Amine exchange in formamidines: An experimental and theoretical study

Article abstract of DOI:10.1002/chem.201002389

N-H-containing formamidines combine a reasonably strong association to carboxylic acids to form complexes of well-defined geometries with a simultaneous proton-induced electrophilicity enhancement that allows for the exchange of their amine portion. The N

Full text of DOI:10.1002/chem.201002389

DOI: 10.1002/chem.201002389
Amine Exchange in Formamidines: An Experimental and Theoretical Study
Marinha dF. Capela, Nicholas J. Mosey, Liyan Xing, Ruiyao Wang, and
Anne Petitjean*[a]
À
Abstract: N H-containing formami-
dines combine a reasonably strong as-
sociation to carboxylic acids to form
complexes of well-defined geometries
same carbon, in which both N-contain-
ing fragments can be exchanged. Con-
sidering the proton-induced sensitisa-
tures by dynamic amine exchange
within formamidines. This study high-
lights three exchange regimes based on
the nature of the incoming amine (ali-
phatic amines, aromatic amines and
alkoxyamines), as well as exchange
rules based on the amine leaving
groups. Following this analysis, a proof
of concept for carboxylic acid templat-
ed macrocycle formation through dy-
namic exchange is provided.
À
tion of both C N units and the well-de-
with
a
simultaneous proton-induced
enhancement that
fined formamidine–carboxylic acid
complex geometry, it should be possi-
ble to use carboxylic acids as templates
for the synthesis of defined architec-
electrophilicity
allows for the exchange of their amine
portion. The N=C(H) NH fragment,
therefore, undergoes “imine-like” ex-
change with N-containing nucleophiles.
Because of the prototropic equilibrium,
À
Keywords: amidines · amines · nu-
cleophilic substitution · self-assem-
bly · supramolecular chemistry
À
the N=C(H) NH fragment may
behave as a “bisimine” centred on the
Introduction
In this context, it is all the more striking that little is
known about exchanges in C=N bonds of amidines. Ami-
dines are popular ligands for mononuclear, as well as dinu-
clear, metal coordination,[12] are among the most popular
DNA groove binders,[13] and have led to elegant self-assem-
bled architectures.[14] Furthermore, amidines have recently
received much attention due to their reversible interaction
with CO2, which has triggered exciting developments in
switchable solvents, as well as switchable catalysts.[15] The
The field of dynamic combinatorial chemistry, in which func-
tional groups involved in covalent bonds are exchanged in a
trigger-sensitive manner, has exploded in the last 10 years.[1]
The initial inspiring concepts[2] have turned into a powerful
method for the evolution of materials,[3a] catalysts[3b] and po-
tential drugs,[2i–j,3c] although calculations still have to define
the limits and potential of such combinatorial methods.[3d]
Several covalent bonds are amenable to exchange, from
À
N=C(R) N fragment of an amidine is extremely diverse in
[1b,4]
À
C C bonds (as in Diels–Alder
and metathesis reac-
1) shape due to the geometrical isomerism around the C=N
tions[1b,5]) to C O bonds (e.g., acetals
and hemiace-
double bond,[16a] combined with the C N s-trans and s-cis
[1b,6a]
À
À
tals[6b]), S S bonds
and C=N bonds,[1b,8,9,10] to name a
conformations, reminiscent of the amide partial double
bond (Scheme 1a); 2) basicity (from weakly basic for aryla-
midines to fairly basic for alkylamidines); and 3) in binding
[1b,7]
À
few. Several reviews highlight the potential and challenges
of each functional group.[1,2] A substantial amount of work
has been deservedly devoted to the dynamic exchange of
C=N double bonds in imines,[8] hydrazones,[9] acylhydra-
zones[10] and oximes,[1b,11] especially considering that the ex-
change of such functional groups can be tuned by coordina-
tion to a cation (from protons to metal cations), is compati-
ble with aqueous solutions (and hence with biomolecular
targeting) and can add geometrical isomerism around the
C=N double bond to the diversity of exchanging species.
À
modes for N H-containing amidines, as the prototropic
equilibrium allows the exchange of the C=N double bond
[16b,c]
À
along the N=C(R) N fragment (Scheme 1a).
Overall,
one would think that the complexity of the amidine frag-
ment, combined with the power of dynamic combinatorial
chemistry, would lead to promising developments.
More specifically in the context of dynamic exchange
within formamidines, the realization that 1) amine exchang-
es may be promoted by protons (in an analogous manner to
the proton-catalyzed exchange in imines), 2) in contrast to
imines, amidines are capable of two exchanges on the elec-
[a] M. dF. Capela, Dr. N. J. Mosey, L. Xing, Dr. R. Wang,
Dr. A. Petitjean
À
trophilic central carbon atom in the N=C(H) N(H) frag-
Department of Chemistry, Queenꢀs University
90 Bader Lane, Kingston, ON K7L 3N6 (Canada)
Fax : (+1)613-533-6669
ment (Scheme 1b), and 3) the binding of the proton from
carboxylic acid, which is directional (6-membered hydrogen-
bonded ring, Scheme 1b), has the potential to template and
direct the exchange of amines toward the selection of partic-
ular architectures. The principle of templated-architecture
E-mail: anne.petitjean@chem.queensu.ca
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201002389.
4598
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4598 – 4612

FULL PAPER
Scheme 2. N,N’-Diarylformamidines used in this manuscript (top: bisme-
thoxyphenyl and aliphatic formamidines, bottom: dipyridylformami-
dines). Abbreviations for the molecules and the proton numbering
scheme are indicated.
Scheme 1. a) Dynamic geometric, conformational and prototropic equili-
bria in the formamidine family; b) carboxylic acid binding and possible
constitutional amine exchanges within the formamidine framework;
c) principle of template-directed amine exchange within the formamidine
framework leading to the formation of macrocycles.
lection and chiral control of helical structures.[14c–h] Polycar-
boxylic acids have been used to assemble several amidine
fragments around them, either quite far from[14c,d,f,g,h] or in
close proximity to each other.[14e] The latter is necessary for
templated exchange between N-containing fragments, and
has never been reported for formamidines. Structural infor-
mation is useful to assess the accessibility of the reactive,
central formamidine carbon atom. In addition, to better un-
derstand the applicability of templated assembly prior to
constitutional exchange in solution, some quantitative data
regarding the strength of the formamidine–carboxylic acid
assembly in organic solvents is necessary and is described
below.
As illustrated in the crystal structures reported in
Figure 1, the respective orientation of carboxylic acid groups
(e.g., para- vs. meta-positions on a phenyl ring) dictate the
spatial distribution of the assembled formamidines. It is
therefore possible to control the relative distance between
electrophilic formamidine CH groups through the use of the
appropriate bis- or triscarboxylic acid partner.
In CDCl3, the self-assembled architectures are main-
tained, as shown by the Nuclear Overhauser effects ob-
served between the aromatic protons of the carboxylic acid
and the protons (a in Scheme 2) of the bisarylformamidine
(see the Supporting Information). This is consistent with the
NMR and CD spectroscopy studies of more complex amidi-
nes.[14c–h] The stability of the carboxylic acid/formamidine
complex in CDCl3 has been assessed by titration (a repre-
sentative example is given in Figure 2) and summarized in
Table 1. The high association constant of the more basic ali-
selection, illustrated for macrocycles, is shown in Scheme 1c.
As far as we know, very few reports deal with the exchange
of the nitrogen-containing fragment in formamidines.[17] All
cases mainly focus on preparative methods to access elec-
tron-rich products, generally starting with N,N-disubstituted
formamidines.[17] In these structures, the reciprocal hydrogen
bonding with a carboxylic acid (complex in Scheme 1b)
does not exist. Herein, a study of the exchange of the
N-containing fragment in N,N’-disubstituted formamidines
(which are hydrogen-bonding donors and acceptors) pro-
moted by carboxylic acids is presented. By varying the
chemical nature of the incoming nucleophile (R3-NH2, R4-
NH2 in Scheme 1b), three different regimes are defined.
The nature of the leaving group (R1-NH2, R2-NH2) also in-
fluences the rate of the exchange reaction, as presented
below. Finally, based on this understanding of the roles of
the nucleophiles and leaving groups, a proof of principle for
the templated dynamic assembly of macrocycles is given.
The formamidines considered in this manuscript, as well as
their abbreviations and proton numbering, are illustrated in
Scheme 2.
Results and Discussion
Self-assembly of formamidines around carboxylic acids:
Amidines have been used as hydrogen-bonding partners for
carboxylic acids for the past 15 years,[14] either through “lat-
eral” interactions in diamidine structures[14a] or, more re-
cently, through frontal interactions in monoamidine motifs
in the solid state,[14e,g] in solution[14c–f] and on surfaces.[14b,f]
The described amidinium–carboxylate interactions have re-
vealed an amazing potential in the elegant development, se-
phatic formamidine
ACHTNUGTERN(NGNU C12)2 with carboxylic acid 1 cannot be
determined by direct titration, but is accessible through
competitive titration of the p2–1 complex with ACTHUNTRGNE(UNG C12)2 (see
the Supporting Information).
As shown in Table 1, the association constants (Kassoc) in
CDCl3 are quite high and follow the relative ranking of the
pKa values. The self-assembly largely results from a proton
Chem. Eur. J. 2011, 17, 4598 – 4612
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4599

