Expedient synthesis of chiral oxazolidinone scaffolds via rhodium-catalyzed asymmetric ring-opening with sodium cyanate

Article abstract of DOI:10.1021/ol4000668

A method for synthesizing chiral oxazolidinone scaffolds from readily available oxabicyclic alkenes is described. The reaction utilizes a domino sequence of Rh(I)-catalyzed asymmetric ring-opening (ARO) with sodium cyanate as a novel nucleophile followed by intramolecular cyclization to generate oxazolidinone products in excellent enantioselectivities (trans stereochemistry).

Full text of DOI:10.1021/ol4000668

ORGANIC
LETTERS
2013
Vol. 15, No. 5
1064–1067
Expedient Synthesis of Chiral
Oxazolidinone Scaffolds via
Rhodium-Catalyzed Asymmetric
Ring-Opening with Sodium
Cyanate
Gavin Chit Tsui, Nina M. Ninnemann, Akihito Hosotani, and Mark Lautens*
Davenport Research Laboratories, Department of Chemistry, University of Toronto,
80 St. George Street, Toronto, Ontario, Canada M5S 3H6
mlautens@chem.utoronto.ca
Received January 9, 2013
ABSTRACT
A method for synthesizing chiral oxazolidinone scaffolds from readily available oxabicyclic alkenes is described. The reaction utilizes a domino
sequence of Rh(I)-catalyzed asymmetric ring-opening (ARO) with sodium cyanate as a novel nucleophile followed by intramolecular cyclization to
generate oxazolidinone products in excellent enantioselectivities (trans stereochemistry).
The use of the cyanate anion in transition-metal-
catalyzed processes has attracted increasing attention. To
date, there haveonlybeena few reports on Ni-,1a Cu-1b and
Pd1c-catalyzed reactions with metal cyanates. For exam-
ple, Buchwald recently reported an efficient cross-coupling
of aryl chlorides and triflates with sodium cyanate to
generate aryl isocyanate intermediates toward the syn-
thesis of unsymmetrical ureas.1c Isocyanates are versatile
intermediates in organic synthesis. They are precursors
to carbamates and ureas by nucleophilic additions with
alcohols and amines. They also participate in various
cycloadditions to generate heterocycles.2 The synthesis
of isocyanates is usually difficult and involves the use of
highly toxic and hazardous reagents.3 However, transition-
metal-catalyzed CꢀN bond formation with metal cyanates
offers more efficient and safe synthetic routes to iso-
cyanate products. To the best of our knowledge, there
has been no example of using cyanates as reagents in
rhodium-catalyzed reactions, particularly in asymmetric
transformations.
We decided to explore the possibility of using metal
cyanate as a nucleophile in Rh(I)-catalyzed asymmetric
ring-opening (ARO) reactions. Over the past decade,
Rh(I)-catalyzed ARO of oxa- and azabicyclic alkenes with
heteroatom nucleophiles has stood out as a selective and
reliable process to generate chiral functionalized hydro-
naphthalene cores and biologically active compounds.4ꢀ6
A mechanistic working model has been established to
account for the formation of the ring-opened products
by an SN20 nucleophilic displacement of the bridgehead
(1) (a) Tkatchenko, I.; Jaouhari, R.; Bonnet, M.; Dawkins, G.;
Lecolier, S. France Patent FR2575467, 1986; U.S. Patent 4,749,806,
1988. (b) Kianmehr, E.; Baghersad, M. H. Adv. Synth. Catal. 2011, 353,
2599. (c) Vinogradova, E. V.; Fors, B. P.; Buchwald, S. L. J. Am. Chem.
Soc. 2012, 134, 11132.
(2) For selected recent examples of applications of isocyanates, see:
(a) Hesp, K. D.; Bergman, R. G.; Ellman, J. A. J. Am. Chem. Soc. 2011,
133, 11430. (b) Oberg, K. M.; Rovis, T. J. Am. Chem. Soc. 2011, 133,
4785. (c) Komine, Y.; Tanaka, K. Org. Lett. 2010, 12, 1312.
(3) For reviews on the synthesis of isocyanates, see: (a) Ozaki, S.
Chem. Rev. 1972, 72, 457. (b) Pal, P. Synlett 2012, 23, 2291.
(4) For reviews, see: (a) Fagnou, K.; Lautens, M. Chem. Rev. 2003,
103, 169. (b) Lautens, M.; Fagnou, K.; Hiebert, S. Acc. Chem. Res. 2003,
36, 48. (c) Hayashi, T.; Yamasaki, K. Chem. Rev. 2003, 103, 2829.
