Intramolecular redox-triggered C-H functionalization

Article abstract of DOI:10.1002/anie.201108639

Sacrifice for the team: A one-pot method achieves remote functionalization at the α-position of an amine moiety through the sacrificial reduction of a neighboring group. The process takes advantage of an intramolecular redox reaction, thereby avoiding the need for any external oxidants. This method was applied to a concise five-step total synthesis of indolizidine 167B.

Full text of DOI:10.1002/anie.201108639

.
Angewandte
Communications
DOI: 10.1002/anie.201108639
Redox Reactions
ꢀ
Intramolecular Redox-Triggered C H Functionalization**
Igor D. Jurberg, Bo Peng, Eckhard Wçstefeld, Maximilian Wasserloos, and Nuno Maulide*
ꢀ
C H functionalization strategies have made tremendous
progress over the past decades. The fast-paced development
of this vibrant area is likely due to the recognition by the
chemical community of the potential that such methodologies
possess to streamline synthetic routes. Indeed, the ability to
ꢀ
selectively manipulate targeted C H bonds precludes sub-
strate prefunctionalization prior to the desired key steps and,
therefore, represents a shift in the logical rationale of organic
synthesis.[1] Particularly appealing for functionalization are
ꢀ
C H linkages located immediately adjacent to a heteroatom:
the contemporary development of cross-dehydrogenative-
coupling (CDC) reactions,[2] where the elegant functionaliza-
tion of certain classes of amine substrates is achieved at the
expense of an external oxidant, attests to this fact.[3]
Scheme 1. a) Prior intramolecular redox methodologies based on the
In this context, the family of redox processes involving the
intramolecular functionalization of one of the a,a’-positions
of certain tertiary amines has witnessed a recent vigorous
revival.[4–6] This transformation is based on the propensity of
these amines to undergo a 1,5-hydride shift to an electrophilic
moiety followed by cyclization of the zwitterionic intermedi-
ate formed (originally called the “tert-amino effect”). Inter-
estingly, the majority of the research described is dominated
by Michael acceptors as electropositive fragments (Sche-
me 1a).[7] Only a limited number of reports have dealt with
aldehydes or iminium ions.[5c,8,9]
We became interested in the design of a synthetic strategy
for coupling such powerful intramolecular redox transforma-
tions with a subsequent, intermolecular functionalization
step. The conversion of readily available, 2-substituted
aminobenzaldehydes 1 to functionalized derivatives 2 was
selected as a model platform to implement this plan
(Scheme 1b). This conversion would amount to a redox-
tert-amino effect. b) Proposed design for intermolecular redox-triggered
ꢀ
C H functionalization.
application of the method to a short total synthesis of a
naturally occuring indolizidine alkaloid.
We screened different solvents and Brønsted/Lewis acids
as catalysts for the envisaged redox process and found the use
of 10 mol% of Sc(OTf)3 in 1,2-DCE at 808C to be the optimal
conditions.[10] We then combined this step with the key C C
ꢀ
bond-forming nucleophilic addition (Scheme 1b).[11] To this
end, a variety of commercially available Grignard reagents
could be successfully employed (Scheme 2).[12] The corre-
sponding a-functionalized amine products, bearing a wide
range of appendages, could be obtained in generally good to
excellent yields. It was possible to introduce alkyl, allyl,
branched alkyl, vinyl, and aryl moieties with only marginal
variations in yield (5a–i).[13] The procedure proved nonethe-
less to be sensitive to the nature of the secondary amine
moiety. Upon the change from a pyrrolidine to a piperidine
ring, the yield for the corresponding redox product dropped
significantly (requiring increased catalyst loading, 5j), whilst
the nucleophilic capture step still proceeded smoothly.[14–16]
Importantly, substitution at the aryl ring was also fully
compatible with this sequence (5k–s).
