Chemoselective N-heterocyclic carbene-catalyzed cross-benzoin reactions: Importance of the fused ring in triazolium salts

Article abstract of DOI:10.1021/ja501772m

Morpholinone- and piperidinone-derived triazolium salts are shown to catalyze highly chemoselective cross-benzoin reactions between aliphatic and aromatic aldehydes. The reaction scope includes ortho-, meta-, and para-substituted benzaldehyde derivatives with a range of electron-donating and -withdrawing groups as well as branched and unbranched aliphatic aldehydes. Catalytic loadings as low as 5 mol % give excellent yields in these reactions (up to 99%).

Full text of DOI:10.1021/ja501772m

Communication
pubs.acs.org/JACS
Chemoselective N‑Heterocyclic Carbene-Catalyzed Cross-Benzoin
Reactions: Importance of the Fused Ring in Triazolium Salts
Steven M. Langdon, Myron M. D. Wilde, Karen Thai, and Michel Gravel*
Department of Chemistry, University of Saskatchewan, 110 Science Place, Saskatoon, Saskatchewan S7N 5C9, Canada
S
* Supporting Information
through O-silyl thiazolium carbinols, though a stoichiometric
amount of the preformed acyl anion equivalent is required.10
ABSTRACT: Morpholinone- and piperidinone-derived
triazolium salts are shown to catalyze highly chemo-
selective cross-benzoin reactions between aliphatic and
Others have effected cross-benzoin reactions using ketones11 or
α-keto esters12 as acceptors, thereby largely avoiding the
aromatic aldehydes. The reaction scope includes ortho-,
meta-, and para-substituted benzaldehyde derivatives with
a range of electron-donating and -withdrawing groups as
well as branched and unbranched aliphatic aldehydes.
Catalytic loadings as low as 5 mol % give excellent yields in
these reactions (up to 99%).
chemoselectivity issues. Analogous ThDP-dependent benzalde-
hyde lyase work has yielded highly chemoselective cross-
couplings between o-substituted and non-o-substituted elec-
tron-deficient benzaldehyde derivatives13 and moderate success
between α-branched aliphatic aldehydes and benzaldehyde.14
Thus, there currently exists no general method to perform
chemoselective cross-benzoin reactions between aldehydes
despite extensive efforts over the last seven decades. Here we
report the first highly chemoselective cross-benzoin reactions
between aromatic and aliphatic aldehydes, utilizing catalyst
rather than substrate control.
rom its humble beginnings in 1943,1 the N-heterocyclic
F
carbene (NHC)-catalyzed benzoin reaction has seen a
variety of approaches to expand its applications.2 Initial successes
focused on the homocoupling of aldehydes and impressive
enantioselective approaches are well-known.3 The addition of a
second aldehyde increases the number of possible products to
four (Scheme 1), and formation of one to the exclusion of others
The primary issue with chemoselectivity in cross-benzoin
reactions is one of reactivity.15 Namely, in a typical reaction the
acyl anion equivalent (Breslow intermediate)16 is formed faster
with the most electrophilic aldehyde. Other factors remaining
constant, it should thus prefer to attack another equivalent of the
same aldehyde, leading to homobenzoin products. If the
electrophilicity of the two aldehydes is similar there is little
impetus for selectivity, and a statistical mixture is generally
obtained. The issue of chemoselectivity is compounded by the
commonly observed reversibility of the benzoin reaction.
Achieving chemoselectivity via kinetic control therefore depends
on the choice of substrates, catalyst, and reaction conditions.
While previous reports have suggested various structural
features of NHCs which aid in achieving chemoselectivity,17
many exciting results are still obtained by serendipity. During our
recent investigations of cross-benzoin reactions using α-
ketoesters, we observed that the use of triazolium catalysts
incorporating a fused morpholine ring resulted in fast reactions
when aliphatic aldehydes are used as substrates.12b In contrast,
aromatic aldehydes were found to be unreactive under the same
conditions. It was surmised that the fused morpholine ring was
perhaps leading to a more facile formation of a Breslow
intermediate with aliphatic aldehydes than with aromatic
aldehydes. With these considerations in mind, we compared
the outcome of benzoin reactions between benzaldehyde and
hydrocinnamaldehyde using a range of representative catalysts.