A. Petitjean et al.
Table 1. Measured logACTHUNRGTNEUN(G Kassoc), in which Kassoc is the association constant
between 1 and formamidines in CDCl3 (fitted by ChemEquili software;
pKa of the formamidines).[18]
Formamidine[a]
log(Kassoc
pKa (in EtOH, 258C) 5.9[a]
o2
3.02Æ0.03 3.7Æ0.5 4.00Æ0.15
6.7 8.0
m2
p2
ACHTNUGRTEN(NNUG C12)2
G
)
8.32Æ0.35
12.4[b]
[a] Estimated from the pKa of the amine precursor.[18b] [b] pKa for
N,N’-dihexylformamidine.[18b]
actions (an initial concentration of the electrophilic forma-
midine and of the carboxylic acid template larger than
10À3 m ensures that most of the formamidine is involved in a
complex).
Exchange of the nitrogen-containing fragment of formami-
dines with nucleophilic amines in the presence of carboxylic
acids:
Regime 1: aliphatic amine nucleophiles: An aliphatic amine
is nucleophilic enough to displace both arylamines in
N,N’-diarylformamidines. As shown in Figure 3, the first ar-
ylamine is displaced very quickly (at mm concentrations and
room temperature in CDCl3). The substitution of the second
arylamine takes days to com-
Figure 1. Molecular views of the self-assembly of p2 with a) terephthalic
acid (0.5 equiv) and b) trimesic acid (0.5 equiv) in the solid state. Select-
À
À
À
ed distances (ꢂ): a) N1 O1: 2.684; N2 O2: 2.668; b) N1 O2: 2.670,
À
À
À
N2 O1: 2.808, O6 N3: 2.782, O5 N4: 2.660.
plete at room temperature. Al-
though bis-substitution is ach-
ieved by the nucleophilic ali-
phatic amine after several days
at room temperature (possibly
making it an appropriate ex-
change partner to elaborate
larger architectures, as in
1
Scheme 1c), the H NMR anal-
ysis of the exchange process
also confirms what the relative
pKa values suggest: the intro-
duction of the aliphatic amine
nucleophile
(pKa (aqueous
EtOH, 258C)=9.7) first dis-
rupts the formamidinium–car-
boxylate complex (this is more
obvious with other carboxylic
acids such as acetic acid; see
Figure 9 below). As a result,
the use of oligocarboxylic acids,
such as isophthalic acid, as tem-
plates for the synthesis of well-
defined oligoformamidines does
not lead to discrete species as
the integrity of the self-assem-
bled oligo(formamidinium–car-
Figure 2. 1H NMR spectroscopic titration (300 MHz, CDCl3, 258C) of p2 with 1: left) aromatic region (fitting in
insert; Hf in grey, a’ and b’ protons on the carboxylic acid are underlined with dotted black lines); right) OCH3
signal. Fitting was performed with ChemEquili software, version 7.23, 2009, V. P. Solov’ev, Moscow University.
exchange that leads to formamidinium–carboxylate com-
plexes, consistent with those reported in the literature for
amidine–carboxylic acids.[14] In addition, the association con-
stants determined herein point to the fact that carboxylic
acids may be used as templates in preparative exchange re-
boxylate) complex is lost, and most of the carboxylate sites
are no longer bound to the formamidinium electrophile, but
to the ammonium ions.
Despite revealing the inapplicability of aliphatic amines
to templated exchange (as proposed in Scheme 1c), the
4600
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4598 – 4612

Amine Exchange in Formamidines
FULL PAPER
Figure 3. Constitutional exchange of amines in formamidines by aliphatic amines promoted by carboxylic acids. a) proton exchange equilibrium followed
by amine exchange; b) aromatic region of the 1H NMR spectrum upon exchange (300 MHz, CDCl3, 258C, [p2]0 =[1]0 =40 mm, [H25C12 NH2]0 =78 mm);
c) OCH3 and NCH2 region of the H NMR spectrum upon exchange (same conditions).
À
1
study of the constitutional exchange with aliphatic amines
tested on the methoxyaryl-based formamidines was able to
highlight different kinetic behaviour: compound o2 under-
goes mono-substitution faster than m2, which itself reacts
faster than p2, as shown by kinetic analysis by using
1H NMR spectroscopy with one equivalent of aliphatic
amine (see Figure 4). Because of the subsequent bis-substi-
tution by the aliphatic amine, pseudo-first-order kinetics
conditions are difficult to achieve to more thoroughly study
the differential reactivity between o2, m2 and p2. In addition,
the initial partitioning of the carboxylate between the form-
amidinium and the aliphatic ammonium species complicates
the thorough analysis of the differential kinetics. To a first
approximation, the reaction half-lives (t1/2) may serve as a
measure of reactivity. In the conditions used in Figure 4,
these are 2, 14 and 21 min for o2, m2 and p2, respectively,
and show an enhanced reactivity for the ortho-substituted
Figure 4. Kinetics of mono-substitution of N,N’-diarylformamidines by
dodecylamine (1H NMR, CDCl3, 400 MHz, [formamidine]0 =[1]0 =
À
[H25C12 NH2]0 =14 mm, 258C). Lines are simply connecting the data
points (no fitting is involved). Stacked 1H NMR spectra are provided in
the Supporting Information.
reagent. The analysis of the differential kinetic behaviour
between different substituent orientations is more easily
quantified with the amines used below.
Chem. Eur. J. 2011, 17, 4598 – 4612
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4601

A. Petitjean et al.
Regime 2: Exchange when bound to carboxylic acids (al-
koxyamines): To preserve the initial formamidinium–carbox-
ylate complex during the constitutional exchange, the use of
a less basic nucleophile is necessary. Alkoxyamines are elec-
tron-rich amines with little basicity. Their pKa (4.6 in H2O at
258C)[18c] should allow the initial formamidinium–carboxyl-
ate complex to remain intact upon addition of the nucleo-
philic amine. Indeed, 1H NMR spectroscopic analysis in
CDCl3 (Figure 5) shows that the alkoxyamine does not dis-
koxyamine fragment and the hydrogen from the arylamine
locks the mono-substituted product into a very weakly basic,
difficult to activate formamidine (see the mono-substitution
product in Figure 5a).[19b] As a result, bis-substitution does
not take place. Although this is a drawback as far as tem-
plated bisamine exchange is concerned (Scheme 1c), this
property is beneficial in the exploration of the relative reac-
tivity of o2, m2 and p2, as pseudo-first-order conditions may
be used in conjunction with UV/Vis spectroscopy for quanti-
tative analysis. Kinetic analysis under pseudo-first-order
conditions was therefore conducted by UV/Vis spectroscopy
and the relative rates were compared for different
N,N’-diaryl- or N,N’-diheteroarylformamidines. Upon reac-
tion, the conjugated diarylformamidine is transformed into a
product with less electronic conjugation (see Figure 5a for
the mono-substitution product) and, as a result, the highest
wavelength of maximum absorption (around 320 nm), which
disappears in the product, may be used to monitor reaction
rates, as shown in Figure 6 for the o2 reagent (in the UV/Vis
Figure 6. Typical UV/Vis spectroscopy reaction profile for mono-substitu-
tion of o2 with dodecylalkoxyamine promoted by acetic acid (258C, fresh-
À
À
ly filtered CHCl3, [o2]0 =[AcOH]0 =1.1 mm, [H25C12 O NH2]0 =11 mm).
Insert: the analysis of the decay of the absorbance at 330 nm gives access
to the pseudo-first-order rate constant which is compared to that of p2 in
Table 2. The related spectral curves for the other formamidines are given
in the Supporting Information.
Figure 5. a) Mono-substitution exchange reaction with alkoxyamines; b)
and c) aliphatic and aromatic portions of the H NMR spectra after addi-
1
tion of alkoxyamine and after the complete reaction (note that the chem-
ical shifts of the Hf and OCH3 protons in p2 are not affected by the initial
alkoxyamine addition, as represented by the black dashed lines at ꢀ8.22
and 3.84 ppm). The spectra were recorded in filtered CDCl3, at 300 MHz,
spectroscopic study, the aromatic carboxylic acid was re-
placed by UV/Vis-silent acetic acid).
Table 2 summarizes the relative rate constants for two
families of formamidines. In the methoxyaryl family (o2, m2,
p2), the relative rates confirm the reactivity trend observed
with alkylamines (Figure 4).
Once again, the ortho-substituted formamidine reacts
faster than its meta- and para-substituted counterparts. How-
ever, their reactivity in terms of rate constants is not
extremely different (the relative rate constants translate to
activation free energy differences in the order of
3–4 kJmolÀ1). In the general exchange mechanism that is in-
spired by transesterification,[20] it is reasonable to consider
three main steps (Scheme 3), among which the proton trans-
fer is probably the fastest (step 2), and the nucleophilic
attack (step 1) and nucleofuge departure (step 3) would
mostly contribute to the overall rate. Step 3 is dictated by
leaving-group ability, which is conventionally related to the
À
À
258C, [p2]0 =[1]0 =[H25C12 O NH2]0 =27 mm.
rupt the formamidinium–carboxylate complex, as anticipat-
ed based on pKa considerations: the chemical shifts of Hf
and OCH3 remain high upon alkoxyamine addition (spectra
1 and 2 in Figures 5b and c), which is consistent with bound
formamidinium (compare to the end points (top spectra) of
the Figure 2 titration).
Amine exchange still occurs within the formamidinium–
carboxylate complex but, in contrast to aliphatic amines, al-
koxyamines do not easily allow the exchange of both aryla-
mines in N,N’-diarylformamidines, even in the presence of
excess alkoxyamine.[19a] Indeed, the first substitution leads to
a change in the formamidine configuration and to the pro-
duction of a mono-substituted product in which a 5-mem-
bered hydrogen-bonded ring involving the oxygen of the al-
4602
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4598 – 4612

