Aerobic Oxidation of Benzyl Alcohols Catalyzed by Aryl Substituted N-Hydroxyphthalimides. Possible Involvement of a Charge-Transfer Complex

Article abstract of DOI:10.1021/jo049887y

A series of aryl-substituted N-hydroxyphthalimides (X-NHPIs) containing either electron-withdrawing groups (4-CH₃OCO, 3-F) or electron-donating groups (4-CH₃, 4-CH₈O, 3-CH ₃O, 3,6-(CH₃O)2) have been used as catalysts in the aerobic oxidation of primary and secondary benzylic alcohols. The selective formation of aromatic aldehydes was observed in the oxidation of primary alcohols; aromatic ketones were the exclusive products in the oxidation of secondary alcohols. O-H bond dissociation enthalpies (BDEs) of X-NHPIs have been determined by using the EPR radical equilibration technique. BDEs increase with increasing the electron-withdrawing properties of the aryl substituent. Kinetic isotope effect studies and the increase of the substrate oxidation rate by increasing the electron-withdrawing power of the NHPI aryl substituent indicate a rate-determining benzylic hydrogen atom transfer (HAT) from the alcohol to the aryl-substituted phthalimide-N-oxyl radical (X-PINO). Besides enthalpic effects, polar effects also play a role in the HAT process, as shown by the negative ρ values of the Hammett correlation with σ ⁺ and by the decrease of the ρ values (from -0.54 to -0.70) by increasing the electron-withdrawing properties of the NHPI aryl substituent. The relative reactivity of 3-CH₃O-C₆H₄CH ₂OH and 3,4-(CH₃O)2-C₆H₃CH ₂OH, which is higher than expected on the basis of the σ ⁺ values, the small values of relative reactivity of primary vs secondary benzylic alcohols, and the decrease of the ρ values by increasing the electron-withdrawing properties of the NHPI aryl substituent, suggest that the HAT process takes place inside a charge-transfer (CT) complex formed by the X-PINO and the benzylic alcohol.

Full text of DOI:10.1021/jo049887y

Aer obic Oxid a tion of Ben zyl Alcoh ols Ca ta lyzed by Ar yl
Su bstitu ted N-Hyd r oxyp h th a lim id es. P ossible In volvem en t of a
Ch a r ge-Tr a n sfer Com p lex
†
†
,†
‡
Claudia Annunziatini, Maria Francesca Gerini, Osvaldo Lanzalunga,* and Marco Lucarini
Dipartimento di Chimica, Universit a` degli Studi di Roma “La Sapienza” and Istituto CNR di
Metodologie Chimiche (IMC-CNR), Sezione Meccanismi di Reazione, P.le A. Moro, 5 I-00185 Rome, Italy,
and Dipartimento di Chimica Organica “A. Mangini”, Universit a` di Bologna,
via S. Donato 15, I-40127 Bologna, Italy
osvaldo.lanzalunga@uniroma1.it
Received J anuary 20, 2004
A series of aryl-substituted N-hydroxyphthalimides (X-NHPIs) containing either electron-withdraw-
ing groups (4-CH OCO, 3-F) or electron-donating groups (4-CH , 4-CH O, 3-CH O, 3,6-(CH O) )
3 3 3 3 3 2
have been used as catalysts in the aerobic oxidation of primary and secondary benzylic alcohols.
The selective formation of aromatic aldehydes was observed in the oxidation of primary alcohols;
aromatic ketones were the exclusive products in the oxidation of secondary alcohols. O-H bond
dissociation enthalpies (BDEs) of X-NHPIs have been determined by using the EPR radical
equilibration technique. BDEs increase with increasing the electron-withdrawing properties of the
aryl substituent. Kinetic isotope effect studies and the increase of the substrate oxidation rate by
increasing the electron-withdrawing power of the NHPI aryl substituent indicate a rate-determining
benzylic hydrogen atom transfer (HAT) from the alcohol to the aryl-substituted phthalimide-N-
oxyl radical (X-PINO). Besides enthalpic effects, polar effects also play a role in the HAT process,
+
as shown by the negative F values of the Hammett correlation with σ and by the decrease of the
F values (from -0.54 to -0.70) by increasing the electron-withdrawing properties of the NHPI aryl
3 6 4 2 3 2 6 3 2
substituent. The relative reactivity of 3-CH O-C H CH OH and 3,4-(CH O) -C H CH OH, which
+
is higher than expected on the basis of the σ values, the small values of relative reactivity of
primary vs secondary benzylic alcohols, and the decrease of the F values by increasing the electron-
withdrawing properties of the NHPI aryl substituent, suggest that the HAT process takes place
inside a charge-transfer (CT) complex formed by the X-PINO and the benzylic alcohol.
In tr od u ction
organic compounds, under mild conditions at moderate
oxygen pressure and temperature.4 Among the catalytic
-6
Catalytic aerobic oxidation of organic compounds is of
fundamental importance in the industrial synthesis of a
large variety of oxy-functionalized chemicals from both
2
processes promoted by the NHPI/O /Co(II) system, the
oxidation of alcohols to carbonyl compounds is of par-
ticular synthetic importance since most of the reagents
which perform this transformation are required in
amounts of 1 equiv or more; moreover some of them are
1
economic and environmental points of view. Therefore,
the search for an efficient system for the catalytic
oxidation of organic substrates with molecular oxygen
under mild conditions remains an important challenge.
