N-tert-butyl triazolylidenes: Catalysts for the enantioselective (3+2) annulation of α,β-unsaturated acyl azoliums

Article abstract of DOI:10.1002/anie.201304081

But(yl) not futile: A range of N-tert-butyl-substituted triazolylidene N-heterocyclic carbenes have been prepared. Of these, the morpholinone-derived catalyst (1) proved best suited to the enantioselective synthesis of cyclopentanes from donor-acceptor cy

Full text of DOI:10.1002/anie.201304081

Angewandte
Chemie
DOI: 10.1002/anie.201304081
N-Heterocyclic Carbene
N-tert-Butyl Triazolylidenes: Catalysts for the Enantioselective (3+2)
Annulation of a,b-Unsaturated Acyl Azoliums**
Lisa Candish, Craig M. Forsyth, and David W. Lupton*
In 1996 the triazolylidene-catalyzed benzoin reaction was
reported by Enders et al.[1] Following this seminal report
triazolylidenes (i.e. 1) have emerged as the N-heterocylic
carbene (NHC) organocatalyst of choice in many contexts.[2]
In 2008 we began studies on NHC catalysis using substrates at
the ester oxidation state.[3] Pivotal to the success of these
studies was the use of electron-rich imidazolylidene NHCs
(i.e., 3 or 4), with triazolylidenes generally proving unsatis-
Scheme 1. a) (CH3)3OBF4, CH2Cl2, 12 h, then tBuNHNH2, 16 h;
b) HC(OEt)3, PhCl, 1208C, 12 h. Yield 37% (two steps).
factory. Partly as a consequence of this limitation, enantio-
selective variants of these reactions have been difficult to
realize.[3a] To address this, access to chiral triazolylidene
NHCs, with imidazolylidene-like reactivity was deemed
valuable.
(2), which have enabled the development of an enantio-
selective all carbon (3+2) annulation.
Studies commenced with the synthesis of the N-tert-butyl
variant of the Rovis triazolium precatalyst, namely A1
(Scheme 1). This triazolium was prepared using methods
originally reported by Knight and Leeper, and later modified
by others.[7] Although A1 was accessible in this way, other N-
tert-butyl triazolium tetrafluoroborates prepared for this
study were generally synthesized as the HCl salt,[4d,7] then
converted into the tetrafluoroborate (see the Supporting
Information).
With A1 in hand its utility was examined in the
enantioselective (3+2) annulation between donor-acceptor
cyclopropane (Æ)-7a and a,b-unsaturated acyl fluoride 8a
(Table 1). This reaction proceeds with high yield and diaste-
reoselectivity, although it is yet to be achieved with enantio-
selectivity.[3e] Using previously reported reaction conditions,
and the known N-aryl triazolylidenes A2, A3, or A4 the
reaction failed (Table 1, entries 1–3), however A1 afforded
the desired cyclopentyl b-lactone 9aa in 8% yield (Table 1,
entries 4). Unfortunately, because of the low yield and purity
of the product, it was not possible to determine the
enantioselectivity.
To improve conversion, less sterically hindered pyrrolidi-
none-derived triazolium catalysts B1–B4 were prepared.[8]
With both B1 and B2, cyclopentane 9aa formed with
moderate enantioselectivity (50% and 63% ee) and yield
(Table 1, entries 5 and 6). When using the more hindered
bis(tert-butyl) catalyst B3, or the tert-butyl variant of the
pyroglutamate catalyst[9] (B4), the reaction failed (Table 1,
entries 7 and 8). Attention was next directed to the morpho-
linone-derived triazolium[10] C1. It was hoped that this catalyst
would display the advantageous features of B2, but enhance
the enantioselectivity through conformational restriction of
the benzyl group. This proved to be the case with cyclopentyl
b-lactone 9aa formed as a single diastereoisomer in 88% ee
and 75% yield (Table 1, entry 9). Finally, reducing the catalyst
loading to 10 mol% of C1 had little effect on the yield, but
The electronic nature of the triazolylidene can be
modulated, to enhance reactivity and/or selectivity, through
selection of the N substituents.[4–6] In this context N-aryl[4] and
N-benzyl[5] groups have been studied. Surprisingly, tertiary or
secondary alkyl N-substituted triazolylidene NHCs have yet
to be examined in enantioselective catalysis, even though
these groups significantly increase the electron density of
NHCs (for example see: 3 versus 4).[6] Herein we introduce
electron-rich homochiral N-tert-butyl triazolylidene catalysts
[*] L. Candish, Dr. C. M. Forsyth, Dr. D. W. Lupton
School of Chemistry, Monash University
Clayton 3800, Victoria (Australia)
E-mail: david.lupton@monash.edu
[**] The authors thank the Australian Research Council (Discovery and
Future Fellowship programs) for financial support. We acknowledge
suggestions and donation of catalyst from Professors Tomislav
Rovis (A4), Michel Gravel (cat. not shown), and Jonathan Morris
(cat. not shown). We acknowledge help with DFT calculations from
Dr. Christopher Thompson.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201304081.