Amine Exchange in Formamidines
FULL PAPER
Table 2. Top: the observed relative rate constants for the reaction of the
formamidinium–acetate complex with dodecylalkoxyamine (the relative
rate constants are reported against that of formamidine p2). The reaction
conditions are the same as for Figure 5 (reactions were run in duplicate).
Bottom: the proton affinity of the respective amines measured as the
binding constant of TFA to the respective amine in CD3OD (1H NMR,
258C, fitted with ChemEquili 2009).
group ability may be involved in the differential reactivity,
and not as definitive evidence; indeed, the relative pKa
values in methanol may not easily be transferred to the
analysis of a reaction performed in chloroform.
Regarding step 1 of the mechanism, one cannot help but
wonder if the presence of an electron-rich site in the vicinity
of the active centre (e.g., o-methoxy in o2) could provide
some assistance in recruiting the nucleophilic amine, and/or
stabilizing the corresponding intermediate (intermediate 1 in
Scheme 3). To investigate this possible effect, pyridyl-based
formamidines, which present a much more basic nitrogen
site than an ether oxygen, were investigated. Two isomeric
bispyridylformamidines containing a nitrogen in the vicinity
Formamidine
k/k(p2)
o2
m2
p2
4.6Æ0.3
3.2Æ0.3
1.0
A
40Æ4
E
48Æ4
G
0.6Æ0.1
Amine
logACHTGUNETRN(NUG Kassoc) TFA
of the reaction centre
unit (4P)2 (see Scheme 2) were therefore prepared and their
reactivity compared. As reported in Table 2, the kinetic be-
haviour of the 2-pyridyl (2P)2 and 4-pyridyl (4P)2 moieties
ACHTUNGERTN(NUNG 2P)2 or away from the formamidine
o-methoxyaniline
m-methoxyaniline
p-methoxyaniline
2.9Æ0.2
3.26Æ0.03
3.6Æ0.2
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
is very different. The substitution of di(4-pyridyl)formami-
dine is slower than that of p2 (roughly half the rate con-
stant), whereas that of the 2-pyridyl isomer is very much
faster (roughly forty times faster), which indeed suggests a
possible facilitating role if the basic nitrogen is close to the
reaction centre.[21] The results of a theoretical study dissect-
ing proximity effects in the reaction pathway are detailed
further below.
Regime 3: Exchange whilst retaining the formamidinium–car-
boxylate complex during bis-substitution (arylamines): De-
spite the simplified analysis of their exchange, alkoxyamines
are only able to perform mono-substitution on the formami-
dine motif. Although this property is interesting in itself, it
lacks the potential illustrated in Scheme 1c, namely the pos-
À
sibility of performing templated exchange of both C N
Scheme 3. Simplified mechanism for the amine exchange within formami-
dines inspired by transesterification.[20] Step 1: amine attack on the acti-
vated formamidinium; step 2: proton exchange; step 3: expulsion of the
amine leaving group; step 4: proton exchange and isomerisation (only op-
erates when an alkoxyamine is the initial nucleophile; note that, in this
case, steps 3 and 4 may merge into one, with the formation of the cis-C=
N bond occurring while the leaving amine departs).
bonds and, consequently, of forming macrocyclic architec-
tures. Arylamines are another family of nucleophiles that
are likely to leave the formamidinium–carboxylate complex
undisturbed (formamidines are typically more basic than
their corresponding amines by at least 2 pKa units, as deter-
mined in EtOH),[18b] and may be nucleophilic enough to
effect amine exchange. In addition, as highlighted in
regime 2, one may take advantage of the faster kinetics of
the ortho-formamidines to substitute the amine fragment
with a para-substituted aniline, which would, in addition,
lead to a more thermodynamically stable formamidinium–
carboxylate complex (Table 1).
pKa of the leaving group (better leaving groups being associ-
ated with less basic groups). To assess a possible qualitative
correlation between the relative rates of reactions among
the methoxyaryl-derived formamidines upon reaction with
the alkoxyamine nucleophile, it was interesting to compare
the pKa values of the released methoxyarylamines. Because
the literature reports a range of pKa values under a variety
of different conditions, an independent, systematic study of
the proton affinity of the free methoxyarylamines was un-
dertaken by direct titration of the free amine with trifluoro-
acetic acid (TFA) in CD3OD. The fitted binding constants
Arylamine exchange was therefore studied by using p-me-
thoxyaniline as the nucleophile, and the o2, m2,
AHCTUNGTRENNUNG
ACHTUNGRTEN(NUNG 2P)2 and
spectroscopy gives important information about this ex-
change reaction, as illustrated in Figure 7b and c, in which
p-methoxyaniline is the incoming nucleophile and m2 is the
formamidine electrophile. First, substitution occurs, and
even one equivalent of p-methoxyaniline leads to significant
bis-substitution (in contrast to the alkoxyamine in regime 2)
as shown by the p2 Hf signal around 8.1 ppm in spectrum 4
(Figure 7b). After 11 h of exchange reaction (Figure 7b,
spectrum 4), the m2/mp/p2 proportions are 30:52:17. Second,
(logACHTUNGTRENNUNG(Kassoc) TFA) are given in Table 2. A stronger binding
constant is observed for p-methoxyaniline, compared to
m-methoxyaniline and to o-methoxyaniline, which correlates
with the observed kinetic effects (p-methoxyaniline, a stron-
ger base, would be a poorer leaving group). However, the
correlation can only be taken as an indication that leaving
Chem. Eur. J. 2011, 17, 4598 – 4612
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4603

A. Petitjean et al.
Figure 7. Exchange reaction with arylamines: a) reaction scheme for the bis-substitution of diarylformamidine by arylamines; b) and c) aromatic and
methyl regions of the 1H NMR spectra of 1) free m2, 2) m2 +p-methoxyaniline (1.0 equiv), 3) m2 +p-methoxyaniline+acetic acid 3 min after addition,
4) 11 h after addition and 5) after the introduction of m-methoxyaniline (1.5 equiv; 45 min after the last addition). Note the binding of m2 to acetic acid
(dashed grey lines between spectra 2 and 3) in the presence of the p-methoxyaniline nucleophile. The spectra are recorded in filtered CDCl3, at
À
À
500 MHz, 258C, [m2]0 =[AcOH]0 =[H25C12 O NH2]0 =57 mm (spectrum 3). The grey arrows in spectrum 5 indicate the change in the measured intensity
upon addition of m-methoxyaniline.
the m2 formamidinium–carboxylate complex remains intact,
even in the presence of the p-methoxyaniline before the ex-
change takes place, as shown by the typical chemical shifts
for bound Hf and OCH3 groups around 8.27 and 3.84 ppm,
respectively (Figure 7b and c, respectively spectrum 3, bind-
ing is highlighted by the dotted grey line; titration of m2
with carboxylic acid 1 is given in the Supporting Information
for comparison). Finally, the exchange is reversible in
chloroform: addition of m-methoxyaniline (1.5 equiv) to the
equilibrated mixture (from spectrum 4 to spectrum 5 in Fig-
ure 7b and c) leads to 1) a decrease in mp Hf intensity (Fig-
ure 7b), 2) an increase in m2 Hf intensity (reverse exchange;
Figure 7b) and 3) an increase in p-methoxyaniline OCH3 in-
tensity (as a leaving group in the reverse exchange, Fig-
ure 7c). After 45 min (following the addition of m-methoxy-
aniline (1.5 equiv)), the m2/mp/p2 distribution changes from
30:52:17 to 40:44:16; although the latter distribution may
not reflect complete equilibration, it is proof of the reversi-
bility of the exchange reaction.
By using a default of nucleophilic p-methoxyaniline, it is
possible to compare the kinetics of mono-substitution for
different diarylformamidines. The graphical results are pre-
sented in Figure 8 and confirm the trend observed for the al-
koxyamine nucleophile by UV/Vis spectroscopy: in the ex-
Figure 8. Comparison of the kinetics of amine exchange between the in-
coming p-methoxyaniline and the N-containing fragment in various for-
mamidines (see Figure 7a for the reaction scheme in which the formami-
dine is m2) measured by 1H NMR spectroscopy. The intensity of the
OCH3 group (normalized to 1) in the product of mono-substitution is
plotted as a function of time. The 1H NMR spectra were recorded in fil-
À
À
tered CDCl3, [formamidine]0 =[AcOH]0 =40 mm, [p-H3CO C6H4
NH2]0 =32 mm, 258C (400 MHz). The experiment was run in duplicate.
Half-lives for the reacting formamidines under these conditions are indi-
cated in the insert.
two main effects may be invoked: 1) a possible stabilization
of intermediate 1 after the addition step (Scheme 3) through
the participation of the nearby O or N lone pairs (“ortho
effect”) and 2) enhanced leaving group ability in the elimi-
nation step. To experimentally investigate these aspects,
mixed formamidines may be of interest. As shown by the
change reaction, o2,
whereas (4P)2 reacts much more slowly (similar to m2).
Once again, the formamidines containing a Lewis base
proximal to the reaction centre (o2, (2P)2 and (2MP)2) are
more reactive than the other formamidines. As in regime 2,
ACHUTNGTRENNG(U 2P)2 and ACHTUTGNREN(NUGN 2MP)2 react very quickly,
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
4604
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4598 – 4612

