, 4.05
aver. 4.34
Dd¼0.59
161.0
151.0
139.8
138.4
140.2
4.78, 4.44
aver. 4.61
Dd¼0.34
4.72, 4.12
aver. 4.42
Dd¼0.60
161.4
151.3
140.3
138.1
140.2
4.93 (vb)
6.71 (vb)
6.72 (vb)
C3
C5
C6
C10
C20 (Cl)
C30 (Cl)
C40
C50
C60
161.8
154.1
139.3
135.4
132.1
133.8
131.4
128.3
130.1
d
162.5
154.5
138.7
132.4
132.0
137.2
131.6
128.9
128.9
N.o.
162.1
154.2
138.3
135.8
131.6
132.0
130.7
128.5
130.6
d
162.1
154.1
138.3
136.8
131.6
132.0
130.6
128.4
130.6
d
161.7
154.4
140.2
140.2
132.6
132.6
129.3
129.3
129.3
d
141.6
130.1
126.9
131.6
140.9
131.6
128.5
131.5
140.8
132.0
128.8
131.6
N1
57.8
38.0
33.9
N2
N4
3-NH2
5-NH2
ꢀ70.6
ꢀ89.2
ꢀ92.6
d
ꢀ97.7
ꢀ176.5
ꢀ305.0a
N.o.
d
d
d
ꢀ183.3
ꢀ316.1
ꢀ314.6
ꢀ185.6
ꢀ314.4
ꢀ310.2
ꢀ185.4
ꢀ314.2
ꢀ308.8
d
d
d
d
d
d
d
d
d
d
d
d
a
1JNH¼88.4 Hz.
Table 4
Chemical shifts (
a mixture of 0.6 mL of CH2Cl2 and 0.4 mL of CF3CO2H (as internal reference CHDCl2 was used)
d
, ppm) and 1He1H coupling constants (Hz) of protonated Lamotrigine. Experimental data: 1H;21 13C.22 The experiments of the present work correspond to
Atom
Calcd gas phase
Calcd gas phase
Calcd gas phase
Calcd DMSO
Calcd DMSO
Calcd DMSO
Exp.
Exp.
DMSO-d6
Exp.
CPMAS
1Hþaa
1Hþab
1Hþad
1Hþaa
1Hþab
1Hþad
CH2Cl2þCF3CO2H
Ref.
H40
This work
8.01
This work
7.93
This work
7.93
This work
7.88
This work
7.82
This work
7.83
This work
7.80
21,22
7.79a
J¼7.8
22
d
3J¼8.2
4J¼1.6
7.52
H50
H60
7.61
7.00
7.55
7.16
7.55
7.31
7.60
7.36
7.54
7.39
7.58
7.41
7.44
d
d
3Jz7.9
3Jz7.9
7.42
7.68a
J¼7.6
3J¼7.7
4J¼1.6
NþH
3-NH2
10.49 (H1)
5.40
8.81 (H2)
5.04
6.07 (NHþ3
)
10.79 (H1)
5.42
9.23 (H2)
5.40
6.41 (NHþ3
6.41
)
d
d
b
6.07 (NHþ3 )
6.29 (b)
6.08a(2H)
(NHþ3 )
5.34
5-NH2
C3
C5
C6
C10
C20 (Cl)
C30 (Cl)
C40
C50
C60
N1
N2
N4
3-NH2
5-NH2
5.21
5.60
5.30
5.38
5.59
7.24 (b)
156.5
154.6
139.7
129.7
132.1
135.1
134.1
129.4
129.8
N.o.
