The thiolate-catalyzed intermolecular crossed tishchenko reaction: Highly chemoselective coupling of two different aromatic aldehydes

Article abstract of DOI:10.1002/anie.201206343

Crossed products: Ortho-substituted benzaldehydes react with other aromatic aldehydes in a highly selective, atom-economical Tishchenko disproportionation (see scheme) in the presence of a readily prepared, inexpensive thiolate-based catalyst. The methodo

Full text of DOI:10.1002/anie.201206343

Angewandte
Chemie
DOI: 10.1002/anie.201206343
Homogeneous Catalysis
The Thiolate-Catalyzed Intermolecular Crossed Tishchenko Reaction:
Highly Chemoselective Coupling of Two Different Aromatic
Aldehydes**
Simon P. Curran and Stephen J. Connon*
In memory of Rory A. More OꢀFerrall
The Tishchenko reaction,[1,2] which has been known for
a century, is a Cannizzaro-reaction-like[3] process with two
aldehyde molecules 1 disproportionating in the presence of
a catalyst (usually based on a metal ion[4–6]) to form an ester
product 2 (Figure 1A). Mechanistically the reaction is
thought to involve hydride transfer (3) to furnish a reduced
alkoxide and an acyl electrophile, which couple to form the
ester adduct in an atom-economic, waste-free fashion.[2]
Despite detailed studies,[2,4] the reaction is generally not
considered an important synthetic methodology, largely
because of the difficulty in achieving selectivity: that is,
when two different carbonyl compounds are involved,
exercising control over which species acts as the hydride
donor and which as the hydride acceptor has proven a near-
intractable problem (Figure 1B). The difficulty resides in
bringing about selective hydride transfer to give one ester
product from four possible outcomes (two crossed products,
two dimers).[7]
In 1993, Ishii and co-workers[4k] exploited the steric and
electronic discrepancies between aliphatic and aromatic
aldehydes in moderately selective Zr-complex-mediated
crossed Tishchenko reactions. In these processes, only trace
amounts of products derived from hydride transfer from 5 to
either 5 or 4 were observed, the homodimerization of 4 could
however not be avoided. Much later, Chan and Scheidt[8]
disclosed the first carbene-catalyzed hydroacylation reactions
between benzaldehydes and 1,2-dicarbonyl compounds such
as 7 (Figure 1B) to afford 8 in high yield. In 2010, our group[9]
utilized a similar strategy in the first thiolate-catalyzed
(crossed) Tishchenko reactions (e.g., product 11). Later,
selenide ions were shown to promote this reaction with
improved efficacy.[10]
Recently, a significant advance in this field was reported
by Ogoshi and co-workers.[11] In a mechanistically distinct
process, a nickel(0)/N-heterocyclic carbene complex was
shown to promote selective crossed Tishchenko reactions
between aliphatic and aromatic aldehydes. For example—the
coupling of 12 and 5 to give the ester 13 occurred with
excellent selectivity and in high isolated yield. While this is
a leap forward in Tishchenko chemistry, thus far only
unfunctionlized aldehydes have been utilized—the scope
with respect to the electronic properties of the aldehyde
and compatibility with either basic/chelating functionality or
groups likely to participate in oxidative addition has not been
established.[12]
Figure 1. The crossed intermolecular Tishchenko reaction.
This cursory survey serves to highlight a key missing
component in the crossed Tishchenko reaction toolbox, that
is, no methodologies for the selective catalytic crossed
coupling of two different aromatic aldehydes are currently
known.[13]
[*] S. P. Curran, Prof. S. J. Connon
Centre for Synthesis and Chemical Biology, Trinity
Biomedical Sciences Institute, School of Chemistry
The University of Dublin, Trinity College, Dublin 2 (Ireland)
E-mail: connons@tcd.ie
The strategies outlined in Figure 1B all rely on a discrep-
ancy in characteristics (either steric or electronic or both)
between very different coupling partners (e.g., between
aliphatic and aromatic aldehydes or between aldehydes and
[**] Financial support from the European Research Council is gratefully
acknowledged.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201206343.
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1
These are not the final page numbers!

.
Angewandte
Communications
ketones etc.). Engendering and then exploiting such a scenario
in a synthetically useful manner, in which both aldehyde
partners are aromatic (flat, conjugated, similar in size) is
a considerable challenge. Herein, we disclose the results of
a study aimed at developing the first such crossed Tishchenko
reactions between aromatic aldehydes (i.e., toward products
16, Figure 1C).