Amine Exchange in Formamidines
FULL PAPER
appearance of mp in Figure 7b, it is possible to prepare
mixed N,N’-diarylformamidines, in which the aryl groups are
different.[17b] Cascade exchanges may, therefore, be studied;
in particular, formamidine o2 may initially be transformed
into mixed formamidine op, which can then undergo amine
exchange with an aliphatic amine, for instance. The inter-
mediate op formamidine should, indeed, have interesting re-
activity; within the same molecule, both leaving groups
(ortho- and para-methoxyaniline) are present, allowing test-
ing of the impact of leaving group on kinetics. At the same
time, reacting molecule op also benefits from a possible sta-
bilization provided by the ortho-oxygen lone pair upon
verted into oC12; the disappearance of o2 despite its low
level testifies to its higher reactivity compared with the
other formamidines present), and 2) op and p2 immediately
dissociate from acetic acid (upfield shift in Hf, Figure 9,
spectrum 4; the chemical shifts move downfield again as the
basic aliphatic amine gets consumed). With time, the signal
intensity for Hf from p2 stays constant (no reaction with do-
decylamine), and most importantly, op is transformed into
oC12 by releasing p-methoxyaniline and into pC12 by releas-
ing o-methoxyaniline. The kinetics of this process can be as-
sessed by following the growth of the oC12 and pC12 Hf
proton signals as a function of time (Figure 9, spectra 4–7).
The appearance of the two products correlates with the
gradual disappearance of their
1
amine addition. Figure 9 reports the H NMR spectroscopic
common
op
precursor
(Figure 9) and the appearance
of the released o- and p-me-
thoxyanilines
(signals
not
shown in Figure 9).
As shown by the change in
intensities, the growth of pC12 is
about twice as fast as that of
oC12. Since both products come
from the same op precursor
and intermediates, within which
the same “ortho effect” would
be active, the observed kinetic
difference is a reflection of the
leaving group ability of o- and
p-methoxyaniline. A factor of 2
in the relative rates therefore
accompanies the contribution
of leaving group ability (under
these conditions). Interestingly,
this observation made by NMR
spectroscopy on a nonsymmet-
rical formamidine (op) being
mono-substituted by an aliphat-
ic amine echoes the results ob-
tained in the UV/Vis spectro-
Figure 9. Cascade of amine exchange studied by 1H NMR spectroscopy: arylamine exchange (formamidine Hf
region). The arylamine exchange between o2 formamidine and p-anisidine (spectrum 1) is activated by acetic
acid, leading to the mixed N-o-methoxyaryl-N’-p-methoxyarylformamidine, op, as the major product accompa-
nied by a very minor amount of N,N’-bis-p-methoxyarylformamidine p2 and remaining starting material o2
(spectra 2 and 3). The mixed formamidine is then exchanged with an alkylamine, and the reaction monitored
1
À
À
by H NMR spectroscopy (spectra 4 to 7). Reaction conditions: [o2]0 =[AcOH]0 =[p-H3CO C6H4 NH2]0 =
À
51 mm (spectra 2 and 3); [H25C12 NH2]0 =32 mm (filtered CDCl3, 218C, 400 MHz). Association of o2 with
acetic acid is highlighted with grey dotted lines between spectra 1 and 2. Dissociation of the op–acetic acid scopic study of mono-substitu-
complex upon addition of the basic dodecylamine is highlighted with grey dotted lines between spectra 3 and
4. The changes in relative intensities upon nucleophilic attack by dodecylamine are indicated in black (with
dotted curved arrows).
tion of symmetrical diarylfor-
mamidines by alkoxyamine.
Formamidine o2 (containing
two o-methoxyphenyl substitu-
monitoring of such a cascade reaction. In Figure 9, spec-
trum 1, only o2 and p-methoxyaniline are present (no ex-
change reaction in the absence of the activating acid). Upon
addition of acetic acid, o2 becomes associated with the acid
(downfield shift for Hf shown in Figure 9, spectrum 2), the
exchange process begins (the appearance of op in Figure 9,
spectrum 2) and the mixture equilibrates after 20 min
(Figure 9, spectrum 3; also consistent with Figure 8). Note
that, in the case of o2, the mixed op formamidine is by far
the major component (o2/op/p2 distribution ꢀ15:75:10; com-
pare with m2 in Figure 7). Upon addition of dodecylamine,
1) the little remaining o2 completely disappears (it is con-
ents) reacted four times faster than p2 towards nucleophilic
À À
attack by H2N O C12H25. Here, the reaction of the “o” por-
tion of op is twice as fast as that of the “p” portion (one
o-methoxyphenyl substituent only). Although these are only
correlations and require a more thorough mechanistic study,
these results suggest that, within the methoxyarylformami-
dine family at least, the leaving group ability is an important
contributor to the relative rates.
Theoretical study of the amine exchange in bisarylformami-
dine by DFT: To gain some insight into the actual mecha-
nism of the exchange reaction in the presence of acetic acid,
Chem. Eur. J. 2011, 17, 4598 – 4612
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4605

A. Petitjean et al.
and into the differential reactivity of formamidines substitut-
ed with Lewis bases in various positions, DFT calculations
were performed, and species along the reaction pathway
identified and compared.
pending upon the particular aryl substituents used. TS2–3
À
leads to the formation of a C N bond between the me-
thoxyamine and the formamidine. The length of the hydro-
gen bond between a hydrogen on the amine group of the
methoxyamine and one of the oxygen atoms in the acetate
component in this transition state has decreased significant-
ly, indicating a strong interaction between these two compo-
Mechanism: Calculations were performed to identify the re-
action mechanisms for the reaction of methoxyamine with
À
formamidines o2, m2, p2,
acetic acid. The only viable reaction pathway identified is
shown in Scheme 4.
(2P)2 and
E
nents. Upon formation of INT3, the C N bond has formed
and one of the hydrogen atoms on the amine group of what
was originally the methoxyamine has been transferred to
the acetate to reform acetic
acid. At this stage in the reac-
À
tion, the O H bond of acetic
acid forms a hydrogen bond to
the nitrogen atom that was
originally in the methoxyamine,
whereas the other oxygen atom
in the acetic acid moiety forms
hydrogen bonds to the hydro-
gen atoms on both of the nitro-
gen atoms in what was original-
ly the formamidine. The system
then proceeds through TS3–4,
which involves a reorientation
of the acetic acid component,
À
such that the O H bond in this
component is shifted from
pointing towards the nitrogen
atom that was initially part of
the methoxyamine in INT3 to
one of the nitrogen atoms in
what was originally the forma-
midine in INT4. TS4–5 involves
the transfer of the proton in
À
this O H bond to one of the ni-
trogen atoms in what was origi-
nally the formamidine, leading
to the elimination of an aryla-
mine. INT5 consists of acetic
acid coordinated to the forma-
midine produced through the
elimination process, which then
decomposes to yield the prod-
ucts.
Scheme 4. Reaction mechanism considered in the calculations. The names given under each structure corre-
spond to labels used throughout the text. R=OMe, and Ar=any of the aryl/heteroaryl groups used in the cal-
culations. Transition states are represented in square brackets with a °. Arrows indicating equilibria on either
side of the transition states are not intended to suggest that an equilibrium exists between the transition states
and minima, but rather indicate the equilibrium established between the intermediates that are connected by
the transition state. The “pre-product” is the result of amine exchange before tautomerism to the final Z prod-
uct (see Figure 5 for a drawing of the final isomer). Hydrogen bonding is represented by grey dotted lines.
The identified reaction mech-
anism shows that acetic acid
plays a key role in shuttling a
The first step involves the transfer of a proton from acetic
acid to the formamidine to yield a hydrogen-bonded ion
pair (INT1). The methoxyamine then approaches this com-
plex to form INT2. In this structure, one of the hydrogen
atoms in the amine group of the methoxyamine forms a hy-
drogen bond with one of the oxygen atoms in the acetate
component. No significant interaction between the nitrogen
atom of the methoxyamine and the central carbon atom of
the formamidine exists at this point along the reaction path-
proton from the incoming methoxyamine to the outgoing ar-
ylamine. This is consistent with experimental data showing
that the acetic acid is bound to the formamidine throughout
the process. This nucleophile activation and proton shuttling
performed by carboxylic acids is very reminiscent of cataly-
sis by enzymes, such as hydrolases.[22]
Energetics: Calculated free-energy profiles for all five sub-
strates are shown in Figure 10. Additional numerical data
regarding the potential and free energies are given in
À
way, with C N distances ranging from 3.30 to 3.85 ꢂ, de-
4606
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4598 – 4612

