Activation of molecular oxygen and its use in stereoselective tetrahydrofuran-syntheses from δ,ε-unsaturated alcohols

Article abstract of DOI:10.1039/b804588g

Bishomoallylic alcohols (pent-4-en-1-ols) underwent efficient oxidative cyclizations, if treated with O₂ and bis{2,2,2-trifluoromethyl-1- [(1R,4S)-1,7,7-trimethyl-2-(oxo-κO)bicyclo[2.2.1]hept-3-yliden] ethanolato-κO}cobalt(ii) in solutions of 2-propanol at 60 °C. Ring closures occurred diastereoselectively and afforded 2,3-trans- (96% de), 2,4-cis- (～60% de), and 2,5-trans-substituted (>99% de) (phenyl)tetrahydrofur-2-ylmethanols as major components. Formation of bicyclic compounds and a 2,3,4,5-substituted oxolane was feasible as exemplified by syntheses of oxabicyclo[4.3.0]nonylmethanols and a derivative of natural product magnosalicin in 61-72% (90-99% de). The effectiveness of tetrahydrofuran synthesis was critically dependent on (i) solvent, (ii) reaction temperature, (iii) initial cobalt concentration, (iv) chain length between hydroxyl and vinyl groups, and (v) substitution at reacting entities. A sequence is proposed for rationalizing observed selectivities.

Full text of DOI:10.1039/b804588g

View Article Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/obc | Organic & Biomolecular Chemistry
Activation of molecular oxygen and its use in stereoselective
tetrahydrofuran-syntheses from d,e-unsaturated alcohols†‡
Ba´rbara Mene´ndez Pe´rez, Dominik Schuch and Jens Hartung*
Received 17th March 2008, Accepted 5th June 2008
First published as an Advance Article on the web 16th July 2008
DOI: 10.1039/b804588g
Bishomoallylic alcohols (pent-4-en-1-ols) underwent efficient oxidative cyclizations, if treated with O2
and bis{2,2,2-trifluoromethyl-1-[(1R,4S)-1,7,7-trimethyl-2-(oxo-jO)bicyclo[2.2.1]hept-3-
yliden]ethanolato-jO}cobalt(II) in solutions of 2-propanol at 60 ◦C. Ring closures occurred
diastereoselectively and afforded 2,3-trans- (96% de), 2,4-cis- (∼60% de), and 2,5-trans-substituted
(>99% de) (phenyl)tetrahydrofur-2-ylmethanols as major components. Formation of bicyclic
compounds and a 2,3,4,5-substituted oxolane was feasible as exemplified by syntheses of
oxabicyclo[4.3.0]nonylmethanols and a derivative of natural product magnosalicin in 61–72% (90–99%
de). The effectiveness of tetrahydrofuran synthesis was critically dependent on (i) solvent, (ii) reaction
temperature, (iii) initial cobalt concentration, (iv) chain length between hydroxyl and vinyl groups, and
(v) substitution at reacting entities. A sequence is proposed for rationalizing observed selectivities.
tetrahydrofuran synthesis had to be developed (Scheme 1). The
Introduction
major results of a study performed on the basis of the given
key notes showed that acceptor-substituted camphor-derived
cobalt(II) chelates were able to catalyze formation of trans-2,3-,
cis-2,4-, and trans-2,5-substituted tetrahydrofur-2-ylmethanols
from underlying bishomoallylic alcohols and O2. Addition of
TBHP was not required for obtaining yields between 60–80%.
The newly developed procedure was furthermore applicable in
syntheses of tetrasubstituted and bicyclic oxolanes. Its success,
In a series of papers it was reported that acceptor-substituted
bis{[2-(oxo-jO)prop-1-yliden]methanolato-jO}cobalt(II) com-
plexes, i.e. compounds derived from b-diketonate anions and
Co(II), were able to induce oxidative cyclization of 1-substituted
bishomoallylic alcohols (pent-4-en-1-ols), if heated with a
1–1.5 molar excess of tert-butyl hydroperoxide (TBHP) in
an atmosphere of molecular oxygen.1–4 The reaction afforded
5-substituted tetrahydrofur-2-ylmethanols with a notable degree
of 2,5-trans diastereoselection and has been applied in a number
of natural product syntheses.3–6
The mechanism of the aerobic tetrahydrofuran synthesis is
subject to debate, since cobalt(II) chelates combine affinity for
dioxygen binding,7–13 with a marked propensity for alkyl hydroper-
oxide decomposition in a Fenton-type manner.14–17
Given the fact that functionalized tetrahydrofurans occur widely
in nature,5,18 we became interested in the scope of aerobic alkenol
oxidation19,20 as a tool in stereoselective heterocycle synthesis.
Four major achievements were considered necessary for a more
profound validation of this reaction. (i) Parameters for conducting
tetrahydrofur-2-ylmethanol syntheses using O2 as terminal oxi-
dant had to be identified and applied to a larger set of 1-, 2-,
and 3-substituted pent-4-en-1-ols. (ii) Guidelines for predicting
selectivity in aerobic oxidative cyclizations had to be derived.
(iii) An extension of the method toward tri- and tetrasubstituted
Fachbereich Chemie, Organische Chemie, Erwin-Schro¨dinger Straße,
Technische Universita¨t Kaiserslautern, D-67663 Kaiserslautern. E-mail:
hartung@chemie.uni-kl.de; Fax: +49 631 205 3921
† The following abbreviations have been used: AIBN
=
a,a-
azobis(isobutyronitrile), CHD = cyclohexa-1,4-diene, DBPO = dibenzoyl
peroxide, TBHP = tert-butyl hydroperoxide. Products of oxidative ring
closure were obtained as racemates. The presented structural formulae of
these molecules were drawn by arbitrarily selecting one of the enantiomers.
‡ Electronic supplementary information (ESI) available: Standard in-
strumentation, protocol for quantitative acetone analysis, procedure for
determining enantiomeric purity via P NMR, and results from aerobic
oxidation in 2-propanol-d8. See DOI: 10.1039/b804588g
Scheme 1 Objectives for the development of sustainable tetrahydrofuran
syntheses via cobalt-catalyzed aerobic alkenol oxidation (A, R = alkyl,
aryl; RE , RZ = H, CH3; H2X = coreductant).
3532 | Org. Biomol. Chem., 2008, 6, 3532–3541
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
however, was critically dependent on solvent, temperature, cobalt
complex concentration, and alkenol constitution.
the IR spectrum of the residue showed apart from two additional
new IR-absorptions at 1205 and 1147 cm−1 similar bands. UV/Vis
absorptions of this material were located at 270 and 600 nm, and
a shoulder at 361 nm (iPrOH). Chromatographic purification
for separating organic components from the residue afforded
3-oxa-1,8,8-trimethylbicyclo[3.2.1]octane-2,4-dione, i.e. camphor
acid anhydride28 (0.08 mmol), a 50 : 50 diastereomeric mix-
ture of 4-isopropoxy-1,8,8-trimethylbicyclo[3.2.1]-3-oxaoctan-2-
one (0.37 mmol, ESI), and at least one additional camphor-derived
product of hitherto unknown constitution.
Results
1. Synthesis and characterization of bis{1-[(1R,4S)-1,7,7-
trimethyl-2-(oxo-jO)bicyclo[2.2.1]hept-3-yliden]methanolato-
jO}cobalt(II) complexes
Auxiliaries 1–4 that allowed a swift and efficient survey of pa-
rameters that control reactivity and selectivity in oxidative cobalt-
catalyzed alkenol ring closure (see also Scheme 1) were prepared
via a-acylation of (1R,4R)-1,7,7-trimethylbicyclo[2.2.1]heptan-2-
one, i.e. (+)-bornan-2-one, (+)-camphor, in extension to literature
procedures.21,22 Treatment of diketones with Co(OAc)2·4H2O in
EtOH (for 1–3, Table 1, entries 1–3)23–25 or in a mixture of aq.
EtOH and aq. NaOH (for 4, Table 1, entry 4)25 furnished cobalt
complexes as red–brown (5, 726), red (6),27 or orange (8) crystalline
solids. In view of its propensity to liberate diketone 1 upon drying
at 20–25 ◦C under reduced pressure, compound 5 was applied
as an in situ preparation. Cobalt complexes were characterized
via MALDI-TOF (5–8), UV/Vis- and IR-spectroscopy (5–8), and
combustion analysis (6, 8). The compounds were readily soluble
in polar organic solvents, e.g., EtOH (except for 8), 2-propanol,
THF, Et2O, acetone, and CH2Cl2. They were insoluble in H2O
or aliphatic hydrocarbons. In contrast to complex 8, cobalt(II)
compounds 5–7 showed a marked affinity for solvate formation,
as evident from combustion analysis.
