Domino rhodium-catalyzed alkyne arylation/palladium-catalyzed N arylation: A mechanistic investigation

Article abstract of DOI:10.1002/anie.201103692

It takes two to tango: A domino multicatalytic synthesis of dihydroquinolines is realized, wherein the products of a rhodium-catalyzed arylation of alkynes are cyclized by a palladium-promoted C-N cross-coupling (see scheme). The combination of catalysts

Full text of DOI:10.1002/anie.201103692

DOI: 10.1002/anie.201103692
Multicatalytic Reactions
Domino Rhodium-Catalyzed Alkyne Arylation/Palladium-Catalyzed
N Arylation: A Mechanistic Investigation**
Jane Panteleev, Lei Zhang, and Mark Lautens*
Domino processes, in which one reagent or catalyst promotes
a series of reactions in a defined order, have become
increasingly used in organic synthesis.[1] In the majority of
cases, the functionality produced from the first step makes the
subsequent step possible, and as a result, single products are
formed rather than complex mixtures. In recent years, the
idea of combining two different catalysts in a single vessel to
promote domino reactions has become more common.[2]
Organo-/metal catalysis, sequential biocatalysis, bio-/metal
catalysis, and metal/metal catalysis have all been reported.[2,3]
In the majority of examples in which two metals catalyze
sequential processes, the presence of the second metal does
not expand the range of transformations that are possible.[3]
We now report an example in which two different metals with
two different phosphine ligands promote two out of three
possible reactions and do so in a time-resolved way, thus
leading to an efficient preparation of dihydroquinolines.[4]
Importantly, we show that one of the two phosphine ligands
is selective in binding to only one of the metals, and the other
ligand is nondiscriminating in its binding properties; yet one
of the two metal–phosphine complexes is far more reactive
and selectivity is observed. This study raised the more general
question of matching ligands and metals in order to efficiently
promote only the desired reaction rather than competing
processes. Thus, complex sequences of reactions that contain
multiple catalytically active species can ultimately be created
in a single reaction vessel.
We began our investigation by optimizing the individual
steps of the sequence (Table 1). Racemic binap was found to
be the optimal ligand for the alkyne arylation of 1, resulting in
product 3 in 60–80% yield (Table 1, entries 3–5).[6,8] Addition
of methanol to the reaction was crucial in improving the yield
and regioselectivity (> 20:1 versus 10:1; Table 1, entries 4 and
5). The methanesulfonyl protecting group gave the best yields
À
(Table 1, entry 5). When we examined the C N bond
Table 1: Optimization of the two-step dihydroquinoline synthesis.[a]
Entry Step Ligand
Additive T [oC] Pg 3 [%][b] 4 [%][b]
1
2
3
4
1
1
1
1
1
2
2
2
2
2
dppp
P(4-FC6H4)3
binap
binap
binap
Xantphos
Davephos
X-Phos
X-Phos
X-Phos
H2O
H2O
H2O
MeOH
MeOH
–
60
60
60
60
60
100
100
100
90
90
90
Ts (24)
–
–
–
–
–
Ts
Ts
Ts
Ms
54
Our work in this area arose from our studies on the
rhodium-catalyzed addition of boronic acids to pyridinyl
alkenes and alkynes.[5] We saw the potential to construct
heterocyclic motifs 4 by starting from appropriately substituted
alkynes 1 [Eq. (1)]. Herein, we report the realization of this idea
through a successful combination of a two-metal/two-ligand
catalyst system, featuring a rhodium-catalyzed alkyne arylation
63[c]
67[d]
77[d]
5
6[e]
7[e]
8[e]
9
Ts (66)
Ts (47)
–
–
–
(6)
70
83
92
69
–
Ts
Ts
Ms
Ms
Ms
0
0
0
8
–
MeOH
MeOH
[6,7]
10
11[f]
12
13
14
À
and palladium-catalyzed C N coupling.
This synthetic
1+2 binap + X-Phos MeOH
sequence proceeds in one vessel, in which both catalysts exist
concurrently and display correct reactivity even though other
reaction pathways, including a Suzuki reaction, are available.