(d) Rayabarapu, D. K.; Cheng, C.-H. Acc. Chem. Res. 2007, 40, 971.
(e) Modern Rhodium-Catalyzed Organic Reactions; Evans, P. A., Ed.;
Wiley-VCH: Weinheim, Germany, 2005.
r
10.1021/ol4000668
Published on Web 02/08/2013
2013 American Chemical Society

leaving group with inversion to give the 1,2-trans products
as a single regio- and diastereomer.7
80 °C using catalytic [Rh(COD)Cl]2/(R,S)-PPF-PtBu2
10
and triethylamine hydrochloride as a proton source
We envisioned that by using the cyanate anion as
a nucleophile in Rh(I)-catalyzed ARO with oxabicyclic
alkene 1, chiral oxazolidinone product 3 would be ob-
tained enantioselectively in a domino ARO/cyclization
sequence via the formation of the isocyanate intermediate
2 (Scheme 1). Chiral oxazolidinones have been widely used
as chiral auxiliaries in asymmetric transformations (Evans’
chiral auxiliaries) and successfully employed in stereose-
lective syntheses of natural products and biologically
active compounds.8
(Table 1).11
Table 1. Optimization of Rh(I)-catalyzed ARO of
Oxabenzonorbornadiene 1a with Sodium Cyanatea
entry Rh catalyst
solvent
additive yieldb (%) eec (%)
1e
2e
3e
4e
5
6e
7e
8e
9e
[Rh(COD)Cl]2 THF
Et3N HCl
17d
42
3
Scheme 1. Using Metal Cyanate in a Domino Rh(I)-catalyzed
ARO/Cyclization Sequence to Synthesize Chiral Oxazolidinone
[Rh(COD)Cl]2 THF/H2O
Et3N HCl
40
41
70
95
93
95
91
92
97
98
99
99
3
3
3
3
[Rh(COD)Cl]2 dioxane/H2O Et3N HCl
33
[Rh(COD)Cl]2 toluene/H2O Et3N HCl
25
[Rh(COD)Cl]2 MeCN/H2O Et3N HCl
[Rh(COD)Cl]2 DCEg/H2O Et3N HCl
30
28
3
[Rh(COD)I]2
DCE/H2O
Et3N HCl
40
3
[Rh(COD)OH]2 DCE/H2O
Rh(COD)2OTf DCE/H2O
Et3N HCl
44
3
Et3N HCl
62
3
10f Rh(COD)2OTf DCE/H2O
11f,h Rh(COD)2OTf DCE/H2O
12f Rh(COD)2OTf DCE
Et3N HCl
66
3
Et3N HCl
69
3
Et3N HCl
57
3
13f Rh(COD)2OTf DCE/H2O
14f Rh(COD)2OTf DCE/H2O
15f Rh(COD)2OTf DCE/H2O
16f Rh(COD)2OTf DCE/H2O
none
28
TFEi
CSAj
tBuOH
<5d
38d
14d
<5d
17f,k none
DCE/H2O
Et3N HCl
3
Some potential challenges included finding suitable cat-
alytic conditions due to the requirement of a protic additive
for catalyst turnover and solubility issues of the metal
cyanate salt in organic solvents. The relative stereochem-
istry of the product that would be formed was also not
easily predicted since the interaction between the cyanate
anion and Rh in the ring-opening process was unknown.9
We began our studies by reacting oxabenzonorbornadiene
1a with excess sodium cyanate in THF or THF/H2O at
a Unless specified otherwise, reactions were run with 1a (0.2 mmol),
Rh catalyst (8 mol % Rh), (R,S)-PPF-PtBu2 (8 mol %), sodium cyanate
(5.0 equiv), additive (5.0 equiv), in organic solvent/H2O (10:1) (0.1 M), at
80 °C for 1ꢀ4 h or at rt for 17 h, under argon atmosphere. b Isolated
yield. c Determined by chiral HPLC analysis. d Determined by 1H NMR
spectroscopy of the crude material. e Reaction was run at 80 °C.
f Reaction was run at rt. g DCE = 1,2-dichloroethane. h Used 2.3 equiv
of sodium cyanate. i TFE = 2,2,2-trifluoroethanol. j CSA = camphor-
sulfonic acid. k Reaction was run without the catalyst and ligand, only
starting material was recovered.