6.08a(2H)
155.0
158.0
138.4
128.0
132.2
138.4
131.6
131.4
130.6
d
d
160.4
155.2
136.4
123.3
140.4
147.2
138.8
130.2
127.0
ꢀ135.7
ꢀ126.4
ꢀ167.1
ꢀ301.4
ꢀ296.7
150.3
153.5
142.8
127.4
139.0
145.8
137.0
129.4
128.4
ꢀ37.8
ꢀ225.2
ꢀ177.8
ꢀ311.0
ꢀ284.6
147.4
152.3
153.6
130.0
138.0
145.2
136.4
129.7
129.3
46.4
ꢀ74.7
ꢀ171.1
ꢀ322.8
ꢀ295.4
160.7
155.7
135.7
126.0
139.0
140.2
142.4
129.8
131.0
ꢀ133.3
ꢀ131.1
ꢀ172.0
ꢀ302.7
ꢀ293.8
151.1
153.4
140.0
131.4
139.6
141.7
134.4
129.4
131.2
ꢀ38.6
ꢀ222.2
ꢀ179.5
ꢀ307.2
ꢀ285.8
151.0
153.1
151.1
133.9
138.8
141.6
133.8
129.5
130.8
30.2
ꢀ79.0
ꢀ162.2
ꢀ325.4
ꢀ295.9
155.4
155.4
139.2
130.0
133.1
139.2
133.1
130.4
130.4
d
N.o.
d
d
ꢀ179.8
ꢀ283.2c
ꢀ279.8d
d
d
d
d
d
d
a
Unassigned.
b
c
Under the TFAA signal.
1JNH¼94.8 Hz.
d
1JNH¼93.8 Hz.

2792
I. Alkorta et al. / Tetrahedron 70 (2014) 2784e2795
a structure with an amino (position 5) and an ammonium group
(position 3). Assuming that the structure is 1Hþab an amino group
and the N(2)þH are not observed. Thus, 1H NMR cannot be used to
establish whether the structure of protonated Lamotrigine as
1Hþab or 1Hþad or even 1Hþaa.
dimethanol] that we have used successfully in a number of pre-
vious works.29e31 Obviously, the use of ABTE prevent using solvents
like DMSO-d6 where Lamotrigine is soluble and solvents like CDCl3
or CD2Cl2 become imperative. In these solvents the solubility of
Lamotrigine is very poor, besides the signal of the CHCl3 present in
CDCl3 interferes so finally CD2Cl2 was selected.
The 13C NMR spectrum of he cyclohexylsulfamate of Lamo-
trigine has been reported by Khan et al.22 (note that the anion has
not aromatic carbons that could interfere). They reported the values
both in DMSO-d6 and in the solid-state but only of eight carbons
instead of nine. We assumed that the signal of C3(Cl) was not ob-
served (it was expected at about 145 ppm) because it appears to-
gether with another signal. The closest signals are those at
138.4 ppm (about 7 ppm, DMSO-d6) and at 139.2 ppm (about
6 ppm, CPMAS). Since the authors do not assign the signals to the
different carbons, those of Table 4 are only tentative, in particular
for the CPMAS spectrum (a typo error reported 55.4 instead of
155.4).22
Using this technique (see Experimental section) we have de-
termined a value of
D
Gz ¼62.4 kJ molꢀ1. With less precise values,
TC
we have calculated in CDCl3
a
D
Gz ¼63 kJ molꢀ1. Compared with
TC
the calculated barriers of 61.9 and 62.1 kJ molꢀ1 for 1a, the agree-
ment is excellent. Similar barriers, between 58 and 68 kJ molꢀ1
,
have been measured for 2-(20-substituted-aryl)pyridines.12
We should note that the barrier was determined on the
ABTE/Lamotrigine complex but since the hydrogen bond linking
both species would take place on N2 (see previously), since these
barriers are essentially of steric origin,12 and since the populations
of both complexes, being diastereomers, are not exactly 50/50 but
very close to equal populations, we can safely assume that the
measured barrier should be very close to that of isolated
Lamotrigine.
According to the calculations (see Supplementary data) there
are only seven isomers that have two signals in the 150e160 ppm
region (between parentheses, the relative energies of Table 2 in
kJ molꢀ1): 1Hþaa (21.2), 1Hþab (0.0), 1Hþad (120.3), 1Hþdd E/Z
(w230) and 1Hþea E/Z (w100).
