Tandem Long Distance Chain-Walking/Cyclization via RuH2(CO)(PPh3)3/Br?nsted Acid Catalysis: Entry to Aromatic Oxazaheterocycles

Article abstract of DOI:10.1021/acs.orglett.5b03499

A novel route to 1,3-oxazaheterocycles based on cooperative Ru-H/Br?nsted acid catalysis is reported. The use of the commercially available RuH₂(CO)(PPh₃)₃ complex allows for an efficient long distance chain-walking process while the Br?nsted acid is responsible for generation of an electrophilic iminium ion which is trapped intramolecularly by an alcohol moiety. The alcohol, besides its nucleophilic function, also plays an important role in the stabilization of the Ru catalyst. (Chemical Equation Presented).

Full text of DOI:10.1021/acs.orglett.5b03499

Letter
pubs.acs.org/OrgLett
Tandem Long Distance Chain-Walking/Cyclization via
RuH2(CO)(PPh3)3/Brønsted Acid Catalysis: Entry to Aromatic
Oxazaheterocycles
Rodrigo Bernardez, Jaime Suarez, Martín Fananas-Mastral, Jesus A. Varela, and Carlos Saa*
́
́
́
́
́
̃
Departamento de Química Organ
Universidad de Santiago de Compostela, 15782 Santiago de Compostela, Spain
́
ica y Centro Singular de Investigacion
́
en Química Biolog
́
ica y Materiales Moleculares (CIQUS),
S
* Supporting Information
ABSTRACT: A novel route to 1,3-oxazaheterocycles based
on cooperative Ru−H/Brønsted acid catalysis is reported. The
use of the commercially available RuH2(CO)(PPh3)3 complex
allows for an efficient long distance chain-walking process
while the Brønsted acid is responsible for generation of an
electrophilic iminium ion which is trapped intramolecularly by
an alcohol moiety. The alcohol, besides its nucleophilic
function, also plays an important role in the stabilization of the Ru catalyst.
Scheme 1. Cooperative Ru−H/Brønsted Acid Catalyzed
Cycloisomerizations
In situ formation of an electrophilic N-acyliminium or N-
sulfonyliminium ion and subsequent trapping with a nucleophile
is a well-known procedure for the synthesis of a wide range of
nitrogen-containing heterocycles.1 Isomerization of enamides
under acidic conditions represents a practical strategy for the
generation of an N-acyliminium ion intermediate. Since the
discovery of the self-condensation of enecarbamates by
Kobayashi,2a enamides have been used as electrophilic precursors
in Pictet−Spengler and Mannich-type reactions which lead to α-
functionalized amines.2
Enamides can be routinely obtained by metal-hydride
catalyzed double bond isomerization of easily available allylic
amides.3,4 Complexes of transition metals such as Ru, Rh, Fe, Co,
or Ir turned out to be active species for the catalytic double bond
isomerization of N- and O-allylic systems.3,5 Among all catalysts,
ruthenium hydride complexes have been particularly useful for
the isomerization of N-allylamides into the corresponding
enamides.6 In all these cases, the reaction is likely to occur
through an olefin coordination, migratory insertion, and β-
hydride elimination sequence.
In 2008 the Terada group reported a tandem isomerization/
C−C bond forming sequence by RuHCl(CO)(PPh3)3/Brønsted
acid cooperative catalysis.7 The method allows for the isomer-
ization of an N-allylamide into a reactive imine which
subsequently undergoes a Brønsted acid catalyzed Friedel−
Crafts type C−C bond forming reaction under relay catalysis.
Following this pioneering work, this methodology was
successfully applied in the synthesis of nitrogenated heterocycles
by intramolecular trapping of the electrophilic iminium ion
intermediate (X = N) with C and O nucleophiles (Scheme 1, eq
1).8,9 More recently, the process has also been extended to
generate the less stable oxocarbenium ions (X = O), starting in
this case from the corresponding allyl ethers.10
leading, in most cases, to a decrease in the reaction yield. Herein
we report that the N-acyl- or N-sulfonyliminium ions can be
efficiently generated by using the commercially available complex
RuH2(CO)(PPh3)3 as the catalyst for long distance chain-
walking olefin isomerization11 under Ru−H/Brønsted acid
cooperative catalysis. We have used this strategy to develop
catalytic methodology for the synthesis of benzoxazines 3
(Scheme 1, eq 2), interesting 1,3-oxazaheterocycles displaying a
wide range of biological activities.12 The reaction reported
proceeds in very good yields and high selectivity and has also
been extended to the synthesis of different oxazaheterocycles.