Amine Exchange in Formamidines
FULL PAPER
Figure 10. Free-energy profiles for the reaction of methoxyamine with a) o2, m2 and p2 and b) ACHTUNRGTENNUG(2P)2 and HCATUNGTERN(NUGN 4P)2.
Table 3. In all cases, the energies are given relative to that
of INT1. The free-energy plots show that the transition
states for addition and elimination, TS2–3 and TS4–5, re-
spectively, have very similar energies. This suggests that
either step is potentially rate limiting. These two steps along
the reaction pathway are explored in greater detail below to
shed light on the experimentally-observed kinetics. It is
noted that limitations exist regarding the calculations. Spe-
cifically, the calculations were performed in the gas phase,
whereas the experiments were performed in solvent. Fur-
thermore, many of the calculated energy differences are on
the order of a few kcalmolÀ1, which is comparable to, or
even below, the point at which definitive comparisons can
be made at the B3LYP/6-31GACTHNURGTNE(UNG d,p) level. As such, the fol-
lowing analysis should be taken to be qualitative in nature.
The free-energy barriers to addition were evaluated by
comparing the free energy of TS2–3 with that of INT2,
which is the preceding intermediate along the reaction path-
way. It should be noted that the data in Figure 10 suggest
that INT2 is not a minimum energy point along the pathway,
with no barrier separating this intermediate from the lower
energy INT1 structure. Normal-mode analysis showed that
INT1 and INT2 were minima on the potential energy sur-
face, and thus must be separated by a barrier. As such, it is
appropriate to use INT2 as the reference point when calcu-
lating the barrier to addition. Note that extensive efforts to
locate the transition state between INT1 and INT2 were un-
successful; however, this step is unlikely to affect the kinet-
ics of the reaction, and thus the absence of the specific
value of the barrier between INT1 and INT2 is unlikely to
alter the conclusions of the following analysis. The calculat-
ed free-energy barriers are 17.7, 18.3 and 19.6 kcalmolÀ1 for
o2, m2 and p2, respectively. By comparison, the free-energy
Table 3. Calculated potential and free energies relative to INT1 for all
groups, along with transition-state-theory rate constants relative to the
“para” form of each “substituent”.
o2
m2
p2
G
ACHTUNGRTEN(NUNG 4P)2
potential energies [kcalmolÀ1
]
reactant
INT1
16.2
0.0
18.1
0.0
18.6
0.0
16.2
0.0
17.9
0.0
INT2
0.1
0.8
0.2
0.5
1.6
TS2–3
INT3
TS3–4
INT4
15.0
0.9
4.6
16.3
2.5
6.7
17.3
4.1
8.0
12.8
À3.1
2.7
15.7
À0.4
4.2
3.8
6.5
7.4
2.5
4.4
TS4–5
INT5
14.4
4.9
16.8
7.5
17.3
8.2
13.5
3.6
18.0
5.8
pre-products
5.3
8.8
10.7
5.7
5.8
free energies [kcalmolÀ1
]
reactant
INT1
3.9
0.0
6.6
0.0
7.4
0.0
5.8
0.0
6.6
0.0
barriers are 14.8 and 17.6 kcalmolÀ1 for
spectively.
ACHUTNGTRNENUG(2P)2 and ACHTUNGTNER(NUGN 4P)2, re-
INT2
TS2–3
INT3
TS3–4
INT4
TS4–5
INT5
9.8
10.9
29.2
14.2
19.1
19.1
29.7
6.6
10.3
29.9
16.2
20.8
19.9
30.3
7.5
11.8
26.7
10.6
16.1
15.9
28.0
4.0
11.0
28.6
11.7
16.6
16.6
30.9
5.3
27.5
13.2
17.0
16.3
26.9
4.1
The calculation of the barrier for elimination introduces
complexities regarding whether the barrier should be evalu-
ated with respect to INT3 or INT4. Essentially, the barrier
preventing the reversion of INT4 to INT3 is less than 1 kcal
molÀ1 in all cases, and thus the question arises as to whether
the system resides in INT4 for a significant portion of time,
or proceeds directly through this intermediate while under-
going the elimination process. In addition, the equilibrium
constants associated with these two intermediates range
from 7.9ꢃ108 to 3.4ꢃ1010 in favour of INT3, suggesting that
only a minor concentration of INT4 will be present at any
pre-products
À7.2
À4.2
À1.1
À4.6
À5.3
relative rate constants[a]
addition
24.5
2.2
0.8
8.6
0.1
0.7
1.0
1.0
1.0
109.4
20.6
41.1
1.0
1.0
1.0
elimination[b]
elimination[c]
[a] Rate constants are given relative to that of the para form of each sub-
stituent. [b] Barrier for elimination calculated relative to INT3. [c] Barri-
er for elimination calculated relative to INT4.
Chem. Eur. J. 2011, 17, 4598 – 4612
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4607

A. Petitjean et al.
time. Given the ambiguity in defining the appropriate refer-
ence point for evaluating the barriers to elimination, both
cases will be considered. The barriers to elimination calcu-
lated with INT4 as the reference state are 10.6, 10.7 and
10.5 kcalmolÀ1 for o2, m2 and p2, respectively, and 12.1 and
14.3 kcalmolÀ1 for
ACHTUNGTRENNUNG(2P)2 and ACHTUNGTERN(NUGN 4P)2, respectively. If INT3 is
used as the reference state, the barriers are 13.7, 15.5 and
14.2 kcalmolÀ1 for o2, m2 and p2, respectively, and 17.4 and
19.2 kcalmolÀ1 for
ACHTUNGTRENNUNG(2P)2 and ACHTUNGTREN(NUGN 4P)2, respectively.
The calculated free-energy barriers indicate that addition
of the alkoxyamine contributes most to the reaction kinetics
of o2, m2 and p2. First, the barrier to addition for each of
these systems is higher than that for the corresponding elim-
ination process, regardless of whether INT3 or INT4 is used
as the reference state. Second, the barriers for the addition
step, and corresponding rate constants, increase as o2 <m2 <
p2, which is consistent with the experimental data given in
Table 2. The experimental trend in rate constants is not re-
covered for these systems if elimination is treated as the
rate-limiting step. Once again, we note that this conclusion
must be taken with caution given the small differences in
the energy barriers for the o2, m2 and p2 systems.
The free-energy barriers do not permit the unambiguous
identification of the rate-limiting step for ACHTNUTRGNEN(UG 2P)2 and ACHTUGNTERN(NUGN 4P)2. If
INT4 is used as the reference state for calculating the elimi-
nation barrier, the addition is the rate-limiting step. Howev-
er, elimination is rate limiting if INT3 is used to calculate
Figure 11. Structures demonstrating hydrogen bonding in the addition
and elimination steps that lead to the observed “ortho” effect. a) INT2
for o2; b) TS2–3 for o2; c) TS4–5 for ACHTNUTRGENNUG(2P)2. Distances relating to the rele-
vant hydrogen bonds are indicated. Oxygen atoms are black, nitrogen
atoms are dark grey, hydrogen atoms are light grey and carbon atoms are
very light grey.
the barrier. In all cases, the rate constants suggest that
reacts faster than (4P)2, which is consistent with experimen-
tal results. The kinetic data in Table 2 suggest that k((2P)2)/
k((4P)2)=80. The data in Table 3 show that the calculated
ACHTUNGRTEN(NUNG 2P)2
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
G
atom, with this distance decreasing from 2.526 ꢂ in INT2 to
2.400 ꢂ in TS2–3. Once again, this will stabilize the transi-
tion state. Formamidines m2, p2 and ACTHNUTRGNEUNG(4P)2 do not form intra-
molecular hydrogen bonds and thus cannot benefit from
these stabilization effects.
The lower elimination barrier for ACTHNUTRGENUNG(2P)2 can also be ac-
relative rate constant for addition (109.4) is closest to the
experimental value. However, the values obtained for the
elimination processes (20.6 and 41.1) are also in agreement
with experiment considering the limited accuracy of rate
constants estimated from quantum chemical calculations.
Overall, the data cannot clearly identify which step is rate
limiting, and in all likelihood both steps influence the kinet-
ics.
One clear feature observed during the experiments that is
consistently reproduced in the calculations is that the barri-
ers are lower if the methoxy groups or nitrogen atoms are in
“ortho” positions than if they are located at other positions
on the ring. This can be attributed to changes in hydrogen
bonding within these systems as they progress along the re-
action coordinate. Structures illustrating this effect for o2 are
shown in Figure 11. In INT2 (Figure 11a), the distance be-
tween the methoxy group and the hydrogen atom on the
nearest nitrogen atom (indicated) is 2.614 ꢂ. This large dis-
tance is observed because the relevant phenyl ring is rotated
out of the plane of the rest of the formamidine in this struc-
ture. Planarity has been restored in TS2–3 (Figure 11b),
which causes this distance to decrease to 2.207 ꢂ. The asso-
ciated increase in hydrogen bonding will stabilize the transi-
counted for through changes in hydrogen bonding. Essen-
tially, during the elimination process rotation occurs about
À
the dissociating C N bond. In TS4–5 (Figure 11c), this rota-
tion brings one of the hydrogen atoms of the amine leaving
group into close proximity to the nitrogen atom in the other
À
pyridine to form a hydrogen bond with an N H distance of
2.311 ꢂ. This hydrogen bond is not found in any other struc-
ture along the reaction pathway. As such, this hydrogen
bond stabilizes the transition state relative to both INT3 and
INT4. The formation of this hydrogen bond does not occur
for
N
ACHTUNGRTENUN(NG 2P)2 is lower
than that for ACHTUNGTRENNUNG
m2 or p2, which is reflected in the fact that the elimination
barriers for these systems do not exhibit a clear trend with
the position of the methoxy groups.
Constitutional exchange between arylamines as a route to
templated macrocycle synthesis: proof of concept: Irrespec-
tive of the driving force behind the higher reactivity of o2
tion state. A similar effect was observed for ACTHNUGTRENUNG(2P)2. In this
case, the relevant interaction involved the nitrogen in the
pyridine ring and the hydrogen on the nearest nitrogen
(or
ACHUTNGTREN(NUG 2P)2 and ACHUTNGTRNEN(GUN 2MP)2), such formamidines 1) bind to car-
boxylic acids and 2) undergo constitutional exchange with
4608
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4598 – 4612

Amine Exchange in Formamidines
FULL PAPER
Figure 12. a) reaction scheme for templated macrocycle synthesis by constitutional arylamine exchange; b) MS (ESI) spectrum of the crude reaction mix-
ture after exchange and washing off of the biscarboxylate template (reaction conditions: [o2]0 =[3]0 =2 mm; [2]0 =1 mm in anhydrous CH2Cl2, reflux,
28 h). R=C16H33.
arylamine while retaining the integrity of the formamidini-
um–carboxylate complex. These are the requirements for
templated constitutional exchange and formation of larger
architectures (as illustrated in Scheme 1). To test the con-
cept, o2 was assembled around and activated with 2,2-dihex-
adecylmalonic acid (2), and the o-methoxyamine displaced
by a bridged bis-functional meta-substituted arylamine (3,
Figure 12a). The MS (ESI) spectrum of the reaction mixture
(after washing the biscarboxylate off with aqueous base)
shows the formation of the bis-substituted macrocycle in the
mono- and bis-protonated states (as well as excess diamine
and intermediates or degraded intermediates on the way to
cyclisation). No oligomerisation products were detected at
higher molecular weight, supporting the templating and acti-
vating roles of the biscarboxylic acid. As a result, it seems
possible to take advantage of the geometrically well-defined
formamidinium–carboxylate complex in order to 1) activate
the formamidine carbon towards nucleophilic attack, and
Scheme 5. Summary of the three exchange regimes observed with alkyla-
mines, alkoxyamines and arylamines.
2) effect the constitutional exchange of arylamines within
the formamidinium–carboxylate complex, taking advantage
of the higher reactivity of the o-methoxyaryl-based form-
amidines.
(regime 1), bis-substitution occurs, but with dissociation of
the formamidinium–carboxylate complex (hence, the direc-
Conclusion
tional aspect of this particular complex cannot be used).
Conversely, alkoxyamines do not disrupt the formamidini-
um–carboxylate complex, do perform exchange reactions,
but do not proceed to bis-substitution (regime 2). Finally, ar-
ylamines are able to perform mono and bis-substitution
Overall, through this study, three regimes of constitutional
exchange in formamidines were identified (Scheme 5). If al-
kylamines are used as the nucleophilic exchange partners
Chem. Eur. J. 2011, 17, 4598 – 4612
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4609