We have compared the nine points of the neutral (see above),
the 18 points of the cation (DMSO and CPMAS) and three possible
cations using the square correlation coefficient, R2, as a criteria:
1Hþaa, 0.62; 1Hþab, 0.91 and 1Hþad, 0.84. If instead of the values
calculated for the gas phase we use those of PCM (DMSO) then the
values are R2, 1Hþaa, 0.59; 1Hþab, 0.94 and 1Hþad, 0.80. In the case
of the best correlation coefficient (1Hþab) we have obtained
Exp.¼ꢀ(21ꢂ8)þ(1.15ꢂ0.06) Calcd DMSO ꢀ(5.1ꢂ1.2) Cl, n¼27,
R2¼0.942. The equation is similar to that found for the neutral from
alone (see before).
Our data (0.6 mL of CH2Cl2 and 0.4 mL of CF3CO2H) are also
reported in Table 4. The signals of C3 and C5 are unambiguously
assigned since both show long distance correlations with the NH2
signal at 6.29 ppm, while C10 and C6 show correlation peaks with
the NH2 group at 7.24 ppm. This also identifies both amino groups
as reported in Table 4, respectively, 3-amino and 5-amino.
The GIAO calculated 15N chemical shifts for the three structures
previously discussed are reported in Table 4. Unfortunately only the
less indicative signals are observed (N4, 3-NH2, and 5-NH2) almost
useless to determine the structure of the cation. Considering that
the calculations correspond to DMSO and the experimental values
to CH2Cl2þCF3CO2H, the agreement is very good. In the literature
we have found the following chemical shifts and coupling con-
stants that are consistent with those we have reported in Tables 3
and 4.28
Since Lamotrigine receptors are obviously chiral, possibly only
one enantiomer will adequately fit (the eutomer) but the other (the
distomer) will rapidly enantiomerize to the eutomer.
2.1.4. Aromaticity. To complete the characterization of Lamotrigine
1a and its conjugated cation 1Hþab we have calculated their aro-
maticity. Two criteria have been used: the geometric using
HOMA32,33 and the NMR using NICS.34 According to HOMA
(benzene¼1.00), Lamotrigine has a value of 0.92 and its cation
a value of 0.66, the last one being much less aromatic. The order of
aromaticity is benzene>1a>>1Hþab. The bond that deviates the
most from an aromatic bond33 is the C5eC6 one (standard value
33
ꢂ
ꢂ
ꢂ
1.388 A) with values of 1.428 A for 1a (Fig. 4) and 1.468 A (Fig. 8)
for 1Hþab. Note that the deviation is greater for the cation in
agreement with its lower aromaticity; also that the character of this
bond corresponds to the resonance form depicted in Fig. 8.
The NICS values are reported in Table 5.
Table 5
ꢂ
ꢂ
ꢂ
Calculated NICS (ppm) values (0 A, 1 A, 2 A). Face a opposite to the Cl substituents;
face b on the same side than the Cl substituents
Benzene35
Neutral 1a
Cation 1Hþab
NICS (0)
ꢀ8.05
ꢀ10.22
ꢀ10.22
ꢀ4.84
ꢀ4.84
ꢀ7.36
ꢀ13.59
ꢀ6.05
ꢀ7.31
ꢀ8.96
ꢀ6.78
NICS (1) face a
NICS (1) face b
NICS (2) face a
NICS (2) face b
ꢀ12.25
ꢀ11.69
ꢀ10.68
ꢀ8.75
NICS(1) is in general a good compromise between the perturbations involved in
NICS(0) and the low values of NICS(2). The order of aromaticity
1a>benzene>>1Hþab, being in reasonable good agreement with the HOMA
conclusions.
3. Conclusions
We have completed an exhaustive survey of Lamotrigine and its
monoprotonated derivatives. Although the most stable structures
are those expected from previous works, particularly X-ray crys-
tallography, we have provided information about less stable tau-
tomers, noting that imino tautomers are protected in some
patents.36 Besides, for Lamotrigine metabolites and derivatives, our
present work should be relevant.
We have reported NMR signals to identify the different struc-
tures emphasizing the usefulness of 15N NMR spectroscopy. Two
new solvated Lamotrigine saccharinates, one with 1,4-dixoane and
the other with tetrahydrofuran, are described.
The solvent effects are not very important particularly for 1Hþab
the most probable structure for the protonated Lamotrigine. On the
other hand the tautomeric structures are easily differentiated.