We began our investigation using N-(but-3-en-1-yl)-N-(2-
(hydroxymethyl)phenyl)-4-methylbenzenesulfonamide 1a as
the model substrate (Table 1).
We first tested Terada’s catalyst system, RuHCl(CO(PPh3)3,
which has shown good iminium ion generation from N-allylic
systems.7,8a Cycloisomerization of 1a to the corresponding 1,3-
Unfortunately, these synthetic approaches have mostly been
Received: December 9, 2015
limited to allyl derivatives, with the use of longer N-substituents
© XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.5b03499
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 1. Optimization of Catalysts and Protecting Group
Table 2. Chain-Walking Olefin Isomerization
a
b
entry
n
RuH2(CO)(PPh3)3 TsOH·H2O time (h) yield (%)
yield
a
b
entry
PG
Ru−H
acid
(%)
1
2
3
4
5
6
0, 1b
1, 1a
2, 1c
3, 1d
4, 1e
9, 1f
2 mol %
2 mol %
2 mol %
2 mol %
4 mol %
8 mol %
4 mol %
4 mol %
4 mol %
4 mol %
6 mol %
10 mol %
24
24
48
48
48
72
93, 3b
88, 3a
82, 3c
76, 3d
80, 3e
41, 3f
c
c
1
2
Ts, 1a RuHCl(CO)(PPh3)3
TsOH·H2O
TsOH·H2O
63, 3a
48, 3a
Ts, 1a [RuH(CO)
(PPh3)2(MeCN)2]BF4
3
Ts, 1a RuH2(CO)(PPh3)3
Ts, 1a RuH2(CO)(PPh3)3
TsOH·H2O
TsOH·H2O
TsOH·H2O
−
HBF4·OEt2
TsOH·H2O
TsOH·H2O
TsOH·H2O
TsOH·H2O
88, 3a
41, 3a
d
c
4
e
5
6
7
Ts, 1a
−
−
a
Conditions: 1a−f (0.25 mmol), RuH2(CO) (PPh3)3, and TsOH·
H2O in 0.5 mL of THF at 95 °C. Isolated yields.
Ts, 1a RuH2(CO)(PPh3)3
Ts, 1a RuH2(CO)(PPh3)3
Ts, 1a RuH2(CO)(PPh3)3
93, 2a
87, 3a
86, 3a
76, 7
b
f
8
loading of catalysts (Ru 8 mol %, TsOH 10 mol %) and a longer
reaction time (72 h) were needed (entry 6).
9
Ns, 4
Ms, 5
RuH2(CO)(PPh3)3
RuH2(CO)(PPh3)3
10
11
80, 8
We next set out to investigate the scope of the reaction with
respect to the alkene-bearing chain (Table 3).
Boc, 6 RuH2(CO)(PPh3)3
90, 9
a
Conditions: 1a (0.25 mmol), Ru−H (0.005 mmol), and Brønsted
acid (0.01 mmol) in 0.5 mL of THF at 95 °C, 24 h. Isolated yields.
N-Crotyl derivative, arising from a single olefin isomerization, was
also formed. 70 °C. Starting material recovered. Toluene as solvent.
