Trans-Symmetric Dynamic Covalent Systems: Connected Transamination and Transimination Reactions

Article abstract of DOI:10.1002/chem.201500520

The development of chemical transaminations as a new type of dynamic covalent reaction is described. The key 1,3-proton shift is under complete catalytic control and can be conducted orthogonally to, or simultaneous with, transimination in the presence of an amine to rapidly yield two-dimensional dynamic systems with a high degree of complexity evolution. The transamination-transimination systems are proven to be fully reversible, stable over several days, compatible with a range of functional groups, and highly tunable. Kinetic studies show transamination to be the rate-limiting reaction in the network. Furthermore, it was discovered that readily available quinuclidine is a highly potent catalyst for aldimine transaminations. This study demonstrates how connected dynamic reactions give rise to significantly larger systems than the unconnected counterparts, and shows how reversible isomerizations can be utilized as an effective diversity-generating element. Constant exchange: The development of chemical transaminations as a new type of dynamic covalent reactions is described (see figure). This study demonstrates how connected dynamic reactions give rise to significantly larger systems than the unconnected counterparts, and shows how reversible isomerizations can be utilized as an effective diversity-generating element.

Full text of DOI:10.1002/chem.201500520

DOI: 10.1002/chem.201500520
Full Paper
&
Systems Chemistry
trans-Symmetric Dynamic Covalent Systems: Connected
Transamination and Transimination Reactions
Fredrik Schaufelberger, Lei Hu, and Olof Ramstrçm*[a]
Abstract: The development of chemical transaminations as
a new type of dynamic covalent reaction is described. The
key 1,3-proton shift is under complete catalytic control and
can be conducted orthogonally to, or simultaneous with,
transimination in the presence of an amine to rapidly yield
two-dimensional dynamic systems with a high degree of
complexity evolution. The transamination–transimination
systems are proven to be fully reversible, stable over several
days, compatible with a range of functional groups, and
highly tunable. Kinetic studies show transamination to be
the rate-limiting reaction in the network. Furthermore, it was
discovered that readily available quinuclidine is a highly
potent catalyst for aldimine transaminations. This study
demonstrates how connected dynamic reactions give rise to
significantly larger systems than the unconnected counter-
parts, and shows how reversible isomerizations can be
utilized as an effective diversity-generating element.
functionalities, mainly imine/acyl hydrazone[4] and disulfide ex-
change,[5] are typically used in dynamic systems, and despite
recent developments of reversible transformations, such as
dynamic thiol ester exchange,[6] the Strecker reaction,[7] nitrone
exchange,[8] the nitroaldol reaction,[9] transamidation,[10] and
alkyne metathesis,[11] there is still a growing demand for new
reversible bonds that can be utilized in the construction of
complex networks and systems.
Introduction
Constitutional dynamic chemistry (CDC) involves studies of
chemical systems that respond to stimuli and adapt to external
or internal pressure.[1] In this context, the systemic properties
that emerge through interactions between components in
a dynamic compound mixture can be very different from those
of the isolated individual members. The systems are also fun-
damentally under thermodynamic control, in which reversible
covalent bonds and noncovalent interactions allow access to
the most energetically stable state, from which adaptation can
occur. The area has expanded rapidly during recent years, and
many applications for dynamic systems within ligand/receptor
interaction studies, materials chemistry, catalysis, and
molecular sensors have been developed.[2] Furthermore, CDC
provides a relevant framework for the emerging field of
systems chemistry.[3]
Dynamic covalent bonds that can operate orthogonally to,
or simultaneous with, other reversible functionalities are, in
this respect, of particular interest. Compared with systems
based on single dynamic reactions, such multidimensional
arrays lead to significantly larger systems with higher diversity,
potentially covering more chemical space. A range of examples
of orthogonal covalent and noncovalent reactions, mostly
based on metal coordination, acyl hydrazones, and disulfide
chemistry, have also been reported.[7,12]
However, a challenge in CDC has been the relatively low
number of suitable reversible covalent bonds that partake in
dynamic systems. To generate large, diverse dynamic systems
for different applications, it is of high importance to develop
more reversible covalent connections capable of exchange
under mild conditions. Only a few dynamic covalent-bond
In this context, we envisaged that the coupling of reversible
isomerization to intermolecular dynamic exchange would pro-
vide rapid entry to highly complex dynamic systems. However,
isomerization reactions constitute an underexplored area of
CDC; the only example involves fluxional systems of bullvalene
derivatives.[13]
The reversibility of azomethine transamination is well
known.[14] In biological systems, transaminases catalyze both
transamination and transimination of amino acids to and from
[a] F. Schaufelberger, Dr. L. Hu, Prof. O. Ramstrçm
Department of Chemistry, KTH - Royal Institute of Technology
Teknikringen 30, 10044 Stockholm (Sweden)
E-mail: ramstrom@kth.se
a-ketoacids with pyridoxal-5’-phosphate as
a
cofactor
(Scheme 1). The transformation is of high industrial interest,
and biocatalytic equilibrium control is also heavily pursued for
the production of chiral amines.[15]
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201500520.
ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA.
This is an open access article under the terms of Creative Commons Attri-
bution NonCommercial-NoDerivs License, which permits use and distribu-
tion in any medium, provided the original work is properly cited, the use is
non-commercial and no modifications or adaptations are made.