A. Petitjean et al.
within the formamidinium–carboxylate complex (regime 3).
The ion-pair complex is 1) not disrupted by the arylamine
nucleophile, 2) activated towards nucleophilic attack and
3) significantly more reactive when the involved formami-
dine contains an ortho-methoxy or 2-pyridyl group (“ortho
effect”). This enhanced reactivity was rationalised by hydro-
gen bonding in the addition and/or elimination transition
states, through DFT calculations. Combining these observa-
tions, it was possible to show a proof of concept for the tem-
plated synthesis of a defined macrocycle around a biscarbox-
ylic acid template and activator.
Clearly, such dynamic arylamine exchange applied to tem-
plated synthesis opens the door to many avenues and future
studies. For example, regarding macrocyclisation, what are
the effects of 1) the spacer length, flexibility and location
(m vs. p positions in 3), and 2) the nature of the carboxylic
acid (aliphatic vs. aromatic, bis- vs. triscarboxylic acid) on
the isolated yields and nature of the products (e.g., if a bis-
carboxylic acid favours a 1+1 macrocycle, does a triscarbox-
ylic acid truly lead to a 1+1+1 macrocycle, as expected?).
Furthermore, the actual structure of the macrocycle/bis-
carboxylate complexes is yet to be determined. In Fig-
ure 12a, if the macrocycle is, on average, in the plane, where
are the R groups pointing to? Would it be possible to use
this arylamine exchange to assemble rotaxanes by building
the macrocycle around a biscarboxylic acid thread (see the
Supporting Information for a crude model of the related
pseudorotaxane)?[23] Applied to the field of topological mol-
ecules, formamidines are unique in the sense that they com-
bine, in one chemical function, both the interaction points
holding units together (e.g., through carboxylic acid binding)
and the reactive units allowing for cyclisation. This is very
different from the traditional dichotomy exemplified by the
coordination/metathesis duo for macrocyclisation.[23]
Finally, the properties of the isolated macrocycles are also
of interest: 1) the presence of the bridging spacer between
two formamidine motifs will probably tune the coordination
modes of the macrocyclic oligoamidines (e.g., with dinuclear
complexes[12]) in a way that is likely to be spacer dependent
(length and nature); 2) multi-functional guests, such as bis-
carboxylates and bis- and trisphosphates, are likely to have
a high affinity for the oligoformamidine macrocycles, and
some selectivity may be imparted by the spacer; 3) amidines
reversibly form ion pairs with carbonic acid generated from
CO2 and water, and this interaction has been used to elabo-
rate switchable solvents;[15] the concentration of multiple
formamidinium sites within the macrocycles may lead to in-
teresting CO2-capture properties; and 4) formamidines are
precursors to imidazolium salts, which themselves may be
converted into N-heterocyclic carbenes, an important family
of ligands for catalysis.[24] Macrocyclic oligoformamidines
may bring some interesting reactivity to the associated car-
bene-based derivatives.
Experimental Section
Synthesis: All formamidines used are known compounds and were syn-
thesised by heating with triethyl orthoformate and characterised by
1H NMR spectroscopy and melting point measurements.[25] Carboxylic
acids 1 and 2 and diamine 3, as well as dodecylalkoxyamine are known
compounds and were synthesised through different routes that are de-
scribed in the Supporting Information.
Physical studies: See the Supporting Information and Figure captions for
1H NMR and UV/Vis spectroscopy reaction conditions, as well as crystal-
lographic details. CCDC-789240 (2 p2 +terephthalic acid) and 789241 (2
p2 +trimesic acid) contain the supplementary crystallographic data for
this paper. These data can be obtained free of charge from The Cam-
bridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif. All calculations were performed by using density functional
theory[26, 27] with the B3LYP hybrid exchange-correlation functional[28,29]
and the 6–31GACTHNUTRGENUGN(d,p) basis set. Tests illustrated that this methodology re-
produces the relative energies of o2, m2 and p2 to within 1.0 kcalmolÀ1 of
results obtained with larger basis sets and MP2 methods. All stationary
points were characterised as minima or first-order saddle points through
harmonic frequency analysis. Vibrational frequency calculations were
also used to evaluate free energies at 298 K and 1 atm. Intrinsic reaction
coordinate calculations were performed in selected cases to ensure the
transition states connected the correct minima. The calculations were
performed by using the Gaussian 03 software package.[30]
Acknowledgements
The authors would like to thank Dr Louis Cuccia, Dr Derek Pratt and
Prof. Stan Brown for useful discussions, and acknowledge Queenꢀs Uni-
versity, the Canadian Foundation for Innovation, the Ministry of Re-
search and Innovation (Ontario) and the Natural Sciences and Engineer-
ing Research Council of Canada (NSERC) for financial support.
[1] For recent reviews, see: a) B. de Bruin, P. Hauwert, J. N. H. Reek,
Angew. Chem. 2006, 118, 2726–2729; Angew. Chem. Int. Ed. 2006,
45, 2660–2663; b) P. T. Corbett, J. Leclaire, L. Vial, K. R. West, J.-L.
Wietor, J. K. M. Sanders, Chem. Rev. 2006, 106, 3652–3711; c) M. M.
Rozenman, B. R. McNaughton, D. R. Liu, Curr. Opin. Chem. Biol.
2007, 11, 259–268; d) O. Ramstrçm, J.-M. Lehn in Comprehensive
Medicinal Chemistry II, Vol 3 (Eds.: J. B. Taylor, D. J. Triggle),
Elsevier, 2006, pp. 959–976; e) J.-M. Lehn, Chem. Soc. Rev. 2007,
36, 151–160; f) S. Ladame, Org. Biomol. Chem. 2008, 6, 219–226;
g) L. Onel, N. J. Buurma, Annu. Rep. Prog. Chem., Sect. B: Org.
Chem. 2009, 105, 363–379; h) Dynamic Combinatorial Chemistry
(Eds.: J. N. H. Reek, S. Otto), Wiley-VCH, Weinheim, 2010.
[2] For early reviews, see: a) A. V. Eliseev, J.-M. Lehn, Curr. Top. Mi-
crobiol. Immunol. 1999, 243, 159–172; b) J.-M. Lehn, Chem. Eur. J.
1999, 5, 2455–2463; c) C. Karan, B. Miller, Drug Discovery Today
2000, 5, 67–75; d) K. M. J. Sanders, Pure Appl. Chem. 2000, 72,
2265–2274; e) M. Albrecht, J. Inclusion Phenom. Macrocyclic
Chem. 2000, 36, 127–151; f) I. Huc, J.-M. Lehn, Actual. Chim. 2000,
51–54; g) J.-M. Lehn, A. V. Eliseev, Science 2001, 291, 2331–2332;
h) I. Huc, R. Nguyen, Comb. Chem. High Throughput Screening
2001, 4, 53–74; i) S. Otto, R. L. E. Furlan, J. K. M. Sanders, Drug
Discovery Today 2002, 7, 117–125; j) O. Ramstrçm, J.-M. Lehn, Nat.
Rev. Drug Discovery 2002, 1, 26–36; k) S. J. Rowan, S. T. Cantrill,
G. R. L. Cousins, J. K. M. Sanders, Angew. Chem. 2002, 114, 938–
993; Angew. Chem. Int. Ed. 2002, 41, 898–952; l) S. Otto, R. L. E.
Furlan, J. K. M. Sanders, Curr. Opin. Chem. Biol. 2002, 6, 321–327;
m) O. Ramstrçm, T. Bunyapaiboonsri, S. Lohmann, J.-M. Lehn, Bio-
chim. Biophys. Acta Gen. Subj. 2002, 1572, 178–186; n) J. K. M.
Sanders, Philos. Trans. R. Soc. London Ser. A 2004, 362, 1239–1245.
4610
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4598 – 4612

