Selective catalytic oxidation of cyclohexylbenzene to cyclohexylbenzene-1-hydroperoxide: A coproduct-free route to phenol

Article abstract of DOI:10.1016/S0040-4020(02)01131-6

A method is described for the highly selective oxidation of cyclohexylbenzene to cyclohexylbenzene-1-hydroperoxide. In the presence of 0.5mol% N-hydroxyphthalimide (NHPI) and 2mol% of the hydroperoxide product, without solvent, a selectivity of ca. 98% to the desired product was obtained at 32% conversion. The use of NHPI increases the selectivity for initial H-abstraction from the 1-position, vs the other positions in the cyclohexyl ring, suppresses byproduct formation via transannular hydrogen abstraction and increases the overall rate of reaction.

Full text of DOI:10.1016/S0040-4020(02)01131-6

Tetrahedron 58 (2002) 9055–9061
Selective catalytic oxidation of cyclohexylbenzene
to cyclohexylbenzene-1-hydroperoxide: a coproduct-free
route to phenol
†
Isabel W. C. E. Arends,a Manickam Sasidharan,a Adolf Ku¨hnle,b Mark Duda,b Carsten Jostb
,
and Roger A. Sheldona
,
*
aLaboratory for Biocatalysis and Organic Chemistry, Department of Biotechnology, Delft University of Technology,
Julianalaan 136, 2628 BL Delft, The Netherlands
¨
CREAVIS Gesellschaft fur Technologie und Innovation mbH, D-45764 Marl, Germany
b
Received 17 May 2002; revised 15 August 2002; accepted 5 September 2002
Abstract—A method is described for the highly selective oxidation of cyclohexylbenzene to cyclohexylbenzene-1-hydroperoxide. In the
presence of 0.5 mol% N-hydroxyphthalimide (NHPI) and 2 mol% of the hydroperoxide product, without solvent, a selectivity of ca. 98% to
the desired product was obtained at 32% conversion. The use of NHPI increases the selectivity for initial H-abstraction from the 1-position,
vs the other positions in the cyclohexyl ring, suppresses byproduct formation via transannular hydrogen abstraction and increases the overall
rate of reaction. q 2002 Elsevier Science Ltd. All rights reserved.
1. Introduction
corresponding tertiary hydroperoxide can be easily con-
verted into phenol and cyclohexanone.3 However, in this
case, the cyclohexanone coproduct can be dehydrogenated
to give a second molecule of phenol,4 thus providing an
overall coproduct-free route to phenol. The complete
process scheme is depicted in Scheme 2. CHB can be
made from benzene via two possible routes (Scheme 2):
(a) selective hydrogenation of benzene to cyclohexene over
a ruthenium catalyst5 followed by Friedel–Crafts alkylation
of a second molecule of benzene or (b) oxidative coupling
of benzene to biphenyl6 followed by selective hydrogen-
ation of the latter.7 Route (a) requires no net consumption of
hydrogen and is 100% atom efficient.8 Route (b), on the
other hand, involves the consumption of one equivalent of
hydrogen and the formation of one equivalent of water.
Recently a palladium based catalytic system was described,
The production of phenol via the Hock process is a well-
established commercial process that accounts for .1
million tons of phenol production annually on a world-
wide basis.1,2 Cumene, which is prepared by reaction of
benzene with propylene, is oxidized with molecular oxygen
to give the corresponding hydroperoxide (Scheme 1). Acid-
catalyzed decomposition of the latter affords one equivalent
of both phenol and acetone. Herein lies the disadvantage of
the Hock-process: coproduction of acetone. A coproduct-
free route to phenol would, therefore, be an economically
attractive alternative.
One possible alternative involves the use of cyclohexyl-
benzene (CHB) (see Scheme 1). Analogous to cumene, the
Scheme 1. Production of phenol via cumene hydroperoxide and cyclohexylbenzene-1-hydroperoxide, respectively.
Keywords: autoxidation; cyclohexylbenzene; phenol; N-hydroxyphthalimide.
