Bis(amino)cyclopropenylidene (BAC) catalyzed aza-benzoin reaction

Article abstract of DOI:10.1021/ol5024807

A bis(amino)cyclopropenylidene (BAC) catalyzed aza-benzoin reaction between aldehydes and phosphinoyl imines has been developed. The reaction is general with a wide range of aromatic aldehydes and aromatic imines. The reaction displays excellent chemoselectivity favoring aza-benzoin products over homobenzoin products. (Chemical Equation Presented).

Full text of DOI:10.1021/ol5024807

Letter
pubs.acs.org/OrgLett
Bis(amino)cyclopropenylidene (BAC) Catalyzed Aza-Benzoin
Reaction
Myron M. D. Wilde and Michel Gravel*
Department of Chemistry, University of Saskatchewan, 110 Science Place, Saskatoon, Saskatchewan S7N 5C9, Canada
S
* Supporting Information
ABSTRACT: A bis(amino)cyclopropenylidene (BAC) catalyzed aza-benzoin
reaction between aldehydes and phosphinoyl imines has been developed. The
reaction is general with a wide range of aromatic aldehydes and aromatic imines. The
reaction displays excellent chemoselectivity favoring aza-benzoin products over
homobenzoin products.
rganocatalysis can be an efficient metal-free way to
Among early work, Miller and Mennen used benzoyl
aldimines for enantioselective aza-benzoin-reactions catalyzed
by a thiazolium-containing tripeptide.6b You’s group developed
a robust reaction using achiral thiazolium salts and N-Ph
aldimines.6f However, the resulting products feature a difficult
to cleave N-phenyl substituent making deprotection and
derivatization difficult. The Rovis group has reported a highly
enantioselective method using well developed chiral triazolium
catalysts with aliphatic aldehydes and N-Boc protected
imines.6k This methodology, though efficient, is reportedly
not tolerant to aromatic aldehydes.
A potential problem encountered when performing aza-
benzoin reactions is the formation of benzoin side products
(Scheme 1). When these products are formed reversibly, the
thermodynamic drive toward aza-benzoin products simplifies
the matter considerably. Ensuring this reversibility however
introduces an unnecessary limitation upon aldehyde substrates,
potential aza-benzoin acceptors, and suitable reaction con-
ditions. Under the conditions reported by Murry, Frantz and
co-workers for instance,6b the benzoin products were formed
irreversibly, underscoring the importance of developing other
methods of governing chemoselectivity in the aza-benzoin
reaction. The presence of benzoin products can also complicate
the purification of the desired aza-benzoin products. The
Scheidt group has reported a methodology using acylsilanes as
aldehyde surrogates, along with the easily removable N-
diphenylphosphinoyl protecting group.12 They posit the
acylsilane is required specifically to avoid the formation of
the benzoin product, thus facilitating purification. Based on our
previous observation that BACs are poor catalysts for the
benzoin reaction, we hypothesized that an appropriate choice of
reactants and conditions would allow the formation of aza-
benzoin products without any of the problems associated with
competing benzoin reactivity.
O
construct complex organic molecules.1 In recent years,
organocatalysis based on the umpolung of functional groups
with N-heterocyclic carbenes (NHCs) has enabled numerous
attractive modes of reactivity and has led to the development of
several reactions accomplishing otherwise challenging bond
connections.2 We recently disclosed the ability of bis(amino)-
cyclopropenylidenes (BACs) to competently catalyze the
intermolecular Stetter reaction between an aldehyde and a
Michel acceptor.3 The BAC catalysts are advantageous, as they
can easily be prepared in a one-pot reaction. These BAC
species are also intriguing because they possess similar
characteristics to NHCs, such as the ability to catalytically
induce acyl anion reactivity in aldehydes.3−5 However, they
exhibit markedly different organocatalytic behavior. In our prior
work, several umpolung reactions were observed using BACs
without the generation of benzoin side products, and no
benzoin products were observed when reaction progress was
3
1
followed by H NMR or thin layer chromatography. Indeed,
the benzoin reaction using BACs is extremely limited, even
under ideal conditions. This is in stark contrast to most popular
NHCs for which benzoin reactivity is often concomitant with
the reaction of interest.
