Intermolecular scandium triflate-promoted nitrene-transfer [5 + 1] cycloadditions of vinylcyclopropanes

Article abstract of DOI:10.1039/c9ob01858a

Sc(OTf)₃-promoted [5 + 1] cycloaddition of vinylcyclopropanes with PhINTs is reported, enabling the regioselective preparation of a range of 1,2,3,6-tetrahydropyridine scaffolds under mild conditions. This represents the second example of a [5

Full text of DOI:10.1039/c9ob01858a

Organic &
Biomolecular Chemistry
View Article Online
View Journal | View Issue
COMMUNICATION
Intermolecular scandium triflate-promoted
nitrene-transfer [5 + 1] cycloadditions of
vinylcyclopropanes†
Cite this: Org. Biomol. Chem., 2019,
17, 9413
Received 23rd August 2019,
Accepted 9th October 2019
Julie E. Laudenschlager, Logan A. Combee ‡ and Michael K. Hilinski
*
DOI: 10.1039/c9ob01858a
rsc.li/obc
Sc(OTf)3-promoted [5 + 1] cycloaddition of vinylcyclopropanes nitrene-transfer cycloaddition leading to six-membered ring
with PhINTs is reported, enabling the regioselective preparation of products (Scheme 1a).9 Specifically, a Rh(II)-catalyzed formal
a range of 1,2,3,6-tetrahydropyridine scaffolds under mild con- [5 + 1] cycloaddition between benzyl (tosyloxy)carbamate and
ditions. This represents the second example of a [5 + 1] nitrene- vinylcyclopropanes (VCPs)10 provided tetrahydropyridine pro-
transfer cycloaddition and exhibits complementary substrate ducts with high regioselectivity. Furthermore, these products
scope to the antecedent work, expanding the range of could be elaborated to complex piperidines, the most common
N-heterocycles accessible via this strategy.
N-heterocycle in FDA-approved drugs.1 This method, though,
is limited to substrates bearing aryl substituents on both the
olefin and cyclopropane ring of the VCP at the specific posi-
tions indicated for generic substrate 1. Due to our continued
interest in this new approach for heterocycle synthesis, we
have endeavored to develop conditions that are compatible
with a more diverse array of VCPs. Herein, we describe a novel
scandium-catalyzed [5 + 1] reaction between VCPs and [N-(p-
toluenesulfonyl)imino]phenyliodinane (PhINTs) that circum-
vents the requirement for aryl substitution of the cyclopropane
ring. This method exhibits complete complementarity to the
previous Rh-catalyzed method, and provides initial mechanis-
tic insight that could help guide the development of [5 + 1]
Nitrogen-containing heterocycles are prevalent in FDA-
approved pharmaceuticals,1 and as a result, new methods and
strategies for their synthesis are needed to facilitate the more
rapid introduction of new therapeutics.2 In particular, control
of regio- and stereoselectivity of substitution patterns not
readily accessible through other means is a prime concern. To
this end, cycloaddition and annulation approaches are pre-
ferred strategies with the potential to exert the desired control.
The use of nitrogen as a one-atom component, in various
forms including amines, amides, carbamates, sulfonamides,
and oxidized forms thereof can be advantageous due to the
wide commercial availability and/or ease of preparation of pre-
cursors bearing these functional groups.3
One class of reactions having these attributes are those that
involve nitrenes or nitrene equivalents as reactive intermedi-
ates. It has been demonstrated, for example, that Cu(II) cataly-
sis,4 Ti(II)/Ti(IV) redox catalysis,5 Ru catalysis,6 vanadium(III)
catalysis,7 and stoichiometric BF3 can enable formal inter-
8
molecular cycloadditions between hydrocarbons or other com-
ponents and pre-oxidized aniline or sulfonamide nitrene pre-
cursors, but these reports focus entirely on the synthesis of
five-membered N-heterocycles. Expanding on this burgeoning
area of research, we recently reported the first examples of
Department of Chemistry, University of Virginia, Charlottesville,
Virginia 22904-4319, USA. E-mail: hilinski@virginia.edu
†Electronic supplementary information (ESI) available. See DOI: 10.1039/
c9ob01858a
‡Present address: Department of Chemistry, University of Michigan, Ann Arbor,
MI 48109.
Scheme 1 [5 + 1] nitrene-transfer cycloaddition.