2.1.3. Determination of the enantiomerization barrier. To determine
the enantiomerization rate constant we have used DNMR. Since
there are not groups giving rise to diastereotopic signals, DNMR
was performed in the presence of a chiral additive in the solvent.
The additive was ABTE [a,a
0-bis (trifluoromethyl)-9,10-anthracen-

I. Alkorta et al. / Tetrahedron 70 (2014) 2784e2795
2793
We have determined by DNMR in the presence of ABTE, the
racemization barrier of Lamotrigine (62.4 kJ molꢀ1) in perfect
agreement with the calculated value. Since enantiomerization is
very fast, the fact that Lamotrigine is a racemate does not seems
relevant for its biological properties.
In the case of LMSC-dio, exchanging the positions of the O-atom
with one of the C-atoms in the 1,4-dioxane molecule not forming
hydrogen bonds does not modify significantly the crystal structure.
R1 [I>2s(I)] increases slightly from 0.0793 to 0.0796 or 0.0799 after
the positions exchange. Thus, the position of the O-atoms in this
molecule could not be ascertained with certainty. The WinGX55
package suite and Mercury56 were used for the preparation of the
publishing material. Further details on the crystal structure solu-
tion and refinement are given in Table 6.
4. Experimental section
4.1. General
The anhydrous crystalline Lamotrigine used for the NMR studies
and for the preparation of the salts used in the crystallographic
studies was obtained as a complimentary sample from a local
pharmaceutical company. The crystals were obtained like in our
previous work on Lamotrigine solvates.18
Table 6
Selected crystallographic data for compounds LMSC-THF and LMSC-dio
LMSC-THF
MoK
¼0.71073 A)
18H16Cl2N6O5S
950.66
Monoclinic
C2/c
26.856(5)
7.852(2)
21.670(4)
LMSC-dio
MoK
C20H20Cl2N6O5S
527.38
Triclinic
P-1
8.0183(4)
11.2415(6)
13.5038(9)
94.338(2)
103.314(2)
92.252(2)
1179.07(12)
2
ꢂ
ꢂ
Radiation
a
(l
a
(
l
¼0.71073 A)
Empirical formula
Formula weight
Crystal system
Space group
C
4.2. Computational methods
ꢂ
a/A
ꢂ
b/A
The geometry of the systems under investigation has been op-
timized at the B3LYP37,38/6-311þþG(d,p)39 computational level.
Frequency calculations have been carried out to ascertain the
structure of the minima (IF¼0) and the transition states (IF¼1). The
Natural Bond Orbital (NBO) method,40,41 within the NBO-6 pro-
gram,42 has been used to calculate the bond indexes of the systems.
All the calculations have been performed with the Gaussian-09
program,43 including the molecular electrostatic potential (MEP)
of neutral Lamotrigine. Absolute chemical shieldings have been
calculated within the GIAO approximation44,45 on the B3LYP/6-
311þþG(d,p) geometries. PCM calculations46 were carried out as
implemented in Gaussian 09. We have transformed the absolute
ꢂ
c/A
ꢁ
ꢁ
a
/
b
/
118.63(2)
ꢁ
g
/
3
ꢂ
Cell volume/A
4010.6(14)
8
1.574
0.466
Z
Density calcd/Mg mꢀ3
Absorption coefficient/
mmꢀ1
1.485
0.409
F(000)
1952
544
Crystal size/mm
Theta range for data
collection/ꢁ
0.25ꢃ0.25ꢃ0.09
0.23ꢃ0.12ꢃ0.07
a
1.73e24.98ꢁ
3.57e22.28ꢁ
Index ranges
ꢀ28ꢄhꢄ31, 0ꢄkꢄ9,
ꢀ25ꢄlꢄ0
ꢀ8ꢄhꢄ8, ꢀ11ꢄkꢄ11,
ꢀ13ꢄlꢄ14
shieldings (s, ppm) into chemical shifts (d, ppm) using two em-
pirical equations we have established previously:
Reflections collected
Independent reflections
Completeness to
Max. and min.
