A structural study of N,N'-bis-aryl-N′′-acylguanidines

Article abstract of DOI:10.1039/c3nj00285c

The N-aryl- and N-acylguanidine structural motifs are essential for the function of several important classes of molecules, including pharmaceuticals, catalysts and natural products. Compounds combining both motifs can exist as different isomers due to tautomerism within the guanidine subunit, E/Z isomerism with reference to the guanidine double bond, and conformational isomerism. This complex phenomenon results in unresolved broad signal NMR spectra that strongly complicate the characterisation of these derivatives. Hence, the present study examines isomerism in N,N′-bis-aryl-N′′-acylguanidines using low temperature NMR spectroscopy in tandem with Density Functional Theory (DFT), Natural Bond Analysis (NBO) and the Gauge-Invariant Atomic Orbital (GIAO) approach for calculating the NMR chemical shifts associated with each isomer. It was found that the structural preference of these compounds is strongly influenced by intramolecular hydrogen bond (IMHB) effects.

Full text of DOI:10.1039/c3nj00285c

NJC
View Article Online
View Journal | View Issue
PAPER
0
00
A structural study of N,N -bis-aryl-N -acylguanidines†
a
a
a
b
Daniel H. O’Donovan,z Brendan Kelly,z Elena Diez-Cecilia,z Martin Kitson and
Cite this: NewJ.Chem., 2013,
a
3
7, 2408
Isabel Rozas*
The N-aryl- and N-acylguanidine structural motifs are essential for the function of several important
classes of molecules, including pharmaceuticals, catalysts and natural products. Compounds combining
both motifs can exist as different isomers due to tautomerism within the guanidine subunit, E/Z
isomerism with reference to the guanidine double bond, and conformational isomerism. This complex
phenomenon results in unresolved broad signal NMR spectra that strongly complicate the
Received (in Montpellier, France)
9th March 2013,
0
0 0
1
characterisation of these derivatives. Hence, the present study examines isomerism in N,N -bis-aryl-N -
Accepted 24th May 2013
acylguanidines using low temperature NMR spectroscopy in tandem with Density Functional Theory
(
DFT), Natural Bond Analysis (NBO) and the Gauge-Invariant Atomic Orbital (GIAO) approach for
DOI: 10.1039/c3nj00285c
calculating the NMR chemical shifts associated with each isomer. It was found that the structural
preference of these compounds is strongly influenced by intramolecular hydrogen bond (IMHB) effects.
www.rsc.org/njc
Moreover, the N-arylguanidino structural motif is found in a
large number of pharmacologically active compounds such as
Introduction
The guanidine functional group exhibits several unique struc- guanabenz, an a -adrenoceptor agonist used clinically as an
2
1
0
tural and electronic properties which have made this motif antihypertensive. The N-arylguanidine moiety is a key feature
ubiquitous both throughout the natural world and in many of a series of DNA minor groove binder agents with potential
synthetic applications. Foremost among these properties is the anticancer applications, as described by our group in a recent
strong basicity of guanidine (pKaH = 13.6 (ref. 1)) and, as a series of publications.1
1–13
Other pharmacological applications
1
4
result, its derivatives are protonated at neutral pH. The impor- include high-affinity serotonin receptor ligands, and the
tance of guanidinium binding to aromatic moieties through anticancer drug imatinib, a tyrosine kinase inhibitor developed
2
15
cation–p interactions is becoming increasingly understood.
by Ciba/Novartis.
Guanidine’s capacity for strong intermolecular binding is
Although N-acyl- and N-arylguanidines have found many
frequently exploited for its catalytic potential, and several applications, the structure of these compounds is non-trivial
asymmetric catalytic transformations utilise guanidine to this and often difficult to predict, as several tautomers and geo-
3
,4
effect.
Anionic and neutral guanidine residues have also metric and conformational isomers potentially exist. The struc-
found applications as ligands for the synthesis of a diverse tural complexity of this class of compounds combined with the
5
,6
range of organometallic complexes.
dependence of their pharmacological, catalytic or other activ-
N-Acylguanidines, which are often used as guanidine analogues ities upon structure has resulted in a number of theoretical and
with diminished polarity, have recently been used in a new experimental studies. For instance, our group has recently
class of neuropeptide Y2 receptor antagonists with improved described an example of conformational control in pyridin-2-yl
7
16
pharmacological profiles. N-Acylguanidines, such as N-Boc or guanidine derivatives, other studies have included an exami-
1
5
17
N-carbamate protected guanidines, are also important synthetic nation of guanidine tautomerism by N NMR spectroscopy
intermediates in the preparation of guanidine derivatives.
8
,9
and an analysis of the conformational preferences of
1
8
N-alkylacylguanidines.
a
Despite these studies, novel guanidine derivatives often
exhibit unexpected tautomeric and conformational prefer-
School of Chemistry, Trinity Biomedical Sciences Institute, Trinity College Dublin,
54-160 Pearse Street, Dublin 2, Ireland. E-mail: odonovdh@tcd.ie,
1
1
13
kellyb2@tcd.ie, diezcece@tcd.ie, rozasi@tcd.ie
ences. In our group, we have observed that the H and
C
b
School of Chemical and Pharmaceutical Sciences, Dublin Institute of Technology,
Kevin Street, Dublin 8, Ireland. E-mail: martin.kitson@dit.ie
0
0 0
NMR spectra of different N,N -bis-aryl-N -Boc protected guani-
dines were always characterised by broad, poorly resolved
signals. It seemed likely that this phenomenon might be
attributable to interconversion between two or more tautomers
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
c3nj00285c
‡
These authors contributed equally to this work.
2
408 New J. Chem., 2013, 37, 2408--2418
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
Paper
NJC
induced by the sterically-congested environment surrounding
The combination of variable temperature NMR and DFT
the central guanidine moiety. However, these poor spectra calculations provides a very strong tool for structural analysis.
made the NMR characterization of these substituted guani- Variable temperature NMR has been used to investigate struc-
dines almost impossible and we felt that a clear explanation of tural phenomena such as axial/equatorial conformational iso-
1
9
20
this observation was required by means of both experimental merism, intermolecular interactions in DNA nucleobases
2
1
and theoretical studies.
and tautomerism in small molecules such as purine and
Hence, continuing with our interest in the guanidine group, malonate derivatives. In addition, DFT methods, and in
and to further investigate this behaviour, we have studied particular B3LYP computations, are often used to complement
2
2
0
0 0
different N,N -bis-aryl-N -Boc protected guanidines by a combi- experimental studies, specifically in tandem with the Gauge-
nation of variable temperature NMR experiments and density Independent Atomic Orbital (GIAO) method for calculating
2
3
functional theory (DFT) calculations. On the one hand, we have NMR shifts.
0
0
0
chosen two model systems (N,N -bis-phenyl- and N-phenyl-N -
pyridin-2-yl)-derivatives 1 and 2, Fig. 1) to analyse the effect of phenylguanidines (note the absence of an acyl substituent) was
different intramolecular hydrogen bonds (IMHB) on the NMR carried out by Kagechika and co-workers; these studies
In one closely-related study, a structural analysis of N,N -bis-
(
2
4
spectra of these guanidines; and, on the other hand we have included extensive X-ray crystal structural data. However, in
selected a more complex example (compound 3, Fig. 1) where the present study the fluxional behaviour of N,N -bis-aryl-N -
0
0 0
two differently substituted guanidines are present, multiplying acylguanidines made the effective use of crystallographic
0
0 0
this congestive phenomenon.
methods impossible. This contrast between N,N -bis-aryl-N -
acylguanidines and their non-acylated counterparts provides a
striking example of how small modifications to the guanidine
subunit can have an enormous impact on the structural and
fluxional behaviour of the resulting compounds, and makes the
outcomes of this study very useful for the guanidine research
community.
Results and discussion
Structural isomers of N,N -bis-aryl-N -acylguanidines
0
0 0
0
0 0
In the neutral form of N,N -bis-aryl-N -acylguanidines, the
CQN double bond of the guanidine moiety is localised and
can conjugate with any of the three N-substituents giving rise
to three distinct tautomers labelled ar1, ac, and ar2 in this
study, according to the position of the double bond (Fig. 2).