Table 1 shows the variation in yield and selectivity between
several azolium salts. Thiazolium catalyst 1 displays limited
selectivity, with 7 being a minor product. While catalyst 2 is more
reactive, morpholinone-derived salt 3 proves far superior in
Scheme 1. General Cross-Benzoin Reaction
becomes the main concern. Even with this challenge intra-
molecular couplings have been achieved with excellent chemo-
and enantioselectivity,4 notably in Mennen and Miller’s
macrocyclization of o-substituted dialdehydes.5 Despite these
advances a general approach to chemoselective intermolecular
cross-benzoin reactions remains elusive. Many methods expand
on the original work of Stetter6 which demonstrates that use of α-
branched aldehydes and/or o-substituted aromatic aldehydes
leads to selectivity.7 Particularly promising work by Connon and
Zeitler shows that o-bromobenzaldehyde can be coupled with a
slight excess of aliphatic aldehyde in up to 90% yield.7c Yang et al.
have shown that an excess of aliphatic aldehyde (10−15 equiv)
can be used to improve yield, though this may not always be
practical in synthetic applications.8 In parallel with these
developments, Kuhl and Glorius reported the hydroxymethyla-
tion of aldehydes, which achieves good levels of chemoselectivity
when electron-poor benzaldehyde derivatives are reacted with
paraformaldehyde.9 The Scheidt group has developed an
ingenious approach toward aliphatic−aliphatic cross-couplings
Received: February 19, 2014
© XXXX American Chemical Society
A
dx.doi.org/10.1021/ja501772m | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
either base or solvent choice (entries 1−5). While surprising, this
also means that the reaction is amenable to a variety of solvents
and bases. Equally as exciting were the observations that both
catalytic loading and reaction time could be halved while
maintaining the yield and selectivity (entries 6−7). A slight
excess of aliphatic aldehyde increased the yield, with the effect
being more pronounced in CH2Cl2 (entries 8−9). Several other
parameters were varied to further enhance the yield with no
effect.18
Table 1. Comparison of Catalysts in Cross-Benzoin Reactions
The reaction scope was explored using the optimized
conditions (Table 3). Both short and long chain aliphatic
a
a
entry
precatalyst
product ratio (6:7:8:9)
yield
Table 3. Scope of the Reaction
b
1
1
2
3
4
5
6:12:21:61
2:59:25:14
2:93:0:5
2
2
3
4
5
43
19
16
26
2:94:0:4
21:68:9:2
a
Yield of compound 7 determined by 1H NMR analysis using
b
dimethyl terephthalate as an internal standard. Performed using 1,8-
diazabicyclo[5.4.0]undec-7-ene (10 mol %) instead of (i-Pr)2NEt.
terms of chemoselectivity (entries 2−3). To explore whether the
oxygen atom plays a role in the observed selectivity, triazolium
salt 4 bearing a fused piperidine ring was also prepared and
screened. As this catalyst displays similar reactivity and
chemoselectivity to 3, it was concluded that selectivity is the
result of the ring size and not the presence of the oxygen atom.
Salt 5 was screened to see if this trend continued with increasing
ring size. A loss of chemoselectivity (entry 5) suggests that the
six-membered ring is optimal. With catalyst 4 readily available on
a gram scale (see Supporting Information (SI)), it was selected
for further study.
An optimization of the reaction was then performed (Table 2).
Elevated temperatures improved the rate of reaction without
affecting chemoselectivity; refluxing conditions were used to
minimize reaction times. Selectivity and yield proved invariant to
a
b
Isolated yields after chromatography. Corrected yield accounting for
c
small amounts of other benzoin products. 5 equiv of isobutyraldehyde
used.
