Four acid-catalysed dehydration reactions proceed without interference

Article abstract of DOI:10.1039/c4cc02990a

Four acid-catalysed dehydration reactions can proceed in one pot, simultaneously and without interference, to yield one imine, one acetal (or boronic ester), one ester and one alkene, even though many other cross-products could be conceived. This advanced

Full text of DOI:10.1039/c4cc02990a

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Four acid-catalysed dehydration reactions
proceed without interference†
Cite this: Chem. Commun., 2014,
50, 9401
ˇ
Rio Carlo Lirag and Ognjen S. Miljanic*
´
Received 22nd April 2014,
Accepted 1st July 2014
DOI: 10.1039/c4cc02990a
www.rsc.org/chemcomm
Four acid-catalysed dehydration reactions can proceed in one pot, partner association preferences even when they are all placed in a
simultaneously and without interference, to yield one imine, one single solution.
acetal (or boronic ester), one ester and one alkene, even though many
We have recently shown that libraries of as many as 25
other cross-products could be conceived. This advanced self-sorting equilibrating members can self-sort8 into a handful of pure products
behaviour is attributed to different dehydration rates, brought about through iterative application of an irreversible chemical9 or
by dissimilar electronic properties of starting materials.
physical10 stimulus. In this contribution, we examine the question
of whether similar reduction of complexity is possible in a system
An ability to synthesize multiple well-defined products from where chemical functionalities of reacting species are different and
complex (‘‘messy’’) mixtures of starting materials is one of the multiple reactions are involved.
hallmarks of living systems. In fact, at the origin of life, it is likely
As the model system, we chose a set of five reactions, all of
that selective reactions needed to occur within complex chemical which are catalysed by Brønsted acids: (1) imine formation
mixtures that did not yet have highly specific enzyme catalysts.1 from aldehydes and primary aromatic amines; (2) acetalization
Parallel synthesis could also yield tremendous benefits to the of aldehydes and alcohols; (3) boronic ester formation from
chemical industry if multiple value-added products could be alcohols and boronic acids; (4) esterification of carboxylic acids
produced simultaneously in a single reactor. Despite these practical with alcohols; and (5) alkene formation through intramolecular
and fundamental benefits, little work has been devoted to the study dehydration of alcohols. All of these reactions are in principle
of multiple reactions occurring within the same laboratory flask. reversible, but equilibria can be forced toward the dehydrated
Otera’s group has been active in the area of ‘‘shotgun’’ processes,2 products by H2O removal via a Dean–Stark trap. In addition,
wherein two reactions occur at rates sufficiently different to allow compounds like aldehydes, alcohols, and diols can engage in
them to proceed in a single flask without interference. This reactions with multiple partners (e.g. an alcohol can form an
principle was exemplified by the parallel operation of aldehyde ester with a carboxylic acid, a boronic ester with a boronic acid,
allylation and aldol reaction,2a multiple Diels–Alder reactions,2b an acetal with an aldehyde or an alkene). This fact allowed the
and in the one-pot combination of Diels–Alder reaction, alcohol exploration of situations wherein multiple species can compete
acetylation and aldehyde allylation.2c It was also applied in total for a reaction partner. We examined the behaviour of various
syntheses of small natural products.2d,e The same group has also libraries of starting materials chosen from Fig. 1.
