Lipophilic N-Hydroxyphthalimide Catalysts for the Aerobic Oxidation of Cumene: Towards Solvent-Free Conditions and Back

Article abstract of DOI:10.1002/chem.201701573

A new class of lipophilic N-hydroxyphthalimide (NHPI) catalysts designed for the aerobic oxidation of cumene in solvent-free conditions was synthesized and tested. The specific strategy proposed for the introduction of lipophilic tails on the NHPI moiety

Full text of DOI:10.1002/chem.201701573

A Journal of
Accepted Article
Title: Lipophilic N-hydroxyphthalimide Catalysts for the Aerobic
Oxidation of Cumene: Towards Solvent-Free Conditions ... and
Back
Authors: Manuel Petroselli, Lucio Melone, Massimo Cametti, and Carlo
Punta
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201701573
Link to VoR: http://dx.doi.org/10.1002/chem.201701573
Supported by

10.1002/chem.201701573
Chemistry - A European Journal
FULL PAPER
Lipophilic N-hydroxyphthalimide Catalysts for the Aerobic
Oxidation of Cumene: Towards Solvent-Free Conditions ... and
Back
Manuel Petroselli,a Lucio Melone,a Massimo Camettia* and Carlo Puntaa*
Abstract: A new class of lipophilic N-hydroxyphthalimides catalysts
designed for the aerobic oxidation of cumene in solvent-free
conditions were synthesized and tested. The specific strategy
proposed for the introduction of lipophilic tails on the NHPI moiety
leads to lipophilic catalysts which, while completely preserving the
activity of the precursor, allow to conduct the catalytic oxidation in
neat cumene for the very first time. The corresponding cumyl
hydroperoxide is obtained in good yields (28-52%) and high
selectivity (95-97%), under mild conditions. Importantly, the
presence of a polar solvent is no longer required to guarantee the
complete solubilization of the catalyst. On the other hand, the
oxidation conducted in neat cumene unearths the unexpected
necessity of using small amounts of acetonitrile in order to fully
promote the hydrogen atom transfer process and prevent the
catalyst from detrimental hydrogen bond (HB) driven aggregation.
NHPI’s catalytic action occurs following a well-known free-
radical mechanism (Scheme 1). Briefly, in an initiation step the
phthalimide N-oxyl (PINO) radical is generated from NHPI (i).
PINO is then capable to undergo HAT from suitable C-H bonds,
promoting the formation of a carbon centered radical (ii). The
latter reacts with O2 under diffusion control, leading to the
corresponding peroxyl radical (iii). The cycle is finally closed by
a second molecule of NHPI which, acting also as a good
hydrogen donor, favors the propagation of the radical chain with
the formation of hydroperoxide and a new PINO unit (iv), while
delaying the termination reaction.
The catalytic efficiency is ascribed to the favorable synergy
among enthalpic, polar, and entropic effects. The bond
dissociation energy (BDE) of the O-H group in the NHPI moiety
(88.1 kcal/mol)[8] renders path i in many cases exothermic or
thermo-neutral. At the same time, the higher electrophilic
character of PINO radical compared to the peroxyl one favors
the stabilization of the transition state for HAT reaction of path
ii.[8] Finally, the relatively high value of kNHPI (7.2 x 103 M-1s-1)[8]
guarantees the efficiency of the propagation chain, which results
in high selectivity towards hydroperoxides, when operating
under proper conditions.
Introduction
For this reason, NHPI has been also widely investigated for
the selective oxidation of alkyl aromatics to the corresponding
hydroperoxides.[9-12] Our group recently focused on this topic as
well, proposing both a new initiation system, capable to generate
PINO at low temperatures, and an unique approach for the
recovery of the catalyst.[13,14]
Research on alkyl C-H bond activation via hydrogen atom
transfer (HAT), an important route for the synthesis of high
added value molecules and materials, is experiencing
a
renewed interest. The selective C-H bond oxidation represents
the most straightforward and versatile strategy for this
purpose,[1] favoring the direct oxygenation,[2] amination[3],
desaturation,[4] or halogenation[5] of a wide range of substrates.
However, this approach implies a demanding task, that of
combining high reactivity with high chemo-selectivity. The design
of new homogeneous catalysts capable both to promote
C(sp3)-H oxidation and limit, or better avoid the formation of side
products is thus highly desirable.
In this context, N-hydroxyphthalimide (NHPI) homogeneous
catalysis has found a starring role and has been particularly
exploited for promoting the aerobic oxidation of different organic
substrates. [6,7]
NHPI catalysis has been always performed in polar
solvents, in order to guarantee the complete solubilization of the
polar catalyst, especially at room temperature. As
a
consequence, the reactivity of this derivative in non-polar
mediums is unknown, and till now experimentally undetermined,
while the huge efforts devoted to the synthesis of more active
derivatives seem not to consider such detrimental aspect.[15,16]
The possible route to overcome this limitation would require
the design of new lipophilic NHPIs. Ishii first opened this way, by
suggesting the introduction of lipophilic chains onto the aromatic
ring of the N-hydroxy derivative.[17] Nevertheless, we recently
demonstrated that the proposed Ishii’s catalyst 1 (Scheme 2),
when used in processes which require a high control of
selectivity in hydroperoxide, is far from ideal.[18]
Indeed, not only its solubility is still low under mild conditions,
in spite of its higher lipophilic character, but the carboxylic group,
by which the alkyl tail is linked to the NHPI moiety, also affects
the NO-H BDE due to its electron-withdrawing character,
determining a value increase of 0.7 kcal/mol. As a result, the
efficiency of 1 as hydrogen donor is significantly reduced.
In the same work, we proposed catalyst 2 (Scheme 2) as a
suitable lipophilic alternative to 1 for the aerobic oxidation of
alkyl aromatics. The catalyst 2 showed an O-H BDE analogous
to that of NHPI and a similar reactivity. This solution did not
[a]
M. Petroselli, L. Melone, M. Cametti, C. Punta
Department of Chemistry, Materials, and Chemical Engineering “G.
Natta”
Politecnico di Milano and INSTM Local Unit
Via Mancinelli 7, I-20131 Milano, Italy
E-mail: carlo.punta@polimi.it; massimo.cametti@polimi.it
Supporting information for this article is given via a link at the end of
the document.((Please delete this text if not appropriate))
This article is protected by copyright. All rights reserved.

