Pyrrolidines and Piperidines by Ligand-Enabled Aza-Heck Cyclizations and Cascades of N-(Pentafluorobenzoyloxy)carbamates

Article abstract of DOI:10.1002/anie.201801109

Ligand-enabled aza-Heck cyclizations and cascades of N-(pentafluorobenzoyloxy)carbamates are described. These studies encompass the first examples of efficient non-biased 6-exo aza-Heck cyclizations. The methodology provides direct and flexible access to carbamate protected pyrrolidines and piperidines.

Full text of DOI:10.1002/anie.201801109

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Pyrrolidines and Piperidines by Ligand-Enabled
Aza-Heck Cyclizations and Cascades of N-
(Pentafluorobenzoyloxy)carbamates
Authors: Ian Hazelden, Rafaela Carmona, Thomas Langer, Paul
Pringle, and John Bower
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201801109
Angew. Chem. 10.1002/ange.201801109
Link to VoR: http://dx.doi.org/10.1002/anie.201801109
http://dx.doi.org/10.1002/ange.201801109

10.1002/anie.201801109
Angewandte Chemie International Edition
COMMUNICATION
Pyrrolidines and Piperidines by Ligand-Enabled Aza-Heck Cycli-
zations and Cascades of N-(Pentafluorobenzoyloxy)carbamates
Ian R. Hazelden, Rafaela C. Carmona, Thomas Langer, Paul G. Pringle, and John F. Bower*
Abstract: Ligand enabled aza-Heck cyclizations and cascades of
N-(pentafluorobenzoyloxy)carbamates are described. These studies
encompass the first examples of efficient non-biased 6-exo aza-Heck
cyclizations. The methodology provides direct and flexible access to
carbamate protected pyrrolidines and piperidines.
Pyrrolidines and piperidines are two of the most common
saturated heterocycles used in pharmaceutical development
(Scheme 1A).[1] Consequently, efficient and general methods for
their preparation are required. A conceptually appealing approach
lies in the intramolecular aza-Wacker process, where oxidative
cyclization of an NH nucleophile with an alkene occurs under
Pd(II)-catalyzed conditions (Scheme 1B).[2] This method has been
developed extensively, but, in general, still requires relatively
acidic NH units, such as sulfonamides (PG = SO2R), for efficient
reactivity.[2,3,4] Aza-Wacker cyclizations of less acidic carbamates
(PG = CO2R) are much slower[3e] and are limited to 5-ring
cyclizations involving more reactive classes of alkene (e.g. cyclic
or sterically undemanding variants).[3c,e,f,4] Because carbamate
protecting groups (e.g. Boc, Cbz) offer the greatest downstream
flexibility, methods that can circumvent these limitations and
provide direct access to protected pyrrolidines and piperidines are
likely to find broad use.
A solution to the aza-Wacker “carbamate problem” potentially
resides in the development of an aza-Heck process where an
activated N-hydroxycarbamate unit (2) is exploited for N-O
oxidative addition (to 3) prior to C-N bond forming migratory
insertion of the alkene (Scheme 1C).[5] The key to this umpoled
approach is that it relies on the electrophilicity of the N-center
rather than on its acidity (cf. Scheme 1B), such that wide scope
might be expected. Further potential benefits include: (a) direct
access to the substrates by Mitsunobu alkylation of bifunctional
amino reagents 1,[6] (b) the avoidance of (hazardous) external
oxidants,[7] which, in turn, should allow highly tunable/stabilizing
phosphine ligands to be used, (c) a compatibility with
Scheme 1. Introduction.
organometallic reagents in cascade processes,[8] and (d)
predictable syn-amino-palladation of the alkene.[4] To date, the
range of catalytically useful N-O oxidative addition processes
developed with Pd(0)-systems is still very limited,[9,11,12] such that
the viability of the approach in Scheme 1C was deemed uncertain.
