Catalytic Synthesis of N-Unprotected Piperazines, Morpholines, and Thiomorpholines from Aldehydes and SnAP Reagents

Article abstract of DOI:10.1002/anie.201505167

Commercially available SnAP (stannyl amine protocol) reagents allow the transformation of aldehydes and ketones into a variety of N-unprotected heterocycles. By identifying new ligands and reaction conditions, a robust catalytic variant that expands the substrate scope to previously inaccessible heteroaromatic substrates and new substitution patterns was realized. It also establishes the basis for a catalytic enantioselective process through the use of chiral ligands. SnAPcat! The identification of new ligands and reaction conditions provides a robust catalytic method for the synthesis of N-unprotected heterocycles using SnAP reagents. This catalytic variant expands the substrate scope to include previously inaccessible piperazines, morpholines, and thiomorpholines and establishes the basis for a catalytic enantioselective process through the use of chiral ligands.

Full text of DOI:10.1002/anie.201505167

.
Angewandte
Communications
International Edition: DOI: 10.1002/anie.201505167
German Edition: DOI: 10.1002/ange.201505167
Cross-Coupling
Catalytic Synthesis of N-Unprotected Piperazines, Morpholines, and
Thiomorpholines from Aldehydes and SnAP Reagents
Michael U. Luescher and Jeffrey W. Bode*
Abstract: Commercially available SnAP (stannyl amine pro-
tocol) reagents allow the transformation of aldehydes and
ketones into a variety of N-unprotected heterocycles. By
identifying new ligands and reaction conditions, a robust
catalytic variant that expands the substrate scope to previously
inaccessible heteroaromatic substrates and new substitution
patterns was realized. It also establishes the basis for a catalytic
enantioselective process through the use of chiral ligands.
reagents are now commercially available, and SnAP chemis-
try is finding widespread use in industry.
Despite the outstanding substrate scope and operational
simplicity of the SnAP method, we have identified two major
limitations. First, SnAP processes have thus far required
a stoichiometric amount of Cu(OTf)2, which both decreases
the efficiency and limits the possibilities of an enantioselec-
tive process. The need for a stoichiometric amount of copper
is also at odds with our current mechanistic hypothesis, which
postulates an overall redox-neutral reaction.[3,5] We attributed
the need for stoichiometric copper to product inhibition—
a common obstacle for metal-catalyzed reactions in which the
products are more basic than the starting materials.[7] Second,
aldehydes with proximal heteroatoms do not undergo cycli-
zation. This is of special interest as the elusive 2-(pyridine-2-
yl)piperazine moiety and related scaffolds are often found in
bioactive small molecules.[8]
We herein report the identification of ligand-accelerated
SnAP reactions that operate with catalytic amounts of cop-
per and also further expand the substrate scope to include
a-heteroaromatic aldehydes. We also introduce a-bis-sub-
stituted SnAP reagents for the synthesis of 2,3-disubstituted
N-heterocycles, a new product class in SnAP chemis-
try (Figure 1), and demonstrate the viability of this system
for the catalytic enantioselective synthesis of N-heterocycles.
