Axially chiral phosphine-oxazoline ligands in silver(I)-catalyzed asymmetric Mannich reaction of aldimines with trimethylsiloxyfuran

Article abstract of DOI:10.1002/adsc.200900550

A new asymmetric catalytic system for the Mannich reaction of aldimines with trimethylsiloxyfuran is described. The combination of an axially chiral phosphine-oxazoline ligand (S)-2-[(R)-2′-(diphenylphosphino)-1,1′- binaphthyl-2-y1]-4-pheny1-4,5dihydrooxa

Full text of DOI:10.1002/adsc.200900550

FULL PAPERS
DOI: 10.1002/adsc.200900550
Axially Chiral Phosphine-Oxazoline Ligands in Silver(I)-
Catalyzed Asymmetric Mannich Reaction of Aldimines with
Trimethylsiloxyfuran
Hong-Ping Deng,a Yin Wei,a and Min Shia,*
a
State Key Laboratory of Organometallic Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of
Sciences, 354 Fenglin Road, Shanghai 200032, Peopleꢀs Republic of China
Fax : (+86)-21-6416-6128; e-mail: mshi@mail.sioc.ac.cn
Received: August 4, 2009; Published online: November 18, 2009
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/adsc.200900550.
Abstract: A new asymmetric catalytic system for the mines with trimethylsiloxyfuran in dichloromethane
Mannich reaction of aldimines with trimethylsiloxy- at À788C, affording the corresponding adducts in up
furan is described. The combination of an axially to 99% yield, 99:1 (dr) and 99% ee (major diastereo-
chiral phosphine-oxazoline ligand (S)-2-[(R)-2’-(di- isomer) under mild conditions.
phenylphosphino)-1,1’-binaphthyl-2-yl]-4-phenyl-4,5-
dihydrooxazole with silver acetate and 2,2,2-tri- Keywords: aldimines; asymmetric Mannich reaction;
fluoroacetic acid is a very effective catalytic system axially chiral phosphine-oxazoline ligands; silver ace-
in the asymmetric Mannich reaction of various aldi- tate; trimethylsiloxyfuran
Introduction
the silver(I)-catalyzed asymmetric Mannich reaction
of aldimines with trimethylsiloxyfuran under mild
The efficient synthesis of highly functionalized g-bute- conditions,[8] in which a wider range of aldimines can
nolides remains an important challenge in organic be applied to afford the corresponding adducts in
chemistry.[1] It has been known that the Lewis acid- good yields and high enantiomeric excesses for anti-g-
catalyzed vinylogous Mannich-type reaction of trime- butenolides.
thylsiloxyfuran with aldimines is a powerful synthetic
protocol to prepare g-butenolide derivatives bearing
an amine functionality.[2,3] However, the catalytic, Results and Discussion
asymmetric version of the vinylogous Mannich-type
reaction has been little explored. Thus far, Martin We initially utilized the Lewis acid AgOAc
and Lopez reported the first catalytic asymmetric ad- (10 mol%) combined with chiral phosphine-oxazoline
dition of trialkylsilyloxyfurans to aldimines in 1999,[2f] ligands L1–L4 (10 mol%) (Figure 1) as the catalysts
affording the adducts in moderate enantioselectivities and the asymmetric vinylogous Mannich-type reaction
and chemical yields. More recently, Hoveyda and of readily available aldimine 1a with the siloxyfuran
Snapper have developed a silver(I)-based catalyst as a model reaction in dichloromethane (DCM) or
using the 2-methyloxyphenyl group as aldimine sub- tetrahydrofuran (THF) containing the additive
stituent, leading to the product of a asymmetric vinyl- CF3CH2OH to develop the optimal reaction condi-
ogous Mannich-type reaction in excellent diastereo- tions (Scheme 1) and the results of these experiments
and enantioselectivity.[4,5] Organocatalyzed asymmet- are summarized in Table 1. It was found that (R,S)-P-
ric vinylogous Mannich-type reactions have also been Oxa-Ph (L1) is the best chiral ligand in this reaction
reported lately.[6] Aside from these pioneering investi- to afford the corresponding product 2a in 95% yield
gations, the exploration of new catalytic systems in with the anti-configuration as the major diastereomer
the asymmetric vinylogous Mannich-type reaction for (dr=99/1) with 95% ee in DCM at À788C for 6 h and
a wide range of aldimines is still a main challenge at then naturally warming up to room temperature
the present time. In this paper, we wish to report ax- (208C) for 10 h (Table 1, entries 1–3). Furthermore,
ially chiral phosphine-oxazoline ligands (L1–L4)[7] in using THF as the solvent also produced 2a in 87%
Adv. Synth. Catal. 2009, 351, 2897 – 2902
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2897