Org. Biomol. Chem., 2019, 17, 9413–9417 | 9413
This journal is © The Royal Society of Chemistry 2019

View Article Online
Communication
Organic & Biomolecular Chemistry
cycloadditions into a general approach for the preparation of
heterocyclic products.
Having established optimized conditions for the [5 + 1]
cycloaddition of 1a, we turned our attention to investigating
We began with VCP 1a, lacking an aryl substituent on the the effect of variations of the VCP structure, focusing on the
vinylcyclopropane ring, and which does not provide the R1, R2, and R3 positions as indicated in Fig. 1. Evaluation of
product 2a using the previously developed Rh2(esp)2-catalyzed the Sc(OTf)3-promoted conditions revealed that yields were
conditions, nor any other Rh-catalyzed previously explored. In highly sensitive to the nature of the R1 substituent. In this
an initially promising result, reaction of 1a with PhINTs under position, aryl substitution was required for successful for-
typical conditions for Cu(II)-promoted olefin aziridination11 mation of the [5 + 1] product. Substitution at the para and
provided tetrahydropyridine 2a as the major product (Table 1, meta positions of the R1 phenyl group was tolerated. Moreover,
entry 1). One limitation of this approach was a requirement to a preference for substitution with strongly deactivating (e.g. 2c)
use the nitrenoid precursor as the limiting reagent to achieve over strongly activating groups was observed, with a p-OMe
reasonable yields (entry 2). This prompted additional investi- group promoting complete and non-selective conversion of the
gations, revealing that a variety of Brønsted and Lewis acids starting material to a complex mixture of products not includ-
(e.g. entries 5–7) could be employed to effect the same overall ing the desired tetrahydropyridine. The use of a trisubstituted
transformation. Ultimately, the use of Sc(OTf)3 proved to be olefin led to reduced yield in the formation of tricyclic product
the most amenable to optimization of other variables.12 The 2i. A key feature of these reactions is that only one olefin regio-
final conditions (2.5 equiv. PhINTs, 0.5 equiv. Sc(OTf)3,
CH2Cl2, 4 °C) employ slow addition of a solution of the VCP
over one hour to achieve the highest yields of the cycloaddition
product in 2 hours of total reaction time (entry 13). To rule out
trace amounts of Brønsted acid as the catalyst under these con-
ditions, we performed a control experiment using 1 mol% of
triflic acid and observed only trace amounts of product for-
mation (entry 17). This result is consistent with the additional
observation that the use metal triflate salts other than
Sc(OTf)3 largely failed to produce reasonable amounts of the
desired product (entries 14–16), pointing toward a critical role
for Sc(III).13
Table 1 Optimization of the [5 + 1] reactiona
Equiv.
Entry Equiv. VCP PhINTs Temp.
Acid (equiv.)
Yield (%)
1
2
3
4
5
6
3
1
1
1
1
1
1
1
1
1
1
1
rt
rt
rt
4 °C
4 °C
4 °C
Cu(OTf)2 (0.5)
Cu(OTf)2 (0.5)
Cu(OTf)2 (1.0)
TfOH (1.0)
TfOH (0.5)
HBF4·OEt2 (2.0) 38
47
20
13
41
20
7
8
9
1
1
1
1
1
1
1
1
1.5
1.5
2.5
2.5
2.5
2.5
1
−10 °C BF3·OEt2 (1.0)
30b
22
24
29
38
45
58
17b
0b
rt
Sc(OTf)3 (1.0)
Sc(OTf)3 (0.5)
Sc(OTf)3 (0.5)
Sc(OTf)3 (0.5)
Sc(OTf)3 (0.5)
Sc(OTf)3 (0.5)
Cu(OTf)2 (0.5)
Zn(OTf)2 (0.5)
Yb(OTf)3 (0.5)
TfOH (0.01)
rt
10
11
12
13
14
15
16
17
4 °C
4 °C
4 °C
4 °C
4 °C
4 °C
4 °C
4 °C
1
1c
1c
1c
1c
1c
1c
5b
<5b
Fig. 1 Scope and limitations of the [5 + 1] cycloaddition. Yields shown
are of isolated products. The substrate was added to a mixture of the
a All reactions were conducted on a 0.2 mmol scale in the limiting
reagent, with isolated yields reported except where noted. b NMR yield.
c A solution of VCP 1a in CH2Cl2 was added dropwise over 1 h.
other reagents over
1 h. Reactions were monitored by TLC and
quenched once full consumption of the substrate was observed.