The aza-benzoin reaction is a carbene catalyzed coupling
between an aldehyde and an imine to generate an α-amino
ketone (Scheme 1).6,7 It is an attractive transformation due to
Scheme 1. NHC vs BAC Catalyzed Aza-Benzoin Reaction
its convergent nature, the convenient accessibility of the
starting materials, and the potential to direct stereoselectivity at
the new stereogenic center. The resulting α-amino ketone motif
is present in a number of medicinal agents,8 and these moieties
can be reduced to useful α-amino alcohols9 or they can be
converted to a number of substituted heterocycles10 such as
oxazoles or imidazoles which can also be bioactive.11
With the goal of finding complementary or otherwise
beneficial reactivity as well as potentially addressing some of
the shortcomings of current methodologies, we set about
Received: August 21, 2014
Published: October 2, 2014
© 2014 American Chemical Society
5308
dx.doi.org/10.1021/ol5024807 | Org. Lett. 2014, 16, 5308−5311

Organic Letters
Letter
exploring suitable approaches for a BAC catalyzed aza-benzoin
reaction. Our first task was to evaluate the competency of BAC
catalysts for the aza-benzoin reaction. Initial efforts with Boc
protected imines or tosyl imines gave no product or ill defined
mixtures. Boc imines were also found to inhibit otherwise
productive Stetter reactions.3 Fortunately, reactions between
aldehydes and N-diphenylphosphinoylimines or their sulfinic
acid adducts gave productive reactions using our reported
bis(diethylamino)cyclopropenium salt 8 as a precatalyst. N-
Diphenylphosphinoylimines13 can be easily formed in one step
by condensation of the corresponding aldehyde and diphenyl-
phosphinamide.13d The corresponding sulfinic acid surrogates
are trivial to prepare and are more practical to use, as they are
less sensitive to moisture.13b
aromatic or ortho substituted aromatic aldehydes led to sluggish
and low yielding reactions (not shown). To better understand
the influence of substituents on the imine acceptor, reactions
were performed with P,P-diphenyl-N-(4-methoxyphenyl(tosyl)-
methyl) phosphinic amide 7c as well as p-P,P-diphenyl-N-(4-
bromophenyl(tosyl)methyl) phosphinic amide 7d. Excellent
reactivity was observed with electron-rich acceptor 7c (16−18).
Using electron-poor acceptor 7d also led to an efficient
reaction, although difficulties were encountered in the isolation
of the aza-benzoin product 20. Meta substitution on the imine
acceptor (7e) led to product 21 in modest yield, though the use
of more electrophilic imines was not explored. The imine
derived from o-chlorobenzaldehyde was also probed, but its use
only led to low conversion (<10%). At this point, modifications
to the current methodology seem required to obtain the
products of ortho-substituted imines in good yield.
In the course of our studies, we explored the effect of using
additional equivalents of aldehyde to the reaction. It was found
to be beneficial to use an excess of aldehyde in some instances
(11, 14, 15, 17, and 18), and this could drive modestly yielding
reactions further to completion. There were no instances where
excess aldehyde was detrimental to the reaction. The excess
aldehyde was still present in the crude reaction mixture, and it
did not form any detectable amounts of benzoin product in the
presence of catalyst 8. The excess aldehyde could even be
recovered, though recovery was not practical on this scale.
There has been discussion in the literature regarding the
reversible formation of adducts between NHC and imines,6e,k
so the direct use of imines instead of their sulfinic acid adducts
was probed in this reaction. Good yields were obtained in the
two cases investigated (10: 75%, 11: 96%), demonstrating the
ability of phospinoylimines to undergo the aza-benzoin reaction
efficiently. Nevertheless, the increased bench stability of the
sulfinic acid adducts makes the latter the preferred reaction
partners in most cases.
