DoE (Design of Experiments) assisted allylic hydroxylation of enones catalysed by a copper-aluminium mixed oxide

Article abstract of DOI:10.1002/ejoc.201301145

The allylic hydroxylation of enones using dioxygen as the oxidant has been studied. The reaction was first examined in the absence of any catalyst, using β-ionone as a model substrate. Then a new copper-aluminium mixed oxide, Cu-Al Ox, was prepared and ch

Full text of DOI:10.1002/ejoc.201301145

FULL PAPER
DOI: 10.1002/ejoc.201301145
DoE (Design of Experiments) Assisted Allylic Hydroxylation of Enones
Catalysed by a Copper–Aluminium Mixed Oxide
Ana Leticia García-Cabeza,[a] Rubén Marín-Barrios,[a] Redouan Azarken,[a]
F. Javier Moreno-Dorado,[a] María J. Ortega,[a] Hilario Vidal,[b] José M. Gatica,[b]
Guillermo M. Massanet,[a] and Francisco M. Guerra*[a]
Keywords: Oxidation / C–H activation / Heterogeneous catalysis / Enones
The allylic hydroxylation of enones using dioxygen as the
oxidant has been studied. The reaction was first examined in
the absence of any catalyst, using β-ionone as a model sub-
strate. Then a new copper–aluminium mixed oxide, Cu–Al
Ox, was prepared and characterized in order to be used as a
catalyst. This oxide showed good activity, and provided the
corresponding γ- or ε-hydroxylated enones, starting from dif-
ferent α,β- or α,β,γ,δ-unsaturated ketones. In all cases, the
yields were significantly improved compared to experiments
run in the absence of the catalyst. The reaction was selective,
and the formation of epoxides or other overoxidation prod-
ucts was detected only to a minor extent. The described pro-
cedure is a technically straightforward synthetic alternative
to those methods described to date involving many reaction
steps or toxic reagents. The reactions were optimized using
design of experiments techniques (DoE).
Bearing this in mind, it is not surprising that this reaction
has not found extensive use in organic synthesis. Usually,
the desired hydroxy enone is accompanied by a plethora of
other oxidation products, mainly epoxides or overoxidation
products, resulting in very low yields.[7]
Evidence for the difficulty of this type of transformation
can be seen in the preparation of (+)-5β-hydroxycarvone (2)
by Yoshikoshi et al.[8] Instead of a direct oxidation of car-
vone (1), it required a nine-step synthetic sequence to pro-
duce the desired hydroxy enone (i.e., 2; Scheme 2a). More
recently, Yang et al. have reported the preparation of this
compound in two steps, using nitrosobenzene to oxidize the
corresponding silyl enol ether of the enone.[9] The toxicity
of nitrosobenzene makes it desirable to find new alternative
ways to perform this transformation (Scheme 2a). In ad-
dition, hydroxy enone 2 is very prone to undergo aromatiza-
tion to produce phenol 3 (Scheme 2b).
Introduction
The γ-hydroxylation of α,β-unsaturated ketones using
oxygen as the oxidant in the presence of a base is a reaction
that has been known since the beginning of the last century
(Scheme 1).[1–4] In the 1950s, Hawkins and the McQuillin
group both published the results of their pioneering studies
into the mechanism of this reaction, using substrates such
as cyperone, mesityl oxide, and isophorone, among
others.[5,6]
Scheme 1. γ-Hydroxylation of α,β-unsaturated ketones.
The reaction of ketones with oxygen is usually sluggish,
and frequently it results in the cleavage of the molecule. It
typically involves species containing O–O bonds, and the
weakness of these bonds results in the characteristic insta-
bility of the molecules. Such species are prone to undergo
reactions such as Hock cleavage, Kornblum–DeLaMare re-
actions, Hock dehydration, rearrangements into dioxetane
rings, etc.[7]
[a] Departamento de Química Orgánica, Universidad de Cádiz,
Puerto Real 11510, Spain
E-mail: francisco.guerra@uca.es
http://www2.uca.es/dept/quimica_organica
[b] Departamento C.M., I.M. y Química Inorgánica,
Puerto Real 11510, Spain
Scheme 2. (a) Synthesis of 5-β-hydroxycarvone (2); (b) oxidative
aromatization of carvone (1). TMS = trimethylsilyl.
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201301145.
Eur. J. Org. Chem. 2013, 8307–8314
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
8307

F. M. Guerra et al.
FULL PAPER
In this paper, we present the results of our efforts to Table 2. Influence of the oxidant on the ε-hydroxylation of β-
ionone (4).[a]
achieve the oxidation of α,β- and α,β,γ,δ-enones, using di-
oxygen (from the air) as the oxidant in the presence of a
copper–aluminium mixed oxide, to give the corresponding
γ-hydroxylated compounds. The products obtained may be
used as building blocks in the synthesis of more complex
molecules such as terpenes, and this represents the final
goal of our investigation.
Run
Oxidant
Yield [%][b]
5 h
2 h
20 h
[c]
1
2
3
4
5
6
7
8
9
KMnO4
KO2
0
0
0
0
5
2
0
9
25
0
5
0
0
6
5
0
16
36
0
14
0
[d]
SeO2
Results and Discussion
DIB[d]
0
tBuOOH
11
15
2
36
45
[e]
NaI/H2O2
NaOCl/H2O2
MnO2
Study of the Influence of Bases, Solvents, and Oxidants in
the Reaction
[d,f]
[g]
O2
Our studies began using β-ionone (4) as a model sub-
strate. Although this compound is an α,β,γ,δ-unsaturated
ketone, it was chosen as a model compound since it is an
affordable commercially available material that has pre-
viously been used in studies of the autoxidation of en-
ones.[10] With this substrate, we carried out a preliminary
screening of solvents and bases. After some experimenta-
tion, it was found that a protic solvent and a strong base
were necessary. The best result was obtained with tBuOK
as the base and EtOH as the solvent (Table 1, Entry 9).
[a] Reaction conditions: 4 (1.18 mmol), base (1.18 mmol), oxidant
(1.18 mmol), solvent (10 mL), room temp. [b] Determined by GC.
[c] Complex mixture. [d] Starting material recovered. [e] NaI
(1.18 mmol)/H2O2 (2.36 mmol). [f] NaOCl (1.18 mmol)/H2O2
(2.36 mmol). [g] Oxygen from air (1 atm).
DoE Optimization of the β-Ionone ε-Hydroxylation
Reaction
To determine the importance of the different factors in-
volved in the reaction, we decided to use a DoE (design of
experiments) approach. This statistic tool is commonly used
in analytical chemistry or chemical engineering, but it is
rarely used in synthetic organic chemistry.[11–15] The design
of experiments approach, in contrast to the classical OVAT
method (changing one variable at a time), allows us to study
a system by changing many variables at the same time. A
matrix delimited by low and high levels of each variable
involved in the reaction is defined. The yield is considered
to be the response of the system to the changes of each
variable, and the final objective is to obtain a model that
accounts for the yield as a function of all of the variables.
Thus, one can take into account interactions between the
variables, and a wide knowledge of the system can be ob-
tained from a reduced number of experiments.[16]
Based on our preliminary screening, we decided to use
EtOH as the solvent and tBuOK as the base. The time was
set to 24 h, and all the runs were carried out at room tem-
perature.
The roles of the substrate concentration and the number
of equivalents of base used were the variables to be opti-
mized by DoE. Since there were only two variables, we de-
Table 1. Influence of the solvent and base on the ε-hydroxylation
of β-ionone.[a]
Entry
Base
Time [h]
Yield [%][b]
EtOH MeOH 2-Propanol
1
2
3
4
5
6
7
8
9
MeOK
MeOK
MeOK
KOH
KOH
KOH
tBuOK
tBuOK
tBuOK
2
5
24
2
5
24
2
19
30
42
18
29
36
25
36
45
5
16
31
37
11
18
21
10
14
17
10
33
5
10
30
6
5
24
13
38
[a] Reaction conditions: 4 (2.36 mmol), base (2.36 mmol), solvent
(10 mL), room temp. [b] Determined by GC. Only alcohol 5 and
starting material 4 were detected.
The reaction time was also studied, and we found that cided to use a three-level full-factorial design. A full-facto-
24 h was an optimal time. Longer reaction times resulted in rial designed experiment includes all possible combinations
lower yields, probably due to overoxidation of the products. of levels for all of the variables.
Although the autoxidation of enones proceeds with O2 as
The range of the substrate concentration spanned from
the oxidant, a series of experiments was performed with 0.10 m (low level) to 1.20 m (high level), with 0.65 m as the
different oxidizing reagents (KMnO4, SeO2, DIB [(di- central point. The values chosen for the amounts of tBuOK
acetoxyiodo)benzene], etc.), but, as expected, the best re- were 1 (low level), 3 (central point), and 5 equiv. (high
sults were obtained with atmospheric O2 (Table 2).
level).
8308
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 8307–8314

DoE-Assisted Allylic Hydroxylation of Enones
The results of the experiments are shown in Table 3. Fig- the number of equivalents of tBuOK, we performed some
ures 1 and 2 show the Pareto plot and the response-surface additional runs outside the established experimental do-
graph, respectively. The vertical line in the Pareto plot (Fig- main. But these experiments using less than 1.0 equiv. of
ure 1) indicates a level of 95% confidence. Horizontal bars tBuOK did not lead to further improvements in the yield.
that cross this line represent variables with greater influence Eventually, we found that when 1.2 equiv. of tBuOK was
on the yield of the reaction for that level of confidence.
used, the yield increased to 55%.
Table 3. Three-level full-factorial design for the oxidation of β-
ionone (4).
Catalytic Activity of Cu–Al Ox, a Copper–Aluminium
Mixed Oxide
Entry
Concentration of β-ionone (4)
[m]
tBuOK
[equiv.]
Yield of 5
[%]
Despite the improvement in the yield resulting from the
statistical study of the reaction, it was clear that the condi-
tions had the potential to be improved further. So we fo-
cussed our attention on hetereogeneous catalysis. A survey
of the literature showed that clays and hydrotalcites have
often been described as catalysts for different organic reac-
tions.[17–22] Commercially sourced hydrotalcite did not af-
fect the outcome of our reaction with any substrate. It has
also been published that the use of some copper sources
may help to decrease the formation of epoxides and other
by-products in the autoxidation of similar enones.[23] After
some experimentation, a copper–aluminium mixed oxide,
(hereafter called Cu–Al Ox), was synthesized. This new
mixed oxide showed remarkable catalytic behaviour in the
hydroxylation of enones. This material was prepared by a
modification of the procedure developed by Guida et al. for
the synthesis of hydrotalcites (see Experimental Section).[24]
To ensure the reproducibility of the Cu–Al Ox prepara-
tion, it was submitted to a characterization process. Fig-
ure 3 shows a typical SEM (scanning electron microscopy)
image of the Cu–Al Ox catalyst prepared in this work. It
has the form of a fine powder, which, although showing
some irregularities in morphology and size, in general con-
sists of micron-sized well-rounded grains. This study was
complemented by granulometric measurements, which indi-
1
2
3
4
5
6
7
8
9
0.10
0.65
1.20
0.10
0.65
1.20
0.10
0.65
1.20
0.65
0.65
0.65
1
1
1
3
3
3
5
5
5
3
3
3
50
48
36
46
4
0
39
0
0
3
4
16
10
11
12
Figure 1. Pareto plot for the oxidation of β-ionone (4).
Figure 2. Calculated response surface for the oxidation of β-ionone.
In this case, the graph shows significant negative effects
for the concentration and the number of equivalents of
tBuOK used. No interaction was observed between these
two variables.
Analysis of the response-surface graph reveals that the
higher the concentration of β-ionone (4), the lower the yield
is. The maximum yield was found at a β-ionone concentra-
tion of 0.1 m, using 1.0 equiv. of tBuOK. Given that the
maximum yield was obtained at the lower limiting level of Figure 3. Scanning electron micrograph of the catalyst Cu–Al Ox.
Eur. J. Org. Chem. 2013, 8307–8314
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
8309

F. M. Guerra et al.
FULL PAPER
cated a mean particle size of 1 μm. In addition, a textural the catalytic activity of Cu–Al Ox, several runs using the
study was performed using N2 physisorption (see Support- Cu–Al Ox precursors under different conditions and with
ing Information).
different substrates were also carried out, with negative re-
The results of compositional analysis performed by ICP sults. Furthermore, blank experiments in the absence of
(inductively coupled plasma mass spectrometry) and XRF Cu–Al Ox led to poorer yields.
(X-ray fluorescence) are summarized in Table 4. There is
The scope of the reaction was investigated with different
good agreement between data obtained by the two tech- α,β-unsaturated ketones. The initial conditions used were
niques. Moreover, they are relatively consistent with the Cu/ those optimized for β-ionone, but we soon discovered that
Al atomic ratio selected for catalyst preparation.
every substrate would have to be studied separately. We also
observed that the yield improved when the Cu–Al Ox was
stirred with ethanol for 10 min prior to the addition of the
substrate and the base. This fact seems to suggest that the
Cu–Al Ox requires preactivation for catalysis to take place.
We used either three-level full-factorial design or Box–
Behnken experiments, now including the amount of the
Cu–Al Ox as a new variable. In all cases except for testos-
terone, the reaction time was limited to 24 h.
The case of carvone (1) was especially interesting. As
mentioned above, this substrate is very reluctant to provide
the corresponding γ-hydroxy derivative 2, in part due to its
tendency to aromatize by elimination of the hydroxy group
(Scheme 2b). To determine the best conditions for its oxi-
dation, a three-level factorial design was performed with
the following range of variables: (a) carvone concentration:
0.10, 0.38, and 0.65 m; (b) amount of Cu–Al Ox: 5.0, 52.5,
and 100.0 mg; (c) amount of tBuOK: 0.50, 1.75, and
3.00 equiv. The resulting yields are shown in Table 5. Under
Table 4. Compositional analysis [wt.-%] of the catalyst studied
(n.a.: not available by means of this technique; n.d.: not detected).
Element
ICP
XRF
EDS[a]
Cu
Al
O
C
Cl
61.2
7.9
n.a.
n.a.
n.a.
64.9
9.5
25.2
n.a.
0.03
63.3Ϯ8.5
9.1Ϯ1.6
22.0Ϯ5.4
5.4Ϯ1.6
n.d.
[a] Data correspond to the average of values obtained for different
analysed areas.
Table 4 also includes the elemental composition accord-
ing to EDS (energy-dispersive X-ray spectroscopy) analysis
performed over different areas of the surface of the catalyst
sample. These data deserve several comments. First of all,
it should be noted that in general, they also agree with the
data obtained by ICP and XRF. Second, although some
copper-rich (77.8 wt.-%) and aluminium-rich (15.4 wt.-%)
areas were found, both elements were detected along with
oxygen in all zones measured. This relatively homogeneous
composition fits well with the formation of a mixed phase.
The presence of a minor amount of carbon is also notable,
and this could reasonably be derived from residual carbon-
ates that might in principle come either from the sodium
carbonate used for the catalyst preparation or from further
reaction of the precipitate with CO2 during its exposure to
air. This finding contrasts with the absence of chlorine,
which confirms that the chlorides originally present in the
starting materials for the catalyst synthesis were completely
removed in the course of the preparation of the material.
A structural study of the catalyst by X-ray diffraction did
not reveal the presence of a hydrotalcite phase. Some small
and not well-defined peaks that could be due to CuO, teno-
rite phase, were detected.[25] These results suggest that an
amorphous copper–aluminium mixed oxide must mainly
have been formed. This fact is in good agreement with
many references in the literature that point out the difficulty
in obtaining crystalline hydrotalcite-like compounds with
Cu2+ compared to with other divalent metals from Mg2+ to
Mn2+, due to the well known Jahn–Teller effect.[26–29]
Table 5. Three-level full-factorial design experiment for the γ-hy-
droxylation of carvone (1).
Entry Concentration of 1 Cu–Al Ox tBuOK Yield of 2
[m]
[mg]
[equiv.]
[%]
1
2
3
4
5
6
7
8
0.65
0.38
0.10
0.38
0.65
0.65
0.38
0.38
0.38
0.65
0.10
0.65
0.38
0.10
0.65
0.65
0.65
0.10
0.38
0.65
0.10
0.38
0.10
0.38
0.10
0.10
0.38
0.10
0.38
5.0
5.0
52.5
100
5.0
52.5
52.5
52.5
5.0
100
52.5
52.5
100
100
100
52.5
100
52.5
52.5
5.0
3.00
0.50
3.00
3.00
1.75
3.00
3.00
1.75
3.00
0.50
1.75
0.50
0.50
1.75
1.75
1.75
3.00
0.50
1.75
0.50
0.50
0.50
0.50
1.75
1.75
3.00
1.75
3.00
1.75
22
23
10
27
17
34
27
25
14
26
11
23
27
12
31
18
0[a]
42
20
18
27
18
24
14
4
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
5.0
52.5
100
52.5
5.0
DoE Optimization of the γ-Hydroxylation Using Cu–Al Ox
as a Catalyst
5.0
4
100
100
5.0
23
13
9
The addition of the Cu–Al Ox catalyst to the reaction
mixture produced a substantial improvement in the yield of
the reaction. To ensure that this improvement was due to [a] Only aromatic phenol 3 was detected.
8310 Eur. J. Org. Chem. 2013, 8307–8314
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

DoE-Assisted Allylic Hydroxylation of Enones
the optimal conditions, a yield of 42% was achieved. The Table 6. Oxidation of different substrates with Cu–Al Ox, tBuOK,
and O2.
resulting γ-hydroxy derivative (i.e., 2) represents an impor-
tant synthon for the synthesis of 8-hydroxylated terpenes.
The Pareto plot shows that there is an important interac-
tion between the concentration of carvone and the amount
of base (Figure 4). Furthermore, the amount of Cu–Al Ox
is significant at the 95% confidence level. In contrast, the
role of tBuOK itself is not considered to be significant. This
suggests the existence of a complex scenario with both
heterogeneous and homogeneous phase reagents present.
The corresponding response-surface graph is shown in Fig-
ure 5.
Figure 4. Pareto plot for the oxidation of carvone.
Figure 5. Calculated response surface for the oxidation of carvone.
The remaining substrates were studied in a similar way.
The results are summarized in Table 6. An inspection of
Table 6 reveals that notable improvements in the yield of
the reaction can be obtained by addition of the appropriate
amount of Cu–Al Ox. For instance, Table 6, Entry 3 shows
that the yield of the reaction of acetylcyclohexene (6) was
almost quadrupled by the addition of Cu–Al Ox.
The oxidation of nor-cyperone (8; Table 6, Entry 4), a
substrate that is easily cleaved under oxidative conditions,
proceeds with almost 50% yield. Cyperone itself (12;
Table 6, Entry 6) proceeded quantitatively, providing 6α-hy-
[a] Concentration of substrate [m]; amount of tBuOK [equiv.];
amount of Cu–Al Ox [mg/mmol substrate]. [b] Yield with Cu–Al
droxycyperone derivative 13 in synthetically useful yields.
Table 6, Entry 7 shows the oxidation of a more complex
ketone such as testosterone (14). The inertness of this sub-
strate is well known, and a low yield (20%) of the product
was isolated, although in the absence of Cu–Al Ox, only
starting material was recovered.
Ox/yield without Cu–Al Ox. [c] Isolated yield. [d] 3 d, 75% yield
based on recovered material.
Some Considerations About the Mechanism
The γ-hydroxylation reaction apparently involves the re-
action of the enone with triplet dioxygen in the presence of
Eur. J. Org. Chem. 2013, 8307–8314
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
8311

F. M. Guerra et al.
FULL PAPER
a base.[7] The reaction would be favoured by the presence cases, Cu–Al Ox catalyses the reaction, and the yields were
of the acidic Al3+ cations, which would coordinate to the higher than those observed in experiments in which the
carbonyl group of the ketone, attaching it to the catalyst mixed oxide was not added. The vicinal presence of the
surface. The experimental data seem to support an anionic Al3+ and Cu2+ cations seems to provide an appropriate en-
mechanism involving the interaction between an enolate vironment for the catalytic process to take place. This study
and molecular oxygen. However, this would mean a viola- was carried out using DoE techniques. This approach is
tion of the spin-conservation rule.[30] Alternatively, the enol- especially suitable for the optimization of reactions in which
ate may be converted first into a free radical by donating the involvement of many different species leads to a com-
an electron to an acceptor, before combining with triplet plex mechanism.
dioxygen (Scheme 3).
This procedure provides a route to valuable synthons
such as (+)-5β-hydroxycarvone (2), a substrate whose prep-
aration by previously described routes is problematic. We
are currently using this compound in the synthesis of more
complex terpenes, and the results of this work will be re-
ported in due course.
Experimental Section
Catalyst Preparation: The procedure for the preparation of the cop-
per–aluminium mixed oxide (Cu–Al Ox) is based on that reported
by Guida et al. for the synthesis of a typical hydrotalcite mate-
rial.[24] The molar ratio [Al/(Cu + Al)] is 0.315. A solution contain-
ing Na2CO3 (1.27 g) and NaOH (5.20 g) in water (100 mL) was
added dropwise over 1.5 h to a second solution containing CuCl2
(5.0 g) and AlCl3·6H2O (4.0 g) in water (50 mL). The resulting blue
suspension was stirred at 70 °C for 22 h. A gradual change in col-
our from blue to black was observed. The black precipitate was
filtered, and the residue was washed with warm water. The solid
was then dried in the oven at 105 °C for 24 h, after which time it
was ground to a uniform consistency. Finally, it was left exposed
to air for 3 d before use. The procedure was repeated several times,
and the different batches of material had identical properties and
catalytic behaviour.
General Procedure for the γ- or ε-Hydroxylation of Enones: The
Cu–Al Ox catalyst (see below for Cu–Al Ox/substrate ratio) was
suspended in EtOH with vigorous stirring for 10 min. The substrate
(1 mmol) and tBuOK (the required amount) were added. The mix-
ture was stirred under air for 24 h, after which time the reaction
mixture was filtered through a Celite pad, which was then rinsed
abundantly with methanol. The solvent was then removed from the
filtrate under vacuum. The resulting oily residue was either ana-
lysed by GC or purified by silica gel column chromatography (mix-
tures of EtOAc/hexanes, according to TLC analysis).
Scheme 3. Proposed mechanism for the oxidation of unsaturated
enones with Cu–Al Ox, tBuOK, and O2.
According to Russell, this acceptor could be dioxygen
itself.[7,31] Nevertheless, in the heterogeneous phase, it would
be more likely that the copper ions may act as an electron
sink to provide the corresponding radicals. Coordination of
the O2 molecule to the copper atoms would provide a way
for the transfer of the oxygen atom to the enone. This trans-
fer could be accomplished by the same copper atom at-
tached to the carbonyl oxygen atom,[32] or it could be coop-
eratively assisted by a second copper atom in a different
active centre. Finally, cleavage of the Cu–O bond would
lead to the hydroxylated product. (Scheme 3).
(E)-4-(3-Hydroxy-2,6,6-trimethylcyclohex-1-en-1-yl)but-3-en-2-one
1
(5): H NMR (400 MHz, CDCl3): δ = 7.17 (br. d, J = 16.5 Hz, 1
H), 6.11 (d, J = 16.5 Hz, 1 H), 4.01 (br. t, J = 4.9 Hz, 1 H), 2.29
(s, 3 H), 1.92 (m, 1 H), 1.84 (br. s, 3 H), 1.72 (m, 1 H), 1.66 (m, 1
H), 1.58 (br. s, OH), 1.44 (m, 1 H), 1.06 (s, 3 H), 1.03 (s, 3 H) ppm.
13C NMR (100 MHz, CDCl3): δ = 198.6, 142.9, 138.8, 134.4, 132.6,
69.4, 34.5, 34.4, 28.5, 28.0, 27.4, 27.0, 18.2 ppm. IR (film): ν =
˜
3422, 2939, 1667, 1607, 1361, 1256, 1022, 996, 577 cm–1. HRMS
(ESI): calcd. for C13H21O2 [M + H]+ 209.1542; found 209.1550.
5β-Hydroxycarvone (2): [α]2D0 = +110.72 (c = 2.00, CHCl3). 1H
NMR (400 MHz, CDCl3): δ = 6.69 (br. s, 1 H), 4.96 (m, 1 H), 4.93
(br. s, 1 H), 4.43 (br. d, J = 9.7 Hz, 1 H), 2.68 (ddd, J = 13.7, 9.7,
4.0 Hz, 1 H), 2.49 (dd, J = 16.3, 4.0 Hz, 1 H), 2.37 (dd, J = 16.3,
13.8 Hz, 1 H), 1.77 (m, 3 H), 1.74 (br. s, 3 H) ppm. 13C NMR
(100 MHz, CDCl3): δ = 198.4, 147.4, 143.0, 135.0, 114.6, 68.4, 52.6,
Conclusions
The treatment of α,β- or α,β,γ,δ-unsaturated enones with
a copper–aluminium mixed oxide, Cu–Al Ox, in the pres-
ence of tBuOK and O2 yields the corresponding γ- or ε-
hydroxylated derivatives in moderate to good yields. In all 40.8, 19.0, 15.3 ppm. IR (film): ν = 3411, 2922, 1668, 1321, 1260,
˜
8312
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 8307–8314

DoE-Assisted Allylic Hydroxylation of Enones
1100, 1037, 893, 796 cm–1. HRMS (ESI): calcd. for C10H15O2 [M
+ H]+ 167.1072; found 167.1084.
Acknowledgments
The authors are grateful to the Junta de Andalucía (FQM-169 and
FQM-110) for their financial support. H. V. and J. M. G. thank the
Ministry of Science and Innovation of Spain/FEDER Program of
the EU (Project MAT2008-00889/NAN). The authors are also
thankful to the Servicios Centrales de Ciencia y Tecnología
(SCCyT) of the University of Cadiz for the use of its X-ray diffrac-
tion, atomic spectroscopy, mass spectrometry, NMR spectroscopy,
and electron microscopy facilities. A. L. G. C. and R. M. B. thank
the Spanish Ministry of Education, Culture and Sport for a fellow-
ship.
1
1-(3-Hydroxycyclohex-1-en-1-yl)ethanone (7): H NMR (400 MHz,
CDCl3): δ = 6.74 (br. s, 1 H), 4.42 (m, 1 H), 2.30 (s, 3 H), 2.20 (m,
2 H), 1.95 (m, 1 H), 1.79 (m, 1 H), 1.56 (m, 2 H) ppm. 13C NMR
(100 MHz, CDCl3): δ = 199.7, 140.8, 140.5, 66.5, 31.4, 25.4, 23.0,
19.0 ppm. IR (film): ν = 3366, 2932, 1662, 1361, 1261, 1025, 795,
˜
698, 487 cm–1. HRMS (ESI): calcd. for C8H13O2 [M + H]+
141.0916; found 141.0926.
6α-Hydroxy-4-nor-methyl-7-epi-cyperone (9): [α]2D0 = –79.43 (c =
0.87, CHCl3). 1H NMR (400 MHz, CDCl3): δ = 5.91 (s, 1 H), 4.86
(s, 1 H), 4.56 (s, 1 H), 4.38 (d, J = 2.5 Hz, 1 H), 2.60 (ddd, J =
17.8, 14.8, 5.3 Hz, 1 H), 2.53 (br. s, 1 H), 2.41 (ddd, J = 4.6, 2.2,
1.0 Hz, 1 H), 2.37 (m, 1 H), 2.25 (m, 1 H), 1.83 (dd, J = 14.4,
4.6 Hz, 1 H), 1.74 (br. s, J = 0.6 Hz, 3 H), 1.69 (ddd, J = 13.2, 5.2,
2.2 Hz, 1 H), 1.60 (m, 1 H), 1.46 (dd, J = 13.6, 4.2 Hz, 1 H), 1.42
(s, 3 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 200.2, 168.2,
144.6, 127.5, 112.5, 74.7, 47.0, 39.1, 35.2, 34.8, 34.4, 25.0, 22.7,
[1] C. Harries, Ber. Dtsch. Chem. Ges. 1901, 34, 2105–2106.
[2] C. D. Harries, F. Gollnitz, H. Tietz, W. S. Mills, T. Warunis, A.
Stahler, Justus Liebigs Ann. Chem. Ann. Chem. 1904, 330, 185–
279.
[3] W. Treibs, Ber. Dtsch. Chem. Ges. 1931, 64B, 2178–2184.
[4] W. Treibs, Ber. Dtsch. Chem. Ges. 1932, 65B, 163–168.
[5] E. G. E. Hawkins, J. Chem. Soc. 1955, 3288–3290.
[6] R. Howe, F. J. McQuillin, J. Chem. Soc. 1958, 1513–1518.
[7] A. A. Frimer in The Oxygenation of Enones (Eds.: S. Patai, Z.
Rappoport), John Wiley & Sons, Chichester, UK, 1989, pp.
781–921.
19.3 ppm. IR (film): ν = 3388, 2925, 1669, 1447, 1261, 1002, 892,
˜
697 cm–1. HRMS (ESI): calcd. for C14H21O2 [M + H]+ 221.1542;
found 221.1552.
(4aR,8S)-8-Hydroxy-1,4a-dimethyl-4,4a,5,6,7,8-hexahydro-
naphthalen-2(3H)-one (11): H NMR (400 MHz, CDCl3): δ = 4.91
(dd, J = 9.4, 6.4 Hz, 1 H), 2.61 (ddd, J = 17.5, 15.0, 5.3 Hz, 1 H),
2.41 (ddd, J = 17.5, 4.7, 2.5 Hz, 1 H), 2.02 (m, 3 H), 1.82 (s, J =
4.8 Hz, 3 H), 1.62 (m, 3 H), 1.46 (m, 2 H), 1.38 (s, 3 H) ppm. 13C
NMR (100 MHz, CDCl3): δ = 200.3, 159.6, 131.3, 66.7, 41.4, 39.1,
1
[8] M. Miyashita, T. Suzuki, A. Yoshikoshi, J. Org. Chem. 1985,
50, 3377–3380.
[9] G.-Q. Tian, J. Yang, K. Rosa-Perez, Org. Lett. 2010, 12, 5072–
5074.
[10] Z. Fang, R. Tang, R. Zhang, K. Huang, Bull. Korean Chem.
Soc. 2009, 30, 2208–2212.
35.2, 34.1, 33.2, 24.6, 15.6, 10.4 ppm. IR (film): ν = 3447, 2924,
˜
[11] P. W. Araujo, R. G. Brereton, TrAC Trends Anal. Chem. 1996,
1652, 1445, 1262, 1198, 1018, 615 cm–1. HRMS (ESI): calcd. for
C12H19O2 [M + H]+ 195.1385; found 195.1404.
15, 26–31.
[12] P. W. Araujo, R. G. Brereton, TrAC Trends Anal. Chem. 1996,
15, 63–70.
6α-Hydroxy-7-epi-α-cyperone (13): [α]2D0 = +44 (c = 0.10, CHCl3).
1H NMR (400 MHz, CDCl3): δ = 4.90 (d, J = 1.3 Hz, 1 H), 4.82
(d, J = 1.7 Hz, 1 H), 4.36 (d, J = 1.7 Hz, 1 H), 2.60 (ddd, J = 17.9,
14.8, 5.4 Hz, 1 H), 2.53 (br. s, 1 H), 2.40 (ddd, J = 17.9, 5.1, 2.2 Hz,
1 H), 2.21 (dddd, J = 13.6, 13.6, 5.3, 2.9 Hz, 1 H), 1.88 (s, 3 H),
1.82 (ddd, J = 14.8, 13.4, 5.1 Hz, 1 H), 1.72 (s, 3 H), 1.60 (ddd, J
= 13.4, 5.4, 2.2 Hz, 1 H), 1.37 (s, 3 H), 1.52–1.36 (m, 1 H) ppm.
13C NMR (100 MHz, CDCl3): δ = 200.5, 160.6, 145.8, 132.5, 111.9,
70.0, 48.2, 39.3, 35.6, 35.6, 34.6, 26.2, 23.5, 19.1, 10.9 ppm. IR
[13] M. A. Bezerra, R. E. Santelli, E. P. Oliveira, L. S. Villar, L. A.
Escaleira, Talanta 2008, 76, 965–977.
[14] N. Bouzidi, C. Gozzi, J. Chem. Educ. 2008, 85, 1544–1547.
[15] R. Marín-Barrios, F. M. Guerra, A. L. García-Cabeza, F. J.
Moreno-Dorado, G. M. Massanet, Tetrahedron 2012, 68, 1105–
1108.
[16] For a general introduction to experimental design in chemistry,
see: a) R. Leardi, Anal. Chim. Acta 2009, 652, 161–172; b) R.
Carlson, J. E. Carlson, “Design and Optimization in Organic
Synthesis” in Data Handling in Science and Technology,
Elsevier, Amsterdam, 2005, vol. 24.
[17] F. Cavani, F. Trifirò, A. Vaccari, Catal. Today 1991, 11, 173–
301.
(film): ν = 3454, 2923, 2862, 1662, 1464, 1343, 1199, 1021, 991,
˜
895, 748 cm–1. HRMS (ESI): calcd. for C15H23O2 [M + H]+
235.1698; found 235.1702.
[18] B. F. Sels, D. E. De Vos, P. A. Jacobs, Catal. Rev. 2001, 43, 443–
6β-Hydroxytestosterone (15): [α]2D0 = +17.20 (c = 0.12, CHCl3). H
1
488.
[19] M. Sychev, R. Prihod’ko, K. Erdmann, A. Mangel, R. A.
van Santen, Appl. Clay Sci. 2001, 18, 103–110.
[20] A. Dubey, V. Rives, S. Kannan, J. Mol. Catal. A 2002, 181,
151–160.
[21] U. Costantino, M. Curini, F. Montanari, M. Nocchetti, O. Ro-
sati, J. Mol. Catal. A 2003, 195, 245–252.
[22] M. Campanati, S. Franceschini, O. Piccolo, A. Vaccari, A. Zic-
manis, Catal. Commun. 2004, 5, 145–150.
[23] W. H. Richardson, J. Am. Chem. Soc. 1966, 88, 975–979.
[24] A. Guida, M. H. Lhouty, D. Tichit, F. Figueras, P. Geneste,
Appl. Catal. A 1997, 164, 251–264.
NMR (400 MHz, CDCl3): δ = 5.81 (s, 1 H), 4.34 (t, J = 2.8 Hz, 1
H), 3.65 (m, 1 H), 2.52 (ddd, J = 17.2, 14.9, 5.0 Hz, 1 H), 2.38 (m,
1 H), 2.05 (m, 1 H), 2.03 (m, 1 H), 2.00 (m, 1 H), 1.98 (m, 1 H),
1.87 (m, 1 H), 1.72 (dd, J = 14.3, 4.4 Hz, 1 H), 1.61 (m, 1 H), 1.55
(m, 1 H), 1.45 (m, 1 H), 1.39 (m, 1 H), 1.38 (s, 3 H), 1.35 (m, 1
H), 1.21 (m, 1 H), 1.09 (m, 1 H), 0.98 (m, 1 H), 0.93 (m, 1 H), 0.81
(s, 3 H) ppm. 13C NMR (100 MHz, CDCl3): δ = 200.4, 168.2,
126.6, 110.2, 81.8, 73.2, 53.9, 50.6, 43.1, 38.2, 37.3, 36.6, 34.4, 30.7,
29.9, 23.4, 20.8, 19.7, 11.2 ppm. IR (film): ν = 3396, 2961, 1662,
˜
1447, 1261, 1020, 802, 598 cm–1. HRMS (ESI): calcd. for C19H29O3
[M + H]+ 305.2117; found 305.2125.
[25] P. Massa, F. Ivorra, P. Haure, R. Fenoglio, Catal. Commun.
2009, 10, 1706–1710.
[26] S. Velu, C. Swamy, Appl. Catal. A 1996, 145, 141–153.
[27] A. Alejandre, F. Medina, P. Salagre, X. Correig, J. E. Sueiras,
Chem. Mater. 1999, 11, 939–948.
[28] A. Alejandre, F. Medina, X. Rodriguez, P. Salagre, J. E.
Sueiras, J. Catal. 1999, 188, 311–324.
Supporting Information (see footnote on the first page of this arti-
1
cle): General experimental methods; H and 13C NMR spectra of
the γ-hydroxy enones; granulometry study; X-ray diffraction phase
identification; N2 adsorption/desorption isotherm of Cu–Al Ox.
Eur. J. Org. Chem. 2013, 8307–8314
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
8313

F. M. Guerra et al.
FULL PAPER
[29]
[32] a) S. E. Allen, R. R. Walvoord, R. Padilla-Salinas, M. C.
Kozlowski, Chem. Rev. 2013, 113, 6234–6458; b) H. C. Volger,
W. Brackman, Recl. Trav. Chim. Pays-Bas 1966, 85, 817–833;
c) H. C. Volger, W. Brackman, J. Lemmers, Recl. Trav. Chim.
Pays-Bas 1965, 84, 1203–1229.
M. M. Selim, N. A. Youssef, Thermochim. Acta 1987, 118, 57–
63.
[30] H. R. Gersmann, H. J. W. Nieuwenhuis, A. F. Bickel, Tetrahe-
dron Lett. 1963, 4, 1383–1385.
[31] G. A. Russell, A. G. Bemis, J. Am. Chem. Soc. 1966, 88, 5491–
5497.
Received: July 30, 2013
Published Online: October 31, 2013
8314
www.eurjoc.org
© 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2013, 8307–8314