Nevertheless, as described below, the identification of
a
privileged ligand set has allowed us to achieve both the envisaged
aza-Heck cyclizations, as well as related cascade processes. The
new method is efficient for both 5-exo and non-biased 6-exo
cyclizations; this latter aspect is particularly significant as prior
aza-Heck protocols cannot achieve cyclizations of this
type.[5,10,11,12] The end result is a highly flexible method that
enables the two-step conversion of bis- or trishomoallylic alcohols
to carbamate protected pyrrolidines or piperidines.
I. R. Hazelden, R. C. Carmona, Prof. P. G. Pringle, Prof. J. F.
Bower
School of Chemistry,
University of Bristol,
Bristol, BS8 1TS, United Kingdom
E-mail: john.bower@bris.ac.uk
At the outset of our studies, only three principal classes of aza-
Heck N-O donor were known: O-FBz ketoxime esters reported by
Narasaka in 1999 (Class 1),[10] O-FBz hydroxysulfonamides
reported by our group in 2016 (Class 2),[11] and O-phenyl
hydroxamates reported by Watson and co-workers in 2016 (Class
3).[12] Each system exhibits prescriptive ligand requirements,
although, in general, electron poor P-based ligands are required
for efficient reactivity. Class 1 and Class 2 N-O donors cyclize via
Dr T. Langer
Pharmaceutical Technology & Development,
AstraZeneca,
Charter Way, Macclesfield,
SK10 2NA, United Kingdom
Supporting information for this article is given via a link at the end
of the document.
This article is protected by copyright. All rights reserved.

10.1002/anie.201801109
Angewandte Chemie International Edition
COMMUNICATION
a cationic aza-Pd(II) intermediate, access to which is driven by
facile protodecarboxylation of the pentafluorobenzoate leaving
group.[10f] Given these considerations, we elected to evaluate the
cyclization of O-FBz carbamate 2a in the presence of Pd-systems
modified by weak donor ligands. Gratifyingly, we found that the
target cyclization was feasible, and, for this non-demanding
system, a variety of triarylphosphine ligands were reasonably
effective using Pd2(dba)3 as the precatalyst (see the SI).
Ultimately, the optimal system was PA-Ph (L-1, PA = 1,3,5,7-
tetramethyl-2,4,8-trioxa-6-phosphaadamantanyl) and, using this
ligand, we were able to access target 4a in 85% yield after
optimization of other reaction parameters. L-1 is a bulky and
electron poor ligand, with the latter facet resulting from its
constrained C-P-C bond angle and inductively withdrawing
oxygen atoms.[13] The bulky tert-butyl unit of 2a is also beneficial,
with less sterically demanding systems 2b and 2c cyclizing in
lower but acceptable yields.
highly effective across a wide range of substrates (Table 1).
Different carbamates are tolerated (4a-d), diastereoselective
processes are readily achieved (4f-h), tetrasubstituted
stereocenters can be constructed (4i, 4k and 4m) and electron
poor alkenes participate efficiently (2j to 4j). The method is
especially powerful for bicyclic ring construction; 5-exo cyclization
onto exocyclic (4k) or cyclic (4l) alkenes provided complex
perhydroindole scaffolds, and spiro (4m) or transannular (4n) C-
N bond formations were also efficient. For demanding systems
(e.g. 4i) L-2 or L-3 provide 10-15% higher yields than L-1
(selected comparisons are given in the SI). The results in Table 1
show that the aza-Heck method offers far greater scope for 5-exo
cyclizations than currently available aza-Wacker protocols.
Table 2. Carbamate protected heterocycles by 6- and 7-exo aza-Heck
cyclization.