transmission
3638
8144
3530 [R(int)¼0.0533]
2953 [R(int)¼0.0355]
s
1H ¼ 31:0e0:97 1H47
q max (%) 99.9
98.5
d
0.959 and 0.892
0.972 and 9.923
Data/restraints/
parameters
3530/0/280
2953/0/307
d
13C ¼ 175:7e0:963
s
13C48
Goodness-of-fit on F2
1.030
1.100
R1, wR2 [I>2
R1, wR2 (all data)
s
(I)]
0.0579, 0.1198
0.1295, 0.1440
0.468 and ꢀ0.330
0.0793, 0.2388
0.0999, 0.2545
1.375 and ꢀ0.472
4.3. X-ray structure determination
Largest diff. peak and
ꢂꢀ3
hole/e A
For LMSC-THF, a 1:1 mixture of Lamotrigine (50 mg) and sac-
charine (36 mg) was dissolved in tetrahydrofuran by heating up to
65 ꢁC. The solution was allowed to cool and evaporate at room
temperature. Crystals suitable for single-crystal X-ray diffraction
analysis were obtained after one day. Single-crystal X-ray diffrac-
tion data was collected at room temperature using a Nonius CAD4
diffractometer. Data reduction was performed with XCAD4,49
crystal structure was solved by direct methods with SHELXS97,50
and the DIFABS51 absorption correction was applied.
For LMSC-dio, a 1:1 mixture of Lamotrigine (50 mg) and sac-
charine (36 mg) was dissolved in 1,4-dioxane by heating up to
65 ꢁC. The solution was allowed to cool and evaporate at room
temperature. Crystals suitable for single-crystal X-ray diffraction
analysis were obtained after one week. Single-crystal X-ray dif-
fraction data was collected at 300(2) K using a Nonius Kappa CCD
diffractometer equipped with an Oxford Cryosystems cryostream.
Data reduction and cell refinement were performed with DENZO
and SCALEPACK.52 An empirical absorption correction was applied
using SORTAV.53 Crystal structure was solved by charge flipping
with SUPERFLIP.54
a
Because of the low quality of the crystals, data could not be collected up to 25ꢁ.
4.4. Static and dynamic NMR (DNMR) experiments and ex-
periments of enantiodifferentiation
A first analysis of 1H NMR spectrum of Lamotrigine on CDCl3
showed the superposition of the signals of the compound with the
residual absorption of non deuterated solvent (CHCl3). 0.7 mL of
CD2Cl2 (99.90% deuteration) were added to 16 mg of Lamotrigine
powder. The suspension was sonicated and heated at 30 ꢁC during
20 min. The suspension was filtered and used to make the NMR
experiments. Experiments of enantiodifferentiation were carried
out after the addition of 6 mg of (S,S)-a,
a0–bis(trifluoromethyl)-
9,10-anthracenedimethanol (S,S-ABTE and S,S-D8-ABTE) (Figs. 13
and 14). The NMR spectrum allows us to determine a ratio
[ABTE]/[Lamotrigine]¼5.4 and calculating a weight of 0.13 mg of
substrate in the tube of NMR. All spectra were obtained and ana-
lyzed at several temperatures between 250 and 290 K. All experi-
ments were acquired on a Bruker AVANCE spectrometer (Bruker
BioSpin, Rheinstetten, Germany) operating at 600.13 MHz proton
frequency, equipped with a 5 mm inverse probe and a z-axis pulsed
field gradient accessory. The NMR spectra (1D and 2D correlation
spectra) of Lamotrigine using DMSO-d6 and a mixture CD2Cl2/
In both cases, crystal structure was refined with the SHELX-97
package.50 H-atoms were introduced in calculated positions with
the exception of the H-atom in the triazine ring, which was located
in the Fourier difference maps. Non-H atoms were refined aniso-
tropically and H-atoms were refined riding on their parent atoms.

2794
I. Alkorta et al. / Tetrahedron 70 (2014) 2784e2795
Fig. 13. 1H NMR (600 MHz, CD2Cl2) of Lamotrigine and effect of addition of ABTE at different temperatures (BeF). The A spectrum corresponds to a sample without ABTE.
Fig. 14. 1H NMR (600 MHz, CD2Cl2) of Lamotrigine and effect of addition of deuterated ABTE at 260 K using ABTE perdeuterated on the anthracene ring (ratio ABTE-d8/
Lamotrigine¼3.6).