3. Oxidative cyclizations
3.1 Establishing parameters for oxidative pent-4-en-1-ol-
cyclization. The oxidation of 2,2-dimethylhept-6-en-3-ol (9a)29
was selected as a reporter reaction for establishing parameters for
efficient aerobic alkenol ring closures. Type, amount, and initial
concentration of cobalt reagents (i.e. 5–8), additive, solvent, and
means of thermal activation were varied.
If stirred in a solution of 2-propanol at 60 ◦C in a stationary O2
atmosphere (>95%), cobalt(II)-complexes 5–8 induced formation
of tetrahydrofuran 10a30 with a marked preference for the trans-
isomer. The yield of product 10a gradually decreased along the
sequence of cobalt(II) complexes 6 (R = CF3) > 8 [R = 3,5-
(CF3)2C6H3] > 7 (R = C6H5) ∼ 5 [R = C(CH3)3] (Table 2, entries
1–4). The product was obtained as a racemate, regardless whether
the material was isolated at conversions below or above 50%. Its
enantiomeric purity was determined by 31P NMR with the aid of a
chiral phosphorous-based derivatization reagent.31 No conversion
of 9a took place, upon substituting Co(OAc)2·4H2O for, e.g.,
complex 6 (GC). Purging of O2 through solutions of 6 and alkenol
9a in 2-propanol at 60 ◦C furnished complete turnover of the
substrate, however, without providing a satisfactory mass balance
(not shown). The use of a monomode microwave instrument§
for inducing thermal activation was associated with shorter
reaction times but also with a significant increase in side product
formation (e.g. Table 2, entry 5).
Table 1 Conversion of diketones 1–4 into cobalt(II) chelates 5–8
Entry
1–4
R
Yield of 5–8 [%]
The yield of product formation was critically dependent on
reactant concentration. In a separate study (not shown) it was
found that a solution containing 1 mmol of 9a in 8 mL of solvent
afforded the most efficient turnover. A precatalyst concentration
of 1.25 × 10−2 M (10 mol%) of 6 posed a reasonable balance
between selectivity and time–yield factor. A further investigation
1a
2a
3a
4c
1 (HL1)
2 (HL2)
3 (HL3)
4 (HL4)
C(CH3)3
CF3
C6H5
5:—b
6: quant.
7: 60
8: quant.
3,5-(CF3)2C6H3
a Reaction in EtOH at 20 ◦C. b In situ-preparation. c In EtOH–H2O, 50 :
50 (v/v) at 20 ◦C.
showed that initial concentrations of complex 6 below 10−2
M
provided slower but more chemoselective conversion of substrate
9a (not shown). Since catalyst deactivation gradually occurred,
a complete conversion of alkenol 9a was not attainable within
15 h, if cobalt concentrations fell below 6.0 × 10−3 M. The
UV/Vis spectrum of the spent catalyst pointed to formation of a
cobalt(III) residue of unknown constitution. UV/Vis absorptions
of the green material [k/nm = 271, 309sh, 362sh, 607] and
strong IR bands at 1203 and 1135 cm−1 were similar but not
identical to spectral information of a product originating from
the reaction between 6 and O2 (see above).32 Addition of typical
Co(III)→Co(II) reductants,33 e.g. formate, hypophosphite, or L-
ascorbate provided a change in color from brown to pink, however,
2. Reactivity of bis[b-diketonato(−1)]cobalt(II) complexes
toward dioxygen
Cobalt(II) complex 6 served for reasons that are outlined below as
a benchmark for probing the reactivity of cobalt(II) compounds
in combination with O2 toward organic substrates. A solu-
tion of bis[3-trifluoroacetylcampherato(−1)]cobalt(II) complex 6
(1.00 mmol, co = 0.10 M) in 2-propanol (20 ◦C) absorbed O2
with a steadily decreasing rate. Gas uptake was paralleled by a
gradual change in color from red (6) via brown to green. It levelled
off after a consumption of 3.21 mmol of the oxidant (∼140 h).
Product analysis indicated formation of acetone (3.21 mmol) and
H2O (3.20 mmol, see also Table 4). Removal of the volatiles from
the solution afforded a green solid. If compared to complex 6,
§ All microwave-assisted experiments were performed using special glass-
ware and microwave equipment.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3532–3541 | 3533
©

View Article Online
Table 2 Solvent and catalyst effects in cobalt(II)-catalyzed oxidations of alkenol 9a
Entry
Co(II) [%]
Solventa
T/◦C
t/h
Convn. [%]b
Yield [%] (cis : trans)b
1
2
5 [10]c
6 [10]
7 [10]
8 [10]
6 [10]
6 [20]
6 [20]
6 [20]
6 [20]
6 [10]
6 [10]
6 [10]
6 [10]
6 [20]
iPrOH
60
60
60
60
60
60
60
60
60
60
60
60
60
20
15
24
91
14
3 (11 : 89)
iPrOH
2.5
3.0
3.0
1.5e
15f
15f
15f
15f
2.5
2.5
2.5
2.5
48f
63 (< 1 : 99)d
4 (< 1 : 99)d
42 (1 : 99)
3
iPrOH
4
5
6
7
8
9
10
11
12
13
14
iPrOH
87
iPrOH
quant.
quant.
quant.
quant.
48
70
42
85
98
85
50 (2 : 98)
EtOH
50 (< 1 : 99)d
32 (< 1 : 99)d
31 (< 1 : 99)d
25 (8 : 92)
tBuOH
cC5H9OH
CF3CH2OH
ꢀ
g
=
RR C
tBuCHO
EtCH(OEt)2
O
9h (15 : 85)
< 1i (< 1 : 99)d
42j (< 1: 99)d
8l (< 1 : 99)d
28 (6 : 94)
k
CHD–C6H6
CH2Cl2
a 8 mL of p.a. grade solvent. b Quantitative GC measurements; er = 50 : 50 for trans-10a. c In situ formation of 5. d cis-10a not detected (GC). e Microwave
f
heating (300 W, monomode instrument). MS 4 A added. g 2-Methylcyclohexanone. h Additional product: 2-tert-butyl-5-methyltetrahydrofuran
˚
11a (cis : trans < 1 : 99, 36%). i additional product: 2-tert-butyl-5-methyltetrahydrofuran trans-11a (40%). j Additional products: 2-tert-butyl-5-
methyltetrahydrofuran trans-11a (13%), 4-tert-butyl-d-valerolactol 14a (10%). k Cyclohexa-1,4-diene : benzene 50 : 50 (v/v). l Additional product: 2-
tert-butyl-5-methyltetrahydrofuran trans-11a (90%).
without restoring catalytic activity for oxidative bishomoallylic
alcohol cyclization. Addition of NH4SCN to the 1-pentanol
soluble fraction of, e.g., hypophosphite-reduced cobalt residue
obtained from aerobic oxidation of 6 in iPrOH afforded an
UV/Vis absorption at 620 nm, which was indicative of Co(SCN)2.
Addition of one of the reductants to an aerated solution of alkenol
9a and complex 6 in 2-propanol completely inhibited formation
of tert-butyltetrahydrofurylmethanol 10a.
but became significant (90%), if substrate 9a was oxidized in a 50 :
50-mixture (v/v) of 1,4-cyclohexadiene (CHD) and C6H6. This
observation called for more complete mass balancing, which is
summarized in section 3.2.
Addition of tert-butyl hydroperoxide (TBHP), either as 70%
(w/w) aqueous solution or as anhydrous 5.5 M solution in nonane
did not improve the efficiency of the conversion 9a→10a (Table 3,
˚
entries 1–2). Molecular sieves (4 A) were added in those instances
Systematic variation showed a gradual increase in yield
of tetrahydrofuran 10a along the series of applied sol-
vents pivalaldehyde < cyclohexa-1,4-diene (CHD)–C6H6 < 2-
methylcyclohexanone < CF3CH2OH < CH2Cl2 < cC5H9OH ∼
tBuOH < 1,1-diethoxypropane < EtOH < iPrOH (Table 2, entries
for comparing data to related values from the literature.1 Addition
of typical radical initiators, such as a,a-azobis(isobutyronitrile)
(AIBN) or dibenzoyl peroxide (DBPO), were ineffective for
improving yields of target compound 10a (Table 3, entries 3–4).
