Precipitation-driven self-sorting of imines

Article abstract of DOI:10.1039/c2ob25736j

Judicious choice of precipitation conditions can lead to self-sorting of equilibrating mixtures of aromatic aldehydes and substituted anilines into a handful of imine products. The selectivity of this process is caused by the solubility differences among possible imines in the EtOH-H₂O solvent mixtures used in precipitation.

Full text of DOI:10.1039/c2ob25736j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Organic &
Biomolecular
Chemistry
Cite this: Org. Biomol. Chem., 2012, 10, 4847
www.rsc.org/obc
COMMUNICATION
Precipitation-driven self-sorting of imines†
Rio Carlo Lirag, Karolina Osowska‡ and Ognjen Š. Miljanić*
Received 16th April 2012, Accepted 21st May 2012
DOI: 10.1039/c2ob25736j
Judicious choice of precipitation conditions can lead to self-
sorting of equilibrating mixtures of aromatic aldehydes and
substituted anilines into a handful of imine products. The
selectivity of this process is caused by the solubility differ-
ences among possible imines in the EtOH–H2O solvent mix-
tures used in precipitation.
Outcomes of chemical reactions can be profoundly influenced
by precipitation and other phase changes. Selective removal of
individual species from the solution can drive their production to
completion, and sequestration within a crystal can also create a
kinetic barrier to undesirable side reactions. Precipitation-driven
selectivity shifts are widely used in academic and industrial set-
tings, and—on account of their simplicity and robustness—they
may have played a role in the prebiotic peptide synthesis1 and
enantiomeric amplification2 processes.
Notable recent work in the area of dynamic combinatorial
chemistry (DCC)3 of imine complexes relied on selective crystal-
lization to achieve diastereomeric amplification of copper–imine
helicates,4 as well as the preparation of interlocked Solomon
knots5 and receptors for CO2.6 A perhaps neglected aspect of
Fig. 1 Anilines and aldehydes discussed in this study. In the text, as
such selectivity-through-precipitation protocols is that precipi-
tation of a certain pure species from a dynamic combinatorial
library (DCL) also simplifies the composition of the solution—
often to a single compound that can be isolated in high yield.
With our interest in self-sorting processes,7 we speculated that
parallel synthesis of multiple imines could be achieved under
precipitative conditions, provided that their low solubilities favor
them over the alternative crossover products. In this Communi-
cation, we report that mixtures of anilines and aromatic alde-
hydes give rise to ordered imine combinations, if their synthesis
is carried out under selective precipitation conditions.
well as in the remaining Schemes, imines formed from these compounds
will be designated by combining the aniline letter with the aldehyde
number—e.g. imine B3 is formed by a dehydrative condensation of
aniline B with aldehyde 3.
be easily oxidized or evaporated, respectively. Precipitation
should present a much broader route to self-sorting, as virtually
any imine can be precipitated—or dissolved—provided that con-
ditions are appropriately adjusted. We thus set out to explore the
self-sorting behaviors of imines constructed from anilines A–D
and aldehydes 1–4 (Fig. 1). These compounds were chosen
because of their widely different electronic and steric character-
istics, which suggested that the formed imines may have solubili-
ties different enough to allow selective self-sorting behavior. In
our first experiment (Scheme 1, left), an equimolar mixture of
aniline (A in Fig. 1), 2,4-dimethoxyaniline (B), benzaldehyde
(1), and 2,4-dinitrobenzaldehyde (2) was dissolved in EtOH,
yielding a 25.5 mM solution with respect to each component.