Recently, particular attention has been given to the use
of N-hydroxyphthalimide (NHPI) in combination with
molecular oxygen and metal salt cocatalysts, like Co-
7
hazardous or toxic. Oxidation of primary aliphatic al-
2
cohols with the NHPI/O /Co(II) system leads to the
corresponding carboxylic acids, while that of secondary
6,8
alcohols leads to the corresponding ketones. Oxidation
of benzyl alcohol by the Ishii system in PhCN at 100 °C
(
OAc)
2 2
or Co(acac) , an oxidizing system developed by
was reported to give a mixture of benzoic acid and
2
,3
Ishii and his group. The NHPI/O
2
/Co(II) system is able
benzaldehyde.2 More recently, however, the selective
to efficiently catalyze the oxidation of a large variety of
(
4) Ishii, Y.; Iwahama, T.; Sakaguchi, S.; Nakayama, K.; Nishiyama,
†
Universit a` degli Studi di Roma “La Sapienza” and Istituto CNR
Y. J . Org. Chem. 1996, 61, 4520.
di Metodologie Chimiche (IMC-CNR).
(5) Yoshino, Y.; Hayashi, Y.; Iwahama, T.; Sakaguchi, S.; Ishii, Y.
J . Org. Chem. 1997, 62, 6810.
‡
Universit a` di Bologna.
(
1) Dioxygen activation and homogeneous catalytic oxidation; Si-
(6) Iwahama, T.; Yoshima, Y.; Keitoku, T.; Sakaguchi, S.; Ishii, Y.
J . Org. Chem. 2000, 65, 6502.
mandi, L. L., Ed.; Elsevier: Amsterdam, 1991. The activation of
dioxygen and homogeneous catalytic oxidation; Barton, D. H. R.,
Martell, A. E., Sawyer, D. T., Eds.; Plenum: New York, 1993.
(7) Highlet, L. J .; Wildman, W. C. J . Am. Chem. Soc. 1955, 77, 4399.
Lee, D. G.; Spitzer, U. A. J . Org. Chem. 1970, 35, 3589. Stevens, R.
V.; Chapman, K. T.; Weller, H. N. J . Org. Chem. 1980, 45, 2030.
Menger, F. M.; Lee, C. Tetrahedron Lett. 1981, 22, 1655.
(8) Iwahama, T.; Sakaguchi, S.; Nishiyama, Y.; Ishii, Y. Tetrahedron
Lett. 1995, 36, 6923.
(
2) Ishii, Y.; Nakayama, K.; Takeno, M.; Sakaguchi, S.; Iwahama,
T.; Nishiyama, Y. J . Org. Chem. 1995, 60, 3934.
3) Ishii, Y.; Sakaguchi, S.; Iwahama, T. Adv. Synth. Catal. 2001,
(
3
43, 393.
1
0.1021/jo049887y CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/20/2004
J . Org. Chem. 2004, 69, 3431-3438
3431

Annunziatini et al.
SCHEME 1
TABLE 1. O-H BDEs of Ar yl-Su bstitu ted NHP Is (1a -g)
a n d Hyp er fin e Sp littin g Con sta n ts of th e Cor r esp on d in g
N-Oxyl Ra d ica ls Mea su r ed a t -10 °C in CH3CN
syntheses of aromatic aldehydes by oxidation of benzylic
alcohols, without appreciable formation of carboxylic
acids, using the NHPI/O /Co(OAc) system has been
2 2
a
-1
catalyst
aN/G
aother/G
BDE /kcal mol
accomplished at room temperature and atmospheric
CN.9
,10
1a
4.55
4.70
4.76
4.84
4.94
4.91
5.01
0.35 (1H)
0.25 (2H), 0.25 (F)
0.50 (2H)
0.56 (4H)
0.83 (1H), 0.14 (4H)
-
-
88.9
88.6
88.1
88.2
87.3
87.9
87.1
pressure in CH
3
The fundamental steps of the
1
1
1
b
c
d
proposed oxidation mechanism are reported in Scheme
9
-11
1
.
Formation, assisted by the Co(II) salt, of the
phthalimide-N-oxyl radical (PINO) (path a) initiates a
free-radical chain. Abstraction of a benzylic hydrogen
atom from the substrate by PINO (path b) leads to an
R-hydroxyl benzyl radical which then reacts with dioxy-
gen to give an R-hydroxyl peroxyl radical (path c). The
latter species abstracts an hydrogen atom from NHPI
giving an hydroperoxide, precursor of the carbonyl prod-
uct, and regenerating PINO (path d).
1e
1
1
f
g
a
Estimated errors are (0.2 kcal mol-1
.
3 3 3 3 2
donating groups (4-CH , 4-CH O, 3-CH O, 3,6-(CH O) ).
The use of NHPI in the catalytic aerobic oxidation
offers the possibility of modifying the catalyst structure
by the introduction of aryl substituents. Such a structural
modification might influence the catalytic efficiency, as
already observed in the oxidation of ethylbenzene to
acetophenone12 and p-xylene to terephthalic acid, the
reaction selectivity as well as the oxidation mechanism.