0
0 0
0
Fig. 1 N,N -Bis-aryl-N -acylguanidines examined in this study; (1) N,N -bis-
0
0
0
0 0
phenyl-N -(tert-butoxycarbonyl)guanidine, (2) N-phenyl-N -(pyridin-2-yl)-N -
(
tert-butoxycarbonyl)guanidine and (3) 3-[2,3-di-(tert-butoxycarbonyl)guanidino]-
0
4
-[2-(tert-butoxycarbonyl)-3-(4-chloro-3-trifluoromethylphenyl)guanidino]-N-
phenylanilino.
0
0 0
Fig. 2 Six possible tautomers and E/Z isomers of N,N -bis-aryl-N -acylguanidines. Stabilising hydrogen bonds (HBs) are shown as hashed lines, while destabilising
steric or electronic interactions are shown as overlapping semicircles.
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 2408--2418 2409

View Article Online
NJC
Paper
Each tautomer also presents the possibility for E and Z isomerism
0
0 0
and thus, N,N -bis-aryl-N -acylguanidines can present at least
six potential isomeric forms (Fig. 2).
There are at least six possible isomers (Fig. 2) and examina-
tion of their intramolecular interactions, such as the stabilising
IMHB between the lone pair of O6 and the slightly acidic proton
H(N3) found in structure ar1.Z, allows identifying the most
stable systems. This hydrogen bonding motif was previously
0
0 0
experimentally observed by us in N,N -bis-Boc protected-N -
arylguanidine derivatives both by NMR spectroscopy and X-ray
1
6
crystallography.
Two structures (ar1.E and ar2.E) suffer from destabilising
steric or electronic interactions, such as the steric clash
between the two aryl groups in ar1.E or the repulsion between
the O6 and N3 lone pairs in ar2.E. Thus, structures ar1.Z, ac.Z,
ar2.Z and ac.E are likely to represent the most energetically
favourable isomers.
A further structural feature to be considered is the rotation
of the two C–N single bonds in the guanidine subunit giving
rise to conformational isomers. In the case of structure ar1.Z,
1
Fig. 3 H NMR spectrum of 1 at 20 1C, 0 1C, ꢁ20 1C and ꢁ40 1C (400 MHz,
CDCl ).
3
1
13
rotation about the C1–N2 and C1–N3 bonds is restricted, as Table 1 Low temperature H (400 MHz, ꢁ40 1C) and C NMR (100 MHz, ꢁ40 1C)
shifts (ppm) observed for tautomers 1.ar.Z and 1.ac in CDCl
3
either rotation would result in loss of the favourable
O6ꢀ ꢀ ꢀH(N3) IMHB, and would force N2 and N3 lone pairs out
of conjugation with the C1QN4 and C5QO6 double bonds.
Structure ar2.Z presents the same restrictions on rotation. In
structure ac.Z, rotation about the C1–N3 single bond would also
lead to loss of the O6ꢀ ꢀ ꢀH(N3) IMHB and of conjugation of the
N3 lone pair with the guanidine system, while rotation about
the remaining C1–N4 single bond would result in a steric clash
1 2
between the Ar and Ar groups; rotation in structure ac.E is
similarly restricted. Thus, the guanidine subunits of structures
ar1.Z, ac.Z, ar2.Z and ac.E can be considered as conformationally
restricted.
Atom number dH (1.ar.Z) dC (1.ar.Z) dH (1.ac)
dC (1.ac)
In cases where Ar
1
= Ar
2
, structures ar1.Z and ar2.Z are
a
a
a
a
a
1
2
3
4
5
6
7
8
9
1
7.13
7.40
6.99
—
123.0
129.9
122.8
146.6
140.2
—
7.66–6.69a 129.9–119.7
7.66–6.69a 129.9–119.7
equivalent due to symmetry, as are structures ac.Z and ac.E.
Thus, for compounds where Ar1 = Ar2 there are only two
energetically favoured isomers (labelled ar.Z and ac), rather
than the four shown in Fig. 2; this consideration simplifies the
characterisation of symmetrical compounds.
7.66–6.69
7.66–6.69
—
129.9–119.7
129.9–119.7
156.2
a
—
9.71
6.76
—
7.73
7.37
7.06
—
6.80
11.07
—
—
—
—
a
a
a
a
138.7
119.9
129.0
122.7
152.9
83.4
129.9–119.7
a
1
13
7.66–6.69a 129.9–119.7
H and C NMR experiments
0
7.66–6.69
7.66–6.69
—
129.9–119.7
129.9–119.7
164.4
a
1
13
11
1
1
At room temperature, the H and C NMR spectra of compounds
, 2 and 3 were found to be highly amorphous, particularly
2
3
1
—
—
79.1
in the aromatic region, suggesting that the N-Boc protected 14
guanidine moiety rapidly interconverts between several possible
tautomers. Using low-temperature NMR spectroscopy, guanidine amorphous.
1.48
28.3
1.56
28.1
a
Despite cooling to ꢁ40 1C, these signals were observed as broad and
1
1 2
(Ar = Ar = Ph in Fig. 3) was cooled in 20 1C increments from
1
20 to ꢁ40 1C, at which point the H NMR spectrum became
1
3
sufficiently resolved to observe two distinct tautomers: 1.ar.Z
and 1.ac (see Fig. 3 and Table 1).
The low temperature C NMR spectrum of 1 also supports
the presence of the two tautomers 1.ar.Z and 1.ac. The Boc CH
3
The tautomer ratio was determined to be 2 : 1 by integration signal (C14) and the quaternary signal of C13 are split into two
of the Boc CH3 signals (see Table 1). The NH signals of peaks corresponding to the major and minor isomers, respec-
1
3
tautomer 1.ar.Z appeared as two sharp singlets at 9.71 and tively (Table 1). However, the aromatic C NMR signals are only
.76 ppm, while those of 1.ac appeared as much broader well-defined in the major isomer whereas in the minor isomer
6
singlets at 11.07 and 6.80 ppm. In both tautomers, the presence (1.ac) these appear as a broad amorphous signal even at ꢁ40 1C;
of one highly shifted NH signal (Table 1) is indicative of IMHB. this peak broadening may be attributable to dynamic equilibration
2
410 New J. Chem., 2013, 37, 2408--2418
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
Paper
NJC
between different conformers of 1.ac due to rotation about the between proton HN7 of the major tautomer 1.ar.Z and the
N6–C4 and N7–C8 bonds. Alternatively, this effect may be C5QN signal at 140.2 ppm (see ESI†). This comparatively low
caused by steric clash in tautomer 1.ac leading to increased chemical shift is due to conjugation between the CQN double
NMR relaxation times for the aromatic nuclei. Notably, this bond and the electron-rich aromatic ring, and sharply contrasts
phenomenon is not observed in compound 2 (Ar = Py, Ar = Ph with the analogous C5 signal in tautomer 1.ac, which occurs at
1
2
in Fig. 3), where IMHB interactions with the pyridine N16 lead 156.2 ppm due to conjugation with the electron-withdrawing
to greater conformational rigidity (see Table 2). Boc carbonyl. The C12QO signals of the Boc group agree with
1
3
The assignment of the C NMR signals of 1 corresponding this assignment since this carbon is more electron-deficient
to C5QN and C12QO was aided by the use of low temperature due to conjugation in tautomer 1.ac (see Table 2).
HMBC experiments, in which a correlation was observed
1 2
In the case of guanidine 2 (Ar = Py, Ar = Ph in Fig. 3), two
distinct tautomers are again observed using low temperature
NMR (ꢁ40 1C), and integration of the Boc CH signals in the
3
1
1
13
Table 2 Low temperature H (400 MHz, ꢁ40 1C) and C NMR (100 MHz,
40 1C) shifts (ppm) of tautomers 2.ar1.Z and 2.ac.Z in CDCl
H NMR spectrum suggests that these tautomers exist in an
ꢁ
3
approximate ratio of 10 : 3. However, in contrast with the tauto-
meric preference of compound 1, the major tautomer of 2 (2.ac.Z)
places the CQN double bond in conjugation with the Boc group
rather than with an aromatic ring. This assignment is supported
13
by the low temperature C NMR spectrum of compound 2
(
ꢁ40 1C, 400 MHz, CDCl ), in which the C5 signal of the major
3
tautomer 2.ac.Z appears at 152.4 ppm due to conjugation between
the CQN double bond and the Boc carbonyl, while in the minor
tautomer 2.ar1.Z the C5 signal is found at 146.4 ppm (Table 2).