Renewed interest has also emerged in nonbiological
systems, for which transamination strategies have been devel-
oped in, for example, asymmetric organocatalysis, synthesis of
fluorinated amines, and selective N-terminal functionalization
Chem. Eur. J. 2015, 21, 9776 – 9783
9776 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
For such
a
system, the
number (N) of generated system
constituents upon addition of n
different
types
of
initial
monofunctional components is
given by Equation (1):
Scheme 1. Reversible transamination of a-ketoacids under transaminase catalysis.
of peptides.[16] The synthetic challenge with transaminations
resides in often unfavorable equilibria, with typical equilibrium
constants near unity. From a CDC viewpoint, however, this
challenge is instead an advantage, since it allows a predictable
constituent expression in the systems.
nðn þ 1Þ
ð1Þ
N ¼
2
For unsymmetric exchange, the two partners belong to dif-
ferent functional groups, such as aldehydes and amines in
imine exchange chemistry. With n monofunctional components
of one class and m of the other, the number (N) of constituents
becomes that given by Equation (2):
Herein, we report the development of reversible
transamination of aromatic imines and orthogonal coupling to
transimination, which yields double dynamic imine systems.
These transamination–transimination (TATI) systems exhibit
interesting properties because each of the two individual
dynamic reactions can be toggled on or off by the addition or
removal of the respective catalyst for each process. Since both
dynamic reactions manipulate the same functional group,
combining the two dynamic exchange processes leads to
connected dynamic systems with unusually large complexity
evolution.
N ¼ mn
ð2Þ
In this study, conditions to dynamically alter the nature of
the components of an unsymmetric reversible connection
have been developed, so that the two functional groups inter-
convert into each other (Figure 1c). This scenario, which we
term trans-symmetric exchange, will in most instances give rise
to
a number (N) of system constituents predicted by
Equation (3):[17]
N ¼ m2 þ 2mn þ n2
ð3Þ
Results and Discussion
From inspection of Equations (1)–(3), it is clear that a TATI
system will generate systems that are at least (for n=m) four
times larger than the analogous unsymmetric imine system.
Thus, the diversity of screening collections generated by trans-
symmetric exchange will be considerably higher than that for
other exchange types.
Theoretical analysis of dynamic exchange
Figure 1a and b illustrates the two main types of dynamic
covalent bonds used in CDC. For symmetric dynamic bonds,
both exchange partners belong to the same class of functional
groups, of which disulfide exchange and alkene metathesis are
two prominent examples.
Transamination catalysis
Imine 1 was initially investigated in an isolated transamination
reaction (Table 1), in which many of the previously reported
conditions proved unsuitable. For example, the use of strong
heating or strong bases, such as tBuOK and 1,5-diazabicy-
clo[5.4.0]undec-5-ene (DBU), instigated rapid decomposition,
which generated complex mixtures well before the transamina-
tion equilibrium could be reached. Finally, successful clean
transamination could be achieved by using the conditions de-
veloped by Soloshonok and co-workers, with a large excess of
NEt3 in MeCN at 508C.[18] After 24 h, about 66% of imine 1 had
undergone transamination to form isomer 2. Extending the re-
action time to 30 h led to arrival at the equilibrium point, with
a product distribution of 25/75 in favor of transamination
product 2. This result is in agreement with theory because the
equilibrium position of benzylic aldimine/aldimine trans-
aminations can be correlated with the Hammett substituent
parameters; more electron-rich systems are favored to exhibit
the imine functionality.[14]
A reversible covalent reaction for CDC needs to display
long-term stability, mild conditions, and rapid kinetics. Thus,
the use of a large excess of NEt3, along with the relatively long
Figure 1. Illustration of the main classes of dynamic bonds in CDC.
a) Symmetric dynamic exchange; b) unsymmetric dynamic exchange;
c) trans-symmetric dynamic exchange.
Chem. Eur. J. 2015, 21, 9776 – 9783
www.chemeurj.org
9777 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 1. Initial optimization of the transamination reaction.[a]
Table 2. Base optimization studies for reversible transamination.[a]
Solvent
T [8C]
Distribution 1/2[b]
Equilibrium 1/2
Entry
Base
Equivalents
Distribution
Equilibrium
1/2[b]
conversion [%][c]
MeCN
MeCN
CHCl3
toluene
THF
20
50
50
50
50
50
50
no reaction
34/66
93/7
n.d.[c]
25/75
n.d.[c]
n.d.[c]
n.d.[c]
27/73
26/74
1
2
3
4
5
6
7
8
NEt3
NEt3
NMI
NMM
NMP
DIPEA
DMAP
DABCO
C1
C2
DABCO
C2
C1
C3
C4
C5
C1
C1
DBU
TMG
4
1
4
4
4
4
4[d]
4
34/66
67/33
no reaction
93/7
92/8
30/70
71/29
25/75
25/75
26/74
71/29
43/57
25/75
98/2
98/2
no reaction
35/65
50/50
88
44
–
94/6
79/21
30/70
34/66
9
DMF
DMSO
11
93
39
100
100
100
39
76
100
3
3
–
87
66
–
[a] Conditions: imine 1 (0.25 mmol), NEt3 (1.0 mmol), 3 Š molecular sieves
(MS; 10 mg), anhydrous solvent (0.25 mL), 24 h, N2. [b] Analyzed by
1H NMR spectroscopy; [c] n.d.=not determined.