Prior classes of aza-Heck process cannot achieve efficient 6-
exo cyclizations of non-biased systems, and a solution to this
issue represents a longstanding challenge of the area.[5a] We
were pleased to find that the present method addresses this, as
demonstrated by the cyclizations of 2o and 2p, which occurred
with good levels of efficiency to afford 4o and 4p, respectively
(Table 2). More highly substituted systems can also be generated
(e.g. 4s and 4t), with the method offering particularly good scope
for the construction of tetrahydroisoquinolines (4r and 4u), as well
as unusual aza-variants (4v and 4w). The process is effective for
cyclizations involving both electron rich (4r) and electron poor
(e.g. 4u) alkenes. For these more demanding 6-exo cyclizations
an N-Boc group is required; cyclization to afford methyl carbamate
system 4q occurred in only 33% yield under optimized conditions.
The use of PA-Ar ligand systems is also critical for the processes
in Table 2, with L-2 or L-3 being the preferred variants.
Triarylphosphines that were effective for 5-exo cyclization
Table 1. Carbamate protected pyrrolidines by aza-Heck cyclization.
The efficacy of L-1 prompted us to undertake the one-step
synthesis of a variety of electronically tuned variants via arylation
of commercially available PA-H (see the SI). These studies
revealed that systems with electron withdrawing groups at the
para-position of the aryl unit were especially effective, such that
L-2 and L-3 emerged as complementary ligands for subsequent
studies. Using this ligand set, we explored the scope of the
catalyst system for 5-exo aza-Heck cyclizations and found it to be
This article is protected by copyright. All rights reserved.

10.1002/anie.201801109
Angewandte Chemie International Edition
COMMUNICATION
generated 4p in less than 10% yield (see the SI). The PA-Ar
ligand system even enabled 7-exo cyclization to afford 4x, albeit
in modest yield.
cyclization of trans-acrylate 2t delivered adduct 4t as a single
geometric isomer, in which the alkene substituents that were
present in the starting material are now in a cis-arrangement. The
observed switch in geometry is consistent with a sequence of syn-
amino-palladation and syn-β-hydride elimination (Scheme 2B); a
similar phenomenon is observed in the conventional Heck
reaction.[15] For the cyclization of 2u and 2v, this geometry
inversion was not observed at full conversion, with 4u and 4v
formed in >25:1 Z:E ratios. However, when the cyclization of 2u
was run to partial conversion (3 hours), 4u was generated in a
14:1 E:Z ratio, such that isomerization of the initially formed
product likely accounts for the geometry of isolated material (see
the SI for details).[16] As with Class 1 and 2 aza-Heck processes,
protodecarboxylation of the pentafluorobenzoate leaving group
likely plays a key role in the processes described here. 19F and 1H
NMR studies revealed that this process is intimately linked to
turnover; in the cyclization of CF3-2i, C6F5H was formed at the
same rate as cyclization product CF3-4i (Scheme 2C).
Accordingly, we suggest that a cationic aza-Pd(II) intermediate is
required for cyclization and access to this is driven by
triethylammonium
mediated
protodecarboxylation
of
pentafluorobenzoate, a process that we have shown to be
facile.[10f] The efficiency of the PA-Ar ligand system is consistent
with studies by Hanley and Hartwig where electron poor and bulky
P-based ligands were found to accelerate alkene aza-palladation
in other contexts.[17] For the current processes, the synergy of a
bulky ligand system and a bulky N-protecting group may be
especially beneficial, and this might account for the higher
efficiencies observed for N-Boc protected systems. The
conformational control that this unit provides is also likely a key
factor.
1
0.8
0.6
0.4
0.2
0
0
1
2
3
4
5
Time / h
Scheme 2. Key mechanistic observations.
For the processes described here, our collective observations
are supportive of an aza-Heck pathway akin to that proposed for
Class 1 and Class 2 N-O donors.[10f,11] Under optimized
conditions, cyclization of OFBz system 2k in the presence of NH
system 5 provided target 4k in 79% yield and aza-Wacker product
4a was not observed (Scheme 2A). This result confirms that the
N-O bond acts as an internal oxidant only. Accordingly, N-O
oxidative addition to 3 should be followed by syn-stereospecific
amino-palladation of the alkene.[14] Consistent with this,
Scheme 3. Cascade processes.
