L. J. Martins et al. / Tetrahedron Letters 55 (2014) 5264–5267
5265
Table 1
Table 2
Preparation of Morita–Baylis–Hillman adducts
Preparation of a-arylated b-keto esters
DABCO
Phosphine-free Mizoroki-Heck reaction
OH
O
ultrasound
R2
Cl
Cl
R1
R1
H
Iodophenol,
O
OH
R2
Nájera's catalyst
DMF, 110 0C, 2h
R2
R2
MBH adducts
R1
R1
N
Pd Cl
4-14
HO
Nájera's catalyst
Entry
MBH adducts
Yielda,b,c (%)
MBH adducts
OH
4-14
-Arylated-β-keto esters
α
1
2
3
4
5
6
7
8
9
4, R1 = C6H5; R2 = CO2CH3
85
96
90
70
71
80
99
77
85
85
90
5, R1 = 4-NO2–C6H4; R2 = CO2CH3
6, R1 = 2-C4H4S; R2 = CO2CH3
15-25
Entry
MBH adducts
Yielda (%)
7, R1 = 3,4-(OCH2O)–C6H3; R2 = CO2CH3
8, R1 = 3,4,5-CH3O–C6H2; R2 = CO2CH3
9, R1 = 3,4,5-CH3O–C6H2; R2 = CO2C2H5
10, R1 = 4-CH3O–C6H4; R2 = CO2CH3
11, R1 = 3-CH3O–C6H4; R2 = CN
1
2
3
4
5
6
7
8
9
4, R1 = C6H5; R2 = CO2CH3
15, 91
16, 71
17, 90
18, 68
19, 82
20, 65
21, 73
22, 80
23, 85
24, 75
5, R1 = 4-O2N–C6H4; R2 = CO2CH3
6, R1 = 2-C4H4S; R2 = CO2CH3
7, R1 = 3,4-(OCH2O)–C6H3; R2 = CO2CH3
8, R1 = 3,4,5-(CH3O)3–C6H2; R2 = CO2CH3
9, R1 = 3,4,5-(CH3O)3–C6H2; R2 = CO2C2H5
10, R1 = 4-CH3O–C6H4; R2 = CO2CH3
11, R1 = 3-CH3O–C6H4; R2 = CN
12, R1 = CH3CH2; R2 = CO2C2H5
10
11
13, R1 = CH3(CH2)7CH2; R2 = CO2CH3
14, R1 = 4-F3C–C6H4; R2 = CO2CH3
a
Yields refer to isolated and purified products.
Spectral data (1H, 13C NMR, IR, HRMS) are all compatible with the proposed
12, R1 = C2H5; R2 = CO2C2H5
b
10
13, R1 = C9H19; R2 = CO2CH3
structures.
c
Reaction times varied from 15 min to 96 h.
a
Yields refer to isolated and purified products.
ward manner, and after isolation, only filtration on silica gel was
required to achieve a high level of purity (P98%). The results are
summarized in Table 1.
With the purified adducts in hand, the b-keto esters can be pre-
pared. a-Substituted b-keto esters are classically prepared by Cla-
The change in the solvent results in a net increase in the yield
and rate of the reaction. The yield increased from 55% to 70%,
and the reaction medium was cleaner than when acetonitrile
was used. We also observed a net effect on the rate of reaction.
After only 3 min, the substrate was completely transformed to
the reaction product based on TLC analysis. Therefore, we decided
to test this experimental protocol with the entire set of b-keto
esters. The results are summarized in Table 3.
For some cases, we observed a net increase in the yield of the
reaction (Table 3, entries 6, 8 and 9). For other cases, no changes
were observed (Table 3, entries 2–6). In all of the cases, no decrease
in the yield was observed. Apparently, the increase in the yields is
substantial for those b-keto esters substituted by alkyl groups
(entries 8 and 9, Table 3). However, for all of the cases, the reaction
times substantially decreased from 10 to 3 min. These results are
due to the high polarity of HFIP. This solvent stabilizes the cationic
intermediate formed in this oxidative process, which contributes
to an increase in the yield and rate of the reaction.14
isen condensation following by alkylation or via a Knoevenagel
reaction between b-keto esters and aldehydes followed by hydro-
genation of a double bond. Although these approaches are very
simple, they present some drawbacks. Di-alkylation can occur,
requiring additional purification. Otherwise, depending on the sub-
stituents present in the alkylating agent or in the aldehyde, a step
involving a protecting group could also be required with both reac-
tions (alkylation or Knoevenagel). In both cases, substituted b-keto
esters can be obtained in 2 or 3 steps, if no protecting step is
involved.
Several years ago, Alonso et al.13 described a Mizoroki–Heck
reaction using new oxime-derived palladacycle catalysts. Based
on this catalyst, which provides an in situ source of palladium(0)
nanoparticles, we have developed a phosphine-free method to pre-
To increase the structural pattern diversity of our spiro-hexa-
dienones, we decided to evaluate the feasibility of this methodol-
ogy for the preparation of brominated spirohexadienones due to
the biological potential exhibited by this particular class of spiro-
hexadienones.1 Bromine atoms can be incorporated into our
sequence using two different approaches. First, the bromination
pare mono-a-arylated b-keto esters in good yields and high selec-
tivity using the Mizoroki–Heck reaction and the MBH adducts as
substrates.11 This approach provides the required b-keto esters in
one step from a MBH without the need of any protecting group.
Therefore, a solution of the MBH adduct in DMF was treated with
iodophenol in the presence of a catalytic amount (0.5 mol %) of
the oxime-derived palladacycle (Nájera’s catalyst) to afford the
required
a-arylated-b-keto esters in good yields, low reaction
O
times and good selectivity. The results are summarized in Table 2.
In our previous communication,11 the key phenolic oxidation
step, which is mediated by a hypervalent iodine reagent, exhibited
only moderate yields, which decreased the overall yields of our
synthetic sequence. Initially, we decided to optimize this experi-
mental condition by changing the solvent used in the oxidative
step. Dohi et al.14 reported the use of hexa-fluoroisopropanol
(HFIP) in phenolic oxidation mediated by hypervalent iodine
reagents. Due to its particular physical and chemical characteristics
(i.e., low nucleophilicity, high polarity and high hydrogen-bond
donor ability), this solvent increases the yield of this oxidative step.
We decided to carry out a phenolic oxidation with HIFP and PIFA
[phenyliodine bis(trifluoroacetate)] and compare this reaction to
the one carried out in acetonitrile. The b-keto ester 15 was used
as a model for both reactions. The results are shown in Scheme 1.
PIFA, CH3CN
O
- 10 0C, 10 min.
55%
O
O
CO2CH3
OCH3
26
O
OH
PIFA, HFIP
15
O
- 10 0C, 3 min.
70%
CO2CH3
26
Scheme 1. Testing a fluorinated solvent for the phenolic oxidation of substituted b-
keto esters.