We have postulated that product inhibition—rather than
mechanistic considerations—has thus far precluded the use of
catalytic amounts of copper, as experiments with 20 mol% of
copper gave the desired product in only approximately 20%
yield (Table 1, entries 1–3). Heating to 908C in the absence of
any ligand increased conversion, but under these conditions,
the reaction displayed a poor substrate scope (entry 4). As
deactivation can often be reversed by additives or by changing
the ligand, we initially focused on ligand screening. A survey
of ligands commonly used in copper-catalyzed reactions, such
as phenanthrolines, bipyridines, or phosphines, were shown to
be of insufficient catalytic activity (entries 5–9).[9] Surpris-
ingly, only a single ligand class, namely Box ligands, led to
appreciable catalysis, and we established that 20 mol% of
Cu(OTf)2 in combination with 20 mol% of L8 promoted full
conversion (entry 10). Further optimization focusing on the
reaction temperature and solvent revealed two crucial
parameters: 1) the integrity of the catalyst[10] and 2) an
increased amount of 1,1,1,3,3,3-hexafluoro-2-propanol
(HFIP).[11] Heating the reaction to induce turnover was
detrimental for catalytic activity (entry 11), a counterintuitive
observation that we attribute to the enhancement of ligand
exchange between bis(oxazoline) ligand L8 and the unpro-
tected product. With HFIP as the sole solvent, we were able to
reduce the catalyst loading to 5 mol% (entry 13).[11] Based on
S
aturated N-heterocycles are privileged scaffolds for the
preparation of bioactive small molecules as they offer several
advantages, including improved solubility, bioavailability, and
pharmacokinetics.[1,2] Their use is currently limited by poor
commercial availability and the paucity of methods for their
preparation, particularly for C-mono- and C-disubstituted
variants. In seeking to provide a cross-coupling approach for
the rapid synthesis of substituted saturated N-heterocycles,
we recently disclosed a cross-coupling approach employing
SnAP (stannyl amine protocol) reagents and aldehydes. This
operationally simple process provides facile, one-step access
to C-substituted thiomorpholines,[3] morpholines, pipera-
zines,[4] diazepanes, and other medium-sized heterocycles[5]
and spirocyclic structures (Figure 1).[6] Many of these SnAP
Figure 1. SnAP reagents for the synthesis of N-heterocycles from
aldehydes and ketones. Boc=tert-butyloxycarbonyl, Tf =trifluorometha-
nesulfonyl.
[*] M. U. Luescher, Prof. Dr. J. W. Bode
Laboratory of Organic Chemistry
Department of Chemistry and Applied Biosciences, ETH Zurich
Vladimir-Prelog-Weg 3, 8093 Zurich (Switzerland)
E-mail: bode@org.chem.ethz.ch
Homepage: http://www.bode.ethz.ch
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201505167.
10884
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 10884 –10888

Angewandte
Chemie
Table 1: Reaction optimization and ligand screening.[14]
carboxaldehyde and related heteroaromatic aldehydes were
excellent substrates, an outcome that we attribute to the use
of a superior ligand and the increased amount of HFIP. For
some substrates, such as those containing additional coordi-
nating heteroatoms, a higher catalyst loading (20 mol%) was
beneficial. Substrate-specific optimization is also possible; for
example, for the gram-scale synthesis of 3c, 5.0 mmol of
SnAP M (1) and a catalyst loading of 5 mol% were used. A
few limitations remain, including the reactions of mesitylal-
dehyde (product 3m) or aliphatic substrates that are prone to
enamine formation (product 4k), which gave mostly the
protodestannylated side products (Scheme 1).
Interestingly, the same conditions did not prove as
effective for piperazine synthesis using SnAP Pip (5) as
increased amounts of the protodestannylated side products
were observed. With these substrates, the use of Cu(OTf)2
(10 mol%) and 2,6-lutidine (10 mol%) as the ligand was
found to be optimal, provided that an increased amount of
HFIP was employed to promote turnover. Acetonitrile was
identified as a beneficial additive that improved the yields of
the unprotected piperazines and also reduced the amounts of
the protodestannylated side products (Scheme 2).[15] Other
additives, such as BF3·Et2O, TMSOTf, Sc(OTf)3, or AgOTf
did not have a significant effect, nor did the addition of
fluoride sources (LiF, KF, CsF, TBAF, CuF2) to facilitate
transmetalation. Applying these conditions to reactions of
SnAP M (1) or SnAP TM (2) resulted mostly in unreacted
starting material, reinforcing the importance of the bis(oxazo-
line) ligand L8 for these transformations.