This is significantly downfield from the analogous signal in the
major tautomer (1.ar.Z) of the bis-phenyl compound (140.2 ppm),
Atom number dH (2.ar1.Z) dC (2.ar1.Z) dH (2.ac.Z) dC (2.ac.Z) suggesting that the double bond in this isomer is conjugated with
the more electron-withdrawing pyridine ring. Notably, none of the
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
7.13
7.38
7.80
—
—
10.13
13.58
—
6.91
7.63
7.06
8.28
—
123.3
129.0
120.8
138.3
146.4
—
7.13
7.38
7.69
—
—
12.28
12.65
—
6.95
7.72
7.06
8.28
—
124.3
129.1
122.4
136.9
152.4
—
other possible structural isomers in which the CQN conjugates
with the phenyl group (2.ar2.Z) or it exhibits the opposite configu-
ration (2.ac.E) are observed by NMR.
Compound 3 represents a more complex structure represen-
tative of arylacylguanidine tautomerism in highly functiona-
lised molecules. Unlike the structurally simpler 1 and 2, in
compound 3 it was possible to observe three different tauto-
mers using low temperature NMR (Fig. 4). High temperature
NMR studies (40, 60 and 80 1C) were also carried out; however,
the resulting spectra were increasingly amorphous due to a
greater degree of equilibration between the various tautomers.
—
—
160.6
117.3
137.5
121.8
145.6
154.0
82.8
155.2
113.7
139.1
118.4
145.9
164.4
79.3
0
1
2
3
4
5
—
1.57
—
1.54
28.3
28.3
1
13
Fig. 4 The three tautomers of 3 observed in low temperature H and C NMR studies.
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 2408--2418 2411

View Article Online
NJC
Paper
The low temperature NMR spectra of compound 3 were of all isomers were fully optimised at the B3LYP theoretical
crowded with overlapping peaks and, as a result, it was not level with the 6-31+G(d,p) basis set as implemented in the
2
5
1
13
possible to unambiguously assign every signal. The assignment Gaussian 09 program. Moreover, H and C NMR chemical
of the NMR spectra is further complicated by the fact that shifts on the optimised geometries were computed, also at
0
25
compound 3 also contains a second N,N -bis-Boc protected B3LYP/6-31+G(d,p) level, using the GIAO method and are
guanidine moiety, and hence many of the signals corres- given relative to the isotropic shifts of tetramethylsilane calcu-
0
0 0
ponding to the N,N -bis-aryl-N -acylguanidine moiety could lated at the same level of theory.
not be examined in isolation. Nonetheless, there is evidence
The differences in the calculated Gibbs energies for com-
for the existence of the three tautomers 3.ac.E, 3.ar1.Z and pound 1 (Table 3) agreed with the experimental NMR data
1
3
(
.ar2.Z (Fig. 4) in the low temperature H NMR spectrum regarding which is the major tautomer. The aryl conjugated
ꢁ
1
ꢁ40 1C). Thus, three signals are observed corresponding to imine tautomer (1.ar.Z) is 7.9 kJ mol more stable than the Boc
the proton of N15 at 6.15, 6.19 and 6.24 ppm; this assignment conjugated one 1.ac (Table 3). This preference for the aryl
1
5
1
was corroborated by examination of the [ N, H] correlation conjugated imine tautomer agrees with the relative integration
1
spectrum. The highly-shifted aromatic signal H(C13) ortho to observed in the H NMR spectrum of 1.
the CF
.99 ppm); integration of the signals of either H(N15) or H(C13) ments revealed compound 2 to favour the tautomer with the
suggests a 4 : 3 : 2 isomer ratio. Although examination of the low imine conjugated to the Boc group (2.ac.Z) instead of the
3
group also appears as three signals (7.93, 7.97 and
In contrast to the tautomeric preference of 1, NMR experi-
7
1
9
temperature F spectrum (ꢁ40 1C) revealed only two distinct 2.ar1.Z one. The third tautomer (2.ar2.Z), in which the imine
peaks (see ESI†), these signals are very broad and it is likely that conjugates with the benzene ring, was not considered as this
one of these peaks consists of two overlapping signals.
configuration negates the highly stabilising IMHB between
Even using extensive correlation spectrometry, it was not the lone pair of the pyridine N16 and the guanidine proton
1
3
possible to assign every peak in the C NMR spectrum of 3, in H(N6). Intriguingly, DFT calculations revealed the opposite
contrast with compounds 1 and 2. Nonetheless, several peaks energetic preference; thus, the energy difference between
were observed supporting the existence of three distinct tauto- the optimised structures of tautomers 2.ac.Z and 2.ar1.Z is
ꢁ
1
mers. For example, three separate signals in the range of 6.9 kJ mol , 2.ac.Z being favoured. Nonetheless, the small
9.8–84.1 ppm are present corresponding to the quaternary magnitude of this difference agrees with the experimental
Boc carbons C17, C27 and C30; this assignment is supported by observation of both tautomers in solution. The small discre-
7
HMBC correlation experiments.
pancy between theory and experiment in predicting which the
major tautomer is may be attributed to the approximations of
DFT, combined with the fact that the NMR experiments are
DFT, NBO and GIAO calculations
In order to confirm the tautomeric preferences of compounds 1, performed in solution while our DFT studies are carried out in
and 3, a theoretical DFT study was undertaken. The geometries the gas-phase.
2
ꢁ1
ꢁ1
Table 3 Gibbs energy difference (DG, kJ mol ), IMHB distance (Å), NBO second order perturbation energy (E(2), kJ mol ) and NBO charges calculated at the B3LYP/
-31+G(d,p) computational level for the experimentally observed tautomers of guanidines 1, 2 and 3
6
NBO charges
N/O
Compound
Isomer
DG
0.0
IMHB distance
2.26
NBO interaction
LP N - BD* CH
E(2)
13.6
43.8
15.0
61.8
H
1
ar1.Z
G
ꢁ0.565
ꢁ0.673
ꢁ0.663
ꢁ0.712
0.280
0.462
0.289
0.485
1
.88
2.24
.78
LP OCO - BD* NH
LP N - BD* CH
ac.Z
ar1.Z
ac.Z
+7.9
G
1
LP OCO - BD* NH
2
0.0
1.87
LP NPy - BD* NH
LP N - BD* CH
G
LP OCO - BD* NH
LP NPy - BD* NH
G
LP N - BD* CH
103.3
14.1
48.9
82.3
15.4
78.5
ꢁ0.539
ꢁ0.659
ꢁ0.718
ꢁ0.554
ꢁ0.609
ꢁ0.679
0.463
0.288
0.480
0.485
0.277
0.463
2
1
.24
.74
+6.9
1.79
2
1
.27
.85
LP OCO - BD* NH
0
a
3
ac.E
0.0
+7.3
1.77
.24
1.84
.23
1.85
.28
1.72
.30
LP OCO - BD* NH
—
—
—
—
—
—
—
—
2
LP N
LP OCO - BD* NH
LP N - BD* CH
LP OCO - BD* NH
LP N - BD* CH
LP OCO - BD* NH
LP N - BD* CH
G
- BD* CH
ar2.Z
ar1.Z
2
G
+7.8
2
G
b
ac.Z
+16.8
2
G
a
A simplified model of compound 3 was used for the computational studies (see Fig. 5); no NBO computation was carried out since compound 1
b
can be taken as a good representation of compound 3 in terms of IMHB. This tautomer was not experimentally observed for compound 3.