*
†
Corresponding author. Tel.: þ31-15-278-2683; fax: þ31-15-278-1415; e-mail: r.a.sheldon@tnw.tudelft.nl
Present address: Catalysis Laboratory—Material Science and Chemistry Engineering, Yokohama National University, 79-5 Tokiwadai, Hodogaya-ku,
Yokohama 240-8501, Japan.
0040–4020/02/$ - see front matter q 2002 Elsevier Science Ltd. All rights reserved.
PII: S0040-4020(02)01131-6

9056
I. W. C. E. Arends et al. / Tetrahedron 58 (2002) 9055–9061
Scheme 2. Process scheme for the production of phenol.
which performs this oxidative coupling of benzene at 1058C
under moderate oxygen pressure.6
effect was ascribed to the replacement of the chain carrying
alkylperoxy radicals by bromine atoms. Bromine atoms
exhibit higher selectivity and retard the termination process
thereby accelerating the rate of oxidation.
Compared to cumene, which has only one reactive tertiary
C–H position to be oxidized, cyclohexylbenzene has 1
tertiary position and 10 secondary C–H positions.9 Earlier
reports10 from patents indicated, however, that in the
presence of azo-initiators air oxidation of cyclohexyl-
benzene resulted in a selectivity for cyclohexylbenzene-1-
hydroperoxide (CHBHP) as high as 90% albeit at low
conversion levels of 3–5%. This encouraged us to embark
on the study of the selective autoxidation of cyclohexyl-
benzene. In order to have practical utility the autoxidation
of CHB should afford CHBHP in .90% selectivity at
conversions of ca. 30%.
Recently Ishii has described the selective oxidation of
a large variety of substrates by the combination of
N-hydroxyphthalimide (NHPI) and metal salts.14 Under
these conditions, alcohols and ketones are the main
products. The use of NHPI/metal resulted in high yields
and selectivities under mild conditions. For example, using
NHPI (10 mol%) and Co(OAc)2 (0.5 mol%) the oxidation
of toluene in acetic acid for 20 h afforded benzoic acid
and benzaldehyde in 81% and 3% yield, respectively.15 The
metal salt functions as an initiator for the autoxidation
reaction, but also catalyzes the decomposition of the
intermediate hydroperoxides. Since the metal salt acts
only as an initiator we envisaged that it could be replaced by
organic initiators leading to an NHPI catalyzed oxidation
that would afford the hydroperoxide in high selectivity. The
validity of this idea was given support by a recent
publication of the group of Ishii, which described the
preparation of hydroperoxides in 75% yield, using 10 mol%
NHPI as the catalyst.16 This publication prompted us to
report our results on the selective oxidation of CHB, using
NHPI as a catalyst.17 We will show that when using NHPI
concentrations as low as 0.1 mol%, CHBHP can be prepared
in 97% yield.
The free radical chain mechanism of hydrocarbon autoxi-
dation is well documented.2,11 The susceptibility of
any substrate to autoxidation is determined by the ratio
kp/(2kt)1/2-referred to as the oxidizability12-which deter-
mines the length of the propagating chain and thus the rate
of the reaction (see Scheme 3). A vital role is played by
alkylperoxy radicals, which are the active propagating
species. In order to improve the performance in
autoxidation processes, a large variety of additives have
been demonstrated to exhibit (minor) advantages, among
which HBr/Br2 is probably most well understood.13 This
The role of NHPI is analogous to that of HBr in metal-
catalyzed autoxidation processes. However, HBr could not
be used for our purposes, as it would lead to acid-catalyzed
decomposition of the hydroperoxide product. Although this
is the final goal, in situ formation of phenol would even
inhibit the autoxidation reaction. The proposed reaction
sequence is shown in Scheme 4. NHPI efficiently traps the
intermediate alkylperoxy radicals (reaction 4) thereby
suppressing competing termination. The thus formed
phthalimide N-oxy radical (PINO) abstracts a hydrogen
from the substrate to regenerate NHPI (reaction 2).
Obviously the ratio of propagation to termination and
hence the hydroperoxide selectivity will be directly
Scheme 3. Standard mechanism autoxidation.