b
c
Table 3. Variations in Alkene-Bearing Chain
d
e
f
benzoxazine 3a took place in the presence of TsOH·H2O (4 mol
%), although in moderate yield which arose from incomplete
olefin isomerization (entry 1). A lower yield was obtained with
the cationic ruthenium monohydride catalyst [RuH(CO-
(PPh3)2(MeCN)2]BF4 (entry 2). To our delight, the ruthenium
dihydride catalyst, RuH2(CO)(PPh3)3 (2 mol %),13 proved to be
more active under the same reaction conditions with cyclo-
isomerization occurring smoothly to give benzoxazine 3a in very
good yield (entry 3). The reaction temperature played a
significant role in the cycloisomerization process, with a slight
decrease of the temperature leading to a significant drop in the
reaction yield (entry 4). Both the ruthenium hydride and
Brønsted acid catalysts were crucial to trigger the cyclo-
isomerization process. Without the Ru catalyst, the starting
material 1a was recovered (entry 5). As expected, in the absence
of acid, the olefin isomerization smoothly occurred to give the
conjugated tosyl enamide 2a, but without subsequent cyclization
(entry 6). Other strong Brønsted acids (HBF4, TfOH) and Lewis
acids (BF3·OEt2, Sc(OTf)3, In(OTf)3) also proved to be efficient
catalysts for this transformation (entry 7; Table S2). Finally, a
survey of solvents showed that toluene could alternatively be
used without yield erosion (entry 8) while the use of chlorinated
solvents or DMF led to a significant decrease in catalyst activity
(Supporting Information (SI)). Our cycloisomerization reaction
could also be satisfactorily performed with other sulfonamide
protecting groups (entries 9−11), such as nosyl and methyl-
sulfonyl amides 4 and 5 (PG = Ns, Ms) in 76% and 80% yields,
respectively. 6 (PG = Boc) was also extremely well tolerated to
give benzoxazine 9 in 90% yield, without formation of any traces
of deprotected amine (entry 11).14
a
Conditions: 1g−j (0.25 mmol), RuH2(CO)(PPh3)3 (0.005 mmol),
and TsOH·H2O (0.01 mmol) in 0.5 mL of THF, 24 h at 95 °C.
b
c
Isolated yields. RuH2(CO) (PPh3)3 5 mol %, TsOH·H2O 7 mol %,
d
in 0.3 mL of toluene, 24 h at 110 °C. 1j (0.15 mmol),
RuH2(CO)(PPh3)3 10 mol %, TsOH·H2O 12 mol %, in 0.3 mL of
toluene, 72 h at 110 °C.
We found that both (E)- and (Z)-1,2-disubstituted alkenes
were competent substrates for this transformation as illustrated
by the synthesis of benzoxazine 3d (Table 3, entry 1). Pleasingly,
the more challenging styryl and conjugated ester derivatives 1h
and 1i were also able to cycloisomerize to the corresponding
benzoxazines 3h and 3i in good yield (entry 2), thus showing the
high activity of this catalytic system. A 1,1-disubstituted alkene,
such as 1j, was likewise compatible with this transformation,
although it showed a diminished reactivity, providing 3j in
moderate yield (entry 3).16
Long distance chain-walking olefin isomerization was analyzed
next (Table 2). As expected, the isomerization of N-allyl
derivative 1b (n = 0) was perfectly accomplished to afford the
benzoxazine 3b in an excellent 93% yield (entry 1). Gratifyingly,
bis-, tris-, and tetra-homoallyl derivatives 1c−e (n = 2, 3, and 4)
smoothly cycloisomerized to benzoxazines 3c−e in fairly good
yields (entries 3−5), although longer reaction times were
needed.15 To our delight, the cycloisomerization also took place
even in a remotely functionalized olefin 1f (n = 9), although extra
Finally, different types of oxygenated nucleophiles were
analyzed (Table 4). Thus, secondary (1k and 1l) and even
B
DOI: 10.1021/acs.orglett.5b03499
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Table 4. Scope of the Oxygenated Nucleophile
Scheme 2. Mechanistic Investigations
the isomeric enamide 17 (eq 5). Therefore, it seems that the
alcohol functionality (in this case, intermolecularly) acts as a
crucial ligand for the ruthenium catalysts while also preventing
the formation of inactive species in the presence of acid.17 On the
other hand, tosyl enamide 2a could be efficiently cycloisomerized
to benzoxazine 3a in the presence of catalytic amounts of TsOH·
H2O (eq 6), which means that both reaction steps, chain-walking
and cycloisomerization, are independently catalyzed by
ruthenium hydride complexes and by acid, respectively.