Table 2. Reaction Optimization
aldehydes afford good to excellent yields (10, 11, and 12). β-
branched aliphatic aldehydes resulted in comparable yields (13),
albeit with extended reaction times. α-branched isobutyralde-
hyde showed decreased reactivity, and both a larger excess and
prolonged reaction time were needed to obtain good yields (14).
The aromatic ring was shown to be tolerant to both electron-
withdrawing and -donating groups at the para position (15 and
16, respectively), as well as functionalization at both meta (19)
and ortho (17 and 18) positions to furnish the products in good
to excellent yields. Notably, heteroaromatic aldehydes were
found to show varied selectivity for cross-benzoin products, but
competing reactions and issues with purification lead to their
exclusion as efficient substrates (details in SI). Compounds 6, 10,
and 14 also displayed unusual susceptibility to decomposition
and/or isomerization on silica gel; characterization was
performed on a mixture containing small amounts of other
benzoin products.
Chiral morpholinone-based triazolium precatalysts are well-
known and easily accessible from amino acids. Valine-derived
triazolium salt 21 was used in an enantioselective synthesis of 7
(Scheme 2). Chemoselectivity remained consistent with that of 4
(6:7:8:9 = 2:77:0:21 ; excess aliphatic aldehyde leads to an
increase in 9 after benzaldehyde has been consumed), though
high enantioselectivity was not achieved. Work with this family of
product ratio
a
solvent
×
base
6
7
8
9
yield (%)
1
2
3
4
5
6
7
8
9
CH2Cl2
THF
10
10
10
10
10
5
(i-Pr)2NEt
(i-Pr)2NEt
(i-Pr)2NEt
6
6
7
3
5
4
3
1
1
82
88
89
86
88
88
89
86
84
8
3
0
8
2
2
2
2
5
4
3
66
69
63
63
66
69
68
75
82
toluene
THF
4
b
DBU
3
THF
NaOAc
5
THF
(i-Pr)2NEt
(i-Pr)2NEt
(i-Pr)2NEt
(i-Pr)2NEt
6
c
THF
5
6
cd
,
THF
5
11
10
cd
,
CH2Cl2
5
a
Yield of compound 7 determined by 1H NMR analysis using
b
dimethyl terephthalate as an internal standard. Performed using 0.05
c
d
equiv DBU. Reaction time decreased to 30 min. Performed using 1.5
equiv of hydrocinnamaldehyde. THF = tetrahydrofuran; DBU = 1,8-
diazabicyclo[5.4.0]undec-7-ene.
B
dx.doi.org/10.1021/ja501772m | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
under kinetic control, chemoselectivity is determined by the
relative energies of the rate-determining step in each of the
possible pathways. With each pathway’s rate-limiting step being
approximately second-order overall, we conclude that selectivity
is determined at or after the C−C bond formation step ([III−
IV] or [IV−I]).22
Scheme 2. Enantioselective Cross-Benzoin Reaction
In summary, a general protocol for the catalyst-controlled
chemoselective coupling of aromatic and aliphatic aldehydes was
developed. Catalysts displaying a fused morpholine (3) or
piperidine (4) ring show dramatically improved chemoselectivity
over commonly employed catalyst 2, displaying a fused
pyrrolidine ring. The reaction is tolerant to a range of solvents
and bases without detrimental effects to either yield or
chemoselectivity. Relatively low catalytic loadings are needed,
and only a slight excess of the aliphatic aldehyde increases the
yield. Most importantly, the reaction features a wide substrate
scope in both the aliphatic and aromatic aldehyde. Based on
crossover experiments and determination of the order of the
reaction, the reaction and product distribution are under kinetic
control. More precise details on the origins of this kinetic bias are
the object of current investigations, as are enantioselective
variants of this cross-benzoin reaction.
a
Determined by HPLC; assignment of absolute configuration based
on order of elution.19
NHCs toward an enantio- and chemo-selective cross-benzoin
reaction is an ongoing area of investigation in our laboratories.