(5) Rh(I)-catalyzed ARO with alcohols and amines: (a) Lautens, M.;
Fagnou, K.; Rovis, T. J. Am. Chem. Soc. 2000, 122, 5650. (b) Lautens,
M.; Fagnou, K.; Taylor, M.; Rovis, T. J. Organomet. Chem. 2001, 624,
259. (c) Lautens, M.; Fagnou, K. J. Am. Chem. Soc. 2001, 123, 7170. (d)
Lautens, M.; Fagnou, K.; Yang, D. J. Am. Chem. Soc. 2003, 125, 14884.
Phenols: (e) Lautens, M.; Fagnou, K.; Taylor, M. Org. Lett. 2000, 2,
1677. Carboxylates: (f) Lautens, M.; Fagnou, K. Tetrahedron 2001, 57,
5067. Thiols: (g) Leong, P.; Lautens, M. J. Org. Chem. 2004, 69, 2194.
Water: (h) Tsui, G. C.; Lautens, M. Angew. Chem., Int. Ed. 2012, 51,
5400. Triethylamine trihydrofluoride: (i) Zhu, J.; Tsui, G. C.; Lautens,
M. Angew. Chem., Int. Ed. 2012, 51, 12353.
(6) Application of Rh(I)-catalyzed ARO in the synthesis of analgesic
compounds: (a) Lautens, M.; Fagnou, K.; Zunic, V. Org. Lett. 2002, 4,
3465. Rotigotine and (S)-8-OH-DPAT: (b) Webster, R.; Boyer, A.;
Fleming, M. J.; Lautens, M. Org. Lett. 2010, 12, 5418.
(7) Lautens, M.; Fagnou, K. Proc. Natl. Acad. Sci. U.S.A. 2004, 101,
5455.
A reaction in aqueous THF (10:1 ratio of THF:H2O)
gave the desired product 3a in 42% isolated yield (>90%
conversion) along with 26% of 1-naphthol from the decom-
position of substrate 1a (entry 2). Using potassium cyanate
gave a similar yield (38%). Added water enhanced the yield
(cf. entry 1). However, the enantioselectivity of the product
was disappointingly low (40% ee). Subsequent screening of
solvents showed a strong influence of the nature of the
solvent on enantioselectivity in the following order: THF ≈
dioxane < toluene < MeCN ≈ 1,2-dichloroethane (DCE)
(entries 3ꢀ6). Very high enantioselectivities (93ꢀ95% ee)
could now be achieved when reactions were run in MeCN/
H2O or DCE/H2O. The chemical yields were still poor
(25ꢀ33%), but by varying the Rh(I) source we could
significantly improve the yield while maintaining high ee
using aqueous DCE as the solvent (entries 7ꢀ9).12 In
particular, the more reactive cationic Rh(COD)2OTf
(8) For reviews, see: (a) Ager, D. J.; Prakash, J.; Schaad, D. R. Chem.
Rev. 1996, 96, 835. (b) Ager, D. J.; Prakash, J.; Schaad, D. R. Aldrichimica
Acta 1997, 30, 3.
(9) Presumably the cis product would arise if Rh(I) transmetallates
with the metal cyanate followed by syn-insertion to the double bond of 1.
Buchwald’s report proposed such a type of transmetalation with Pd(II),
see ref 1c. The trans product would arise if the reaction follows the
above-mentioned SN20 pathway where Rh(I) does not transmetallate
with metal cyanate, see ref 7.
(10) (R,S)-PPF-PtBu2 = (R)-1-[(S)-2-(Diphenylphosphino)-ferrocenyl]-
ethyl-di-tert-butylphosphine, a Josiphos family ligand.
(11) We have previously found that Et3N HCl was a crucial proton
source in Rh(I)-catalyzed ARO with sodium acetate salt, see reference 5f.
3
Org. Lett., Vol. 15, No. 5, 2013
1065

catalyst gave the highest yield (62%) and good enantio-
selectivity (92% ee).13 A reaction run at room temperature
gave improved yield and enantioselectivity (entry 10). In
fact, only 2.3 equiv of sodium cyanate were needed to give
complete conversion at room temperature to furnish the
desired product in 69% yield and 98% ee (entry 11). Under
these conditions, the formation of the 1-naphthol side
product was also suppressed. Further reduction of the
NaOCN equivalents (1.2 equiv) gave a lower yield (60%).
