Dynamic aminal-based TPA ligands

Article abstract of DOI:10.1002/chem.201500105

The use of dynamic covalent reactions (DCRs) is gaining popularity for the construction of self-assembling architectures. We have recently introduced DCRs that exchange alcohols and aldehydes to create hemiaminal ethers within tri(2-picolyl)amine (TPA) li

Full text of DOI:10.1002/chem.201500105

DOI: 10.1002/chem.201500105
Full Paper
&
Dynamic Covalent Reactions
Dynamic Aminal-Based TPA Ligands**
[a, b]
[b]
[a]
[c]
Yuntao Zhou,
Yaofeng Yuan, Lei You,* and Eric V. Anslyn*
Abstract: The use of dynamic covalent reactions (DCRs) is
gaining popularity for the construction of self-assembling
architectures. We have recently introduced DCRs that ex-
change alcohols and aldehydes to create hemiaminal ethers
within tri(2-picolyl)amine (TPA) ligands, all of which are tem-
the equilibrium constants from imines to aminals were cor-
+
related by a linear free energy relationship (LFER) with s
values. Dynamic component exchange was investigated as
a means of probing multiple equilibriums quantitatively in
the system. Further, the mechanism was analyzed with
a qualitative kinetics study. NMR spectra reveal the different
extents of two competing pathways for assembly depending
upon whether the aromatic amine has electron-withdrawing
or electron-donating groups on the ring. Finally, mass spec-
tral evidence supports the presence and differing extents of
dominance of the two pathways as a function of the
substituents.
II
plated by Zn . To expand the scope of this assembly, aromat-
ic imines derived from pyridine-2-carboxyaldehyde were ex-
plored as dynamic covalent receptors for di(2-picolyl)amine
II
in the presence of Zn to create TPA ligands that contain
aminal linkages. This represents another metal-templated in
situ multicomponent assembly. The stability of the assembly
was successfully modulated through substituent effects, and
Introduction
thermodynamic and kinetic properties often need to be
[9]
optimized for the development of DCRs for specific purposes.
The field of dynamic covalent chemistry (DCC) has been bur-
The chemistry of carbonyl compounds is highly diverse, en-
compassing condensations, alkylations, cyclizations, and so on,
and hence provides an excellent platform for the discovery of
new DCRs. Indeed, the aforementioned imine and hydrazone
DCRs are based on carbonyl reactivity, and are in part thermo-
dynamically driven by the loss of a water molecule. In contrast
to condensation with primary amines, the addition reactions of
aldehydes/ketones with alcohols, thiols, and secondary amines
are more sluggish and not as thermodynamically favored, and
[
1]
geoning in the past decade, with the ultimate goal of build-
[
2]
ing dynamic combinatorial libraries, constructing molecular
[
3]
assemblies and nanomaterials, and modulating biological
[
4]
structures and functions. Reversible covalent bonding can
lead to dynamic component exchange as does its noncovalent
counterpart, and dynamic covalent reactions (DCRs) lead to in
situ-generated molecular diversity and complexity with a much
higher efficiency than traditional stepwise covalent synthesis.
In spite of the increasing popularity in chemical and biological
applications, the chemical space for current DCRs is rather
limited. The most used DCRs are imine condensation and ex-
[10]
therefore, the formation of organic heterocycles or metal-
[11]
templated macrocycles are commonly used to stabilize the
products: Hemiacetals/acetals, hemithioacetals/thioacetals, and
hemiaminals/aminals, respectively.
[
5]
[6]
change, hydrazone condensation and exchange, disulfide
[
7]
[8]
exchange, and cyclic boronate ester formation. Both the
Recently, we reported a strategy of stabilizing hemiaminal
ethers for a broad range of secondary alcohols using in situ-
generated tri(2-picolyl)amine (TPA)-like metal complexes
[a] Y. Zhou, Prof. Dr. L. You
[12]
(
Scheme 1a).
The thermodynamic driving force resulting
State Key Laboratory of Structural Chemistry
Fujian Institute of Research on the Structure of Matter
Chinese Academy of Sciences
Fuzhou, 350002 (P.R. China)
E-mail: lyou@fjirsm.ac.cn
from metal coordination shifts the equilibrium toward the
desired product, which is otherwise unfavorable. Herein, we
describe that this four-component assembly that incorporates
alcohols creating hemiaminal ethers can be expanded to incor-
porate aromatic amines, thereby yielding aminals. The experi-
ments reveal the dependence of the assembly equilibrium and
mechanism on changes in substituents, as well as the ability to
exchange components of multiple equilibria in the assembly
mixture, further expanding the scope of DCRs.