Our investigation began with an attempt to bias the
reaction outcome through manipulation of the electronic
characteristics of the substrate (Scheme 1).[14] Accordingly, we
ity-determining processes the aldehyde component is the
electrophile. Therefore, if one attempts to render the process
selective by further increasing the electrophilicity of one
component relative to the other, homodimerization of the
more electrophilic aldehyde is likely to result.
We therefore attempted to influence the reaction through
the modification of the steric properties of one aldehyde. Our
hypothesis was simple: while a solution to the problem
outlined above was not obvious right away, the greatest
difference between the two selectivity-determining steps is
the bulk of the attacking nucleophilic components (i.e., the
magnesium thiolate catalyst and the hemithioacetal conjugate
base 27). Therefore, the perhaps most promising tactic to
influence the process is to modulate the bulk of the electro-
philic aldehyde component.[17] Gratifyingly, exchange of 20
for o-tolylaldehyde (30) led to the formation of the crossed
product 31 as the major component of the crude material
(Scheme 2).
Scheme 1. Preliminary investigation into the effects of electronic char-
acteristics on chemoselectivity.
Scheme 2. Preliminary investigation into the effects of steric character-
istics on chemoselectivity.
reacted p-anisaldehyde (20) with the more activated p-
chlorobenzaldehyde (21) in the presence of the magnesium
thiolate derived from thiol 22 (20 mol%). This inexpensive
thiol (considerably less smelly than benzyl mercaptan) was
previously found to serve as a convenient precatalyst for the
promotion of the Tishchenko homodimerization of benzalde-
hydes.[9] All four possible benzyl ester products were formed
in a relatively unselective process.
Since 31 derives from 30, which acts as a hydride acceptor
in the reaction, we next evaluated the corresponding tri-
fluoro-derivative 32, which is more electrophilic than 30 but
possesses similar steric characteristics. A very selective
reaction between this aldehyde and 21 occurred, which
furnished the expected crossed product 33 in 80% yield.
While these examples represent a step toward high
selectivity in crossed Tishchenko reactions involving aromatic
aldehydes, the process would not be synthetically useful,
while the scope is limited by a requirement for permanent
bulky substituents at the ortho position of one of the reacting
aldehydes. The precise origin of the observed chemoselectiv-
ity was also unclear. Hindering one aldehyde would certainly
reduce the rate of its dimerization; however, it might also be
expected to increase the relative rate of dimerization of the
other, less-hindered aldehyde, which could then (relatively)
easily participate in both the 1,2 addition and hydride transfer
steps of the catalytic cycle (see Figure 2).
The inherent difficulties associated with this process are
apparent from an examination of the proposed catalytic cycle
(Figure 2).[9a,15] From a chemoselectivity standpoint, the two
If both these steps are reversible, the chemoselectivity
could (at least in part) be due to the difference in stability of
the sp3-hybridized adducts arising from these steps. For
example, initial addition of the thiolate to the hindered
aldehyde 34 would generate 35, a species that would be
unable to avoid the steric strain associated with interaction of
the ortho substituent and the methine group (Figure 3A).
Thus, one would expect the formation of the hemithioacetal
conjugate base, derived from attack of the catalyst on the less-
hindered aldehyde 21, to be more energetically favorable. In
the hydride-transfer step (Figure 3B), the alkoxide adduct 37
is less sterically congested than 35, and thus steric effects may
Figure 2. Proposed catalytic cycle for the thiolate-catalyzed crossed
Tishchenko reaction.
key steps in the catalytic cycle are: a) 1,2 addition of the
thiolate (e.g., 14), and b) hydride transfer from the resultant
adduct 27 to 15 (in a process reminiscent of the hydride
transfer catalyzed by the glycolytic enzyme glyceraldehyde-3-
phosphate dehydrogenase,[16] that is, 27a), to give alkoxide 28
and thioester 29. An obvious impediment to achieving
selectivity in such a reaction is that in both chemoselectiv-
2
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
yields of isolated products. When both aldehyde partners
contained chelating functionality in close proximity to the
aldehyde, the resulting selectivity was poor (43 and 44,
entries 4–5). However, heteroaromatic aldehydes that were
not predisposed to chelation, such as thiophene-3-carbalde-
hyde (which contains a sulfur atom that is potentially
incompatible with soft-transition-metal-based catalysts; 45,
entry 6) and the highly electrophilic (and hence dimerizable)
pyridine-4-carbaldehyde (46, entry 7) were compatible with
the reaction.
Figure 3. Rationalization of the observed chemoselectivity and the
proposed influence of an ortho chelating group.
