Multicomponent-multicatalyst reactions (MC)2R: Efficient dibenzazepine synthesis

Article abstract of DOI:10.1021/ol4030925

A Rh^I/Pd⁰ catalyst system was applied to the multicomponent synthesis of aza-dibenzazepines from vinylpyridines, arylboronic acids, and amines in a domino process with no intermediate isolation or purification.

Full text of DOI:10.1021/ol4030925

Letter
pubs.acs.org/OrgLett
Multicomponent-Multicatalyst Reactions (MC)2R: Efficient
Dibenzazepine Synthesis
Jennifer Tsoung, Jane Panteleev, Matthias Tesch, and Mark Lautens*
Department of Chemistry, University of Toronto, 80 St George Street, Toronto, Ontario, Canada, M5S 3H6
S
* Supporting Information
ABSTRACT: A RhI/Pd0 catalyst system was applied to the
multicomponent synthesis of aza-dibenzazepines from vinyl-
pyridines, arylboronic acids, and amines in a domino process
with no intermediate isolation or purification.
Table 1. Optimization of Pd0-Catalyzed Amination
n recent years, significant efforts have been made toward
Ideveloping new concepts in catalysis that more closely
mimic enzymatic systems where multiple catalysts each execute
a transformation selectively. While progress has been made
toward developing combinations of organocatalysts1 or
biocatalysts2 with transition-metal catalysts, the use of two or
more transition metals in domino sequences is less studied as
orthogonal reactivity and time resolution are necessary.3−5 We
recently reported the synthesis of aza-dibenzoxepines using a
two catalyst-two component domino reaction.4k As the
analogous dibenzazepine motif is prevalent in many pharma-
ceutically relevant molecules,6 and few de novo syntheses exist,7
we were interested in applying our methodology to the
synthesis of these interesting structures. However, the
attempted synthesis using ortho-aminophenylboronic acids
with our previously developed method failed to give the
desired product (Scheme 1). We speculated that a strategy
using two aryl halides followed by formation of the azepine ring
with two Pd0-catalyzed C−N bond formations with an external
amine might circumvent this problem.
a
a
entry
[Pd]
ligand(s)
4a (%)
5a (%)
1
2
3
4
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
RuPhos (5 mol %)
RuPhos (10mol %)
BrettPhos (10 mol %)
RuPhos (5 mol %)
XPhos (5 mol %)
XPhos (5 mol %)
RuPhos (5 mol %)
19
73
−
41
−
31
89
−
5
6
6
7
88
73
−
−
a
Determined by 1H NMR spectroscopy using p-nitroacetophenone as
an internal standard.
Not only would this allow further complexity to be added to
the desired product in a highly modular fashion, but also would
expand the field of multicatalyst-multicomponent reactions
((MC)2R). Inclusion of three components in a multimetal
catalyzed domino process would increase the complexity of this
strategy as the three RhI- and Pd0-catalyzed transformations
must operate independently under the same reaction
conditions without one catalyst interfering with the others.
Furthermore, as the ortho-aminophenylboronic acid does not
react with 1, the order of the reaction sequence is critical for
success, as the C−N bond must be formed after the other
couplings. We speculated that the C−N bond would form first
on the pyridyl ring, followed by the intramolecular amination.
Herein, we report the successful application of RhI/Pd0-
catalysis in a multicomponent, domino reaction to access aza-
dibenzazepines. To the best of our knowledge, this is the first
example of a three catalyst-three component domino
reaction.8,9
Scheme 1. Proposed Catalytic Steps
Received: October 28, 2013
Published: December 13, 2013
© 2013 American Chemical Society
110
dx.doi.org/10.1021/ol4030925 | Org. Lett. 2014, 16, 110−113

Organic Letters
Letter
Table 2. Optimization of Domino Process
Table 4. Scope of Domino Reaction (2)
ligand(s)
5a
a
b
b
entry
1
[Pd]
(5 mol %)
base
4a (%)
(%)
Pd(OAc)2
RuPhos
XPhos
NaOtBu
57
15
c
2
3
4
5
6
Pd(OAc)2
RuPhos
K2CO
52
11
−
52
6
6
6
6
XPhos
K2CO3
K2CO3
K2CO3
K2CO3
71 (67)
46
−
RuPhos
XPhos
54
34
d
56
a
5 mol % of each ligand is added to the reaction so that the Pd/L ratio
b
is 1:2. Determined by 1H NMR spectroscopy using p-nitro-
acetophenone as an internal standard. Isolated yields in parentheses.
c
d
10 mol % RuPhos was used. Reaction was performed on a 1 mmol
scale. Isolated yield.
a
Isolated yields.