Although the specific origin of chemoselectivity remains
uncertain, it is clear the size of the fused ring plays a critical role.
Scheme 3 below highlights the generally accepted mechanism of
Scheme 3. Mechanism of the Cross-Benzoin Reaction
ASSOCIATED CONTENT
* Supporting Information
Experimental details and characterization data. This material is
available free of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
michel.gravel@usask.ca
■
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank the Government of Saskatchewan for an Innovation
and Opportunity Scholarship (to S.M.L.), the Natural Sciences
and Engineering Research Council of Canada, the Canada
Foundation for Innovation, and the University of Saskatchewan
for financial support. We wish to also thank the reviewers for
helpful suggestions and Dr. Ian Burgess for useful discussions.
the benzoin reaction. First, attack of the carbene on an aldehyde
[I−II] results in the formation of a tetrahedral intermediate. A
proton transfer [II−III] yields the Breslow intermediate, which
acts as an acyl anion equivalent and attacks another aldehyde
[III−IV]. A proton-transfer step at this stage may be concerted
or stepwise. Collapse of the intermediate releases the benzoin
product and turns-over the catalyst [IV−I]. A pair of crossover
experiments were performed to determine the reversibility of the
formation of the homobenzoin products. Subjection of
compound 6 and hydrocinnamaldehyde, or 9 and benzaldehyde,
to reaction conditions results only in formation of 9 or 6,
respectively (see SI).20 These results suggest that both
homobenzoin products are formed irreversibly under reaction
conditions. This further implies that, under reaction conditions,
the observed chemoselectivity results from kinetic control.
Following the rates of consumption of benzaldehyde and
hydrocinnamaldehyde in their respective homobenzoin reactions
show that these reactions are higher than first order with respect
to the aldehyde.21 Additionally, the cross-benzoin reaction shows
at least first order dependence in each aldehyde (details in SI)
and is at least second order overall. Given that the reaction is
REFERENCES
■
(1) Ukai, T.; Dokawa, T.; Tsubokawa, S. J. Pharm. Soc. Jpn. 1943, 63,
296.
(2) For reviews on N-heterocyclic carbenes as organocatalysts:
(a) Enders, D.; Niemeier, O.; Henseler, A. Chem. Rev. 2007, 107,
5606. (b) Moore, J. L.; Rovis, T. Top. Curr. Chem. 2010, 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 Efficient Synthetic Tools; S. Díez-Gonzalez, Ed.;
Royal Society of Chemistry: Cambridge, UK, 2010; p 339. (e) Campbell,
C. D.; Ling, K. B.; Smith, A. D. In N-Heterocyclic Carbenes in Transition
Metal Catalysis and Organocatalysis; Cazin, C. S. J., Ed.; Springer,
Dortrecht, Germany, 2011; Vol 32, 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: Catalysts, Reactions, and
Applications; Dalko, P. I., Ed.; Wiley-VCH: Weinheim, Germany, 2013;
p 477.
(3) For a historical summary of enantioselective homocouplings, see
ref 2b. For selected recent examples: (a) Enders, D.; Kallfass, U. Angew.
Chem., Int. Ed. 2002, 41, 1743. (b) Ma, Y.; Wei, S.; Wu, J.; Yang, F.; Liu,
B.; Lan, J.; Yang, S.; You, J. Adv. Synth. Catal. 2008, 350, 2645.
C
dx.doi.org/10.1021/ja501772m | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Communication
(c) Enders, D.; Han, J. Tetrahedron: Asymmetry 2008, 19, 1367.
(d) Baragwanath, L.; Rose, C. A.; Zeitler, K.; Connon, S. J. J. Org. Chem.
2009, 74, 9214. (e) Soeta, T.; Tabatake, Y.; Inomata, K.; Ukaji, Y.
Tetrahedron 2012, 68, 894.
Harada, S.; Kang, B.; Oriez, R.; Yamaoka, Y.; Takasu, K.; Yamada, K. J.
Am. Chem. Soc. 2013, 135, 11485. See also ref 8.