[b] Y. Zhou, Prof. Dr. Y. Yuan
College of Chemistry, Fuzhou University
Fuzhou, 350116 (P.R. China)
[c] Prof. Dr. E. V. Anslyn
Department of Chemistry
The University of Texas at Austin
Austin, Texas, 78712 (USA)
E-mail: anslyn@austin.utexas.edu
[**] TPA=Tri(2-picolyl)amine.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201500105.
Chem. Eur. J. 2015, 21, 1 – 8
1
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
2
+
ed if no Zn was present (see the Supporting Information), in
agreement with our previous results demonstrating that the
metal ion is crucial to shift the equilibrium toward the hemi-
[12a]
aminal 2.
However, there were no corresponding peaks of
2
+
complex 2 or 3 even with the presence of Zn in the mixture,
although all aldehyde converted to imine 1. We reasoned that
there is not enough driving force to offset the relative
inertness of imine 1 under these conditions.
II
The lack of a coordinating counterion to the Zn in 3 means
that the complex must form solely with a solvent occupying
the open site on zinc. This paves the way for manipulating the
equilibrium using exogenous ligands. The assembly reaction
was hence conducted with 1 equivalent of tetrabutylammoni-
1
um chloride. After 24 h, no further changes in the H NMR
spectra were observed, and a similar pattern of resonances as
the tripodal TPA complexes when using alcohols were present.
The appearance of two sets of doublets at d=5.55 and
5
.70 ppm (H and H , J=12 Hz), as well as four sets of doublets
a b
between d=3.75 and 4.50 ppm (H , J=16 Hz, germinal cou-
d
pling: one at d=4.45 ppm; one at 4.25 ppm; the other two
around d=4.00 ppm overlapped with other peaks), indicates
the formation of complex 3, in which a new stereocenter
makes the four methylene protons diastereotopic (Figure 1).
Scheme 1. a) The multicomponent assembly with alcohols; b) The possible
pathways for multicomponent assembly with primary aromatic amines.
Results and Discussion
Design of a model system
The reaction of pyridine-2-carboxaldehyde (2-PA), di(2-picolyl)-
amine (DPA), zinc triflate, and an alcohol, creates hemiaminal
ether complexes in the presence of
a
Brønsted acid
[
12b]
(Scheme 1a).
We envisioned that analogous assemblies
could be performed by using primary aromatic amines (4), or
potentially from isolated imines. In contrast to the alcohol as-
sembly, in which the reaction proceeds through a tris(pyri-
dine)-substituted iminium ion 5, we reasoned that the use of
aromatic amines would proceed primarily through imine inter-
mediates due to the higher nucleophilicity of 4 relative to sec-
ondary alcohols. Thus, in Scheme 1b the prediction was that
the pathway from 4 and 2-PA to 3 would be dominated by
imines (1) rather than the iminium. Moreover, such an in situ
assembly of 2-PA and 4 would minimize the synthetic efforts
by avoiding the isolation of the imine, but would afford similar
equilibrium mixtures because the reaction is under thermo-
dynamic control.
1
Figure 1. a) The dynamic covalent assembly and b) its H NMR spectrum.
II
The resonances of methylene protons in DPA-Zn (6) are between d=3.75
and 4.25 ppm and overlap with another two peaks of benzylic protons (H )
d
in 3.
The resonances of methine (d=5.70 ppm, H ) and NH proton
a
Reaction screening
(
d=5.55 ppm, H ) were further assigned by proton exchange
b
To test our hypothesis, we screened the reaction of 2-PA, DPA,
after the addition of D O. The aromatic portion of the spec-
2
II
and 4-methylaniline under
a
variety of conditions. The
trum reveals the existence of imine 1, as well as the DPA-Zn
reactions were stirred with 3 ꢁ molecular sieves in deuterated
acetonitrile at room temperature overnight and studied by
complex (6), and hence these components are in equilibrium
with 3. The ratio of 1 and 3 is 1.35, reflecting a yield of 43%.
The formation of 3 was further validated by its corresponding
m/z peak of 494.4 [3+Cl] in the ESI mass spectrum (the Sup-
1
H NMR spectroscopy. As expected, the aldehyde peak at d=
10 ppm completely disappeared, and only imine 1 was detect-
&
&
Chem. Eur. J. 2015, 21, 1 – 8
www.chemeurj.org
2
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
ꢀ
ꢀ
ꢀ
porting Information, Figure S17). Br , I , and OAc counterions
activating and deactivating the ring, respectively. We postulat-
ed that the reactivity of imine 1 could be similarly fine-tuned
by the substituent on the aromatic amine due to conjugation
also increase the formation of 3, but to a lesser extent than
Cl . The enhancement effect of counteranions is likely due to
ꢀ
[14]
stabilization of zinc complex through ligand coordination in
of the benzene ring and the C=N functional group.