[g]
1
1
2
no ligand
X-Phos
binap
MeOH
MeOH
MeOH
60
90
90
Ms (5)
Ms (84)
–
(4)
[a] See Table 3 for representative reaction conditions. [b] Yields of
isolated products; yields in parentheses indicate yields determined by
NMR spectroscopy. [c] 10:1 regioisomeric ratio (r.r.) determined by NMR
spectroscopy of the crude product. In the minor isomer, arylboronic acid
is added to the opposite end of the alkyne. [d]>20:1 r.r. [e] Cs2CO3
(1.5 equiv) was used. [f] Domino reaction using K2CO3 (2.2 equiv). [g] A
complex mixture of by-products was observed. binap=racemic 2,2’-
bis(diphenylphosphino)-1,1’-binaphthyl, cod=1,5-cyclooctadiene,
DavePhos=2-dicyclohexylphosphino-2’-(N,N-dimethylamino)biphenyl,
dppp=propane-1,3-diylbis(diphenylphosphane), Ms=methanesul-
fonyl, Ts=toluene-4-sulfonyl, X-Phos=2-dicyclohexylphosphino-2’,4’,6’-
triisopropylbiphenyl, XantPhos=4,5-bis(diphenylphosphino)-9,9-dime-
thylxanthene.
[*] J. Panteleev, L. Zhang, Prof. Dr. M. Lautens
Davenport Research Laboratories
Department of Chemistry, University of Toronto
80 St George St. Toronto, Ontario, M5S 3H6 (Canada)
E-mail: mlautens@chem.utoronto.ca
[**] The authors wish to thank the Natural Sciences and Engineering
Research Council (NSERC), Merck for an Industrial Chair, and the
University of Toronto for financial support. J.P. thanks NSERC for a
CGSD scholarship.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201103692.
Angew. Chem. Int. Ed. 2011, 50, 9089 –9092
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9089

Communications
formation we obtained excellent yields (92%) of 4 by using
the X-Phos ligand (Table 1, entries 8–10).[9] Notably, both of
the reactions proceeded in 1,4-dioxane as the solvent when
using a weak carbonate base.
We were interested in combining the two steps in a single
reaction vessel, and in order to simplify the system, the
ligand–metal solutions were prepared separately prior to the
addition of the substrates.[10] To our gratification, the overall
transformation proceeded in a combined yield that was
similar to the two independent reaction steps (69% versus
71%; Table 1, entry 11). Examples of domino reactions
catalyzed by two transition metals are rare, thus we elected
to further investigate the interactions between the two
catalyst systems.
To survey the reactivity of the possible metal–ligand
combinations, we carried out several control experiments on
each step [Table 1, Eq. (2)]. Use of phosphine-free [Rh-
(cod)OH]2 led primarily to the decomposition of 1a (Table 1,
entry 12). The arylation reaction using X-Phos as a ligand led
to a low yield of 3a (5%) and a significant amount of
decomposition (40%; Table 1, entry 13). Subjecting substrate
1a to only [Pd(binap)] or [Pd(X-Phos)] did not lead to 3a or
4a; instead, [Pd(X-Phos)] furnished the Suzuki cross-coupling
product 5 in good yield [76%; Eq. (2)].[11,12] The reaction of 3a
with [Pd(binap)] yielded only trace amounts of 4a (Table 1,
entry 14). From these control experiments it was apparent
that use of [Rh(binap)] resulted in the desired product, and
phosphine-free rhodium led to decomposition of 1a. The
ability to form [Pd(X-Phos)] was equally important since
Figure 1. Effect of step 2 components on step 1 conversion. L2 =
X-Phos.[14]
in the optimized reaction, thus indicating that even if an
excess of the aryl boronic acid is used, the Suzuki coupling is
À
slower than the intramolecular C N coupling in 3a.
À
Similarly, we examined the C N coupling step (Figure 2).
While 92% yield of product was achieved when using [Pd(X-
Phos)], only 4% of product was obtained when [Pd(binap)]
was utilized (Table 1, entry 14). This observation had a
significant implication on the domino process, considering
that any ligand interchange between rhodium and palladium
À
[Pd(binap)] was catalytically inactive in C N coupling.
À
could be deleterious to the C N bond formation. In fact,
when the reaction was carried out using a premixed [Pd(X-
Phos)] catalyst solution with 5 mol% of binap added, the
yield plummeted from 92% to 3% (Figure 2, entry 2). When
we looked at the reaction of 3a with 5 mol% of [Rh(binap)] as
the additive, the yield returned to 91% (Figure 2, entry 4).
With higher rhodium loading (10 mol%, Figure 2, entry 5) a
decrease in conversion was observed, possibly resulting from
binap inhibition.
Ligand interference was further confirmed when we
examined the Rh/Pd ratio in the domino reaction
(Figure 3). When increasing the loading of the [Rh(binap)]
catalyst, progressively less product 4a was formed and the
reaction stalled at the intermediate stage. Varying the ligand
Both reactions utilize specialized phosphine ligands,
which have the potential to exchange between the metal
centers. We observed that Pd(OAc)2 forms discrete com-
plexes with both binap and X-Phos.[13] Based on 31P NMR
spectroscopy, it was apparent that [Pd(binap)] and [Pd(X-
Phos)] exist in an equilibrium. While rhodium is known to
bind binap, no complexation of rhodium and X-Phos was
indicated by 31P NMR spectroscopy.[10,13] Our NMR experi-
ments suggested that palladium could bind both phosphine
ligands, however rhodium could only complex binap.