Once optimal conditions were established (see Supporting
Information), a scope exploration was undertaken starting with
the p-tolyl bearing acceptor 7b (Scheme 2). The reactivity of
Scheme 2. Scope of the BAC Catalyzed Aza-Benzoin
a
Reaction
To compare the effectiveness of various catalysts in this
transformation, a number of parallel experiments were
undertaken (Table 1). Our previous experience with triazolium
Table 1. Catalyst Comparison
recovered imine recovered
aza-benzoin
a
benzoin
a
a
a
entry catalyst
22 (%)
6a (%)
11 (%)
23 (%)
a
b
N-Dpp = NPOPh2. Reactions performed on a 0.15 mmol scale.
Yields of isolated pure products. c Reaction done with the imine of 7b.
1
2
3
8
8
35
1
79
91
70
−
67
d 3 equiv of aldehyde were used. The corresponding imine was used
e
24
−
−
as substrate on a 0.3 mmol scale. f 5 equiv of aldehyde were used. g 1
NMR conversion in parentheses.
H
25
−
47
a
Determined from 1H NMR integration relative to a known amount of
internal standard.
this acceptor was found to be similar to that of substrate 7a (9
and 10). It was found that a good yield could be obtained with
P,P-diphenyl-N-(4-methylphenyl(tosyl)methyl) phosphinic
amide 7b and a number of para and meta substituted
electron-deficient aromatic aldehydes (10, 12, 13, and 14) as
well as heteroaromatic aldehydes (11). Benzaldehyde was also
shown to be a competent reaction partner (15). As is typically
the case in NHC organocatalysis, the use of electron-rich
salts indicated their reactions with imines give much superior
results to their reactions with sulfinic acid adducts. To establish
a valid comparison between catalysts, we therefore opted to use
phosphinoylimine 22 as a reaction partner. It was found that
precatalyst 8 gave a suitable reaction with excess furfural (6a),
with a large amount of aldehyde remaining after workup despite
its volatility (entry 1). Thiazolium 24 gave a very high
5309
dx.doi.org/10.1021/ol5024807 | Org. Lett. 2014, 16, 5308−5311

Organic Letters
Letter
conversion as well, with a large amount of accompanying
benzoin side-product 23 (entry 2). Triazolium 25 also afforded
good conversion, again with substantial benzoin contamination,
along with other unidentified side products (entry 3).
Scheme 4. Retro-Aza-Benzoin Stetter Reaction
To elucidate whether the absence of benzoin side products
when using catalyst 8 is due to a facile retro-benzoin/aza-
benzoin reaction, benzoin compound 26 was subjected to
reaction conditions as an aldehyde equivalent (Scheme 3). As
no aza-benzoin product was observed, we conclude the benzoin
reaction does not occur with catalyst 8 under these conditions.
At this stage in the reaction development, efforts were
undertaken to induce enantioselectivity in the reaction using
chiral BAC catalyst 29 (Scheme 5). However, the product
Scheme 3. Retro-Benzoin Aza-Benzoin with BAC 8
Scheme 5. Aza-Benzoin Reaction Using Chiral BAC Catalyst
As BACs were shown previously to be effective catalysts for
the Stetter reaction, competition experiments were designed to
explore the relative reactivity of phosphinoyl imines and
Michael acceptors toward the putative Breslow intermediate
formed between the aldehyde and catalyst (Table 2). In these
obtained was found to be racemic under our optimal
conditions. Miller and Mennen6d reported a high incidence of
postreaction epimerization over time, and Rovis’ group6k
reported cooling the reaction to be necessary to minimize
epimerization. As the same chiral BAC catalyst was found to
induce modest enantioselectivities in the Stetter reaction,3 it
was hypothesized that the aza-benzoin product was undergoing
racemization. This racemization could occur via a retro-aza-
benzoin process or base-induced epimerization. In an attempt
to minimize the risk of epimerization, the reaction was
performed using the imine substrate and substoichiometric
amounts of base. Unfortunately, the product obtained under
these conditions proved to be racemic as well. NMR
experiments showed no deuterium incorporation at the
stereogenic center in the presence of cesium carbonate and
D2O, suggesting a facile retro-aza-benzoin is the more likely
scenario. Efforts to obtain enantiomerically enriched aza-
benzoin products using different chiral BAC catalysts are
continuing.