We felt it is of interest to address this problem, and to
this purpose we carried out a kinetic and product study
of the aerobic oxidation of primary and secondary ben-
zylic alcohols catalyzed by aryl-substituted N-hydroxy-
phthalimides. For a better interpretation of the results
we also investigated the effect of the aryl ring substituent
on the NHPI O-H bond dissociation energy (BDE) by
using the EPR technique. The aryl-substituted N-hy-
droxyphthalimides used in this study (1a -g) contained
13
Resu lts
3
either electron-withdrawing (4-CH OCO, 3-F) or electron-
EP R Mea su r em en ts. The nitroxide radicals were
generated in CH
light in the presence of about 10% of di-tert-butyl
peroxide. An examination of the measured nitrogen
N
hyperfine splitting constants, a , reported in Table 1,
3
CN at -10 °C by photolysis with UV
(
9) Minisci, F.; Punta, C.; Recupero, F.; Fontana, F.; Pedulli, G. F.
Chem. Commun. 2002, 688.
10) Minisci, F.; Recupero, F.; Cecchetto, A.; Gambarotti, C.; Punta,
C.; Faletti, R.; Paganelli, R.; Pedulli, G. F. Eur. J . Org. Chem. 2004,
09.
11) Minisci, F.; Recupero, F.; Pedulli, G. F.; Lucarini M. J . Mol.
Catal. A 2003, 204-205, 63.
12) Wentzel, B. B.; Donners, M. P. J .; Alsters, P. L.; Feiters, M. C.;
Nolte, R. J . M. Tetrahedron 2000, 56, 7797.
13) Saha, B.; Nobuyoshi, K.; Espenson, J . H. J . Phys. Chem. A 2004,
08, 425.
(
1
shows that they increase by decreasing the electron-
withdrawing character of the ring substituent.
(
(
The determination of the BDEs of ring-substituted
NHPIs were done by using the EPR radical equilibration
technique that has been previously used in order to
(
1
3
432 J . Org. Chem., Vol. 69, No. 10, 2004

Aerobic Oxidation of Benzyl Alcohols
TABLE 2. In ter m olecu la r Deu ter iu m Kin etic Isotop e
Effects (kH/kD) in th e Oxid a tion s of Ben zyl Alcoh ol w ith
1
a , 1c, 1e, or 1g/Co(II)/O2 System s
catalyst
kH/kDa
1
1
1
1
a
c
e
g
14.3
15.5
17.7
19.5
a
Molar ratio measured via GC-MS by the ratio between the
corrected signal intensities of the two molecular ions at m/z ) 107
and 106. Average of at least three determinations; the error is
(2%.
SCHEME 2
mM) and 1e (6.2 mM), and its computer simulation (d).
BDEs for the investigated NHPIs are reported in Table
F IGURE 1. EPR spectra obtained in CH CN at -10 °C by
3
1
.
photolysis of 1c (a) and 1e (b) with UV light in the presence
of about 10% of di-tert-butyl peroxide. (c) EPR spectrum
obtained, under the same experimental conditions, by pho-
tolysis of a solution containing 1c (15 mM) and 1e (6.2 mM).
Kin etic Isotop e Effect Stu d ies. Intermolecular ki-
netic isotope effects have been determined in CH CN (3
mL) at 25 °C under oxygen by mixing equimolar amounts
3
2
(d) Computer simulation of spectrum (c).
2
(0.3 mmol) of benzyl alcohol and [1- H ]-benzyl alcohol,
the catalyst 1a , 1c, 1e, or 1g (10 mol %), m-chlorobenzoic
acid (MCBA) (5 mol %), and Co(OAc) (0.5 mol %). k /k
measure the O-H BDE of the unsubstituted NHPI.14
This technique consists of measuring the equilibrium
constant for the hydrogen atom transfer reaction between
the aryl-substituted NHPIs, an appropriate reference
compound, whose BDE is already known, and the corre-
sponding radicals. In the present case, the reference
derivative used was the unsubstituted NHPI (eq 1),
whose BDE (88.1 kcal/mol) has been already measured
in tert-butyl alcohol.14
2
H
D
have been calculated by the molar ratio between the
2
benzaldehyde and [1- H]-benzaldehyde measured via
GC-MS by the ratio between the corrected signal inten-
sities of the two molecular ions at m/z ) 107 and 106.
The results are reported in Table 2.
Effect of NHP I Ar yl Su bstitu en ts on th e Ca ta lytic
Activity. This effect was evaluated in the oxidation of
1-(4-methoxyphenyl)ethanol (2). The oxidations were
carried out in AcOEt/dioxane 1:1 (10 mL)17 at 25 °C under
oxygen by mixing 2 (3 mmol), the catalyst 1a -f (10 mol
2
%), MCBA (5 mol %), and Co(OAc) (0.5 mol %). GC
analysis of 1 mL aliquots of the reaction mixtures taken
at different times revealed that, with all of the X-NHPI,
the oxidation led to the formation of 4-methoxyacetophen-
one as the exclusive reaction product (Scheme 2).
The regular increase of the yields of 4-methoxyac-
etophenone during the first 30 min of the reaction is
shown in Figure 2. The plot is linear, and the initial rates
of formation of 4-methoxyacetophenone have been cal-
culated within 10% of substrate conversion (Table 3).
Com p etitive Oxid a tion s. The effect of the NHPI aryl
substituents on the selectivity and mechanism of the
oxidation of benzylic alcohols has been evaluated by the
analysis of the competitive oxidations of substituted
primary benzylic alcohols and the competitive oxidations
of primary vs secondary benzylic alcohols.