ꢀ
triggered C H functionalization through the sacrificial reduc-
tion of the neighboring carboxaldehyde group. We report
herein our results on the development of this strategy, a
number of relevant preliminary mechanistic studies, and an
[*] Dr. I. D. Jurberg, Dr. B. Peng, E. Wçstefeld, M. Wasserloos,
Dr. N. Maulide
Given the intrinsic versatility and synthetic usefulness of
the alkyne functional group,[17] we became intrigued by the
possibility of achieving C(sp3)–C(sp) bond formation and
eagerly probed lithium alkynyl trifluoroborates[18] as nucleo-
philes. In the event (Scheme 3), addition of excess lithium
alkynyl borates following the redox transformation delivered
the desired substituted amines 6 directly in good to very good
yields for this one-pot operation. As indicates in Scheme 3,
acetylides carrying alkyl, tert-butyl, silyl, and aryl moieties
were generally well tolerated (6a–d). Functionalized alkyl
moieties (6e) and densely substituted aryl residues (6 f–h)
were also tolerated by this process, as well as substitution at
the aromatic ring of the substrate (6i–o). It is noteworthy that
products 5 and 6 result from the overall addition of a strong
Max-Planck-Institut fꢀr Kohlenforschung
Kaiser-Wilhelm-Platz 1, 45470 Mꢀlheim an der Ruhr (Germany)
E-mail: maulide@mpi-muelheim.mpg.de
Homepage: http://www.kofo.mpg.de/maulide
[**] We are grateful to the Max-Planck-Society and the Max-Planck-
Institut fꢀr Kohlenforschung for generous support of our research
programs. This work was funded by the Ecole Polytechnique
(fellowship to I.D.J.) and the Fonds der Chemischen Industrie
(Sachkostenzuschuss to N.M.). We further acknowledge the
invaluable support of our NMR, HPLC, and GC Departments, Dr. R.
Goddard (MPI Mꢀlheim) for crystallographic analysis, and Dr. M.
Klußmann (MPI Mꢀlheim) for helpful discussions.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201108639.
1950
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1950 –1953

Angewandte
Chemie
ꢀ
Scheme 3. Redox-mediated C C coupling employing alkynylborate
nucleophiles. General conditions: amino aldehyde (0.2 mmol),
Sc(OTf)3 (0.02 mmol), 1,2-DCE (2 mL), and lithium alkynylborate
(1.0 mmol). TMS=trimethylsilyl, TBS=tert-butyldimethylsilyl.
ꢀ
Scheme 2. Redox-mediated C C bond formation employing Grignard
nucleophiles. General conditions: amino aldehyde (0.3 mmol),
Sc(OTf)3 (0.03 mmol), 1,2-DCE (3 mL), and Grignard reagent
(0.6 mmol). [a] Reaction performed with 40 mol% catalyst.
DCE=dichloroethane, Cy=cyclohexyl, 2-Np=2-naphthyl.
nucleophile to the amine functionality of the starting amino-
aldehydes 3, in clear contrast to what would be expected from
the well-established, classical reactivity of aldehydes towards
such nucleophiles.
At this stage, preliminary mechanistic experiments aimed
at elucidating details of the redox event were performed
(Scheme 4). As expected, the perdeuterated aminoaldehyde
[D8]-3a afforded the corresponding benzoxazine [D8]-4a in
89% yield, with complete transfer of the deuterium label to
the benzylic position (Scheme 4a). In addition, a crossover
experiment (Scheme 4b) involving equimolar amounts of
aminoaldehydes [D8]-3a and 3b revealed no deuterium
incorporation at the benzylic position of 4b (i.e., no crossover
products). This observation is in full agreement with a
postulated strictly intramolecular process.
We then examined the regioselectivity of the redox step in
more detail. It is generally accepted that in nonsymmetrically
substituted donors, the hydride-transfer pathway leading to
the most stabilized carbocation should dominate.[6,7h] In the
case of the simple substrate 3g, the regioselectivity appeared
to be particularly sensitive to the reaction conditions. The
Scheme 4. a) Labeling experiment. b) Crossover experiment.
Angew. Chem. Int. Ed. 2012, 51, 1950 –1953
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1951

.
Angewandte
Communications
reaction of aminoaldehyde 3g with increasing amounts of
Sc(OTf)3 afforded increasing amounts of the “expected”
(more highly substituted) regioisomer 4g relative to regio-
isomer 4h (Table 1). At lower conversions of 3g (5–10 mol%
of catalyst, entries 1 and 2), a nearly 1:1 mixture of isomers
Table 1: Regioselectivity experiments concerning the redox step.
Entry
x
3g/4g/4h[a]
ꢀ
Scheme 5. CAN-mediated C N bond cleavage of a “PMP-like” protect-
1
2
3
4
5
7:1.8:1
ing group. General conditions: arylamine (0.2 mmol), (NH4)2Ce(NO3)6
(1 mmol) in H2O/CH3CN (3 mL/2 mL). [a] Isolated as the free amine
owing to its higher boiling point. CAN=cerium(IV) ammonium
nitrate.