To account for the observed data, the following catalytic cycle
is proposed (Scheme 3). Following the formation of the active
Scheme 3. Proposed Catalytic Cycle
a
Conditions: Substrate (0.25 mmol), RuH2(CO)(PPh3)3 (2 mol %,
0.005 mmol), and TsOH·H2O (4 mol %, 0.01 mmol) in 0.5 mL of
b
c
THF, 24 h, 95 °C. Isolated yields. 1l or 1m (0.25 mmol),
d
RuH2(CO)(PPh3)3 5 mol %, TsOH·H2O 7 mol %, 110 °C. 12 (0.15
mmol), RuH2(CO)(PPh3)3 5 mol %, TsOH·H2O 7 mol %, 95 °C.
tertiary (1m) alcohols cycloisomerized to their corresponding
benzoxazines 3k−m in fairly good yields, albeit with modest
diastereoselectivities (entries 1−3). Phenol 10 was also an
efficient substrate and provided benzoxazine 13 in excellent yield
(entry 4). Gratifyingly, seven-membered benzoxazepines 14 and
15 could also be synthesized by this procedure by elongating the
hydroxyl chain (11, entry 5) or using a benzyl amide derivative
(12, entry 6).
Reactions performed with N-(but-3-en-1-yl)-N-phenyl-4-
methylbenzenesulfonamide 16, the parent dehydroxy derivative,
provided valuable insight into the mechanism of the reaction
(Scheme 2). The chain-walking process is triggered by the
catalyst RuH2CO(PPh3)3 without the need for external acid,
since 16 gave the isomeric enamide 17 (mixture of E/Z isomers)
in 83% yield after heating in THF at 95 °C for 24 h (eq 3). In
contrast, the reaction of 16 with the same catalyst, but in the
presence of catalytic amounts of TsOH, gave the isomeric N-
(but-2-en-1-yl)tosylamide 18 in 80% yield (mixture of E/Z
isomers), in which only one “step-walk” took place (eq 4).
Therefore, the presence of a protic acid is deleterious for the
chain-walking event possibly by forming inactive catalytic
ruthenium species. Interestingly, adding 1 equiv of benzylic
alcohol to the last mixture allowed the recovery of the catalytic
activity of the ruthenium hydride species giving again efficiently
Ru(II) hydride complex, Ru(H)X(CO)L3,13 coordination
through the alcohol unit of alkenylamide 1a facilitates the
chain-walking event (inter- or intramolecularly) to give enamide
2a with recovery of the catalytic Ru(II)−H species. Sub-
sequently, 2a enters into an independent acid-catalyzed cycle to
give the cycloisomerized benzoxazine 3a with recovery of the
proton carrier.18 On the other hand, in the absence of the alcohol
ligand and under acidic conditions, the Ru(H)X(CO)L3 catalyst
might be slowly deactivated to give, most likely, the Ru(II)
complex RuX2(CO)L3 which stops the chain-walking process. In
the absence of acid, the starting Ru(II) complex RuH2(CO)L3 is
able to isomerize the starting alkenylamide 1a to 2a through a
typical chain-walking isomerization process (iterative insertions
of an alkene into a Ru−H bond followed by β-elimination to give
a new Ru−H and an alkene until the formation of the more
thermodynamically stable olefin).
Preliminary studies on an enantioselective variant with a chiral
Brønsted acid of this tandem isomerization/cyclization showed
promising results (Scheme 4).19,20 It was found that the use of
C
DOI: 10.1021/acs.orglett.5b03499
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
Scheme 4. Enantioselective Cycloisomerization
REFERENCES
■
(1) (a) Maryanoff, B. E.; Zhang, H.-C.; Cohen, J. H.; Turchi, I. J.;
Maryanoff, C. A. Chem. Rev. 2004, 104, 1431. (b) Royer, J.; Bonin, M.;
Micouin, L. Chem. Rev. 2004, 104, 2311. (c) Kataja, A.; Masson, G. O.
Tetrahedron 2014, 70, 8783.