The
[
12]
the axial position.
physical organic tool of choice to study this would be the
Hammett linear free energy relationship (LEFR), which quanti-
[15]
fies substituent electronic effects. This LFER has been exten-
sively employed to correlate kinetic and thermodynamic data
Equilibrium constants
[16]
Having confirmed the formation of 3 by a multicomponent as-
sembly reaction, the next step was to measure the equilibrium
constants. Because multiple equilibria exist in the solution, we
set out to simplify the system. First, none or only a residue of
to elucidate associated mechanisms. However, the utilization
of Hammett equation for the development of new DCRs has
been rarely reported.
A series of aromatic amines with para or meta substituents,
including -OCH , -CH , -F, -Br, -CF , -CN, and -NO , were used in
2
-PA was detected after the equilibrium was reached. Second,
3
3
3
2
although imine 1 and the remaining DPA can both coordinate
the assembly reaction with 2-PA, DPA, zinc triflate, and tetrabu-
tylammonium chloride. The mixture was allowed to stir for
24 h, and the equilibrium constant was determined as de-
scribed above. The x, y, z, and K values are listed in Table S1
(the Supporting Information). In all cases, the corresponding
imine 1 and complex 3 are present at equilibrium, but to dif-
fering extents. Generally, electron-withdrawing groups facilitate
the formation of 3, whereas electron-donating groups favor
[
13]
zinc, the assumption was made that the non-reactive zinc is
predominately bound to DPA to form complex 6. Thus, we
chose the following reaction to derive the equilibrium
constants [Eq. 1], as this two-component system will directly
correlate with the reactivity of imine 1. All concentrations in
1
Equation (1) can be deduced from the H NMR integrals, as
well as the mass balance of 2-PA and zinc. By defining the inte-
gral ratio of 1 to 3 as x, and the total integral of other
possible products from 2-PA (i.e., 2 and 7, see Scheme 1b) to 3
as y (for 4-methylaniline, yꢁ0), the specific expression of K as
given in Equation (2) can be derived (see the details in the
Supporting Information).
ꢀ
1
imine 1. 4-Nitroaniline afforded the largest K value (304m ),
ꢀ1
whereas p-methoxy gave the worst (K=8.4m ).
With all equilibrium constants in hand, they were then sub-
jected to a Hammett analysis. An attempt to correlate log K
with s did not give a linear relationship, with p-OCH signifi-
3
cantly deviating from the line (the Supporting Information, Fig-
ure S4), indicative of different stabilizing or activating mecha-
nism. To quantify the resonance effect and separate it from the
+
inductive effect, s values were employed. This parameter ac-
counts for resonance effects as a consequence of direct reso-
nance between the reaction site and the para electron-donat-
[15b]
ing or electron-withdrawing group, respectively.
Since these
interactions are impossible when the substituent is attached to
the meta position, the original s values of meta substituents
K ¼ ½3ꢂ=f½1ꢂ½6ꢂg
ð1Þ
ð2Þ
+
are still valid. A plot of log K versus s indeed afforded
2
a linear line (Figure 2, r =0.981). The substituent p-F can be
K ¼ð1 þ x þ yÞ=fxðx þ yÞ½2-PAꢂtotal
g
+
modestly electron-donating (s =ꢀ0.07), and this is confirmed
ðx ¼ ½1ꢂ=½3ꢂ, y ¼ ð½2ꢂ þ ½7ꢂÞ=½3ꢂÞ
by the K values for 4-fluoroaniline and aniline derived
ꢀ
1
assembly, respectively (43 and 65m ).
For 4-methylaniline, the apparent equilibrium constant was
ꢀ
1
determined to be 25m , a value much smaller than the
II
assembly of 2-PA and DPA-Zn to create 2 (K around
ꢀ
1 [12a]
6
600m ).
This is in agreement with the higher stability of
an imine (starting point for assembly to 3 in [Eq. 1]) than its
parent aldehyde (starting point for the alcohol assembly,
Scheme 1a). Nevertheless, the equilibrium does reveal the
potential of aromatic imines for the development of new
DCRs.
LFER-based modulation
The next goal was to increase the stability of the assembly by
varying the electronic nature of the building blocks. It is well
known that functional groups on the aromatic rings signifi-
cantly affect their reactivity in electrophilic substitution reac-
tions, with electron-donating and electron-withdrawing group
II
Figure 2. Hammett plot for the reaction of imine 1 with DPA-Zn .
Chem. Eur. J. 2015, 21, 1 – 8
www.chemeurj.org
3
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
+
The linear correlation of log K with s was rationalized by
using resonance structures (Scheme 2). Because an equilibrium
exists between imine 1 and assembly 3, resonance effects on
either of these species will be reflected in the Hammett plot.