9
4
4
10
11
12
13
14
15
16
17
18
19
20
0.2
0.2
0.2
0.2[d]
0.2
0.2
0.1
0.05
0.1
0.1
equilibrium time, highlighted the need for an improved proto-
col. The initial screening (Table 1) indicated that the highest
rates occurred in the presence of NEt3 in polar aprotic solvents;
anhydrous MeCN, DMSO, and DMF provided access to equilib-
rium compositions in slightly longer than 24 h. Since the use
of the last two solvents led to complications in product
isolation and reaction monitoring, MeCN was adopted in the
next optimization step.
decomposition
decomposition
–
[a] Conditions: imine 1 (0.25 mmol), base (n equivalents), 3 Š MS (10 mg),
anhydrous MeCN (0.25 mL), 508C, N2, 24 h. NMI=N-methylimidazole,
NMM=N-methylmorpholine, NMP=N-methylpyrrolidine, DIPEA=diiso-
propylethylamine, DMAP=4-dimethylaminopyridine, DABCO=1,4-diaza-
bicyclo[2.2.2]octane, TMG=N,N,N’,N’-tetramethyl-1,3-propanediamine; for
the structures of C1–C5 (see Figure 2). [b] Analyzed by 1H NMR
spectroscopy. [c] Conversion towards the equilibrium position, 25/75 for
1/2. [d] Low solubility.
The use of less NEt3 resulted in significant retardation of the
equilibration rate (Table 2, entry 2). To be able to decrease the
catalyst loading, a more suitable Brønsted base was thus
deemed necessary.
Mechanistic studies have indicated that many base-catalyzed
1,3-proton shifts may proceed through a concerted proton
shuffling mechanism (Scheme 2).[18] The transamination rate
should therefore not only be dependent on the base strength,
but also on steric congestion around the basic site. A more
accessible nitrogen center should be able to provide more
efficient proton shuffling.
could be utilized for the transformation, albeit at the cost of
equilibration time.
Due to their successful application in the asymmetric trans-
amination of a-ketoesters and the high similarity of the active
basic site to that of C1, cinchona alkaloids were also evaluated
as catalysts (Figure 2).[16b] However, quinine C3 was almost
completely inactive, possibly due to low solubility (Table 2,
entry 14). Alkylation of the hydroxyl functionality yielded solu-
ble catalyst C4, but no increase in activity could be observed
(Table 2, entry 15). Compound C5 was furthermore evaluated,
since demethylation of the quinoline methoxy group could
reveal a stabilizing hydrogen-bond-donor functionality, but this
was also found to be completely inactive (Table 2, entry 16).
Because cinchona alkaloids are significantly bulkier around the
quinuclidine nitrogen than C1, these results again indicate that
the transamination reaction is strongly dependent on the
steric environment around the
Thus, a range of bases with less steric bias were screened.
As expected, the position of the equilibrium did not change
significantly upon variation of base or base loading. Weak
bases, such as NMI and NMM, showed low activity (Table 2, en-
tries 3 and 4), as did the stronger base NMP, which somewhat
surprisingly provided only a modest equilibration rate (Table 2,
entry 5). However, the use of 4 equivalents of quinuclidine C1,
3-hydroxyquinuclidine C2, or DABCO led to efficient transami-
nation, reaching equilibria within 24 h. Drastically lowering the
loadings of these three catalysts to 0.2 equivalents still led to
decent rates, with C1 providing the best performance with an
equilibrium reached in around 20 h. As little as 5 mol% C1
basic site.
Because system stability is of
utmost importance in dynamic
chemistry applications, it was
gratifying to observe that C1
catalysis did not induce any deg-
Scheme 2. Proposed concerted proton shuffling during the chemical transamination of aldimines.[18]
Chem. Eur. J. 2015, 21, 9776 – 9783 www.chemeurj.org
radation, even six days after
9778 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
basic amine A1 into the imine system. Increasing the amount
of A1 to 0.5 and 1.0 equivalents led to a clear, gradual increase
in the proportion of imine 4 up to 50% of the total imine con-
tent (Figure 3b and c). However, by omitting amine A1 and in-
stead adding 0.5 equivalents of the trifluoromethyl-substituted
benzylamine A2, the relative amount of compound 4 was dras-
tically reduced and the expression of the doubly trifluorometh-
yl-substituted imine 3 instead increased (Figure 3d). Curiously,
the ratios of compounds 1 and 2 were relatively stable during
these system manipulations, only varying slightly despite dras-
tic changes in the overall system composition. This indicates
that homosubstituted imines 3 and 4 are acting as “sinks” that
modulate and buffer the concentration of imines 1 and 2 by
preferential incorporation of newly added amine.
Figure 2. Quinuclidine-based catalysts evaluated for transamination activity.
equilibrium had been reached. Given the general sluggishness
of aldimine transamination, catalyst C1 seems to be a remarka-
bly effective and mild catalyst, even from a synthetic perspec-
tive.