The aza-Heck process can also be adapted to cascade
sequences where the alkyl-Pd(II) intermediate formed upon
alkene amino-palladation is diverted to a subsequent C-C bond
forming event. For example, aza-Heck-Heck cyclization of bis-
This article is protected by copyright. All rights reserved.

10.1002/anie.201801109
Angewandte Chemie International Edition
COMMUNICATION
[6] The O-based activating group facilitates this step. NH-carbamates required
for aza-Wacker cyclization are not prepared directly by this method
because of the low acidity of primary carbamates and competing
iminophosphorane formation: Bittner, S.; Assaf, Y.; Krief, P.; Pomerantz,
M.; Ziemnicka, B. T.; Smith, C. G. J. Org. Chem. 1985, 50, 1712.
[7] The avoidance of an external oxidant is beneficial for scale-up.
[8] Organometallic reagents effect reduction of Pd(II)-catalysts used in aza-
Wacker reactions and so are not compatible with this approach. For a
process that is stoichiometric in Pd(II), see: Ambrosini, L. M.; Cernak, T. A.;
Lambert, T. H. Synthesis 2010, 870.
alkenyl system 2y delivered spirocycle 4y in 68% yield (Scheme
3A). We have also assessed the feasibility of partially
intermolecular cascade processes as a means of providing a
modular and flexible approach to alkene 1,2-carboamination
(Scheme 3B).[10f,18,19] Cyclization of 6a in the presence of N-methyl
indole-2-boronic acid pinacol ester (200 mol%) provided 1,2-
amino-arylation product 7aa in 73% yield. Other electron rich
heteroaryl boronic esters were also able to trap the alkyl-Pd(II)
intermediate efficiently to give 1,2-amino-arylation products 7ab-
7ad.
[9] To the best of our knowledge, this step has only been confirmed for oxime
ester N-O bonds. See: (a) Tan, Y.; Hartwig, J. F. J. Am. Chem. Soc. 2010,
132, 3676; (b) Hong, W. P.; Iosub, A. V.; Stahl, S. S. J. Am. Chem. Soc.
2013, 135, 13664. For a process where N-O oxidative addition of Pd(0) is
invoked but not confirmed, see: c) J. He, T. Shigenari, J.-Q. Yu, Angew.
Chem. Int. Ed. 2015, 54, 6545.
In summary, we outline highly efficient aza-Heck cyclizations of
activated N-hydroxycarbamates. The chemistry is reliant on PA-
Ar ligand systems, and, importantly, these allow, for the first time,
efficient non-biased 6-exo cyclizations. Further generalization of
the approach, including the development of asymmetric
variants[20] and other classes of cascade reaction, will be reported
in due course. In broader terms, the studies described here have
uncovered a new entry to aza-Pd(II) intermediates via N-O
oxidative addition. Given the now broad utility of oxime ester
derived imino-Pd(II) intermediates,[10,21] application of this unusual
initiation mode[9] in the design of other redox neutral C-N bond
formations can be anticipated.
[10] Seminal studies on aza-Heck cyclizations using oxime esters: (a) Tsutsui,
H.; Narasaka, K. Chem. Lett. 1999, 28, 45; (b) Tsutsui, H.; Kitamura, M.;
Narasaka, K. Bull. Chem. Soc. Jpn. 2002, 75, 1451. Contributions from our
group: (c) Faulkner, A.; Bower, J. F. Angew. Chem. Int. Ed. 2012, 51, 1675;
(d) Race, N. J.; Bower, J. F. Org. Lett. 2013, 15, 4616; (e) Faulkner, A.;
Scott, J. S.; Bower, J. F. Chem. Commun. 2013, 49, 1521; (f) Faulkner, A.;
Scott, J. S.; Bower, J. F. J. Am. Chem. Soc. 2015, 137, 7224; (g) Race, N.