Angew. Chem. Int. Ed. 2013, 52, 9149 –9152
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9149

.
Angewandte
Communications
Table 1: Selected catalyst optimization.
electron rich than might be expected, a situation arising as
a consequence of a twisted arrangement between the
triazolylidene and the N-aryl group. In terms of catalytic
activity the least electron-rich NHCs were inactive (Table 1,
entries 15 and 16) or moderately active (Table 1, entries 13
and 14), whereas the more-electron-rich N-alkyl and aryl
catalysts (C1–C3) gave the expected cyclopentane 9aa in
reasonable yields (Table 1, entries 9–12). When the yield is
modest the product is accompanied by ring-opened cyclo-
propane,[12] presumably its formation is kinetically viable as
the desired reaction slows. While catalysts C2–C5 achieved
similar enantioinduction (67–79% ee), the most sterically
encumbered[13] and electron-rich catalyst C1 gave the highest
enantioselectivity (Table 1, entry 10). Although a qualitative
connection between NHC electronics and reaction outcome
has been established, quantitative studies are ongoing.
The scope of the reaction was initially examined with the
annulation of a series of cinnamoyl fluorides 8a–i with
cyclopropane (Æ)-7a (Table 2). While electron-rich ortho- or
para-substituted cinnamoyls routinely gave high enantio-
selectivity (9ab, ac, ad), electron-poor substrates resulted in
decreased enantioselectivity (9ae, af). The reaction was
sensitive to ortho-disubstitution, with the mesityl cinnamoyl
fluoride 8g, providing the expected product 9ag along with
the diastereomeric 9ag’, both with excellent enantioselectiv-
ity.[14] The electronic nature of the aryl ester has an impact on
enantioselectivity, thus the 2,6-dimethoxy phenol cyclopro-
pane (Æ)-7b derived products are more enantioenriched than
(Æ)-7a-derived products, with 9ba forming in 93% ee (9aa;
90% ee), 9bb forming in 97% ee (9ab; 89% ee), and 9be with
79% ee (9ae; 74% ee). In the case of 9ba single-crystal X-ray
analysis was used to determine the absolute configuration of
the cyclopentanes.[15] The reaction tolerated b-alkyl a,b-
unsaturated acyl fluorides, however both yield and enantio-
selectivity were eroded (9ah; Table 2). In contrast a,b,g,d-
unsaturated acyl fluorides reacted smoothly and with high
enantioselectivity to provide styrenyl cyclopentane 9ai with
87% ee. Finally variation in the cyclopropane was inves-
tigated, providing dimethyl cyclopentane 9cb (83% ee), bis-
cyclopentane 9db (94% ee), and cycloheptane-containing
9ea and 9eb (98% ee and 96% ee) smoothly.
Derivatization of the b-lactone-containing products was
investigated under a number of reaction conditions. Partial
reduction of 9aa using LiAlH4 provided diol 10 (Scheme 2).
Ring opening with alcohols proved to be more challenging,
and led to mixtures of products, except when the reaction was
performed in the presence of NaBH4. Finally ring opening
with benzyl amine gave amide ester 12 in 93% yield. In all
cases the enantiopurity was maintained.
Mechanistically, the reaction likely commences with
formation of the a,b-unsaturated acyl azolium I and ester
enolate II from acyl fluoride 8 and cyclopropane (Æ)-7
(Scheme 3).[16] The union of I and II then occurs either via 1,2
adduct III with subsequent ester enolate Claisen rearrange-
ment to afford IV, a process resembling pathways proposed by
Bode and co-workers.[17] Or alternatively a diastereoselective
and enantioselective Michael addition allows the direct
conversion of I and II into IV. The latter pathway is preferred
by Studer, Mayr and co-workers,[18] while computational
Entry
Catalyst
13C NMR
Yield [%][b]
ee[c]
d [ppm][a]
1
2
3
4
A2, R=C6F5
A3, R=Mes
A4, R=Ph
A1, R=tBu
–
–
–
–
no reaction
no reaction
no reaction
8
–
–
–
n.d.