These optimized conditions represent a general procedure
for the synthesis of functionalized 1,4-diazacyclohexanes (6a–
6l) in good yields using SnAP Pip (5; Scheme 2). No special
precautions were necessary for the reaction setup as the
reactions are not particularly air- or moisture sensitive. All
experiments were performed using identical reaction condi-
tions without substrate-specific optimization. Aldehydes
containing various functional groups, such as ester (6k), aryl
halide (6c), or pinacol boronate (6d) moieties, or diverse
heterocycles (6e–6j) enabled the synthesis of scaffolds
suitable for further elaboration. The tolerance of these
conditions towards heterocyclic carboxaldehydes with heter-
oatoms in the ortho position (6e–6h) is noteworthy as no
product was observed under the previously reported stoi-
chiometric conditions with one equivalent of Cu(OTf)2.[3-6]
The catalytic conditions were unfortunately not applicable to
the formation of larger rings, such as diazepanes, which were
easily obtained under the stoichiometric conditions. Ketones
also did not prove to be viable substrates under these
conditions and afforded the desired spirocycles in low yields.
An advantage of SnAP chemistry is the ability to
incorporate various substituents into the reagents themselves,
facilitating the synthesis of substituted N-heterocycles, often
as single diastereomers.[3,4,6b] To date, we have explored
substitution at every position except for that adjacent to the
tributyltin moiety. To complete the study on ring substitution,
we investigated the use of a-bis-substituted SnAP reagents 7
and 9, which would afford two adjacent stereocenters upon
ring closure. Using the described standard conditions, these
reagents were coupled with representative aldehydes to
Entry
Cu(OTf)2 [mol%]
Ligand
Conv. [%] Yield[a] [%]
1
2
3
100
20
20
20
20
20
20
20
20
20
5
2,6-lutidine
2,6-lutidine
–
–
L1 or L2
L3
L4 or L5
L6 or L7
100
33
89
25
24
15
4[b]
5
92
71
ca. 10
30
<5
20
6
7
8
9
10
11[c]
12
13[d]
ca. 20
<10
0
0
<10
–
PPh3 or BINAP
L8
L8
L8
L8
100
ca. 15
25
86
ca. 10
19
5
5
88
76
[a] Yield of product 3c determined by NMR spectroscopy using 1,3,5-
trimethoxybenzene as the internal standard. [b] At 908C. [c] At 458C.
[d] HFIP as the solvent.
literature precedents,[3,11] we attribute the beneficial effect of
HFIP, a solvent of broad utility owing to its strong hydrogen-
bond-donor abilities,[12,13] to the complexation of the
N-heterocycle product, thereby promoting turnover of the
Lewis acidic copper catalyst (Figure 2).
Under the optimized conditions, morpholines and thio-
morpholines were readily obtained from the corresponding
SnAP reagents; aromatic, heteroaromatic, and branched
aliphatic aldehydes generally gave the corresponding prod-
ucts in excellent yields. We were pleased to find that—unlike
for the stoichiometric variant—all regioisomers of pyridine-
Figure 2. Proposed catalytic cycle for the copper-catalyzed cyclization.
Angew. Chem. Int. Ed. 2015, 54, 10884 –10888
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 10885

.
Angewandte
Communications
Scheme 1. Catalytic synthesis of morpholines and thiomorpholines. Reactions were performed on a 0.5 mmol scale using the indicated SnAP
reagent (1.0 equiv) and aldehyde (1.0 equiv). Yields of isolated, analytically pure compounds after chromatography are given. [a] Cu(OTf)2
(10 mol%), (Æ)-PhBox (10 mol%), HFIP (0.1m). [b] Cu(OTf)2 (20 mol%), (Æ)-PhBox (20 mol%), HFIP (0.05m). [c] Reaction on a 5.0 mmol scale:
Cu(OTf)2 (5 mol%), (Æ)-PhBox (5 mol%), HFIP (0.1m), RT, 36 h. Cbz= benzyloxycarbonyl, M.S.=molecular sieves.
Figure 3. Synthesis of disubstituted morpholines. [a] Cu(OTf)2
(10 mol%), (Æ)-PhBox (10 mol%), HFIP (0.1m), 20 h, RT. [b] Cu(OTf)2
(20 mol%), (Æ)-PhBox (20 mol%), HFIP (0.05m), 20 h, RT. [c] The
relative stereochemistry was confirmed by X-ray analysis of (Æ)-8b; all
others were assigned by analogy.