9414 | Org. Biomol. Chem., 2019, 17, 9413–9417
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Communication
isomer of the [5 + 1] product is observed. Evaluating the effect
of alkyl substitution at R2, we observed that this regioisomeric
preference extends to a high selectivity for the formation of the
2,5-disubstituted rather than 3,5-disubstituted product (e.g.
2j), arising from cleavage of the more substituted cyclopropane
C–C bond. In this case, the observation of only one regio-
isomer is notable given that cleavage of the less substituted
bond is clearly feasible in other cases (vide supra). In a related
experiment, a VCP bearing a phenyl group at the R1 position
and a cyclohexyl group at the R2 position (not shown) gener-
ated the [5 + 1] product in very low yield (7%), suggesting that
the reaction is highly sensitive to steric effects at the R2 posi-
tion. Substitution at the R3 position is tolerated, allowing
access to a 2,3-disubstituted product (as in 2k), albeit in
reduced yield. Adding to the synthetic utility of this reaction,
methods have been reported that are capable of stereoselec-
tively elaborating both monosubstituted tetrahydropyridines
such as 2a (via enantioselective hydrogenation)14 and 2,5-di-
substituted tetrahydropyridines such as 2j (via diastereo-
selective epoxidation)9 to complex bio-relevant piperidines.
Intriguingly, this Sc(OTf)3-promoted nitrene-transfer cyclo-
addition exhibits substrate scope that completely differentiates
it from our previously reported Rh-catalyzed method.9 In the
case where R1 and R2 are both phenyl groups, the Rh2(esp)2-
catalyzed reaction successfully produces product 4a, whereas
use of Sc(OTf)3/PhINTs as reported here results in degradation
of the starting material to a complex mixture of products. This
remains true for all other bis-aryl substrates of this class that
were evaluated (Scheme 2a). Conversely, substrate 1a as well as
all other substrates evaluated as shown in Fig. 1 fail to provide
any products using the Rh-catalyzed conditions. To evaluate
the role of the nitrene precursor in these complementary
results, we also investigated the use of PhINTs/Rh2(esp)2
and TsONHCbz/Sc(OTf)3 reagent/catalyst combinations
(Scheme 2b). In each of the four cases evaluated, this resulted
in a failed reaction, providing no isolable amounts of the
desired products and, in three out of the four cases, leading to
complete non-selective consumption of the starting material.
This indicates that the differences in reactivity are not due
solely to either the choice of catalyst or nitrene precursor, but
are reliant on the specific combination of reagents discovered
via reaction optimization. Collectively, these results would
seem to be indicative of substantial mechanistic differences
between the two conditions. Therefore, we set out to further
understand the reasons behind these contrasting results.
The Rh-nitrenoid generated as the active oxidant in Rh(II)-
catalyzed nitrene transfer reactions is postulated to possess
sufficient Lewis acidity to promote ring-opening of N-tosyl azir-
Scheme 2 Control experiments to investigate the complementarity of
Sc-promoted and Rh-catalyzed conditions and the role of the nitrene
precursor.
idines to zwitterionic intermediates,15 which we exploited in nism of the Sc(OTf)3-promoted [5 + 1]. As demonstrated in
the development of the Rh-catalyzed [5 + 1]. That reaction is Table 1, Cu(OTf)2 is also capable of promoting the [5 + 1] reac-
proposed to proceed through a cyclopropylcarbinyl cation,16 tion of 1a. Like in the case of Sc(OTf)3, we observe the same
generated upon nitrene transfer to the olefin of the VCP, that complementarity of substrate scope when comparing the
subsequently undergoes rearrangement to the observed copper-promoted and rhodium-catalyzed conditions. In the
product.9 In contrast, there are no prior reports of scandium- case of Cu(OTf)2, evidence gathered by Evans suggests that the
promoted nitrene transfer and, by extension, no prior mechan- non-stereospecific reaction of styrenes with Cu-nitrenoids pro-
istic studies on which to base our investigations of the mecha- ceeds through radical intermediates.11 To evaluate whether
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 9413–9417 | 9415

View Article Online
Communication
Organic & Biomolecular Chemistry
this is also the case for Sc(OTf)3 we attempted a [5 + 1] reaction studies indicating the stepwise, radical nature of this trans-
of 1a in the presence of the radical scavenger TEMPO formation help to rationalize these improvements in scope.