We further examined the effect of Pd(OAc)2 or X-Phos
addition on the rate of the alkyne arylation reaction
(Figure 1). When the reaction was carried out with X-Phos
present in the catalyst solution, a minimal effect on the rate of
the reaction was observed (Figure 1, entry 2 versus entry 1).
In contrast, addition of 5 mol% of Pd(OAc)2 led to the
formation of only a trace of 3a (5%; Figure 1, entry 3), while
the conversion of 1a returned to more than 70% with the
addition of 5 mol% of [Pd(X-Phos)] solution to the reaction
mixture (Figure 1, entry 4). With a higher [Pd(X-Phos)]
loading, the Suzuki product 5 was formed in substantial
amounts (Figure 1, entry 5). Interestingly, no Suzuki products
containing the arylated alkene, such as 6, were ever observed
Figure 2. Effect of step 1 components on step 2 yield. L1 =binap.[14]
9090 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 9089 –9092

Figure 3. Effect of the Rh/Pd Ratio on the yield of the reaction.[14]
1.05 equivalents of binap with respect to [Rh] were used.
equivalents showed similar trends (Table 2). With binap
loading exceeding 5.5 mol%, the reaction stalled at the
intermediate stage (Table 2, entries 1–4). Addition of an
excess of X-Phos did not remedy this effect; in fact, further
equivalents of X-Phos resulted in lower yields (Table 2,
entries 4–6). While counterintuitive at first, this result is
consistent with the inverse rate dependence that is often
observed with bulky phosphine ligands and is caused by
saturation of the coordination sites on the palladium
center.[15]
Scheme 1. Proposed mechanism of the domino dihydroquinoline
synthesis.
The experiments described above gave us some insight
into the mechanism of the reaction (Scheme 1). It appears
that the two catalytic cycles occur independently, and no
direct interaction between the two active metal complexes
exists. Initially, substrate 1 reacts rapidly under rhodium
catalysis to yield 3, which can then cyclize through a
The domino reaction proceeded in an overall yield of
69%, in comparison to the two-step combined yield of 71%.
However, isolation of the intermediate in good yield and
purity was often problematic because of the presence of
minor by-products. Preliminary investigations of the scope
show that electron-rich and electron-poor aryl and heteroaryl
boronic acids could be used to give good overall yields
(Table 3). Performing the reaction in two-stage heating (608C
for 1.5 h, then 908C) had a subtle beneficial effect (81%
versus 78%; Table 3, entries 5 and 6). We observed signifi-
cantly higher yields when utilizing 3-thiophenylboronic acid
2b as the nucleophile (Table 3, entry 1 versus entry 2). A
competition experiment indicated that this boronic acid is
approximately five times more reactive than 2a.[16]
Domino reactions try to address the need for less wasteful
and more time- and cost-efficient synthetic pathways. Cur-
rently, the use of multiple transition-metal-catalyzed trans-
formations in domino processes is rare. A domino multi-
catalytic synthesis of dihydroquinolines was realized, wherein
the products of a rhodium-catalyzed arylation were cyclized
À
palladium-catalyzed C N coupling to result in product 4.
An alternative reaction pathway occurs where 1 undergoes
Suzuki cross-coupling, but with optimized catalyst ratios this
À
pathway is suppressed. Additionally, inhibition of the C N
coupling by binap is observed. We propose that the [Rh-
(binap)OH]2 solution is a source of trace amounts of free
binap, but Rh/binap binding largely remedies the inhibitory
effect of the free ligand.
Table 2: Effect of ligand loading on the domino reaction.[a]
Entry
binap
[mol%]
X-Phos
[mol%]
3a [%][b]
4a [%][b]
3a:4a
À
by palladium-promoted C N coupling. While some inhibition
À
of the palladium-catalyzed C N coupling by the rhodium
1
2
3
4
5
6
7
5.2
5.5
7.5
10
10
10
4
4
4
4
2
6
8
6
69
47
33
27
36
21
6
1:11
arylation ligand was observed, the domino reaction pro-
ceeded in equivalent yields as the independent reaction steps.
Our work provides a rare example of a system where two
transition-metal complexes with different phosphine ligands
capable of dissociation function along a desired pathway, even
when other reaction pathways are available. We look forward
to a more thorough investigation of this concept and its
application to other synthetic systems.