2
412 New J. Chem., 2013, 37, 2408--2418
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
Paper
NJC
Taking into account the complexity of compound 3, a However, in the case of compound 2, a third type of interaction
simplified model was considered for the theoretical study occurs from the lone pair of the pyridine N atom to the
0
(
compound 3 , Fig. 5). There are at least six possible isomers antibonding molecular orbital of a N–H bond of guanidine
available to the molecule, namely the E and Z isomer of each (LP NPy - BD* NH), and the E(2) of this interaction is much
ꢁ1
tautomer of the guanidine subunit. Four of these structures can higher than the other two (around 82 to 103 kJ mol , Table 3)
incorporate two IMHB, the other two structures which do not indicating that this type of IMHB is the strongest.
include these stabilising interactions were discarded from
this study.
The E(2) difference for the Nꢀ ꢀ ꢀCH interaction between
tautomers is minimal and no preference could be extrapolated
DFT calculations indicate that three of these four isomers from that. However, the E(2) difference for the Oꢀ ꢀ ꢀNH inter-
ꢁ
1
are within an energy range of 7.8 kJ mol , while the fourth, actions for the tautomers in compounds 1 and 2 is more
0
ꢁ1
3
.ac.Z is 9.0 kJ mol less stable than the other three (Table 3). significant indicating in both cases that a stronger IMHB is
All four isomers incorporate two IMHB in their optimised formed in the acyl conjugated isomers (ac.Z). Only compound 2
structures, shown to be minima by the absence of imaginary can present Nꢀ ꢀ ꢀNH interactions and these show very high E(2)
frequencies in frequency calculations. As was observed for values, the strongest being that of the 2.ar1.Z tautomer.
compound 1, isomers in which the imine is in conjugation
with an aryl system are stable. In this case 3 .ac.E is also theoretical level for compounds 1 and 2 (Table 4). Both H and
GIAO calculations were carried out at the B3LYP/6-31+G(d,p)
0
1
13
C
favoured as it contains a HB from the imine nitrogen lone-pair shifts (given relative to those of tetramethylsilane calculated using
to the relatively acidic proton on the ortho position of the the same level of theory) strongly agreed with the experimental
p-Cl-m-CF3 phenyl ring. The corresponding interaction in values recorded and proved useful in confirming the assignments
0
3
.ac.Z is weaker due to the electron rich nature of the p-NHR made from the NMR spectra. In particular, the NH protons of the
phenyl ring, a fact demonstrated by the longer HB distance. guanidine system, and the CQ(N/O) signals of the guanidine and
This interaction has previously been observed in N-phenyl- Boc groups, reinforced our assignments of compounds 1 and 2,
0
00
16
N ,N -bis-Boc-guanidines and it is likely to stabilise the despite the tautomeric preference being reversed for compound 2.
molecule significantly. As described in the previous section, As can be observed in Table 4, the computed NH signals of
the experimental NMR data for the actual compound 3 strongly tautomer 1.ar.Z appeared at 7.22 and 10.60 ppm (experimental:
agrees with the calculations on the simplified model (Table 3), 6.76 and 9.71 ppm), while the NH signals of tautomer 1.ac
and there is evidence that the corresponding three isomers appeared at 12.40 and 7.00 ppm (experimental: 11.07 and
(3.ac.Z, 3.ar1.Z and 3.ar2.Z) co-exist in solution.
6.80 ppm). The unusually low shift of the guanidine C5 experi-
To analyze the strength of the IMHB established in the mentally observed for 1.ar.Z (140.2 ppm) was replicated in the
tautomers of these molecules Natural Bond Order (NBO) calcu- GIAO calculations (137.3 ppm). The experimental NMR signal
lations have been carried out for compounds 1 and 2 and we of the Boc carbonyl (C12) from this tautomer (152.9 ppm) was
have considered that compound 1 represents those IMHB also consistent with the GIAO value (153.5 ppm).
1
3
found in compound 3 (see Table 3). In both molecules 1 and
2
In the case of tautomer 1.ac.Z, the C shifts obtained were
0
two common types of interactions were found, from a lone similar to those more commonly reported for N,N -di-(tert-
pair of the imine N in the guanidine moiety to the antibonding butoxycarbonyl)arylguanidines. Experimentally, the guanidine
1
6
molecular orbital of an aromatic C–H bond (LP N - BD* CH) C5 and Boc carbonyl C12 appeared at 156.2 and 164.4 ppm,
G
and from the lone pair of the O atom of the CQO group in the respectively. Their GIAO shifts correlated strongly to these
Boc moiety to the antibonding molecular orbital of a N–H bond values, appearing at 151.4 and 165.6 ppm, respectively
of guanidine (LP OCO - BD* NH). Looking at the second order (Table 4). Further confirmation for the existence of tautomers
perturbation energy (E(2)) values obtained for compounds 1 1.ar.Z and 1.ac came from the ipso carbon C4. In the low
1
3
and 2 for the LP N
1
G
- BD* CH, it shows values around 14 to temperature C NMR of 1.ar.Z, this signal appears at 146.6 ppm,
ꢁ
1
5 kJ mol (Table 3), this being the weakest IMHB for all which is very close to the GIAO value of 148.3 ppm (Table 4).
tautomers in both compounds. In the case of the LP OCO
BD* NH interactions, E(2) varies from around 44 to 79 kJ mol
-
This is highly shifted compared to the same C4 in 1.ac, which
was calculated to be 139.7 ppm (experimental: between 119.7
ꢁ
1
(Table 3), being the strongest type of IMHB for compound 1. and 129.9 ppm).
0
Fig. 5 Simplified structures (3 ) used as models for the tautomers of compound 3.
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 2408--2418 2413

View Article Online
NJC
Paper
1
13
1
13
Table 4 H and C NMR chemical shifts of compound 1 at the B3LYP/
-31+G(d,p) theoretical level using the GIAO method
Table 5 H and C NMR chemical shifts (ppm) of compound 2 at the B3LYP/
6
6-31+G(d,p) theoretical level using the GIAO method
Atom
Atom
number dH (1.ar.Z)
number dH (2.ar1.Z)
dC (2.ar1.Z)
dH (2.ac.Z)
dC (2.ac.Z)
dC (1.ar.Z)
dH (1.ac)
dC (1.ac)
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
7.83
8.07/8.21
7.71/10.24
—
121.0
112.0
127.0
128.5
145.1
—
7.85
8.09/8.21
7.78/10.23
—
121.7
126.7/128.9
117.8/122.0
140.4
151.9
—
1
2
3
4
5
6
7
8
9
1
1
1
1
1
7.91
120.8
7.83
121.5
8.16/8.20
7.60/7.89
—
—
10.60
7.22
128.6/129.4
119.8/123.5
148.3
137.3
—
—
140.5
117.0/118.5
126.4/128.7
120.8
7.99/8.20
7.15/10.27
—
—
7.00
12.40
—
8.00/8.20
8.32/8.32
8.06
—
—
126.9/129.3
115.7–120.6
139.7
151.4
—
—
138.3
123.0/124.4
128.8–129.7
124.8
—
—
11.16
14.49
—
8.00
8.34
7.49
9.04
—
12.71
13.71
—
7.52
8.37
7.60
9.14
—
—
—
161.9
122.7
136.1
115.2
145.1
154.9
87.8
153.9
112.0
136.9
115.3
146.2
165.2
84.5
—
7.71/9.72
8.10/8.19
7.80
0
1
2
3
4
5
0
1
2
3
4
—
153.5
88.5
165.6
84.3
—
—
—
1.82/2.28/2.32 28.3/28.3/34.5 2.03/2.34/2.37 28.4/28.5/35.2
2.03/2.40/2.40 28.4/28.4/34.7 2.03/2.37/2.37 28.3/28.3/35.2
Shifts are given relative to the isotropic values of TMS calculated at the
same level of theory. Multiple values for chemically equivalent protons
arise as GIAO is a single point calculation.
Shifts are given relative to the isotropic values of TMS calculated at the
same level of theory. Multiple values for chemically equivalent protons
arise as GIAO is a single point calculation.