These data strongly indicate that a combination of steric
bulk and chelating ability in just one of the aldehyde
components is sufficient to render the crossed Tishchenko
process highly selective. In order to improve the utility of the
process, we next probed the use of removable chelating
functionality. This would allow the aspiration toward the use
of an ortho substituent as a chemoselectivity-directing group,
which could be either removed later or utilized as a handle for
further elaboration.[20] Upon reaction with 21, o-chloro-, o-
bromo-, and o-tosyl-benzaldehyde (47, 48, and 49 respec-
tively) participated in selective crossed couplings to form 50,
51, and 52; with the latter two substrates providing the
product in more than 90% yield (Scheme 3).[21] The use of
a hydrogenolysis-labile sulfide chelating group (i.e., 49a) is
also feasible, but produces the product in lower yield.[21]
have less influence over this reaction than in the catalyst-
addition step. This argument explains the formation of 31 and
33 (Scheme 2), however, it fails to account for the fact that the
homodimers of 21 are not formed as the major product.[18]
We wished to develop a process that was both selective
but also of the broadest possible scope and potential synthetic
utility. To this end, we speculated that the use of a group
capable of chelating magnesium ions in the ortho position
could be advantageous (38, Figure 3C). Such a group could
provide the requisite steric bulk, yet also serve to stabilize the
hydride-transfer adduct (39, Figure 3C),[19] which we hoped
would reinforce the previously observed bias toward the
hindered aldehyde acting as the hydride acceptor.
To test this hypothesis, we evaluated the use of o-
anisaldehyde as a coupling partner (Table 1). The expected
crossed products of reactions with either activated (product
40, entry 1), electron-neutral (41, entry 2), or deactivated
aldehydes (42, entry 3) were obtained in good to excellent
Table 1: Selective crossed Tishchenko reactions with o-anisaldehyde.
Scheme 3. Use of removable chelating substituents.
Entry
1
Product
t [h]
Yield [%][a]
88
To demonstrate that this is a general phenomenon, we
carried out crossed Tishchenko reactions in which one of the
benzaldehyde partners contained two substituents, one of
which was an ortho chelating group (Table 2). Chemoselec-
tivity again was determined by the presence of the ortho
substituent, irrespective of the overall electronic character-
istics of the aldehyde, with possible yields of isolated products
of more than 90%, even using challenging heteroaromatic
and relatively electron-rich coupling partners (53–55,
entries 1–3). The electrophilic pyridine-4-carbaldehyde was
again compatible with the reaction, allowing the formation of
the highly manipulatable esters 56 and 57 (entries 4 and 5).
The tolerance toward chelating functionality is underlined by
the efficient formation of 58, which contains an acetal-
protected aldehyde and a benzylated phenol group (entry 6).
It was also found that the ortho substituent could facilitate
crossed coupling with aliphatic aldehydes (59 and 60, entries 7
and 8) in moderate to good yields. The sulfide group, which
has the potential to poison soft-metal-ion catalysts, could also
be tolerated without loss of catalytic efficiency (i.e., 61,
entry 9). Thus, while this thiolate-mediated reaction (the
40
41
42
43
44
45
46
24
2
3
4
5
6
7
24
24
24
24
24
36
73
69
16[b]
38[b]
71
63
[a] Yields of isolated products after chromatography. [b] Determined by
1H NMR spectroscopy using an internal standard.
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
3
These are not the final page numbers!

.
Angewandte
Communications
Table 2: Expansion of the reaction scope.
Tishchenko reaction detailed in Scheme 1, which produces 23
in only 27% yield as part of a mixture of four products.
In summary, we have developed the first highly selective
Tishchenko coupling reactions between two different aro-
matic aldehydes. The protocol uses a readily prepared,
inexpensive catalyst and produces crossed products in excel-
lent yields. The reaction relies on the use of an ortho
substituent (usually bromo), however, this can either be
efficiently removed after reaction, or provide opportunities
for further synthetic manipulation.[20] In the reactions detailed
above, a very wide range of aromatic aldehydes are tolerated,
and the methodology can be also utilized to bring about
selective couplings between aliphatic and functionalized
aromatic aldehydes.
Entry
1
Product
Yield [%][a]
93
53
54
2
3
4
5
6
95
99
72
69
92
55
56
57
58
Received: August 7, 2012
Published online: && &&, &&&&
Keywords: disproportionation · esters · hydride transfer · thiols ·
.
Tishchenko reaction
[1] a) W. Tischtschenko, J. Russ. Phys. Chem. 1906, 38, 355; b) W.
Tischtschenko, Chem. Zentralbl. 1906, 771, 1309; c) L. Claisen,
Ber. Dtsch. Chem. Ges. 1887, 20, 646.
[2] For recent reviews, see: a) T. Seki, T. Nakajo, M. Onaka, Chem.