I. Alkorta et al. / Tetrahedron 70 (2014) 2784e2795
13. Guidelines on the Use of INNs for Pharmaceutical Substances, 1997.
2795
CF3CO2H (3:2) were obtained on a Bruker AVANCE spectrometer
(Bruker BioSpin, Rheinstetten, Germany) operating at 500.13 MHz
proton frequency, equipped with a 5 mm CryoProbe triple reso-
nance and a z-axis pulsed field gradient accessory.57
Using the calculation formula for the rate constant for a coupled
site (Eq. 1) and the Eyring equation (Eq. 2)58 we have calculated the
barrier for Lamotrigine plus ABTE in CD2Cl2,
14. Elguero, J.; Marzin, C.; Katritzky, A. R.; Linda, P. The Tautomerism of Heterocycles;
Academic: New York, NY, 1976; p 165.
15. Stanovnik, B.; Tisler, M.; Katritzky, A. R.; Denisko, O. V. Adv. Heterocycl. Chem.
2006, 91, 1e134.
16. CSD Database Version 5.35 (Nov. 2013) Allen, F. H. Acta Crystallogr., Sect. B 2002,
58, 380e388 Allen, F. H.; Motherwell, W. D. S. Acta Crystallogr., Sect. B 2002, 58,
407e422.
17. Cheney, M. L.; Shan, N.; Healey, E. R.; Hanna, M.; Wojtas, L.; Zaworotko, M. J.;
Sava, V.; Song, S.; Sanchez-Ramos, J. R. Cryst. Growth Des. 2010, 10, 394e405.
h
i
0:5
ꢃꢃ ꢀ
ꢀ ꢀ
18. Galcera, J.; Friscic, T.; Hejczyk, K. E.; Fabian, L.; Clarke, S. M.; Day, G. M.; Molins,
E.; Jones, W. CrystEngComm 2012, 14, 7898e7906.
kC
¼
p
=2 Dn2 þ 6 JA2B
(1)
(2)
19. Sridhar, B.; Ravikumar, K. Acta Crystallogr., Sect. C 2009, 65, o460eo464.
20. Egli, M.; Sarkhel, S. Acc. Chem. Res. 2007, 40, 197e205.
21. Qian, Y.; Lv, P.-C.; Shi, L.; Fang, R.-Q.; Song, Z.-C.; Zhu, H.-L. J. Chem. Sci. 2009, 121,
463e470.
D
GzTC ¼ 19:12ꢅTCꢅð10:32 þ log TC=kCÞ
22. Rahman, Z.; Zidan, A. S.; Samy, R.; Sayeed, V. A.; Khan, M. A. AAPS PharmSciTech
2012, 13, 793e801.
23. Venkanna, G.; Nagender, D.; Venkateswarlu, P.; Shekhar, K. C.; Madhusudhan,
G.; Mukkanti, K. Der Pharm. Chem. 2012, 4, 100e105.
24. Pinto, J.; Silva, V. L. M.; Silva, A. M. S.; Claramunt, R. M.; Sanz, D.; Torralba, M. C.;
Torres, M. R.; Reviriego, F.; Alkorta, I.; Elguero, J. Magn. Reson. Chem. 2013, 51,
203e221.
with Dn¼56.175 Hz and JAB¼7.85 Hz measured at 260 K and a co-
alescence temperature TC¼300 K. This lead to
a value of
D
Gz ¼62.4 kJ molꢀ1. With less precise values (the signal of CHCl3
TC
interferes) we have calculated in this solvent a
D
Gz ¼63 kJ molꢀ1
.
TC
25. Claramunt, R. M.; Santa María, M. D.; Sanz, D.; Alkorta, I.; Elguero, J. Magn.
Reson. Chem. 2006, 44, 566e570.
Acknowledgements
26. Alkorta, I.; Alvarado, M.; Elguero, J.; García-Granda, S.; Goya, P.; Jimeno, M. L.;
Thanks are also given to the Ministerio de Economía y Com-
petitividad of Spain (Projects CTQ2012-13129-C02-02 and
ꢀ
Menendez-Taboada, L. Eur. J. Med. Chem. 2009, 44, 1864e1869.