Amine Exchange in Formamidines
FULL PAPER
[3] a) N. Giuseppone, J.-M. Lehn, Angew. Chem. 2006, 118, 4735–4740;
Angew. Chem. Int. Ed. 2006, 45, 4619–4624; b) S. Otto, Curr. Opin.
Drug. Discovery Dev. 2003, 6, 509–520; c) G. Gasparini, L. J. Prins,
P. Scrimin, Angew. Chem. 2008, 120, 2509–2513; Angew. Chem. Int.
Ed. Engl. 2008, 47, 2475–2479; d) A. V. Eliseev, Pharm. News 2002,
9, 207–215; e) A. Lee, J. G. Breitenbucher, Curr. Opin. Drug Dis-
covery Dev. 2003, 6, 494–508; f) M. Cano, S. Balasubramanian,
Drugs Future 2003, 28, 659–678; g) R. M. Sanchez-Martin, S.
Mittoo, M. Bradley, Curr. Top. Med. Chem. 2004, 4, 653–669;
h) J. L. Reymond, Angew. Chem. 2004, 116, 5695–5697; Angew.
Chem. Int. Ed. 2004, 43, 5577–5579; i) L. Weber, Drug Discovery
Today 2004, 1, 261–267; j) A. G. Orrillo, R. L. E. Furlan, J. Org.
Chem. 2010, 75, 211–214.
[4] a) S. Otto, R. L. E. Furlan, J. K. M. Sanders, Science 2002, 297, 590–
593; b) P. J. Boul, P. Reutenauer, J.-M. Lehn, Org. Lett. 2005, 7, 15–
18; c) P. Reutenauer, E. Buhler, P. J. Boul, S. J. Candau, J.-M. Lehn,
Chem. Eur. J. 2009, 15, 1893–1900.
[5] a) D. G. Hamilton, N. Feeder, S. J. Teat, J. K. M. Sanders, New J.
Chem. 1998, 22, 1019–1021; b) B. R. McNaughton, K. M. Bucholtz,
A. Camaano-Moure, B. L. Miller, Org. Lett. 2005, 7, 733–736;
c) P. G. M. van Gerven, J. A. A. W. Elemans, J. W. Gerritsen, S.
Speller, R. J. M. Nolte, A. E. Rowan, Chem. Commun. 2005, 3535–
3537; d) P. Wipf, G. S. Mahler, K. Okumura, Org. Lett. 2005, 7,
4438–4486; e) S.-A. Poulsen, L. F. Bornaghi, Bioorg. Med. Chem.
2006, 14, 3275–3284; f) P. Edwards, Drug Discovery Today 2007, 12,
497–498.
5864–5866; Angew. Chem. Int. Ed. 2007, 46, 5762–5764; d) M.-K.
Chung, C. M. Hebling, J. W. Jorgenson, K. Severin, S. J. Lee, M. R.
Gagne, J. Am. Chem. Soc. 2008, 130, 11819–11827; e) L. A. Inger-
man, M. L. Waters, J. Org. Chem. 2009, 74, 111–117; f) M. G. Simp-
son, M. Pittelkow, S. P. Watson, J. K. M. Sanders, Org. Biomol.
Chem. 2010, 8, 1173–1180; g) M. G. Simpson, M. Pittelkow, S. P.
Watson, J. K. M. Sanders, Org. Biomol. Chem. 2010, 8, 1181–1187.
[10] For recent examples involving acylhydrazones, see references [8f],
[9f] and [9g], as well as: a) N. Sreenivasachary, J.-M. Lehn, Proc.
Natl. Acad. Sci. USA 2005, 102, 5938–5943; b) D. T. Hickman, N.
Sreenivasachary, J.-M. Lehn, Helv. Chim. Acta 2008, 91, 1–20.
[11] a) N. Nazarpack-Kandlousy, J. Zweigenbaum, J. Henion, A. V. Eli-
seev, J. Comb. Chem. 1999, 1, 199–206; b) V. A. Polyakov, M. I.
Nelen, N. Nazarpack-Kandlousy, A. V. Eliseev, J. Phys. Org. Chem.
1999, 12, 357–363; c) V. Goral, M. I. Nelen, A. V. Eliseev, J.-M.
Lehn, Proc. Natl. Acad. Sci. USA 2001, 98, 1347–1352.
[12] a) J. Barker, M. Kilner, Coord. Chem. Rev. 1994, 133, 219–300;
b) F. A. Cotton, C. Lin, C. A. Murillo, Acc. Chem. Res. 2001, 34,
759–771; c) M. H. Chisholm, A. M. Macintosh, Chem. Rev. 2005,
105, 2949–2976; d) P. C. Junk, M. L. Cole, Chem. Commun. 2007,
1579–1590; e) F. T. Edelmann, Adv. Organomet. Chem. 2008, 57,
183–352.
[13] a) C. Bailly, I. O. Donkor, D. Gentle, M. Thornalley, M. J. Waring,
Mol. Pharmacol. 1994, 46, 313–322; b) F. Animati, F. M. Arcamone,
M. R. Conte, P. Felicetti, A. Galeone, L. Gomez-Paloma, P. Lombar-
di, L. Mayol, C. Rossi, J. Med. Chem. 1995, 38, 1140–1149; c) T. C.
Jenkins, A. N. Lane, Biochim. Biophys. Acta, Gene Struct. Expres-
sion 1997, 1350, 189–204; d) A. N. Lane, T. C. Jenkins, T. A. Fren-
kiel, Biochim. Biophys. Acta Gene Struct. Expression 1997, 1350,
205–220; e) L. Wei, C. Carrasco, A. Kumar, C. E. Stephens, C.
Bailly, D. W. Boykin, W. D. Wilson, Biochemistry 2001, 40, 2511–
2521; f) K. Bielawski, S. Wolczynski, A. Bielawska, Biol. Pharm.
Bull. 2001, 24, 704–706; g) F. A. Tanious, W. D. Wilson, D. A. Pat-
rick, R. R. Tidwell, P. Colson, C. Houssier, C. Tardy, C. Bailly, Eur.
J. Biochem. 2001, 268, 3455–3464; h) B. Nguyen, M. P. H. Lee, D.
Hamelberg, A. Joubert, C. Bailly, R. Brun, S. Neidle, W. D. Wilson,
J. Am. Chem. Soc. 2002, 124, 13680–13681; i) B. Nguyen, D. Hamel-
berg, C. Bailly, P. Colson, J. Stanek, R. Brun, S. Neidle, W. D.
Wilson, Biophys. J. 2004, 86, 1028–1041.
[14] a) M. W. Hosseini, R. Ruppert, P. Schaeffer, A. De Cian, N. Kyritsa-
kas, J. Fischer, J. Chem. Soc. Chem. Commun. 1994, 2135–2136;
b) F. Auer, G. Nelles, B. Sellergen, Chem. Eur. J. 2004, 10, 3232–
3240; c) M. Ikeda, T. Tanaka, T. Hasegawa, Y. Furusho, E. Yashima,
J. Am. Chem. Soc. 2006, 128, 6806–6807; d) T. Hasegawa, K.
Morino, Y. Tanaka, H. Katagiri, Y. Furusho, E. Yashima, Macromo-
lecules 2006, 39, 482–488; e) H. Katagiri, Y. Tanaka, Y. Furusho, E.
Yashima, Angew. Chem. 2007, 119, 2487–2491; Angew. Chem. Int.
Ed. 2007, 46, 2435–2439; f) T. Maeda, Y. Furusho, S.-I. Sakurai, J.
Kumaki, K. Okoshi, E. Yashima, J. Am. Chem. Soc. 2008, 130,
7938–7945; g) E. Yashima, K. Maeda, Y. Furusho, Acc. Chem. Res.
2008, 41, 1166–1180; h) H. Iida, M. Shimoyama, Y. Furusho, E. Ya-
shima, J. Org. Chem. 2010, 75, 417–423.
[15] a) T. Endo, D. Nagai, T. Monma, H. Yamaguchi, B. Ochiai, Macro-
molecules 2004, 37, 2007–2009; b) E. R. Pꢄrez, R. H. A. Santos,
M. T. P. Gambardella, L. G. M. De Macedo, U. P. Rodrigues-Filho,
J.-C. Launay, D. W. Douglas, J. Org. Chem. 2004, 69, 8005–8011;
c) Y. Liu, P. G. Jessop, M. Cunningham, C. A. Eckert, C. L. Liotta,
Science 2006, 313, 958–960; d) T. Yamada, P. J. Lukac, M. George,
R. G. Weiss, Chem. Mater. 2007, 19, 967–969; e) T. Yamada, P. J.
Lukac, T. Yu, R. G. Weiss, Chem. Mater. 2007, 19, 4761–4768; f) D.
Nagai, T. Endo, J. Polym. Sci., Part A: Polym. Chem. 2009, 47, 653–
657; g) S. L. Desset, D. J. Hamilton, Angew. Chem. 2009, 121, 1500–
1502; Angew. Chem. Int. Ed. 2009, 48, 1472–1474.
[16] a) L. Xing, C. Wiegert, A. Petitjean, J. Org. Chem. 2009, 74, 9513–
9516; b) S. Patai, The Chemistry of Amidines and Imidates, Vol. 1,
Wiley, New York, 1975; c) S. Patai, Z. Rappoport, The Chemistry of
Amidines and Imidates, Vol. 2, Wiley, New York, 1991.
[17] a) For amine exchange in N-acyl and N-tosyl N,N’-disubstituted ami-
dines, see: M. Ono, S. Tamura, Chem. Pharm. Bull. 1990, 38, 590–
[6] a) D. Berkovich-Berger, G. N. Lemcoff, Chem. Commun. 2008,
ˇ
1686–1688; b) D. Drahonovsky, J.-M. Lehn, J. Org. Chem. 2009, 74,
8428–8432.
[7] a) L. Milanesi, C. A. Huner, S. E. Sedelnikova, J. P. Waltho, Chem.
Eur. J. 2006, 12, 1081–1087; b) B. Danieli, A. Giardini, G. Lesma, D.
Passarella, B. Peretto, A. Sacchetti, A. Silvani, P. Alessandra, G.
Pratesi, F. Zunino, J. Org. Chem. 2006, 71, 2848–2853; c) Z. Pei, R.
Larsson, T. Aastrup, H. Anderson, J.-M. Lehn, O. Ramstrçm, Bio-
sens. Bioelectron. 2006, 22, 42–48; d) S. Andrꢄ, Z. Pei, H.-C. Siebert,
O. Ramstrçm, H.-J. Gabius, Bioorg. Med. Chem. 2006, 14, 6314–
6326; e) B. M. R. Liꢄnard, N. Selevsek, N. J. Oldham, C. J. Schofield,
ChemMedChem 2007, 2, 175–179; f) B. M. R. Liꢄnard, R. Hueting,
P. Lassaux, M. Galleni, J.-M. Frere, C. J. Schofield, J. Med. Chem.
2008, 51, 684–688; g) R. Pꢄrez-Fernꢅndez, M. Pittelkow, A. M. Be-
lenguer, J. K. M. Sanders, Chem. Commun. 2008, 1738–1740;
h) M. T. Cancilla, M. M. He, N. Viswanathan, R. L. Simmons, M.
Taylor, A. D. Fung, K. Cao, D. A. Erlanson, Bioorg. Med. Chem.
Lett. 2008, 18, 3978–3981; i) K. R. West, R. F. Ludlow, P. T. Corbett,
P. Besenius, F. M. Mansfeld, P. A. G. Cormack, D. C. Sherrington,
J. M. Goodman, M. C. A. Stuart, S. Otto, J. Am. Chem. Soc. 2008,
130, 10834–10835; j) H. Y. Au-Yeung, P. Pengo, G. D. Pantos, S.
Otto, J. K. M. Sanders, Chem. Commun. 2009, 419–421; k) H. Y.
Au-Yeung, G. D. Pantos, J. K. M. Sanders, J. Am. Chem. Soc. 2009,
131, 16030–16032.
[8] For recent examples involving imines, see: a) C. Godoy-Alcꢅntar,
A. K. Yatsimirsky, J.-M. Lehn, J. Phys. Org. Chem. 2005, 18, 979–
985; b) N. Giuseppone, J.-M. Lehn, Chem. Eur. J. 2006, 12, 1715–
1722; c) A. Valade, D. Urban, J.-M. Beau, ChemBioChem 2006, 7,
1023–1027; d) S. Xu, N. Giuseppone, J. Am. Chem. Soc. 2008, 130,
1826–1827; e) K. Ziach, J. Jurczak, Org. Lett. 2008, 10, 5159–5162;
f) A. Herrmann, N. Giuseppone, J.-M. Lehn, Chem. Eur. J. 2009, 15,
117–124; g) A. Herrmann, Org. Biomol. Chem. 2009, 7, 3195–3204;
h) G. Nasr, E. Petit, C. T. Supuran, J.-Y. Winum, M. Barboiu,
Bioorg. Med. Chem. Lett. 2009, 19, 6014–6017; i) J. Leclaire, G.
Husson, N. Dervaux, V. Delorme, L. Charles, F. Ziarelli, P. Desbois,
A. Chaumonnot, M. Jacquin, F. Fotiadu, G. Buono, J. Am. Chem.
Soc. 2010, 132, 3582–3593; j) A. Le-Gresley, N. Kuhnert, J. Chem.
Res. 2010, 61–67.
[9] For recent examples involving hydrazones, see: a) J. Liu, K. R. West,
C. R. Bondy, J. K. M. Sanders, Org. Biomol. Chem. 2007, 5, 778–
786; b) F. Bulos, S. L. Roberts, R. L. E. Ricardo, J. K. M. Sanders,
Chem. Commun. 2007, 3092–3093; c) R. F. Ludlow, J. Liu, H. Li,
S. L. Roberts, J. M. K. Sanders, S. Otto, Angew. Chem. 2007, 119,
Chem. Eur. J. 2011, 17, 4598 – 4612
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
4611