Table 5. Reaction Scope with Aliphatic Amines
Figure 1. In situ monitoring of the reaction of 1a, 2, and p-toluidine.
Conditions: 6 (5 mol %), XPhos (5 mol %), K2CO3 (3 equiv),
dioxane-d8/D2O (10:1, 0.909 M), 95−110 °C, 400 MHz.14
Table 3. Scope of Domino Process (1)
a
entry
product
R1
X
yield (%)
1
2
3
4
4a
4b
4c
CF3
CN
H
CH
CH
CH
N
67
49
b
41
4d
H
53
a
b
Isolated yields. Isolated using a one-pot procedure: 1c, 2,
[Rh(cod)OH]2, and K2CO3 were stirred at 110 °C in dioxane/H2O
for 30 min until 1c was fully consumed, and then p-toluidine, 6, and
XPhos were added to the reaction and stirring resumed at 110 °C for
16 h.
a
b
Isolated yields. Isolated as an inseparable mixture of 9c and
monoaminated product.
(R1CF3). Using Pd(OAc)2 and an equimolar amount of
RuPhos (which has been shown to be effective for C−N bond
formations between 2° amines and aryl chlorides),11 we found
that formation of the dibenzazepine ring was incomplete, giving
predominantly intermediate 5a, where the amine reacted only
with the more electronically favored pyridyl chloride (Table 1,
Studies showed that the RhI-catalyzed arylation reaction with
ortho-halogenated arylboronic acids can occur in high yields in a
short reaction time.10 This encouraging result indicates that a
time-resolved reaction sequence would be possible. A study of
the Pd0-catalyzed amination steps was then conducted using 3a
111
dx.doi.org/10.1021/ol4030925 | Org. Lett. 2014, 16, 110−113

Organic Letters
Letter
also compatible, and electron-poor anilines are generally higher
yielding. Substituents on the arylboronic acids are also
permitted at the 4- (4i−j) and 5-position (4k−l), but 6-
substituted arylboronic acids lead to only trace amounts of the
arylation product.
Scheme 2. Deprotection of 9f and 9g
For aliphatic amines, the reaction was best run in a
sequential, one-pot fashion.16 Full conversion of the RhI-
catalyzed step can generally be observed within 30 min, after
which the Pd0 catalyst, additional base, and amine can be added
to the reaction vessel.17,18 Increasing the electron density of aryl
chloride 2 leads to decreasing yields of the cyclized product 9
(Table 5, entries 9a−9c). Notably, heterocycles such as
pyridines (9d) and thiophenes (9e) can be incorporated.
Protected amines such as 2,4-dimethoxy-benzyl (9f) and allyl
(9g) amines are also compatible, which can then be
deprotected readily to provide the secondary amine 10
(Scheme 2).
entry 1). Adding an excess of ligand partially resolved this
problem, giving a 73% yield of the desired product (entry 2).
Based on reports from Buchwald and co-workers,12 we also
examined multiligand based palladium catalysts, as C−N
coupling with both 1° and 2° amines is required in our
sequence. We found that using a 1:1 ratio of RuPhos and
XPhos gave us the best results, yielding 89% of the desired
dibenzazepine product (entry 4).