27. Santa María, M. D.; Claramunt, R. M.; Herranz, F.; Alkorta, I.; Elguero, J. J. Mol.
Struct. 2009, 920, 323e326.
ꢀ
ENE2012-36368-C02-02), the Comunidad Autonoma de Madrid
28. Berger, S.; Braun, S.; Kalinowski, H.-O. NMR Spectroscopy of the Non-metallic
Elements; John Wiley & Sons: Chichester, UK, 1997; see Chapter 4.
(Project MADRISOLAR2, Ref. S2009/PPQ-1533) and the Generalitat
de Catalunya (Grant 2009SGR-203). Gratitude is also due to the CTI
(C.S.I.C.) and CESGA for an allocation of computer time. Financial
support to MICINN (project CTQ2012e32436) is gratefully ac-
knowledged. J.G. thanks CSIC and European Social Fund for a PhD
grant. We thank to the Servei de Ressonncia Magnetica Nuclear,
ꢀ
29. Pomares, M.; Sanchez-Ferrando, F.; Virgili, A.; Alvarez-Larena, A.; Piniella, J. F. J.
Org. Chem. 2002, 67, 753e758.
30. Comelles, J.; Estivill, C.; Moreno-Manas, M.; Virgili, A.; Vallribera, A. Tetrahedron
~
2004, 60, 11541e11546.
31. Estivill, C.; Pomares, M.; Kotev, M.; Ivanov, P.; Virgili, A. Tetrahedron: Asymmetry
2005, 16, 2993e2997.
32. Kruszewski, J.; Krygowski, T. M. Tetrahedron Lett. 1972, 13, 3839e3842.
33. Krygowski, T. M. J. Chem. Inf. Comput. Sci. 1993, 33, 70e78.
34. Schleyer, P. V. R.; Maerker, C.; Dransfield, A.; Jiao, H.; van Eikema Hommes, N. J.
R. J. Am. Chem. Soc. 1996, 118, 6317e6318.
35. Claramunt, R. M.; Alkorta, I.; Elguero, J. Comput. Theor. Chem. 2013, 1019,
108e115.
ꢀ
Universitat Autonoma de Barcelona, for allocating instrument time.
Dr. John E. Davies is acknowledged for single crystal X-ray diffrac-
tion data collection.
Supplementary data
36. Congreve, M. S.; Andrews, S. P.; Mason, J. S.; Richardson, C. M.; Brown, G. A. US
Patent 2013/0029963.
37. Becke, A. D. J. Chem. Phys. 1993, 98, 5648e5652.
Supplementary data related to this article (1H, 13C, and 15N NMR.
Spectra of Lamotrigine in CD2Cl2 at different temperatures). Opti-
mized Cartesian coordinates, corresponding energies and absolute
shieldings for species discussed in the text. CCDC-969425 (for
LMSC-dio), and CCDC-969426 (for LMSC-THF) contain the supple-
mentary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif. Supplementary
data related to this article can be found at http://dx.doi.org/10.1016/
j.tet.2014.02.075.
38. Lee, C. T.; Yang, W. T.; Parr, R. G. Phys. Rev. B 1988, 37, 785e789.
39. Frisch, M. J.; Pople, J. A.; Binkley, J. S. J. Chem. Phys. 1984, 80, 3265e3269.
40. Reed, A. E.; Curtiss, L. A.; Weinhold, F. Chem. Rev. 1988, 88, 899e926.
41. Foster, J. P.; Weinhold, F. J. Am. Chem. Soc. 1980, 102, 7211e7218.
42. Glendening, E. D.; Badenhoop, J. K.; Reed, A. E.; Carpenter, J. E.; Bohmann, J. A.;
Morales, C. M.; Landis, C. R.; Weinhold, F. NBO 6.0; University of Wisconsin:
Madison, WI, 2013.
43. Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.;
Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Na-
katsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng,
G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.;
Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery,
J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K.
N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.;
Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; Klene, M.;
Knox, J. E.; 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.;
Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dan-
References and notes
1. Engel, J.; Pedley, T. A. In Epilepsy. A Comprehensive Textbook Lippincott Williams &
Wilkins, Wolters Kluwer: Philadelphia, 2008.