˚
2, 6–14). The use of molecular sieves (4 A) had no effect (not
3.2 Product selectivity and mass balancing. In a second
part of the study, mass balancing was performed for obtaining
additional information on product- and chemoselectivity of
aerobic cobalt-catalyzed oxidation of alkenols 9a–b.34,35 Thorough
shown).
2-tert-Butyl-5-methyltetrahydrofuran (11a)29 was consistently
found as a side product. Its yield generally remained below 15%
Table 3 Effect of additives in aerobic cobalt(II)-catalyzed oxidations of alkenol 9a
Entry
Additive
T/◦C
t/h
6 [mol%]
Convn. of 9a [%]a
10a [%] (cis : trans)a
b
d
56 (< 1 : 99)c
62 (< 1 : 99)c
63 (< 1 : 99)f
35 (1 : 99)g
˚
˚
1
2
3
4
TBHP–MS 4 A
50
50
60
60
5
5
2.5
2.5
20
20
10
10
82
quant.
94
TBHP–MS 4 A
AIBNe
DBPOe
83
a Quantitative GC measurements. b 1.0 equiv. of a 70% (w/w) aqueous solution. c cis-10a was not detected (GC). d 1.0 equiv. of a 5.5 M anhydrous solution
in nonane. e 1.0 equiv. of a,a-bis(isobutyronitrile) (AIBN) or dibenzoyl peroxide (DBPO). f Additional product: 4-tert-butyl-d-valerolactol 14a (13%).
g Additional products: 2-tert-butyl-5-methyltetrahydrofuran 11a (15%), 4-tert-butyl-d-valerolactol 14a (7%).
3534 | Org. Biomol. Chem., 2008, 6, 3532–3541
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
Table 4 Mass balancing in aerobic cobalt(II)-catalyzed oxidations of 1-substituted pent-4-en-1-ols 9a and 9b
Yield [%]b
Entry
9–17
R
[O2]a [%]
Conv. [%]b
10
11
12
13
14
15
16
17
d
d
d
d
1c
2f
3c
4c
5f
6c
a
a
b
b
b
b
tBu
tBu
Ph
Ph
Ph
>95
>95
>95
50
91
98
75
99
99
98
63
60
49
63
59
51
—
2
—
5
9
20
6
5
4
5
3
4
1e
11
15
4
6
6
4
6
—
—
—
—
3
2
13
13
17
7
—
—
4
2
2
7g
d
d
d
d
d
d
d
d
—
—
—
—
50
Ph
∼21h
4
1
a Gas phase oxygen concentration (v/v). b Quantitative GC measurements. c 10 mol% of 6. d Not detected. e dr = 62 : 38. f Catalyst generated in situ from
10 mol% of Co(OAc)2·4 H2O and 20 mol% of 2. g dr = 55 : 45. h Air.
Table 5 Quantitative analysis of H2O and acetone formed in aerobic
chromatographic separation was supplemented by independent
syntheses to furnish the set of compounds 10–1736–39 that consis-
tently accounted for 97–99% of products (Table 4, entries 1 and
3). The yields were dependent on initial O2 concentration in the
gas phase. A gradual increase in yield of target compound 10b was
noted along the series of applied O2 volume percentages ∼95% <
∼21% < 40% ∼ 60% (the latter two not shown) < 50% (Table 4,
entries 3–6). In situ preparations or independently synthesized
cobalt reagent 6 afforded comparable results (Table 4, entries 1–2,
4–5). Treatment of 5-phenylpent-1-ene40 with O2 and 10 mol% of
catalyst 6 in 2-propanol (60 ◦C) furnished 91% of recovered olefin
and no detectable compounds of oxidative conversion (GC/MS,
not shown).
cobalt(II)-catalyzed oxidations in 2-propanol
Yield [%]
Conv. of
Entry
9
R
pO2/bar 1 [%]a
10a
Acetoneb H2Oc
1
2
3
4
9ad tBu 3.0
9a tBu 1.5
9bd Ph 3.0
9b Ph 1.5
80
82
66
71
1
4
10a: 50
10a: 53
10b: 25
10b: 20
2
2
60
53
47
46
6
5
78
71
64 10
67
8
a Determined via GC. b HPLC analysis of derived 2,4-dinitro-
3.3 Probing for secondary products—acetone and H2O. Reac-
tion mixtures originating from aerobic cobalt-catalyzed oxidation
of alkenols 9a and 9b in 2-propanol were subjected to quantitative
acetone and H2O analysis using appropriate analytical techniques
(Table 5, entries 1–4).41,42 The yields of both products, were
corrected for trace amounts of acetone and water that were present
in substrates 9a–b, analytical grade iPrOH, and a solution of
cobalt complex 6 that was stirred for 2.5 (9a)–3 h (9b) in analytical
grade 2-propanol at 60 ◦C in an atmosphere of O2.
phenylhydrazone.41 c Karl–Fischer-titration.42 d Mean value
deviation as determined from 5 independent runs.
standard
2-Methyl-substituted tetrahydrofurans, i.e. 11b–d,34 were iden-
tified in all instances as side products. Diastereoselectivity of
the latter corresponded to values determined for hydroxyl-
substituted derivatives 10b–d. Enantiomeric purities of alcohols
10b–d were checked via 31P-NMR, after derivatization with
a chiral dioxaphospholane.31 Products 10b–d were consistently
formed as racemates. This result was independent from whether
samples from an early phase of the reaction or after complete
consumption of substrates 9b–d were collected for derivatization.
In a similar manner, diastereomeric ratios of 2,4-disubstituted
tetrahydrofurans 10c and 11c were independent from the degree of
conversion of 9c.43 A decrease in amount of cobalt complex 6 from
20 to 5 mol% generally reduced rates of turnover of 9b–d but also
improved overall selectivity by inhibiting 2-methyltetrahydrofuran
formation (Table 6, entries 2–3, 5–7, 9–10).
4. Stereoselectivity
4.1 Terminal unsubstituted alkenols. Variation of phenyl
substituent positioning furnished the set of substrates 9b–d for
exploring stereoselectivity in aerobic cobalt-catalyzed alkenol ring
closures associated with groups attached to position 1, 2, and 3
in pent-4-en-1-ols. If exposed for 15 h to an atmosphere of O2
◦
in a solution of 2-propanol (60 C) containing cobalt(II) reagent
6, alkenols 9b–d furnished disubstituted tetrahydrofurans 10b–
d30 as major products (Table 6, entries 1, 4, 8). All oxidations
proceeded diastereoselectively and afforded trans-10b (>99% de),
cis-10c (56% de), and trans-10d (96% de) as major components.
In a further set of experiments, the effects of an additional
substituent attached to position 1 and chain elongation between
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3532–3541 | 3535
©

View Article Online
Table 6 Product formation in aerobic oxidation of phenyl-substituted bishomoallylic alcohols
Entry
R1
R2
R3
6 [mol l−1] (mol%)a
9–11
Conv. of 9 [%]b
Yield of 10 [%]b (cis : trans)
Yield of 11 [%]b (cis : trans)
1
2
3
4
5
6
7
8
9
Ph
Ph
Ph
H
H
H
H
H
H
H
H
H
H
Ph
Ph
Ph
Ph
H
H
H
H
H
H
H
H
H
H
Ph
Ph
Ph
0.025 (20)
0.013 (10)
0.008 (6.5)
0.025 (20)
0.013 (10)
0.008 (6.5)
0.006 (5.0)
0.025 (20)
0.013 (10)
0.009 (7.5)
b
b
b
c
c
c
c
d
d
d
quant.
99
86
quant.
quant.
quant.
83
quant.
98
48 (< 1 : 99)c,d
63 (< 1 : 99)c,d
56 (< 1 : 99)c,d
55 (78 : 22)d
61 (78 : 22)d
76 (78 : 22)d
72 (80 : 20)d
44 (2 : 98)d
24 (< 1 : 99)c,e
5 (< 1 : 99)c,e
f
—
31 (66 : 34)e
23 (71 : 29)e
2 (< 1 : 99)g
f
—
16 (< 1 : 99)c,e
5 (< 1 : 99)c,e
63 (2 : 98)d
f
10
91
59 (2 : 98)d
—
a 8 mL of reagent grade iPrOH; 15 h reaction time, except for entry 2 (3 h). b GC. c cis Isomer not detected (GC). d er for trans-10b, cis/trans-10c, and
trans-10d = 50 : 50. e er not determined. f Not detected (GC). g trans-11c not detected (GC).
reacting entities were investigated (Scheme 2). In this case, aerobic
oxidation required elevated temperatures in order to occur. Almost
no conversion was detected, if 2-phenylhex-5-en-2-ol (9e)44 was
aerated for 3 h in a solution of 2-propanol at 60 ◦C in the presence
of complex 6. In 2-butanol at 98 ◦C, 83% conversion of 9e was
attainable after a period of 15 h. The reaction afforded 15% of
trisubstituted tetrahydrofuran 10e (cis : trans = 20 : 80) and 39%
of diol 1845 (Scheme 2). In a similar way, oxidation of alkenol
1946 required elevated temperature for substrate conversion to
occur, and extended reaction time to afford a new product in
substantial amounts. The material was isolated and identified as
diol 20 (Scheme 2).