I. W. C. E. Arends et al. / Tetrahedron 58 (2002) 9055–9061
9057
Scheme 4. Reaction scheme for the NHPI catalyzed oxidation of cyclohexylbenzene.
dependent on the NHPI concentration. A decrease in NHPI
concentration should lead to a decrease in both rate and
selectivity of RO2H formation.
5% conversion, a selectivity of 87% was achieved, while at
9% conversion the selectivity for the 1-hydroperoxide
decreased to 65%. The main byproducts were the cyclo-
hexylbenzene-4-hydroperoxide (around 8% selectivity) and
the 1,3-disubstituted products A and B (total selectivity
ranging from 4 to 15%, for structures see Scheme 5 and
Experimental). The latter two products arise from trans-
annular abstraction of hydrogen from the 3 position of
the cyclohexane ring in the intermediate alkylperoxy
radical 1 as depicted in Scheme 5. Mass spectral data (see
Experimental) are in accordance with the assigned struc-
tures of A and B. Furthermore, molecular mechanics
calculations showed that for the 1-phenylcyclohexylperoxy
radical (1) the interatomic distance between the distal
2. Results and discussion
Oxidation reactions were performed using neat CHB with
1 bar of pure oxygen. In initial experiments we investigated
the oxidation of CHB in the presence of 1.5–2 mol% of tert-
butyl hydroperoxide (TBHP) or CHBHP and commercially
available azo compounds as initiators at different tempera-
tures. The results are given in Table 1. The selectivity
decreased rapidly with increasing conversion. At 1008C and
Table 1. Autoxidation of cyclohexylbenzene using azo-initiators and hydroperoxide
Initiator
Temperature (8C)
Conv.% CHB
Selectivity products (%)a
Ratiob attack at 1 vs 4-position
1-ROOH
2-ROOH
4-ROOH
A
B
AMDNc
AMDNc
ADVNd
ACCNe,f
AIBNg
ACCNe
Noh
40
50
60
95
95
95
100
110
120
0.96
1.9
3.1
5.1
6.0
9.5
3.2
13
94.8
94.0
87.5
86.7
76.2
65.0
86.0
78.7
72.2
–
–
–
–
–
–
0.9
0.5
0.5
2.4
2.5
4.0
8.0
8.3
8.0
7.4
9.4
9.1
0.50
0.65
1.8
10
6.8
10
5.1
7.7
12
–
–
40
39
23
10
9.2
10
11
8.8
8.8
1.0
5.0
3.6
5.0
0.7
1.2
1.0
Noh
Noh
20
Conditions: no solvent, 60 mmol CHB, 0.19 mmol initiator, 0.9 mmol TBHP, 8 h reaction time, 1 atm O2, hydroperoxide determined by GC as well as
iodometric titration.
a
Selectivity of products formed, missing balance is accounted for by several unidentified products in low quantities.
Ratio of initial attack at 1-position vs attack at 4-position. Calculated as {[1-ROOH]þ[A]þ[B]}/{[2-ROOHþ[4-ROOH]}.
b
c
Initiator is¼2,20-azobis(4-methoxy,2,4-dimethylvaleronitrile)¼AMDN, t1/2¼150 min at 608C and 37 min at 508C.
d
Initiator is 2,20-azobis(2,4-dimethyl valeronitrile)¼ADVN, t1/2¼200 min at 608C.
e
1,10-azobis(cyclohexane carbonitrile) (ACCN) was used as initiator, t1/2¼150 min at 1008C.
f
No TBHP was added.
g
2,20-azobis(2-methylpropionitrile) (AIBN) was used as initiator, t1/2¼10 min at 1008C.
h
No azo-initiator and 1.2 mmol CHBHP was used.

9058
I. W. C. E. Arends et al. / Tetrahedron 58 (2002) 9055–9061
Scheme 5. Formation of products in the radical oxidation of cyclohexylbenzene.
peroxy oxygen and the H-atom at the 3-position is
selectivity as high as 96% at 29% conversion, which
remained 94% at 37% conversion. In the presence of NHPI
but without extra CHBHP (entries 1–3, Table 2) the
selectivity for oxidation at the 1-position was improved but
the transannular hydrogen abstraction was still significant.