demonstrated parallel recognition:3 non-interfering parallel reac-
tions which occur at sites that react completely independently from (Scheme 1: top left and entries #1 and 2 in the table), we
each other. combined equimolar amounts (0.32 mmol) of aldehyde 1a,
Three sets of reactions were performed. In the first set
In the domain of dynamic combinatorial chemistry,4 multiple boronic acid 5a or 5b,11 aniline 2a, diol 6a, carboxylic acid 3a
non-interfering metal–ligand binding events have been utilized by and alcohol 4c, along with a threefold excess of alcohol 4a
Nitschke5 and Schmittel,6 among others. In a seminal paper, Isaacs7 (0.96 mmol).12 A catalytic amount of p-toluenesulfonic acid was
has shown that eight hydrogen-bonding ensembles maintain their added and the solution was heated at reflux for 2 d with H2O
removal. Within such a library of starting materials, many
Department of Chemistry, University of Houston, 112 Fleming Building, Houston,
possible products could have formed: one imine, four acetals,
four boronic esters, four esters and two alkenes, and possibly
other dehydration products (e.g. amides or boroxines). However,
Texas 77204-5003, USA. E-mail: miljanic@uh.edu; Fax: +1 (713)743-2709;
Tel: +1 (832)842-8827
† Electronic supplementary information (ESI) available: Complete reaction pro-
only four discrete products were observed, corresponding to
the reactions highlighted by thick red lines in Scheme 1, top left.
tocols, annotated NMR spectra and characterization data for all new compounds.
See DOI: 10.1039/c4cc02990a
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 9401--9404 | 9401

View Article Online
Communication
ChemComm
The top table in Scheme 1 details this selectivity: one amine
(from 1a and 2a, pathway A), one boronic ester (from 5a/5b and
6a, pathway B), one ester (from 3a and 4a, pathway C) and one
alkene (from 4c, pathway D) were formed, in yields that
typically exceeded 90% as determined by 1H NMR spectroscopy
with an internal standard.
In the second set of thirteen experiments (Scheme 1: top
centre and entries #3–15), boronic acid was replaced by a
second aldehyde, setting up a competition between the two
aldehydes—as each could form either an imine or an acetal, or
both. In total, at least 16 products could be formed: two imines,
eight acetals, four esters and four alkenes. In the benchmark
reaction (entry #3), electron-poor aldehyde 1a reacted with
amine 2a to form an imine (pathway A, 84%) and with 6a to give
a miniscule amount of acetal (pathway E, 0.4%). Electron-rich 1c
was dominantly converted into an acetal with 6a (pathway B, 77%),
Fig. 1 Compounds examined in this study.
Scheme 1 Self-sorting of complex reaction mixtures during acid-catalysed dehydrations. Dominant reaction pathways are highlighted with thick red
lines, minor pathways with full black lines. Pathways that were not observed but were theoretically possible are shown in dashed black lines.
9402 | Chem. Commun., 2014, 50, 9401--9404
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
but a minor imine product (with 2a, pathway F, 9%) was also at reflux (to simulate the conditions at the end of self-sorting
observed. As in the previous set of experiments, carboxylic acid reactions) in the presence of p-toluenesulfonic acid, no equilibration
3a and alcohol 4a formed an ester (pathway C, 96%) and alcohol was observed even after 4 d—suggesting that the mixture
4c exclusively dehydrated into an alkene (pathway D, 99%).
composition more likely corresponds to kinetic distribution
We subsequently probed the effect of replacing individual of the products. No equilibration was observed in a similar
components of this self-sorting system. Aldehydes can be experiment where ‘‘wrong’’ ester and acetal partners were
switched without erosion of selectivity (entries #4 and 5), but heated for 2 d in the presence of an acid catalyst.15
once they become too close in their electronic properties (entry
In conclusion, we have demonstrated that dehydration
#6, where both 1b and 1f are electron-poor), imines and acetals reactions yielding imine, acetal/boronate ester, ester and alkene
are formed indiscriminately. Switching the primary amine products may be accomplished in one-pot with a common—and
(entries #7–9) does not lower self-sorting selectivity. With very simple—acid catalyst. A single set of dehydrated products
alkylamine 2c, esterification of 3a with 4a proceeds in only stems from different mechanisms for the participant reactions
43% yield, which can be rationalized by the formation of small and electronic properties of starting materials, which together
amounts of amide between 2c and 3a (as well as other unidentified translate into sufficiently different rates of competing reactions.