Scheme 3 Stereoselective synthesis of oxabicyclo[4.3.0]nonanes 10f–g
(top), and tetrasubstituted tetrahydrofuran 10h (bottom). acis : trans-ratios
refer to the relative configuration of substituents attached at positions 6
and 8. b1H NMR.
Scheme 2 Aerobic oxidation of 2-phenylhex-5-en-2-ol (9e) and 2,2-
dimethyloct-7-en-3-ol (19).
4 h) using cobalt(II) catalyst 6 (10 mol%, c0 = 5.0 × 10−3 M) and
For investigating whether stereoselectivity in ring closures
prevailed in the case of constrained or multiply substituted
substrates, appropriate alkenols were prepared and aerated in the
a second batch of cobalt reagent 6 (10 mol%) at 75 ◦C (15 h)
provided diastereomerically pure (1H NMR) tetrasubstituted
tetrahydrofuran 10h (72%, Scheme 3, bottom).
presence of complex 6. Treatment of cis- and trans-isomers of
2-allylcyclohexanol 9f–g47 under established conditions afforded
8-substituted oxabicyclo[4.3.0]nonanes 10f (68%) and 10g (61%)
in a 6,8-trans-selective manner (Scheme 3, top).48,49 Oxidation of
diastereomerically pure alkenol 9h50 with O2 in iPrOH (60 ◦C,
4.2 Oxidation of alkenols having methyl substituents attached
at the terminal position of the p-bond. In a final set of oxi-
dations, terminally methyl-substituted alkenols were applied as
3536 | Org. Biomol. Chem., 2008, 6, 3532–3541
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
substrates. It was the aim to screen for polar and steric substituent
effects, since CH3-substitution is known to control stereoselectivity
in other cyclizations, e.g. alkenoxyl radical ring closures and
peroxide-mediated tetrahydrofuran syntheses.14,29 Treatment of a
solution of (E)-1-phenylhex-4-en-1-ol (9i)51 and cobalt reagent
6 for 4 h in iPrOH at 60 ◦C with O2 furnished 34% of 2,5-
trans-disubstituted tetrahydrofuran 10i as a 56 : 44 mixture of
diastereomers with respect to the stereogenic center located in the
side chain. In addition, 17% of dicarbonyl compound 16b was
detected (GC-MS) (Scheme 4).
5-phenylpent-1-ene was applied as substrate. The olefin was inert,
whereas 1-phenylpentenol 9b underwent swift and (stereo)selective
5-exo-ring closure under standard conditions (O2, iPrOH, 60 ◦C,
3 h, 10 mol% of 6). Oxidative disintegration of cobalt reagent
6 in 2-propanol occurred, if exposed to an atmosphere of O2.
It is supposed that this background reaction accounted for
a considerable fraction of catalyst deactivation in the present
investigation.
2. Oxidation catalysis
Yields of tetrahydrofur-2-ylmethanols were dependent on solvent,
reaction temperature, initial cobalt concentration, and alkenol
constitution.
(a) Solvent. Adequate time–yield factors for tetrahydrofur-2-
ylmethanol formation were attainable using 2-propanol as solvent.
The reaction provided acetone and water as secondary products.
Their yields slightly diverged but were clearly correlated with those
of tetrahydrofuran formation. Autoxidations that were identified
as side reactions, were considered to increase the percentage of
reaction water without affecting the acetone balance.
Quantitative alkenol conversion was paralleled by 2-
methyltetrahydrofuran formation. An experiment performed in
2-propanol-d8 at 75 ◦C (see ESI) provided 18% of 2-phenyl-5-
methyltetrahydrofuran 11b with a 50% degree of deuteration,
exclusively at the methyl substituent (1H NMR), which clearly
pointed to the solvent as one of the H-atom sources. The role of
the solvent in the synthesis of these compounds was furthermore
evident from an experiment performed in CHD–C6H6, which
provided 90% of 2-tert-butyl-5-methyltetrahydrofuran 11a. If
bond dissociation energies (381 4 kJ mol−1 for 2-C,H in iPrOH,
and 318 5 kJ mol−1 for 3-C,H in CHD)53 were taken as a guide
for interpretation, CHD is expected to serve as H-atom donor in
homolytic substitutions. The fact that initiators, such as AIBN a◦nd
DBPO, failed to significantly alter product selectivities at 60 C
strongly pointed to a radical trapping and thus terminating effect
of CHD, and not to an active role in radical generation.
Scheme 4 Aerobic oxidation of (E)-1-phenylhex-4-en-1-ol (9i).
1-Phenyl-5-methylhex-4-en-1-ol (9j)30 underwent almost no
conversion if treated with cobalt complex 6 in 2-propanol at
60 ◦C. A change of solvent to 2-butanol allowed the reaction
to be conducted at 98 ◦C, which led to formation of 2,5-
disubstituted tetrahydrofuran 10j (cis : trans = 75 : 25),30 diol
21,52 and hydroxyketone 22 (Scheme 5, top). tert-Butyl-substituted
derivative 9k was oxidized, if treated with O2 and catalyst 6 in a
solution of EtOH (60 ◦C), to furnish 13% of tetrahydrofuran 10k
(cis : trans = 57 : 43) along with 3% of tetrahydropyran 23 (cis :
trans < 1 : 99) (Scheme 5, bottom).
(b) Temperature dependence and initial cobalt(II) concentration.
Effects of temperature and initial cobalt(II) complex concentration
were correlated in terms of rate and selectivity. Temperatures
below 60 ◦C and initial cobalt(II) concentrations of less than
10−2 M reduced rates of oxidative cyclization but improved product
selectivity. Conducting oxidations at T > 60 ◦C, regardless of the
applied solvent (e.g. iPrOH or iBuOH) and at initial cobalt(II)
concentrations of >10−2 M increased rates of alkenol oxidation
but also provided notable amounts of 2-methyltetrahydrofurans,
acyclic alcohols, and products of autoxidation.
Scheme 5 Aerobic oxidation of terminal dimethyl-substituted alkenols
9j–k. acis-23 not detected (GC).
Discussion
(c) Stereoselectivity. Diastereoselection gradually decreased
along substituent positions 1 > 3 > 2. The observed trans-2,3- and
trans-2,5-selectivity of the cobalt method was notable. Explanation
on the lack of enantioselectivity in this chemistry, however, has
to await profound structural investigations. The cobalt atom in,
e.g., complex 6 poses a stereogenic center.27 The configuration at
this site, either in the reagent or in the active oxidant, is entirely
unknown. Formation of diastereomerically pure all-trans-3,5-
diaryl-4-methyl-2-hydroxymethyl-tetrahydrofuran 10h, a deriva-
tive of the natural product magnosalicin,54,55 indicated that the
concept of oxidative ring closure was extendable to the formation
1. Cobalt complex reactivity toward O2
Acceptor substituted [e.g. CF3 or 3,5-(CF3)2C6H3] bis[b-dike-
tonato(−1)]cobalt(II) complexes were able to activate O2 for its
application in diastereoselective synthesis of alkyl- and aryl-
substituted tetrahydrofurylmethanols e.g. 10a–d from bishomoal-
lylic alcohols 9a–d. The fact that O2-pressures above 1 bar
decreased rates of alkenol oxidation pointed to competitive
substrate and oxidant binding to cobalt (see Table 5). Support
for this interpretation originated from an experiment, in which
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3532–3541 | 3537
©

View Article Online
of multiple substituted tetrahydrofurans. Studies associated with
the synthesis of 8-substituted oxabicyclo[4.3.0]nonane 10f revealed
that a reversal of from 2,4-cis to 2,4-trans is feasible, if the former
would interfere with an obviously more dominating 2,5-trans-
stereocontrol.
of diastereoisomers with respect to side chain configuration of a
product obtained from (E)-configured alkenol 9i (see Scheme 4). A
5-exo-mode of hydroxyl O-atom attack onto the oxidized p-bond
in this model constitutes the first of two sequential C,O bond
forming steps associated with the conversion 9→10.