We next looked into the compatibility of the three catalyst
systems in a domino process. When all three starting materials
were combined with the catalyst systems, we were able to form
the desired aza-dibenzazepine product 4a as the major product,
with intermediate 5a and the amination/protodeboronation
byproduct 8 produced in minor amounts. By using Buchwald’s
precatalyst 6 as the palladium source,12 we were able to
produce 4a in good yield while also simplifying the reaction
setup (Table 2, entry 3). Further studies show that, without the
addition of XPhos, product yields were significantly lower
(entries 4−5). This suggests that both palladium complexes are
critical for optimal results. The domino process was performed
on a 1 mmol scale, giving a yield of 56% (entry 6).
In summary, we successfully implemented a RhI/Pd0-
catalyzed multicomponent reaction methodology in a domino
process to access N-aryl aza-dibenzazepines in good yields. This
method was further adapted into a pot-economical protocol in
order to generate N-alkyl aza-dibenzazepines in moderate to
good yields. These convergent processes can allow the rapid
creation of complex and structurally diverse compound libraries
from simple starting materials. These processes also exemplify
that multiple transition-metal complexes can operate independ-
ently in one pot, completing several bond-forming steps
without any intermediate workup or purifications. Further
studies to extend this (MC)2R concept are ongoing in our
group.
ASSOCIATED CONTENT
■
S
* Supporting Information
A study of the reaction by 1H NMR spectroscopy (Figure 1)
showed that the vinylpyridine substrate 1a is first rapidly
consumed to give arylation product 3a. The dihalogenated
intermediate is then converted more slowly to form both the
amination intermediate 5a and the final cyclized product 4a.
This study indicates that the domino reaction occurs in a time-
resolved fashion, where the first catalytic cycle creates the
product that is fed into the next catalytic cycle. The
intermolecular C−N bond formation on the pyridyl chloride
occurs faster than the intramolecular amination, and no
intermolecular C−N bond forms on the phenyl chloride.
While the reaction did not reach completion within the same
time as that under standard reaction conditions, heating for
another 24 h showed full conversion to the product 4a.13
With the optimized conditions in hand, we examined the
scope of the (MC)2R. Electron-poor vinylpyridines work best
and gave the desired products in good yields (Table 3, entries
1−2). When the vinylpyridine is not sufficiently electrophilic,
the arylated product, 5, and unidentified byproducts were
observed in the domino process.
When there are no substituents on the vinylpyridine, the aza-
dibenzazepine product can be accessed in a sequential, one-pot
procedure (entry 3).15 Other N-heterocycles can also be used,
such as vinylpyrazines (entry 4) in good yields.
The electronically biased vinylpyridine 1a was used to further
explore the reaction scope of the domino process (Table 4) due
to its higher reactivity.
Ortho-, meta-, and para-substituents are tolerated on the
aniline partner, giving moderate to good yields (4e−h).
Functional groups such as ketones (4f) and nitriles (4h) are
Experimental procedures and full spectroscopic data for all new
compounds. This material is available free of charge via the
Internet at http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
*E-mail: mlautens@chem.utoronto.ca.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Merck and the Natural Science and Engineering
Research Council (NSERC) for an Industrial Research Chair,
and the University of Toronto for financial support. J.T. thanks
the Ontario Student Assistance Program for an Ontario
Government Scholarship. J.P. thanks NSERC for a CGS-D
scholarship. M.T. thanks DAAD for a PROMOS scholarship.
Special thanks to the NMR staff for their help with the NMR
studies.
REFERENCES
■
(1) (a) Park, Y. J.; Park, J.-W.; Jun, C.-H. Acc. Chem. Res. 2008, 41,
222. (b) Shao, Z.; Zhang, H. Chem. Soc. Rev. 2009, 38, 2745.
(c) Rueping, M.; Koenigs, R. M.; Atodiresei, I. Chem.Eur. J. 2010,
16, 9350. (d) Zhong, C.; Shi, X. Eur. J. Org. Chem. 2010, 16, 2999.
(2) (a) Zimmermann, F. T.; Schneider, A.; Schorken, U.; Sprenger,
̈
G. A.; Fessner, W.-D. Tetrahedron: Asymmetry 1999, 10, 1643.
(b) Pamies, O.; Backvall, J. E. Chem. Rev. 2003, 103, 3247.