products). As entries #10–12 show, diol can also be varied rather The system is modular, but within limits. While individual
freely, although 6c (entry #11) shows a lower yield, presumably on components can be swapped, examples of aldehydes with similar
account of the Thorpe–Ingold effect. Switching the carboxylic electronic properties and diols of similar chain lengths showed
acid (entries #14 and 15) and carboxylic acid along with alcohol that selectivity quickly erodes when substrates are offered a
(entry #13) still keeps the selectivity largely unaffected, although ‘‘confusing’’ choice of reagents. Future work will focus on applica-
esterification of benzoic acid (3d, entry #15) proceeded only to tions of this process, including testing whether similar parallel
11% conversion.13
reactions occur in mixtures containing activated amino acid
This high preference for the dominant products observed in derivatives, to establish whether privileged oligopeptide structures
experiments #3–5 and #7–15 can tentatively be explained as follows could form in the absence of enzymatic catalysis.
(entry #3 will be used as an example). Donor–acceptor imines are
This research was supported by the National Science Foundation
both more stable and formed faster than their donor–donor (award CHE-1151292) and the Welch Foundation (award E-1768).
ˇ
counterparts. Therefore, the electron-poor 1a reacts quickly with O. S. M. is a Cottrell Scholar of the Research Corporation for Science
amine 2a, forcing aldehyde 1c into an acetal with diol 6a—which is Advancement.
entropically favoured relative to acetals formed from 1c and two
molecules of a monool. Unimolecular dehydration of tertiary
Notes and references
1 Origin of Life: Chemical Approach, ed. P. Herdewin and M. V. Kısaku¨rek,
alcohol 4c into an alkene is quick,14 leaving 4a with no choice
but to slowly (in a bimolecular reaction) form an ester with 3a.
Wiley-VCH, Weinheim, 2008.
Curiously, excess of 4a still does not compete in acetal formation:
it is observed to be unchanged at the end of the reaction.
2 (a) Y. Nagano, A. Orita and J. Otera, Bull. Chem. Soc. Jpn., 2003, 76,
2183–2189; (b) Y. Nagano, A. Orita and J. Otera, Tetrahedron, 2002,
58, 8211–8217; (c) A. Orita, Y. Nagano, K. Nakazawa and J. Otera,
Synlett, 2000, 599–602; (d) Y. Nagano, A. Orita and J. Otera, Synlett,
2003, 684–688; (e) Y. Nagano, A. Orita and J. Otera, Tetrahedron,
2003, 59, 5569–5578.
3 (a) J. Chen and J. Otera, Angew. Chem., Int. Ed., 1998, 37, 91–93; (b) J. Chen,
K. Sakamoto, A. Orita and J. Otera, Tetrahedron, 1998, 54, 8411–8420.
4 (a) J. N. H. Reek and S. Otto, Dynamic Combinatorial Chemistry,
Wiley-VCH, Wenheim, 2010; (b) 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.
5 (a) A. G. Salles Jr., S. Zarra, R. M. Turner and J. R. Nitschke, J. Am.
Chem. Soc., 2013, 135, 19143–19146; (b) 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.
6 (a) M. L. Saha and M. Schmittel, J. Am. Chem. Soc., 2013, 135,
17743–17746; (b) M. L. Saha, J. W. Bats and M. Schmittel, Org.
Biomol. Chem., 2013, 11, 5592–5595; (c) K. Mahata, M. L. Saha and
M. Schmittel, J. Am. Chem. Soc., 2010, 132, 15933–15935.
7 A. Wu and L. Isaacs, J. Am. Chem. Soc., 2003, 125, 4831–4835.
The final set of experiments (Scheme 1: top right and entries
#16 and 17) examined what happens when one of the monools
is replaced with a diol—thus setting up a competition between
two aldehydes and two diols. Thirteen products: two imines,
four acetals, five esters, and two alkenes could be expected.