(Scheme 3a). This additive completely shut down the [5 + 1] Overall, this report further cements nitrene-transfer [5 + 1] as
pathway. In addition, we obtained NMR and HRMS evidence an emerging class of useful reactions for tetrahydropyridine
of TEMPO adduct 5.17 Based on these observations, we synthesis, and, along with the previous Rh-catalyzed method,
propose the mechanism outlined in Scheme 3b. Scandium- provides a firm basis on which to continue to develop cycload-
promoted nitrene transfer to the VCP could lead, either ditions of this type as a general entry to the regiocontrolled
directly or via an aziridine intermediate,18 to benzylic radical synthesis of heterocycles.
6, as is evident from the formation of 5. Cyclopropylcarbinyl
rearrangement19 would then provide intermediate 7, which,
after ring closure, would provide the observed product. This Conflicts of interest
mechanism is consistent with the observed regioselectivity of
There are no conflicts to declare.
olefin formation as well as in the substitution pattern observed
for product 2j, presumably due to favorable C–C bond homo-
lysis to the more stable secondary radical in the latter case.
Moreover, the possibility that this class of reactions could
Acknowledgements
occur on a spectrum of polar vs. radical character might
provide a satisfactory explanation for the complementary
selectivity observed for the Rh(II) vs. Sc(III)-promoted reactions.
In summary, we have developed a new nitrene-transfer
[5 + 1] cycloaddition between vinylcyclopropanes and PhINTs
Partial financial support of this work from the National
Science Foundation (CAREER 1845219) is gratefully
acknowledged.
promoted by Sc(OTf)3. This approach eliminates the require-
ment for aryl substituted cyclopropanes and allows access to
Notes and references
products distinct from our previously reported method,
thereby expanding the structural diversity of nitrogen-contain-
ing heterocycles accessible via this strategy. Mechanistic
1 E. Vitaku, D. T. Smith and J. T. Njardarson, J. Med. Chem.,
2014, 57, 10257.
2 (a) D. C. Blakemore, L. Castro, I. Churcher, D. C. Rees,
A. W. Thomas, D. M. Wilson and A. Wood, Nat. Chem.,
2018, 10, 383; (b) F. Lovering, J. Bikker and C. Humblet,
J. Med. Chem., 2009, 52, 6752.
3 For recent examples see: (a) J. A. Malone, C. E. Toussel,
F. R. Fronczek and R. Kartika, Org. Lett., 2019, 21, 3610;
(b) W. Liu, C. Chen, P. Zhou and H. Tan, Org. Lett., 2017,
19, 5830; (c) S. Kramer and G. C. Fu, J. Am. Chem. Soc.,
2015, 137, 3803; (d) Q. Zheng and R. Hua, Tetrahedron Lett.,
2010, 51, 4512; (e) W. Liu, H. Jiang and L. Huang, Org. Lett.,
2010, 12, 312.
4 Q. Wu, J. Hu, X. Ren and J. Zhou, Chem. – Eur. J., 2011, 17,
11553.
5 (a) Z. W. Gilbert, R. J. Hue and I. A. Tonks, Nat. Chem.,
2016, 8, 63; (b) H.-S. Chiu and I. A. Tonks, Angew. Chem.,
Int. Ed., 2018, 57, 6090; (c) Z. W. Davis-Gilbert, X. Wen,
J. D. Goodpaster and I. A. Tonks, J. Am. Chem. Soc., 2018,
140, 7267.
6 K. Matsui, M. Shibuya and Y. Yamamoto, Commun. Chem.,
2018, 1, 21.
7 K. Kawakita, E. P. Beaumier, Y. Kakiuchi, H. Tsurugi,
I. A. Tonks and K. Mashima, J. Am. Chem. Soc., 2019, 141,
4194.
8 A. Saito, Y. Kambara, T. Yagyu, K. Noguchi, A. Yoshimura
and V. V. Zhdankin, Adv. Synth. Catal., 2015, 357, 667.