The BDEs for the aryl substituted NHPIs were calcu-
lated, in the assumption that the entropic term can be
neglected,15 by means of eq 2 from Keq and the known
BDE of NHPI.16
BDE(X-NHPI) ) BDE(NHPI) - RTln(K ) (2)
eq
(1) Com p etitive Oxid a tion of Su bstitu ted P r i-
m a r y Ben zylic Alcoh ols. The competitive experiments
As an example, in Figure 1 are reported the EPR
spectra of 1c (a) and 1e (b), the EPR spectrum (c)
obtained by photolysis of a solution containing 1c (15
(15) Lucarini, M.; Pedulli, G. F.; Cipollone, M. J . Org. Chem. 1994,
5
9, 5063. Lucarini, M.; Pedrielli, P.; Pedulli, G. F.; Cabiddu, S.;
Fattuoni, C. J . Org. Chem. 1996, 61, 9259.
16) It has been assumed that the BDE of NHPI is not significantly
altered by passing from tert-butyl alcohol to CH CN.
(
(
14) Amorati, R.; Lucarini, M.; Mugnaini, V.; Pedulli, G. F.; Minisci,
F.; Recupero, F.; Fontana, F.; Astolfi, P.; Greci, L. J . Org. Chem. 2003,
8, 1747.
3
(17) All the catalysts were completely soluble in AcOEt/dioxane 1:1
except for 1g which was thus not used in the kinetic studies.
6
J . Org. Chem, Vol. 69, No. 10, 2004 3433

Annunziatini et al.
SCHEME 3
F IGURE 3. Hammett plot for the competitive oxidation of
F IGURE 2. Formation of 4-methoxyacetophenone in the
oxidation of 1-(4-methoxyphenyl)ethanol (2) by the 1a -f/Co-
(II)/MCBA/O systems.
2
ring substituted benzylic alcohols with 1a /Co(II)/MCBA/O
g/Co(II)/MCBA/O systems (the points for the m-MeO and 3,4-
MeO) groups are excluded from the correlations).
2
and
1
2
(
2
TABLE 3. In itia l Ra te of F or m a tion of
TABLE 4. G Va lu es Ca lcu la ted for th e Ha m m ett
4
1
-Meth oxya cetop h en on e in th e Oxid a tion of
Cor r ela tion in th e Oxid a tion of Rin g-Su bstitu ted
-(4-Meth oxyp h en yl)eth a n ol (2) w ith th e 1a -f/Co(II)/
MCBA/O2 System s
Ben zylic Alcoh ols w ith th e 1a -g/Co(II)/MCBA/O2 System s
catalyst
F
catalyst
initial rates (mmol/min)
1
1
1
1
1
1
1
a
b
c
d
e
f
-0.70
-0.69
-0.68
-0.67
-0.60
-0.60
-0.54
-3
-3
-3
-3
-3
-3
1
1
1
1
1
1
a
b
c
d
e
f
(12 ( 1) × 10
(11 ( 1) × 10
(10 ( 1) × 10
(8.7 ( 0.9) × 10
(6.0 ( 0.6) × 10
(5.1 ( 0.5) × 10
g
3
were carried out in CH CN (3 mL) at 25 °C under oxygen
by mixing equimolar amounts (0.3 mmol) of ring-
substituted benzyl alcohol (3b-k ) and of benzyl alcohol
are excluded from the plots (vide infra). As an example
in Figure 3 are shown the Hammett plots for the
oxidation catalyzed by 4-CH
3
OCO-NHPI (1a ) and 3,6-
(
3a ), the catalyst 1a -g (10 mol %), MCBA (5 mol %), and
(
CH O) -NHPI (1g). The Hammett plots for the oxida-
3
2
Co(OAc)
2
(0.5 mol %) and determining the relative
tions catalyzed by the other aryl-substituted NHPIs are
amounts of the benzaldehydes formed which are the only
reaction products (Scheme 3).
The relative reactivities (k /k ) of the alcohols 3b-k
Y H
with respect to benzyl alcohol (3a ), determined for all the
catalysts 1a -g, are reported in Table S1 (Supporting
Information).
available in the Supporting Information. The F values,
2
0
which range from -0.70 to -0.54, are listed in Table 4.
(
2) Com p etitive Oxid a tion of P r im a r y vs Secon d -
a r y Ben zylic Alcoh ols. The relative reactivity of pri-
mary vs secondary benzylic alcohols was studied with 1c
and 1e as catalysts. The oxidations were carried out in
When the log(k
lyst were plotted against the substituent constants σ ,
good Hammett correlations were obtained (r > 0.98),
provided the points for the 3-MeO and 3,4-(MeO) groups
Y H
/k ) values determined for each cata-
+
CH
lar amounts (0.3 mmol) of benzyl alcohol (3a ) and
-phenylethanol (4a ) or equimolar amounts (0.3 mmol)
3
CN (3 mL) at 25 °C under oxygen by mixing equimo-
2
18
1
2
of 3,4-dimethoxybenzyl alcohol (3b) and 1-(3,4-dimethoxy-
(
18) The correlation with the σ constants was not satisfactory (r2
.79); moreover, no improvement of the correlation is observed on
<
0
(19) Dust, J . M.; Arnold, D. R. J . Am. Chem. Soc. 1983, 105, 1221.
(20) A F value (-0.69) very close to that here reported was
determined by Minisci et al. for the aerobic oxidation of benzylic
+
replacing the single parameter Hammett equation (log k
with the dual parameter equation (log k
by Arnold and his associates.
Y
/k
H
) Fσ )
•
•
+
Y
/k
H
) F σ
R
+ Fσ ) proposed
1
9
10
alcohols by the unsubstituted NHPI.