For compound 2, all the calculated GIAO shifts for the NH being adjacent to a NH group and strongly suggests that the
protons and guanidine, Boc carbonyl and C8 ipso carbons Boc-conjugated imine species 2.ac.Z comprises the major
(
Table 5) strongly agree with experimental NMR values. The tautomer.
computed NH signals of tautomer 2.ar1.Z (Table 5) appeared at
1.16 and 14.49 ppm (experimental: 10.13 and 13.58 ppm),
while the NH signals of 2.ac.Z appeared at 12.71 and 13.71 ppm
experimental shifts: 12.28 and 12.65 ppm). The high d values A structural study of N,N -bis-aryl-N -acylguanidines has been
1
Conclusions
0
00
(
1
13
relative to 1.ar.Z and 1.ac can be explained by the increased carried out by means of low temperature H and C NMR
acidity of H6 in 2.ac.Z as well as by the presence of a HB analysis and theoretical DFT, NBO and GIAO calculations.
interaction with the strongly basic pyridine nitrogen. Similar to Thus, experimental and computational observations support
compound 1, an unusually low NMR shift for guanidine C5 is the prevalence of the aryl-conjugated Z structural isomer 1.ar.Z
0
experimentally observed in tautomer 2.ar1.Z (146.4 ppm). The for bis-phenyl compound 1. In the case of the N-phenyl-N -
GIAO value (Table 5) is consistent with this, as is the value for pyridin-2-yl derivative 2, the same isomer 2.ar.Z is predicted to
the ipso carbon C8 at 161.9 ppm (experimental: 160.6 ppm). The be the most stable one using DFT calculations; however, using
shifts of both of these carbons indicate an aryl-conjugated low temperature NMR experiments, the 2.ac.Z isomer was
1
3
imine as in tautomer 2.ar1.Z. The C NMR signal of the Boc found to predominate due to a strong IMHB between the
carbonyl C13 in 2.ar1.Z (154.0 ppm) is also consistent with the pyridine N lone pair and an adjacent NH proton within the
calculated GIAO value (Table 5).
central guanidine moiety (confirmed by the NBO calculations).
Tautomer 2.ac.Z, the major tautomer for this compound, A third bis-phenyl guanidine 3, exhibiting a more complex
1
3
0
exhibits C shifts more commonly seen for N,N -di-(tert-butoxy- structure, was found to show a preference similar to compound
1
6
carbonyl)pyridylguanidines. The guanidine (C5) and Boc 1; however, several other structural isomers are also evident
carbonyl (C13) carbons experimentally appeared at 152.4 and within the low temperature NMR spectra.
164.4 ppm, respectively. Their GIAO shifts correlated strongly to
The combination of experimental low temperature NMR and
these values (Table 5). The ipso carbon C8 experimentally computational studies presented here has helped to better
appeared at 155.2 ppm and computationally at 153.9 ppm. understand the complex and unresolved experimental NMR
This decreased shift compared to 2.ar1.Z agrees with C8 now spectra that are usually obtained for arylacylguanidines,
2
414 New J. Chem., 2013, 37, 2408--2418
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
Paper
NJC
facilitating the assignment of all experimental signals and the (3 ꢂ 20 mL). It was dried over MgSO
4
, filtered and concentrated
comprehension of all possible tautomers and isomers of these in vacuo to yield a residue which was purified by column
compounds due to fluxional effects.
chromatography over silica, eluting with hexane : EtOAc (10 to
Therefore, the design and structural interpretation of com- 40% hexane–EtOAc, gradient elution) to afford the product as a
0
00
1
pounds containing the N,N -bis-aryl-N -acylguanidine motif clear oil (83 mg, 62%). H NMR (400 MHz, CDCl , ꢁ40 1C, 10 : 3
3
can benefit from preliminary analysis of which major structural mixture of isomers): major isomer d 12.65 (s, 1H, NH), 12.28
form predominates under experimental conditions. The incor- (s, 1H, NH), 8.28 (s, 1H, CH Ar), 7.72 (m, 1H, CH Ar), 7.69 (d, J =
poration of IMHB features can be used to influence which 8.0 Hz, 2H, CH Ar), 7.38 (m, 2H, CH Ar), 7.13 (m, 1H, CH Ar),
structure is the major isomer; however, in more complex 7.06 (m, 1H, CH Ar), 6.95 (d, J = 8.2 Hz, 1H, CH Ar), 1.54 (s, 9H,
examples such as that of compound 3 it may be difficult to Boc). Minor isomer d 13.58 (s, 1H, NH), 10.13 (s, 1H, NH), 7.80
fully control which structural form predominates under experi- (d, J = 7.2 Hz, CH Ar), 7.63 (m, 1H, CH Ar), 7.38 (m, 2H, CH Ar),
mental conditions.
7.13 (m, 1H, CH Ar), 7.06 (m, 1H, CH Ar), 6.91 (m, 1H, CH Ar),
1
3
1
.57 (s, 9H, Boc). C NMR (100 MHz, CDCl , ꢁ40 1C, 10 : 3
3
mixture of isomers): major isomer d 164.4 (CQO), 155.2 (Cq Ar),
52.4 (CQN), 145.9 (CH Ar), 139.1 (CH Ar), 136.9 (Cq Ar), 129.1
CH Ar), 124.3 (CH Ar), 122.4 (CH Ar), 118.4 (CH Ar), 113.7 (CH
Boc). Minor isomer d 160.6 (Cq Ar),
a solution of aniline (99 mL, 1.09 mmol), N-tert-butoxycarbonyl- 154.0 (CQO), 146.4 (CQN), 145.6 (CH Ar), 137.5 (CH Ar), 138.3
Experimental
1
(
Preparation of compounds 1–7
0
0 0
N,N -Bis-phenyl-N -(tert-butoxycarbonyl)guanidine (1). Over Ar), 79.3 (Cq Boc), 28.3 (CH
3
0
10
N -phenylthiourea (250 mg, 0.99 mmol), and Et N (456 mL, (Cq Ar), 129.0 (CH Ar), 123.3 (CH Ar), 121.8 (CH Ar), 120.8
3
3
1
.27 mmol) in CH Cl (5 mL), HgCl2 was added (296 mg, (CH Ar), 117.3 (CH Ar), 82.8 (Cq Boc), 28.3 (CH Boc). IR (neat,
2 2 3
.09 mmol) at 0 1C. The resulting mixture was stirred at 0 1C cm ): nmax 3290 (NH), 2979, 1711, 1646 (CQO), 1587 (CQN),
ꢁ
1
for 0.5 h and 14 h more at room temperature. The reaction was 1561, 1498, 1448, 1417, 1357, 1292, 1245, 1157, 1097, 1010, 809,
+
filtered through celite and the pad was washed with EtOAc 690, 752. HRMS (m/z-ES): found: 335.1494 (M
20 mL). The filtrate was concentrated in vacuo and purified by Na requires: 335.1484.
column chromatography over silica, eluting with hexane: EtOAc 3-[2,3-di-(tert-Butoxycarbonyl)guanidino]-4 -[2-(tert-butoxy-
+ Na ).