112
dx.doi.org/10.1021/ol4030925 | Org. Lett. 2014, 16, 110−113

Organic Letters
Letter
(3) For reviews on multicatalytic processes, see: (a) Bruggink, A.;
Schoevaart, R.; Kieboom, T. Org. Process Res. Dev. 2003, 7, 622.
(b) Lee, J. M.; Na, Y.; Han, H.; Chang, S. Chem. Soc. Rev. 2004, 33,
302. (c) Wasilke, J. C.; Obrey, S. J.; Baker, R. T.; Bazan, G. C. Chem.
Rev. 2005, 105, 1001. (d) Ambrosini, L. M.; Lambert, T. H.
ChemCatChem 2010, 2, 1373. (e) Ramachary, D. B.; Jain, S. Org.
Biomol. Chem. 2011, 9, 1277.
(4) For selected examples of multiple transition metal catalysts in
domino reactions, see: (a) Zimmermann, B.; Herwig, J.; Beller, M.
Angew. Chem. 1999, 111, 2515; Angew. Chem., Int. Ed. 1999, 38, 2372.
(b) Jeong, N.; Seo, S. D.; Shin, J. Y. J. Am. Chem. Soc. 2000, 122,
10220. (c) Park, K. H.; Son, S. U.; Chung, Y. K. Org. Lett. 2002, 4,
4361. (d) Nishibayashi, Y.; Yoshikawa, M.; Inada, Y.; Milton, M. D.;
Hidai, M.; Uemura, S. Angew. Chem. 2003, 115, 2785; Angew. Chem.,
Int. Ed. 2003, 42, 2681. (e) Cossy, J.; Bargiggia, F.; BouzBouz, S. Org.
Lett. 2003, 5, 459. (f) Goldman, A. S.; Roy, A.; Huang, Z.; Ahuja, R.;
Schinski, W.; Brookhart, M. Science 2006, 312, 257. (g) Kammerer, C.;
Prestat, G.; Gaillard, T.; Madec, D.; Poli, G. Org. Lett. 2008, 10, 405.
(h) Takahashi, K.; Yamashita, M.; Ichihara, T.; Nakano, K.; Nozaki, K.
Angew. Chem. 2010, 122, 4590; Angew. Chem., Int. Ed. 2010, 49, 4488.
(i) Panteleev, J.; Zhang, L.; Lautens, M. Angew. Chem. 2011, 123,
9255; Angew. Chem., Int. Ed. 2011, 50, 9089. (j) Zhang, L.; Sonaglia,
L.; Stacey, J.; Lautens, M. Org. Lett. 2013, 15, 2128. (k) Friedman, A.;
Panteleev, J.; Tsoung, J.; Hyunh, V.; Lautens, M. Angew. Chem., Int. Ed.
(13) The reaction mixture in the NMR tube shows two distinct
solvent phases from lack of stirring, which explains why the reaction is
not complete in the same time as in the conventional setup.
(14) We were unable to obtain spectral data at 110 °C due to poor
shimming. Initial studies at 95 °C showed no formation of product 4a;
thus, Scheme 3 refers to data obtained from running the reaction in a
sealed NMR tube at 110 °C, with cooling to 95 °C to collect data.
(15) With electron-rich substituents such as morpholine, the Pd0-
catalyzed C−N bond formation does not occur.
(16) Attempts to extend the domino “all-in-one” methodology to
aliphatic amines using the optimized conditions led to primarily the
monoaminated intermediate. The origin of the problem is likely the
need for water in the first step and the need for a strong base such as
NaOtBu in the second.
(17) Optimization studies showed that using precatalyst 6 (5 mol%)
and SPhos (5 mol%) gave better results for the aliphatic amines than
with XPhos used for the aromatic amines. See Supporting Information
for further details.
(18) Isolated yields for product 9a were 60% with 4 Å MS versus
48% without.
2013, 52, 9755. (l) Nahra, F.; Mace,
Angew.Chem.Int.Ed. 2013, 52, 3208.
́
Y.; Lamin, D.; Riant, O.