5
6
7
8
B1, R=iPr
B2, R=Bn
B3, R=tBu
B4, R=C(OH)Ph2
–
–
–
–
51
40
50
63
–
no reaction
no reaction
–
9
C1, R=tBu
C1, R=tBu
C2, R=iPr
C3, R=4-CH3OC6H4
C4, R=Ph
C5, R=Mes
C6, R=2,6-(CH3O)2C6H3
C7, R=C6F5
206.9
206.9
207.0
209.3
211.0
212.8
214.0
217.4
75
72
23
53
26
18
88
90
69
79
77
67
–
10[d]
11
12
13
14
15
16
no reaction
no reaction
–
[a] 13C NMR shift of the carbenic carbon atom (C5) in C6D6; see the
Supporting Information. [b] Yield of the product isolated after flash
column chromatography. [c] Determined by HPLC using Daicel AD-H
stationary phases. [d] Reaction performed with both 10 mol% C1 and
KHMDS. n.d.=not determined. M.S.=molecular sieves,
KHMDS=potassium hexamethyldisilazide, THF=tetrahydrofuran,
TMS=trimethylsilyl.
cyclopentane 9aa was isolated with 90% ee (Table 1,
entry 10).
To examine in more detail the impact of catalyst
electronics on the reaction, seven triazoliums (C1–C7) were
prepared using this chiral scaffold by varying the N substitu-
ent. In addition to assessing their utility in the synthesis of
cyclopentane 9aa the 13C NMR spectrum of the free carbene
was determined to provide a measure of electron density at
the carbenic carbon atom.[11] 13C NMR analyses of the NHCs
derived from C1–C7 indicate that the tert-butyl NHC
(Table 1, entry 9) is most electron rich, followed by iPr
(Table 1, entry 11) and then the aryl NHCs ranging from
p-CH3OC6H4 through to C6F5 (Table 1, entries 12–16). The
ortho-disubstituted NHCs (Table 1, entries 14 and 15) are less
9150
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 9149 –9152

Angewandte
Chemie
Table 2: Scope of the enantioselective cyclopentane 9 synthesis.[a,b]
Scheme 2. Derivitization of cyclopentanes 9aa and 9ab.
[a] Yield of product isolated after flash column chromatography. [b] The
ee value was determined by HPLC analysis using AD-H or OD-H
stationary phases. [c] Enantiomeric excess determined by analysis of
diester 11. [d] For X-ray data[15]
.
studies by Schoenebeck and co-workers highlight challenges
discriminating between these mechanisms.[19] After formation
of IV, rotation and aldol cyclization provides the cyclopentyl
alkoxide V, which undergoes b-lactonization to provide the
cyclopentane 9 and regenerate the catalyst. Modeling of
azolium I indicates that the carbonyl group is rotated out of
conjugation with the azolium, with two low-energy rotamers,
anti-I and syn-I, identified with very similar energies.[20] In
both, the Re face appears more open than the Si face, thus
allowing enolate approach which ultimately provides the
stereochemistry observed in cyclopentane 9aa.
Scheme 3. Plausible mechanism and computational modeling.
NHC catalysis has grown into a vibrant and diverse aspect
of organocatalysis. This growth has been achieved using
a small family of chiral catalysts which display exquisite
flexibility. In this report N-tert-butyl triazolylidene catalysts,
which combine the chiral environments of known triazolyli-
denes, with the electron-rich behavior of imidazolylidene
NHCs, are introduced. Through 13C NMR analysis it is clear
that the electronics of the catalyst play a significant role in
their success. We believe that these electron-rich triazolyli-
Angew. Chem. Int. Ed. 2013, 52, 9149 –9152
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
9151

.
Angewandte
Communications
[7] This approach is based on the reports of a) R. L. Knight, F. J.
Leeper, J. Chem. Soc. Perkin Trans. 1 1998, 1891; b) H. U. Vora,
S. P. Lathrop, N. T. Reynold, M. S. Kerr, J. Read del Alaniz, T.
Rovis, Org. Synth. 2010, 87, 350; c) J. R. Struble, J. W. Bode, Org.
Synth. 2010, 87, 362.
[8] For the N-C6F5 variant of B2, see: Q. Liu, S. Perreault, T. Rovis,
J. Am. Chem. Soc. 2008, 130, 14066.
[9] For early reports on pyroglutamate-derived NHCs, see: a) D.
Enders, J. Han, Tetrahedron: Asymmetry 2008, 19, 1367; b) H.
Lv, Y.-R. Zhang, X.-L. Huang, S. Ye, Adv. Synth. Catal. 2008,
350, 2715.