Hong-Ping Deng et al.
FULL PAPERS
Table 2. Temperature effect in the combination of L1/
AgOAc/CF3CH2OH in DCM.
[a]
Reaction was kept at this temperature for 6 h, then the
reaction mixture was warmed up to room temperature
naturally for 10 h under the standard conditions.
Reaction was kept at this temperature for 2 h and then
Figure 1. Axially chiral phosphine-oxazoline ligands L1–L4.
[b]
the reaction was warmed up to room temperature natu-
rally for 14 h under the standard conditions.
yield as the anti-configuration as the major diastereo-
mer (dr=99/1) with 85% ee (Table 1, entry 1). The
(S,S)-P-Oxa-Ph L4, a diastereomeric isomer of L1, af-
forded low yield and poor stereoselectivity under
identical conditions (Table 1, entry 4).
[c]
Yields of the purified syn and anti products.
[d]
[e]
1
Determined from H NMR spectroscopic data.
Determined by chiral HPLC.
Next, we further optimized the reaction conditions reaction time at À788C to 2 h afforded 2a in lower
using L1 as the ligand to examine the temperature, yield and diastereoselectivity (Table 2, entry 6). The
solvent and additive effects in this reaction and the examination of solvents effect revealed that DCM is
results of these experiments are summarized in the best solvent to afford 2a in higher yield as well as
Table 2, Table 3 and Table 4, respectively. It was diastereoselectivity and enantioselectivity (Table 3,
found that the reaction temperature should be kept at entries 1–5). The additives are also crucial to this re-
À788C for 6 h and then warming up to room tempera- action and we found that 2a could be obtained in
ture (208C) naturally for 10 h to give the adduct in higher enantioselectivity in the presence of
higher yield as well as higher diastereoselectivity and CH3CH2OH, CF3CH2OH or BnOH under the stan-
enantioselectivity (Table 2, entries 1–5). Reducing the dard conditions (Table 4, entries 1–6).[9] Using
Scheme 1. Reaction of aldimines with trimethylsiloxyfuran.
Table 1. Examination of ligands L1–L4 in DCM or THF.
[a]
Yields of the purified syn and anti products.
Determined from H NMR spectroscopic data.
Determined by chiral HPLC.
Reaction was performed in DCM.
Reaction was performed in THF.
[b]
[c]
[d]
[e]
1
2898
asc.wiley-vch.de
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2897 – 2902

Axially Chiral Phosphine-Oxazoline Ligands in Silver(I)-Catalyzed Asymmetric Mannich Reaction
FULL PAPERS
Table 3. Solvent effect in the combination of L1/AgOAc/
CF3CH2OH.
conditions are those using 10 mol% of AgOAc and
10 mol% of chiral ligand L1 as the catalyst and carry-
ing out the reaction in DCM at À788C for 6 h and
then warming up to room temperature naturally for
10 h in the presence of CF3CH2OH.
Under these optimum conditions, we next exam-
ined the generality of this reaction with various aldi-
mines 1, without 2-methyloxyphenylaldimine which
has been an essential effect in Hoveydaꢀs asymmetric
vinylogous Mannich-type reaction system,[4a] with the
siloxyfuran and the results are summarized in Table 5
and Table 6, respectively. Aldimines 1b–1f having a
G1 substituent at the ortho-, meta- or para-position of
the phenyl ring gave the corresponding asymmetric
vinylogous Mannich-type products 3b–3f in 80–86%
yields, drs of 2:1–20:1 and 93–99% enantiomeric ex-
cesses for the anti-diastereoisomers (Table 5, en-
tries 1–5). On the other hand, as for aldimines 1g–1k
having the G2 substituent at the meta- or para-posi-
tion of the phenyl ring, the corresponding adducts 2g–
2k were formed in 64–95% yields, drs of 5:1-99:1 as
well as 87–99% enantiomeric excesses for the anti-
diastereoisomers under the standard conditions
(Table 6, entries 1–5). The absolute configuration of
products 2 was unequivocally assigned as (R,S) by an
X-ray diffraction study of 2g bearing a bromine atom
at the benzene ring (Figure 2, see the Supporting In-
formation).[10] In the case of aldimine 1l having sub-
stituents on the both of the phenyl ring (G1 =p-Br
and G2 =p-Br), the corresponding adduct 2l was ob-
tained in 76% yield as well as 2:1 dr and 90% ee for
the major diastereoisomer under identical conditions,
indicating the broad substrate generality of this novel
asymmetric catalytic system (Table 6, entry 6).
[a]
Reaction was kept at this temperature for 6 h, then the
reaction mixture was warmed up to room temperature
naturally for 10 h under the standard conditions.
Reaction was kept at this temperature for 16 h under the
standard conditions.
[b]
[c]
Yields of the purified syn and anti products.
[d]
[e]
1
Determined from H NMR spectroscopic data.
Determined by chiral HPLC.
Table 4. Additive effect in the combination of L1/AgOAc in
DCM.
[a]
Yields of the purified syn and anti products.
[b]
[c]
1
Determined from H NMR spectroscopic data.
Determined by chiral HPLC.
CF3CH2OH as the additive, 2a was produced in higer
The synthetic utility of the product can be dis-
yield and enantioselctivity. Thus, the best reaction played by removal of the N-aryl group with ceri-
Table 5. Examination of imines having different substituents (G1) on the aromatic ring.
[a]
Yields of the purified syn and anti products.
Determined from H NMR spectroscopic data.
Determined by chiral HPLC.
[b]
[c]
1
Adv. Synth. Catal. 2009, 351, 2897 – 2902
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2899