Table 2. Aza-Benzoin Stetter Competition Reaction
a
ratio of products
yield of 10
yield of 28
b
b
entry catalyst
(10:28)
(%)
(%)
1
2
3
8
0.36:1
1:0.13
1:0
10
18
8
50
4
25
26
−
a
Determined from 1H NMR integration on the crude reaction
mixture. Homobenzoin product 26 was not detected, but could be
b
present in small amounts. Yields of isolated pure products.
There has been a range of conditions reported for the
deprotection of phosphinic amides,12,14 so to further
demonstrate the usefulness of this methodology a representa-
tive aza-benzoin product 11 was deprotected (Scheme 6). It
reactions, we had a limiting amount of aldehyde 6b compete for
p-tolyl phosphinoyl imine 22 and chalcone 27 to find out which
reaction was favored with BAC catalysts. Under these
conditions, the Stetter reaction was favored. To determine
whether this reaction was under catalyst or substrate control,
we then performed reactions with thiazolium 24 and triazolium
25. Both catalysts were less efficient and favored the aza-
benzoin reaction. The reversibility of the aza-benzoin reaction
and the thermodynamic preference for the Stetter product was
subsequently established (Scheme 4). Specifically, we found
that subjecting aza-benzoin product 10 to the reaction
conditions with chalcone in the absence of any aldehyde led
to the Stetter product (28) and imine 22 when using 8 as the
catalyst, in addition to recovered starting material 10. This
result shows the reversible nature of the aza-benzoin reaction
when using catalyst 8. Whether the selectivities observed in
Table 2 are the result of kinetic or thermodynamic control (or
both) is unclear and is the object of further investigations.
Scheme 6. Removal of Protecting Group
was found that stirring the product at rt in a mixture of THF
and 2 N hydrochloric acid (1:1) was sufficient to effect
complete deprotection. However, the product was not stable as
the free base and it proved necessary to reprotect the amine
with a Boc group to obtain a good yield.
To conclude, we report the first examples of the BAC
catalyzed aza-benzoin reaction. As was the case for BAC
catalyzed Stetter reactions, no competing benzoin dimerization
5310
dx.doi.org/10.1021/ol5024807 | Org. Lett. 2014, 16, 5308−5311

Organic Letters
Letter
Am. Chem. Soc. 2008, 130, 418. (h) Wu, K.-J.; Li, G.-Q.; Li, Y.; Dai, L.-
X.; You, S.-L. Chem. Commun. 2011, 47, 493. (i) Sun, F.-g.; Ye, S. Org.
Biomol. Chem. 2011, 9, 3632. (j) Uno, T.; Kobayashi, Y.; Takemoto, Y.
Beilstein J. Org. Chem. 2012, 8, 1499. (k) DiRocco, D. A.; Rovis, T.
Angew. Chem., Int. Ed. 2012, 51, 5904.
was observed. The reaction is effective for various aromatic and
heteroaromatic aldehydes and a wide range of aromatic
phosphinoyl imines (as their sulfinic acid adducts). The
phosphinoyl activating group was found to be compatible
with this methodology and cleavable under mild conditions.
(7) Aza-benzoin reactions of ketimines: (a) Enders, D.; Henseler, A.;
Lowins, S. Synthesis 2009, 24, 4125. (b) Sun, L.-H.; Liang, Z.-Q.; Jia,
W.-Q.; Ye, S. Angew. Chem., Int. Ed. 2013, 52, 5803. (c) Xu, J.; Mou,
C.; Zhu, T.; Song, B.-A.; Chi, Y. R. Org. Lett. 2014, 16, 3272.
(8) (a) Gioia, D.; Leggio, M. L.; Liguori, A.; Napoli, A.; Siciliano, A.;
Sindona, C. J. Org. Chem. 2001, 66, 7002. (b) Beguin, C.; Andurkar, S.
V.; Jin, A. Y.; Stables, J. P.; Weaver, D. F.; Kohn, H. Bioorg. Med. Chem.