(
17 20 4 2
C H N O
0
(
10 to 40% hexane–EtOAc, gradient elution) to afford the carbonyl)-3-(4-chloro-3-trifluoromethylphenyl)guanidino]-N-
0
product as a white solid (284 mg, 92%, m.p. 146–148 1C). phenylaniline (3). Over a solution of 3-amino-4 -[2-(tert-butoxy-
1
H NMR (400 MHz, CDCl , rt, highly amorphous mixture of carbonyl)-3-(4-chloro-3-trifluoromethylphenyl)guanidino]-N-phenyl-
3
0
26
isomers): d 9.62 (br s, 1H, NH), 8.86 (br s, o1H, NH, minor anilino 4 (203 mg, 0.39 mmol), N,N -di-Boc protected thiourea
isomer), 7.74–6.75 (m, 11H, CH Ar + NH), 1.51 (s, 9H, Boc). (129.5 mg, 0.47 mmol), and Et
3
N (168.7 mL, 1.21 mmol) in
1
H NMR (400 MHz, CDCl
3 2 2 2
, ꢁ40 1C, 2 : 1 mixture of isomers): CH Cl (2 mL), HgCl (203 mg, 0.39 mmol) was added. The
major isomer d 9.71 (br s, 1H, NH), 7.73 (d, J = 7.8 Hz, 2H, CH resulting mixture was stirred at room temperature for 5 h. The
Ar), 7.40 (t, J = 7.3 Hz, 2H, CH Ar), 7.37 (t, J = 7.3 Hz, 2H, CH Ar), filtrate was washed with saturated brine solution (20 mL) and
7
.13 (t, J = 7.2 Hz, 1H, CH Ar), 7.06 (t, J = 7.2 Hz, 1H, CH Ar), 6.99 water (20 mL), extracting with EtOAc (3 ꢂ 20 mL). It was dried
(d, J = 7.6 Hz, 2H, CH Ar), 6.76 (s, 1H, NH), 1.48 (s, 9H, Boc). over MgSO , filtered and concentrated in vacuo to yield a
4
Minor isomer d 11.07 (br s, 1H, NH), 7.42–7.00 (m, 10H, CH Ar), residue which was purified by column chromatography over
1
3
6
.80 (br s, 1H, NH), 1.56 (s, 9H, Boc). C NMR (100 MHz, silica, eluting with hexane : EtOAc (10 to 40% hexane–EtOAc,
CDCl
, ꢁ40 1C, 2 : 1 mixture of isomers): major isomer d 152.9 gradient elution) to afford the product as a light dark solid
CQO), 146.6 (Cq Ar), 140.2 (CQN), 138.7 (Cq Ar), 129.9 (CH (144 mg, 41.6%, m.p. 108–110 1C). H NMR (600 MHz, CDCl
Ar), 129.0 (CH Ar), 123.0 (CH Ar), 122.8 (CH Ar), 122.8 (CH Ar), rt, amorphous mixture of isomers): d 11.7 (bs s, 1H, NH),
3
1
(
3
,
1
1
7
3
1
1
19.8 (CH Ar), 83.4 (Cq Boc), 28.3 (CH
64.4 (CQN), 156.2 (CQO), 129.9–119.7 (br m, CH Ar + Cq Ar), (bs m, 11H, CH Ar), 5.94 (bs s, 1H, NH), 1.55 (s, 9H, Boc), 1.51
3
Boc). Minor isomer d 10.1–10.3 (bs s, 1H, NH), 9.35–9.94 (bs, 1H, NH), 8.03–6.43
ꢁ
1
1
9.1 (Cq Boc), 28.1 (CH Boc). IR (neat, cm ): nmax 3402 (NH), (s, 9H, Boc), 1.49 (s, 9H, Boc). H NMR (400 MHz, CDCl3,
3
294 (NH), 3261, 2976 (Aryl), 1717, 1663, (CQO), 1599 (CQN), ꢁ40 1C, 4 : 3 : 2 mixture of isomers): d 11.7 (bs s, NH), 11.7
588, 1552, 1462, 1446, 1367, 1241, 1208, 1144 (C–O), 1091, (bs s, NH), 10.7 (bs s, NH), 10.3 (bs s, NH), 10.3 (bs s, NH), 10.2
+
078, 1091. HRMS (m/z-ES): found: 334.1527 (M + Na ), (bs s, NH), 9.90 (bs s, NH), 9.64 (bs s, NH), 9.55 (bs s, NH), 7.99
C
18
H
21
N
3
O
2
Na requires: 334.1531.
(s, H, CH Ar), 7.97 (s, H, CH Ar), 7.93 (s, H, CH Ar), 7.74 (bs s,
N-(Phenyl)-N -(pyridin-2-yl)-N -(tert-butoxycarbonyl)-guanidine NH), 7.49 (d, H, J = 7.8 Hz, CH Ar), 7.45 (d, H, J = 7.8 Hz, CH Ar),
2). Over a solution of 2-aminopyridine (40 mg, 0.43 mmol), 7.34–7.40 (m, H, CH Ar), 7.21–7.29 (m, H, CH Ar), 6.97–7.13 (m,
0
00
(
0
27
N-tert-butoxycarbonyl-N -phenylthiourea (113 mg, 0.45 mmol), H, CH Ar + NH), 6.78–6.93 (m, H, CH Ar + NH), 6.73 (s, H, CH
and Et N (178 mL, 1.28 mmol) in CH Cl (3 mL), HgCl was Ar), 6.60 (s, H, CH Ar), 6.48 (bs s, H, NH), 6.41 (d, H, J = 7.4 Hz,
3
2
2
2
added (121 mg, 0.45 mmol) at 0 1C. The resulting mixture was CH Ar), 6.24 (bs s, H, NH linker), 6.19 (bs s, H, NH linker), 6.15
stirred at 0 1C for 0.5 h and 16 h more at room temperature. The (bs s, H, NH linker), 1.53 (s, 9H, Boc), 1.49 (s, 9H, Boc), 1.45
1
3
reaction was filtered through celite and the pad was washed (s, 9H, Boc), 1.44 (s, 9H, Boc), 1.39 (s, 9H, Boc). C NMR
with 20 mL EtOAc. The filtrate was washed with saturated brine (150 MHz, CDCl , rt, highly amorphous mixture of isomers):
solution (20 mL) and water (20 mL), extracting with EtOAc d 144–149 (quaternary signals in this range could not
3
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 2408--2418 2415

View Article Online
NJC
Paper
be assigned as they were broad and highly amorphous), 1240, 1135, 1111, 1080, 1031, 991, 831, 771, 688, 666. HRMS
+
1
1
1
26 5 2 3
39.6 (Cq Ar), 138.3 (Cq Ar), 137.9 (Cq Ar), 137.5 (Cq Ar), (m/z-ES): found: 520.1727 (M + H). C25H N O F Cl requires:
32.0 (CH Ar), 131.4 (CH, Ar), 129.3 (CH, Ar), 126.6 (CH, Ar), 520.1727.
25.3 (CH, Ar), 124.1 (q, J = 5.2 Hz, CH Ar), 124.4 (CH, Ar), 124.6
3-(4-Aminoanilino)aniline (5). To a Parr shaker flask, the
(
(
(
(
(
Cq, d, J = 30 Hz, CF Ar), 123.6 (CH, Ar), 123.5 (CH, Ar), 122.8 di-nitro derivative 7 (61.0 mg, 0.24 mmol), 10% of Pd–C mixture
3
CH, Ar), 121.7 (Cq, c, J = 274 Hz, CF
CH, Ar), 118.8 (CH, Ar), 113.9 (CH, Ar), 112.7 (CH, Ar), 110.7 vigorously stirred at room temperature under 3 atm of H
CH, Ar), 83.5 (Cq, Boc), 79.4 (Cq, Boc), 79.3 (Cq, Boc), 28.0 pressure for 24 h. The reaction mixture was filtered and the
3
), 120.2 (CH, Ar), 119.2 (12.0 mg) and EtOH (20.0 mL) were added. The mixture was
2
1
3
CH
3
, Boc), 27.9 (2 CH
3
, Boc). C NMR (400 MHz, CDCl
3
, ꢁ40 solvent was removed under vacuum to afford the product
1
1
C, 4 : 3 : 2 mixture of isomers): d 163.7 (CQN), 163.6 (CQN), as a brown-purple solid (47.0 mg, 99% yield, m.p. 99 1C.
63.6 (CQN), 163.6 (CQN), 163.5 (CQN), 154.4 (CQO), 154.0 [lit. 97 1C] ). H NMR (600 MHz, DMSO): d 7.13 (bs s, 1H,
2
7
1
(CQO), 153.3 (CQO), 153.2 (CQO), 153.2 (CQO), 153.1 (CQO), NH), 6.79 (d, 2H, J = 8.6 Hz, CH Ar), 6.74 (t, 1H, J = 7.9 Hz, CH
1
1
52.8 (CQO), 152.7 (CQO), 147.0 (Cq), 145.9 (Cq), 144.9 (Cq), Ar), 6.50 (d, 2H, J = 8.6 Hz, CH Ar), 6.05 (t, 1H, J = 2.0 Hz, CH Ar),
44.1 (Cq), 144.0 (Cq), 141.5 (Cq), 140.3 (Cq), 139.9 (Cq), 139.2 5.98 (dd, 1H, J = 7.9, J = 2.0, CH Ar), 5.87 (dd, 1H, J = 7.9 Hz, J =
(
1
Cq), 138.2 (Cq), 137.82 (Cq), 137.74 (Cq), 137.2 (Cq), 137.0 (Cq), 2.0 Hz, CH Ar), 4.76 (bs s, 2H, NH
2
), 4.71 (bs s, 2H, NH
2
).