(Figure 3). The trans relative configuration was confirmed by
NMR spectroscopy and X-ray analysis. The facile formation
of vicinally disubstituted heterocycles from these sterically
more demanding SnAP reagents represents a promising
approach to more congested structures that are difficult to
access by other methods.[16]
The identification of substoichiometric reaction condi-
tions presents the exciting opportunity to develop catalytic
ligand-controlled enantioselective variants. Catalytic asym-
metric methods to generate optically active saturated
N-heterocycles remain quite rare and have various limita-
tions, such as the need for complex linear precursors or
protecting groups.[17] In preliminary experiments, we were
pleased to find that using racemic SnAP 6-Et-M (7) in
Scheme 2. Catalytic piperazine synthesis. Reactions were performed on
a 0.5 mmol scale using SnAP Pip (5, 1.0 equiv) and an aldehyde
(1.0 equiv). Yields of isolated, analytically pure compounds after
chromatography are given.
provide the 2,3-disubstituted morpholines 8a and 8b or 10a–
10c in good to excellent yields and high diastereoselectivity
10886 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 10884 –10888

Angewandte
Chemie
combination with enantiopure (S)-PhBox as the ligand,
enantioenriched 2,3-disubstituted morpholine 8c was deliv-
ered in 82% yield and a promising enantiomeric ratio of 80:20
(Scheme 3). Furthermore, the observed stereoconvergence
[1] For selected reviews, see: a) R. D. Taylor, M. MacCoss, A. D. G.
Lawson, J. Med. Chem. 2014, 57, 5845 – 5859; b) E. Vitaku, D. T.
Smith, J. T. Njardarson, J. Med. Chem. 2014, 57, 10257 – 10274;
c) S. D. Roughley, A. M. Jordan, J. Med. Chem. 2011, 54, 3451 –
3479; d) M. E. Welsch, S. A. Snyder, B. R. Stockwell, Curr. Opin.
Chem. Biol. 2010, 14, 347 – 361; e) R. Wijtmans, M. K. S. Vink,
H. E. Schoemaker, F. L. van Delft, R. H. Blaauw, F. P. J. T.
Rutjes, Synthesis 2004, 641 – 662; for recent examples, see: f) L.
Guandalini et al., Bioorg. Med. Chem. Lett. 2015, 25, 1700 –
1704; g) K. Fukunaga et al., Bioorg. Med. Chem. Lett. 2015, 25,
1086 – 1091.
[2] a) F. Lovering, Med. Chem. Commun. 2013, 4, 515 – 519; b) T. J.
Ritchie, S. J. F. Macdonald, S. Peace, S. D. Pickett, C. N. Lus-
combe, Med. Chem. Commun. 2012, 3, 1062 – 1069; c) N. A.
Meanwell, Chem. Res. Toxicol. 2011, 24, 1420 – 1456; d) T. J.
Ritchie, S. J. F. Macdonald, R. J. Young, S. D. Pickett, Drug
Discovery Today 2011, 16, 164 – 171; e) P. D. Leeson, S. A. St-
Gallay, M. C. Wenlock, Med. Chem. Commun. 2011, 2, 91 – 105;
f) T. J. Ritchie, S. J. F. Macdonald, Drug Discovery Today 2009,
14, 1011 – 1020; g) F. Lovering, J. Bikker, C. Humblet, J. Med.
Chem. 2009, 52, 6752 – 6756.
[3] C.-V. T. Vo, G. Mikutis, J. W. Bode, Angew. Chem. Int. Ed. 2013,
52, 1705 – 1708; Angew. Chem. 2013, 125, 1749 – 1752.
[4] M. U. Luescher, C.-V. T. Vo, J. W. Bode, Org. Lett. 2014, 16,
1236 – 1239.