2003, 11, 4275. (c) Shishido, Y.; Jinno, M.; Ikeda, T.; Ito, F.; Sudo, M.;
Makita, N.; Ohta, A.; Iki-Taki, A.; Ohmi, T.; Kanai, Y.; Tamura, T.;
Shimojo, M. Bioorg. Med. Chem. Lett. 2008, 18, 1072. (d) Mehta, N. B.;
Yeoweil, D. A. Biologisch wirksame ketone. DE2059618, Oct 7, 1971.
(e) Canning, H.; Goff, D.; Leach, M. J.; Miller, A. A.; Tateson, J. E.;
Wheatley, P. L. Br. J. Pharmacol. 1979, 66, 104P.
ASSOCIATED CONTENT
* Supporting Information
■
S
Experimental details and full spectroscopic data are available for
all substrates and aza-benzoin products, as well as their
1
derivatives. H NMR spectra of the crude reaction mixture
are included for all competition experiments. This material is
available free of charge via the Internet at http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
■
(9) Ager, D. J.; Prakash, I.; Schaad, D. R. Chem. Rev. 1996, 96, 835.
(10) (a) Taguchi, H.; Yokoi, T.; Tsukatani, M.; Okada, Y.
Tetrahedron 1995, 51, 7361. (b) Medaer, B. P.; Hoornaert, G. J.
Tetrahedron 1999, 55, 3987. (c) Adam, I.; Orain, D.; Meier, P. Synlett
*E-mail: michel.gravel@usask.ca.
Notes
The authors declare no competing financial interest.
2004, 11, 2031. (d) Bodtke, A.; Pfeiffer, W.-D.; Gorls, H.; Dollinger,
̈
H.; Langer, P. Tetrahedron 2007, 63, 11287.
ACKNOWLEDGMENTS
■
(11) (a) Fitzi, K.; Pfister, R. 2-Alkyl- and 2-cycloalkyl-4,5-bis-phenyl-
imidazoles. 3,901,908, Aug 26, 1975. (b) Barreau, M.; Kryvenko, M.
Nouveaux derives de l′oxazole et leur preparation. EP 0 517 591 A1, Sep
́ ́
12, 1992. (c) Mercep, M.; Mesic, M.; Pesic, D.; Benko, I. 1-Oxa-3-azo-
dibenzoazulenes as inhibitors of tumour necrosis factor production and
intermediates for the production thereof. WO 03/084964 A1, Oct 16,
2003. (d) Yamamoto, H.; Ishida, J.; Tanabe, D.; Satoh, S.; Sawada, Y.;
Ohkawa, T.; Imamura, K.; Nakamura, K. Oxazole derivatives as
inhibitors of cyclooxygenase. WO 2004/065374, Aug 5, 2004.
(e) Cananzi, S.; Merlini, L.; Artali, R.; Beretta, G. L.; Zaffaroni, N.;
Dallavalle, S. Bioorg. Med. Chem. 2011, 19, 4971.
(12) The use of acylsilanes as aldehyde surrogates was shown to
avoid the competing benzoin reaction: (a) Mattson, A. E.; Scheidt, K.
A. Org. Lett. 2004, 6, 4363. (b) Mattson, A. E.; Bharadwaj, A. R.; Zuhl,
A. M.; Scheidt, K. A. J. Org. Chem. 2006, 71, 5715.
(13) (a) Ramage, R.; Atrash, B.; Hopton, D.; Parrott, M. J. J. Chem.
Soc., Perkin. Trans. 1 1985, 1217. (b) Jennings, W. B.; Lovely, C. J.
We thank Dr. Yun Yan Hou (TC Scientific Inc.) for helpful
discussions. We thank the Natural Sciences and Engineering
Research Council (NSERC) of Canada, the Canada Founda-
tion for Innovation (CFI), and the University of Saskatchewan
for financial support.
REFERENCES
■
(1) (a) Dalko, P. I. Comprehensive Enantioselective Organocatalysis:
Catalysts, Reactions, and Applications; Wiley-VCH: Weinheim, 2013.