Using anhydrous DCE caused a considerable decrease
in yield (entry 12).14 The effects of the proton source were
Table 2. Scope of Oxabicyclic Alkenes 1bꢀg in Rh(I)-catalyzed
ARO with Sodium Cyanatea
entry
R1
R2
yieldb (%)
eec (%)
product
1
H
-OCH2O-
52
53
65
62
73
35
98
96
97
94
96f
90
3b
3c
3d
3e
3f
also investigated. In the absence of Et3N HCl (i.e., water
3
2
H
OMe
Me
Br
H
as the proton source), reaction was incomplete and gave
a poor yield (entry 13). Using trifluoroethanol (TFE) did
not afford 3a but yielded 70% of the TFE-induced ring-
opened product,15 and using camphorsulfonic acid (CSA)
and tert-butanol gave low yields (entries 15ꢀ16). It was
3
H
4d
5
Me
OMe
H
6e
F
3g
a Unless specified otherwise, reactions were run with 1bꢀg (0.2
found that 5.0 equiv of Et3N HCl was optimal. Reaction
3
mmol), Rh(COD)2OTf (8 mol %), (R,S)-PPF-PtBu2 (8 mol %), sodium
was not complete at 4 mol % Rh loading, and DPPF was
not an effective ligand in this transformation. The control
experiment showed no background ring-opening process
(entry 17).16
cyanate (2.0 equiv), Et3N HCl (5.0 equiv), DCE/H2O (10:1) (0.1 M), at
3
rt for 17 h, under argon atmosphere. b Isolated yield. c Determined by
chiral HPLC analysis. d Reaction was run at 80 °C, 1.5 h. e Reaction was
run at 30 °C, 17 h. f The ee was measured from the N-Ts derivative of 3f.
The scope of oxabicyclic alkenes was studied under the
optimized conditions (Table 2). The electron-rich sub-
strates 1b, 1c, 1d and 1f gave moderate yields and excellent
enantioselectivities (entries 1ꢀ3 and 5). The elctron-poor
1g gave much lower yield and a diminished ee (entry 6).
The more hindered substrate 1e required a higher reaction
temperature to reach complete conversion though main-
taining satisfactory yield and ee (entry 4). The bromo-
functionalgroupwastoleratedanddidnotpreventcatalyst
turnover considering the reactive cationic Rh could
potentially insert into the ArꢀBr bond. Unfortunately,
less reactive oxabicyclic alkenes such as nonbenzofused,
bridgehead-substituted substrates and azabicyclic alkenes
did not afford the desired products.
products 3h and 3h0 were obtained in a total yield of 60%
with >90% ee in each product (eq 1).17
The absolute configuration of the oxazolidinone prod-
uct was unambiguously confirmed by single-crystal X-ray
analysis of compound 3b (Figure 1). The structure clearly
showed a trans stereochemistry at the ring junction.
By using regiodivergent resolution of the racemic un-
symmetrical oxabicyclic alkene 1h, the two regioisomeric
(12) We have previously demonstrated strong halide effects in Rh(I)-
catalyzed ARO reactions, see refs 5c, 5d, 5f, and a review: Fagnou, K.;
Lautens, M. Angew. Chem., Int. Ed. 2002, 41, 26. The use of a Rh iodide
catalyst (generated in situ using tetrabutylammonium iodide) often gave
drastically improved yields and enantioselectivites. In this case, the
effects were moderate (cf. Table 1, entries 6 and 7).
(13) For examples of the superior reactivity of cationic Rh catalyst
€
over Rh halide catalysts in ARO reactions, see: (a) Webster, R.; Boing,
C.; Lautens, M. J. Am. Chem. Soc. 2009, 131, 444. (b) Preetz, A.; Kohrt,
C.; Drexler, H. -J.; Torrens, A.; Buschmann, H.; Lopez, M. G.; Heller,
D. Adv. Synth. Catal. 2010, 352, 2073.
(14) We observed improved solubility of NaOCN and Et3N HCl
3
when water was added to the reaction at room temperature. Dissolving
NaOCN and Et3N HCl in water then adding to the reaction did not
improve the yield. The reaction mixture was biphasic.
3
(15) TFE was a known nucleophile in the ARO reaction, see ref 5a.
(16) See Supporting Information for full details of the optimization
including the screening of various chiral ligands. We have also found
that polar solvents such as DMSO and DMF were ineffective. Other
chlorinated solvents such as dichloromethane was effective (62% yield,
94% ee, rt, 17 h) but chloroform gave no conversion. Using cationic Rh
catalyst in aqueous DCE gave a higher yield compared to MeCN. At a
higher temperature (80 °C), reactions tend to give lower yields over
prolonged reaction times. Increasing concentration (0.2 M) led to lower
yields. Adding more water (DCE/H2O = 1:1) resulted in a decreased
yield. We have also used NaSCN and CuSCN as the nucleophile but
observed no reaction.