1
Figure 3. H NMR spectra of dynamic covalent assembly derived from a) 4-
methylaniline and c) 4-methoxyaniline and b) their component exchange.
Analogous component exchange was also conducted with
p-Br and p-CF -substituted amines (the Supporting Information,
3
Figures S8 and S9).
Scheme 2. Resonance structures of aromatic imines with a) p-OCH
b) p-NO
3
and
To study the associated dynamic covalent bonding within
the multicomponent assembly quantitatively, equilibrium
constants of a series of dynamic component exchanges were
calculated. Because imine 1, tripodal complex 3, as well as
their corresponding aromatic amines were present in the equi-
librium mixture, imine/amine exchange [Eq. (3)], aminal/amine
exchange [Eq. (4)], and imine/aminal exchange [Eq. (5)] were
investigated (Table 1). For example, the equilibrium constant
2
.
There is no resonance interaction between the substituents
and the aminal center of 3, but instead only induction would
be relevant. However, as with the quinoidal resonance interac-
tions in benzylic carbocations and phenoxides, resonance
structures can be drawn for imine 1. p-OCH donates a pair of
3
electrons to the electron-deficient pyridine to stabilize imine
1
(Scheme 2a). Moreover, the partial negative charge on the
Table 1. Equilibrium constants of exchange reactions.
imine carbon from a contributing structure explains the rela-
tive inertness of 1 (i.e., less of complex 3). In contrast, for the
electron-withdrawing nitro group, a quinoidal resonance inter-
action leads to an imine that is more electrophilic (Scheme 2b).
[a]
[b]
X
K
1
K
2
K
3
K
3
p-OCH
3
3.47
1.00
0.132
0.021
1.01
1.00
0.349
0.181
0.289
1.00
2.65
8.59
0.339
1.00
4.36
p-CH
p-Br
3
The high reactivity of 1(p-NO ) is further supported by a small
amount of aminal 7 in the reaction mixture. In another way of
2
p-CF
3
11.3
considering the data, because the resonance is in the reactant,
[a] Derived from exchange reactions [Eq. (5)] and used for LFER analysis.
[b] Calculated from Equation (6).
+
the s value is appropriate for correlating the dissociation of
complex 3 into imine 1 and 6 with a negative 1 value because
the reactant imine carbon is partially positively charged. The
reverse reaction (the reaction of 1 and 6 to create 3) would
thereby have a positive 1 value, as we observed.
(K ) for the reaction of 1(p-CH ) with p-methoxyaniline is 3.47,
1
3
therefore favoring 1(p-OCH ). The K values for other amines
3
1
are listed in Table 1, and their trend (p-OCH >p-CH >p-F>p-
3
3
CF ) is consistent with the nucleophilicity of the amine and
3
Dynamic component exchange
hence its reactivity toward 2-PA. Moreover, a linear relationship
+
Having modulated the DCRs through LFER analysis, compo-
nent exchange was conducted to verify the reversibility of the
assembly system. The reaction was first conducted with 3
equivalents of 4-methylaniline, followed by the addition of
equal equivalents of 4-methoxyaniline. The equilibrium was
was afforded when the value of log K was plotted against s
1
2
(slope=ꢀ1.63, r =0.985, see Figure 4). The negative 1 value is
due to stabilization of the product imine through a resonance
effect by electron-donating groups.
For the exchange of amines within aminal complex 3, there
1
reached after 24 h, and H NMR spectroscopy revealed a
is a decrease in equilibrium constant (K ) as the amine
2
decrease of 4-methylaniline-derived complex 3 with a concomi-
tant increase in the analogous assembly incorporating 4-me-
thoxyaniline (Figure 3). Upon the reversal of the sequence of
addition of amines, the same distribution of components was
observed. Moreover, the competition experiments afforded the
same results as the component exchange, thereby further
confirming the reversibility of the multicomponent assembly.
becomes less nucleophilic (p-OCH ꢁp-CH >p-F>p-CF ) as
3
3
3
may be expected. However, the difference between electron-
rich and electron-deficient amines on K2 (for example, K (p-
2
OCH )/K (p-CF )=5.58) is significantly smaller than that for
3
2
3
imine/amine exchange (in comparison, K (p-OCH )/K (p-CF )=
1
3
1
3
165). The influence of the nucleophilic character of the amine
should indeed be far less in complex 3 than in 1 because one
&
&
Chem. Eur. J. 2015, 21, 1 – 8
www.chemeurj.org
4
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Due to the differential response of K and K toward substi-
1
2
tution, the value of K increases as the arene becomes more
3
electron-deficient (p-OCH <p-CH <p-F<p-CF ). It is notable
3
3
3
that the equilibrium constant obtained from exchange reac-
tions is comparable to that calculated from Equations (2) and
(
6), considering that several approximations were made to
derive Equation (2). Analogous to the K value, the plot of log
+
2
K3 versus s is linear (R =0.998, Figure 4) with a positive
value (1.05).