Apparently the presence of high concentrations of CHBHP
in the beginning of the reaction instead of azo-initiator
is essential to prevent the latter rearrangement. Thus, only
in the presence of both NHPI and CHBHP, was the
selectivity for oxidation at the 1-position improved and
the rearrangement of the intermediate alkylperoxy radical
suppressed.
˚
substantially shorter (2.4 A) than for the 2- (3.0 A) or the
˚
˚
4-position (4.0 A). We conclude, therefore, that trans-
annular hydrogen abstraction in 1 occurs from the
3-position, resulting in the formation of A and B.
The other byproducts—the 2- and 4-hydroperoxides were
identified by comparison with authentic samples of the
corresponding alcohols (after reduction with triphenyl-
phosphine). Hence, lower selectivities result from two
competing processes: oxidation at the 1 vs other positions in
the cyclohexane ring and transannular hydrogen abstraction
in the intermediate alkylperoxy radical. Competition from
both processes increased with increasing temperature.
A possible explanation for this effect may be that CHBHP
undergoes homolytic decomposition at an optimum rate
under these conditions (100–1108C) during the early stages
of reaction. It is well known2 that in order to achieve smooth
autoxidations the rate of initiation should have an optimum
value. If the rate is too low, little or no reaction occurs while
if it is too fast the selectivity decreases owing to higher
The oxidations were also performed in the presence of
1 mol% NHPI. The results are shown in Table 2. Our first
attempts using a combination of azo-initiator and NHPI
did not give satisfactory results. However, the combination
of CHBHP and NHPI between 100 and 1108C gave a
Table 2. Oxidation of cyclohexylbenzene catalyzed by NHPI
Entry
Temperature (8C)
Initiator
Conv.% CHB
Selectivity products (%)a
Ratiob attack at 1 vs 4-position
1-ROOH
2-ROOH
4-ROOH
A
B
1
2
3
4
5
6
60
95
95
95
100
110
ADVNc
AIBNd
ACCNe
CHBHP
CHBHP
CHBHP
2.3
9.5
17
8.4
29
96.9
65.2
58.1
98.0
96.2
93.9
–
1.1
2.9
0.8
1.8
2.1
2.5
1.3
13
9.8
–
1.6
2.7
0.7
8.0
12
–
0.1
0.2
90
25
44
49
47
36
0.4
1.0
0.2
–
37
0.2
Conditions: no solvent, 60 mmol CHB, 0.6 mmol NHPI (1 mol%), 0.03 mmol azo-initiator (0.05%) or 1.2 mmol CHBHP (2 mol%), 8 h reaction time,1 atm
O2, internal standard used is naphthalene; hydroperoxide determined by GC as well as iodometric titration.
a
Selectivity of products formed, missing balance is accounted for by several unidentified products in low quantities.
Ratio of initial attack at 1-position vs attack at 4-position. Calculated as {[1-ROOH]þ[A]þ[B]}/{[2-ROOHþ[4-ROOH]}.
b
c
See footnote d Table 1.
See footnote g Table 1.
See footnote e Table 1.
d
e

I. W. C. E. Arends et al. / Tetrahedron 58 (2002) 9055–9061
9059
Table 3. Oxidation of cyclohexylbenzene at 1008C using CHBHP as initiator: dependence on NHPI concentration
Entry
Ratio CHB/NHPI
Conv.% CHB
Selectivity products (%)a
1-ROOH
2-ROOH
4-ROOH
A
1
2
3
4
5
No NHPI
2000
1000
200
3.2
14
19
32
29
86.0
94.1
96.7
97.6
96.2
0.9
0.3
0.1
–
6.0
4.0
3.2
1.2
2.1
2.9
1.7
–
0.4
1.6
100
–
Conditions: no solvent, 60 mmol CHB, 0.6 mmol NHPI (1 mol%), 1.2 mmol CHBHP (2 mol%), 8 h reaction time, 1 atm O2, internal standard used is
naphthalene; hydroperoxide determined by GC as well as iodometric titration.
a
See note a, Table 2. Yield of product B, was ,0.5% in all cases.
concentrations of alkylperoxy radicals resulting in more
termination vs propagation (which can also lead to a
decrease in overall rate).