10
20
50
1:13.9:10
0:2.1:1
0:24:1[b]
[a] Ratio determined by GC analysis. The peaks of the diastereoisomers
of 4h overlap in the GC traces (ratio not determined). [b] The yield of
isolated compound 4g was 90%.
4g/4h was produced, whereas at close to full conversion, 4g
was the dominant product (20–50 mol% of catalyst, entries 3
and 4). Interestingly enough, when the isolated product
mixture obtained with 20 mol% catalyst (4g/4h = 2.1:1,
entry 3 of Table 1) was resubmitted to the reaction conditions
with 50 mol% of the scandium salt, a new mixture signifi-
cantly enriched in 4g (new ratio of 4g/4h = 32:1) was formed.
These observations suggest two intriguing and unusual points:
1) the formation of 4g/4h from 3g is subject to thermody-
namic control, with 4g being the thermodynamic product;
and 2) the formation of 4h (and possibly even 4g) is
reversible under the reaction conditions, implying the exis-
tence of a mechanism for interconversion between 4g and 4h.
To the best of our knowledge, this is the first observation of
such a phenomenon in redox processes of this kind.[19] We also
could measure a significant primary kinetic isotope effect of
2.9, which strongly supports hydride transfer as the rate-
determining step.[20,21]
Importantly, if the aromatic substituent bears a para-
methoxy substituent (such as in substrates 5o–s and 6o),
smooth cleavage of the aromatic residue can be achieved
under mild conditions,[22] thus providing suitably functional-
ized, synthetically useful amines 7a–f in high yields
(Scheme 5).[23]
Finally, the synthetic utility of the method outlined herein
is showcased in the context of the total synthesis of the natural
product (ꢁ)-indolizidine 167B.[24] As depicted in Scheme 6,
alkylation of pyrrolidine 8 with 2-fluoro-4-methoxybenzalde-
hyde 9, followed by sequential treatment with 20 mol% of
Sc(OTf)3 and the Grignard reagent 10 afforded the anisidine
intermediate 11,[25] with an overall yield of 43% for the three
steps. Subsequent cleavage of the aryl protecting group with
CAN was accompanied by partial deprotection of the
dioxolane moiety.[26] Therefore, the crude reaction mixture
was directly subjected to hydrogenolysis at ambient pressure
in aqueous HCl (1m) to achieve the total synthesis of (ꢁ)-
indolizidine 167B (12) in a combined 57% yield for these two
Scheme 6. Total synthesis of (ꢁ)-indolizidine 167B (12).
last steps (Scheme 6). This final transformation is a domino
reaction consisting of acetal cleavage and a diastereoselective
intramolecular reductive amination. The target indolizidine
was synthesized from pyrrolidine in only five steps and 25%
overall yield. Importantly, the initial operations of this short
synthetic sequence can be readily conducted on a gram scale,
highlighting the robustness of the procedure.
ꢀ
In summary, we have developed a one-pot C H function-
alization process for cyclic tertiary amines, in which the
sacrificial reduction of a neighboring carboxaldehyde group
directs the addition of Grignard reagents and lithium alkynyl
trifluoroborates to the a-position of the amine moiety.
Important mechanistic results pertaining to this class of
internal redox reactions relying on the tert-amino effect were
obtained, showcasing the potential for kinetic versus thermo-
dynamic control and pointing towards an unprecedented
isomerization of the reactive intermediates involved. The
removal of the sacrificial substituent under mild conditions at
the end of the sequence provides a-functionalized amine
products of clear synthetic value, which was laid out in a
concise and direct total synthesis of (ꢁ)-indolizidine 167B.
The concepts and insights disclosed herein have obvious
potential for further applications and progress in this area
shall be reported in due course.
1952
www.angewandte.org
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2012, 51, 1950 –1953

Angewandte
Chemie
[10] See the Supporting information for more details.
Received: December 7, 2011
[11] For selected examples on the chemistry of N,O-acetals, see:
a) L. E. Burgess, A. I. Meyers, J. Am. Chem. Soc. 1991, 113,
9858; b) R. W. Bates, Y. Lu, M. P. Cai, Tetrahedron 2009, 65,
7852; c) N. Kise, H. Yamazaki, T. Mabuchi, T. Shono, Tetrahe-
dron Lett. 1994, 35, 1561.