Intriguingly, in neither of the two experiments, electron-poor
aldehyde 1a formed an acetal—even though it was offered a
choice of diols (therefore, pathway E is omitted from the last
table in Scheme 1). In the first experiment (entry #16), two diols
(6a and 6d) were of the same length and aldehyde 1c could not
distinguish them (pathway B: 54%, pathway H: 46%). Carboxylic
acid then reacted with the excess of 6d to produce an ester
(pathway C, 95%). The second experiment replaced 6d with
longer 6e, restoring the original selectivity: only four products
from pathways A–D were observed.
Observed reaction preferences appear to be kinetic in character.
By focusing on reaction #3 in Scheme 1 and observing its progress
by 1H NMR spectroscopy, we were able to estimate rates of
formation of dehydrated products following the trend: alkene E
imine 4 acetal 4 ester. We then independently prepared the
crossover imine and acetal species not observed in the actual
reaction: imine formed from 1c and 2a and acetal from 1a and 6a.
When these two ‘‘wrong’’ products were heated in anhydrous PhMe
ˇ
´
8 (a) Q. Ji, R. C. Lirag and O. S. Miljanic, Chem. Soc. Rev., 2014, 43,
1873–1884; (b) M. M. Safont-Sempere, G. Fernandez and F. Wu¨rthner,
ˇ
´
Chem. Rev., 2011, 111, 5784–5814; (c) K. Osowska and O. S. Miljanic,
Synlett, 2011, 1643–1648.
ˇ
´
9 K. Osowska and O. S. Miljanic, J. Am. Chem. Soc., 2011, 133, 724–727.
ˇ
´
10 (a) Q. Ji, N. El-Hamdi and O. S. Miljanic, J. Chem. Educ., 2014, 91, 830–833;
ˇ
´
(b) Q. Ji and O. S. Miljanic, J. Org. Chem., 2013, 78, 12710–12716;
(c) R. C. Lirag, K. Osowska and O. S. Miljanic, Org. Biomol. Chem., 2012,
10, 4847–4850; (d) K. Osowska and O. S. Miljanic, Angew. Chem., Int. Ed.,
2011, 50, 8345–8349.
ˇ
´
ˇ
´
11 Formation of imines and boronate esters has been shown to occur
orthogonally, and has been used in the synthesis of cage compounds.
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 9401--9404 | 9403

View Article Online
Communication
ChemComm
¨
See, for example: (a) B. Içli, N. Christinat, J. Tonnemann, C. Schu¨ttler, 14 A carboxylic acid can add to the alkene formed by the dehydration of
R. Scopelliti and K. Severin, J. Am. Chem. Soc., 2009, 131, 3154–3155;
(b) N. Christinat, R. Scopelliti and K. Severin, Angew. Chem., Int. Ed.,
2008, 47, 1848–1852.
the tertiary alcohol (see: J. Otera and J. Nishikido, Esterification.
Methods, Reactions, and Applications, Wiley-VCH, Weinheim, 2nd edn,
2010, p. 10), but we did not observe this process in the examined
reactions. A secondary alcohol (1,2-diphenylethanol) was also used in
place of 4c: while 1H NMR analysis showed no ester formation,
dehydration of the secondary alcohol did not go to completion after
several days at reflux, reaching a maximum conversion of 32%.
12 Alcohol was always added in excess (3 eq. compared to other starting
materials) to ensure complete consumption of carboxylic acid and
to remain available to compete with diol for acetal formation.
13 (a) A. S.-Y. Lee, H.-C. Yang and F.-Y. Su, Tetrahedron Lett., 2001, 42,
301–303; (b) M.-Y. Chen and A. S.-Y. Lee, J. Chin. Chem. Soc., 2003, 15 If either an acetal or the ester bore a free –OH group (e.g. in a
50, 103–108.
monoester of 6d and 3a), equilibration did proceed.
9404 | Chem. Commun., 2014, 50, 9401--9404
This journal is ©The Royal Society of Chemistry 2014