3. Conclusions
We have developed a highly efficient method for the
selective autoxidation of cyclohexylbenzene to cyclo-
hexylbenzene-1-hydroperoxide using NHPI as the catalyst
and the CHBHP product as the initiator. Commercially
attractive results (ca. 98% selectivity at 20–30% conver-
sion) were obtained with CHB/NHPI molar ratios as high
as 200–1000. These results provide the basis for a new
coproduct-free route to phenol. We are currently investi-
gating the use of this methodology for the synthesis of other
industrially useful alkyl hydroperoxides and will report our
results in due course.
The results of studies of selectivity and conversion as a
function of NHPI concentration are presented in Table 3.
Without NHPI the conversion after 8 h was as low as 3%,
with 86% selectivity for cyclohexylbenzene-1-hydro-
peroxide. However, the presence of even 0.05% NHPI
was sufficient to increase conversion to 13%, with 94%
selectivity for CHBHP. The optimum result (32% CHB
conversion and 97.6% selectivity to CHBHP) was achieved
using 0.5% of NHPI and 2 mol% of the CHBHP product as
the initiator. The use of 1 mol% NHPI did not lead to further
improvements. This is most likely due to the limited
solubility of NHPI in cyclohexylbenzene, which is close to
0.5 mol% at 1008C.18 Fig. 1 summarizes all the results
obtained, by depicting the selectivity obtained as a function
of conversion. It shows that only in the presence of both
NHPI and CHBHP does the selectivity remain constantly
high with conversions of up to 37%. These results clearly
demonstrate the validity of our idea that NHPI should
increase both the rate and selectivity of alkyl hydroperoxide
formation by increasing the rate of propagation and/or
decreasing the rate of termination.
4. Experimental
4.1. Safety caution
The production of hydroperoxides should always be
performed with caution. In our case with 100% molecular
oxygen, we worked above the explosion limit, and
monitored the oxygen uptake by a burette, making sure
that the conversion stayed below 40%. Laboratory glass-
ware was used behind safety screens, and the scale was
Figure 1. Conversion vs selectivity plot for all data reported in Tables 1–3.

9060
I. W. C. E. Arends et al. / Tetrahedron 58 (2002) 9055–9061
limited to 10 ml solutions. Hydroperoxides were stored
at 48C as dilute solutions, with the corresponding alkyl-
benzenes as solvent.
hexylbenzene and phenylcyclohexanol were 1.03 and 1.09
with respect to naphthalene, respectively.
The structures of products A and B (after reduction with
triphenylphosphine) were assigned qualitatively on the basis
of GC/MS data. Product B (rt was 18.2 min in the gas
chromatogram, see above) showed a molecular ion peak at
Mz 190 (1% relative intensity). The NIST MS database gave
a good match for the spectrum of 3-hydroxy-3-phenyl-
cyclohexanone (CAS: 25444-79-5; NIST entry 8202).
Major peaks: Mz 51 (19.5%); Mz 77 (55%); Mz 105
(100%); Mz 106 (9%); Mz 120 (38%); Mz 133 (6.5%); 146
(6%); Mz 172 (5%). The product A (rt was 19.3 in the in the
GC) gave a MI peak of 192 (2%) and as major peaks Mz 51
(16%); Mz 77 (53%); Mz 105 (100%); Mz 106 (11%); 120
(64%); 133 (4.5%); 174 (8.5%). Based on the fact that this
fragmentation pattern is very close to that of product B
(Mþ2), we assigned the spectrum of B to 3-hydroxy-3-
phenylcyclohexanol (although this could not be verified due
to the absence of this compound in the NIST database). For
qualitative analysis of products A and B a Rf of 1.0 with
respect to naphthalene was used.