[12] A. Alberola, M. A. Alvarez, C. Andrꢃs, A. Gonzꢀlez, R.
Pedrosa, Synth. Commun. 1990, 20, 1149.
[13] CCDC 840594 (5g) 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.
Published online: January 19, 2012
ꢀ
Keywords: amines · C H functionalization · indolizidine 167B ·
reaction mechanisms · redox reactions
.
[1] a) K. Godula, D. Sames, Science 2006, 312, 67, and references
therein.
[2] For reviews on CDC couplings, see: a) C.-J. Li, Acc. Chem. Res.
2009, 42, 335; b) C. J. Scheuermann, Chem. Asian J. 2010, 5, 436;
c) M. Klussmann, D. Sureshkumar, Synthesis 2011, 353.
[3] For an excellent review on the a-functionalization of amines,
see: a) K. Campos, Chem. Soc. Rev. 2007, 36, 1069.
[14] The yield for the benzoxazine derived from 2-piperidinobenzal-
dehyde was 40%. See the Supporting Information for details on
isolated benzoxazines.
[4] The first example of such a process dates back to 1895: a) J.
Pinnow, Ber. Dtsch. Chem. Ges. 1895, 28, 3039. The term “tert-
amino effect” was first used in 1972: b) O. Meth-Cohn, H.
Suschitzky, Adv. Heterocycl. Chem. 1972, 14, 211. For other
reviews, see: c) O. Meth-Cohn, Adv. Heterocycl. Chem. 1996, 65,
1; d) P. Mꢀtyus, O. ꢁliꢀs, P. Tapolcsꢀnyi, ꢂ. Polonka-Bꢀlint, B.
Halꢀsz-Dajka, Synthesis 2006, 2625.
[5] More rare, 1,6-hydride shifts have also been reported: a) D. N.
Reinhoudt, G. W. Visser, W. Verboom, P. H. Benders, M. L. M.
Pennings, J. Am. Chem. Soc. 1983, 105, 4775; b) E. O. M.
Orlemans, B. H. M. Lammerink, F. C. J. M. van Veggel, W.
Verboom, S. Harkema, D. N. Reinhoudt, J. Org. Chem. 1988,
53, 2278; c) C. Zhang, C. K. De, R. Mal, D. Seidel, J. Am. Chem.
Soc. 2008, 130, 416; d) K. Mori, S. Sueoka, T. Akiyama, J. Am.
Chem. Soc. 2011, 133, 2424.
[15] These findings are in agreement with previous reports, where the
ease of hydride migration was established to follow the order:
pyrrolidine > piperidine > morpholine. See: a) F. Effenberger, P.
Fischer, W. W. Schoeller, W. D. Stohrer, Tetrahedron 1978, 34,
2409.
[16] Other substrates bearing an acyclic amine moiety (such as N,N-
dipropylamino or N,N-dibenzylamino) failed to produce the
corresponding benzoxazines under these conditions.
[17] For leading references on the use of alkynes as versatile motifs,
see: a) R. Chinchilla, C. Nꢀjera, Chem. Rev. 2007, 107, 874; b) A.
Fꢄrstner, P. W. Davies, Chem. Commun. 2005, 2307; c) B. M.
Trost, Acc. Chem. Res. 2002, 35, 695; d) B. M. Trost, A. H. Weiss,
Angew. Chem. 2007, 119, 7808; Angew. Chem. Int. Ed. 2007, 46,
7664.
[18] a) M.-J. Wu, D.-S. Yan, H.-W. Tsai, S.-H. Chen, Tetrahedron Lett.
1994, 35, 5003.
[6] There is still considerable uncertainty whether 1,5-hydride
migration in such molecules occurs as a sigmatropic shift or
“through space”, as “an intramolecular variant of a Meerwein–
Pondorf–Verley reaction”, cf.: a) W. H. N. Nijhuis, W. Verboom,
A. A. El-Fadl, S. Harkema, D. N. Reinhoudt, J. Org. Chem. 1989,
54, 199.
[19] Sames and co-workers have established reversibility between
starting material and redox product in the case of O,O-
spiroacetals (see Ref. [8c]). The case presented herein concerns
the reversibility between two different redox products. See the
Supporting Information for details.
[7] a) W. H. N. Nijhuis, W. Verboom, D. N. Reinhoudt, J. Am. Chem.
Soc. 1987, 109, 3136; b) K. M. McQuaid, J. Z. Long, D. Sames,
Org. Lett. 2009, 11, 2972; c) S. J. Pastine, K. M. McQuaid, D.