[10] For the introduction of this scaffold and the triazoliums C4 and
C5, see: a) B. E. Maki, A. Chan, E. M. Phillips, K. A. Scheidt,
Org. Lett. 2007, 9, 371; for C6, see: [4d]; for C7, see: b) E.
Sꢁnchez-Larios, K. Thai, F. Bilodeau, M. Gravel, Org. Lett. 2011,
13, 4942.
[11] For a discussion of the tensors defining the chemical shifts of the
carbenic carbon atom in NHCs, see: a) A. J. Arduengo III, Acc.
Chem. Res. 1999, 32, 913; for a review on NHC 13C NMR studies,
see: b) D. Tapu, D. A. Dixon, C. Roe, Chem. Rev. 2009, 109,
3385.
dene NHCs will be useful for a range of NHC-catalyzed
reactions which are poorly suited to conventional triazolyli-
dene catalysts.
Received: May 13, 2013
Revised: June 12, 2013
Published online: July 14, 2013
Keywords: cyclization · enantioselectivity ·
.
N-heterocyclic carbene · organocatalysis · synthetic methods
[1] a) D. Enders, K. Breuer, J. H. Teles, Helv. Chim. Acta 1996, 79,
1217; for the properties and stoichiometric reactions of this
carbene see: b) D. Enders, K. Breuer, G. Raabe, J. Runsink, J. H.
Teles, J.-P. Melder, K. Ebel, S. Brode, Angew. Chem. 1995, 107,
1119; Angew. Chem. Int. Ed. Engl. 1995, 34, 1021.
[2] For a comprehensive review on NHC organocatalysis, see: a) D.
Enders, O. Niemeier, A. Henseler, Chem. Rev. 2007, 107, 5606;
for homoenolate chemistry, see: b) V. Nair, R. S. Menon, A. T.
Biju, C. R. Sinu, R. R. Paul, A. Jose, V. Sreekumar, Chem. Soc.
Rev. 2011, 40, 5336; for acyl azolium enolates, see: c) H. U. Vora,
P. Wheeler, T. Rovis, Adv. Synth. Catal. 2012, 354, 1617; d) J.
Douglas, G. Churchill, A. D. Smith, Synthesis 2012, 2295; for
cascade catalysis, see: e) A. Grossmann, D. Enders, Angew.
Chem. 2012, 124, 320; Angew. Chem. Int. Ed. 2012, 51, 314; for
acyl anion chemistry, see: f) X. Bugaut, F. Glorius, Chem. Soc.
Rev. 2012, 41, 3511; for applications in total synthesis, see: g) J.
Izquierdo, G. E. Hutson, D. T. Cohen, K. A. Scheidt, Angew.
Chem. 2012, 124, 11854; Angew. Chem. Int. Ed. 2012, 51, 11686;
for acyl-anion-free catalysis, see: h) S. J. Ryan, L. Candish, D. W.
Lupton, Chem. Soc. Rev. 2013, 42, 4906.
[12] When a modest yield of 9 is observed, the ring-opened cyclo-
propane 13 accounts for the remaining substrate. When the
reaction fails unreacted starting material is re-isolated.
[13] For quantification of the sterics of ItBu (3), see: R. Dorta, E. D.
Stevens, N. M. Scott, C. Costabile, L. Cavallo, C. D. Hoff, S. P.
Nolan, J. Am. Chem. Soc. 2005, 127, 2485.
[3] For an isolated example of enantioselective triazolylidene
catalysis (50% ee) with ester oxidation state substrates, see:
a) S. J. Ryan, L. Candish, D. W. Lupton, J. Am. Chem. Soc. 2009,
131, 14176; for other reports from our group, see: b) S. J. Ryan,
L. Candish, D. W. Lupton, J. Am. Chem. Soc. 2011, 133, 4694;
c) L. Candish, D. W. Lupton, Chem. Sci. 2012, 3, 380; d) S. J.
Ryan, A. S. Stasch, M. N. Paddon-Row, D. W. Lupton, J. Org.
Chem. 2012, 77, 1113; e) L. Candish, D. W. Lupton, J. Am. Chem.
Soc. 2013, 135, 58.
[4] For selected examples of the N-aryl group effecting reaction
outcome, see: a) M. S. Kerr, J. Read de Alaniz, T. Rovis, J. Am.
Chem. Soc. 2002, 124, 10298; b) J. Read de Alaniz, T. Rovis, J.