4.2. Materials
AMDN (2,20-azobis(4-methoxy-2,4-dimethylvaleronitrile)
and ADVN (2,20-azobis(2,4-dimethyl valeronitrile) were
purchased from WACO, Pure Chemical Industries Ltd,
Japan (trade names V-65 and V-70, respectively). Cyclo-
hexylbenzene-1-hydroperoxide was prepared from the
corresponding alcohol using H2O2 and H2SO4, adopting a
similar reported procedure.19 All other chemicals were
from commercial sources. Cyclohexylbenzene was distilled
before use.
4.3. Typical experimental procedure
(For variations in amount of reagents etc. see footnotes in
tables.) Reactions were performed in a stirred three-necked
flask equipped with a condenser. A typical reaction mixture
consisted of 60 mmol cyclohexylbenzene (10.1 ml),
0.6 mmol of NHPI, 1.2 mmol of the cyclohexylbenzene-1-
hydroperoxide, and 4 mmol of naphthalene as internal
standard (quantities for specific experiments can be found in
the tables). 1 atm of oxygen was applied by first purging the
reaction mixture and protecting the reaction atmosphere
with a gas-burette filled with water. In this way the oxygen
uptake could be followed in time. After this the reaction was
brought to the reaction temperature and run for 8 h. Samples
were taken during and after the reaction. Samples were
cooled down, diluted with dichloromethane and enough
triphenylphosphine was added to reduce all the alkyl-
hydroperoxides present in the liquid to the corresponding
alcohols. Analysis followed with the GC. Iodometric
titration was performed to determine the amount of
hydroperoxide present in the reaction mixture. The amounts
of hydroperoxides given in Tables 1–3 correspond to the
total amount of the alcohols produced after reduction.
A close correspondence of the amount of hydroperoxide,
determined by iodometric titration, with the amount of
alcohols determined by GC, after reduction with triphenyl-
phosphine, was taken as evidence that only minor amounts
of alcohols are formed as primary products in reactions in
the presence of NHPI.
Given the fact that products A and B were analyzed after
reduction with triphenylphosphine, and that they were most
likely formed via intramolecular H-abstraction (see text)
in the 1-phenyl-cyclohexylperoxy radical (see structure 1 in
Scheme 5), their structure corresponds to 1,3-dihydro-
peroxy-1-phenylcyclohexane (A) and 1-hydroperoxy-1-
phenyl-3-cyclohexanone (B). To rationalize the formation
and substitution pattern (1,3 vs e.g. 1,2 or 1,4) of products A
and B, we performed a molecular mechanics calculation
using HyperChemw 7 (see also text).
Acknowledgements
The authors thank CREAVIS for financial support.
References
1. Hock, H.; Lang, B. Chem. Ber. 1944, 77, 257–264.
2. Sheldon, R. A.; Kochi, J. K. Metal-catalyzed Oxidations of
Organic Compounds. Academic: New York, 1981.
4.4. Analysis
3. Using a non-optimized procedure we were able to decompose
1-phenyl-1-cyclohexyl hydroperoxide completely in 3 h in
refluxing acetone, using 10 wt% of K10 montmorillinite
(a solid acid) to phenol and cyclohexanone (1:1).
Products were analyzed after reduction with triphenyl-
phosphine by GC, using naphthalene as an internal standard
and a CP-Sil-5 CB (50 m£0.53 mm) column. Qualitative
and quantitative analysis of 1-phenylcyclohexanol, 2-phenyl-
cyclohexanol and 4-phenylcyclohexanol (obtained by
reduction of 1-, 2- and 4-substituted cyclohexylbenzene
hydroperoxide) was achieved by comparison with authentic
samples. 1- and 2-phenylcyclohexanol were obtained from
commercial sources, while 4-phenylcyclohexanol was
synthesized by catalytic reduction of 4-phenylcyclohexa-
none. The retention times of cyclohexylbenzene, 1-, 2- and
4-phenylcyclohexanol were 13.7, 16.0, 16.7 and 16.9 min,
respectively (temp. program: init. temp 1008C. for 5 min,
followed by 58/min to 2208C). The molar responses Rf
(defined as (Asubstr/AIS)/(Csubstr/CIS)) determined for cyclo-
4. The dehydrogenation of cyclohexanone to phenol is a standard
process, see e.g. Jordan, W.; Barneveld, H.; van Gerlich, O.;
Kleine-Boymann, M.; Ullrich, J. Ullmann’s Encyclopedia of
Industrial Chemistry, VCH: Weinheim, 1991; Vol. A19. p
307.