(b) List, B. Topics in Current Chemistry, Asymmetric Organocatalysis;
Springer-Verlag: Berlin, Heidelberg, 2010. (c) Waser, M. Asymmetric
Organocatalysis in Natural Product Synthesis; Springer-Verlag: Wien,
2012. (d) Berkessel, A.; Groger, H. Asymmetric Organocatalysis: From
̈
Biomimetic Concepts to Applications in Asymmetric Synthesis; Wiley-
VCH: Weinheim, 2005.
(2) For general reviews on NHC organocatalysis, see: (a) Enders, D.;
Niemeier, O.; Henseler, A. Chem. Rev. 2007, 107, 5606. (b) Moore, J.
L.; Rovis, T. Top. Curr. Chem. 2009, 291, 77. (c) Phillips, E. M.; Chan,
A.; Scheidt, K. A. Aldrichimica Acta 2009, 42, 55. (d) Chiang, P.-C.;
Bode, J. W. In N-Heterocyclic Carbenes: From Laboratory Curiosities to
Tetrahedron 1991, 29, 5561. (c) Desrosiers, J. N.; Co
̂ ́
te, A.; Boezio, A.
A.; Charette, A. B. Org. Synth. 2006, 83, 5. (d) Cote, A.; Boezio, A. A.;
̂
́
Charette, A. B. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 5405.
(e) Morales, S.; Guijarro, F. G.; Ruano, J. L. G.; Cid, M. B. J. Am.
Chem. Soc. 2014, 136, 1082. (f) Appel, R.; Mayr, H. J. Am. Chem. Soc.
2011, 133, 8240.
́
Efficient Synthetic Tools; Díez-Gonzalez, S., Ed.; Royal Society of
Chemistry: Cambridge, U.K., 2010; p 339. (e) Campbell, C. D.; Ling,
K. B.; Smith, A. D. In N-Heterocyclic Carbenes in Transition Metal
Catalysis and Organocatalysis, Vol. 32; Cazin, C. S. J., Ed.; Springer:
Dortrecht, 2011; p 263. (f) Grossmann, A.; Enders, D. Angew. Chem.,
Int. Ed. 2012, 51, 314. (g) Bugaut, X.; Glorius, F. Chem. Soc. Rev. 2012,
41, 3511. (h) Thai, K.; Sanchez-Larios, E.; Gravel, M. In Comprehensive
Enantioselective Organocatalysis; Dalko, P., Ed.; Wiley-VCH: Weinheim,
2013.
(14) Greene, T. W.; Wuts, P. G. M. Protecting Groups in Organic
Synthesis, 3rd ed.; John Wiley & Sons, Inc.: New York, 1999.
(3) Wilde, M. M. D.; Gravel, M. Angew. Chem., Int. Ed. 2013, 52,
12651.
(4) Holschumacher, D.; Hrib, C. G.; Jones, P. G.; Tamm, M. Chem.
Commun. 2007, 3661.
(5) Melaimi, M.; Soleilhavoup, M.; Bertrand, G. Angew. Chem., Int.
Ed. 2010, 49, 8810.
(6) Aza-benzoin reactions of aldimines or aldimine equivalents:
(a) Castells, J.; Lopez-Calahorra, F.; Bassedas, M.; Urrios, P. Synthesis
́
1988, 314. (b) Murry, J. A.; Frantz, D. E.; Soheili, A.; Tillyer, R.;
Grabowski, E. J. J.; Reider, P. J. J. Am. Chem. Soc. 2001, 123, 9696.
(c) Frantz, D. E.; Morency, L.; Soheili, A.; Murry, J. A.; Grabowski, E.
J. J.; Tillyer, R. D. Org. Lett. 2004, 6, 843. (d) Mennen, S. M.; Gipson,
J. D.; Kim, Y. R.; Miller, S. J. J. Am. Chem. Soc. 2005, 127, 1654.
(e) He, M.; Bode, J. W. Org. Lett. 2005, 7, 3131. (f) Li, G.-Q.; Dai, L.-
X.; You, S.-L. Chem. Commun. 2007, 852. (g) He, M.; Bode, J. W. J.
5311
dx.doi.org/10.1021/ol5024807 | Org. Lett. 2014, 16, 5308−5311