Hong-Ping Deng et al.
Table 6. Examination of imines having different substituents (G2 and G1) on the aromatic
FULL PAPERS
ring.
[a]
Yields of the purified syn and anti products.
Determined from H NMR spectroscopic data.
Determined by chiral HPLC.
[b]
[c]
1
AHCTUNGTREGUNiNN mine substrate during the reaction. We have opti-
mized the structure of this kind of activated complex
using L1, substrate 1a and Ag(I) at HF/3-21G* level
(Figure 3, see the Supporting Information). The opti-
mized structure shows that substrate 1a is indeed as-
sociated with Ag(I) and the phenyl group in substrate
1a is close to the phenyl group on the oxazoline ring
in L1. The optimized structure of the activated com-
plex throws light on explanation of the enantio-/dia-
stereoselectivity. Herein, we propose several plausible
mechanistic models (Scheme 3) to rationalize the
Figure 2. ORTEP drawing drawing of the X-ray crystal
structure of 2g.
um(IV) ammonium nitrate (CAN) in MeCN/H2O
under mild conditions,[4a,11] affording the correspond-
ing amino compound 3i in 61% yield and high ee
value (Scheme 2).
Based on previous mechanistic studies by Hoveyda
and co-workers,[4a,12] an activated complex is initially
associated through chelation of chiral Ag(I) and ald-
Figure 3. The structure of the activated complex optimized
at the HF/3-21G* level.
Scheme 2. Removal of the N-aryl group with CAN.
2900
asc.wiley-vch.de
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2897 – 2902

