Iron(III) 2-ethylhexanoate as a novel, stereoselective hetero-Diels-Alder catalyst

Article abstract of DOI:10.1039/a706194c

Iron(III) 2-ethylhexanoate has been used as a novel, mild Lewis acid catalyst for the stereoselective Diels-Alder reaction of ethyl (E)-4-oxobutenoate with alkyl vinyl ethers to stereoselectively produce cis-2-alkoxy-3,4-dihydro-2H-pyran-4-carboxylic acid, ethyl esters with diastereoisomeric excesses (de) as high as 98%.

Full text of DOI:10.1039/a706194c

View Article Online / Journal Homepage / Table of Contents for this issue
Iron(iii) 2-ethylhexanoate as a novel, stereoselective hetero-Diels–Alder
catalyst
David B. Gorman* and Ian A. Tomlinson
The Dow Chemical Company, Midland, Michigan 48674, USA
Iron(III) 2-ethylhexanoate has been used as a novel, mild
Lewis acid catalyst for the stereoselective Diels–Alder
reaction of ethyl (E)-4-oxobutenoate with alkyl vinyl ethers
to stereoselectively produce cis-2-alkoxy-3,4-dihydro-
2H-pyran-4-carboxylic acid, ethyl esters with diastereoiso-
meric excesses (de) as high as 98%.
90% at 80 °C (entries 2–5 vs. entries 7–10 vs. entry 15). For
reaction mixtures at elevated temperatures, yields could be
improved by adding a trace amount of Et3N as an inhibitor; at
levels above 1 wt% Et3N in the reaction of 2, yields and
selectivity began to decrease, indicating Et3N was poisoning the
catalyst (entries 15–19). Reaction times could also be shortened
by adding additional catalyst, but optimal yields of !80% were
obtained using 0.20–0.25 equiv. of 1 (entries 3, 8 and 13).
Modest yield increases also were realized by increasing the
amount of vinyl ether from 6 to 10 equiv. (entries 12 and 14 vs.
entries 11 and 13).
Since its discovery over 65 years ago,1 the Diels–Alder reaction
has been one of the most versatile reactions for assembling six-
membered rings. The use of Lewis acid catalysts has led to
improved regio- and stereo-selectivity for many of these
reactions. Lanthanide reagents have been used as very mild
Lewis acid catalysts for the Diels–Alder reaction of a,b-
unsaturated aldehydes or esters with vinyl ethers to give
substituted dihydropyrans.2 In a search for cost-effective
alternative catalysts, we have discovered that iron(iii) 2-ethyl-
hexanoate 1 is a novel, mild and economical Lewis acid catalyst
for carrying out highly stereoselective Diels–Alder reactions
between ethyl (E)-4-oxobutenoate 23 and alkyl vinyl ethers
3a–3c to give 2,4-disubstituted pyrans 4a–c and 5a–c
(Scheme 1).
Based on analogous Diels–Alder transformations to give
2,4-substituted tetrahydropyrans, the major products have been
assigned as the endo isomers.4 Effectively an inverse electron
demand hetero-Diels–Alder reaction, cyclization occurs se-