(2) (a) Kobayashi, S.; Gustafsson, T.; Shimizu, Y.; Kiyohara, H.;
Matsubara, R. Org. Lett. 2006, 8, 4923. (b) Terada, M.; Sorimachi, K. J.
Am. Chem. Soc. 2007, 129, 292. (c) Jia, Y.-X.; Zhong, J.; Zhu, S.-F.;
Zhang, C.-M.; Zhou, Q.-L. Angew. Chem., Int. Ed. 2007, 46, 5565.
(d) Terada, M.; Tanaka, H.; Sorimachi, K. Synlett 2008, 2008, 1661.
(3) (a) Escoubet, S.; Gastaldi, S.; Bertrand, M. Eur. J. Org. Chem. 2005,
2005, 3855. (b) Krompiec, S.; Krompiec, M.; Penczek, H.; Ignasiak, H.
Coord. Chem. Rev. 2008, 252, 1819.
(4) For a non-Ru−H catalyzed isomerization, see: Larsen, C. R.;
Grotjahn, D. B. J. Am. Chem. Soc. 2012, 134, 10357.
(5) Kuznik, N.; Krompiec, S. Coord. Chem. Rev. 2007, 251, 222.
(6) (a) Stille, J. K.; Becker, Y. J. Org. Chem. 1980, 45, 2139.
(b) Krompiec, S.; Pigulla, M.; Kuznik, N.; Krompiec, M.; Marciniec, B.;
Chadyniak, D.; Kasperczyk, J. J. Mol. Catal. A: Chem. 2005, 225, 91.
(c) Arisawa, M.; Terada, Y.; Takahashi, K.; Nakagawa, M.; Nishida, A. J.
Org. Chem. 2006, 71, 4255.
(7) Sorimachi, K.; Terada, M. J. Am. Chem. Soc. 2008, 130, 14452.
(8) (a) Toda, Y.; Terada, M. Synlett 2013, 24, 752. (b) Hansen, C. L.;
Clausen, J. W.; Ohm, R. G.; Ascic, E.; Le Quement, S. T.; Tanner, D.;
Nielsen, T. E. J. Org. Chem. 2013, 78, 12545. (c) For a review, see:
Ishoey, M.; Nielsen, T. E. Chem. - Eur. J. 2014, 20, 8832.
(9) For tandem cycloisomerizations catalyzed by Hoveyda−Grubbs
complexes, see: (a) Ascic, E.; Jensen, J. F.; Nielsen, T. E. Angew. Chem.,
Int. Ed. 2011, 50, 5188. (b) Cai, Q.; Liang, X.-W.; Wang, S.-G.; You, S.-L.
Org. Lett. 2012, 14, 5022. (c) Cai, Q.; Liang, X.-W.; Wang, S.-G.; You, S.-
L. Org. Biomol. Chem. 2013, 11, 1602. For RhCl(PPh3)3 catalyzed
cycloisomerizations: (d) Ascic, E.; Hansen, C. L.; Le Quement, S. T.;
Nielsen, T. E. Chem. Commun. 2012, 48, 3345.
(10) (a) Lombardo, V. M.; Thomas, C. D.; Scheidt, K. A. Angew. Chem.,
Int. Ed. 2013, 52, 12910. (b) Ascic, E.; Ohm, R. G.; Petersen, R.; Hansen,
M. R.; Hansen, C. L.; Madsen, D.; Tanner, D.; Nielsen, T. E. Chem. - Eur.
J. 2014, 20, 3297.
chiral phosphoric acid (R)-19 in combination with RuH2(CO)-
PPh3 (SI) gave rise to benzoxazine 3a with encouraging levels of
enantioselectivity (81:19 er). Interestingly, this phosphoric acid
also allowed the reaction to be performed under reduced
temperature (40 °C) with full conversion and selectivity.21 This
positive result paves the way for an asymmetric method to
synthesize 1,3-oxazaheterocycles and represents the first example
of an enantioselective tandem Ru−H/chiral Brønsted acid
catalyzed process in which an N-homoallyl amide is used.