[5] C.-V. T. Vo, M. U. Luescher, J. W. Bode, Nat. Chem. 2014, 6,
310 – 314.
Scheme 3. Catalytic enantioselective morpholine synthesis. For addi-
tional examples, see the Supporting Information. [a] The enantiomeric
ratio (e.r.) was determined by HPLC analysis of the purified product on
a chiral stationary phase. [b] Absolute stereochemistry not assigned.
supports our current mechanistic conjecture that the organo-
stannane leads to a carbon radical upon oxidation by copper-
(II),[3,5] as also observed by Falck and co-workers.[18] In
contrast, most cross-coupling methods of stereodefined
nucleophiles with copper proceed by stereospecific trans-
metalation, and enantioenriched products are accessed only
from non-racemic, configurationally stable precursors, which
are often difficult to access.[19] Although improvements in the
enantioselectivity are clearly needed, the observation of
significant enantioinduction invites further studies on ligand
optimization and investigations into the stereochemistry-
determining step.
[6] a) W.-Y. Siau, J. W. Bode, J. Am. Chem. Soc. 2015, 137, 17726 –
17729; b) K. Geoghegan, J. W. Bode, Org. Lett. 2015, 17, 1934 –
1937.
[7] For a review, see: R. H. Crabtree, Chem. Rev. 2015, 115, 127 –
150.
[8] a) R. Kakarla et al., J. Med. Chem. 2014, 57, 2136 – 2160; b) S.
Mente et al., J. Med. Chem. 2013, 56, 6819 – 6828; c) C. Gardelli
et al., J. Med. Chem. 2007, 50, 4953 – 4975; d) V. Summa, A.
Petrocchi, V. G. Matassa, C. Gardelli, E. Muraglia, M. Rowley,
O. Gonzalez Paz, R. Laufer, E. Monteagudo, P. Pace, J. Med.
Chem. 2006, 49, 6646 – 6649; e) F. Rondu et al., J. Med. Chem.
1997, 40, 3793 – 3803.
In summary, we have identified catalytic methods for the
one-step synthesis of substituted N-unprotected piperazines,
morpholines, and thiomorpholines. This robust and air- and
moisture-tolerant procedure is a valuable addition to SnAP
chemistry and expands the substrate scope to include
2-(pyridine-2-yl)piperazines and related substrates. This
work also introduces a new class of SnAP reagents, with
substitution adjacent to the heteroatom, resulting in the
diastereoselective synthesis of 2,3-disubstituted N-heterocy-
cles. Our preliminary findings on a catalytic enantioselective
variant provide a promising start for the development of new
routes to enantioenriched N-heterocycles.
[9] A. Alexakis, N. Krause, S. Woodward, Copper-Catalyzed Asym-
metric Synthesis, Wiley-VCH, Weinheim, 2014.
[10] Cu(OTf)2 (5–20 mol%), (Æ)-PhBox (5–20 mol%), HFIP (2–
4 mL), RT, 6 h. The reaction should be kept below 308C to
prevent catalyst deactivation during the process of the reaction.
[11] For some examples of the use of hexafluoroisopropanol or
trifluoroethanol to overcome product inhibition, see: a) H. F.
Motiwala, C. Fehl, S. W. Li, E. Hirt, P. Porubsky, J. AubØ, J. Am.
Chem. Soc. 2013, 135, 9000 – 9009; b) S. Kobayashi, H. Kiyohara,
Y. Nakamura, R. Matsubara, J. Am. Chem. Soc. 2004, 126, 6558 –
6559; c) D. A. Evans, D. S. Johnson, Org. Lett. 1999, 1, 595 – 598;
d) D. A. Evans, T. Rovis, M. C. Kozlowski, J. S. Tedrow, J. Am.
Chem. Soc. 1999, 121, 1994 – 1995.
[12] For selected reviews, see: a) I. A. Shuklov, N. V. Dubrovina, A.
Bçrner, Synthesis 2007, 2925 – 2943; b) J.-P. BØguØ, D. Bonnet-
Delpon, B. Crousse, Synlett 2004, 18 – 29.