0
36.9 (Cq), 132.6 (Cq), 131.7 (CH Ar), 131.4 (CH Ar), 129.7
N-(4-Chloro-3-trifluoromethylphenyl)-N -tert-butoxycarbonyl-
0
(
(
(
CH Ar), 129.6 (CH Ar), 127.2 (CH Ar), 125.3 (CH Ar), 124.3 thiourea (6). A solution of N,N -di-Boc protected thiourea
CH Ar), 124.1 (CH Ar), 123.7 (CH Ar), 123.6 (CH Ar), 123.2 (2000 mg, 7.25 mmol) in THF (58.5 mL) was treated with NaH
CH Ar), 122.1 (CH Ar), 121.4 (CH Ar), 120.0 (CH Ar), 119.0 (CH (435 mg, 10.9 mmol) under argon. The reaction mixture was
Ar), 117.5 (CH Ar), 114.2 (CH Ar), 113.8 (CH Ar), 113.0 (CH Ar), stirred at 0 1C for 1 h and 20 min followed by the addition of
10.8 (CH Ar), 110.6 (CH Ar), 84.1 (Cq Boc), 84.0 (Cq Boc), 83.9 trifluoroacetic anhydride (1.11 mL, 8.0 mmol). 30 min later this
Cq Boc), 83.9 (Cq Boc), 83.8 (Cq Boc), 80.0 (Cq Boc), 80.0 was followed by the addition of 4-chloro-3-(trifluoromethyl)-
Cq Boc), 79.9 (Cq Boc), 79.8 (Cq Boc), 28.1 (CH Boc). IR aniline (1.63 g, 8.34 mmol). The reaction mixture was stirred at
neat, cm ): nmax 3264 (NH), 2978, 2929, 2049, 1719 (CQO), 0 1C for 1 h, and overnight at room temperature. The reaction
1
(
(
(
3
ꢁ
1
1596 (CQN), 1560, 1509, 1479, 1368, 1320, 1234, 1146, 1106, was quenched with H
2
O (20 mL) followed by extraction with
1
057, 1029, 806, 771, 688, 665. HRMS (m/z-ES): found: 762.2999 EtOAc (3 ꢂ 20 mL). The combined organic phases were washed
+
(M + H), C H N O F Cl requires: 762.2994.
with saturated brine solution (30 mL) and dried over MgSO ,
4
3
6
44
7
6
3
0
3
-Amino-4 -[2-(tert-butoxycarbonyl)-3-(4-chloro-3-trifluoro- followed by removal of solvents in vacuo. Purification pro-
methylphenyl)guanidino]-N-phenylaniline (4). Over a solution ceeded by silica gel chromatography using gradient elution
of diamine 5 (403.8 mg, 2.03 mmol), N-(4-chloro-3-trifluoro- (hexane : EtOAc), followed by recrystallisation from boiling
0
methylphenyl)-N -tert-butoxycarbonylthiourea
6
(239 mg, hexane afforded the product as white needles (1680 mg,
1
0
.676 mmol), and Et
3.5 mL), HgCl (220.2 mg, 0.81 mmol) was added at 0 1C. (bs s, NH Boc), 8.02 (s, 1H, CH Ar), 7.99 (br s, NH Ph), 7.94
The resulting mixture was stirred at 0 1C for 1 h and further 3 h (d, 1H, J = 8.5 Hz, CH Ar), 7.53 (d, 1H, J = 8.5 Hz, CH Ar), 1.56
3
N (291.7 mL, 2.10 mmol) in CH
2
Cl
2
3
91.7%, m.p. 140–142 1C). H NMR (600 MHz, CDCl ): d 11.7
(
2
1
3
at room temperature. The filtrate was washed with saturated (s, 9H, Boc). C NMR (150 MHz, CDCl ): d 178.0 (CQS), 152.0
3
brine solution (20 mL) and water (20 mL), extracting with EtOAc (CQO), 136.5 (Cq Ar), 129.5 (Cq Ar), 128.8 (Cq, d, J = 30 Hz, CF3
(
3 ꢂ 20 mL). It was dried over MgSO , filtered and concentrated Ar), 126.2 (CH Ar), 123.3 (q, Cq, J = 274 Hz, CF ), 123.2 (q, J =
4
3
ꢁ1
in vacuo to yield a residue which was purified by column 5.2 Hz, CH Ar), 84.9 (Cq Boc), 28.0 (CH
chromatography over silica, eluting with hexane : EtOAc (10 to max 3172 (NH), 2979, 2900, 1727, 1710, 1589, 1527 (CQO),
0% hexane–EtOAc, gradient elution) to afford the product as a 1478, 1425, 1324, 1258, 1239, 1128 (CQS), 1113, 1033, 1015,
3
Boc). IR (neat, cm ):
n
4
1
dark solid (262.6 mg, 86%, m.p. 78–80 1C). H NMR (600 MHz, 972, 838, 820, 775, 691, 663. HRMS (m/z-ES): found: 353.0332
CDCl
(
+
3
, rt, highly amorphous mixture of isomers): d 9.40–10.0 (M ꢁ H). C13
13 2 2 3
H N O ClF S requires: 353.0338.
bs m, 1H, NH), 7.89–8.13 (m, 1H, CH Ar), 6.76–7.47 (m, 9H,
3-Nitro-N-(4-nitrophenyl)aniline (7). An oven-dried Schlenk
CH Ar + NH), 6.55–6.73 (m, 1H, CH Ar), 6.43 (d, 1H, J = 7.2 Hz, flask was evacuated and backfilled with Ar. Afterwards, it was
CH Ar), 6.39 (s, 1H, CH Ar), 6.27 (d, 1H, J = 7.2 Hz, CH Ar), charged with Pd (dba) (27.5 mg, 3.0 mol%), (2-diphenyl)-
2
3
5
.44–5.73 (bs s, 1H, NH), 3.72 (bs s, 2H, NH
2
), 1.52 (s, 9H, Boc). dicyclohexylphosphine (10.5 mg, 3.0 mol%) and 1,2-dimethoxy-
1
3
ꢁ1
C NMR (150 MHz, CDCl
3
, rt, highly amorphous mixture of ethane (DME) (2 mL mmol of halide). The mixture was left
3 4
isomers): d 147.5 (CQO), 134.4–160.0 (quaternary signals in to react for 5 min without heating. Then, K PO (297.2 mg,
this range could not be assigned as they were broad and 1.4 mmol), 1-chloro-3-nitrobenzene (157.5 mg, 1 mmol) and
highly amorphous), 132.2 (CH Ar), 131.5 (CH Ar), 130.2 1,4-nitroaniline (138.1 mg, 1 mmol) were added as solids
(
(
(
(
CH Ar), 126.6 (CH Ar), 124.1 (q, J = 5.2 Hz, CH Ar), 124.6 maintaining the flow of Ar. The mixture was stirred at 100 1C
d, Cq, J = 30 Hz, CF Ar), 123.6 (CH Ar), 123.0 (CH Ar), 121.8 for 24 h. It was then cooled to room temperature, diluted with
3
q, Cq, J = 274 Hz, CF ), 120.7 (CH Ar), 116.0 (CH Ar), 108.1 1 : 1 ether : EtOAc (40 mL), filtered through celite, and concen-
3
CH Ar), 107.6 (CH Ar), 103.2 (CH Ar), 83.5 (Cq Boc), 28.0 (CH
3
trated in vacuo. The crude material was purified by chromato-
ꢁ
1
Boc). IR (neat, cm ): nmax 3340 (NH), 3333 (NH), 2978, 2933, graphy on silica gel using 1 : 1 (EtOAc : hexane) to afford the
162, 1720 (CQO), 1591 (CQN), 1556, 1509, 1479, 1368, 1318, product as an orange solid (181.5 mg, 70%, m.p. 217–218 1C
2
2
416 New J. Chem., 2013, 37, 2408--2418
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013

View Article Online
Paper
NJC
28 1
[
lit. 214–218 1C] ). H NMR (600 MHz, DMSO): d 9.67 (bs s, 1H, NH),
4 D. Leow and C.-H. Tan, Chem.–Asian J., 2009, 4, 488.
8
.17 (d, 2H, J = 8.0 Hz, CH Ar), 8.00 (s, 1H, CH Ar), 7.85 (d, 1H,
5 E. G. Ijpeij, M. A. Zuideveld, H. J. Arts and G. H. J. van
Doremaele, PCT Int. Appl., WO 2007071317 A1 20070628,
2007.