Lett. 2006, 35, 824; b) O. P. Tçrmꢀkangas, A. M. P. Koskinen,
Recent Res. Dev. Org. Chem. 2001, 5, 225.
58 (72)[b]
63
[3] S. Cannizzaro, Justus Liebigs. Ann. 1853, 88, 129.
7
8
59
60
[4] Aluminium alkoxides: a) W. C. Child, H. Adkins, J. Am. Chem.
Soc. 1925, 47, 798; b) F. J. Villani, F. Nord, J. Am. Chem. Soc.
1947, 69, 2605; c) L. Lin, A. R. Day, J. Am. Chem. Soc. 1952, 74,
5133; d) T. Saegusa, T. Ueshima, J. Org. Chem. 1968, 33, 3310;
e) T. Ooi, T. Miura, K. Takaya, K. Maruoka, Tetrahedron Lett.
1999, 40, 7695; f) T. Ooi, K. Ohmatsu, K. Sasaki, T. Miura, K.
Maruoka, Tetrahedron Lett. 2003, 44, 3191; g) Y. Hon, C. Chang,
Y. Wong, Tetrahedron Lett. 2004, 45, 3313; transition-metal-
based catalysts (selected examples): h) M. Yamashita, Y. Wata-
nabe, T. Mitsudo, Y. Takegami, Bull. Chem. Soc. Jpn. 1976, 49,
3597; i) T. Ito, H. Horino, Y. Koshiro, A. Yamamoto, Bull. Chem.
Soc. Jpn. 1982, 55, 504; j) S. H. Bergens, D. P. Fairlie, B. Bosnich,
Organometallics 1990, 9, 566; k) K. Morita, Y. Nishiyama, Y.
Ishii, Organometallics 1993, 12, 3748; l) M. Yamashita, T. Ohishi,
Appl. Organomet. Chem. 1993, 7, 357; m) P. Barrio, M. A.
Esteruelas, E. Onate, Organometallics 2004, 23, 1340; n) T.
Suzuki, T. Yamada, T. Matsuo, K. Watanabe, T. Katoh, Synlett
2005, 1450; o) S. Ogoshi, Y. Hoshimoto, M. Ohashi, Chem.
Commun. 2010, 46, 3354; alkali metal salts: p) M. M. Mojtahedi,
E. Akbarzadeh, R. Sharifi, M. S. Abaee, Org. Lett. 2007, 9, 2791;
q) D. C. Waddell, J. Mack, Green Chem. 2009, 11, 79; r) T.
Werner, J. Koch, Eur. J. Org. Chem. 2010, 6904; alkali-/earth-
metal-based catalysts: s) M. R. Crimmin, A. G. M. Barrett, M. S.
Hill, P. A. Procopiou, Org. Lett. 2007, 9, 331; t) B. M. Day, N. E.
Mansfield, M. P. Coles, P. B. Hitchcock, Chem. Commun. 2011,
47, 4995.
9
61
62
[a] Yields of isolated products after chromatography. [b] Using
2.0 equivalents of the o-bromobenzaldehyde.
primary application of which is in aromatic–aromatic alde-
hyde couplings) is not as efficient as the Ni0 [11] system in
aliphatic–aromatic coupling reactions, it could serve as
a complementary strategy for use with functionalized alde-
hydes.[22]
Finally, we demonstrated that the directing bromine atom
can be easily removed after reaction to provide highly
efficient access to products that were hitherto formed only
through unselective Tishchenko chemistry. Benzyl ester 53
was prepared through a selective crossed Tishchenko reaction
(Table 2, entry 1) and subsequently debrominated by using
ammonium formate in the presence of catalytic Pd0 to give 23
in 95% yield (88% over two steps from the parent aldehydes,
Scheme 4). This compares very favorably with the unselective
[5] For the use of boric acid as a catalyst, see: P. R. Stapp, J. Org.
Chem. 1973, 38, 1433.
[6] a) D. Ulrich, H. Jankowski, Chem. Tech. 1988, 40, 393; b) K.
Weissermel, H. J. Arpe, Industrial Organic Chemistry, 4th ed.,
Wiley-VCH, Weinheim, 2003; c) A. S. Hester, K. Himmler, Ind.
Eng. Chem. 1959, 51, 1424.
[7] For intramolecular crossed Tishchenko reactions, see: a) J. L.
Hsu, J. M. Fang, J. Org. Chem. 2001, 66, 8573; b) L. Lu, H. Y.
Chang, J. M. Fang, J. Org. Chem. 1999, 64, 84.
Scheme 4. Removal of the chelating substituent.