Figure 1. Molecular structure of 3b showing 30% displacement
ellipsoids.
The double bond of oxazolidinone 3a can be reduced
under mild hydrogenation conditions to afford 4
(17) For applications of regiodivergent resolution in Rh(I)-catalyzed
ARO reactions, see refs 6b, 13 and Nguyen, T. D.; Webster, R.; Lautens,
M. Org. Lett. 2011, 13, 1370.
1066
Org. Lett., Vol. 15, No. 5, 2013

Scheme 2. Further Transformations of Product 3a
Scheme 3. Proposed Mechanism for Rh(I)-catalyzed ARO with
Cyanate Anion and Intramolecular Cyclization
containing the tetrahydronaphthalene core (Scheme 2).
The nitrogen can be protected with Ts and Boc groups in
high yields. Subsequent hydrolysis of the cyclic carbamate
moiety of 5 offered the amino alcohol product 7. We
observed no deterioration of ee in these operations.
Based on the X-ray structure of the product and the
previously established mechanistic model,7 we proposed the
following pathway for the formation of oxazolidinone
3a (Scheme 3). The trans stereochemistry of the product
implied an SN20 pathway that is common with heteroatom
nucleophile and ruled out the possibility of transmetala-
tion between Rh and sodium cyanate. The key event is the
enantioselective formation of the highly reactive cationic Rh
complex A, facilitated by protonation from the proton
source.18 Nucleophilic attack by the cyanate anion in an
SN20 fashion leads to the trans isocyanate intermediate B,
which is susceptible to cyclization furnishing 3a. Alterna-
tively, a Rh-alkoxide species C can be generated after the
ring-opening. Insertion of RhꢀO bond to the isocyanate
moiety leads to N-bound and/or O-bound Rh complexes
D and/or E.19 Finally, protonolysis gives product 3a and
regenerates the catalyst.20
In conclusion, we have developed a highly enantio-
selective cationic Rh(I)-catalyzed ARO reaction using
sodium cyanate as the nucleophile. Chiral oxazolidinone
scaffolds with well-defined stereocenters can be synthe-
sized in one step from readily available oxabicyclic
alkenes via a domino ARO/cyclization sequence. We are
currently investigating the potential of using other metal
salts to generate novel chiral building blocks via ARO
technology.
Acknowledgment. We gratefully thank NSERC/Merck
for an Industrial Research Chair. We also thank the
University of Toronto for support of our program, Solvias
AG for the generous gift of chiral ligands and Umicore
for the generous gift of rhodium catalyst. G.C.T. thanks
NSERC for a postgraduate scholarship. N.M.N. is grate-
ful for the PROMOS scholarship granted by the Freie
Universitaet Berlin and the DAAD. A.H. thanksthe Japan
Society for the Promotion of Science (JSPS) for providing
a Research Fellowship for Young Scientists.
(18) Since the pKa of the conjugate acid of cyanate is lower than the
conjugate acid of Et3N, sodium cyanate will remain deprotonated in the
presence of Et3N HCl. In the absence of Et3N HCl, water can also act
3
3
as the proton source albeit less effectively (cf. Table 1, entry 13). In the
absence of both Et3N HCl and water, no reaction occurred (see
Supporting Information).
3
(19) A similar mechanism has been proposed in Rh(I)-catalyzed
addition reactions of isocyanates, see: (a) Miura, T.; Takahashi, Y.;
Murakami, M. Chem. Commun. 2007, 3577. (b) Hoshino, Y.; Shibata,
Y.; Tanaka, K. Angew. Chem., Int. Ed. 2012, 51, 9407.
(20) Attempts to trap the isocyanate intermediates B or C with an
external nucleophile such as phenol, tBuOH or aniline failed to give the
corresponding noncyclized carbamate or urea products. Only 3a and the
competing nucleophile-induced ARO products were observed, which
showed that the intramolecular cyclization is a much faster process than
intermolecular nucleophilic addition.
Supporting Information Available. Experimental pro-
cedures, full characterization for all compounds and
crystallographic data for 3b. This material is available
free of charge via the Internet at http://pubs.acs.org.
The authors declare no competing financial interest.
Org. Lett., Vol. 15, No. 5, 2013
1067