A. Petitjean et al.
596; b) for amine exchange in N,N’-disubstituted amidines, see: M.
Ono, R. Todoriki, S. Tamura, Chem. Pharm. Bull. 1990, 38, 866–
873; c) for amine exchange in N-formamidine ureas, see: D. D. Dꢆaz,
M. G. Finn, Chem. Eur. J. 2004, 10, 303–309; D. D. Dꢆaz, M. G.
Finn, Lett. Org. Chem. 2005, 2, 621–627; d) for amine exchange
starting from N,N-dimethylformamidines, see: D. D. Dꢆaz, W. G.
Lewis, M. G. Finn, Synlett 2005, 14, 2214–2218.
[23] For selected recent reviews, see: a) R. L. E. Furlan, S. Otto, J. K. M.
Sanders, Proc. Natl. Acad. Sci. USA 2002, 99, 4801–4804; b) J.-C.
Chambron, J.-P. Collin, V. Heitz, D. Jouvenot, J.-M. Kern, P.
Mobian, D. Pomeranc, J.-P. Sauvage, Eur. J. Org. Chem. 2004, 1627–
1638; c) N. Gimeno, R. Vilar, Coord. Chem. Rev. 2006, 250, 3161–
3189; d) M. S. Vickers, P. D. Beer, Chem. Soc. Rev. 2007, 36, 211–
225; e) S. J. Loeb, Chem. Soc. Rev. 2007, 36, 226–235; f) C. D.
Meyer, C. S. Joiner, J. F. Stoddart, Chem. Soc. Rev. 2007, 36, 1705–
1723; g) K. E. Griffiths, S. J. Fraser, Pure Appl. Chem. 2008, 80,
485–506; h) P. C. Haussmann, J. F. Stoddart, Chem. Rec. 2009, 9,
136–154; i) J. D. Crowley, S. M. Goldup, A.-L. Lee, D. A. Leigh,
R. T. McBurney, Chem. Soc. Rev. 2009, 38, 1530–1541; j) P. Gavina,
S. Tatay, Curr. Org. Synth. 2010, 7, 24–43.
[24] K. Hirano, S. Urban, C. Wang, F. Glorius, Org. Lett. 2009, 11, 1019–
1022.
[25] a) H. Komber, H.-H. Limbach, F. Bçhme, C. Kuner, J. Am. Chem.
Soc. 2002, 124, 11955–11963; b) A. Petitjean, L. Xing, R. Wang,
CrystEngComm 2010, 12, 1397–1400.
[26] W. Kohn, L. J. Sham, Phys. Rev. 1965, 140, A1133–A1138.
[27] P. Hohenberg, W. Kohn, Phys. Rev. 1964, 136, B864–B871.
[28] A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652.
[29] C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785–789.
[30] Gaussian 03, Revision C.02, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
´
[18] a) J. Oszczapowicz, R. Orlinski, E. Hejchman, Pol. J. Chem. 1979,
´
53, 1259–1265; b) J. Oszczapowicz, R. Orlinski, Pol. J. Chem. 1980,
54, 1901–1906; c) T. C. Bissot, R. W. Parry, D. H. Campbell, J. Am.
Chem. Soc. 1957, 79, 796–800.
À
À
[19] a) Up to 15 equivalents of H2N O R were used, and only mono
substitution was observed (even after several days of contact at
room temperature); b) the conformation in Figure 5a is consistent
with the reported crystal structures such as in: G. I. Birnbaum, B.
Kierdaszuk, D. Shugar, Nucleic Acids Res. 1984, 12, 2447–2460; J. B.
Noar, U. V. Venkataran, T. C. Bruice, G. Bollag, R. Whittle, D. Sam-
mons, Bioorg. Chem. 1986, 14, 17–27; J. E. Johnson, J. A. Maia, K.
Tan, A. Ghafouripour, P. de Meer, S. S. C. Chu, J. Heterocycl. Chem.
1986, 23, 1861–1868; T. Fujii, T. Saito, T. Date, Y. Nishibata, Chem.
Pharm. Bull. 1990, 38, 912–916; L. Van Meervelt, Acta Crystallogr.
Sect. C 1991, 47, 2635–2637; R. Hoos, A. B. Naughton, W. Thiel, A.
Vasella, W. Weber, K. Rupitz, S. G. Withers, Helv. Chim. Acta 1993,
76, 2666–2686; B. L. Booth, F. A. T. Costa, Z. Mahmood, R. G.
Pritchard, M. F. Proenca, J. Chem. Soc. Perkin Trans. 1 1999, 1853–
1858; B. L. Booth, F. A. T. Costa, R. Pritchard, M. F. Proenca, Syn-
thesis 2000, 1269–1278; W. Szczepankiewicz, T. Borowiak, M. Ku-
AHCTUNGTREGeNNUN ry, Jr., T. Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S.
Iyengar, J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani,
N. Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P. Hratchian,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J.
Dannenberg, V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C.
Strain, O. Farkas, D. K. Malick, A. D. Rabuck, K. Raghavachari,
J. B. Foresman, J. V. Ortiz, Q. Cui, A. G. Baboul, S. Clifford, J. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng,
A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wall-
ingford CT, 2004.
´
bicki, J. Suwinski, P. Wagner, Pol. J. Chem. 2002, 76, 1137–1142; T.
Kurz, K. Widyan, C. Wackendorff, K. Schluter, Synthesis 2004,
1987–1990; T. Kurz, D. Geffken, K. Widyan, Tetrahedron 2004, 60,
2409–2416; T. Kurz, Tetrahedron 2005, 61, 3091–3096; J. SaÅzewski,
Z. Brzozowski, M. Gdaniec, Tetrahedron 2005, 61, 5303–5309.
[20] a) N. C. Om Tapanes, D. A. Gomes Aranda, J. W. De Mesquita Car-
neiro, O. A. Ceva Antunes, Fuel 2008, 87, 2286–2295; b) D. M.
Alonso, M. L. Granados, R. Mariscal, A. Douhal, J. Catal. 2009, 262,
18–26.
[21] Within the pyridylformamidine family, knowledge of the pKa values
is not very relevant as the basicity of aminopyridines typically re-
flects that of the pyridine nitrogen, not that of the exocyclic amine
(which is the group departing in step 3 in Scheme 3).
[22] For selected examples, see: a) A. Maurady, A. Zdanov, D. de Moi-
ssac, D. Beaudry, J. Sygusch, J. Biol. Chem. 2002, 277, 9474–9483;
b) K. P. Bzymek, R. C. Holz, J. Biol. Chem. 2004, 279, 31018–31025;
c) D. Xu, H. Guo, J. Am. Chem. Soc. 2009, 131, 9780–9788.
Received: August 19, 2010
Published online: February 17, 2011
4612
www.chemeurj.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 4598 – 4612