4
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
These are not the final page numbers!

Angewandte
Chemie
[8] A. Chan, K. A. Scheidt, J. Am. Chem. Soc. 2006, 128, 4558.
[9] a) L. Cronin, F. Manoni, C. J. OꢁConnor, S. J. Connon, Angew.
Chem. 2010, 122, 3109; Angew. Chem. Int. Ed. 2010, 49, 3045;
b) C. J. OꢁConnor, F. Manoni, S. P. Curran, S. J. Connon, New J.
Chem. 2011, 35, 551.
[10] S. P. Curran, S. J. Connon, Org. Lett. 2012, 14, 1074.
[11] Y. Hoshimoto, M. Ohashi, S. Ogoshi, J. Am. Chem. Soc. 2011,
133, 4668.
[12] For a Highlight article, see: W. I. Dzik, L. J. Gooßen, Angew.
Chem. 2011, 123, 11241; Angew. Chem. Int. Ed. 2011, 50, 11047.
[13] For example, Ogoshi and co-workers (Ref. [11]) attempted both
these types of reactions under conditions previously found to be
beneficial to selective aliphatic–aromatic Tishchenko chemistry.
However, reactions involving either two different aliphatic or
two different aromatic aldehydes were found to be unselective.
[14] Eisen had already shown that actinide-complex-catalyzed
crossed Tishchenko reactions involving aromatic aldehydes of
similar electronic characteristics were unselective: T. Andrea, E.
Barnea, M. S. Eisen, J. Am. Chem. Soc. 2008, 130, 2454.
[15] The thioester- and alkoxide-derived alcohol can be observed
during the reaction by 1H NMR spectroscopy and both have
been shown to serve as competent intermediates in the cycle. See
Ref. [9a].
[17] Our group in collaboration with Zeitler and co-workers
employed a similar strategy to develop highly selective (mech-
anistically distinct) crossed acyloin reactions. S. E. OꢁToole,
C. A. Rose, S. Gundala, K. Zeitler, S. J. Connon, J. Org. Chem.
2011, 76, 347.
[18] It is tempting to ascribe this observation to reduced conjugation
between the carbonyl group and the electron-rich aromatic ring,
however, attempts to rationalize this observation are perhaps
best left until after the emergence of a clearer mechanistic
picture.
[19] Naturally, the chelating substituent would also stabilize the
hemithioacetal conjugate base. However, as we have pointed
out, steric effects are likely to make more of a contribution to the
stability (or lack thereof) of that species than to the hydride-
transfer adduct.
[20] For example, Nicolaou et al. have employed benzyl benzoates
containing o-bromo substituents (similar to 51) as key compo-
nents in the total synthesis of balanol (and analogues thereof):
K. C. Nicolaou, K. Koide, M. E. Bunnage, Chem. Eur. J. 1995, 1,
454.
[21] The coupled product 47 could not be cleanly isolated from the
crude material. Therefore the yield of this compound was
determined unambiguously by 1H NMR spectroscopy using
styrene as an internal standard. The yields given for 48, 49,
and 49a are those of isolated products. For a review of the
hydrogenolysis of sulfur-containing compounds using Raney
nickel, see: H. Hauptmann, W. F. Walter, Chem. Rev. 1962, 62,
347.
[16] a) J. M. Berg, J. L. Tymoczko, L. Stryer, Biochemistry, 5th ed.,
Freeman, New York, 2002; b) A. Soukri, A. Mougin, C. Corbier,
A. Wonacott, C. Branlant, G. Branlant, Biochemistry 1989, 28,
2586; c) K. DꢁAmbrosio, A. Pailot, F. Talfournier, C. Didierjean,
E. Benedetti, A. Aubry, G. Branlant, C. Corbier, Biochemistry
2006, 45, 2978.
[22] Aliphatic aldehydes containing a-unbranched aldehydes
undergo preferential aldol condensation under these conditions.
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
5
These are not the final page numbers!

.
Angewandte
Communications
Communications
Homogeneous Catalysis
S. P. Curran,
S. J. Connon*
&&&& — &&&&
The Thiolate-Catalyzed Intermolecular
Crossed Tishchenko Reaction: Highly
Chemoselective Coupling of Two
Different Aromatic Aldehydes
Crossed products: Ortho-substituted
benzaldehydes react with other aromatic
aldehydes in a highly selective, atom-
economical Tishchenko disproportiona-
tion (see scheme) in the presence of
a readily prepared, inexpensive thiolate-
based catalyst. The methodology is of
exceptionally wide scope and exhibits
a high functional-group tolerance.
6
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1 – 6
These are not the final page numbers!