J.; Faulkner, A.; Shaw, M. H.; Bower, J. F. Chem. Sci. 2016, 7, 1508; (h)
Race, N. J.; Faulkner, A.; Fumagalli, G.; Yamauchi, T.; Scott, J. S.; Rydén-
Landergren, M.; Sparkes, H. A.; Bower, J. F. Chem. Sci. 2017, 8, 1981.
[11] Aza-Heck cyclizations using N-(pentafluorobenzoyloxy)sulfonamides:
Hazelden, I. R.; Ma, X.; Langer, T.; Bower, J. F. Angew. Chem. Int. Ed.
2016, 55, 11198.
Acknowledgements
[12] Aza-Heck cyclizations using O-phenyl hydroxamates: Shuler, S. A.; Yin, G.;
Krause, S. B.; Vesper, C. M.; Watson, D. A. J. Am. Chem. Soc. 2016, 138,
13830.
We thank AstraZeneca and EPSRC (EP/M506473/1;
studentship to I.R.H.), FAPESP (Grant no. 2016/00422-0;
studentship to R.C.C.) and the Royal Society (URF to J.F.B.). Dr
David Whittaker and Dr Michael Nunn (AstraZeneca) are thanked
for assistance.
[13] (a) Baber, R. A.; Clarke, M. L.; Heslop, K. M.; Marr, A. C.; Orpen, A. G.;
Pringle, P. G.; Ward, A.; Zambrano-Williams, D. E. Dalton Trans. 2005,
1079. (b) Le, C. M.; Hou, X.; Sperger, T.; Schoenebeck, F.; Lautens, M.
Angew. Chem. Int. Ed. 2015, 54, 15897.
[14] Hanley, P. S.; Hartwig, J. F. Angew. Chem. Int. Ed. 2013, 52, 8510.
[15] Littke, A. F.; Fu, G. C. J. Am. Chem. Soc. 2001, 123, 6989.
[16] Attempted isolation of the (E)-isomer of 4u was not possible due to facile
isomerization to the (Z)-isomer during chromatography.
Keywords: palladium, aza-Heck, N-heterocycle, cascade
[1] Vitaku, E.; Smith, D. T; Njardarson, J. T. J. Med. Chem. 2014, 57, 10257.
[2] Reviews: (a) Minatti, A.; Muñiz, K. Chem. Soc. Rev. 2007, 36, 1142; (b)
McDonald, R. I.; Liu, G.; Stahl, S. S. Chem. Rev. 2011, 111, 2981.
[17] Hanley, P. S.; Hartwig, J. F. J. Am. Chem. Soc. 2011, 133, 15661.
[18] For selected examples of Pd(II)-catalyzed alkene 1,2-carboaminations
under oxidative conditions, see: (a) Cernak, T. A.; Lambert, T. H. J. Am.
Chem. Soc. 2009, 131, 3124; (b) Sibbald, P. A.; Rosewall, C. F.; Swartz,
R. D.; Michael, F. E. J. Am. Chem. Soc. 2009, 131, 15945; (c) Cheng, J.;
Qi, X.; Chen, P.; Liu, G. J. Am. Chem. Soc. 2015, 137, 2480 and references
cited therein.
[3] Selected contributions: (a) Hegedus, L.S.; Allen, G. F.; Waterman, E. L. J.
Am. Chem. Soc. 1976, 98, 2674; (b) Hegedus, L. S.; McKearin, J. M. J.
Am. Chem. Soc. 1982, 104, 2444; (c) Rönn, M.; Bäckvall, J.-E.; Andersson,
P. G. Tetrahedron Lett. 1995, 36, 7749; (d) Larock, R. C.; Hightower, T. R.;
Hasvold, L. A.; Peterson, K. P. J. Org. Chem. 1996, 61, 3584; (e) Fix, S.