lectively through the endo transition state, regio- and stereo-
selectively producing the cis-2,4-disubstituted pyrans 4). This
endo selectivity is enhanced using 1 as a catalyst (Table 1),
giving diastereoisomeric excesses (de) as high as 98% based on
relative product peak ratios obtained by gas chromatography.
Similar yields and selectivities have been obtained when ethyl,
isobutyl, and n-butyl vinyl ethers are used.
To isolate gram quantities, the product was removed from the
catalyst by selective precipitation followed by column chroma-
Table 1 Comparison of vinyl ethersa
Entry
Ether
T/°C
t/d
Yield (%)
De (%)
1
2
3
3a
3b
3c
23
23
23
14
14
14
78
80
79
98
98
98
a
Each reaction mixture contained ethyl (E)-4-oxobutenoate, 6 equiv. of
vinyl ether and 0.1 euqiv. of iron(iii) 2-ethylhexanoate. Ethyl
(E)-4-oxobutenoate was inhibited with a trace amount of hydroquinone.
Reported yields are GLC yields based on an internal standard. Yields at 66
h were calculated to be 72, 79 and 71% for entries 1, 2 and 3,
respectively.
Table 2 Optimization reactionsa
Entry
T/°C
t/h
1 (equiv.) Yieldb (%)
De (%)
1
2
3
23
23
23
23
23
60
60
60
60
60
60
60
60
60
80
80
80
80
80
96
96
96
96
96
18
18
18
18
18
16
16
16
16
20
20
20
20
20
0
2
66
81
78
70
13
72
80
79
67
72
75
80
84
64
73
67
56
47
—
98
98
98
98
66
96
96
98
98
96
96
96
96
90
92
90
82
78
The reaction of 2 and 3b has been examined with varying
amounts of 1 at various temperatures (Table 2). Clearly,
selectivity and reaction rate were enhanced by the addition of 1
(entry 6 vs. entries 7–10). The highest selectivity was seen at
room temperature. Increasing reaction temperature shortened
the reaction time from a few days to a few hours, but the de
decreased from 98% at room temperature to 96% at 60 °C and
0.10
0.25
0.50
1.00
0
4
5
6c
7c
0.10
0.25
0.50
1.00
0.10
0.10
0.20
0.20
0.10
0.10
0.10
0.10
0.10
8c
9c
10c
11
12d
13
14d
15e
16e
17e
18e
19e
O
OR
+
CO2Et
2
3a–c
Fe(BuEtCHCO2)3
1
O
OR
O
OR
a Each reaction mixture contained ethyl (E)-4-oxobutenoate, isobutyl vinyl
ether and iron(iii) 2-ethylhexanoate. Ethyl (E)-4-oxobutenoate was in-
hibited with a trace amount of hydroquinone. Unless otherwise specified, 6
equiv. of isobutyl vinyl ether were used. b Reported yields are GLC yields
based on an internal standard. c The reaction mixtures were stirred at room
temperature for 18 h before heating to 60 °C for 18 h. d 10 equiv. of isobutyl
vinyl ether were added instead of 6 equiv. e Et3N was added in the following
weight percents relative to 2: 0% for entry 15, 1% for entry 16, 5% for entry
17, 15% for entry 18 and 25% for entry 19.
+
CO2Et
CO2Et
99:1
4a–c
5a–c
a R = Et
b R = Bui
c R = Bun
Scheme 1
Chem. Commun., 1998
25