(b) Stereochemical model. A chair like folding of the alkenol
chain in II and III would direct substituents preferentially into
positions that are reminiscent of equatorial sites in cyclohexane.
Intramolecular C,O bond formation III→IV in this model would
afford 2,5-trans-, 2,4-cis-, and 2,3-trans-disubstituted products,
reflecting experimentally observed diastereoselectivities. Steric
encroachment imposed by cobalt coordination (for R1) and vinyl
substitution (for R3), would not be similarly experienced by R2.
3. Mechanistic interpretation
Selectivity in cobalt-catalyzed alkenol oxidation was critically
dependent on the chain length and substitution pattern at
reactive sites. Subtle changes associated with auxiliary, solvent,
temperature, and alkenol constitution were reflected in yields
and/or product profiles. In combination with available data from
the literature dealing with related transformations, a mechanism
was proposed (Schemes 6–8) with the aim to address, the origin
of the preferred 5-exo-ring closure, stereodisintegration associated
with oxygenation of (E)-configured alkenol 9i, stereoselection, and
product selectivity.
(c) Termination of the sequence and cobalt(II) regeneration. In
extension to the well-known driving force of cobalt(II) compounds
for alkyl hydroperoxide reduction,61 intermediate IV is expected
to furnish tetrahydrofur-2-ylmethanol 10 and hydroxycobalt(III)
complex V in the presence of a suitable H-atom donor, e.g. a
solvent molecule.62 Cobalt(II) regeneration is considered to occur
via hydride shift from coordinated 2-propanol. This step would
yield acetone, H2O, and hydridocobalt(III) complex VI (Scheme 7).
This interpretation closely follows the chemistry of hydroxo- and
alkanolato[bis(acetylacetonato)]cobalt(III).63,64
(a) Oxidative 5-exo-ring closure. Oxygen activation is con-
sidered to occur via binding to paramagnetic complex 6.43 Due
to effects exerted by substituents R in terms of reactivity and
selectivity, it is expected that at least one of the auxiliaries
remained attached to cobalt in O2 adduct I (Scheme 6). EPR
data56 and computational investigations of dioxygen adducts of
cobalt(II) complexes with macrocyclic N2O2-donor ligands57 had
been previously interpreted in terms of a doublet ground state.58,59
If a similar structure and bonding applied for O2 adducts of
(b-diketonato)cobalt complexes, a superoxo-mode of binding is
expected to provide a strong Lewis acid and a metal based oxidant.
In extension to the propensity of bis(trifluoroacetato)cobalt(III)
complexes to convert olefins into radical cations,60 an approach
of reacting entities in proposed intermediate II is considered to
furnish radical cation III. A decrease in p-bond order lowers the
barrier to rotation toward an almost unhindered torsional move-
ment of entities that were previously connected via a double bond.
Evidence for this interpretation originated from a ∼ 50 : 50 ratio
Scheme 7 Proposed sequence for cobalt(II) regeneration.
(d) Olefin hydration and reversal of diastereoselectivity. Sec-
ondary and tertiary alcohols were formed in instances where
oxidative cyclization for unknown reasons was slow (see Schemes 2
and 5). The chemistry of olefin hydration using the combination
of O2 and cobalt(II) Schiff base complexes has been extensively
investigated.65 It proceeds via alkylperoxy cobalt(III) complex
formation and decomposition according to an alkylperoxyl radical
pathway (not shown).
Since superoxo cobalt(III) complexes, e.g. I, have a strong affinity
for homolytic H-atom substitution, preferentially from molecules
with weak C,H-bonds as present in 2-propanol, hydroperoxy
cobalt(III) complex VII (Scheme 8) is expected to be formed
in instances of slow oxidative cyclization.17 It is proposed that
this intermediate accounts for the reversal of stereoselectivity
in tetrahydrofuran formation in going from substrate 9a–b to
dimethyl-substituted derivatives 9j–k. Alkenol oxygenation in
this case is considered to follow a mechanism that has been
investigated in tert-butylperoxy vanadium(V) chemistry for alkenol
9j in detail.30 In extension to this sequence, stereoselective epoxy
alcohol formation and isomerization would furnish cis-10j as
major cyclic ether.
Scheme 6 Proposed mechanistic scheme for O2 activation (top) and
dioxygen utilization in aerobic cobalt-catalyzed tetrahydrofuran synthesis
(bottom).
3538 | Org. Biomol. Chem., 2008, 6, 3532–3541
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
C28H38O6F6Co (643.53): C, 52.26; H, 5.95%. Found: C, 52.36; H,
5.86%.
Bis{2,2-dimethyl-1-[(1R,4S)-1,7,7-trimethyl-2-(oxo-jO)bicyclo-
[2.2.1]hept-3-yliden]propanolato-jO}cobalt(II) (5). Compound
5 was prepared from Co(OAc)2·4H2O (47.2 mg, 190 lmol)
and (1R,4R)-3-[(2,2-dimethyl-1-oxo)prop-1-yl]-1,7,7-trimethyl-
bicyclo[2.2.1]heptan-2-one (1) (89.4 mg, 380 lmol) (ESI) in
iPrOH in extension to the procedure outlined for cobalt complex
6. Compound 5 liberated auxiliary 1 upon drying under reduced
pressure and was therefore used as obtained in solution (red-
colored). [a]2D0 +43.5 (c 1.00 in MeOH). kmax (MeOH)/nm 281sh
and 314 (log e 2.82). MALDI-TOF (DHB) calc. for C30H46O4Co:
529.6. Found: 529.6 [M+].
Scheme 8 Proposed mechanistic scheme for stereoselective formation
of 2,5-cis-configured tetrahydrofurans from terminal dimethyl-substituted
pent-4-en-1-ols 9j–k (R1 = Ph or tBu).
Bis{1-phenyl-1-[(1R,4S)-1,7,7-trimethyl-2-(oxo-jO)bicyclo-
[2.2.1]hept-3-yliden]methanolato-jO}cobalt(II) (7). Compound 7
was obtained from Co(OAc)2·4H2O (126 mg, lmol) and (1R,4R)-
3-benzoyl-1,7,7-trimethylbicyclo[2.2.1]heptan-2-one (3) (253 mg,
1.00 mmol) (ESI) as described for deriva◦tive 6. Yield: 175 mg
(60%), red–brown crystalline solid, mp 325 C (EtOH). [a]2D0 +95.5
(c 0.90 in CHCl3). kmax (EtOH)/nm 233 (log e 2.76), 327 (2.89).
mmax (KBr)/cm−1 2957 and 2870 (CH), 1755 and 1609 (CO), 1575,
1506, 1471, 1374, 1179, 1078, 800, 781 and 700. MALDI-TOF
(DHB) calc. for C34H38O4Co: 569.6. Found: 569.3 [M+] and 591.4
[(M + Na)+].
Conclusions
Secondary and primary bishomoallylic alcohols (pent-4-en-1-
ols), e.g. 9a–d, underwent efficient oxidative ring closures if
treated with molecular oxygen and catalytic amounts (10−2 M,
10 mol%) of bis{2,2,2-trifluoromethyl-1-[(1R,4S)-1,7,7-trimethyl-
2-(oxo-jO)bicyclo[2.2.1]hept-3-yliden]ethanolato- jO}cobalt (6)
in 2-propanol at 60 ◦C. The reaction proceeded diastereoselectively
and was applied for sustainable syntheses of 2,3-trans- (96%
de), 2,4-cis- (56% de), and 2,5-trans-di- (>99 de), tri-, and
tetrasubstituted tetrahydrofurans. The procedure therefore has the
potential to open perspectives for natural product synthesis.
Bis{1-[3,5-Bis-(trifluoromethyl)benzoyl]-1-[(1R,4S)-1,7,7-tri-
methyl-2-(oxo-jO)bicyclo[2.2.1]hept-3-yliden]methanolato-jO}-
cobalt(II) (8). Compound 8 was prepared from Co(OAc)2·
4H2O (124 mg, 500 lmol) and (1R,4R)-3-{1-[3,5-bis-(trifluo-
romethyl)phenyl]-1,7,7-trimethylbicyclo[2.2.1]heptan-2-one (4)
(392 mg, 1.00 mmol) (ESI) in extension to a published procedure.67
Yield: 419 mg (quant.), orange crystalline solid, mp 173 ◦C
(EtOH–H2O). [a]D20 +32.4 (c 1.00 in CHCl3). kmax (CH3CN)/nm
228sh, 262sh and 342 (log e 3.14). mmax (KBr)/cm−1 2964 and 2867
(CH), 1634 (CO), 1523, 1387, 1367, 1278, 1172, 1137, 1107, 1078,
1053, 900, 847 and 680. dF (565 MHz; CDCl3) −59.5 and −63.0
(6 : 1). MALDI-TOF (DHB) calc. for C38H34O4F12Co: 841.6.