The high level of adaptability strongly indicates that the
connected system was under thermodynamic control.
However, further support that the TATI system had reached
equilibrium was provided by dual-entry-point analysis. Two
identical systems were thus generated, starting from either
compound 1 or 2, as displayed in Figure 4. Upon addition of
0.5 equivalents of amine A1 to a solution of imine in MeCN
with C1 and ZnBr2 as catalysts, both systems resulted in identi-
cal distributions, which remained stable over several days.
Connected TATI
With conditions for the reversible transamination reaction at
hand, transimination was next investigated.[19] A range of Lewis
acids were evaluated as catalysts (see the Supporting Informa-
tion), with most catalysts providing complete transimination of
all tested substrates within 10 min. The most robust trans-
imination conditions were achieved with ZnBr2,[7] although
transimination also worked well with primary amine catalysis.
The time to reach equilibrium did, however, increase to about
1 h in this case, compared with a few minutes for ZnBr2.
Connected TATI systems could be generated simply through
the addition of both catalysts to the same mixture, as shown
in Figure 3. Mixing imine 1 with benzylamine A1, ZnBr2, and
base C1 in MeCN, in the presence of 3 Š MS, led to the genera-
tion of four imines, as evidenced by both 1H NMR spectroscopy
and GC/FID analysis. After around 30 h, no further system
composition changes could be observed.
Figure 4. Results obtained from dual-entry-point equilibration analysis. The
double dynamic systems were identical when generated from either
direction. See the Supporting Information for further details and more
control experiments.
The system distribution at equilibrium could be easily tuned
through amine addition. The use of 0.1 equivalents of benzyl-
amine A1 yielded a system of constituents 1–4 (Figure 3a).
Here, compound 2 constituted nearly 43% of the total imine
content, which indicated a preferred systemic expression of 2
over the other constituents under these conditions. In general,
the system indicated a preference for releasing the less-basic
free amine A2 and incorporating higher amounts of more
Kinetic studies
The kinetic profile of the dynamic system generation was next
evaluated (Figure 5). Since the system based on imines 1–4
required GC analysis for full quantification due to overlapping
signals in the NMR spectra, compound 5 and benzylamine A3
were instead subjected to the TATI conditions. Careful reaction
monitoring by NMR spectroscopy during the equilibration
Figure 3. TATI equilibrium perturbation experiments. The relative system composition was dependent on the amine added: a) 0.1 equivalents of amine A1,
1
b) 0.5 equivalents of amine A1, c) 1.0 equivalents of amine A1, d) 0.5 equivalents of amine A2. The system composition was analyzed by both H NMR
spectroscopy and GC. Conditions: imine 1 (0.25 mmol), amine, C1 (0.05 mmol), ZnBr2 (0.0125 mmol), 3 Š MS (10 mg), MeCN (0.25 mL).
Chem. Eur. J. 2015, 21, 9776 – 9783
www.chemeurj.org
9779 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
process revealed that transamination of compound 5 was the
limiting step of the reaction system, in which equilibrium was
reached after around 48 h. The concentration profile of
compound 5 indicated an exponential decrease, which was
symptomatic of first-order behavior with respect to the re-
agent. Immediately upon transamination, the new hetero-
substituted imine 6 was formed. Transimination with amine A3
immediately occurred to form compound 7, releasing benzyla-
mine A2, which, in turn, underwent transimination with start-
ing compound 5 to yield the second homosubstituted imine 3.
Interestingly, the concentration profile of direct transamination
product 6 turned out to be almost sigmoidal in shape. During
a long induction period, the system seemed to settle into local
transimination equilibria, in which compound 6 was an under-
expressed constituent. Only once a certain ratio of free amines
A2 and A3 had been formed did compound 6 start to accumu-
late, eventually reaching approximately the same concentra-
tion as imine 3. NMR spectroscopy analysis confirmed that the
free amine ratio stayed relatively constant after 18 h, which
again provided support for a model in which the homo-
substituted imines acted as amine sinks during equilibration.
This unexpected effect is an example of the intricate kinetic
behavior of interconnected chemical systems and highlights
the need for dynamic reaction networks to be analyzed from
a broader systems chemistry perspective.
Dynamic pathway control
The individual transimination and transamination equilibria of
model compound 5 were easily accessible by selection of the
reagent, and represent two orthogonal reversible reaction
pathways in the dynamic covalent reaction network (Figure 6).
However, the key diversity generation originates from a combi-
nation of the two pathways. When mixing compound 5 with
transamination catalyst C1 and benzylamine A1 (1 equiv),
a system of nine imine compounds and three amines was effi-
ciently generated (Figure 6). This experiment verifies the trans-
symmetric exchange mode of the system and illustrates the
potential in the TATI protocol for rapid systemic complexity
generation. The system size growth relative to the amount of
starting compounds is thus demonstrated to be significantly
higher than that in non-trans-symmetrical protocols utilizing
monofunctional exchange partners.
Figure 5. Kinetic profile for the four different imines in the TATI system (5
1
~
&
^
*
(
); 6 ( ); 7 ( ); 3 ( )), as measured by H NMR spectroscopy. Conditions:
imine 1 (0.25 mmol), amine A3 (0.025 mmol), catalyst C1 (0.05 mmol), ZnBr2
(0.0125 mmol), 3 Š MS (10 mg), MeCN (0.25 mL). Data were obtained from
duplicate experiments.
