Phthalides by rhodium-catalyzed ketone hydroacylation

Article abstract of DOI:10.1021/ja907711a

(Chemical Equation Presented) Phthalides are biologically relevant five-membered lactones found in herbs, fruits, and vegetables. Herein we communicate the first atom-economical approach to phthalides by using enantioselective ketone hydroacylation. In the presence of Rh[(Duanphos)]X (X = NO₃, OTf, OMs), various 2-ketobenzaldehydes undergo intramolecular hydroacylation to produce phthalide products in good yields and 92-98% ee's. Our study highlights the key role counterions play in controlling both reactivity and enantioselectivity. A concise asymmetric total synthesis of the celery extract (S)-(-)-3-n-butylphthalide is also presented.

Full text of DOI:10.1021/ja907711a

Published on Web 10/08/2009
Phthalides by Rhodium-Catalyzed Ketone Hydroacylation
Diem H. T. Phan, Byoungmoo Kim, and Vy M. Dong*
Department of Chemistry, UniVersity of Toronto, 80 St. George Street, Toronto, Ontario M5S 3H6, Canada
Received September 10, 2009; E-mail: vdong@chem.utoronto.ca
Scheme 1. Mechanistic Rationale for Ketone Hydroacylation
Phthalide natural products show a range of bioactivity, from
enhancing flavors to slowing memory loss.1 Although diverse in
function, these phytochemicals share a common structure: a benzene
ring fused to a γ-lactone. Significant effort has been focused on
synthesizing phthalides,2 but most asymmetric methods require
chiral auxiliaries2a,b or chiral organometallics,2c,d few are catalytic,2e-i
and none is atom-economical.3 Toward addressing this challenge,
herein we report an efficient and complementary route to phthalides
by enantioselective ketone hydroacylation. While hydroacylation
of olefins to form cyclopentanones is well-studied,4 hydroacylation
of ketones to form five-membered lactones remains to be explored.5
Table 1. Developing a Rh Catalyst for Phthalide Synthesis:
We recently reported that intramolecular hydroacylation is a
promising approach to chiral lactones.6 However, our catalyst,
[Rh(DTBM-Segphos)]BF4, was tailored for making seven-mem-
bered lactones, specifically benzodioxepinones (eq 1 in Scheme 1).
Mechanistic studies revealed the turnover-limiting step to be
insertion of the ketone CdO bond to the rhodium hydride via
transition state A.6b In A, the ether oxygen is coordinated to Rh,
and this coordination is critical for promoting insertion over
competitive decarbonylation. Indeed, our previous protocol was
limited in scope to ketoaldehydes bearing an ether linkage. On the
basis of these studies, we imagined 2-ketobenzaldehydes 1 under-
going hydroacylation to form phthalides by an analogous pathway
via a distinct transition state B (eq 2 in Scheme 1). In B, one
coordination site would be available for solvent, ligand, or
counterion binding. Thus, finding a new catalyst would be critical
for promoting the desired reactivity over decarbonylation.
Counterion Effects on Reactivitya
With this mechanism in mind, we set out to develop a Rh catalyst
for the preparation of phthalides by varying both the bisphosphine
ligand and counterion. Our initial study focused on cyclizing model
substrate 1a into lactone 2a preferentially over the decarbonylation
product 3a (eq 3 in Table 1). In the absence of catalyst, 1a was
recovered (Table 1, entry 1). After evaluating many chiral phosphines,
we identified Duanphos as a promising ligand.7 With this bulky,
electron-rich phosphine, we studied counterions with various coordi-
a Reaction scale: 0.1 mmol of 1a. b Based on 1H NMR integration.
c Using 0.2 mmol of 1a at 100 °C. d No silver salt was added. e Isolated
yields. f Determined by chiral HPLC.
required 3 days with 10 mol % catalyst (97% yield, 97% ee; entry
8). We chose AgNO3 as the optimal additive; nitrate is less strongly
coordinating than chloride (giving shorter reaction times) but
coordinating enough to suppress decarbonylation and assist in
enantioinduction.8b-d Use of 5 mol % AgNO3 and 5 mol % Rh
afforded phthalide 2a in 97% yield and 97% ee (entry 7).
To examine catalyst scope, we prepared a series of 2-ketoben-
zaldehydes 1 with varying substitution on the aromatic backbone.
Substituents were tolerated at the 4-, 5-, and 6-positions of the ring
(Table 2, entries 2-10). Substrates bearing electron-donating (e.g.,
t-Bu, MeO) and electron-withdrawing (e.g., Cl, NO2, CO2Me)
substituents underwent hydroacylation in 67-97% yields and
92-98% ee’s. However, steric bulk at the 3-position of the
nating strengths (SbF6- < BF4-
(entries 2-8).
<
-OTf < -OMs < NO3- < Cl-)8
This study revealed marked counterion effects on reactivity.
Catalysts with more strongly coordinating counterions gave better
selectivity for hydroacylation over decarbonylation, with yields
increasing from 30 to 97%. These counterions also impacted
asymmetric induction, with ee’s ranging from 29 to 97%. The
cationic complexes appeared to be more reactive than the neutral
Rh chloride catalyst. For example, with triflate, hydroacylation was
complete in less than 30 min with 10 mol % catalyst (>95% yield,
81% ee; entry 4). In contrast, with chloride, the same transformation
9
15608 J. AM. CHEM. SOC. 2009, 131, 15608–15609
10.1021/ja907711a CCC: $40.75  2009 American Chemical Society