Found: 841.2 [M+]. Calc. for C38H34O4F12Co (841.60): C, 54.23;
H, 4.07%. Found: C, 54.00; H, 4.10%.
Experimental
1. General remarks
Standard instrumentation and general remarks have been dis-
closed previously (see also ESI). All solvents and reagents were pu-
rified following recommended standard procedures.66 Petroleum
ether refers to the fraction boiling between 30–75 ◦C.
2. Synthesis of bis[3-acyl-1,7,7-trimethylbicyclo[2.2.1]heptan-2-
onato(−1)]cobalt(II) complexes
Bis{2,2,2-Trifluoromethyl-1-[(1R,4S)-1,7,7-trimethyl-2-(oxo-
3. Aerobic cobalt-catalyzed oxidations
jO)bicyclo[2.2.1]hept-3-yliden]ethanolato-jO}cobalt(II) (6).
A
solution of (1R,4R)-3-trifluoroacetyl-1,7,7-trimethylbicyclo-
[2.2.1]heptan-2-one (2) (1.99 g, 8.00 mmol) (ESI) in EtOH (6 mL)
was added in an atmosphere of nitrogen to a suspension of
Co(OAc)2·4H2O (996 mg, 4.00 mmol) in EtOH (12 mL). The
mixture was stirred for 30 min at 20 ◦C. The volatiles were
removed under reduced pressure◦to afford complex 6 (2.50 g,
quant.) as a red solid, mp > 300 C (from EtOH), this material
was applied in aerobic oxidations. [a]2D0 +91.9, (c 0.96 in CHCl3).
mmax (KBr)/cm−1 2965, 1654, 1560, 1389, 1326, 1270, 1227, 1199,
1132 and 806. dF (565 MHz; EtOD) −76.44. MALDI-TOF
(DHB) calc for C24H28O4F6Co: 553.4. Found m/z 553.3 [M+].
Calc for C24H28O4F6Co (553.41): C, 52.09; H, 5.10. Found: C,
50.40; H, 5.35%. An analytically pure sample was obtained by
slowly allowing pentane to diffuse into a solution of 6 in wet
THF to furnish hydrated monotetrahydrofuran adduct of 6 as
red crystalline solid. kmax (iPrOH)/nm 309 (log e 3.25). Calc. for
General procedure. A solution of a substrate (0.40 mmol of
9h; 0.91 mmol of 9i; 0.99–1.02 mmol of 9a–d, 9k, 19; 1.55 mmol
of 9j; 2.03 mmol of 9e and 3.00 mmol of 9f–g) and a cobalt(II)
complex 5–8 (0.05–0.20 equivalents) in a solution of a given solvent
(20 mL mmol−1 for 9h; 8 ml/mmol for of 9a–d, 9i–k, 19 and
6 mL mmol−1 for 9e–g) was stirred at 20–98 ◦C for 2–48◦h in a
stationary oxygen atmosphere. After cooling down to 20 C, the
reaction mixture was diluted with dichloromethane (6 mL mmol−1)
and Et2O (4 mL mmol−1). The solution was washed with satd. aq.
Na2S2O3 soln. (3 × 5 mL mmol−1). Combined Na2S2O3 washings
were extracted with Et2O (2 × 4 mL mmol−1). (i) GC Analysis: n-
decane was added as internal standard to combined organic layers
to afford a solution which was quantitatively analyzed by GC
via independently measured responsivity factors. (ii) Preparative
scale experiments: combined organic layers were dried (MgSO4)
to afford a solution that was concentrated under reduced pressure.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3532–3541 | 3539
©

View Article Online
The remaining residue was purified by column chromatography
(SiO2).
4.06–4.10 (1 H, m, 1ꢀ-H), 4.14 (1 H, mc, 2-H), 5.02 (1 H, dd, J 8.7,
6.1, 5-H), 7.27–7.30 (1 H, m, Ph-H) and 7.33–7.34 (4 H, m, Ph-H);
dC (150 MHz; CDCl3) 17.9 (C2ꢀ), 25.5 (C3), 35.6 (C4), 68.1 (C1ꢀ),
81.6 (C5), 83.7 (C2), 125.5, 127.3, 128.4 and 143.3 (Ph); NOESY
(cross peaks) 2-H ↔ Ph. m/z (EI, 70 eV) 192 (M+, 10%), 174 (10),
147 (91), 129 (45), 117 (22), 104 (100), 91 (99), 77 (27) and 51 (15).
Tetrahydro-5-methyl-5-phenylfur-2-ylmethanol (10e). Yield:
60.0 mg (15%, cis : trans = 20 : 80) from 357 mg (2.03 mmol)
of 9e. Rf 0.35 [SiO2, Et2O–pentane = 2 : 1 (v/v)], colorless oil.
C12H16O2 (192.25). trans-10e: dH (600 MHz; CDCl3) 1.55 (3 H, s,
Me), 1.80–1.86 (2 H, m, 3-H, 4-H), 2.02–2.09 (1 H, m, 3-H),
2.24–2.30 (1 H, m, 4-H), 3.58 (1 H, dd, J 11.4, 5.5, CH2OH), 3.77
(1 H, dd, J 11.4, 3.3, CH2OH), 4.17 (1 H, ddd, J 12.5, 6.9, 3.3,
2-H), 7.20–7.25 (1 H, m, Ph), 7.31–7.35 (2 H, m, Ph), 7.38–7.41 (2
H, m, Ph). dC (150 MHz; CDCl3) 27.3 (C3), 30.4 (Me), 39.4 (C4),
65.4 (CH2OH), 79.0 (C2), 85.2 (C5), 124.5, 126.4, 128.2, 147.7
(Ph). NOESY (cross peaks) CH2OH ↔ Me, 2-H ↔ Ph. m/z (EI,
70 eV) 192 (M+, 1), 177 (100), 161 (78), 143 (87), 128 (38), 117
(52), 105 (71), 91 (85), 77 (48). cis-10e: dH (600 MHz; CDCl3) 1.52
(3 H, s, Me), 1.80–2.15 (3 H, m, 3-H, 4-H), 2.21–2.31 (1 H, m,
4-H), 3.54 (1 H, dd, J 11.6, 6.4, CH2OH), 3.71 (1 H, dd, J 11.5,
3.4, CH2OH), 4.32 (1 H, ddd, J 13.5, 6.9, 3.3, 2-H), 7.20–7.44 (5
H, m, Ph). dC (150 MHz; CDCl3) 27.5 (C3), 29.6 (Me), 39.2 (C4),
65.6 (CH2OH), 79.4 (C5), 85.0 (C2), 124.5, 126.5, 128.2, 148.2
(Ph). m/z (EI, 70 eV) 177 (M+ − CH3, 95%), 161 (93), 143 (97),
128 (42), 117 (54), 105 (85), 91 (100), 77 (60).
7,7-Dimethyloctane-2,6-diol (20). Yield: 55.6 mg (31%, dr =
50 : 50) from 160 mg (1.02 mmol) of 19, Rf 0.28 [SiO2, acetone–
petroleum ether = 2 : 5 (v/v)], colorless oil. dH (400 MHz; CDCl3)
0.89 (2 × 9 H, s, 7-Me), 1.19 (2 × 3 H, d, J 6.2, 1-H), 1.23–1.68
(2 × 6 H, m, 3-H, 4-H, 5-H), 1.72 (2 × 2 H, s, OH), 3.20 (2 × 1
H, d, J 10.4, 6-H), 3.83 (2 × 1 H, mc, 2-H). dC (100 MHz; CDCl3)
23.1 and 23.2 (C4), 23.5 and 23.6 (C1), 25.7 (C8), 31.2 and 31.4
(C5), 34.9 (C7), 39.2 and 39.3 (C3), 68.0 and 68.2 (C2), 79.9 (C6).
m/z (EI, 70 eV) 141 (M+ − CHO, 1%), 117 (8), 99 (63), 87 (16), 81
(100), 71 (18), 57 (58), 43 (36), 29 (9). Calc. for C10H22O2 (174.28):
C, 68.92; H, 12.72. Found: C, 68.60; H, 12.63%.