3
434 J . Org. Chem., Vol. 69, No. 10, 2004

Aerobic Oxidation of Benzyl Alcohols
SCHEME 4
TABLE 5. Rela tive Rea ctivities (k1°/k2°) in th e
Com p etitive Oxid a tion of P r im a r y vs Secon d a r y Ben zylic
Alcoh ols w ith th e 1c- or 1e/Co(II)/MCBA/O2 System s
k1°/k2°a
substrates
1c
1e
3
3
a /4a
b/4b
0.6
1.1
0.5
1.0
a
Relative reactivities statistically corrected for the number of
benzylic hydrogens.
F IGURE 4. Correlation between O-H BDEs of aryl-substi-
tuted NHPIs and nitrogen hyperfine splitting constants, a
SCHEME 5
N
.
contribution of the structure II in Scheme 5 and therefore
a greater stability of the N-oxyl radical and a consequent
smaller O-H BDE.
Oxid a tion Ra tes a n d Rea ction Mech a n ism . The
data reported in Table 2 show that the aerobic oxidation
of benzylic alcohols catalyzed by aryl-substituted NHPIs
is characterized by high kinetic isotope effects (KIE)
which increase from k /k ) 14.3 for 1a to k /k ) 19.5
phenyl)ethanol (4b), the catalyst (10 mol %), MCBA (5
mol %), and Co(OAc) (0.5 mol %). The oxidation led
2
H
D
H
D
for 1g.22 Interestingly, the KIE determined in the com-
selectively to the formation of benzaldehydes from the
primary benzylic alcohols and of acetophenones from the
secondary benzylic alcohols (Scheme 4).
petitive experiment catalyzed by NHPI (k /k ) 15.5) was
H
D
similar to that (16.9) recently measured in a kinetic study
of the hydrogen abstraction by PINO, generated by oxi-
The relative reactivity values (k1°/k2°), statistically
corrected for the number of benzylic hydrogens, are
reported in Table 5.
dation of NHPI with Pb(OAc) , from benzyl alcohol and
4
2
24
[1- H ]-benzyl alcohol in AcOH. This observation is very
2
important as it indicates that, despite the multistep na-
ture of the aerobic oxidations catalyzed by NHPIs (Scheme
1
), the competitive experiments carried out by us provide
Discu ssion
reliable information on the relative reactivity of the
benzylic substrates in the step of hydrogen atom abstrac-
tion by the aryl substituted PINO (Scheme 1, path b).
The data in Table 3 show that the initial rate of the
oxidation of 2 increases by increasing the electron-
withdrawing power of the aryl substituent in NHPI, and
therefore, the reactivity order parallels the increase of
O-H BDE in the substituted NHPIs (Table 1). This is
in line with a rate-determining step involving the ben-
zylic HAT from the alcohol to the aryl-substituted
PINO.25 Thus, a stronger O-H bond in the NHPIs favors
the reaction from the enthalpic point of view. As expected,
since the substituent effects on the BDEs are not very
large, there is not much difference in the initial rates of
oxidation with the various NHPIs, as shown by the fact
3
O-H Bon d Dissocia tion En er gies a n d EP R Ex-
p er im en ts. The results reported in Table 1 indicate that
in the aryl-substituted NHPIs the O-H BDEs regularly
decrease by increasing the electron-donating properties
of the substituent. The effects are, however, small, the
difference between the stronger O-H bond in 4-CH
NHPI (88.9 kcal/mol) and the weaker O-H bond in 3,6-
MeO) -NHPI (87.1 kcal/mol) being only 1.8 kcal/mol.
An examination of the EPR data reported in Table 1
shows also that the nitrogen hyperfine splitting constant,
, increases by increasing the electron-donating char-
3
OCO-
(
2
a
N
acter of the ring substituent. This observation can be
interpreted on the basis of the two canonical forms I and
II of the nitroxide function (Scheme 5).1
0,21
The greater the electron-donating nature of the sub-
stituent the more favored will be the structure II which
bears larger electron density on oxygen and spin density
on nitrogen. As a consequence, the nitrogen hyperfine
that the highest rate observed with 4-CH OCO-NHPI
(22) The increase of KIE on going from 1a to 1g can be explained
by the fact that the maximum magnitude of the KIE should be obtained
2
3
when the HAT reaction is thermoneutral. Since the reaction of the
splitting constant, a
Another interesting notation is that the a
correlate with the BDEs of the NHPIs, as shown in
Figure 4. Namely, the BDE decreases as a increases.
As said before, an increase in a indicates a greater
N
, increases.
23
PINO with benzyl alcohol is exothermic for ca. 7-8 kcal/mol, an
N
values
increase of the KIE should be expected by decreasing the O-H BDE
of the aryl substituted NHPI, i.e., by introducing electron-donating
substituents.
N
(23) Nobuyoshi, K.; Cai, Y.; Espenson, J . H. J . Phys. Chem. A 2003,
1
07, 4262.
24) Such a high isotope effect value was interpreted as due to
N
(
quantum mechanical tunneling taking part in the benzylic HAT from
the alcohol to the PINO. Nobuyoshi, K.; Saha, B.; Espenson, J . H. J .
Org. Chem. 2003, 68, 9364.
(
21) Minisci, F.; Punta, C.; Recupero, F.; Fontana, F.; Pedulli, G. F.
J . Org. Chem. 2002, 67, 2671.
J . Org. Chem, Vol. 69, No. 10, 2004 3435

Annunziatini et al.
