reaction stopped at 55% yield in less than 6 h. At 80 °C it stopped
at 33% yield in less than 1 h. These results clearly indicate that
catalysis by NHTPPI is more rapid at higher temperature, but that
better turnover numbers of the catalyst are obtained at lower
temperature.
decomposition of the radical. The absorbance change was mon-
itored at 425 nm (ESI†). Second-order kinetics fitting2c gave
second-order rate constant kd = 0.162 L mol21 s21 at 35 °C. For
[10b]0 = 0.5 mmol L21, the half life (t1/2) of 10b is 205 min.
For a comparison, self-decomposition of 1b (PINO) in the same
conditions of temperature and concentration, monitored at 380 nm
gave kd = 0.777 L mol21 s21 and t1/2 = 43 min (ESI†). So, the half
life of 10b was found to be 4.7 times longer than that of PINO at 35
°C. This result has to be compared with recent results of Espenson:
he showed that PINO radicals substituted either by electron
donating or by electron withdrawing substituents have lower
kinetic stabilities than unsubstituted PINO, accounting for lower
efficiency of the corresponding NHPI catalysts.2e In our case
indeed, enhanced kinetic stability of the radical species is
accompanied by enhanced catalytic activity of NHTPPI.
A series of benzylic compounds has next been oxidized using 1
or 2 mol% NHTPPI and 5 mol% CuCl (Table 2). Indan-1-one or
2-methylindan-1-one have been obtained in 77 and 65% isolated
yields, which represent more than 90% yields based on indan or
2-methylindan reacted (runs 1 and 2). Tetralin was similarly
oxidized to tetralone with 74% isolated yield (run 3). In the same
conditions, the oxidation of indan-1-ol was very slow, yielding only
15% of indan-1-one after 24 h. Therefore, a direct pathway leading
from indan to indan-1-one seem likely (run 4).8 4-Methoxy-
1-ethylbenzene and fluorene gave 60% and 42% conversions after
6 h and 24 h oxidation. An additional 1 mol% NHTPPI raised the
conversions to 79 and 84% and allowed isolation of 70% of
4-methoxyacetophenone and of 79% fluorenone (runs 5 and 6).
Oxidation of a methyl substituent was less efficient: 4-methylanisol
gave 25% of p-anisaldehyde, accompanied by 9% of p-anisic acid
after 48 h reaction (run 7). Selective oxidation of one ethyl group of
1,4-diethylbenzene was observed, resulting in the isolation of 52%
4-ethylacetophenone using 1 mol% NHTPPI (run 8). This results,
clearly indicates that a carbonyl substituent deactivates benzylic
sites to oxidation. Finally, xanthone and isochromanone were
obtained with excellent 96 and 86% isolated yields from xanthene
and isochroman with only 1 mol% NHTPPI (run 9 and 10). In each
case the main oxidation compound was accompanied by only small
amounts (1–3%) of the corresponding alcohol.
To obtain some information on the cause of enhanced catalytic
properties of NHTPPI compared to NHPI, we investigated
generation and self decomposition of the corresponding radical 10b
in acetonitrile at 35 °C in the absence of substrate, measuring the
absorbance changes by UV-vis spectrophotometry. Ammonium
hexanitratocerate(IV) (CAN)9 has been found to be the most
convenient oxidant to generate 10b in CH3CN: when CAN was
added progressively to a solution of 10a a new absorption band with
a maximum at 425 nm appeared. It increased until 1 equiv. of CAN
had been added.
In conclusion we have developed a new catalytic system for
aerobic oxidation using NHTPPI/CuCl. Enhanced kinetic stability
of the corresponding radicals allows efficient catalysis with low
catalyst loading. Further extention of this approach is underway.
We thank Professor J.-L. Pierre for fruitful discussions.
Notes and references
‡ To our knowledge the use of CuCl in association with N-hydroxyimides
for catalytic oxidations has been restricted to our own work on NHPI chiral
analogues.5 Its precise role has not yet been established. However, this role
can tentatively be related to the formation of m-oxocopper(II) species by
reaction of CuCl with dioxygen in acetonitrile.6 These species may generate
PINO or 10b radicals from NHPI or NHTPPI. Further investigations
concerning this point are underway.
§ Valuable NHPI analogues have been described by Ishii, bearing several
N-hydroxyimide functionalities in the same molecule,4d,f or a masked N-
hydroxyimide functionality.4e In the present study, we have restricted our
choice to NHPI analogues with a single free N-hydroxyimide moiety.