5-Hydroxy-5-methyl-1-phenylhexan-1-one (22). Yield: 86.8 mg
(27%) from 295 mg (1.55 mmol) of 9j, Rf 0.43 [SiO2, acetone–
pentane = 1 : 3 (v/v)], colorless oil. C13H18O2 (206.28). dH
(400 MHz; DMSO-d6) 1.07 (2 × 3 H, s, Me), 1.35–1.42 (2 H,
m, 3-H), 1.64 (2 H, mc, 4-H), 2.98 (2 H, t, J 7.3, 2-H), 4.13 (1 H, s,
OH), 7.47–7.54 (2 H, m, Ph), 7.58–7.64 (1 H, m, Ph), 7.92–7.97 (2
H, m, Ph). dC (150 MHz; DMSO-d6) 19.0 (C3), 29.3 (Me), 38.5,
43.0 (C2 and C4), 68.7 (C5), 127.8, 128.7, 133.0, 136.7 (Ph), 200.2
(C1). m/z (EI, 70 eV) 192 (M+ − CH2, 2%), 188 (14), 170 (5), 147
(4), 133 (27), 120 (55), 105 (100), 77 (56), 59 (29), 55 (21).
3-trans-4-cis-5-trans-Tetrahydro-3,5-bis(2,4,5-trimethoxyphe-
nyl)-4-methylfur-2-ylmethanol (10h). Yield: 130 mg (72%, cis :
trans < 1 : 99) from 173 mg (0.40 mmol) of 9h. Rf 0.42 [SiO2,
◦
acetone–Et2O = 1 : 3 (v/v)], colorless solid, mp 46 C [acetone–
Et2O = 1 : 3 (v/v)]. C24H32O8 (448.51). dH (400 MHz; DMSO-d6)
0.72 (3 H, d, J 6.5, 4-Me), 2.28 (1 H, mc, 4-H), 3.10 (1 H, dd, J 10.6,
9.5, 3-H), 3.27–3.33 (1 H, m, CH2OH), 3.39 (1 H, ddd, J 11.7, 5.6,
2.4, CH2OH), 3.69 (3 H, s, OMe), 3.73 (3 H, s, OMe), 3.75 (2 ×
3 H, s, OMe), 3.76 (3 H, s, OMe), 3.78 (3 H, s, OMe), 4.19 (1 H,
ddd, J 8.6, 5.9, 2.4, 2-H), 4.62 (1 H, t, J 5.7, OH), 4.84 (1 H, d, J
9.5, 5-H), 6.66 (1 H, s, Ph), 6.67 (1 H, s, Ph), 6.84 (1 H, s, Ph), 7.06
(1 H, s, Ph). dC (100 MHz; CDCl3) 14.1 (4-Me), 48.4, 48.9 (C3 and
C4), 56.2, 56.3, 56.6, 56.7, 56.9, 57.0 (6 × OMe), 63.8 (CH2OH),
81.9 (C5), 84.7 (C2), 98.2, 98.6, 112.0, 113.0, 119.0, 121.5, 143.6,
143.7, 148.7, 149.4, 151.9, 152.6 (Ph). NOESY (cross peaks) 2-H
↔ Ph, 2-H ↔ 4-H, 3-H ↔ 5-H, 3-H ↔ 4-Me, 4-H ↔ Ph, 5-H ↔
4-Me; m/z (EI, 70 eV) 448 (M+, 45%), 388 (15), 357 (6), 221 (16),
209 (100), 181 (64), 151 (19).
Acknowledgements
This work was supported by the Deutsche Forschungsgemein-
schaft (Schwerpunktprogramm 1118-Sekunda¨re Wechselwirkun-
gen, grant Ha1705/9-1) and the Fonds der Chemischen Industrie
(scholarship for D.S.). We also express our gratitude to Dipl.-
Chem. Georg Stapf for helpful advice on determining enan-
tiomeric purity of oxidation products via 31P NMR, Nathalie
Laponche and Markus Weyland for technical assistance asso-
ciated with quantitative water and acetone analysis, and Dr
Simone Drees for performing preliminary experiments22 on cobalt-
catalyzed oxidations.
trans-1-(5-Phenyltetrahydrofur-2-yl)ethanol trans-(10i).
Notes and References
Isomer 1. Yield: 33.0 mg (19%) from 160 mg (0.91 mmol) of
9i, Rf 0.38 [SiO2, acetone–pentane = 1 : 5 (v/v)], colourless oil.
dH(400 MHz; CDCl3) 1.18 (3 H, d, J 6.4, 2ꢀ-H), 1.68–1.77 (1 H, m,
3-H), 1.89 (1 H, mc, 4-H), 2.11 (1 H, mc, 3-H), 2.37 (1 H, mc, 4-H),
2.61 (1 H, s, OH), 3.68 (1 H, dq, Jd 6.7, Jq 6.4, 1ꢀ-H), 4.00 (1 H,
ddd, J 7.3, 2-H), 4.96 (1 H, dd, J 8.8, 5.9, 5-H), 7.26–7.29 (1 H,
m, Ph-H) and 7.34–7.35 (4 H, m, Ph-H); dC (150 MHz; CDCl3)
18.5 (C2ꢀ), 28.7 (C3), 35.8 (C4), 70.7 (C1ꢀ), 80.7 (C5), 84.5 (C2),
125.6, 127.4, 128.3 and 142.7 (Ph); NOESY (cross peaks) 2-H ↔
Ph, 1-Hꢀ ↔ 5-H, 2-H ↔ OH. m/z (EI, 70 eV) 192 (M+, 8%), 174
(12), 147 (85), 129 (51), 117 (24), 104 (99), 91 (100), 77 (31) and 51
(16).
1 S. Inoki and T. Mukaiyama, Chem. Lett., 1990, 67–70.
2 For a review see: T. Mukaiyama and T. Yamada, Bull. Chem. Soc. Jpn.,
1995, 68, 17–35.
3 Z.-M. Wang, S.-K. Tian and M. Shi, Tetrahedron Lett., 1999, 40, 977–
980.
4 Z.-M. Wang, S.-K. Tian and M. Shi, Tetrahedron: Asymmetry, 1999,
10, 667–670.
5 Z.-M. Wang, S.-K. Tian and M. Shi, Eur. J. Org. Chem., 2000, 349–356.
6 For a recent review on stereoselective tetrahydrofuran synthesis: J. P.
Wolfe and M. B. Hay, Tetrahedron, 2007, 63, 261–290.
7 A. Werner and A. Z. Z. Mylius, Z. Anorg. Allg. Chem., 1898, 16, 245–
267.
8 T. Tsumaki, Bull. Chem. Soc. Jpn., 1938, 13, 252–260.
9 R. H. Bailes and M. Calvin, J. Am. Chem. Soc., 1947, 69, 1886–1893.
10 G. Henrici-Olive´ and S. Olive´, Angew. Chem., 1974, 86, 1–12; G.
Henrici-Olive´ and S. Olive´, Angew. Chem., Int. Ed. Engl., 1974, 13,
29–38.
Isomer 2. Yield: 25.9 mg (15%) from 160 mg of 9i, Rf 0.29
[SiO2, acetone–pentane = 1 : 5 (v/v)], colourless oil. dH (600 MHz;
CDCl3) 1.18 (3 H, d, J 6.4, 2ꢀ-H), 1.86–1.94 (1 H, m, 3-H), 1.94–
1.99 (1 H, m, 4-H), 2.01–2.07 (1 H, m, 3-H), 2.39 (1 H, mc, 4-H),
11 F. Basolo, B. M. Hoffman and J. A. Ibers, Acc. Chem. Res., 1975, 8,
384–392.
3540 | Org. Biomol. Chem., 2008, 6, 3532–3541
This journal is
The Royal Society of Chemistry 2008
©

View Article Online
12 R. S. Drago and B. B. Corden, Acc. Chem. Res., 1980, 13, 353–360.
13 J. W. Buchler, Angew. Chem., 1978, 90, 425–441; J. W. Buchler, Angew.
Chem., Int. Ed. Engl., 1978, 17, 407–423.
14 J. Hartung and M. Greb, J. Organomet. Chem., 2002, 661, 67–84.
15 W. H. Richardson, J. Am. Chem. Soc., 1965, 87, 247–253.
16 L. I. Sama´ndi, in Advances in Catalytic Activation of Dioxygen by Metal
Complexes, ed. L. I. Sima´ndi, Kluwer Academic Publishers, Dordrecht,
2003, pp. 265–328.