2. Wyllie, E.; Cascino, G. D.; Gidal, B. A.; Goodkin, H. P. In Wyllie’s Treatment of
Epilepsy: Principles and Practice Lippincott Williams & Wilkins, Wolters Kluwer:
Philadelphia, 2011.
€
nenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, O.; Foresman, J. B.; Ortiz, J. V.;
Cioslowski, J.; Fox, D. J. Gaussian 09, Revision A.1; Gaussian: Wallingford CT,
2009.
44. Ditchfield, R. Mol. Phys. 1974, 27, 789e807.
45. London, F. J. Phys. Radium 1937, 8, 397e409.
3. International League Against Epilepsy (ILAE), www.ilae.org.
ꢀ
~
4. Gil-Nagel, A.; Lopez-Munoz, F.; Serratosa, J. M.; Moncada, I.; García-García, P.;
46. Tomasi, J.; Mennucci, B.; Cammi, R. Chem. Rev. 2005, 105, 2999e3093.
47. Silva, A. M. S.; Sousa, R. M. S.; Jimeno, M. L.; Blanco, F.; Alkorta, I.; Elguero, J.
Magn. Reson. Chem. 2008, 46, 859e864.
48. Blanco, F.; Alkorta, I.; Elguero, J. Magn. Reson. Chem. 2007, 45, 797e800.
49. Harms, K.; Wocadlo, S. CAD4 Data Reduction; University of Marburg: Marburg,
Germany, 1995.
50. Sheldrick, G. M. Acta Crystallogr., Sect. A 2008, 64, 112e122.
51. Walker, N.; Stuart, D. Acta Crystallogr., Sect. A 1983, 39, 158e166.
52. Otwinowski, Z.; Minor, W. Macromol. Crystallogr., Part A 1997, 276, 307e326.
53. Blessing, R. H. Acta Crystallogr., Sect. A 1995, 51, 33e38.
54. Palatinus, L.; Chapuis, G. J. Appl. Crystallogr. 2007, 40, 786e790.
55. Farrugia, L. J. J. Appl. Crystallogr. 1999, 32, 837e838.
56. Macrae, C. F.; Bruno, I. J.; Chisholm, J. A.; Edgington, P. R.; McCabe, P.; Pidcock,
E.; Rodriguez-Monge, L.; Taylor, R.; van de Streek, J.; Wood, P. A. J. Appl. Crys-
tallogr. 2008, 41, 466e470.
ꢀ
Alamo, C. Seizure 2006, 15, 142e149.
5. Goodwin, F. K. J. Clin. Psychiatry 2003, 64, 161e174.
6. Hirschfeld, R. M. A. Guideline Watch: Practice Guideline for the Treatment of
Patients with Bipolar Disorder, 2nd ed.; American Psychiatric Association: Ar-
lington, VA, 2005.
7. Calabrese, J. R.; Huffman, R. F.; White, R. L.; Edwards, S.; Thompson, T. R.;
Ascher, J. A.; Monaghan, E. T.; Leadbetter, R. A. Bipolar Disord. 2008, 10,
323e333.
8. Goodwin, F. K.; Jamison, K. R.; Ghaemi, S. N. Manic-depressive Illness: Bipolar
Disorders and Recurrent Depression, 2nd ed.; Oxford University: New York, 2007.
9. Beattie, K.; Phadke, G.; Novakovic, J. Lamotrigine, Profiles of Drug Substances,
Excipients, and Related Methology, 2012, Vol. 37, pp 245e284.
10. Seridi, L.; Boufelfel, A. J. Mol. Liq. 2011, 158, 151e158.
11. Ramya, T.; Gunasekaran, S.; Ramkumaar, G. R. Spectrochim. Acta, Part A 2013, 114,
277e283.
57. http://www.bruker-biospin.com/cryofit.html.
12. Alkorta, I.; Elguero, J.; Roussel, C.; Vanthuyne, N.; Piras, P. Adv. Heterocycl. Chem.
2012, 105, 1e188.
ꢁ
58. Sandstram, J. Dynamic NMR Spectroscopy; Academic: London, UK, 1982.