[13] A. Berkessel, J. A. Adrio, D. Hüttenhain, J. M. Neudçrfl, J. Am.
Chem. Soc. 2006, 128, 8421 – 8426.
Acknowledgements
Financial support was provided by an ETH Research Grant
(ETH-12 11-1) and the European Research Council (ERC
Starting Grant 306793 - CASAA). We thank the LOC Mass
Spectrometry Service, the LOC NMR Service, the Small
Molecule Crystallography Center of ETH Zurich for the X-
ray measurement, and Thomas Pieth for experimental
assistance.
[14] A more detailed list can be found in the Supporting Information
(Table S1).
[15] The reaction reaches full conversion using SnAP Pip (5) with
protodestannylated side products as the main constituents when
the standard conditions reported in Ref. [4] are applied in
a catalytic manner: Cu(OTf)2/2,6-lutidine (10 mol%) in HFIP/
CH2Cl2 (1:4, 0.05m) at room temperature. An increased amount
of HFIP and CH3CN as a co-solvent were vital to obtain
piperazines in meaningful yields.
Keywords: aldehydes · cross-coupling · homogeneous catalysis ·
nitrogen heterocycles · SnAP reagents
How to cite: Angew. Chem. Int. Ed. 2015, 54, 10884–10888
Angew. Chem. 2015, 127, 11034–11038
Angew. Chem. Int. Ed. 2015, 54, 10884 –10888
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 10887

.
Angewandte
Communications
[16] For an example of the addition of a sterically hindered tin
nucleophile to an imine, see: H. Kagoshima, K. Shimada, Chem.
Lett. 2003, 32, 514 – 515.
Lu, Y. G. Zhou, Chem. Rev. 2012, 112, 2557 – 2590; g) Y.-G.
Zhou, Acc. Chem. Res. 2007, 40, 1357 – 1366.
[18] R. K. Bhatt, J. Ye, J. R. Falck, Tetrahedron Lett. 1996, 37, 3811 –
3814.
[19] a) L. Li, C.-Y. Wang, R. Huang, M. R. Biscoe, Nat. Chem. 2013,
5, 607 – 612; b) J. R. Falck, P. K. Patel, A. Bandyopadhyay, J. Am.
Chem. Soc. 2007, 129, 790 – 793; c) J. R. Falck, R. K. Bhatt, J. Ye,
J. Am. Chem. Soc. 1995, 117, 5973 – 5982; d) J. Ye, R. K. Bhatt,
J. R. Falck, J. Am. Chem. Soc. 1994, 116, 1 – 5.
[17] For selected recent examples, see: a) K. M. Korch, C. Eidam-
shaus, D. C. Behenna, S. Nam, D. Horne, B. M. Stoltz, Angew.
Chem. Int. Ed. 2015, 54, 179 – 183; Angew. Chem. 2015, 127, 181 –
185; b) Y. Numajiri, G. JimØnez-OsØs, B. Wang, K. N. Houk,
B. M. Stoltz, Org. Lett. 2015, 17, 1082 – 1085; c) D. C. Behenna,
Y. Liu, T. Yurino, J. Kim, D. E. White, S. C. Virgil, B. M. Stoltz,
Nat. Chem. 2012, 4, 130 – 133; d) H. Zhai, A. Borzenko, Y. Y.
Lau, S. H. Ahn, L. L. Schafer, Angew. Chem. Int. Ed. 2012, 51,
12219 – 12223; Angew. Chem. 2012, 124, 12385 – 12389; e) Y.
Uozumi, A. Tanahashi, T. Hayashi, J. Org. Chem. 1993, 58,
6826 – 6832; for reviews on the catalytic asymmetric hydro-
genation of heteroarenes, see: f) D. S. Wang, Q. A. Chen, S. M.
Received: June 6, 2015
Revised: June 25, 2015
Published online: July 23, 2015
10888 www.angewandte.org
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 10884 –10888