C O M M U N I C A T I O N S
Table 2. Hydroacylation of Substituted Ketoaldehydesa
Scheme 2. Total Synthesis of Celery Extract
(S)-(-)-3-n-Butylphthalide (6)a
a Reagents and conditions: (a) MeN(OMe)(CdO)n-Bu, 1.6 M n-BuLi,
THF, -78 °C to rt, overnight, then 2 M HCl(aq), rt, 3 h; (b) 5 mol %
[Rh(cod)Cl]2, 10 mol % (S,S,R,R)-Duanphos, 10 mol % AgNO3, toluene,
75 °C, 3 days.
entry
R
iso. yield of 2 (%)
ee (%)d
time (day)
1
H (1a)
97 (2a)
88 (2b)
84 (2c)
91 (2d)
84 (2e)
67 (2f)
94 (2g)
94 (2h)
5 (2i)
97
96
96
92
98
95
93
95
-
1
2
3
2
2
1
1
1
1
3
1
2
4-Me (1b)
4-OMe (1c)
5-Me (1d)
5-t-Bu (1e)
5-Cl (1f)
5-NO2 (1g)
5-CO2Me (1h)
6-OMe (1i)
6-OMe (1i)
3-Me (1j)
3b
4
for treating strokes.9 As shown in Scheme 2, we converted
commercial acetal 4 to ketoaldehyde 5 in 71% yield in one pot. In
the presence of [Rh((S,S,R,R)-Duanphos)]NO3, 5 cyclized to
phthalide 6 in 93% yield and 97% ee.
In conclusion, we have reported an atom-economical approach
to phthalides by enantioselective C-H bond functionalization. A
hydroacylation catalyst for making five-membered lactones has been
discovered. The appropriate choice of counterion was crucial in
suppressing decarbonylation and controlling enantioselectivity.
Mechanistic studies to better understand the counterion effects and
develop future carbonyl hydroacylations are underway.
5
6
7
8
9
10c
11
78 (2i)
<5 (2j)
97
-
a Conditions: 0.2 mmol of substrate. b Using 0.1 mmol of substrate
and 10 mol % catalyst; the 1H NMR yield is given. c Using 10 mol %
catalyst and no AgNO3. d Determined by chiral HPLC.
Table 3. Intramolecular Hydroacylation of Various Ketonesa
Acknowledgment. We thank the University of Toronto, the
Canada Foundation for Innovation, the Ontario Research Fund, the
National Science and Engineering Council of Canada (NSERC),
and Boehringer Ingelheim (Canada) Ltd for funding.
entry
R′
X
iso. yield (%) ee (%)d T (°C) time (day)
Note Added after ASAP Publication. The yield of 2a was
corrected in Table 2 on October 14, 2009.
1b
2c
3
Et (1k)
i-Pr (1l)
C6H5 (1m)
NO3
NO3
OMs
94 (2k)
83 (2l)
81 (2m)
93 (2n)
88 (2o)
22 (2o)
48 (2o)
92 (2p)
25 (2p)
96
97
93
96
92
89
13
96
-
100 2.5
75 3.5
90
90
90
90
90
75
90
3
3
3
3
3
3
3
Supporting Information Available: Experimental procedures,
characterization data for new compounds, and chiral chromatographic
analyses. This material is available free of charge via the Internet at
http://pubs.acs.org.
4
5
6
7
4-OMeC6H4 (1n) OMs
4-CH3C6H4 (1o)
4-CH3C6H4 (1o)
4-CH3C6H4 (1o)
OMs
NO3
OTf
8
9
4-NO2C6H4 (1p) OTf
4-NO2C6H4 (1p) OMs
References
(1) Ge, L.; Chan, S.-K. S.; Chung, H.-S.; Li, S.-L. Studies in Natural Products
Chemistry; Rahman, A., Ed.; Elsevier: Amsterdam, 2005; Vol. 32, pp 611-
671.
a Conditions: 0.1 mmol of substrate. b Using 0.2 mmol of substrate
and 7 mol % catalyst. c Using 0.2 mmol of substrate and 15 mol %
catalyst. d Determined by chiral HPLC.
(2) For selected examples, see: (a) Kosaka, M.; Sekiguchi, S.; Naito, J.; Uemura,
M.; Kuwahara, S.; Watanabe, M.; Harada, N.; Hiroi, K. Chirality 2005,
17, 218. (b) Pedrosa, R.; Sayalero, S.; Vincente, M. Tetrahedron 2006,
62, 10400. (c) Ramachandran, P. V.; Chen, G.-M.; Brown, H. C.
Tetrahedron Lett. 1996, 37, 2205. (d) Knepper, K.; Ziegert, R. E.; Brase,
S. Tetrahdedron 2004, 60, 8591. (e) Kitamura, M.; Okhuma, T.; Inoue, S.;
Sayo, N.; Kumobayshi, H.; Akutagawa, S.; Takaya, H.; Noyori, R. J. Am.
Chem. Soc. 1998, 110, 629. (f) Witulski, B.; Zimmermann, A. Synlett 2002,