In summary, we have reported a novel catalytic procedure for
the synthesis of six- and seven-membered 1,3-oxazaheterocycles
through a simple, redox-economical dual Ru−H/Brønsted acid
catalyzed cycloisomerization of 2-hydroxy(alkyl) substituted N-
alkenyl aniline and benzylamine derivatives. The use of the
commercially available RuH2(CO)(PPh3)3 complex enables an
efficient long-distance chain-walking process, thus allowing a
wide range of 2-alkylsubstituted 1,3-oxazaheterocycles to be
obtained in very good yields. We have found that the hydroxyl
unit, besides its nucleophilic function, plays also an important
role in the stabilization of the Ru catalyst. In addition, preliminary
studies show that 1,3-oxazaheterocycles can be obtained in an
enantioselective fashion by using a chiral phosphoric acid, which
provides space for further study of this asymmetric variant.
(11) For Pd−H catalyzed long distance chain walking of alkenes and
alkenyl alcohols, see: (a) Kochi, T.; Hamasaki, T.; Aoyama, Y.;
Kawasaki, J.; Kakiuchi, F. J. Am. Chem. Soc. 2012, 134, 16544.
(b) Larionov, E.; Lin, L.; Guenee, L.; Mazet, C. J. Am. Chem. Soc.
́ ́
2014, 136, 16882. (c) Mei, T.-S.; Patel, H. H.; Sigman, M. S. Nature
2014, 508, 340.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.orglett.5b03499.
(12) (a) Schwenker, G.; Chen, J. Arch. Pharm. 1991, 324, 891.
(b) Zhang, P.; Terefenko, E. A.; Fensome, A.; Zhang, Z.; Zhu, Y.;
Cohen, J.; Winneker, R.; Wrobel, J.; Yardley, J. Bioorg. Med. Chem. Lett.
2002, 12, 787. (c) Ouberai, M.; Asche, C.; Carrez, D.; Croisy, A.; Dumy,
P.; Demeunynck, M. Bioorg. Med. Chem. Lett. 2006, 16, 4641.
(13) Zbieg, J. R.; Yamaguchi, E.; McInturff, E.; Krische, M. J. Science
2012, 336, 324.
■
S
Experimental procedures, spectral data of all products
(PDF)
(14) Carboxybenzyl (Cbz) and acetyl (Ac) protected amides and failed
to give the corresponding cyclization products.
(15) Starting with 1c, isomerization of the N-4-penten-1-yl to N-[(E)-
3-penten-1-yl] isomer was observed when using nonhydride ruthenium
catalysts (only one “step-walk”). See ref 4.
AUTHOR INFORMATION
Corresponding Author
■
(16) No cycloisomerization occurred when a substrate containing a N-
(3-methyl-3-butenyl) chain was used.
*E-mail: carlos.saa@usc.es.
Notes
(17) For a related stabilization, see: Perdriau, S.; Chang, M.-C.; Otten,
E.; Heeres, H. J.; de Vries, J. G. Chem. - Eur. J. 2014, 20, 15434.
(18) For other catalytic combinations using lower amounts of a
Brønsted acid, incomplete cycloisomerizations were found (see SI).
(19) For a review on the use of chiral Brønsted acids in asymmetric
transformations, see: Parmar, D.; Sugiono, E.; Raja, S.; Rueping, M.
Chem. Rev. 2014, 114, 9047.
(20) For examples on enantioselective tandem ruthenium-hydride/
chiral Brønsted acid catalyzed cycloisomerizations of N-allyl derivatives,
see refs 8a, 8b, 9b, and 9c.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by MINECO (Spain) (Projects
CTQ2011-28258 and CTQ2014-59015R), Xunta de Galicia
(Project GRC2014/032), and the European Regional Develop-
ment Fund (Projects CTQ2014-59015R and GRC2014/032).
R.B. thanks MEC for a predoctoral FPU fellowship, and J.S.
thanks XUGA for a postdoctoral contract. M.F.-M. thanks
(21) Heating the enantiomeric mixture of 3a under the reaction
conditions for 24 h gave no epimerization (see SI).
́
MINECO for a “Ramon y Cajal” contract.
D
DOI: 10.1021/acs.orglett.5b03499
Org. Lett. XXXX, XXX, XXX−XXX