(18) Selectivity and yield were not significantly affected by the
following: temperature and concentration variation, addition of
molecular sieves, addition of 1 equiv of H2O, and increased base
loading. Carbonate bases, protic solvents, and use of nitrogen gas in
place of argon each led to a decrease in yield.
(19) Chen, C.; Kao, J.; Salunke, S. B.; Lin, Y. Org. Lett. 2011, 13, 26.
(20) Attempts to subject a mixture of all benzoin compounds with
excess hydrocinnamaldehyde (as formed by 1) resulted in no
consumption of the aldehyde. Presumably some component of the
mixture inhibited catalytic activity. Attempts to purify the mixture using
flash column chromatography prior to use resulted in decomposition
and mass loss.
(4) (a) Hachisu, Y.; Bode, J. W.; Suzuki, K. J. Am. Chem. Soc. 2003, 125,
8432. (b) Enders, D.; Niemeier, O.; Balensiefer, T. Angew. Chem., Int. Ed.
Engl. 2006, 45, 1463. (c) Enders, D.; Niemeier, O.; Raabe, G. Synlett
2006, 2431. (d) Takikawa, H.; Hachisu, Y.; Bode, J. W.; Suzuki, K.
Angew. Chem., Int. Ed. Engl. 2006, 45, 3492. (e) Li, Y.; Feng, Z.; You, S.-L.
Chem. Commun. 2008, 2263. (f) Ema, T.; Oue, Y.; Akihara, K.; Miyazaki,
Y.; Sakai, T. Org. Lett. 2009, 11, 4866. (g) Takada, A.; Hashimoto, Y.;
Takikawa, H.; Hikita, K.; Suzuki, K. Angew. Chem., Int. Ed. 2011, 50,
2297. (h) For a review highlighting the use of NHC-catalyzed
intramolecular cross-benzoin reactions in total synthesis: Izquierdo, J.;
Hutson, G. E.; Cohen, D. T.; Scheidt, K. A. Angew. Chem., Int. Ed. 2012,
51, 11686.
(21) Breslow, R.; Kim, R. Tetrahedron Lett. 1994, 35, 699.
(22) A rate-limiting Breslow intermediate formation involving two
molecules of aldehyde cannot be rigorously excluded based on our data.
(5) Mennen, S. M.; Miller, S. J. J. Org. Chem. 2007, 72, 5260.
(6) Stetter, H.; Dambkes, G. Synthesis 1977, 6, 403.
̈
(7) For recent examples: (a) Piel, I.; Pawalczyk, M. D.; Hirano, K.;
Frolich, R.; Glorius, F. Eur. J. Org. Chem. 2011, 5475. (b) Rose, C. A.;
̈
Gundala, S.; Connon, S. J.; Zeitler, K. Synthesis 2011, 190. (c) O’Toole,
S. E.; Rose, C. A.; Gundala, S.; Zeitler, K.; Connon, S. J. J. Org. Chem.
2011, 76, 347.
(8) (a) Jin, M. Y.; Kim, S. M.; Han, H.; Ryu, D. H.; Yang, J. W. Org. Lett.
2011, 13, 880. (b) Jin, M. Y.; Kim, S. M.; Hui, M.; Ryu, D. H.; Song, C.
E.; Yang, J. W. Org. Biomol. Chem. 2014, 12, 1547.
(9) (a) Kuhl, N.; Glorius, F. Chem. Commun. 2010, 47, 573. See also
(b) Matsumoto, T.; Ohishi, M.; Inoue, S. J. Org. Chem. 1985, 50, 603.
(10) Mathies, A. K.; Mattson, A. E.; Scheidt, K. A. Synlett 2009, 377.
(11) (a) Enders, D.; Henseler, A. Adv. Synth. Catal. 2009, 351, 1749.
(b) Enders, D.; Grossmann, A.; Fronert, J.; Raabe, G. Chem. Commun.
2010, 46, 6282. See also refs 4a−g.