1855. (g) Tanaka, K.; Nishida, G.; Wada, A.; Noguchi, K. Angew. Chem.,
Int. Ed. 2004, 43, 6510. (h) Chang, H.-T.; Jeganmohan, M.; Cheng, C.-H.
Chem.sEur. J. 2007, 13, 4356. (i) Everaere, K.; Mortreux, A.; Carpentier,
J.-F. AdV. Synth. Catal. 2003, 345, 67.
2-ketobenzaldehyde prohibited reactivity (<5% yield; entry 11). In
the case of aldehyde 1i, using the Rh nitrate catalyst resulted in
poor reactivity (5% yield; entry 9), while the Rh chloride catalyst
was effective (78% yield, 97% ee; entry 10). With Rh nitrate, we
imagine that the o-methoxy group competes with the ketone
carbonyl by coordinating to Rh and preventing the proper geometry
for insertion. With Rh chloride, this methoxy coordination must
be more dynamic, thus allowing for ketone insertion.
Next, we varied the substituent (R′) on the prochiral ketone (eq
5 in Table 3). Ketones with substituents larger than methyl
underwent hydroacylation with high enantioselectivity but required
increased catalyst loading (Table 3). Ethyl ketone 1k cyclized to
phthalide 2k (94% yield, 96% ee; entry 1) and isopropyl ketone 1l
to 2l (83% yield, 97% ee; entry 2). In studying biaryl ketones, we
observed counterion effects that were remarkably substrate-specific.
For the biaryl ketones 1m (R′ ) Ph) and 1n and 1o (R′ ) phenyl
with electron-donating groups), mesylate was the best counterion
(81-93% yield, 92-96% ee; entries 3-5), while nitrate and triflate
resulted in poor results (entries 6 and 7). In contrast, ketone 1p (R′
) phenyl with an electron-withdrawing group) forms lactone 2p
more efficiently with AgOTf (92% yield, 96% ee; entry 8) than
AgOMs (25% yield; entry 9).10
(3) Trost, B. M. Science 1991, 254, 1471.
(4) For a review, see: (a) Fu, G. C. In Modern Rhodium-Catalyzed Reactions;
Evan, P. A., Ed.; Wiley-VCH: New York, 2005; pp. 79-91, and references
therein. For enantioselective indanone synthesis, see: (b) Kundu, K.;
McCullagh, J. V.; Moorehead, A. T. J. Am. Chem. Soc. 2005, 127, 16042.
(5) For Rh-catalyzed aldehyde hydroacylation, see: (a) Bergens, C. H.; Fairlie,
D. P.; Bosnich, B. Organometallics 1990, 9, 566. (b) Fuji, K.; Tsutsumi,
K.; Kakiuchi, K.; Morimoto, T. Chem. Commun. 2005, 3295. For NHC-
catalyzed ketone hydroacylation, see: (c) Chan, A.; Scheidt, K. J. Am. Chem.
Soc. 2006, 128, 4558.
(6) (a) Shen, Z.; Khan, H. A.; Dong, V. M. J. Am. Chem. Soc. 2008, 130,
2916. (b) Shen, Z.; Dornan, P. K.; Khan, H. A.; Woo, T. K.; Dong, V. M.
J. Am. Chem. Soc. 2009, 131, 1077.
(7) Toluene was found to be the best solvent. See the Supporting Information
for more details on ligand and solvent effects.
(8) For a review of ion pairing in transition metals, see: (a) Macchioni, A.
Chem. ReV. 2005, 105, 2039, and references therein. For selected examples
of counterion effects on asymmetric catalyses, see: (b) Ashimori, A.;
Bachand, B.; Poon, D. J.; Overman, L. E. J. Am. Chem. Soc. 1998, 120,
6477. (c) Nandi, M.; Jin, J.; RajanBabu, T. V. J. Am. Chem. Soc. 1999,
121, 9899. (d) Moreau, C.; Hague, C.; Weller, A. S.; Frost, C. G.
Tetrahedron Lett. 2001, 42, 6957.
(9) Zhao, C.; Cui, S.; Zhang, R.; He, Z. Biomed. Chromatogr. 2003, 17, 391.
(10) Phthalide 2p was observed to epimerize in polar solvent over time (see the
Supporting Information for details).
Lastly, we present an asymmetric synthesis of the natural product
(S)-(-)-3-n-butylphthalide (6).2a This phthalide is responsible for
the flavor of celery, and its racemate was in phase-III clinical trials
JA907711A
9
J. AM. CHEM. SOC. VOL. 131, NO. 43, 2009 15609