Initially clear solution quickly turned cloudy, indicating the
onset of precipitation of an imine product. Two crops of precipi-
tate were collected and combined, and their analysis revealed
virtually pure imine B2, in 77% total yield. The residual solution
contained imine A1 as the dominant component (82% yield),
As our previous work demonstrated, freely equilibrating
dynamic imine libraries3 spontaneously simplify (i.e. self-sort)
when exposed to an irreversible chemical (oxidation)8 or physical
(distillation) stimulus.9 These methods require substrates that can
University of Houston, Department of Chemistry, 136 Fleming Building,
Houston, TX 77204-5003, USA. E-mail: miljanic@uh.edu;
Fax: +1 (713) 743-2709; Tel: +1 (832) 842-8827
†Electronic supplementary information (ESI) available: Detailed exper-
imental procedures and spectral characterization data for all new com-
pounds. See DOI: 10.1039/c2ob25736j
‡Current address: Selvita S. A., ul. Bobrzyńskiego 14, 30-348 Kraków,
Poland.
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 4847–4850 | 4847

View Article Online
Scheme 1 Self-sorting of two [2 × 2] aniline–aldehyde mixtures. Only
the yields of major component imines are given; for the yields of other
imines, see the text and the ESI.†
with minor contamination by B1 (11%), A2 (7%), and
B2 (2%).10 The mixture was effectively self-sorted, as products
A2 and B1 were strongly disfavored in both solution and the
precipitate—although they are in principle competitive with the
dominant B2 and A1.
This initial result encouraged us to target a mixture of anilines
and aldehydes with smaller structural differences (Scheme 1,
right). Upon dissolution of compounds B, 2, 4-methoxyaniline
(C), and 4-nitrobenzaldehyde (3) in EtOH, precipitation begun
almost instantaneously. Preliminary analysis of the precipitate
revealed a combination of B2 and C3; while the two imines
co-precipitated, they were curiously uncontaminated by the
crossover products B3 or C2. However, analysis of the remaining
solution after 24 h indicated significant presence of all four poss-
ible imines, suggesting that the precipitation was not complete.
To circumvent this problem, we devised a protocol in which
H2O was slowly injected into the EtOH solution during the
course of the reaction. The purpose of this modification was to
turn EtOH into a poor solvent for imines, but to do so gradually
in order to avoid precipitating starting anilines or aldehydes. At
the end of H2O addition, the final H2O–EtOH volume ratio was
1.33 : 1. Using this new strategy, overnight precipitation resulted
in the solid that contained B2 (78%) and C3 (84%) as major
components, with minor amounts of crossover imines B3 (10%)
and C2 (11%). Among three independent trials, yields of B2 and
C3 in this precipitation protocol varied by less than 2%.
Importantly, the addition of H2O did not lower imine yields;
once precipitated, imines examined in this study appeared stable
to hydrolysis on the time scale of the performed experiments.
High selectivities observed in these two [2 × 2] experiments
suggested that self-sorting of imines was possible during their
precipitation and that it was not limited to a strict fractionation
between a solution and a precipitate—i.e. selectivity can be
achieved even during co-precipitation. Low solubility of precipi-
tated imines appears to control the selectivities of these reactions.
This notion was confirmed by the transmutation experiment in
which pure imines B1 and A2 were dissolved in EtOH and sub-
jected to the slow addition of H2O. After overnight stirring,
imine B2 precipitated as a virtually pure compound (90% yield),
while the solution contained chiefly A1, which could be isolated
in 75% yield.
Scheme 2 Self-sorting of two [3 × 3] aniline–aldehyde mixtures. Only
the yields of major component imines are given; for the yields of other
imines, see the text and the ESI.†
aldehydes 1–3 in EtOH (Scheme 2, top), and subjected the
mixture to slow addition of H2O. An orange solid was isolated
after overnight precipitation. Its spectroscopic analysis revealed a
mixture dominated by imines B2 (77%) and C3 (79%). Small
amounts (1–15%) of five additional imines could be identified
1
from the crude H NMR spectrum of the mixture. The filtrate
contained A1 as the main component (81%) and low amounts
(0.2–9%) of the eight remaining possible imines.
Next [3 × 3] experiment saw the aldehyde 1 replaced with
4-phenylbenzaldehyde (4), and aniline Awith 3-phenylaniline (D),
while keeping B, C, 2, and 3 as the components of the mixture.