SCHEME 6
tions, the reactivity of the substrates with 3-CH
3
O and
3
,4-(CH O) substituents resulted significantly higher
3
2
than expected (see Figure 3 for 1a and 1g and Figures
S1-S5 in the Supporting Information for the catalysts
1
b-f). This result is surprisingly since in reactions
occurring by a HAT mechanism, even though with a
significant polar character, the 3-CH O group always
3
exhibits a little rate-retarding kinetic effect due to the
slight electron-withdrawing inductive effect, which de-
stabilizes the partial positive charge developing in the
transition state.28
To explain the rate-enhancing effect of the 3-CH
3
O and
3
,4-(CH O) groups observed in our reactions, it might
3
2
be suggested that the HAT process takes place inside a
charge-transfer (CT) complex formed by the X-PINO and
2
9
the benzylic alcohol, as illustrated in Scheme 6. Several
examples of hydrogen atom transfer process taking place
inside a CT complex are reported in the literature, such
F IGURE 5.
1.2 × 10 mmol min-1) is about twice the lowest value
-2
(
3
1
-3
as in the photoreduction of aromatic ketones and the
side-chain oxidation of benzyltrimethylsilanes by iodo-
sylbenzene in the presence of iron porphyrin.3 In the CT
complex a partial positive charge resides on the aromatic
ring of the alcohol, and therefore, some stabilization by
a ring methoxy group is expected, whatever its position
with respect to the side chain.3
which is observed with 3-CH
3
O-NHPI (5.1 × 10 mmol
-
1
min ).
2
Besides enthalpic effects, also polar effects must play
a role as shown by the negative F values of the Hammett
correlations for the aerobic oxidation of benzylic alcohols
(3a -k ) by X-NHPI. Thus, the transition state of the HAT
2,33
process can be represented as hybrid of the two canonical
resonance forms a and b (Figure 5). The increase of the
substrate reactivity by increasing the electron-donating
power of its substituent26 is explained by the stabilization
of the partial positive charge on the benzylic carbon atom
in the resonance form b of the transition state.27
Further support for the hypothesis of the formation of
a CT complex comes from the analysis of the competitive
oxidation of primary vs secondary benzylic alcohols. From
the data reported in Table 5, we can observe that the
Even considering the polar effects, the mechanistic
pattern described so far cannot take into account for all
the experimental data. In fact, in the Hammett correla-
(
28) Walling, C.; Rieger, A. L.; Tanner, D. D. J . Am. Chem. Soc.
1963, 85, 3129. Martin, M. M.; Gleicher, G. J . J . Org. Chem. 1963, 28,
266. Gleicher, G. J . J . Org. Chem. 1968, 33, 332.
29) The rate enhancing effect of the 3-CH O and 3,4-(CH
3
(
3
3 2
O) groups
could be also explained by an electron transfer mechanism from the
alcohol to the X-PINO. We have excluded this possibility by the fact
that the oxidation of the mechanistic probe 1-(3,4-dimethoxyphenyl)-
2,2-dimethyl-1-propanol led to the formation of 1-(3,4-dimethoxyphen-
yl)-2,2-dimethyl-1-propanone as the only reaction product and no traces
of 3,4-dimethoxybenzaldehyde have been detected (see the Experi-
(
25) Rate-determining hydrogen atom abstraction from C-H bonds
by PINO (Scheme 1, path b) in aerobic oxidations catalyzed by NHPI
is well documented.3
,9-11
The possibility of a rate-determining forma-
tion of aryl-substituted PINO (Scheme 1, path d) should be considered
1
4
if a higher alcohol/X-NHPI molar ratio is used.
26) Similar substituent effects have been reported for the benzylic
3
0
(
mental Section).
5
12
aerobic oxidation of substituted toluenes and arylacetic esters by
the NHPI/Co(II)/O system.
27) A minor contribution of radical stabilization (resonance form
a ) vs polar effects is suggested by the lack of improvement of the
(30) Baciocchi, E.; Belvedere, S.; Bietti, M.; Lanzalunga, O. Eur. J .
Org. Chem. 1998, 299. Baciocchi, E.; Belvedere, S.; Bietti, M. Tetra-
hedron Lett. 1998, 39, 4711.
(31) Cohen, S. G.; Parola, A.; Parsons, G. H. Chem. Rev. 1973, 73,
141. Coenjarts, C.; Scaiano, J . C. J . Am. Chem. Soc. 2000, 122, 3635.
(32) Baciocchi, E.; Lanzalunga, O. Tetrahedron 1993, 49, 7267.
2
(
•
•
+
correlation using the dual parameter equation (log k
Y
/k
H
) F σ
R
+ Fσ )
proposed by Arnold.1
9
3
436 J . Org. Chem., Vol. 69, No. 10, 2004

Aerobic Oxidation of Benzyl Alcohols
k
1°/k2° relative reactivities of 3,4-dimethoxybenzyl alcohol
and 1-(3,4-dimethoxyphenyl)ethanol with the NHPI/Co-
II)/O and 4-MeO-NHPI/Co(II)/O systems are close to
phenyl)-2,2-dimethyl-1-propanol was prepared according to a
previously described procedure.30 4-Methoxycarbonyl-N-hy-
droxyphthalimide (1a ), 3-fluoro-N-hydroxyphthalimide (1b),
(
2
2
4
-methyl-N-hydroxyphthalimide (1d ), 4-methoxy-N-hydroxy-
unity. This unexpected result is in accordance with the
occurrence of a CT complex since its rate of formation
should be mainly determined by the electronic properties
of the aromatic ring of the two alcohols, which are the
same, and not by the side-chain substitution pattern.