1
Mechanism studies
+
Figure 4. Hammett plot for imine/amine exchange (log K
amine exchange (log K
1
vs. s ), aminal/
+
2
vs. s), and imine/aminal exchange (log K
3
vs. s ).
We next set out to further distinguish the two possible mecha-
nistic pathways given in Scheme 1b. Although DCRs are ther-
modynamically controlled, the elucidation of mechanism is
beneficial for future design as well as optimization of reaction
rates that are crucial to the practicality of DCRs. Qualitative ki-
netics experiments were performed to determine relative rates,
and to potentially spectroscopically observe the intermediates
and their transformations. The assembly reactions were run
with p-OCH , m-F, and p-NO -substituted amines in the
would expect the sigma bond strengths in 3 to be less affect-
ed by the identity of the amine than the combination of sigma
and pi bonds in 1. A second rationalization comes from the
thermodynamic driving force of zinc coordination, which mini-
[
17]
mizes the impact of the amines.
Because only inductive
effects exist between substituents X and the aminal center of
3
2
3
, log K was plotted against the standard Hammett parameter
absence and presence of 1 equiv of pyridinium triflate, and
2
2
1
s, and a linear line was obtained (r =0.981, Figure 4). The
lower sensitivity toward substitution for aminal/amine ex-
change than imine/amine exchange was also supported by the
smaller absolute value of the slope 1 (ꢀ0.972).
H NMR spectra were recorded after 1.5, 5, and 24 h, respec-
tively. These three aromatic amines were chosen to cover
+
a broad range of s values. The Brønsted acid was utilized as
a means of catalyzing the assembly as well as modulating the
reaction pathway. For electron-donating p-OCH , the equilibri-
3
um was reached after 1.5 h with or without acid (the Support-
ing Information, Figures S10 and S11). Both imine 1 and
complex 3 were observed, but no other intermediates were
apparent. These results indicate the domination of the imine
pathway, which is in agreement with the high nucleophilicity
of 4-methoxyamine.
For the 3-fluoroaniline assembly without acid, the hemiami-
nal 2 was present after 1.5 and 5 h (Figure 5b). But after 24 h,
only a trace amount of 2 was detected, suggesting
that 2 gradually transforms to complex 3 (the Sup-
porting Information, Figure S12). In the presence of
the acid, hemiaminal 2 was in the mixture after
K ¼ ½1ðp-XÞꢂ½4ðp-CH Þꢂ=f½1ðp-CH Þꢂ½4ðp-XÞꢂg
ð3Þ
1
3
3
1.5 h, but to a lesser extent (Figure 5c). After 5 h,
equilibrium was reached and no hemiaminal 2 was
observed (the Supporting Information, Figure S13).
We rationalize these observations as acceleration of
both pathways through acid catalysis, although the imine
mechanism still outweighs the competing iminium route.
When using the highly elec-
K ¼ ½3ðp-XÞꢂ½4ðp-CH Þꢂ=f½3ðp-CH Þꢂ½4ðp-XÞꢂg
ð4Þ
2
3
3
tron-withdrawing p-NO2 group
on the aromatic amine, aminal
3
was the major component
after 1.5 h, although significant
amounts of 2-PA and complex 2
existed (Figure 5d). However, at
this time period the percentage
of imine 1 was much smaller than with p-OCH and m-F-substi-
3
K ¼ ½3ðp-XÞꢂ½1ðp-CH Þꢂ=f½3ðp-CH Þꢂ½1ðp-XÞꢂg ¼ K =K
ð5Þ
ð6Þ
tuted amines. When the reaction was conducted for a longer
period (i.e., 5 and 24 h), the corresponding peaks of 2-PA as
well as hemiaminal 2 decreased while the amount of assembly
3
3
3
2
1
K ¼ ½Kðp-XÞꢂ=½Kðp-CH Þꢂ
3
3
Chem. Eur. J. 2015, 21, 1 – 8
www.chemeurj.org
5
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
plex 3 was observed in the assembly mixture with 4-methoxy-
aniline (the Supporting information, Figure S18) with an m/z of
510.0 [3+Cl], but not complex 2, which is in good agreement
with the NMR spectroscopic studies. For 3-fluoroaniline (the
Supporting Information, Figure S20), the corresponding 3 (m/
z=498.4, [3+Cl]) is also the major species, although both
hemiaminal 2 (m/z=405.1, [2+Cl]) and iminium 5 (m/z=
2
89.8) were detected, indicative of the emergence of the imini-
um pathway. In the case of 4-nitroaniline-derived assembly
the Supporting Information, Figure S19), complex 3 (m/z=
25.0, [3+Cl]) was present. However, significant abundance of
(m/z=405.1, [2+Cl]) and iminium 5 (m/z=289.3) were pres-
(
5
2
ent. These findings suggest that 5 is a viable intermediate, and
are also consistent with the low nucleophilicity of 4-nitro-
aniline.