22
38
55
47
61
75
1:2.1
1.2:1
2:1
1.3:1
2.9:1
12.5:1
10
[a] See Table 3 for reaction conditions. [b] Yields determined by NMR
spectroscopy using 1,3,5-trimethoxybenzene as internal standard.
Angew. Chem. Int. Ed. 2011, 50, 9089 –9092
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9091

Communications
Table 3: Preliminary scope of the domino reaction.[a]
Komon, G. M. Diamond, M. K. Leclerc, V. Murphy, M. Okazaki,
G. C. Bazan, J. Am. Chem. Soc. 2002, 124, 15280 – 15285; c) S.
Ko, C. Lee, M.-G. Choi, Y. Na, S. Chang, J. Org. Chem. 2003, 68,
1607 – 1610; d) J. Cossy, F. Bargiggia, S. BouzBouz, Org. Lett.
2003, 5, 459 – 462; e) Y. Nishibayashi, M. Yoshikawa, Y. Inada,
M. Hidai, S. Uemura, J. Am. Chem. Soc. 2004, 126, 16066 –
16072; f) A. S. Goldman, A. H. Roy, Z. Huang, R. Ahuja, W.
Schinski, M. Brookhart, Science 2006, 312, 257 – 261; g) C.
Kammerer, G. Prestat, T. Gaillard, D. Madec, G. Poli, Org.
Lett. 2008, 10, 405 – 408; h) T. A. Cernak, T. H. Lambert, J. Am.
Chem. Soc. 2009, 131, 3124 – 3125; i) K. Takahashi, M. Yama-
shita, T. Ichihara, K. Nakano, K. Nozaki, Angew. Chem. 2010,
122, 4590 – 4592; Angew. Chem. Int. Ed. 2010, 49, 4488 – 4490.
For selected examples of multimetallic cooperative catalysis,
see: j) L. J. Gooßen, G. Deng, L. M. Levy, Science 2006, 313,
662 – 664; k) J. Z. Deng, D. V. Paone, A. T. Ginnetti, H. Kur-
ihara, S. D. Dreher, S. A. Weissman, S. R. Stauffer, C. S. Burgey,
Org. Lett. 2009, 11, 345 – 347; l) J. Huang, J. Chan, Y. Chen, C. J.
Borths, K. D. Baucom, R. D. Larsen, M. M. Faul, J. Am. Chem.
Soc. 2010, 132, 3674 – 3675; m) B. M. Trost, X. Luan, J. Am.
Chem. Soc. 2011, 133, 1706 – 1709.
Entry Product
Yield Entry
[%][b]
Product Yield
[%][b]
5
78
81
1
69
6[c]
2
3
78
68
7
8
70
73
[4] For a recent example of a one-metal two-ligand catalytic system,
see: B. P. Fors, S. L. Buchwald, J. Am. Chem. Soc. 2010, 132,
15914 – 15917.
[5] a) M. Lautens, A. Roy, K. Fukuoka, K. Fagnou, B. Martin-
Matute, J. Am. Chem. Soc. 2001, 123, 5358 – 5359; b) M. Lautens,
M. Yoshida, J. Org. Chem. 2003, 68, 762 – 769; c) M. Lautens, M.
Yoshida, Org. Lett. 2002, 4, 123 – 125; d) G. C. Tsui, M. Lautens,
Angew. Chem. 2010, 122, 9122 – 9125; Angew. Chem. Int. Ed.
2010, 49, 8938 – 8941.
[6] a) T. Hayashi, K. Inoue, N. Taniguchi, M. Ogasawara, J. Am.
Chem. Soc. 2001, 123, 9918 – 9919; b) A. Acardi, M. Aschi, M.
Chiarini, G. Ferrara, F. Marinelli, Adv. Synth. Catal. 2010, 352,
493 – 498.
4
65
9
45
[a] Stock catalyst solutions ([Rh]2 (0.005m) with BINAP (1.05 equiv with
respect to [Rh]), and Pd(OAc)2 (0.008m) with X-Phos (2 equiv with
respect to [Pd])), were mixed separately in dioxane at 508C for 15 min.
0.5 mL of each solution was added to a vial containing 1 (0.2 mmol,
49 mg), 2 (0.4 mmol, 49 mg), K2CO3 (61 mg), and MeOH (100 mL) in
dioxane (1 mL). The mixture was stirred at 908C for 16 h. [b] Yield of
isolated product for the domino process. [c] Reaction carried out at 608C
over 1.5 h, then at 908C for 14.5 h.