Cyano, trifluoromethyl, chloro, and bromo substituents all
worked with the system, and different ortho-, meta-, and para-
substituted compounds were all well tolerated. The strongly
electron-withdrawing nitro-substituted imine 14 decomposed
under the reaction conditions (Table 3, entry 6). Problematic
behavior of nitro-substituted compounds in aldimine transami-
nations has been observed before.[14] Reverting to the use of
less reactive NEt3 at room temperature led to quick equilibra-
tion, although the system decomposed over the next 48 h
(Table 3, entry 7). Heteroaromatic systems generally performed
well, albeit without the addition of ZnBr2 and with a lower
amount of free amine with pyridine-based structures due to
complexation and degradation of the picolylamines formed
(Table 3, entries 8 and 9). With some compounds, such as
imidazole imine 17 and furyl imine 18, small amounts of
unidentified side products appeared during equilibration.
Generally, compounds with only electron-donating substi-
tuents underwent very slow transamination (Table 3, entries
12–14). An exception was imine 12, which reacted smoothly to
provide a full TATI system in a short time (Table 3, entry 4). One
Substrate scope and synthetic considerations
Finally, the substrate scope of the TATI system formation was
investigated in detail. As displayed in Table 3, a range of
substrates were compatible with the optimized reaction con-
ditions. Generally, the system needed at least one electron-
withdrawing component on one of the aromatic rings for
acceptable equilibration times. This can be understood when
considering the reaction mechanism, in which the initial trans-
amination proton shuffling is the rate-limiting step. Because
electron-withdrawing groups on the aromatic rings increase
the acidity of the benzylic protons, more electron-poor
systems should lead to faster transamination.
Chem. Eur. J. 2015, 21, 9776 – 9783
www.chemeurj.org
9780 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
reason for this reactivity could
be intramolecular hydrogen
bonding between the phenolic
proton and the imine nitrogen.
Furthermore, benzylic imines
with substituents on both aro-
matic rings were compatible
with the system (Table 3, en-
tries 16–19). Even methyl- and
methoxy-substituted aromatics
participated readily in the ex-
change processes, as long as at
least one other component had
a sufficiently electron-withdraw-
ing functional group attached.
The high compatibility range of
these model substrates indicates
that TATI exchange can be uti-
lized with a range of substituted
aromatic and heteroaromatic
exchange partners, covering
a
wide part of the relevant
chemical space in terms of
aromatic compounds.
An advantage of TATI systems
from a practical standpoint is
the dual nature of the involved
functional groups. If a particularly
interesting aldehyde or amine is
commercially unavailable or diffi-
cult to access synthetically, the
reverse compound can instead
be employed and the desired
compound created for the
screening collection in situ. This
is, for example, the case for en-
tries 10 and 15 in Table 3, for
which very expensive benzyl-
amines are created in situ from
cheap, widely available alde-
hydes. It is also straightforward
to form the benzylic imine in the
flask before equilibration to gen-
erate a one-pot protocol that
omits the extra imine formation
step. Simply condensing the al-
dehyde and benzylamine in the
presence of 3 Š MS, as shown in
Scheme 3, followed by addition
of the catalysts, led to the evolu-
tion of an identical system to
that observed with preformed
Figure 6. 1H NMR spectra illustrating selective access to each equilibration mode of imine 5: a) starting material 5; imine 1.
b) transimination equilibrium (15 min) with imine 5 (1.0 equiv),
Furthermore, simple aqueous
washing led to the removal of all
catalysts and benzylamines,
which allowed the isolation of
benzylamine A1 (1.0 equiv), ZnBr2 (0.05 equiv); c) transamination equilibrium (24 h) between imines 5 and 6, with
C1 (0.2 equiv); d) coupled transamination and transimination equilibrium (45 h) with A1 (1.0 equiv), 5 (1.0 equiv),
C2 (0.2 equiv), and ZnBr2 (0.05 equiv). e) Enlarged view of the characteristic imine resonance region. Resonances
corresponding to compounds 1 and 2 overlap.
Chem. Eur. J. 2015, 21, 9776 – 9783
www.chemeurj.org
9781 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
patible with a wide range of dif-
ferent functional groups, which
were well tolerated under the re-
action conditions.
Table 3. Substrate scope of TATI equilibration.[a]
The generated dynamic sys-
tems could potentially be used
for in situ dynamic screening or
as starting point systems for
combinatorial multicomponent
reactions. Efforts to expand the
scope of TATI systems for metal
sensing and double dynamic
materials are currently underway.