Axially Chiral Phosphine-Oxazoline Ligands in Silver(I)-Catalyzed Asymmetric Mannich Reaction
FULL PAPERS
Scheme 3. Mechanistic models.
enantio-/diastereoselectivity. The siloxylfuran may ap- yield and 99% ee (major diastereoisomer). Current ef-
proach the activated complex from one face of the forts are in progress to extend this chiral phosphine-
imine as shown in model I or another face of the oxazoline ligand/AgOAc/BnOH combination to other
imine as shown in model II. If the siloxyfuran reaches catalytic enantioselective reactions in our laboratory.
the activated complex as shown in the model II, a
strong steric repulsion may arise between the phenyl
group on the oxazoline ring in L1, the phenyl group
of 1a and siloxyfuran. Thus, the siloxyfuran prefers to
Experimental Section
approach the activated complex as shown in model I,
resulting in good enantioselectivity. Similarly, we en-
AgOAc (0.02 mmol) and chiral phosphine-oxazoline ligand
visaged that the steric repulsion may exist between
the substrate and siloxyfuran in model III, and the
model I is still preferable, which leads to the anti-dia-
stereoisomer as the major product. However, the dia-
Typical Procedure
L1 (0.02 mmol) were added into a Schlenk tube and then
DCM (0.5 mL) was added into the reaction vessel. The re-
sulting solution was stirred for 0.5 h at room temperature.
Aldimine 1a (0.20 mmol) and CF3CH2OH (0.36 mmol) were
stereoselectivity is not so good with respect to some
added followed by another 0.5 mL of DCM. The mixture
substrates, probably due to the weak steric effects in
was cooled to À788C, and stirred for 0.5 h, then siloxyfuran
these cases.
(0.36 mmol) was added. The mixture was allowed to stir at
À788C for 6 h, then warmed to room temperature naturally
for 10 h. The reaction was quenched by the addition of a sa-
turated aqueous solution of NaHCO3 (1.0 mL). After the
mixture had been stirred for 10 min at room temperature,
the resulted mixture was extracted by DCM for three times
and the organic layer was dried over anhydrous Na2SO4.
Then the solvent was removed under reduced pressure and
the residue was purified by flash column chromatography
on silica gel (elution with petroleum ether/EtOAc=6:1) to
give previously known 2a[2] as a plae yellow solid; yield:
95%, 96% ee. 1H NMR (400 MHz, CDCl3, TMS): d=4.38
(1H, d, J=6.4 Hz, NH), 4.84 (1H, dd, J=4.4, 6.4 Hz, CH),
Conclusions
In summary, we have developed a new catalytic,
asymmetric vinylogous Mannich-type reaction system
applicable to a wide range of aldimines which do not
possess a 2-methyloxyphenyl group and siloxyfuran
using an axially chiral phosphine-oxazoline ligand
[(R,S)-P-Oxa-Ph]/AgOAc/CF3CH2OH combination.
The catalytic system reported here afforded asymmet-
ric vinylogous Mannich-type products 3 in up to 99% 5.41 (1H, ddd, J=1.2, 2.0, 4.4 Hz, CH), 6.04 (1H, dd, J=
Adv. Synth. Catal. 2009, 351, 2897 – 2902
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
asc.wiley-vch.de
2901