View Article Online
Table 3 Comparison of Iron-Based Catalystsa
Entry
Catalyst
T °C
t/h
2 (%)
Yieldb (%)
De (%)
1
2
3
4
5
6
7
Iron(iii) 2-ethylhexanoate
Iron(ii) 2-ethylhexanoate
Iron(iii) naphthenate
Iron(iii) benzoate
Iron(ii) stearate
70
70
70
70
70
70
70
6
6
5
5
4
4
6
15
29
21
44
59
71
71
58
38
30
29
18
14
10
96
92
92
86
86
80
64
Iron(iii) stearate
—
a All reactions were carried out in sealed glass tubes containing 2, 6 equiv. of ethyl vinyl ether and 0.1 equiv. of catalyst. b Reported yields are GLC yields
based on an internal standard.
layer was separated and dried (MgSO4). The solution was decanted and the
tography on silica gel which had been deactivated by mixing
MeCN evaporated. The remaining brown oil was flushed through a 0.75B 3
12B deactivated silica gel column, eluting with hexane, followed by 20%
EtOAc in hexane. After concentrating, a clear yellow oil (3.36 g) was
obtained, assaying at 97% of 4a as a single isomer for a calculated yield of
3.26 g (81%).
with Et3N in CH2Cl2 and evaporating to dryness. For larger
samples, the product was isolated either by chromatography or
wiped-film evaporation.
Although the exact binding nature of the catalyst is unknown,
the choice of metal ligands and metal oxidation state is critical
to the success of the catalytic reaction. Similar catalysts
containing lower metal oxidation states and/or ligands which
are bulkier and/or more rigid give lower selectivity, as well as
lower reaction yields (Table 3). The bulkier/more rigid ligands
may cause steric interference with the ethyl (E)-4-oxobutenoate,
where complexation is believed to take place. A lower oxidation
state may decrease the catalyst electrophilicity enough that it
does not coordinate well with any of the oxygens from 2.
In summary, iron(iii) 2-ethylhexanoate is a novel, mild and
relatively inexpensive Lewis acid catalyst for the stereoselec-
tive Diels–Alder reaction of ethyl (E)-4-oxobutenoate with
alkyl vinyl ethers to give cis-2,4-disubstituted pyrans, extending
the range of known catalysts for these types of trans-
formations.
(b) purification by wiped-film evaporation: A mixture of of 52% iron(iii)
2-ethylhexanoate (96.5 g) in mineral spirits and (±)-bis(2-ethylhexyl)
sebacate (50.3 g) was dripped down a wiped-film still over 2 h operating at
a vacuum of 0.05 mmHg, a heating temperature of 60 °C, a condensing
temperature of 4 °C and a spinning rate of 300 rpm. The recovered bottoms
(99.4 g) were assumed to be 50% catalyst by weight. A mixture of ethyl
(E)-4-oxobutenoate (25.7 g, 0.20 mol) n-butyl vinyl ether (120.4 g, 1.20
mol) Et3N (0.3 g), hydroquinone (0.026 g), and 50% iron(iii) 2-ethylhex-
anoate in (±)-bis(2-ethylhexyl) sebacate (18.6 g, 0.02 mol) was heated at
60 °C for 12 h. After evaporating excess vinyl ether, the residue was dripped
at 1 drop per second down a wiped-film still operating at a vacuum of 0.1
mmHg, a heating temperature of 90 °C, a condensing temperature of 2 °C
and a spinning rate of 300 rpm. The overheads (30.2 g) assayed at 93% of
4a for a calculated yield of 28 g (62%).
1 For selected reviews of the Diels–Alder reaction, see U. Pindur, G. Lutz
and C. Otto, Chem. Rev., 1993, 93, 741; H. B. Kagan and O. Riant, Chem.
Rev., 1992, 92, 1007; G. Desimoni and G. Tacconi, Chem. Rev., 1975, 75,
651; A. Wasserman, Diels–Alder Reactions, Elsevier, New York, 1965;
H. L. Holmes, Org. React., 1948, 4, 60.
2 J.-Y. Merour, A.-S. Bourlot and E. Desarbre, Tetrahedron Lett., 1995, 36,
3527; I. E. Marko, G. R. Evans, J.-P. Declercq, B. Tinant, J. Feneau-
Dupont, Acros Organics Acta, 1995, 1, 2, 63; S. Danishefsky,
M. Bednarski, J. Am. Chem. Soc., 1983, 105, 3716; S. Danishefsky,
M. Bednarski, Tetrahedron Lett., 1984, 25, 721.
We thank Don Zakett for determining mass spectral data for
4a–c and 5a–c.
Footnotes and References
* E-mail: dbgorman@dow.com
† Iron(iii) 2-ethylhexanoate was purchased as a 52% solution in mineral
spirits from Johnson Matthey Alfa Aesar. Ethyl (E)- 4-oxobutenoate was
purchased from Lonza. Representative experimental procedures. (a)
Purification by chromatography: A mixture of ethyl (E)-4-oxobutenoate
(2.56 g, 0.02 mol): n-butyl vinyl ether (8.65 g, 0.12 mol) and 52% iron(iii)
2-ethylhexanoate (1.86 g, 0.002 mol) in mineral spirits was stirred at room
temperature for 44 h. The catalyst precipitated after adding 20 ml of 20%
MeCN in water. The slurry was filtered through filter agent, rinsing the cake
with 20% MeCN in water. The filtrate was saturated with NaCl. The organic
3 For preparation of ethyl (E)-4-oxobutenoate by ozonolysis of ethyl
sorbate, see V. Tarik, L. A. Mitscher and J. K. Dep, Synthesis, 1980, 10,
807.
4 W. G. Dauben and H. O. Krabbenhoft, J. Org. Chem., 1977, 42, 282.
Received in Corvallis, OR, USA, 25th August 1997; 7/06194C
26
Chem. Commun., 1998