Entry
Imine
Ar1
Ar2
t [h][b]
Comment
1
2
3
4
5
6
7
8
1
4-CF3-C6H4
3-CN-C6H4
4-Br-C6H4
2-OH-C6H4
2,4-Cl-C6H3
4-NO2-C6H4
4-NO2-C6H4
4-pyridyl[d]
2-pyridyl[d]
1-N-Me-imidazoyl
2-furyl[e]
2-MeO-C6H4
4-Me-C6H4
4-F-C6H4
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
30
30
80
24
48
–
12
16
16
24
96
–
–
–
–
–
10
11
12
13
14
14
15
16
17
18
19
20
21
22
23
24
25
5
–
rapid decomposition
degrades within 48 h
–
–
[c]
Experimental Section
General TATI procedure
9
10
11
12
13
14
15
16
17
18
19
–
ZnBr2 (2.8 mg, 0.0125 mmol) was
added to a dry 2 mL screw-cap vial
with activated 3 Š MS (ground,
10 mg) and the mixture was left
under N2 for 1 h. Benzylic imine
(0.25 mmol) was dissolved in anhy-
drous MeCN (0.25 mL), and C1
(5.6 mg, 0.05 mmol) was added.
The resulting mixture was trans-
ferred to the vial and amine was
added. The solution was stirred
under N2 at 508C in a sand bath.
Reaction monitoring was per-
formed by removing an aliquot
(10 mL) of the reaction mixture, fil-
degradation during equilibration
very slow
no reaction
no reaction
–
–
2-naphthyl[e]
4-CF3-C6H4
4-CF3-C6H4
4-CF3-C6H4
4-CF3-C6H4
168
36
42
42
48
degradation during equilibration
4-F-C6H4
4-MeO-C6H4
3,4,5-MeO-C6H2
2-Me-C6H4
–
–
–
–
[a] Conditions: imine (0.25 mmol), amine (0.125 mmol), C1 (0.05 mmol), ZnBr2 (0.0125 mmol), 3 Š MS (10 mg),
MeCN (0.25 mL). [b] Approximate time for the system to reach equilibrium. [c] At RT with four equivalents of
NEt3 instead of compound C1. [d] With 0.1 equivalents of amine, no ZnBr2. [e] With 0.5 equivalents of
compound C1.
tering through a pad of cotton, and diluting with CDCl3, after
which an NMR spectrum was recorded.
One-pot protocol for imine condensation–TATI
Scheme 3. One-pot protocol for the direct creation of TATI dynamic systems
from aldehydes and amines.
4-Trifluoromethylbenzaldehyde (66.8 mL, 87.1 mg, 0.5 mmol) was
dissolved in anhydrous MeCN (0.5 mL) in a dry 2 mL screw-cap vial
with activated 3 Š MS (ground, 100 mg), and benzylamine (81.9 mL,
80.4 mg, 0.75 mmol) was added. The mixture was stirred at RT
under N2 for 24 h, after which time C1 (11.1 mg, 0.1 mmol) and
ZnBr2 (5.6 mg, 0.025 mmol) were added in one batch. The system
was subsequently generated in about 24 h, and analyzed as
described in the general procedure.
imine mixtures in high purity and yield. The imines can also be
readily hydrolyzed, leading to aldehyde mixtures. This protocol
could thus in theory also be of use for the preparation of non-
dynamic imine or aldehyde systems for high-throughput
screening, although this would not utilize the advantages for
in situ screening that a system provides.
General procedure for imine synthesis
Aldehyde (3.0 mmol) was dissolved in anhydrous CH2Cl2 (30 mL) in
a dry round-bottomed flask in the presence of activated 4 Š MS
(whole beads, ca. 3.0 g). Amine (3.0 mmol) was added dropwise
under N2, and the solution was stirred slowly under N2. The reac-
Conclusion
We have developed the first example of trans-symmetric rever-
sible exchange in dynamic systems. The results showed that
reversible transamination and transimination reactions could
be carried out orthogonally or connected to provide access to
catalyst-controlled dynamic systems with high level of com-
plexities from simple starting materials. New conditions for al-
dimine transaminations were also developed, with which qui-
nuclidine was found to be an effective and mild catalyst for
the transformation. The TATI system furthermore proved com-
1
tion was monitored by H NMR spectroscopy sampling, and upon
consumption of starting materials the reaction mixture was filtered
through a pad of Celite and concentrated in vacuo to directly
obtain the imine in a typical purity of 98–99%.
Isolation protocol
A dynamic imine system was equilibrated from compound 25
(88.3 mg, 0.25 mmol) and 3,4,5-trimethoxybenzylamine (4.3 mL,
Chem. Eur. J. 2015, 21, 9776 – 9783
www.chemeurj.org
9782 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Mondal, N. Radeva, H. Kçster, A. Park, C. Potamitis, M. Zervou, G. Klebe,
A. K. H. Hirsch, Angew. Chem. Int. Ed. 2014, 53, 3259–3263; Angew.
Chem. 2014, 126, 3324–3328; i) H. Fanlo-Virgós, A.-N. R. Alba, S.
Hamieh, M. Colomb-Delsuc, S. Otto, Angew. Chem. Int. Ed. 2014, 53,
11346–11350; Angew. Chem. 2014, 126, 11528–11532; j) G. Gasparini,
M. Dal Molin, L. J. Prins, Eur. J. Org. Chem. 2010, 2429–2440; k) L. Hu, O.
Ramstrçm, Chem. Commun. 2014, 50, 3792–3794; l) S. Mihai, Y. Le Duc,
D. Cot, M. Barboiu, J. Mater. Chem. 2010, 20, 9443–9448; m) N. Sreeni-
vasachary, J.-M. Lehn, Proc. Natl. Acad. Sci. USA 2005, 102, 5938–5943;
n) L. You, J. S. Berman, E. V. Anslyn, Nat. Chem. 2011, 3, 943–948.