R.; Brice, J. L.; Stahl, S. S. Angew. Chem. Int. Ed. 2002, 41, 164; (f) Mori,
M.; Nakanishi, M.; Kajishima, D.; Sato, Y. J. Am. Chem. Soc. 2003, 125,
9801; (g) Liu, G. S.; Stahl, S. S. J. Am. Chem. Soc. 2007, 129, 6328.
[4] Aza-Wacker processes proceed via condition dependent syn- or anti-
aminopalladation pathways. The acidity of the NH unit is an important factor
in the former (see reference 3g), whereas the nucleophilicity of this
component is an important factor in the latter; as such, sulfonamides are
more effective than carbamates in both scenarios. The prevalence of two
competing and stereochemically divergent amino-palladation pathways
impacts diastereo- and/or enantioselective processes. The SI contains a
comparison of aza-Heck vs. aza-Wacker protocols for the synthesis of 4h.
[5] The term “aza-Heck” describes a Pd-catalyzed process encompassing
steps analogous to the conventional Heck reaction: (1) oxidative initiation
at nitrogen, (2) C-N bond forming alkene migratory insertion and (3) β-
hydride elimination. For reviews, see: (a) Race, N. J.; Hazelden, I. R.;
Faulkner, A.; Bower, J. F. Chem. Sci. 2017, 8, 5248; (b) Vulovic, B.;
Watson, D. A. Eur. J. Org. Chem. 2017, 4996.
[19] For selected Pd-catalyzed alkene 1,2-carboaminations that use internal C-
based oxidants, see: (a) Ney, J. E.; Wolfe, J. P. Angew. Chem., Int. Ed.
2004, 43, 3605; (b) Mai, D. N.; Wolfe, J. P. J. Am. Chem. Soc. 2010, 132,
12157; (c) Liu, Z.; Wang, Y.; Wang, Z.; Zeng, T.; Liu, P.; Engle, K. M. J.
Am. Chem. Soc. 2017, 139, 11261 and references cited therein.
[20] In efforts towards this, we have found that 4i is generated in 44% yield and
48% e.e. when L-2 is replaced by (S,R,R)-(+)-(3,5-dioxa-4-
phosphacyclohepta[2,1-a:3,4-a′]dinaphthalen-4-yl)bis(1-
phenylethyl)amine.
[21] Since its introduction in 1999, this initiation mode has been used for: (a)
alkene aziridinations: Okamoto, K.; Oda, T.; Kohigashi, S.; Ohe, K. Angew.
Chem. Int. Ed. 2011, 50, 11470; (b) alkene 1,2-iodoamination: Chen, C.;
Hou, L.; Cheng, M.; Su, J.; Tong, X. Angew. Chem. Int. Ed. 2015, 54, 3092;
(c) aryne aminofunctionalizations: Gerfaud, T.; Neuville, L.; Zhu, J. Angew.
Chem. Int. Ed. 2009, 48, 572; (d) β-carbon eliminations: Nishimura, T.;
Uemura, S. J. Am. Chem. Soc. 2000, 122, 12049; (e) enantioselective
alkene 1,2-carboaminations: Bao, X.; Wang, Q.; Zhu, J. Angew. Chem. Int.
Ed. 2017, 56, 9577.
This article is protected by copyright. All rights reserved.

10.1002/anie.201801109
Angewandte Chemie International Edition
COMMUNICATION
COMMUNICATION
Ian R. Hazelden, Rafaela C. Carmona,
Thomas Langer, Paul G. Pringle and
John F. Bower*
Page No. – Page No.
Pyrrolidines and Piperidines by
Ligand-Enabled Aza-Heck
Cyclizations and Cascades of
N-(Pentafluorobenzoyloxy)-
carbamates
Ligand
enabled
aza-Heck
cyclizations
and
cascades
of
N
(pentafluorobenzoyloxy)carbamates are described. These studies encompass the
first examples of efficient non-biased 6-exo aza-Heck cyclizations. The methodology
provides direct and flexible access to carbamate protected pyrrolidines and
piperidines.
This article is protected by copyright. All rights reserved.