Hong-Ping Deng et al.
FULL PAPERS
2.0, 6.0 Hz, CH), 6.54 (2H, d, J=7.6 Hz, ArH), 6.70 (1H, t,
J=8.0 Hz, ArH), 7.10 (2H, dd, J=7.6, 8.0 Hz, ArH), 7.24–
7.31 (5H, m, ArH), 7.34 (1H, dd, J=1.2, 6.0 Hz, CH); the
enantiomeric excess was determined by HPLC with a Chir-
alcel OD-H column (l =254 nm; eluent: hexane/isopropyl
alcohol=80/20; flow rate: 0.9 mLminÀ1): tmajor =25.73 min,
Casiraghi, Tetrahedron: Asymmetry 1997, 8, 2975–
2987; f) S. F. Martin, O. D. Lopez, Tetrahedron Lett.
1999, 40, 8949–8953; g) L. Battistini, G. Rassu, L.
Pinna, F. Zanardi, G. Casiraghi, Tetrahedron: Asymme-
try 1999, 10, 765–773; h) M. V. Spanedda, M. Ourꢃ-
vitch, B. Crousse, J.-P. Bꢃguꢃ, D. Bonnet-Delpon, Tetra-
hedron Lett. 2004, 45, 5023–5025; i) S. Oudeyer, B.
Dudot, J. Rorer, Heterocycles 2005, 65, 823–841.
[3] For a review, see: S. F. Martin, Acc. Chem. Res. 2002,
35, 895–904.
[4] a) E. L. Carswell, M. L. Snapper, A. H. Hoveyda, J.
Tang, Angew. Chem. 2006, 118, 7388–7391; Angew.
Chem. Int. Ed. 2006, 45, 7230–7233; b) H. Mandai, K.
Mandai, M. L. Snapper, A. H. Hoveyda, J. Am. Chem.
Soc. 2008, 130, 17961–17969; c) Z.-L. Yuan, J.-J. Jiang,
M. Shi, Tetrahedron 2009, 65, 6001–6007.
t
minor =36.64 min; ee%=96%; [a]2D0: À134.8 (c 1.20, CHCl3).
Supporting Information
Spectroscopic data and chiral HPLC traces of the com-
pounds shown in Tables 1–6 and the detailed descriptions of
experimental proceduresare available as Supporting Infor-
mation.
[5] Also see: A. S. Gonzꢄlez, R. G. Arrayꢄs, M. R. Rivero,
Acknowledgements
J. C. Carretero, Org. Lett. 2008, 10, 4335–4337.
[6] a) T. Y. Liu, H. L. Cui, J. Long, B. J. Li, Y. Wu, L. S.
Ding, Y. C. Chen, J. Am. Chem. Soc. 2007, 129, 1878–
1879; b) T. Akiyama, Y. Honma, J. Itoh, K. Fuchibe,
Adv. Synth. Catal. 2008, 350, 399–402; c) D. S. Giera,
M. Sichert, C. Schneider, Org. Lett. 2008, 10, 4259–
4262; d) M. Sickert, C. Schneider, Angew. Chem. 2008,
120, 3687–3690; Angew. Chem. Int. Ed. 2008, 47, 3631–
3634.
Financial support from the Shanghai Municipal Committee
of Science and Technology (06XD14005 and 08dj1400100-2),
National Basic Research Program of China
ACHTUNGTRNE(NUNG 973)-
2010CB833300, and the National Natural Science Foundation
of China (20872162, 20672127, 20732008, 20821002 and
20702013) and Professor Xiaoli Zhao for performing X-ray
diffraction are greatly acknowledged.
[7] For the preparation of ligands L1–L4, see: M. Ogasa-
wara, K. Yoshida, H. Kamei, K. Kato, Y. Uozumi, T.
Hayashi, Tetrahedron: Asymmetry 1998, 9, 1779–1787.
[8] Reviews for the silver-catalyzed asymmetric reactions,
please see: a) M. Naodovic, H. Yamamoto, Chem. Rev.
2008, 108, 3132–3148; b) A. Yanagisawa, T. Arai,
Chem. Commun. 2008, 1165–1172.
[9] For a review on the effect of additives in asymmetric
catalysis, please see: a) E. M. Vogl, H. Grçger, M. Shi-
basaki, Angew. Chem. 1999, 111, 1672–1680; Angew.
Chem. Int. Ed. 1999, 38, 1570–1577; b) D. S. Giera, M.
Sichert, C. Schneider, Org. Lett. 2008, 10, 4259–4262.
[10] CCDC 735984 contains the supplementary crystallo-
graphic data for compound 2g in 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.
References
[1] Selected reviews on the synthesis of g-butenolides; a) P.
Langer, Synlett 2006, 3369–3381; b) G. Romeo, D. Ian-
nazzo, A. Piperno, R. Romeo, A. Corsaro, A. Rescifi-
na, U. Chiacchio, Mini-Rev. Org. Chem. 2005, 2, 59–77;
c) M. Ito, Pure Appl. Chem. 1991, 63, 13–22; d) R.
Bruckner, Curr. Org. Chem. 2001, 5, 679–718; e) P. A.
Jacobi, Advances in Heterocycl. Nat. Prod. Synth. 1992,
2, 251–298; f) for reviews for catalytic asymmetric
Mannich-type reactions, please see: g) S. Kobayashi, H.
Ishitani, Chem. Rev. 1999, 99, 1069–1094; h) A. Cꢂrdo-
va, S. Cicchi, F. M. Cordero, A. Goti, Acc. Chem. Res.
2004, 37, 102–112; i) G. K. Friestad, A. K. Mathies, Tet-
rahedron 2007, 63. 2541–2569.
[2] a) T. Tsukamoto, T. Kitazume, Chem. Lett. 1992, 1377–
1378; b) G. Rassu, L. Pinna, P. Spanu, N. Culeddu, G.
Casiraghi, Tetrahedron 1992, 48, 727–742; c) G. Casira-
ghi, G. Rassu, P. Spanu, L. Pinna, F. Ulgheri, J. Org.
Chem. 1993, 58, 3397–3400; d) G. Casiraghi, G. Rassu,
Synthesis 1995, 607–626; e) L. Battistini, F. Zanardi, G.
Rassu, P. Spanu, G. Pelosi, G. Fava, M. B. Ferrari, G.
[11] a) T. Fukuyama, R. K. Frank, C. F. Jewell, J. Am.
Chem. Soc. 1980, 102, 2122–2123; b) K. Nakayama,
H. C. Kawato, H. Inagaki, R. Nakajima, A. Kitamura,
K. Someya, T. Ohta, Org. Lett. 2000, 2, 977–980.
[12] L. C. Wieland, E. M. Vieira, M. L. Snapper, A. H. Hov-
eyda, J. Am. Chem. Soc. 2009, 131, 570–576.
2902
asc.wiley-vch.de
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Adv. Synth. Catal. 2009, 351, 2897 – 2902