Conclusion
Figure 5. The detection of intermediate 2. a) The control experiment: The
creation of 2 by the assembly of 2-PA and DPA-Zn ; b) and c) The assembly
with 3-fluoroaniline in the absence and presence of pyridinium triflate after
Dynamic covalent reactions (DCRs) of aromatic amines to
create imines that convert to TPA-like ligands were studied in
detail. The multicomponent assembly was conducted as
a means of creating imines in situ and controlling the compet-
ing pathway. The equilibrium reaction of 2-PA, DPA, zinc tri-
flate, and aromatic primary amines can be modulated by coun-
teranions and substituents. Halogen anions facilitate the as-
sembly by increasing the thermodynamic stability of the zinc
complex through coordination on the axial position of the
metal center. The stability of the assembly was further opti-
mized through substituent effects and correlated by a linear
II
1.5 h; d) and e) The assembly with 4-nitroaniline in the absence and
presence of pyridinium triflate after 1.5 h.
3
increased (the Supporting Information, Figure S14). In addi-
tion, the percentage of imine 1 gradually increased while the
peaks for aminal 7 appeared. We rationalize these results as
following: 1) Due to the relatively low nucleophility of 4-nitroa-
niline, its addition to 2-PA is slow and rivaled by DPA; 2) Once
the imine 1 forms, however, it rapidly reaches equilibrium with
aminals 3 and 7 as a result of its high reactivity; 3) Hemiaminal
+
free energy relationship (LFER) with s values. Electron-donat-
ing groups, such as p-OCH , stabilize the imine through quinoi-
3
2
also slowly transforms to 3. Thus, in summary, electron-do-
dal resonance interactions, whereas electron-withdrawing
nating groups on the aromatic amine direct the assembly pri-
marily through an imine, albeit with lower equilibrium con-
stants for full assembly. Electron-withdrawing groups on the
aromatic amine seem to access both mechanistic pathways
groups, such as p-NO and p-CN, destabilize the imine and in-
2
crease its reactivity. Moreover, the reversibility of the assembly
is validated by dynamic component exchange, and the mecha-
nism is elucidated by kinetics and mass spectral analysis. The
competing iminium pathway is more pronounced for aromatic
amines bearing an electron-withdrawing group. The equilibri-
um and mechanism insights revealed here should be beneficial
to other DCRs involving carbonyl derivatives.
(imine and iminium) and give overall more assembly.
The reaction pathways were also confirmed by the multi-
component assembly reaction in the absence of molecular
sieves. With p-OCH , m-F, and p-NO -substituted amines, hemi-
3
2
aminal 2 was observed in all cases, although its amount
increases as the arene becomes more electron-deficient (the
Supporting Information, Figure S16). This is reasonable because Experimental Section
without molecular sieves assisting imine formation the direct
All assembly reactions were performed in situ in acetonitrile with-
DPA addition to 2-PA is more involved. Hence, we have a
dynamic multicomponent assembly whose pathway can be
modulated by substitution, Brønsted acid, or the addition of
molecular sieves.
out isolation and purification. Activated 3 ꢁ molecular sieves (4 to
8 mesh), di-(2-picolyl)amine (DPA, 1.2 equiv), primary aromatic
amine (3.0 equiv), and tetrabutylammonium salts (1.0 equiv) were
added to a stirred solution of pyridine-2-carboxyaldehyde (2-PA,
5
0–55 mm, 1 equiv) and zinc triflate ([Zn(OTf) ], 1.0 equiv) in
2
[
D ]acetonitrile (0.6 mL). The mixture was stirred at room tempera-
3
Mass spectral analysis
ture overnight. The assembly solution was characterized by
H NMR spectroscopy and ESI-MS.
1
To further elucidate the assembly mechanism, ESI-MS experi-
ments were performed to characterize potential intermediates.
Because of the dynamic nature of these multicomponent as- Acknowledgements
sembly reactions and the lability of the components, fragmen-
tation is likely under mass spectral conditions. Nevertheless,
their information sheds light on possible intermediates. Com-
L.Y. thanks The Recruitment Program of Global Youth Experts
and Natural Science Foundation of Fujian Province, China
&
&
Chem. Eur. J. 2015, 21, 1 – 8
www.chemeurj.org
6
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
(
2014J05024) for support. E.V.A. acknowledges financial
Commun. 2013, 49, 5820–5822; e) K. R. West, K. D. Bake, S. Otto, Org.