Clearly, the role of a CT complex should be less important
with the unsubstituted benzylic alcohols, and accordingly,
phthalimide (1e), and 3-methoxy-N-hydroxyphthalimide (1f)
12,34
were synthesized as described in the literature.
3,4-
Dimethoxy-N-hydroxyphthalimide (1g) was synthesized by
reacting 3,6-dimethoxyphthalic anhydride with NH OH, in
2
anhydrous pyridine.35 For the synthesis of 3,6-dimethoxy-
phthalic anhydride 1,4-dimethoxy-1,3-cyclohexadiene was re-
acted with dimethyl acetylenedicarboxylate to give dimethyl
3
,6-dimethoxyphthalate36 which was heated for 2 h under
1
-phenylethanol is twice more reactive than benzyl
alcohol.
Finally, the formation of a CT complex can also account
reflux in methanol/water with KOH to give 3,6-dimethoxy-
37
phthalic acid. 3,6-Dimethoxyphthalic anhydride was then
obtained by reaction of 3,6-dimethoxyphthalic acid with acetic
for the variation of F values as a function of the aryl
substituent in the NHPI. The F values regularly decrease
by increasing the electron-withdrawing properties of the
aryl substituent ranging from F ) -0.70 for the most
selective 4-CH
selective catalyst 3,6-(CH
Table 4). Accordingly, electron-withdrawing substituents
in NHPI stabilize the negative charge developing in the
CT complex, thus increasing the degree of charge transfer
from the benzylic alcohol to the X-PINO, as a consequence
the oxidation is more sensitive to the effect of the aryl
substituent of the alcohol.
anhydride.35
Physical properties are identical to those reported
in the literature.38
EP R Mea su r em en ts. The equilibration experiments for
the determinations of the bond dissociation enthalpies (BDEs)
of ring-substituted NHPIs were done by preparing solutions
3
OCO-NHPI to F ) -0.54 for the least
O) -NHPI (see Figure 3 and
of 1c (ca. 0.1 M) and of 1a -1b, 1d -1g (ca. 0.1 M) in CH
containing 10% of di-tert-butyl peroxide as photoinitiator. The
nitroxide radicals were generated in CH CN at -10 °C by
3
CN
3
2
3
photolysis with UV light. Since these experiments were
performed on concentrated solutions of the radical precursors,
the initial concentration of NHPIs could be used in the
calculation of Keq, while the relative concentrations of the
equilibrating radical species were determined by simulation
of the EPR spectra.
Kin etic Isotop e Effect Stu d ies. Intermolecular kinetic
Con clu sion s
isotope effects have been determined at 25 °C under oxygen
2
by mixing 0.3 mmol of benzyl alcohol, 0.3 mmol of [1- H
2
]-
Among the catalytic processes promoted by the NHPI/
O /Co(II) system the oxidation of alcohols to carbonyl
2
compounds is of particular synthetic importance. In this
work, a kinetic study of the oxidation of a benzylic alcohol
2
with the X-NHPI/Co(II)/O systems has shown that it is
possible to increase the catalytic efficiency by introducing
electron-withdrawing substituents in the NHPI aryl ring.
O-H bond dissociation energies (BDEs), determined by
using the EPR technique, increase by increasing the
electron-withdrawing properties of the NHPI substituent.
An interesting correlation between the hyperfine splitting
benzyl alcohol, 0.03 mmol of the catalyst (1a , 1c, 1e, or 1g),
0
3
.015 mmol of MCBA, and 0.0015 mmol of Co(OAc)
2
‚4H
2
O in
mL of CH CN. k /k have been calculated by the molar ratio
3
H
D
2
between the benzaldehyde and [1- H]-benzaldehyde measured
via GC-MS by the ratio between the corrected signal intensi-
ties of the two molecular ions at m/z ) 107 and 106.
Oxid a tion of 1-(4-Meth oxyp h en yl)eth a n ol (2). Solutions
of 3 mmol of 1-(4-methoxyphenyl)ethanol, 0.3 mmol of X-NHPI,
0
.15 mmol of MCBA, and 0.015 mmol of Co(OAc)
mL of a AcOEt/dioxane 1:1 v/v were placed in a Schlenk tube
equipped with a rubber baloon filled with O and vigorously
2
‚4H
2
O in 10
2
stirred at 25 °C. One milliliter aliquots of the reaction mixtures
were taken at different times and, after addition of a solution
of an internal standard (bibenzil) for GC analysis, were washed
constants a and the O-H BDEs has been observed. The
N
parallel increase of O-H BDEs and of the catalytic
efficiency in the presence of electron-withdrawing NHPI
substituents indicates that benzylic HAT from the alcohol
to the aryl substituted PINO is the rate-determining step.
This hypothesis is further supported by kinetic isotope
effects studies. Besides enthalpic effects, polar effects also
play a role in the HAT process, as shown by the negative
2 4
with water, dried over anhydrous Na SO , and analyzed by
GC. The oxidations with all the X-NHPI lead to the formation
of 4-methoxyacetophenone as the only reaction product. The
initial rates of product formation were determined at low
<10%) substrate conversion.