À
[7] For reviews on C N coupling, see: a) D. Prim, J. M. Campagne,
D. Joseph, B. Andrioletti, Tetrahedron 2002, 58, 2041 – 2075;
b) B. Schlummer, U. Scholz, Adv. Synth. Catal. 2004, 346, 1599 –
1626.
[8] Optimization studies of the rhodium-catalyzed arylation show
that some decomposition of ortho-halogenated aryl-alkynes
occurs under the reaction conditions. This behavior accounts
for the remainder of the mass balance of the reaction.
Received: May 30, 2011
Published online: August 24, 2011
À
Keywords: chemoselectivity · dihydroquinolines ·
domino catalysis · heterocycles · multicatalytic reactions
[9] For the use of X-Phos in C N coupling, see: a) K. W. Anderson,
.
R. E. Tundel, T. Ikawa, R. A. Altman, S. L. Buchwald, Angew.
Chem. 2006, 118, 6673 – 6677; Angew. Chem. Int. Ed. 2006, 45,
6523 – 6527; b) M. D. Charles, P. Schultz, S. L. Buchwald, Org.
Lett. 2005, 7, 3965 – 3968; c) E. R. Strieter, D. G. Blackmond,
S. L. Buchwald, J. Am. Chem. Soc. 2003, 125, 13978 – 13980.
[10] For formation of [Rh(binap)OH]2, see: T. Hayashi, M. Takaha-
shi, Y. Takaya, M. Ogasawara, J. Am. Chem. Soc. 2002, 124,
5052 – 5058.
[1] L. F. Tietze, G. Brasche, K. Gericke, Domino Reactions in
Organic Synthesis, Wiley-VCH, Weinheim, 2006.
[2] For reviews on combinations of multiple catalytic reactions into
domino processes, see: a) A. Bruggink, R. Schoevaart, T.
Kieboom, Org. Process Res. Dev. 2003, 7, 622 – 640; b) J. M.
Lee, Y. Na, H. Han, S. Chang, Chem. Soc. Rev. 2004, 33, 302 –
312; c) J.-C. Wasilke, S. J. Obrey, R. T. Baker, G. C. Bazan,
Chem. Rev. 2005, 105, 1001 – 1020; d) L. M. Ambrosini, T. H.
Lambert, ChemCatChem 2010, 2, 1373 – 1380; e) D. B. Ramach-
ary, S. Jain, Org. Biomol. Chem. 2011, 9, 1277 – 1300; For
transition metal and biocatlaysts, see: f) O. Pamies, J.-E. Back-
vall, Chem. Rev. 2003, 103, 3247 – 3261. For combinations of
organocatalysts and transition metal catlysts, see: g) Y. J. Park,
J.-W. Park, C.-H. Jun, Acc. Chem. Res. 2008, 41, 222 – 234; h) Z.
Shao, H. Zhang, Chem. Soc. Rev. 2009, 38, 2745 – 2755; i) M.
Rueping, R. M. Koenigs, I. Atodiresei, Chem. Eur. J. 2010, 16,
9350 – 9365; j) C. Zhong, X. Shi, Eur. J. Org. Chem. 2010, 2999 –
3025.
À
À
[11] For the use of Pd/binap in C N/C O coupling, see: a) M.
Palucki, J. P. Wolfe, S. L. Buchwald, J. Am. Chem. Soc. 1996, 118,
10333 – 10334; b) S. Shekhar, P. Ryberg, J. F. Hartwig, J. S.
Mathew, D. G. Blackmond, E. R. Strieter, S. L. Buchwald, J.
Am. Chem. Soc. 2006, 128, 3584 – 3591.
[12] For the use of Pd/X-Phos in Suzuki reactions, see: T. E. Barder,
S. D. Walker, J. R. Martinelli, S. L. Buchwald, J. Am. Chem. Soc.
2005, 127, 4685 – 4696.
[13] See the Supporting Information for 31P NMR spectra; in
addition, in situ hydrogenation of the cod ligand with H2 did
not yield any observable Rh/X-Phos complexation.
[14] Reactions were carried out after premixing the Rh/binap and the
Pd/X-Phos catalyst solutions separately for 15 min at 508C.
[15] L. M. Alcazar-Roman, J. F. Hartwig, J. Am. Chem. Soc. 2001,
123, 12905 – 12906.
[3] For selected examples of the use of two transition-metal catalysts
in a domino transformation, see: a) N. Jeong, S. D. Seo, J. Y.
Shin, J. Am. Chem. Soc. 2000, 122, 10220 – 10221; b) Z. J. A.
[16] See the Supporting Information.
9092 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 9089 –9092