[3] a) N. Wagner, G. Ashkenasy, Chem. Eur. J. 2009, 15, 1765–1775;
b) R. A. R. Hunt, S. Otto, Chem. Commun. 2011, 47, 847–858.
[4] a) I. Huc, J.-M. Lehn, Proc. Natl. Acad. Sci. USA 1997, 94, 2106–2110;
b) G. R. L. Cousins, S.-A. Poulsen, J. K. M. Sanders, Chem. Commun. 1999,
1575–1576; c) T. Bunyapaiboonsri, O. Ramstrçm, S. Lohmann, J.-M.
Lehn, L. Peng, M. Goeldner, ChemBioChem 2001, 2, 438–444.
[5] a) H. Hioki, W. C. Still, J. Org. Chem. 1998, 63, 904–905; b) O. Ramstrçm,
J.-M. Lehn, ChemBioChem 2000, 1, 41–48.
4.9 mg, 0.025 mmol) for 48 h under the conditions described in the
general procedure. After equilibrium was reached, the reaction so-
lution was diluted with diethyl ether (2 mL) and the organic phase
was washed with a saturated aqueous NH4Cl solution (2”1 mL)
and brine (1 mL). The colorless solution was dried with MgSO4, fil-
tered, and concentrated to afford a clean mixture of the four
imines (79.1 mg, 90% mass recovery). No amine or quinuclidine
was present in the sample according to NMR spectroscopy analy-
sis, and hydrolysis of the imines under the workup conditions was
less than 1%.
Hydrolysis of the imine system
The imine systems could also be hydrolyzed to provide mixtures of
aldehydes. The imine mixture (0.25 mmol imine content, either
crude or purified according to the procedure outlined above) was
dissolved in MeOH (1 mL) and a 1m aqueous solution of HCl
(1 mL) was added dropwise. The resulting solution was stirred at
RT under air for 2 h, and the aldehydes were subsequently extract-
ed with diethyl ether (3”2 mL). Drying with MgSO4, filtering, and
concentration yielded the pure aldehydes without any side
products in typically 95–99% mass balance.
[6] R. Larsson, Z. C. Pei, O. Ramstrçm, Angew. Chem. Int. Ed. 2004, 43,
3716–3718; Angew. Chem. 2004, 116, 3802–3804.
[7] P. Vongvilai, O. Ramstrçm, J. Am. Chem. Soc. 2009, 131, 14419–14425.
[8] S. M. Turega, C. Lorenz, J. W. Sadownik, D. Philp, Chem. Commun. 2008,
4076–4078.
[9] P. Vongvilai, M. Angelin, R. Larsson, O. Ramstrçm, Angew. Chem. Int. Ed.
2007, 46, 948–950; Angew. Chem. 2007, 119, 966–968.
[10] Y. Ruff, V. Garavini, N. Giuseppone, J. Am. Chem. Soc. 2014, 136, 6333–
6339.
Acknowledgements
[11] Q. Wang, C. Zhang, B. C. Noll, H. Long, Y. Jin, W. Zhang, Angew. Chem.
Int. Ed. 2014, 53, 10663–10667; Angew. Chem. 2014, 126, 10839–10843.
[12] a) Y. Zhang, O. Ramstrçm, Chem. Eur. J. 2014, 20, 3288–3291; b) Y.
Zhang, L. Hu, O. Ramstrçm, Chem. Commun. 2013, 49, 1805–1807; c) J.
Leclaire, L. Vial, S. Otto, J. K. M. Sanders, Chem. Commun. 2005, 1959–
1961; d) A. V. Gromova, J. M. Ciszewski, B. L. Miller, Chem. Commun.
2012, 48, 2131–2133; e) R. J. Sarma, S. Otto, J. R. Nitschke, Chem. Eur. J.
2007, 13, 9542–9546; f) Z. Rodriguez-Docampo, S. Otto, Chem.
Commun. 2008, 5301–5303; g) L. A. Ingerman, M. L. Waters, J. Org.
Chem. 2009, 74, 111 –117; h) N. Christinat, R. Scopelliti, K. Severin,
Angew. Chem. Int. Ed. 2008, 47, 1848–1852; Angew. Chem. 2008, 120,
1874–1878; i) V. Goral, M. I. Nelen, A. V. Eliseev, J.-M. Lehn, Proc. Natl.
Acad. Sci. USA 2001, 98, 1347–1352; j) Y.-M. Legrand, A. van der Lee, M.
Barboiu, Inorg. Chem. 2007, 46, 9540–9547.
This study was, in part, supported by the Swedish Research
Council and the Royal Institute of Technology. F.S. thanks the
Royal Institute of Technology Excellence Award for financial
support. L.H. thanks the China Scholarship Council for a special
Scholarship Award.
Keywords: dynamic covalent chemistry · isomerization · Schiff
bases · systems chemistry · transamination
[1] a) B. L. Miller, Dynamic Combinatorial Chemistry, in Drug Discovery, Bioor-
ganic Chemistry and Materials Science, Wiley, Hoboken, 2010; b) J. N. H.