Com p etitive Oxid a tion of Su bstitu ted P r im a r y Ben -
(
zylic Alcoh ols. Solutions of 0.3 mmol of ring-substituted
benzylic alcohols, 0.3 mmol of benzyl alcohol, 0.03 mmol of
+
F values of the Hammett correlation with σ . The higher
than expected relative reactivity of 3-CH
X-NHPI, 0.015 mmol of MCBA, and 0.0015 mmol of Co(OAc)
O in 3 mL of CH CN were placed in a Schlenk tube
equipped with a rubber balloon filled with O and vigorously
2
‚
3
O-C
6
H
4
CH
2
OH
4
H
2
3
and 3,4-(CH O) -C CH
3
2
6
H
3
2
OH, the surprisingly small
2
values of relative reactivity of primary vs secondary
benzylic alcohols and the decrease of the F values by
increasing the electron-withdrawing properties of the
NHPI aryl substituent suggest that the HAT process
takes place inside a charge-transfer (CT) complex formed
by the X-PINO and the benzylic alcohol.
stirred at 25 °C for 2 h. After addition of a solution of an
internal standard (4-methoxyacetophenone) for GC analysis
and water (3 mL), the reaction mixtures were extracted with
2 2 2 4
CH Cl . The organic phase was dried over anhydrous Na SO
and analyzed by GC. The oxidations with all the X-NHPI led
to the formation of the corresponding benzaldehydes as the
only reaction products. A good recovery of materials (> 95%)
was observed in all the experiments.
Exp er im en ta l Section
Su bstr a tes a n d Rea gen ts. The benzylic alcohols and
N-hydroxyphthalimide are commercially available and were
used as received without further purification. 1-(3,4-Dimethoxy-
(34) Gorgy, K.; Lepretre, J .-C.; Saint-Aman, E.; Einhorn, C.; Ein-
horn, J .; Marcadal, C.; Pierre, J .-L. Electrochim. Acta 1998, 44, 385.
(
35) Einhorn, C.; Einhorn, J .; Marcadal-Abbadi, C.; Pierre, J .-L. J .
Org. Chem. 1999, 64, 4542.
36) Harland, P. A.; Hodge, P Synth. Commun. 1982, 223.
(
(
33) Fukuzumi, S.; Kochi, J . K. J . Am. Chem. Soc. 1981, 103, 7240.
Baciocchi, E.; Crescenzi, M.; Fasella, E.; Mattioli, M. J . Org. Chem.
992, 57, 4684.
(37) Corrie, J . E. T. J . Chem. Soc., Perkin Trans. 1 1994, 2981.
(38) Parrick, J .; Ragunathan, R. J . Chem. Soc., Perkin Trans. 1 1993,
211.
1
J . Org. Chem, Vol. 69, No. 10, 2004 3437

Annunziatini et al.
Com p etitive Oxid a tion of P r im a r y vs Secon d a r y Ben -
zylic Alcoh ols. Solutions of 0.3 mmol of unsubstituted or ring-
substituted benzyl alcohol, 0.3 mmol of unsubstituted or ring-
substututed 1-phenylethanol, 0.03 mmol of X-NHPI, 0.015
anhydrous Na
2
SO
4
and analyzed by 1HNMR. The oxidation
led to the formation of 1-(3,4-dimethoxyphenyl)-2,2-dimethyl-
1-propanone as the only reaction product. No traces of 3,4-
dimethoxybenzaldehyde were observed with all the catalysts.
mmol of MCBA, and 0.0015 mmol of Co(OAc)
of CH CN were placed in a Schlenk tube equipped with a
rubber balloon filled with O and vigorously stirred at 25 °C
for 2 h. The reaction was performed in a similar way as
described for the competitive oxidation of substituted primary
benzylic alcohols. The oxidations with all the X-NHPI led
selectively to the formation of benzaldehydes from the primary
benzylic alcohols and of aromatic ketones from the secondary
benzylic alcohols. A good recovery of materials (>95%) was
observed in all the experiments.
2
‚4H
2
O in 3 mL
3
Ack n ow led gm en t . We thank the Ministero dell’
Istruzione, dell’Universit a` e della Ricerca (MIUR), and
the Consiglio Nazionale delle Ricerche (CNR) for finan-
cial support. We also thank Prof. Enrico Baciocchi for
helpful discussions.
2
Su p p or tin g In for m a tion Ava ila ble: Instrumentation.
Determination of relative rates in the competitive oxidation
of substituted primary benzylic alcohols and in the competitive
oxidation of primary vs secondary benzylic alcohols. Relative
Oxid a tion of 1-(3,4-Dim eth oxyp h en yl)-2,2-d im eth yl-1-
p r op a n ol. The oxidation of the mechanistic probe 1-(3,4-
dimethoxyphenyl)-2,2-dimethyl-1-propanol was carried out at
reactivities (k
primary benzylic alcohols with the 1a -g/Co(II)/MCBA/O
systems. Hammett plots for the competitive oxidation of ring
Y H
/k ) in the competitive oxidation of substituted
2
2
5 °C by mixing 0.3 mmol of the substrate, 0.03 mmol of the
catalyst (1a , 1c, or 1g), 0.015 mmol of MCBA, and 0.0015 mmol
of Co(OAc) in 3 mL of AcOEt/dioxane 1:1 v/v in a Schlenk
tube equipped with a rubber balloon filled with O . After 24
h, water (3 mL) was added, and the reaction mixture was
extracted with CH Cl . The organic phase was dried over
substituted benzylic alcohols with the 1b-f/Co(II)/MCBA/O
2
2
systems. This material is available free of charge via the
Internet at http://pubs.acs.org.
2
2
2
J O049887Y
3
438 J . Org. Chem., Vol. 69, No. 10, 2004