5. (a) Struijk, J.; Scholten, J. J. F. Appl. Catal. A 1992, 82,
277–287. (b) Augustine, R. L. Heterogeneous Catalysis for
the Synthetic Chemist. Marcel Dekker: New York, 1996; p 405.
6. Mukhopadhyay, S.; Rothenberg, G.; Lando, G.; Agbaria, K.;
Kazanci, M.; Sasson, Y. Adv. Synth. Catal. 2001, 343, 455–458.
7. (a) Rylander, P. N. Catalytic Hydrogenation in Organic
Syntheses. Academic: New York, 1979; p 179. (b) Steele, D. R.
U.S. Patent 3,387,048 to Engelhard Ind., 1968.

I. W. C. E. Arends et al. / Tetrahedron 58 (2002) 9055–9061
9061
8. Sheldon, R. A. CHEMTECH 1994, 24, 38–47. 12. Howard, J. A. Adv. Free-Radical Chem. 1972, 4, 55–173.
9. As an approximation, one could consider the rate constants for
chain propagation kp for cyclohexane and cumene (which
should closely resemble the reactivity of ring hydrogens vs the
tertiary benzylic C–H position in cyclohexylbenzene) with
13. Walling, C. In Active Oxygen in Chemistry. Foote, C. S.,
Valentine, J. S., Greenberg, A., Liebman, J. L., Eds.; Blackie
Acad.&Professional: London, 1995; pp 24–65.
14. Ishii, Y.; Sakaguchi, S.; Iwahama, T. Adv. Synth. Catal. 2001,
343, 393–427, and refs. cited herein.
tert-butyl peroxy which are 0.00028 and 0.22 M21 s21
,
respectively at 308C (ref: Howard, J. A. Free Radicals,
Kochi, J. K., Ed.; Wiley: New York, 1973; Vol. 2. Chapter 12.
This would correspond to a ratio for attack at the 1 vs the other
(total of 10) positions in the cyclohexyl ring of 79:1 at 308C,
and 19:1 at 1108C (assuming the difference in kp is accounted
for by the difference in Ea). These ratios are in the same order
of magnitude (regarding the number of assumptions made) as
the actual values observed in Table 1, namely a ratio of 40:1 at
408C, and 8.8:1 at 1108C.
15. Yoshino, Y.; Hayashi, T.; Iwahama, T.; Sakaguchi, S.; Ishii,
Y. J. Org. Chem. 1997, 62, 6810–6813.
16. Fukuda, O.; Sakaguchi, S.; Ishii, Y. Adv. Synth. Catal. 2001,
343, 809–813.
17. (a) Ku¨hnle, A.; Duda, M.; Tanger, U.; Sheldon, R. A.;
Manickham, S.; Arends, I. W. C. E. DE 10,015,874 2001.
(b) Ku¨hnle, A.; Duda, M.; Tanger, U.; Sheldon, R. A.; Arends,
I. W. C. E.; Manickam, S. WO 01/74767 A1, 2001.
18. Visual observation. After around 30 min at reaction tempera-
ture up to 0.5 mol% of NHPI in cyclohexylbenzene, gradually
all the NHPI solubilizes, whereas at 1 mol% this is no longer
the case.
10. Dai, S.-H. A.; Lin, C.-Y.; Stuber, F. A. U.S. Patent 4,282,383,
1981, to Upjohn Company.
11. (a) Reich, L.; Stivala, S. S. Autoxidation of Hydrocarbons and
Polyolefins. Dekker: New York, 1969. (b) Emmanuel, N. M.;
Denisov, E. T.; Maizus, Z. K. In Liquid Phase Oxidation
of Hydrocarbons. Hazzard, B. J., Ed.; Plenum: New York,
1967.
19. Davies, A. G.; Foster, R. V.; White, A. M. J. Chem. Soc. 1953,
1541–1547.