41 F. Lipari and S. J. Swarin, J. Chromatogr., 1982, 247, 297–306.
42 G. Wieland, Wasserbestimmung durch Karl-Fischer-Titration, GIT-
Verlag, Darmstadt, 1985.
43 B. Mene´ndez Pe´rez, D. Schuch and J. Hartung, in Activating Unreactive
Substrates, ed. C. Bolm and E. Hahn, Wiley-VCH, Weinheim, accepted
for publication.
44 A. C. Dupont, V. H. Audia, P. P. Waid and J. P. Carter, Synth. Commun.,
1990, 20, 1011–1021.
17 R. A. Sheldon and J. K. Kochi, Metal-Catalyzed Oxidations of Organic
45 J. L. Princival, S. M. G. de Barros, J. V. Comasseto and A. A. Dos
Santos, Tetrahedron Lett., 2005, 46, 4423–4425.
Compounds, Academic Press, New York, 1981.
18 M. M. Faul and B. E. Huff, Chem. Rev., 2000, 100, 2407–2473.
19 For recent accounts on aerobic oxidations see: (a) T. Punniyamurthy,
S. Velusamy and J. Iqbal, Chem. Rev., 2005, 105, 2329–2363; (b) F.
Recupero and C. Punta, Chem. Rev., 2007, 107, 3800–3842.
20 For a comprehensive review on dioxygen activation: D. Schro¨der and
H. Schwarz, in Essays in Contemporary Chemistry, ed. G. Quinkert and
M. V. Kisaku¨rek, Wiley-VCH, Weinheim, 2001, pp. 131–156.
21 K. R. Kopecky, D. Nonhebel, G. Morris and G. S. Hammond, J. Org.
Chem., 1962, 27, 1036–1037.
22 S. Drees, PhD Thesis, Universtita¨t Wu¨rzburg, 2002.
23 M. D. McCreary, D. W. Lewis, D. L. Wernick and G. M. Whitesides,
J. Am. Chem. Soc., 1974, 96, 1038–1054.
24 H. L. Goering, J. N. Eikenberry, G. S. Koermer and C. J. Lattimer,
J. Am. Chem. Soc., 1974, 96, 1493–1501.
25 A. Togni, Organometallics, 1990, 9, 3106–3113.
46 (a) C. Meyer, I. Marek, G. Courtemanche and J.-F. Normant, Tetra-
hedron, 1994, 50, 11665–11692; (b) A. L. J. Beckwith, M. D. Cliff and
C. H. Schiesser, Tetrahedron, 1992, 48, 4641–4648.
47 (a) V. Spe´ziale, M. M. Amat and A. Lattes, J. Heterocycl. Chem., 1976,
13, 349–355; (b) M. B. Hay, A. R. Hardin and J. P. Wolfe, J. Org.
Chem., 2005, 70, 3099–3107; (c) Y.-S. Hon, Y.-W. Liu and C.-H. Hsieh,
Tetrahedron, 2004, 60, 4837–4860.
48 S. Tang and R. M. Kennedy, Tetrahedron Lett., 1992, 33, 5299–
5302.
49 M.-C. Benhamou, G. Etemad-Moghadam, V. Spe´ziale and A. Lattes,
J. Heterocycl. Chem., 1978, 15, 1313–1318.
50 O. Muraoka, T. Sawada, E. Morimoto and G. Tanabe, Chem. Pharm.
Bull., 1993, 41, 772–774.
51 M. Kimura, A Ezoe, M. Mori, K. Iwata and Y. Tamaru, J. Am. Chem.
Soc., 2006, 128, 8559–8568.
26 P. M. O’Neill, S. Hindley, M. D. Pugh, J. Davies, P. G. Bray, B. K. Park,
D. S. Kapu, S. A. Ward and P. A. Stocks, Tetrahedron Lett., 2003, 44,
8135–8138.
27 E. C. Hsu and G. Holzwarth, J. Am. Chem. Soc., 1973, 95, 6902–6906.
28 M. G. Moloney, D. R. Paul, R. M. Thompsom and E. Wright,
Tetrahedron: Asymmetry, 1996, 7, 2551–2562.
29 J. Hartung and F. Gallou, J. Org. Chem., 1995, 60, 6706–6716.
30 J. Hartung, S. Drees, M. Greb, P. Schmidt, A. Murso, D. Stalke, I.
Svoboda and H. Fuess, Eur. J. Org. Chem., 2003, 2388–2408.
31 J.-M. Brunel, O. Pardigon, M. Maffei and G. Buono, Tetrahedron:
Asymmetry, 1992, 3, 1243–1246.
32 For related experiments see: J. Masłowska and J. B. Baranowski, J. Mol.
Struct., 1978, 47, 405–416.
33 A. F. Holleman and E. Wiberg, Lehrbuch der Anorganischen Chemie,
ed. W. de Gruyter, 101. Auflage, Berlin, 1995, pp. 1548–1574.
34 J. Hartung, M. Hiller and P. Schmidt, Chem.–Eur. J., 1996, 2, 1014–
1023.
35 J. Iqbal and R. R. Srivastava, Tetrahedron, 1991, 47, 3155–3170.
36 A. Schmitt and H.-U. Reißig, Chem. Ber., 1995, 128, 871–876.
37 S. Hu¨nig and M. Scha¨fer, Chem. Ber., 1993, 126, 177–189.
38 G. A. Molander and K. O. Cameron, J. Am. Chem. Soc., 1993, 115,
830–846.
52 F. Foubelo, S. A. Saleh and M Yus, J. Org. Chem., 2000, 65, 3478–
3483.
53 Y.-R. Luo, Handbook of Bond Dissociation Energies in Organic Com-
pounds, CRC Press, Boca Raton, 2003.
54 K. Mori, M. Komatsu, M. Kido and K. Nakagawa, Tetrahedron, 1986,
42, 523–528.
ˇ
55 M. Greb, J. Hartung, F. Ko¨hler, K. Spehar, R. Kluge and R. Csuk,
Eur. J. Org. Chem., 2004, 3799–3812.
56 M. Valko, R. Klement, P. Pelika´n, R. Bocˇa, L. Dlha´n, A. Bo¨ttcher, H.
Elias and L. Mu¨ller, J. Phys. Chem., 1995, 99, 137–143.
57 N. J. Henson, P. J. Hay and A. Redondo, Inorg. Chem., 1999, 38, 1618–
1626.
58 B. S. Tovrog and R. S. Drago, J. Am. Chem. Soc., 1974, 96, 6765–
6766.
59 D. Getz, E. Melamud, B. L. Silver and Z. Dori, J. Am. Chem. Soc.,
1975, 97, 3846–3847.
60 R. T. Tang and J. K. Kochi, J. Inorg. Nucl. Chem., 1973, 35, 3845–3856.
61 R. Hiatt, K. C. Irwin and C. W. Gould, J. Org. Chem., 1968, 33, 1430–
1435.
62 J. H. Tocher and J. P. Fackler, Jr., Inorg. Chim. Acta, 1985, 102, 211–215.
63 J. A. Bertrand and D. Caine, J. Am. Chem. Soc., 1964, 86, 2298–2299.
64 B. P. Cotsoradis and R. D. Archer, Inorg. Chem., 1967, 6, 800–805.
65 D. E. Hamilton, R. S. Drago and A. Zombeck, J. Am. Chem. Soc.,
1987, 109, 374–379.
39 C. E. Elliott, D. O. Miller and D. J. Burnell, J. Chem. Soc., Perkin Trans.
1, 2002, 217–226.
40 (a) J. C. Anderson and R. H. Munday, J. Org. Chem., 2004, 69, 8971–
8974; (b) G. C. R. Ellis-Davies, A. Gilbert, P. Heath, J. C. Lane, J. V.
Warrington and D. L. Westover, J. Chem. Soc., Perkin Trans. 2, 1984,
1833–1841.
66 D. D. Perrin, W. L. F. Armatego and D. R. Perrin, Purification of
Laboratory Chemicals, Pergamon Press, Oxford, 2nd. edn, 1980.
67 K. J. Eisentraut and R. E. Sievers, J. Am. Chem. Soc., 1965, 87, 5254–
5256.
This journal is
The Royal Society of Chemistry 2008
Org. Biomol. Chem., 2008, 6, 3532–3541 | 3541
©