1 Y. Ishii, S. Sakaguchi and T. Iwahama, Adv. Synth. Catal., 2001, 343,
379.
2 (a) R. Arnaud, A. Milet, C. Adamo, C. Einhorn and J. Einhorn, J. Chem.
Soc., Perkin Trans., 2002, 2, 1967; (b) R. Amorati, M. Lucarini, V.
Mugnaini, G.-F. Pedulli, F. Minisci, F. Recupero, F. Fontana, P. Astolfi
and L. Greci, J. Org. Chem, 2003, 68, 1747; (c) N. Koshino, B. Saha and
J. H. Espenson, J. Org. Chem, 2003, 68, 9364; (d) N. Koshino, Y. Cai and
J. H. Espenson, J. Phys. Chem. A, 2003, 107, 4262; (e) B. Saha, N.
Koshino and J. H. Espenson, J. Phys. Chem. A, 2004, 108, 425.
3 C. Ueda, M. Noyama, H. Ohmori and M. Masui, Chem. Pharm. Bull.,
1987, 35, 1372.
This band has been attributed to radical 10b, as its fast
appearance was followed by slower decreasing accounting for self-
Table 2 Oxidation of various substrate catalysed by NHTPPI/CuCla
Conv.b
(%)
Run
Substrate
Time/h
Yieldc (%)
4 (a) K. Gorgy, J.-C. Lepretre, E. Saint-Aman, C. Einhorn, J. Einhorn, C.
Marcadal and J.-L. Pierre, Electrochem. Acta, 1998, 44, 385; (b) B. B.
Wentzel, M. P. J. Donners, P. L. Alster, M. C. Feiters and R. J. M. Nolte,
Tetrahedron, 2000, 56, 7797; (c) N. Sawatari, T. Yokota, S. Sakaguchi
and Y. Ishii, J. Org. Chem., 2001, 66, 7889; (d) A. Shibamoto, S.
Sakaguchi and Y. Ishii, Tetrahedron Lett., 2002, 43, 8859; (e) N.
Sawatari, S. Sakaguchi and Y. Ishii, Tetrahedron Lett., 2003, 44, 2053; (f)
N. Hirai, N. Sawatari, N. Nakamura, S. Sakaguchi and Y. Ishii, J. Org.
Chem., 2003, 68, 6587.
1
2
3
4
5
6
7
8
9
Indan
2-Methylindan
Tetralin
Indan-1-ol
4-Methoxy-1-ethylbenzene
Fluorene
4-Methylanisol
1,4-Diethylbenzene
Xanthene
6
27
18
24
23
48
48
8
83
73
84
15
79
84
34
69
98
93
77 (93)
65 (90)
74 (88)
15 (100)
70 (88)d
79 (94)d
25 (70)e
52 (75)
96 (98)
86 (92)
18
7
5 C. Einhorn, J. Einhorn, C. Marcadal-Abbadi and J.-L. Pierre, J. Org.
Chem., 1999, 64, 4542.
10
Isochroman
6 (a) M. Maumy and P. Capdevielle, Bull. Soc. Chim. Fr., 1995, 132, 734;
(b) P. Capdevielle and M. Maumy, Tetrahedron Lett., 1983, 24, 5611.
7 C. Einhorn, J. Einhorn and C. Marcadal-Abbadi, Synth. Commun., 2001,
31, 741.
8 For similar observations, see: C. Einhorn, J. Einhorn, C. Marcadal and
J.-L. Pierre, Chem. Commun., 1997, 447.
a Standard procedure: 1 mmol of substrate, 0.01 mmol of NHTPPI, 0.05
mmol of CuCl in 10 ml of acetonitrile with O2 at atmospheric pressure and
at 35 °C. b Determined by GC. c Yield of pure oxidation product after
column chromatographic purification. Numbers in parentheses give the
yield based on the substrate reacted. d 0.02 mmol of NHTPPI were used. e In
addition to 25% of p-anisaldehyde, 9% of p-anisic acid was also
obtained.
9 S. Sakaguchi, T. Hirabayashi and Y. Ishii, Chem. Commun., 2002,
516.
C h e m . C o m m u n . , 2 0 0 4 , 1 5 0 0 – 1 5 0 1
1501