Am. Chem. Soc. 2005, 127, 6284; c) H. Lv, X.-Y. Chen, L.-H. Sun,
S. Ye, J. Org. Chem. 2010, 75, 6973; d) F. Liu, X. Bugaut, M.
Schedler, R. Frçhlich, F. Glorius, Angew. Chem. 2011, 123,
12834; Angew. Chem. Int. Ed. 2011, 50, 12626; e) P. Zheng, C. A.
Gondo, J. W. Bode, Chem. Asian J. 2011, 6, 614; f) J. Mahattha-
nanchai, J. W. Bode, Chem. Sci. 2012, 3, 192; g) D. A. DiRocco,
T. Rovis, J. Am. Chem. Soc. 2012, 134, 8094; h) S. C. Allen, J.
Mahatthananchai, J. W. Bode, M. C. Kozlowski, J. Am. Chem.
Soc. 2012, 134, 12098; i) C. J. Collett, R. S. Massey, O. R.
Maguire, A. S. Batsanov, A. C. OꢀDonoghue, A. D. Smith,
Chem. Sci. 2013, 4, 1514; j) M. Schedler, D.-S. Wang, F. Glorius,
Angew. Chem. 2013, 125, 2645; Angew. Chem. Int. Ed. 2013, 52,
2585.
[14] Relative configuration assigned by nOe analysis of 9ag and 9ag’
and comparison of coupling constants.
[15] Single-crystal X-ray analysis was performed using CuKa radia-
tion for assignment of absolute configuration. The crystal
analyzed was subjected to HPLC and found to correlate to the
major enantiomer present in the bulk. CCDC 937492 contains
the supplementary crystallographic data for this paper. These
data can be obtained free of charge from The Cambridge
Crystallographic Data Centre via www.ccdc.cam.ac.uk/data_
request/cif.
[16] For a review covering retro-aldol reactions of donor-acceptor
cyclopropanes, see: a) H.-U. Reissig, R. Zimmer, Chem. Rev.
2003, 103, 1151; for ring opening using fluoride see: b) E.
Kunkel, I. Reichelt, H.-U. Reissig, Liebigs Ann. Chem. 1984, 802.
[17] a) J. Kaeobamrung, J. Mahatthananchai, P. Zheng, J. W. Bode, J.
Am. Chem. Soc. 2010, 132, 8810; b) J. Mahatthananchai, P.
Zheng, J. W. Bode, Angew. Chem. 2011, 123, 1711; Angew.
Chem. Int. Ed. 2011, 50, 1673; c) J. Mahatthananchai, J.
Kaeobamrung, J. W. Bode, ACS Catal. 2012, 2, 494; d) L.
Candish, D. W. Lupton, Org. Biomol. Chem. 2011, 9, 8182.
[18] For reaction discovery and preliminary mechanistic studies see:
a) S. De Sarkar, A. Studer, Angew. Chem. 2010, 122, 9452;
Angew. Chem. Int. Ed. 2010, 49, 9266; for mechanistic studies
b) R. C. Samanta, B. Maji, S. De Sarkar, K. Bergander, R.
Froehlich, C. Mueck-Lichtenfeld, H. Mayr, A. Studer, Angew.
Chem. 2012, 124, 5325; Angew. Chem. Int. Ed. 2012, 51, 5234.
[19] E. Lyngvi, J. W. Bode, F. Schoenebeck, Chem. Sci. 2012, 3, 2346.
[20] Low-energy conformers were identified using approaches com-
municated by Shoenebeck and co-workers[19] using the
wB97XD/6-31G(d) level of theory and the polarization contin-
uum model for THF. The isomers anti-I and syn-I were found to
be separated by 0.2 kJmolÀ1 with the former being more stable.
In depth computational analysis of this reaction is ongoing.
[5] For selected examples of the N-benzyl group effecting reaction
outcome, see: a) D. Enders, J. Han, A. Henseler, Chem.
Commun. 2008, 3989; b) P.-L. Shao, X.-Y. Chen, S. Ye, Angew.
Chem. 2010, 122, 8590; Angew. Chem. Int. Ed. 2010, 49, 8412;
c) L.-H. Sun, Z.-Q. Liang, W.-Q. Jia, S. Ye, Angew. Chem. 2013,
125, 2803; Angew. Chem. Int. Ed. 2013, 52, 5803.
[6] For a useful review and entry to the literature on the physical
properties of NHCs, see: a) T. Drçge, F. Glorius, Angew. Chem.
2010, 122, 7094; Angew. Chem. Int. Ed. 2010, 49, 6940.
9152
www.angewandte.org
ꢀ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2013, 52, 9149 –9152