Reek, S. Otto, in Dynamic Combinatorial Chemistry, Wiley-VCH, Wein-
heim, 2010; c) M. Barboiu, in Constitutional Dynamic Chemistry, Springer,
Heidelberg, 2012; d) A. Herrmann, Chem. Soc. Rev. 2014, 43, 1899–
1933; e) M. C. Misuraca, E. Moulin, Y. Ruff, N. Giuseppone, New J. Chem.
2014, 38, 3336–3349; f) J.-M. Lehn, Angew. Chem. Int. Ed. 2013, 52,
2836–2850; Angew. Chem. 2013, 125, 2906–2921; g) L. Hu, F. Schaufel-
berger, B. J. J. Timmer, M. A. Flos, O. Ramstrçm, Constitutional Dynamic
Chemistry, in Kirk-Oethmer Encyclopedia of Chemical Technology, Wiley,
2014; h) Y. Jin, C. Yu, R. J. Denman, W. Zhang, Chem. Soc. Rev. 2013, 42,
6634–6654; i) A. Wilson, G. Gasparini, S. Matile, Chem. Soc. Rev. 2014,
43, 1948–1962.
[13] a) O. Ramstrçm, J.-M. Lehn, Nat. Rev. Drug. Discovery 2002, 1, 26–36;
b) A. R. Lippert, J. Kaeobamrung, J. W. Bode, J. Am. Chem. Soc. 2006,
128, 14738–14739.
[14] P. A. S. Smith, D. Chi Van, J. Org. Chem. 1976, 41, 2013–2015.
[15] a) C. K. Savile, J. M. Janey, E. C. Mundorff, J. C. Moore, S. Tam, W. R.
Jarvis, J. C. Colbeck, A. Krebber, F. J. Fleitz, J. Brands, P. N. Devine, G. W.
Huisman, G. J. Hughes, Science 2010, 329, 305–309; b) B. Wang, H.
Land, P. Berglund, Chem. Commun. 2013, 49, 161–163; c) A. P. Green,
N. J. Turner, E. O’Reilly, Angew. Chem. Int. Ed. 2014, 53, 10714–10717;
Angew. Chem. 2014, 126, 10890–10893; d) E. Busto, R. C. Simon, B. Gri-
schek, V. Gotor-Fernµndez, W. Kroutil, Adv. Synth. Catal. 2014, 356,
1937–1942.
[16] a) R. A. Scheck, M. T. Dedeo, A. T. Iavarone, M. B. Francis, J. Am. Chem.
Soc. 2008, 130, 11762–11770; b) X. Xiao, Y. Xie, C. Su, M. Liu, Y. Shi, J.
Am. Chem. Soc. 2011, 133, 12914–12917; c) S. Bandyopadhyay, W. Zhou,
R. Breslow, Org. Lett. 2007, 9, 1009–1012; d) T. Ono, V. P. Kukhar, V. A.
Soloshonok, J. Org. Chem. 1996, 61, 6563–6569.
[17] By assuming that all individual components m and n have different
substitution patterns, excluding the reversible linkage; otherwise the
number of different unique compounds naturally decreases.
[18] P. Nagy, H. Ueki, D. O. Berbasov, V. A. Soloshonok, J. Fluorine Chem.
2008, 129, 409–415.
[2] a) R. Caraballo, H. Dong, J. P. Ribeiro, J. P. JimØnez-Barbero, O. Ram-
strçm, Angew. Chem. Int. Ed. 2010, 49, 589–593; Angew. Chem. 2010,
122, 599–603; b) M. Demetriades, I. K. H. Leung, R. Chowdhury, M. C.
Chan, M. A. McDonough, K. K. Yeoh, Y.-M. Tian, T. D. W. Claridge, P. J. Rat-
cliffe, E. C. Y. Woon, C. J. Schofield, Angew. Chem. Int. Ed. 2012, 51,
6672–6675; Angew. Chem. 2012, 124, 6776–6779; c) A. J. Clipson, V. T.
Bhat, I. McNae, A. M. Caniard, D. J. Campopiano, M. F. Greaney, Chem.
Eur. J. 2012, 18, 10562–10570; d) Y. Zhang, L. Hu, O. Ramstrçm, in Con-
stitutional Dynamic Chemistry for Bioactive Compounds, Vol. 13 (Ed.: H. J.
Schneider), 2013, pp. 397–418; e) A. G. Santana, E. JimØnez-Moreno,
A. M. Gómez, F. Corzana, C. Gonzµlez, G. JimØnez-Oses, J. P. JimØnez-Bar-
bero, J. L. Asensio, J. Am. Chem. Soc. 2013, 135, 3347–3350; f) E.
Moulin, G. Cormos, N. Giuseppone, Chem. Soc. Rev. 2012, 41, 1031–
1049; g) L. Marin, S. Moraru, M.-C. Popescu, A. Nicolescu, C. Zgardan,
B. C. Simionescu, M. Barboiu, Chem. Eur. J. 2014, 20, 4814–4821; h) M.
[19] S. Di Stefano, M. Ciaccia, Org. Biomol. Chem. 2015, 13, 646–654.
Received: February 8, 2015
Published online on June 3, 2015
Chem. Eur. J. 2015, 21, 9776 – 9783
www.chemeurj.org
9783 ꢀ 2015 The Authors. Published by Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim