Copper-catalyzed homolytic and heterolytic benzylic and allylic oxidation using tert-butyl hydroperoxide

Article abstract of DOI:10.1039/a805324c

Allylic and benzylic alcohols were oxidized in good yields to the respective ketones by tert-butyl hydroperoxide (TBHP) in the presence of copper salts under phase-transfer catalysis conditions. This dehydrogenation was found to proceed via a heterolytic mechanism. CuCl₂, CuCl, and even copper powder were equally facile as catalysts, as they were all transformed in situ to Cu(OH)Cl which was extracted into the organic phase by the phase-transfer catalyst (PTC). Deuterium labeling experiments evidenced the scission of the benzylic C-H bond in the rate-determining step. Nonproductive TBHP decomposition was not observed in the presence of the alcohol substrates. Conversely, the oxygenation of π-activated methylene groups in the same medium was found to be a free radical process, and the major products were the appropriate tert-butyl peroxides. Catalyst deactivation, solvent effects, and extraction effects are discussed. By applying Minisci's postulations concerning the relative reactivity of TBHP molecules towards tert-butoxyl radicals in protic and nonprotic environments, the coexistence of the homolytic and the heterolytic pathways can be explained. A complete reaction mechanism is proposed, wherein the free-radical oxidation obeys Kochi's mechanism, and the heterolytic dehydrogenation is based on either a high-valent Cu^IV=O species or a [Cu(OH)Cl]₂ species.

Full text of DOI:10.1039/a805324c

View Article Online / Journal Homepage / Table of Contents for this issue
Copper-catalyzed homolytic and heterolytic benzylic and allylic
oxidation using tert-butyl hydroperoxide
Gadi Rothenberg, Liron Feldberg, Harold Wiener and Yoel Sasson*
Casali Institute of Applied Chemistry, Hebrew University of Jerusalem, Jerusalem 91904, Israel
Received (in Cambridge) 9th July 1998, Accepted 17th September 1998
Allylic and benzylic alcohols were oxidized in good yields to the respective ketones by tert-butyl hydroperoxide
(TBHP) in the presence of copper salts under phase-transfer catalysis conditions. This dehydrogenation was found to
proceed via a heterolytic mechanism. CuCl2, CuCl, and even copper powder were equally facile as catalysts, as they
were all transformed in situ to Cu(OH)Cl which was extracted into the organic phase by the phase-transfer catalyst
(PTC). Deuterium labeling experiments evidenced the scission of the benzylic C–H bond in the rate-determining step.
Nonproductive TBHP decomposition was not observed in the presence of the alcohol substrates. Conversely, the
oxygenation of π-activated methylene groups in the same medium was found to be a free radical process, and the
major products were the appropriate tert-butyl peroxides. Catalyst deactivation, solvent effects, and extraction
effects are discussed. By applying Minisci’s postulations concerning the relative reactivity of TBHP molecules
towards tert-butoxyl radicals in protic and nonprotic environments, the coexistence of the homolytic and the
heterolytic pathways can be explained. A complete reaction mechanism is proposed, wherein the free-radical
oxidation obeys Kochi’s mechanism, and the heterolytic dehydrogenation is based on either a high-valent
CuIV᎐O species or a [Cu(OH)Cl] species.
᎐
2
or ketones was observed with 70% aqueous TBHP. Good con-
versions and excellent selectivities were observed for various
substrates, as shown in Table 1. Primary and unactivated
alcohols could also be oxidized, but yields were lower. 1-
Introduction
The mechanism of oxidation using Fenton-type reagents has
been the subject of recent debate, with specific regard to the
participation of free-radical intermediates in the oxidation of
aliphatic and aromatic compounds with H2O2–M and ROOH–
Phenylethanol 1 was chosen as a model substrate for a mech-
anistic investigation of the oxidation of benzylic alcohols with
M reagent systems. Sawyer et al. have stated that free carbon
TBHP [eqn. (1)].
radicals are not produced in Fenton oxidation systems,1 while
Walling, and MacFaul et al. have argued that Fenton and
Fenton-type reagents (i.e. H2O2 or TBHP with Fe/Co/Cu cata-
lysts) would have to include free-radical intermediates if they
are to provide a satisfactory explanation of the observed kinetic
and product studies.2,3
CuCl2 1 mol%
TBAB 3 mol%
O
OH
+ TBHP
(1)
+
t-BuOH + H2O
25 oC, CH2Cl2, 24 h
Ph
Ph
1
2
Allylic and benzylic oxidation reactions are the foundation of
many important current industrial and fine-chemical pro-
cesses,4 covering a diverse field from natural product chemistry
to “buckybowl” synthesis.5 Traditionally, these reactions have
been performed with stoichiometric chromium reagents, e.g.
CrO3,6a or Cr–pyridine complexes.6b,6c However, today’s
environmental restrictions render most of the stoichiometric
metal oxidants obsolete. Using TBHP as an alternative oxygen
source7a has been investigated with several transition-metal
catalysts, including Cr, Ru, Rh, and Cu.7b–g The latter is inter-
esting because it is both cheap and relatively benign.8 Recent
oxidations with TBHP and copper catalysts show that the
immediate surroundings of the Cu atom can influence reaction
enantioselectivity,9a,9b and maybe tell us something about the
activity of copper-containing monooxygenases.9c
Remarkably, no unproductive decomposition of TBHP was
observed in this system. Yields were nearly quantitative and no
excess of TBHP was required. Consequently, the oxidant can be
added in one batch without overheating risks. It should be
noted, however, that when the reagents were mixed without the
alcohol substrate, a free-radical chain decomposition of TBHP
was observed.11
Examination of various organic solvents showed that CH2Cl2
and PhCN were superior to PhNO2 and cyclohexanone (Fig. 1).
This probably reflects different partition coefficients for the
catalytic species between the organic and aqueous phases.
We have also carried out the reaction in a dry mono-
phasic system, using t-BuOH or MeCN as solvents. The copper
catalysts were soluble in these media, which was an advantage,
and high conversions and selectivities were obtained (GC
indicated over 95% conversion with 2 being the sole product
after 45 h). The main drawback of the monophasic approach
was that it made product separation difficult, while from the
biphasic medium the product could be purified by simply
separating the phases and washing the organic phase with
water. A further disadvantage of the monophasic system was
that unproductive decomposition of TBHP occurred.
Herein we present the results of product studies and kinetic
investigations of the title reactions, and explain the funda-
mental mechanistic differences between the oxidation of alco-
hols and activated methylene groups in the TBHP–Cu–PTC
system.
Results
Oxidation of alcohols10
The presence of both catalysts was found to be essential. No
reaction was observed in the absence of copper, and very low
conversions were obtained in the absence of TBAB. Bubbling
O2 or Ar through the reaction mixture (with TBHP) did not
In the presence of catalytic amounts of CuCl2 and a phase-
transfer catalyst such as n-Bu4NϩBrϪ (TBAB), a smooth oxid-
ative dehydrogenation of alcohols to the appropriate aldehydes
J. Chem. Soc., Perkin Trans. 2, 1998, 2429–2434
2429

View Article Online
Table 1 Copper-catalyzed oxidation of alcohols with TBHPa
Entry
Substrate
% Conversionb
Product
% Selectivityb
100
OH
O
1
35b
2
3
80 (17)
96 (4)
100 (100)
100 (100)
OH
O
O
OH
Ph
OEt
OEt
Ph
O
O
OH
O
OEt
4
5
6
60b
OEt
100
O
O
OH
O
75 (3)
71 (4)
100 (100)
100 (100)
Ph
Ph
OH
O
OH
O
7
67 (3)
97 (100)
8
9
38 (7)
6c
100 (100)
66
Ph
Ph
Ph
OH
O
OH
OH
O
Ph
O
10
20c
100
C6H13
C6H13
OH
OH
O
11
12
6c
100
C6H13
C6H13
O
53 (24)
59 (58)
Ph
Ph
a Reaction conditions: 25.0 mmol substrate, 25.0 mmol 70% aq. TBHP, 24 h, l mol% CuCl2, 3 mol% TBAB, 25 ЊC, 10 ml CH2Cl2. b The conversion
and selectivity in the absence of TBAB are indicated in brackets. c No conversion was detected in the absence of TBAB.
alter the results, while substitution of the TBHP with H2O2 gave
no products, and a nonproductive decomposition of H2O2 was
observed.
Several phase-transfer catalysts were tested. Similar results
were obtained with n-Bu4NϩClϪ, TBAB, n-Bu4NϩIϪ, (C10H21)2-
Me2NϩBrϪ, C16H33(Me)3NϩBrϪ, and Aliquat 336, indicating
that cation symmetry and halide anion type do not influence
the catalytic activity. Ammonium salts that lack a halide coun-
terion, e.g. n-Bu4NϩHSO4Ϫ, were inactive. Nevertheless, it is not
the halide ion alone that is responsible for the catalysis, as KBr
and Br2 failed to catalyze the reaction. The hydrophilic PTC
Me4NϩBrϪ was also inactive, as expected for biphasic systems
in which the reaction takes place in the organic phase.12
A series of blank reactions were run in order to elucidate the
nature and structure of the active catalytic species. We dis-
covered that the oxidation can be carried out with equal facility
Fig. 1 Oxidation of 1 using different solvents. Reaction conditions:
26.0 mmol 1, 26.0 mmol TBHP, 2 mol% CuCl2, 3 mol% TBAB, 25 ЊC,
10 ml solvent.
using CuCl2, CuCl, and even plain copper metal as the catalyst
(cf. reaction profiles in Fig. 2). A colloidal dispersion of copper
in CH2Cl2 had the same catalytic activity as ordinary copper
2430
J. Chem. Soc., Perkin Trans. 2, 1998, 2429–2434

View Article Online
Fig. 2 Oxidation of 1 using ᭺, CuCl2; ᭹, CuCl; ᭡, Cu. Reaction
conditions: 15.4 mmol 1, 15.4 mmol TBHP, 2 mol% catalyst, 3 mol%
TBAB, 25 ЊC, 18 ml CH2Cl2.
Fig. 4 Oxidation of 1 and 1-d. Reaction conditions: 26.0 mmol sub-
strate, 26.0 mmol TBHP, 0.33 mol% CuCl2, 1 mol% TBAB, 25 ЊC, 10 ml
CH2Cl2.
presence of Cu2Cl(OH)3 (d-values: 5.46, 2.76, 2.27) and CuCl2
(d-values: 5.44, 4.02, 2.64), but no CuI species were detected.
The binuclear species Cu2Cl(OH)3 has been proposed previ-
ously as the deactivated catalyst in copper-catalyzed oxid-
ations.14
In order to gain additional insight into the reaction mechan-
ism, the deuterated substrate Ph-CDOH-CH3 1-d was prepared
and oxidized. Fig. 4 shows the reaction profiles for 1 and 1-d,
respectively. An uncommonly large kinetic isotopic effect (KIE)
was observed, kH/kD = 12.9. This asserts that the RDS involves
breaking the benzylic C–H bond.
We have found that the RDS in the oxidation of 1 and analo-
gous compounds in this system is heterolytic, i.e. involves the
transfer of two electrons rather than free radicals. This was
supported by several observations: (1) Addition of a radical
scavenger (1,6-di-tert-butyl-4-methylphenol, BHT) to the
reaction mixture had no effect on the reaction rate.15 (2) No
hydroperoxide or peroxide intermediates were detected by
iodometric titration. (3) UV–Visible and XR diffraction
spectra indicated the presence of Cu() species, but not Cu(),
making a one-electron transfer harder to envisage. (4) Single
products were obtained in high selectivities, which is uncom-
mon in free-radical processes. (5) No changes were observed
when oxygen was bubbled through the reaction mixture,
indicating that the presence of alkyl radicals is unlikely.
Significantly, no benzylic oxidation was observed at the 4-
and 3-positions of 1-tetralol (tetrahydro-1-naphthol) and
indan-1-ol, respectively. This finding suggests a mechanistic
difference between the oxidation of alcohols and that of π-
activated methylenes in this system.
Fig. 3 Dependence of k and kd on the initial concentration of CuCl2.
Reaction conditions: 15.0 mmol 1, 15.0 mmol TBHP, 0.5–8 mol%
CuCl2, 2 mol% TBAB, 25 ЊC, 18 ml CH2Cl2.
powder. Atomic absorption (AA) and UV–visible spectra of the
aqueous and the organic phases showed that CuCl2 was not
extracted into CH2Cl2 by TBAB. Conversely, Cu(OH)2 and
Cu(OH)Cl were extracted efficiently into CH2Cl2.13
Note that the final conversion in Fig. 2 is lower than that in
Fig. 1. This system is sensitive to concentration changes, espe-
cially of the metallic catalyst. The catalyst undergoes a
deactivation process, which is much more pronounced at low
concentrations. Fig. 3 shows the dependence of the reaction
rate constant k, and the catalyst deactivation rate constant kd,
on the initial amount of the copper catalyst.
The kinetics of the oxidation reaction were found to fit a
first-order profile with first-order deactivation of the catalyst
[eqn. (2), where [A]t is the concentration of the substrate at
Oxygenation of ð-activated methylenes16
[A]t
k
ln
ͩ
ln
ͪ
= ln Ϫ kdt
kd
(2)
The oxidation of 1,2,3,4-tetrahydronaphthalene (tetralin, 3)
proceeded to 50% substrate conversion when a 1:1 molar
ratio of TBHP:Substrate was employed [eqn. (3)]. The major
[A]∞
time t, [A]∞ is the concentration of the substrate at time ∞, k
is the reaction rate constant, and kd is the catalyst deactiv-
ation rate constant; see experimental section for the derivation
of eqn. (2)]. Experimental measurements (for 1 mol% CuCl2)
OOt-Bu
1 mol% CuCl2
3 mol% TBAB
gave k = 3.88 × 10Ϫ5 Ϫ1, and kd = 2.77 × 10Ϫ5 sϪ1 (r2 = 0.996).
s
+ TBHP
50% conversion
25 oC, CH2Cl2
Further rate measurements indicated 1st order in the sub-
strate and zero order for the oxidant (r2 = 0.983). Altering the
stirring rate (50–1000 rpm) did not affect the reaction, indicat-
ing that the rate-determining step (RDS) is chemically, rather
than diffusively, controlled. This was supported by calculating
the Arrhenius energy of activation, which gave Ea = 50 kJ
molϪ1 (12 kcal molϪ1, r2 = 0.991 for four measurements at 1,
11, 28 and 39 ЊC), a typical value for a chemically controlled
RDS.
3
4
(75%)
O
O
OH
+
+
+ t-BuOH + H2O
+
(3)
XR diffraction spectroscopy of the powder residue con-
taining the catalyst at the end of the reaction indicated the
OH
7 (4%)
5 (17%)
6
(2%)
J. Chem. Soc., Perkin Trans. 2, 1998, 2429–2434
2431

View Article Online
Table 2 Oxidation of π-activated methylenes with TBHPa
Entry Substrate
Time/h
Conversion (%)
Product (% selectivity)
1
2
3
4
5
6
7
8
9
Tetralin
4
4
6
24
4
48
4
50
8
1-t-Bu-peroxytetralin (75), 1-tetralone (17), 4-hydroxytetralone (8), 1-tetralol (2)
1-t-Bu-peroxytetralin (50), 1-tetralol (50)
Tetralinb
Indane
50
18
32
10
50
6
17
18
12
36
1-t-Bu-peroxyindane (70), indan-1-one (25), indan-1-ol (4), 1-indene (1)
1-t-Bu-peroxyindane (28), indan-1-one (55), indan-1-ol (17)
1-t-Bu-peroxydiphenylmethane (16), benzophenone (78), diphenylcarbinol (6)
1-t-Bu-peroxydiphenylmethane (40), benzophenone (55), diphenylcarbinol (5)
1-t-Bu-peroxyfluorene (tr), fluorenone (86), fluorenol (14)
1-t-Bu-peroxyfluorene (14), fluorenone (86)
Indaneb
Diphenylmethane
Diphenylmethaneb
Fluorene
Fluoreneb
4
(R)-Limonene
α-Pinene
14
14
4
t-Bu-peroxylimonenec (17), carvone (20), limonene oxide (26)
t-Bu-peroxypinenec (15), verbenone (22), pinene oxide (29)
t-Bu-peroxycumene (17), 1-phenylpropan-2-ol (83)
10
11
12
Cumened
Ethylbenzened
9
1-Phenyl-1-t-Bu-peroxyethane (10), acetophenone (72), 1-phenylethanol (18)
a Reaction conditions: 25.0 mmol substrate, 25.0 mmol 70% aq. TBHP, 1 mol% CuCl2, 3 mol% TBAB, 25 ЊC, 10 ml CH2Cl2. b No TBAB. c The exact
structure could not be determined. d No conversion was detected in the absence of TBAB.
product was 1-tert-butylperoxytetralin 4 (75%), together with
1-tetralone 5 (17%), 1-tetralol 6 (2%), naphthalene (2%), and
4-hydroxy-1-tetralone 7 (4%). Using a 2:1 ratio of TBHP to
tetralin, as the stoichiometry of the reaction demands, led to a
higher conversion but with poorer selectivity for 4.
posed earlier by Kochi [eqn. (4)–(8)], who used a series of redox
reactions to explain the formation of the tert-butyl peroxides.17
II
I
ϩ
ؒ
Cu ϩ TBHP → Cu ϩ t-BuOO ϩ H
(4)
(5)
(6)
(7)
(8)
I
II
ؒ
Cu ϩ TBHP → Cu OH ϩ t-BuO
CuIIOH ϩ TBHP → CuII-OO-t-Bu ϩ H2O
Separate oxidation experiments performed on an isolated
sample of the peroxide 4 showed that it could be an inter-
mediate in the formation of 5 and 7 (6 was not detected). This
indicates that there are two possible pathways for the formation
of the ketone 5, i.e. 4→5 and 6→5. The latter is not sufficiently
fast for the reaction to proceed by the pathway 4→6→5, as
some 6 would have to accumulate. The decomposition of 4 to 5
and 7 was found to be temperature-sensitive. In the absence of
TBHP at 25 ЊC, 4 was stable in the presence of Cu–TBAB, but
decomposed when fresh TBHP was added. Conversely, at
60 ЊC, the presence of Cu–TBAB was sufficient to transform 4
to 5 and 7. A commercial sample of 5 was oxidized (with an
excess of 4 equiv. of TBHP) to 7, but the conversion was only
9% after 3 d.
Similar results were obtained with several other benzylic and
allylic olefins. The results are summarized in Table 2. Again, the
presence of both catalysts was required. It is significant that
some tertiary peroxide was obtained in all cases, because this
could only result from the participation of the tert-butylperoxy
radical. As expected, we found that BHT addition inhibits the
oxidation of 3.
ؒ
ؒ
t-BuO ϩ RH → t-BuOH ϩ R
II
I
ؒ
Cu -OO-t-Bu ϩ R → R-OO-t-Bu ϩ Cu
The mechanism of the oxidative dehydrogenation of alcohols
to the appropriate aldehydes or ketones is somewhat more elu-
sive. Experimental data point to a heterolytic process, and the
large kinetic isotope effect indicates cleavage of the benzylic
C–H bond in the rate-determining step. Fortunately, the
mechanistic picture can be elucidated by taking into account
the extraction of the copper species between the aqueous and
organic phases and Minisci’s reasoning concerning the hydro-
gen abstraction from TBHP by the tert-butoxyl radical [eqn. (9)]
in the presence and in the absence of alcohols. Minisci has
postulated18 that the high rate constant measured19 for eqn. (9)
ؒ
ؒ
t-BuO ϩ TBHP → t-BuOH ϩ t-BuOO
(9)
(k = 2.5 × 108 MϪ1 sϪ1 at 22 ЊC) is dramatically lowered in a
medium that allows hydrogen bonding with the TBHP, in
accordance with the fact that free-radical TBHP decomposition
was not observed in the presence of alcohol substrates.
Hydrogen-bonding between the quaternary ammonium salt
and the TBHP can also occur,20 but its effect would be limited
due to the small amount (1–3 mol%) of ammonium salt pres-
ent. Additional experiments showed that the rate of tetralin
oxidation diminished by a factor of 3.5 in the presence of an
equimolar amount of tetralol.
Discussion
The TBHP–Cu–TBAB system involves a complex set of inter-
related reactions (Scheme 1): (a) benzylic/allylic tert-butyl per-
OOt-Bu
O
(b)
(a)
R
R'
R
R'
R
R'
R = aryl, allyl
OH
O
(d)
We therefore suggest that when 70% aqueous TBHP is added
to a mixture of substrate and catalytic amounts of CuCl2 and
TBAB in CH2Cl2, eqn. (10) and (11) take place in the aqueous
(c)
(e)
R
R'
R'
R
R'
R
R'
O
II
I
ؒ
Cu Cl2 ϩ TBHP → Cu Cl ϩ t-BuOO ϩ HCl (10)
R
R'
allylic subs.
only
I
II
ؒ
Cu Cl ϩ TBHP → Cu (OH)Cl ϩ t-BuO
(11)
Scheme 1 Oxidation of allylic and benzylic olefins.
phase. With substrates that cannot form a hydrogen bond with
the TBHP molecule (e.g. tetralin), the formation of the tert-
butoxyl radical would lead to instantaneous abstraction of the
peroxidic hydrogen from another TBHP molecule, generating
the tert-butylperoxy radical and eventually producing the
peroxide 4.
oxide formation; (b) peroxide decomposition affording the
ketone as the main product; (c) formation of the allylic/
benzylic alcohol; (d) oxidative dehydrogenation of the alcohol
to the ketone; (e) double-bond epoxidation (allylic substrates
only).
Our data pertaining to the oxidation of allylic and benzylic
olefins can be explained by adaption of the mechanism pro-
Conversely, in the case of alcohol oxidation, e.g. 1-tetralol,
hydrogen bonding between the peroxidic hydrogen of the
2432
J. Chem. Soc., Perkin Trans. 2, 1998, 2429–2434

View Article Online
TBHP and the alcohol would inhibit the abstraction of that
hydrogen by the tert-butoxyl radical. CuII(OH)Cl which was
generated in eqn. (11) would be extracted into the organic phase
by TBAB [note that ammonium cations paired with a soft
anion, e.g. HSO4Ϫ, cannot form hydrogen bonds with CuII-
(OH)Cl, and evidently do not catalyze the reaction], becoming
a precursor for an active catalytic species: one possibility would
be interaction of CuII(OH)Cl with the hydrogen-bonded pair
Conclusions
In the Fenton-type TBHP–Cu system under PTC conditions,
the oxygenation of π-activated methylenes is indeed a classic
free-radical Kochi17a process. Conversely, the oxidation of
alcohols in the same medium follows a heterolytic mechanism.
Alcohol substrates interact by hydrogen bonding with the
TBHP oxidant, and this lessens the free-radical peroxidic
hydrogen abstraction by the tert-butoxyl radical. In this case, the
TBHP–ROH to form an active CuIV᎐O catalytic species21
᎐
active catalyst may be a high-valent CuIV᎐O species, generated
by oxidation of Cu(OH)Cl by TBHP.
᎐
(Scheme 2, top). Ketone formation and regeneration of the
CuII(OH)Cl species would then be possible by abstraction of
the geminal hydrogen from the alcohol as a hydride via a
seven-membered pericyclic reaction (Scheme 2, bottom).
Experimental
1H NMR spectra were measured in CDCl3 on a Bruker AMX
300 instrument at 300.13 MHz. δH values are reported in ppm
downfield from TMS. GC and GCMS analysis were performed
using an HP-5790 gas chromatograph with a 5% diphenyl 95%
dimethyl capillary column (25 m/0.31 mm). HPLC analysis was
performed on a Hewlett-Packard series 1050 instrument, using
H
O
H
O
R
R
H
O
O
H
O
O
Cu
Cu
an RP18 column with 80:20 MeCN:H2O as eluent (1 ml minϪ1
,
Cl
O
H
Br
Cl
O H
Br
NBu4
NBu4
UV detector, 250 nm). Atomic absorption measurements were
performed on a GBC 903 single beam spectrophotometer.
Powder XR diffraction spectra were measured with a Philips
PW1830 diffractometer. Unless stated otherwise, reagents were
purchased from Aldrich, Fluka, and Merck, and were used as
received. 70% TBHP (remainder water) was purchased from
Sigma. 1-d was synthesized and identified by its spectral proper-
ties. Products were either isolated and identified by comparison
R1
R2
H
O
H
O
O
Cu
CuII(OH)Cl + TBAB-H2O +
Cl
O
H
Br
R1
R2
NBu4
Scheme 2 Suggestions for the formation of an active CuIV species
(top) and the oxidation of the alcohol substrate (bottom).
1
of their H NMR spectra to commercial samples, or identified
by MS data and comparison of their GC retention times with
commercial samples.
However, we have not been able to prove the existence of the
CuIV᎐O species.
᎐
Preparation of a colloidal dispersion of copper in CH2Cl2
Another possible pathway† involves the dimerization of
CuII(OH)Cl and subsequent coordination of the TBHP and
ROH molecules. As depicted in Scheme 3, this mechanism
An aqueous solution of 0.056 g (1.5 mmol) NaBH4 was added
to a flask containing 40 mg (0.3 mmol) CuCl2, 108 mg (0.3
mmol) C16H33(Me)3NϩBrϪ, and 10 ml CH2Cl2. The mixture was
stirred for 2 h at 25 ЊC. An intense red colour indicated the
presence of colloidal Cu0 in the organic phase. The phases were
separated when the green color disappeared from the aqueous
phase, indicating that all of the CuCl2 was reduced.
TBHP
R1R2CHOH
active dimer species
Cl
Cu
2H2O
Cl
OH
HO
Cu
Preparation of 1-phenyl-1-deuteroethanol 1-d
Cl
A suspension of 0.8 g (21 mmol) LiAlD4 in 25 ml of dry ether
was added dropwise to a gently boiling solution of 0.49 g (41
mmol) 2 in 6.5 ml dry ether. The mixture was kept at a gentle
boil for 30 min, after which the LiAlD4 remnants were neutral-
ized with EtOAc, and the solution was brought to pH = 7 with
aqueous HCl. Separation of the organic phase, drying over
Na2SO4, filtration, and evaporation of the solvent in vacuo
afforded 4.5 g (89 mol% based on 2) of pure 1-d. δH: 7.22 (5H,
m), 2.01 (1H, s), 1.38 (1H, s,). [cf. with 1: 7.22 (5H, m), 4.78
(1H, d,), 1.79 (1H, s), 1.38 (3H, d)]; m/z: 123 (cf. with 1, 122).
Isotopic purity was >99.9% according to 1H NMR.
Cl
Cu
OH
O
O
Cu
TBHP
Cl
Cu
Cu
2 HO-Cu-Cl
- t-BuO•
OH
O
HO
R2
H
eqn.
(10) and (11)
deactivated
catalyst
R1
HCl
[from eqn. (10)]
CuCl2
+
Cu(II)=O
t-BuOH
R1R2C=O
General procedure for oxidation of allylic and benzylic alcohols
and olefins
Scheme 3 Suggestions for alcohol oxidation via a dimeric catalytic
species.
Example: tetralin 3. 3.46 g (26.2 mmol) of 3, 35 mg (0.26 mmol.
1 mol%) of CuCl2, 246 mg (0.78 mmol, 3 mol%) of TBAB and
10 ml CH2Cl2 were stirred at 25 ЊC. 3.4 g of TBHP (70% aq.;
26.2 mmol) and 0.2 g of m-C6H4Cl2 (internal standard) were
added and the mixture was stirred for 4 h. Reaction progress
was monitored by GC/HPLC. Products were either identified
by GCMS or purified from the organic phase by flash chrom-
atography on silica (Merck, 0.040–0.063 mm) and identified by
could also account for the catalyst deactivation, through form-
ation of the inactive dimer species. Further study of these
mechanistic possibilities is in progress.
The oxidation of alcohols would not be affected by the
presence of free-radical scavengers such as BHT, because the
generation of the extractable catalytic species would occur
in the aqueous phase. The in situ transformation of Cu0 to
soluble compounds by TBHP has been envisaged.7g
1
their H NMR spectra, e.g. 4 was isolated in 75% yield and
>99.5% purity using a gradient of petroleum ether:EtOAc
from 100:0 to 90:10 with a drip rate of 90 drops minϪ1. δH: 1.27
(9H, s), 1.7–1.82 (2H), 1.93 (1H, m), 2.33 (1H, m), 2.58–2.9
† We thank a referee for this suggestion.
J. Chem. Soc., Perkin Trans. 2, 1998, 2429–2434
2433

View Article Online
(2H), 5.46 (1H, t), 7.1–7.6 (4H). Good agreement was found
with lit. values.22
adding metal catalysts (we have had a minor explosion when
adding RuCl3 to a 70% TBHP soln). Small-scale reactions are
preferable.
Experimental procedure for kinetic studies
References
Example: 1.88 g (15.4 mmol) of 1, 26 mg (0.15 mmol, 1 mol%)
of CuCl2, 90 mg (0.28 mmol, 1.9 mol%) of TBAB, 0.5 g of m-
C6H4Cl2 (internal standard) and 18 ml CH2Cl2 were mechanic-
ally stirred (300 rpm) at 25 ЊC for 24 h. Reaction progress was
monitored by GC. The following parameters were studied: sub-
strate concentration (5 expts., with [1] from 0.19 to 3.08 M);
CuCl2 amount (7 expts., with 0, 0.5, 1, 2, 3, 4 and 8 mol% of
CuCl2); TBAB amount (8 expts., with 0, 1, 2, 3, 4, 5, 6 and 8
mol% of TBAB); stirring rate (7 expts., at 0, 180, 270, 480, 584,
680 and 820 rpm); and temperature (4 expts. at 1, 11, 28 and
39 ЊC). The best fit (r2 = 0.993 for 8 data points) was obtained
with eqn. (2) (vide infra).
1 D. T. Sawyer, A. Sobkowiak and T. Matsushita, Acc. Chem. Res.,
1996, 9, 409.
2 C. Walling, Acc. Chem. Res., 1998, 31, 155.
3 P. A. MacFaul, D. D. M. Wayner and K. U. Ingold, Acc. Chem. Res.,
1998, 31, 159.
4 (a) For a monograph on oxidation reactions, see R. A. Sheldon and
J. K. Kochi, Metal-Catalyzed Oxidations of Organic Compounds,
Academic Press, New York, 1981; (b) for recent studies on the
oxidation of allylic and benzylic alcohols, see K. P. Peterson and
R. C. Larock, J. Org. Chem., 1998, 63, 3185, and refs. cited therein.
5 (a) H. H. Szmant, Organic Building Blocks of the Chemical Industry,
Wiley, New York, 1989, pp. 386–391; (b) M. D. Clayton,
Z. Marcinow and P. W. Rabideau, J. Org. Chem., 1996, 61, 6052.
6 (a) M. Nakayama, S. Shinke, Y. Matsushita, S. Ohira and
S. Hayashi, Bull. Chem. Soc. Jpn., 1979, 52, 184; (b) W. G. Dauben,
M. Lorber and D. S. Fullerton, J. Org. Chem., 1969, 34, 3587;
(c) E. J. Parish, S. Chitrakorn and T.-Y. Wei, Synth. Commun., 1986,
16, 1371.
7 (a) K. B. Sharpless and T. R. Verhoeven, Aldrichim. Acta, 1979, 12,
63; (b) J. Muzart and A. Naît-Ajjou, J. Mol. Catal., 1991, 66, 155;
(c) Synthesis, 1993, 785; (d) J. Muzart, Tetrahedron Lett., 1986, 27,
3139; (e) 1987, 28, 2131; ( f ) S. Uemura and S. R. Patil, Tetrahedron
Lett., 1982, 23, 4353; (g) J. A. R. Salvador, M. L. Sa e Melo and
A. S. C. Neves, Tetrahedron Lett., 1997, 38, 119.
The derivation of eqn. (2)
Assuming that the oxidation of 1 and the catalyst deactivation
are both first-order reactions, the following rate laws apply: eqn.
(12) and (13), where [A] is the concentration of the substrate at
Ϫd[A]
ϪrA =
= kr[A]a
(12)
(13)
dt
Ϫda
8 (a) I. E. Marko, P. R. Giles, M. Tsukazaki, S. M. Brown and C. J.
Urch, Science, 1996, 274, 2044; (b) J. Muzart, J. Mol. Catal., 1991,
64, 381.
= kda
dt
9 (a) M. T. Rispens, C. Zondervan and B. L. Feringa, Tetrahedron:
Asymmetry, 1995, 6, 661; (b) A. Levina and J. Muzart, Tetrahedron:
Asymmetry, 1995, 6, 147; (c) Z.-R. Lu, Y.-Q. Yin and D.-S. Jin,
J. Mol. Catal., 1991, 70, 391.
10 Preliminary communication: L. Feldberg and Y. Sasson, J. Chem.
Soc., Chem. Commun., 1994, 1807.
11 cf. (a) R. A. Miller, W. Li and G. R. Humphrey, Tetrahedron Lett.,
1996, 37, 3429; (b) M. Harre, R. Haufe, K. Nickish, P. Weinig,
H. Weinmann, W. A. Kinney and X. Zhang, Org. Proc. Res. Dev.,
1998, 2, 100; (c) G. Rothenberg, H. Wiener and Y. Sasson, J. Mol.
Catal., 1998, 136, 251; (d) See also J. Muzart and A. Naît-Ajjou, in
The Activation of Dioxygen and Homogenous Catalytic Oxidation,
D. H. R. Barton, A. R. Martell and D. T. Sawyer, ed., Plenum Press,
New York, 1993, p. 471.
12 For a monograph on the fundamentals of PTC, see C. M. Starks,
C. L. Liotta and M. Halpern, Phase-Transfer Catalysis, Chapman
and Hall, New York, 1994.
13 For a review on PTC extraction by hydrogen-bonding, see Y. Sasson
and R. Neumann, in Handbook of Phase Transfer Catalysis,
Y. Sasson and R. Neumann, ed., Chapman and Hall, London, 1997,
pp. 510–546.
14 J. B. Sharkey and S. Z. Lewin, Thermochim. Acta, 1972, 3, 189.
15 I. W. C. E. Arends, K. U. Ingold and J. Lusztyk, J. Am. Chem. Soc.,
1993, 115, 466.
16 Preliminary communication: L. Feldberg and Y. Sasson,
Tetrahedron Lett., 1996, 12, 2063.
17 (a) J. K. Kochi, Tetrahedron, 1962, 18, 483; (b) J. Am. Chem. Soc.,
1962, 84, 1572.
18 (a) F. Minisci, F. Fontana, S. Araneo, F. Recupero, S. Banfi and
S. Quici, J. Am. Chem. Soc., 1995, 117, 226; (b) F. Minisci,
F. Fontana, S. Araneo, F. Recupero and L. Zhao, Synlett, 1996, 119.
19 H. Paul, R. D. Small and J. C. Scaiano, J. Am. Chem. Soc., 1978,
102, 4520.
20 E. Napadensky and Y. Sasson, J. Chem. Soc., Chem. Commun.,
1991, 65.
21 Similar transient high oxidation state copper species have been
envisaged: (a) D. H. R. Barton, S. D. Bévière, W. Chavasiri, É.
Csuhai and D. Doller, Tetrahedron, 1992, 48, 2895; (b) S.-I.
Murahashi, Y. Oda, T. Naota and M. Komiya, J. Chem. Soc., Chem.
Commun., 1993, 139.
time t, a is the activity of the catalyst, kr is the reaction rate
constant with no deactivation and kd is the catalyst deactivation
rate constant.
Likewise, the change in the concentration of the substrate
with time is given by eqn. (14) and (15), where NA is the number
Ϫd[A] Ϫ1 dNA
=
(14)
(15)
dt
V
dt
Ϫd[A]
dt
W
V
Ϫ1 dNA
=
ͩ ͪ= k[A]a
W
dt
of moles of A, V is the reaction volume, W is the weight of the
catalyst, and k is the reaction rate constant with the deactiv-
ation taken into account.
Integration of eqn. (13) gives eqn. (16), where a0 is the activ-
ity of the catalyst at time t = 0 (it is assumed that a0 = 1).
Substituting eqn. (16) into eqn. (15) gives, after integration,
a = a0 exp (Ϫkdt)
(16)
eqn. (17) for the reaction at time t, and eqn. (18) for the reaction
at time t = ∞. Simple mathematical operations lead from eqn.
(17) and (18) to eqn. (2).
[A]0
k
ln
=
[1 Ϫ exp(Ϫkdt)]
(17)
(18)
[A] kd
[A]0
k
ln
=
[A]∞ kd
CAUTION! Although relatively safe to work with, TBHP,
like almost all substances containing peroxidic bonds, should
be handled cautiously. Concentrations above 95% purity should
be avoided. Containers should be kept at 5–10 ЊC, to avoid
layer separation. Strong acids should not be added to high-
strength TBHP solutions. Extreme care should be taken when
22 J. Muzart and A. Naît-Ajjou, J. Mol. Catal., 1994, 92, 277.
Paper 8/05324C
2434
J. Chem. Soc., Perkin Trans. 2, 1998, 2429–2434