(5) An alternate approach is flow chemistry. For recent reviews, see:
(a) Mason, B. P.; Price, K. E.; Steinbacher, J. L.; Bogdan, A. R.;
McQuade, B. T. Chem. Rev. 2007, 107, 2300. (b) Ley, S. V.; Baxendale,
I. R. Chimia 2008, 62, 162. (c) Seeberger, P. Nat. Chem. 2009, 1, 258.
(d) Miller, P. W.; Jennings, L. E.; deMello, A. J.; Gee, A. D.; Long, N.
J.; Vilar, R. Adv. Synth. Catal. 2009, 351, 3260. (e) Webb, D.; Jamison,
T. F. Chem. Sci. 2010, 1, 675. (f) Yoshida, J. I. Chem. Rec. 2010, 10,
332. (g) Wegner, J.; Ceylan, S.; Kirschning, A. Chem. Commun. 2011,
47, 4583. (h) Wiles, C.; Watts, P. Chem. Commun. 2011, 47, 6512.
(i) Hartman, R. L.; McMullen, J. P.; Jensen, K. F. Angew. Chem., Int.
Ed. 2011, 50, 7502. (j) Baumann, M.; Baxendale, I. R.; Ley, S. V. Mol.
Diversity 2011, 15, 613. (k) Glasnov, T. N.; Kappe, C. O. Chem.Eur.
J. 2011, 17, 11956.
(6) For the biologically activity of tricyclic, nitrogen-containing
scaffolds, see: (a) Gillman, P. K. J. Pharmacol. 2007, 151, 737.
(b) Lednicer, D. In Strategies for Organic Drugs Synthesis and Design,
2nd ed.; WILEY-VCH: Hoboken, NJ, 2009. (c) Thansandote, P.;
Lautens, M. Chem.Eur. J. 2009, 15, 5874.
(7) For selected de novo syntheses of substituted dibenzazepines, see:
(a) Bergmann, E. D.; Shana, I.; Aisenshtat, Z. Tetrahedron Lett. 1968,
9, 3469. (b) Jorgensen, T. K.; Andersen, K. E.; Lau, J.; Madsen, P.;
Huusfeldt, P. O. J. Heterocycl. Chem. 1999, 36, 57. (c) Tsvelikhovsky,
D.; Buchwald, S. L. J. Am. Chem. Soc. 2010, 132, 14048.
(d) Christensen, H.; Schjøth-Eskesen, C.; Jensen, M.; Sinning, S.;
Jensen, H. H. Chem.Eur. J. 2011, 17, 10618. (e) Bozinovic,
́
N.;
̌
̌
Opsenica, I.; Solaja, B. A. Synlett 2013, 24, 49.
(8) While we have only two metals, the two phosphine ligands each
bind to palladium to form two catalytic systems, as evident from the
differences in product yields when both RuPhos and XPhos are used
versus only RuPhos.
(9) For examples of a three-catalyst/two-component reaction, see:
(a) Komon, Z. J. A.; Diamond, G. M.; Leclerc, M. K.; Murphy, V.;
Okazaki, M.; Bazan, G. C. J. Am. Chem. Soc. 2002, 124, 15280.
(b) Huff, C.; Sanford, M. S. J. Am. Chem. Soc. 2011, 133, 18122.
(10) Typical reaction conditions for RhI-catalyzed arylation:
Vinylpyridine, o-chloroboronic acid (1.2 equiv), [Rh(cod)OH]2 (2
mol%), K2CO3 (2 equiv), dioxane/H2O (10:1), 110 °C, 2 h. Yields are
between 81% and 91%.
(11) (a) Biscoe, M. R.; Fors, B. P.; Buchwald, S. L. J. Am. Chem. Soc.
2008, 130, 6686. (b) Fors, B. P.; Davis, N. R.; Buchwald, S. L. J. Am.
Chem. Soc. 2009, 131, 5766.
(12) Fors, B. P.; Buchwald, S. L. J. Am. Chem. Soc. 2010, 132, 15914.
113
dx.doi.org/10.1021/ol4030925 | Org. Lett. 2014, 16, 110−113