Sames, J. Am. Chem. Soc. 2005, 127, 12180; d) K. M. McQuaid,
D. Sames, J. Am. Chem. Soc. 2009, 131, 402; e) M. C. Haibach, I.
Deb, C. K. De, D. Seidel, J. Am. Chem. Soc. 2011, 133, 2100; f) S.
Murarka, C. Zhang, M. D. Konieczynska, D. Seidel, Org. Lett.
2009, 11, 129; g) K. Mori, T. Kawasaki, S. Sueoka, T. Akiyama,
Org. Lett. 2010, 12, 1732.
[8] To the best of our knowledge, there are only two examples in the
literature closely related to the transformation 3!4. One of
them employs harsh conditions and provides very low yields:
a) N. Kaval, B. Halasz-Dajka, G. Vo-Thanh, W. Dehaen, J.
van der Eycken, P. Mꢀtyus, A. Loupy, E. van der Eycken,
Tetrahedron 2005, 61, 9052; the second method employs the
corresponding trifluoromethyl ketones as more reactive 1,5-
hydride acceptors: b) W. Verboom, B. G. van Dijk, D. N. Rein-
houdt, Tetrahedron Lett. 1983, 24, 3923; for 1,5-hydride shifts to
aldehydes, see: c) S. Pastine, D. Sames, Org. Lett. 2005, 7, 5429;
for an example using imines, see: d) C. Zhang, S. Murarka, D.
Seidel, J. Org. Chem. 2009, 74, 419.
[20] See the Supporting Information for more details and additional
experiments with deuterium-labeled compounds.
[21] A similar KIE value of 3.0 ꢁ 0.3 at 91.28C in [D6]DMSO was
measured for [D4]-2-(2-(pyrrolidin-1-yl)benzylidene) malononi-
trile. See: a) L. C. Groenen, W. Verboom, W. H. N. Nijhuis, D. N.
Reinhoudt, G. J. Van Hummel, D. Feil, Tetrahedron 1988, 44,
4637.
[22] a) D. R. Kronenthal, C. Y. Han, M. K. Taylor, J. Org. Chem.
1982, 47, 2765; b) J. M. M. Verkade, L. J. C. van Hemert,
P. J. L. M. Quaedflieg, P. L. Alsters, F. L. van Delft, F. P. J. T.
Rutjes, Tetrahedron Lett. 2006, 47, 8109.
[23] This route appears competitive with previously reported meth-
ods towards racemic a-alkylated pyrrolidines. For selected
examples, see the Supporting Information. For a related strategy,
see: G. Han, M. G. LaPorte, M. C. McIntosh, S. M. Weinreb, J.
Org. Chem. 1996, 61, 9483.
[24] For selected previous total syntheses of indolizidine 167B, see
the Supporting Information.
[25] Grignard reagent 10 was prepared in three steps (see the
Supporting Information for details). As 10 is not commercially
available, we found it more practical in this case to isolate the
benzoxazine intermediate prior to the Grignard addition.
[26] For prior reports on acetal deprotection using CAN, see: a) N.
Maulide, J.-C. Vanherck, A. Gautier, I. E. Markꢅ, Acc. Chem.
Res. 2007, 40, 381; b) N. Maulide, I. E. Markꢅ, Synlett 2005, 2195;
c) A. Ates, A. Gautier, B. Leroy, J.-M. Plancher, Y. Quesnel, J.-
C. Vanherck, I. E. Markꢅ, Tetrahedron 2003, 59, 8989, and
references therein.
[9] For the use of other hydride acceptors such as alkynes, see: a) B.
Bolte, Y. Odabachian, F. Gagosz, J. Am. Chem. Soc. 2010, 132,
7294; b) I. D. Jurberg, Y. Odabachian, F. Gagosz, J. Am. Chem.
Soc. 2010, 132, 3543; c) P. A. Vadola, D. Sames, J. Am. Chem.
Soc. 2009, 131, 16525; d) D. Shikanai, H. Murase, T. Hata, H.
Urabe, J. Am. Chem. Soc. 2009, 131, 3166; For a leading
reference on the use of allenes as hydride acceptors, see: e) B.
Bolte, F. Gagosz, J. Am. Chem. Soc. 2011, 133, 7696.
Angew. Chem. Int. Ed. 2012, 51, 1950 –1953
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1953