J = 8.0 Hz, CH Ar), 7.70 (d, 1H, J = 8.0 Hz, CH Ar), 7.65 (t, 1H, J =
.0 Hz, CH Ar), 7.23 (d, 2H, J = 9.5 Hz, CH Ar).
8
6
M. P. Coles, S. E. Soezerli, J. D. Smith and P. B. Hitchcock,
Organometallics, 2007, 26, 6691.
Low temperature NMR studies: experimental details
The solvent used for the low-temperature NMR measurements
7 N. Pluym, A. Brennauer, M. Keller, R. Ziemek, N. Pop,
G. Bernhardt and A. Buschauer, ChemMedChem, 2011,
6, 1727.
8 D. H. O’Donovan and I. Rozas, Tetrahedron Lett., 2011,
52, 4117.
was CDCl , purchased from Apollo. NMR spectra were recorded
3
on Bruker DPX-400 Avance spectrometers, operating at
1
4
1
00.13 MHz and 600.1 MHz for H NMR; 100.6 MHz and
1
3
50.9 MHz for C-NMR. Shifts are referenced to the internal
solvent signals. Initially, H-NMR and C-NMR spectra were
1
13
9 B. R. Linton, A. J. Carr, B. P. Orner and A. D. Hamilton,
J. Org. Chem., 2000, 65, 1566.
obtained at room temperature (20 1C). The probe temperature
was then reduced by cooling with a liquid nitrogen/air stream 10 D. T. Nash, J. Clin. Pharmacol., 1973, 13, 416.
ꢁ
1
with a gas flow rate of around 535 L h . The probe temperature 11 P. S. Nagle, F. Rodriguez, B. Nguyen, W. D. Wilson and
was lowered in 20 1C increments over 30 min, and spectra were
I. Rozas, J. Med. Chem., 2012, 55, 4397.
obtained at 20, 0, ꢁ20 and ꢁ40 1C. When the desired tempera- 12 P. S. Nagle, S. J. Quinn, J. M. Kelly, D. H. O’Donovan,
ture was reached, the spectrometer was left to equilibrate for
0–15 min. The readings are estimated to be accurate within
A. R. Khan, F. Rodriguez, B. Nguyen, W. D. Wilson and
I. Rozas, Org. Biomol. Chem., 2010, 8, 5558.
1
ꢃ1 1C. The spectrometer was tuned and shimmed at each 13 P. S. Nagle, F. Rodriguez, A. Kahvedzic, S. J. Quinn and
1
13
new temperature prior to running the H-NMR and C-NMR
I. Rozas, J. Med. Chem., 2009, 52, 7113.
spectra. After the ꢁ40 1C NMR spectra were recorded, the 14 R. A. Glennon, M. K. Daoud, M. Dukat, M. Teitler,
temperature was brought back to room temperature, by slowly
removing the cooling source. The NMR spectra were loaded
into the Bruker TOPSPIN software for data analysis.
K. Herrick-Davis, A. Purohit and H. Syed, Bioorg. Med.
Chem., 2003, 11, 4449.
15 E. Buchdunger, J. Zimmermann, H. Mett, T. Meyer,
M. Mueller, B. J. Druker and N. B. Lydon, Cancer Res.,
Computational details
1996, 56, 100.
1
1
1
1
6 B. Kelly, D. H. O’Donovan, J. O’Brien, T. McCabe, F. Blanco
and I. Rozas, J. Org. Chem., 2011, 76, 9216.
7 I. Ghiviriga, B. El-Gendy, M. E.-D. P. J. Steel and
A. R. Katritzky, Org. Biomol. Chem., 2009, 7, 4110.
8 R. Kleinmaier, M. Keller, P. Igel, A. Buschauer and
Geometries were fully optimized at the B3LYP theoretical level
with the 6-31+G(d,p) basis set as implemented in the Gaus-
2
8
sian09 program. Harmonic frequency calculations verified
the nature of the stationary points as minima (all real frequen-
cies). No zero-point correction to the energy values was applied.
NMR calculations were performed using the Gauge-Invariant
Atomic Orbital (GIAO) protocol at the same level of theory.
Chemical shifts were taken with reference to the tetramethyl-
silane standard. Where chemically equivalent signals gave
different GIAO values these were reported as obtained, though
this occurs as GIAO is a single point calculation on the
optimised structure, which considers chemically equivalent
signals as existing in distinct environments.
R. M. Gschwind, J. Am. Chem. Soc., 2010, 132, 11223.
´
´
9 S. O. Wallevik, R. S. Bjornsson, A. Kvaran, S. Jonsdottir,
G. V. Girichev, N. I. Giricheva, K. Hassler and I. Arnason,
J. Mol. Struct., 2010, 978, 209.
2
0 K. Weisz, J. J ¨a hnchen and H.-H. Limbach, J. Am. Chem. Soc.,
1997, 119, 6436.
2
1 P. n. Se ˇc k ´a ˇr ov ´a , R. Marek, K. Mali nˇ ´a kov ´a , E. Kolehmainen,
D. Hockov ´a , M. Hocek and V. Sklen ´a ˇr , Tetrahedron Lett.,
2004, 45, 6259.
2
2 E. M. B. L. Janke, S. Schlund, A. Paasche, B. Engels, R. d.
Dede, I. Hussain, P. Langer, M. Rettig and K. J. Weisz, J. Org.
Chem., 2009, 74, 4878.
Acknowledgements
This work was supported by studentships from the Irish
Research Council (DHOD, EDC) and the HEA-PRTLI-4 (BK). 23 K. Wolinski, J. F. Hilton and P. Pulay, J. Am. Chem. Soc.,
Thanks are given to Dr Martin Peters (ICHEC) for allocation of
computer time.
1990, 112, 8251.
24 A. Tanatani, K. Yamaguchi, I. Azumaya, R. Fukutomi,
K. Shudo and H. Kagechika, J. Am. Chem. Soc., 1998,
120, 6433.
Notes and References
2
5 M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,
B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato,
X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, R. Fukuda,
J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao,
H. Nakai, T. Vreven, J. A. Montgomery Jr, J. E. Peralta,
1
D. D. Perrin, Dissociation Constants of Organic Bases in
Aqueous Solution, Butterworths, London, 1965, supplement,
1972.
2
3
F. Blanco, B. Kelly, I. Alkorta, I. Rozas and J. Elguero, Chem.
Phys. Lett., 2011, 511, 129.
E. J. Corey and M. J. Grogan, Org. Lett., 1999, 1, 157.
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013
New J. Chem., 2013, 37, 2408--2418 2417

View Article Online
NJC
F. Ogliaro, M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin,
Paper
G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dapprich,
A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz,
J. Cioslowski and D. J. Fox, Gaussian 09, Revision A.1,
¨
V. N. Staroverov, R. Kobayashi, J. Normand, K. Raghavachari,
A. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi,
N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross,
Gaussian, Inc., Wallingford, CT, 2009.
V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. 26 K. S. Kim and L. Qian, Tetrahedron Lett., 1993, 34, 7677.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, 27 A. Albert, J. Chem. Soc., 1948, 1225.
J. W. Ochterski, R. L. Martin, K. Morokuma, V. G. Zakrzewski, 28 Z. Wang, H. Fu, Y. Jiang and Y. Zhao, Synlett, 2008, 2540.
2
418 New J. Chem., 2013, 37, 2408--2418
This journal is c The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2013