Lett. 2005, 7, 2615–2618.
8] a) S. D. Bull, M. G. Davidson, J. M. H. van den Elsen, J. S. Fossey, A. T. A.
Jenkins, Y.-B. Jiang, Y. Kubo, F. Marken, K. Sakurai, J. Zhao, T. D. James,
Acc. Chem. Res. 2013, 46, 312–326; b) J. Tçnnemann, R. Scopelliti, K. O.
Zhurov, L. Menin, S. Dehnen, K. Severin, Chem. Eur. J. 2012, 18, 9939–
support from the National Science Foundation (CHE-1212971).
[
Keywords: imines · ligands · molecular recognition · self-
assembly · substituent effects · zinc
9
945; c) R. Nishiyabu, Y. Kubo, T. D. James, J. S. Fossey, Chem. Commun.
011, 47, 1124–1150.
9] a) Q. Ji, R. C. Lirag, O. S. Miljanic, Chem. Soc. Rev. 2014, 43, 1873–1884;
2
[
[1] a) S. J. Rowan, S. J. Cantrill, G. R. L. Cousins, J. K. M. Sanders, J. F. Stod-
ˇ
´
b) Q. Ji, O. S. Miljanic, J. Org. Chem. 2013, 78, 12710–12716; c) K. Osow-
dart, Angew. Chem. Int. Ed. 2002, 41, 898–952; Angew. Chem. 2002, 114,
ˇ
´
ska, O. S. Miljanic, J. Am. Chem. Soc. 2011, 133, 724–727; d) W. R. Dich-
938–993; b) J.-M. Lehn, Chem. Soc. Rev. 2007, 36, 151–160.
ˇ
´
tel, O. S. Miljanic, W. Zhang, J. M. Spruell, K. Patel, I. Aprahamian, J. R.
[
2] a) P. T. Corbett, J. Leclaire, L. Vial, K. R. West, J.-L. Wietor, J. K. Sanders, S.
Otto, Chem. Rev. 2006, 106, 3652–3711; b) E. Moulin, G. Cormos, N. Giu-
seppone, Chem. Soc. Rev. 2012, 41, 1031–1049; c) A. Herrmann, Chem.
Soc. Rev. 2014, 43, 1899–1933.
Heath, J. F. Stoddart, Acc. Chem. Res. 2008, 41, 1750–1761.
[
10] a) A. Star, I. Goldberg, B. Fuchs, Angew. Chem. Int. Ed. 2000, 39, 2685–
689; Angew. Chem. 2000, 112, 2797–2801; b) P. Wipf, S. G. Mahler, K.
2
Okumura, Org. Lett. 2005, 7, 4483–4486; c) C. Saiz, P. Wipf, E. Manta, G.
Mahler, Org. Lett. 2009, 11, 3170–3173; d) B. Buchs nꢃe Levrand, G.
Godin, A. Trachsel, J.-Y. de Saint Laumer, J.-M. Lehn, A. Herrmann, Eur. J.
Org. Chem. 2011, 2011, 681–695.
[
3] a) J.-M. Lehn, Science 2002, 295, 2400–2403; b) N. Giuseppone, Acc.
Chem. Res. 2012, 45, 2178–2188; c) F. Dumitru, Y.-M. Legrand, E. Petit,
A. van der Lee, M. Barboiu, Dalton Trans. 2012, 41, 11860–11865; d) K.
Kobayashi, M. Yamanaka, Chem. Soc. Rev. 2015, 44, 449–466; e) G.
Zhang, M. Mastalerz, Chem. Soc. Rev. 2014, 43, 1934–1947; f) Y. Jin, Q.
Wang, P. Taynton, W. Zhang, Acc. Chem. Res. 2014, 47, 1575–1586.
4] a) V. T. Bhat, A. M. Caniard, T. Luksch, R. Brenk, D. J. Campopiano, M. F.
Greaney, Nat. Chem. 2010, 2, 490–497; b) I. M. Serafimova, M. A. Pufall,
S. Krishnan, K. Duda, M. S. Cohen, R. L. Maglathlin, J. M. McFarland, R. M.
Miller, M. Frçdin, J. Taunton, Nat. Chem. Biol. 2012, 8, 471–476; c) S.
Ulrich, P. Dumy, Chem. Commun. 2014, 50, 5810–5825; d) I. K. H. Leung,
T. Brown Jr, C. J. Schofield, T. D. W. Claridge, MedChemComm 2011, 2,
[
11] a) B. Fuchs, A. Nelson, A. Star, J. F. Stoddart, S. Vidal, Angew. Chem. Int.
Ed. 2003, 42, 4220–4224; Angew. Chem. 2003, 115, 4352–4356; b) R.