9 L. A. Combee, S. L. Johnson, J. E. Laudenschlager and
M. K. Hilinski, Org. Lett., 2019, 21, 2307.
10 For other recent examples of the use of vinylcyclopropanes
in N-heterocycle synthesis, see: (a) X.-B. Huang, X.-J. Li,
T.-T. Li, B. Chen, W.-D. Chu, L. He and Q.-Z. Liu, Org. Lett.,
2019, 21, 1713; (b) K. Speilmann, E. Tosi, A. Lebrun,
Scheme 3 TEMPO trapping experiment and proposed mechanism for
the formation of [5 + 1] products.
9416 | Org. Biomol. Chem., 2019, 17, 9413–9417
This journal is © The Royal Society of Chemistry 2019

View Article Online
Organic & Biomolecular Chemistry
Communication
G. Niel, A. van der Lee, R. M. de Figueiredo and
G. Dequirez, P. Retailleau, V. Gandon and P. Dauban,
J.-M. Campagne, Tetrahedron, 2018, 74, 6497; (c) J. Ling,
Chem. – Eur. J., 2016, 22, 9338.
M. Laugeois, V. Ratovelomanana-Vidal and M. R. Vitale, 16 (a) G. A. OIah, V. P. Reddy and G. K. S. Prakash, Chem. Rev.,
Synlett, 2018, 29, 2288; (d) Q. Wang, C. Wang, W. Shi,
Y. Xiao and H. Guo, Org. Biomol. Chem., 2018, 16, 4881;
(e) C.-H. Liu, F. Li, Y. Yuan, M. Dou and Z.-X. Yu, Asian
J. Org. Chem., 2018, 7, 1609.
1992, 92, 69; (b) Z. Goldschmidt and B. Crammer, Chem.
Soc. Rev., 1988, 17, 229; (c) W. J. Hehre and P. D. Hiberty,
J. Am. Chem. Soc., 1974, 96, 302; (d) K. B. Wiberg, D. Shobe
and G. L. Nelson, J. Am. Chem. Soc., 1993, 115, 10645;
(e) C. D. Poulter and S. Winstein, J. Am. Chem. Soc., 1969,
91, 3649; (f) C. D. Poulter and S. Winstein, J. Am. Chem.
Soc., 1969, 91, 3650.
11 D. A. Evans, M. M. Faul and M. T. Bilodeau, J. Am. Chem.
Soc., 1994, 116, 2742.
12 Brief investigations into the effect of ligands (e.g. BOX,
PyBOX, etc.) on the Sc-promoted reaction revealed no 17 See ESI† for details.
advantages over the use of Sc(OTf)3 alone. See the ESI† for 18 Alternatively, Sc(OTf)3 has been shown to catalyze dipolar
details.
cycloadditions of styrene-derived aziridines via C–N bond
cleavage. See: J. S. Yadav, B. V. S. Reddy, S. K. Pandey,
P. Srihari and I. Prathap, Tetrahedron Lett., 2001, 42, 9089.
19 (a) T. A. Halgren, J. D. Roberts, J. H. Horner,
F. N. Martinez, C. Tronche and M. Newcomb, J. Am. Chem.
Soc., 2000, 122, 2988; (b) M. Newcomb and D. L. Chestney,
J. Am. Chem. Soc., 1994, 116, 9753; (c) M. Newcomb,
C. C. Johnson, M. B. Manek and T. R. Varick, J. Am. Chem.
Soc., 1992, 114, 10915; (d) A. L. J. Beckwith and
V. W. Bowry, J. Am. Chem. Soc., 1994, 116, 2710.
13 An expanded optimization table including information on
the effect of solvents other than CH2Cl2 can be found in
the ESI.†
14 (a) J. J. Verendel, T. Zhou, J.-Q. Li, A. Paptchikhine,
O. Lebedev and P. G. Andersson, J. Am. Chem. Soc., 2010,
132, 8880; (b) K. Meng, J. Xia, Y. Wang, X. Zhang, G. Yang
and W. Zhang, Org. Chem. Front., 2017, 4, 1601.
15 (a) G. Dequirez, J. Ciesielski, P. Retailleau and P. Dauban,
Chem.
– Eur. J., 2014, 20, 8929; (b) J. Ciesielski,
This journal is © The Royal Society of Chemistry 2019
Org. Biomol. Chem., 2019, 17, 9413–9417 | 9417