After a slightly modified procedure, in which H2O addition to
EtOH solution was followed by slow (10 d) evaporation of
mixture to dryness, the analysis of the thus formed orange solid
revealed exclusive presence of imines B2 (73%), C4 (82%), and
D3 (84%). Surprisingly—in light of our previous experiments—
no significant amounts of either C3 or D4 were detected. This
precipitation protocol was also repeated three times to ensure
reproducibility, and only minor variations in yields were
observed (up to 6% for B2, 4% for C4 and 2% for D3).
Successful self-sorting observed in experiments presented in
Schemes 1 and 2 is readily explained after examination of solubi-
lity data for all possible imines (Table 1). In the first experiment
(Scheme 1, left), imine B2 is clearly the compound with the
lowest molar solubility: ∼18 times less soluble than A2, and >40
times less soluble than either A1 or B2. Thus, its precipitation is
“uncontested” and essentially complete in pure EtOH, removing
all B and 2 from the solution. With these two components
sequestered in the precipitate, remaining A and 1 have no choice
but to form A1 as the solution-phase second product.
Examination of the relative solubilities of possible imines in
the second [2 × 2] self-sorting experiment (Scheme 1, right)
again suggests B2 as the least soluble compound. However, the
After demonstrating successful self-sorting of [2 × 2] mixtures
of anilines and aldehydes, we analysed the behavior of more
complex [3 × 3] mixtures. First, we combined anilines A–C with
4848 | Org. Biomol. Chem., 2012, 10, 4847–4850
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 1 Solubilities of imines discussed in this study
B, C, 2, and 3 (Scheme 1, right) is performed in EtOH at
10 times lower component concentrations (4.17 mM) than
initially described, all four imines remain soluble throughout the
course of the reaction. In this case, no self-sorting is observed at
equilibrium: relative molar ratios of B2 (1.00), B3 (1.15),
C2 (1.08), and C3 (1.25) are close to random distribution.11
When a solvent was changed to MeCN—in which all imines are
soluble—similarly random mixtures of four or nine components
were obtained in all [2 × 2] and [3 × 3] experiments. These two
control experiments virtually exclude the possibility that self-
sorting was caused by electronic factors in solution.12 On the
other hand, concentrations of all components can be increased
up to 125 mM (3×) without a large loss in the fidelity of self-
sorting (83% B2, 74% C3). However, at component concen-
trations of 209 mM (5×), dramatic deterioration in fidelity is
observed: B2 is formed in 51% yield, C2 in 42%, and C3
in 34%.13
Solubilitya [mmol L−1] in:
Imine
EtOH
EtOH–H2O (1 : 2.4, v/v)
A1
A2
A3
B1b
B2
B3
B4
C1
C2
C3
C4
D2
D3
D4
>36.8
11.1
29.5
>27.6
0.6
>23.3
8.7
>31.6
5.2
10.9
2.3
13.3
1.5
3.5
>16.6
<0.1
2.8
1.6
8.5
0.3
0.2
1.7
2.9
1.0
0.3
8.6
12.1
7.4
a All solubilities were measured at 22 °C. b Compound B1 was the only
Experiments performed in this study suggest that self-sorting
of DCLs can proceed under precipitative conditions. During pre-
cipitation, some imine libraries sorted into segregated solution/
solid ensembles, while others produced co-precipitated mixtures
—but all reactions proceeded with high selectivities. These four
experiments illustrate an important concept in the study of self-
sorting systems, namely that collective properties take pre-
cedence over the properties of individual components studied in
isolation. This switch is not absolute: dominant members of the
potential imine pool—such as B2 in all our examples—will
invariably occur in the self-sorted state because of their extreme
individual insolubility. On the other hand, library members with
higher individual solubilities are forced to compete with each
other for resources (aldehyde and aniline components). In those
cases, the influence of the remainder of the mixture can be
crucial, as a library member (e.g. C2) can be strongly disfavored
if it competes for a resource with a dominant member (B2). Con-
versely, a library member can be favored beyond its individual
“merit”, if its composition is orthogonal to the dominant
member, while that of its less soluble competitors is not. This
primitive form of internal regulation of a synthetic process
appears to be a general feature of self-sorting systems and the
one that we are currently intensively exploring.