Cacciapaglia, S. Di Stefano, L. Mandolini, J. Am. Chem. Soc. 2005, 127,
[
13666–13671; c) J. A. Berrocal, R. Cacciapaglia, S. D. Stefano, Org.
Biomol. Chem. 2011, 9, 8190–8194.
[
12] a) L. You, S. R. Long, V. M. Lynch, E. V. Anslyn, Chem. Eur. J. 2011, 17,
11017–11023; b) L. You, J. S. Berman, E. V. Anslyn, Nat. Chem. 2011, 3,
943–948.
[
[
13] M. Kruppa, B. Kçnig, Chem. Rev. 2006, 106, 3520–3560.
14] a) E. H. Cordes, W. P. Jencks, J. Am. Chem. Soc. 1963, 85, 2843–2848;
b) K. Koehler, W. Sandstrom, E. H. Cordes, J. Am. Chem. Soc. 1964, 86,
390–395.
[
5] a) M. E. Belowich, J. F. Stoddart, Chem. Soc. Rev. 2012, 41, 2003–2024;
b) P. Kova rˇ ꢂ cˇ ek, J.-M. Lehn, J. Am. Chem. Soc. 2012, 134, 9446–9455;
c) M. d. Capela, N. J. Mosey, L. Xing, R. Wang, A. Petitjean, Chem. Eur. J.
2413–2419; c) R. L. Reeves, J. Org. Chem. 1965, 30, 3129–3135; d) Y.
Ogata, A. Kawasaki, J. Org. Chem. 1974, 39, 1058–1061.
2011, 17, 4598–4612; d) N. Wilhelms, S. Kulchat, J.-M. Lehn, Helv. Chim.
[
[
15] a) H. H. Jaffe, Chem. Rev. 1953, 53, 191–261; b) C. Hansch, A. Leo, R. W.
Taft, Chem. Rev. 1991, 91, 165–195; c) H. C. Brown, Y. Okamoto, J. Am.
Chem. Soc. 1958, 80, 4979–4987.
16] a) T. T. Tsou, J. K. Kochi, J. Am. Chem. Soc. 1979, 101, 6319–6332; b) C.
Amatore, F. Pfluger, Organometallics 1990, 9, 2276–2282; c) V. P. W.
Bçhm, W. A. Herrmann, Chem. Eur. J. 2001, 7, 4191–4197; d) J. A. Muel-
ler, M. S. Sigman, J. Am. Chem. Soc. 2003, 125, 7005–7013.
Acta 2012, 95, 2635–2651; e) G. Wang, G. Wu, Z. Wang, X. Zhang, Lang-
muir 2014, 30, 1531–1535.
[6] a) A. Dirksen, S. Dirksen, T. M. Hackeng, P. E. Dawson, J. Am. Chem. Soc.
2006, 128, 15602–15603; b) J. F. Folmer-Andersen, J.-M. Lehn, J. Am.
Chem. Soc. 2011, 133, 10966–10973; c) S. R. Beeren, M. Pittelkow,
J. K. M. Sanders, Chem. Commun. 2011, 47, 7359–7361; d) L. A. Tatum,
X. Su, I. Aprahamian, Acc. Chem. Res. 2014, 47, 2141–2149.
[
17] H. H. Jo, R. Edupuganti, L. You, K. N. Dalby, E. V. Anslyn, Chem. Sci. 2015,
[7] a) S. P. Black, J. K. M. Sanders, A. R. Stefankiewicz, Chem. Soc. Rev. 2014,
6, 158–164.
43, 1861–1872; b) Y. Zhang, P. Vongvilai, M. Sakulsombat, A. Fischer, O.
Ramstrçm, Adv. Synth. Catal. 2014, 356, 987–992; c) N. Ponnuswamy,
F. B. L. Cougnon, J. M. Clough, G. D. Panto s¸ , J. K. M. Sanders, Science
Received: January 10, 2015
2012, 338, 783–785; d) A. R. Stefankiewicz, J. K. M. Sanders, Chem.
Published online on && &&, 2015
Chem. Eur. J. 2015, 21, 1 – 8
www.chemeurj.org
7
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
&
Dynamic Covalent Reactions
Y. Zhou, Y. Yuan, L. You,* E. V. Anslyn*
& – &&
&
Dynamic Aminal-Based TPA Ligands
Probing pathways: Dynamic covalent
reactions (DCRs) of aromatic amines to
create imines that convert to tri(2-pico-
lyl)amine (TPA)-like ligands were studied
in detail. The multicomponent assembly
was conducted as a means of creating
imines in situ and controlling the
competing pathway (see figure;
DPA=di(2-picolyl)amine).
&
&
Chem. Eur. J. 2015, 21, 1 – 8
www.chemeurj.org
8
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!