(12) (a) Rose, C. A.; Gundala, S.; Fagan, C.-L.; Franz, J. F.; Connon, S.
J.; Zeitler, K. Chem. Sci. 2012, 3, 735. (b) Thai, K.; Langdon, S. M.;
Bilodeau, F.; Gravel, M. Org. Lett. 2013, 15, 2214.
(13) Dunkelmann, P.; Kolter-Jung, D.; Nitsche, A.; Demir, A. S.;
̈
Siegert, P.; Lingen, B.; Baumann, M.; Pohl, M.; Muller, M. J. Am. Chem.
̈
Soc. 2002, 124, 12084.
(14) (a) Muller, C. R.; Per
Org. Biomol. Chem. 2013, 11, 2000. For the use of a large excess of
́
ez-San
́
chez, M.; Domínguez de María, P.
̈
̈
̧esȩ noglu, O.; Dunkelmann, P.;
̈
aliphatic aldehyde: (b) Demir, A. S.; S
Muller, M. Org. Lett. 2003, 5, 2047.
̈
(15) For an excellent explanation of this problem, see ref 7b.
(16) Breslow, R. J. Am. Chem. Soc. 1958, 80, 3719.
(17) For discussions on the significance of the bicyclic backbone:
(a) Knight, R. L.; Leeper, F. J. Tetrahedron Lett. 1997, 38, 3611.
(b) Dvorak, C.; Rawal, V. Tetrahedron Lett. 1998, 39, 2925. (c) Knight,
R. L.; Leeper, F. J. J. Chem. Soc., Perkin Trans. 1 1998, 1891. See also ref
3a. For discussions on the significance of the N-aryl substituent:
(d) Rovis, T. Chem. Lett. 2008, 37, 2. (e) Enders, D.; Han, J.; Henseler,
A. Chem. Commun. 2008, 3989. (f) Read de Alaniz, J.; Rovis, T. Synlett
2009, 1189. (g) Mahatthananchai, J.; Bode, J. W. Chem. Sci. 2012, 3, 192.
(h) Schedler, M.; Frohlich, R.; Daniliuc, C.-G.; Glorius, F. Eur. J. Org.
̈
Chem. 2012, 4164. (i) Cohen, D. T.; Eichman, C. C.; Phillips, E. M.;
Zarefsky, E. R.; Scheidt, K. A. Angew. Chem., Int. Ed. 2012, 51, 7309.
(j) Massey, R. S.; Collett, C. J.; Lindsay, A. G.; Smith, A. D. J. Am. Chem.
Soc. 2012, 134, 20421. (k) Candish, L.; Forsyth, C. M.; Lupton, D. W.
Angew. Chem., Int. Ed. 2013, 52, 9149. (l) Collett, C. J.; Massey, R. S.;
Maguire, O. R.; Batsanov, A. S.; O’Donoghue, A. C.; Smith, A. D. Chem.
Sci. 4, 1514. (m) Zhao, X. D.; Glover, G. S.; Oberg, K. M.; Dalton, D. M.;
Rovis, T. Synlett 2013, 1229. (n) Mahatthananchai, J.; Bode, J. W. Acc.
Chem. Res. 2014, 47, 696. See also refs 3d and 7a. For an example of
backbone conformational control through substitution: (o) DiRocco, D.
A.; Oberg, K. M.; Dalton, D. M.; Rovis, T. J. Am. Chem. Soc. 2009, 131,
10872. For reports showing the importance of the counterion: (p) Ma,
Y.; Wei, S.; Wu, J.; Yang, F.; Liu, B.; Lan, J.; Yang, S.; You, J. Adv. Synth.
Catal. 2008, 350, 2645. (q) Kaeobamrung, J.; Mahatthananchai, J.;
Zheng, P.; Bode, J. W. J. Am. Chem. Soc. 2010, 132, 8810. (r) Kuwano, S.;
D
dx.doi.org/10.1021/ja501772m | J. Am. Chem. Soc. XXXX, XXX, XXX−XXX