liquid among the studied imines.
differences are not as dramatic as in the previous case: B2 is
∼9 times less soluble than C2, ∼18 times less than C3, and
>35 less than B3. These smaller differences (along with the
lower absolute solubility of C3, vide infra) explain why B2
could not be precipitated alone, leaving the more soluble imines
in the solution. The formation of C3 as the second product is
rationalized by the fact that it does not share either the aldehyde
or the aniline component with the favored B2. Thus, precipita-
tive self-sorting can express the more soluble C3 (which does
not compete with B2), rather than the less soluble C2—which
must compete with B2 for a resource, aldehyde 2.
Sorting of the more complex [3 × 3] aniline–aldehyde mix-
tures could have potentially generated nine imines, but resulted
in the high-yield formation of only three. These findings can
also be rationalized using the solubility data. In the first [3 × 3]
study (Scheme 2, top), possible products included imines A1–3,
B1–3, and C1–3. Among these, imine B2 is ∼9 times less
soluble than next most-insoluble compound, C2. Precipitation of
B2 will thus be favored, sequestering B and 2 from all other
imines that contain them. Within the reduced pool of imines that
do not share a component with B2—that is, compounds A1, A3,
C1, and C3—the only imine that can be precipitated under the
examined conditions is C3. The absolute solubility of A1, A3,
and C1 is too high for them to effectively compete during the
precipitation process. Exclusive precipitation of C3 removes all
C and 3 from the solution, and the soluble A1 must form as the
third product.
Results presented in this Communication were obtained with
the generous financial support from the National Science Foun-
dation CAREER program (CHE-1151292), the donors of the
American Chemical Society Petroleum Research Fund
(ACS-PRF), the Welch Foundation (grant no. E-1768), the
University of Houston (UH) and its Grant to Advance and
Enhance Research, and the Texas Center for Superconductivity
at UH. K.O. acknowledges Drs Joan and Herman Suit for the
Eby Nell McElrath Postdoctoral Fellowship.
Similar reasoning explains the outcome of the last [3 × 3]
experiment. Introduction of the new aldehyde component 4
allows the generation of a highly insoluble imine C4, which is
now the second least-soluble compound (after B2). Since B2 and
C4 do not share constituents, their precipitation can proceed
orthogonally and simultaneously, depleting the solution’s supply
of aldehydes 2 and 4, as well as anilines B and C. This leaves
D3 as the only possible imine in the mother liquor; addition of
H2O ensures its complete precipitation, completing the self-
sorting event.
Notes and references
1 B. M. Rode, D. Fitz and T. Jackschitz, in Origin of Life: Chemical
Approach, ed. P. Herdewijn and M. V. Kısakürek, Helvetica Chimica
Acta/Wiley-VCH, Zürich/Weinheim, 2008, pp. 185–214.
2 M. Klussmann, H. Iwamura, S. P. Mathew, D. H. Wells Jr., U. Pandya,
A. Armstrong and D. G. Blackmond, Nature, 2006, 441, 621–623.
3 (a) J. N. H. Reek and S. Otto, Dynamic Combinatorial Chemistry, Wiley-
VCH, Weinheim, 2010; (b) Dynamic Combinatorial Chemistry in Drug
Discovery, Bioorganic Chemistry, and Materials Science, ed.
Lastly, we explored the effects of concentration and solvent on
the compositions of final imine libraries. If the reaction between
This journal is © The Royal Society of Chemistry 2012
Org. Biomol. Chem., 2012, 10, 4847–4850 | 4849

View Article Online
B. L. Miller, Wiley, Hoboken, 2010; (c) P. T. Corbett, J. Leclaire, L. Vial,
K. R. West, J.-L. Wietor, J. K. M. Sanders and S. Otto, Chem. Rev., 2006,
106, 3652–3711; (d) S. J. Rowan, S. J. Cantrill, G. R. L. Cousins,
J. K. M. Sanders and J. F. Stoddart, Angew. Chem., Int. Ed., 2002, 41,
898–952. See also: (e) R. F. Ludlow and S. Otto, Chem. Soc. Rev., 2008,
37, 101–108.
4 M. Hutin, C. J. Cramer, L. Gagliardi, A. R. M. Shahi, G. Bernardinelli,
R. Cerny and J. R. Nitschke, J. Am. Chem. Soc., 2007, 129, 8774–8780.
5 (a) C. D. Meyer, R. S. Forgan, K. S. Chichak, A. J. Peters,
N. Tangchaivang, G. W. V. Cave, S. I. Khan, S. J. Cantrill and
J. F. Stoddart, Chem.–Eur. J., 2010, 16, 12570–12581;
(b) C. D. Pentecost, K. S. Chichak, A. J. Peters, G. W. V. Cave,
S. J. Cantrill and J. F. Stoddart, Angew. Chem., Int. Ed., 2007, 46, 218–222.
6 J. Leclaire, G. Husson, N. Devaux, V. Delorme, L. Charles, F. Ziarelli,
P. Desbois, A. Chaumonnot, M. Jacquin, F. Fotiadu and G. Buono, J. Am.
Chem. Soc., 2010, 132, 3582–3593.
7 (a) M. M. Safont-Sempere, G. Fernandez and F. Würthner, Chem. Rev.,
2011, 111, 5784–5814; (b) K. Osowska and O. Š. Miljanić, Synlett, 2011,
1643–1648; (c) S. Ghosh and L. Isaacs, in Dynamic Combinatorial
Chemistry in Drug Discovery, Bioorganic Chemistry, and Materials
Science, ed. B. L. Miller, Wiley, Hoboken, NJ, 2010, ch. 4, pp. 155–168;
(d) B. H. Northrop, Y.-R. Zheng, K.-W. Chi and P. J. Stang, Acc. Chem.
Res., 2009, 42, 1554–1563; (e) J. R. Nitschke, Acc. Chem. Res., 2007,
40, 103–112.
8 K. Osowska and O. Š. Miljanić, J. Am. Chem. Soc., 2011, 133, 724–727.
9 K. Osowska and O. Š. Miljanić, Angew. Chem., Int. Ed., 2011, 50, 8345–
8349.
10 Yields of individual imines were calculated from the overall mass
recovery of the mixture and 1H NMR spectra integration data. Isolation of
individual imines would have likely altered the distribution of the
products and was not performed.
11 When H2O was slowly added to this diluted mixture of imines, slow and
incomplete precipitation occurred, yielding mixtures of imines in both the
precipitate and the mother liquor.
12 Imine mixtures precipitated from EtOH–H2O exhibit a certain degree of
kinetic stability after re-dissolution in MeCN: mixture obtained in exper-
iment shown in Scheme 2 (bottom) did not change in composition even
after 7 days in MeCN solution at 25 °C. Equilibration into the random
mixture of products was achieved only after heating in MeCN at reflux
for 24 h. This kinetic stability is perhaps unsurprising, as rapid imine
exchange often requires catalysis by free anilines. See, for example:
(a) A. Dirksen, T. M. Hackeng and P. E. Dawson, Angew. Chem., Int. Ed.,
2006, 45, 7581–7584; (b) A. Dirksen, S. Dirksen, T. M. Hackeng and
P. E. Dawson, J. Am. Chem. Soc., 2006, 138, 15602–15603.
13 Concentrations higher than 209 mM were not studied because they led to
the formation of undesired non-imine products. Most significant of these
(29%) was the acetal formed between aldehyde C and EtOH which was
used as the solvent.
4850 | Org. Biomol. Chem., 2012, 10, 4847–4850
This journal is © The Royal Society of Chemistry 2012