10.1002/chem.201701573
Chemistry - A European Journal
FULL PAPER
allow to completely remove the polar co-solvent, but it was
possible to run cumene (CU) oxidations under homogeneous
could open the route to further substitution, while 4c-e
progressively increase the lipophilic character of the tail.
The
consisted
second
into
step
the
hydrolysis of compounds
5a-e to the corresponding
phthalic acid derivatives
6a-e,.[19]
Finally, the closure of
the
corresponding N-hydroxy
phthalic derivatives
cycle
to
the
occurred following a two-
step procedure previously
reported.
intermediates
The
7a-e
were
obtained
substrates
acetic
by
6a-e
anhydride
reacting
with
in
microwave at 140 °C for
just
h.[20] The crude
1
anhydride products were
then let to react, without
further
purification, in
microwave
with
hydroxylamine
hydrochloride,[18] leading
Scheme 1. General improved catalytic mechanism for NHPI-mediated aerobic oxidation of hydrocarbons
to the formation of 8a-e
potential catalysts, with 85% yield after purification.
Solubility tests
O
O
O
N
OH
O
5
O
1
2
O
O
The effect of lipophilic tails on the solubility of 8a-e in CU
was investigated in detail. Table 1 reports the minimum amount
of MeCN required for having 4 % mol of each N-hydroxy
derivative completely solubilized, when added to 0.5 ml of CU.
NHPI, 1, and 2 were also investigated for comparison.
O
HO
N
N OH
O
O
Scheme 2. Ishii’s catalyst 1 and organocatalyst 2.
Table 1. Solubilization of N-hydroxy derivatives in CU at room
temperature by addition of the minimum quantity of MeCN as co-
solvent.a)
conditions even at 45 °C with 1% of 2, drastically reducing the
CU/MeCN(v/v) ratio up to 2.67. Under the same solvent
composition, NHPI resulted almost completely insoluble. As
expected, the new conditions also led to an increment in cumyl
hydroperoxide (CHP) selectivity.
Following this proof of concept, we herein report on the
design and synthesis of a new class of lipophilic NHPI
derivatives, reaching the final goal of achieving catalysts
completely soluble in neat CU even at room temperature.
MeCN
Entry
Compound
Volume (µl)
% v/v
73
33
26
26
36
41
/
1
NHPI
1
1375
250
175
175
275
350
/
2
3b)
2
4
8a
8b
8c
8d
8e
5
6
Results and Discussion
7
Synthesis of the new lipophilic NHPI catalysts
8
65
12
a)
General conditions: 0.5 mL of CU (3.6 mmol) added with 4%
mol of selected compound at 25 °C. b) 2% of catalyst 2 was
added.
The general multi-step procedure for the synthesis of the
new lipophilic NHPI catalysts is summarized in Scheme 3.
In a first step, 4-nitrophthalonitrile (3) was reacted with
phenols bearing different substituents on the ortho and/or para
positions (4a-e). 4a was selected in order to reproduce the
mono-functional derivative analogous to the double-NHPI
compound 2. 4b, which presents a halogen in para position,
As expected, while NHPI requires a huge amount of MeCN
(73% v/v) to remain in solution (entry 1), the presence of the
lipophilic chain in catalyst 1 induces a considerable increase of
solubility in CU (entry 2), with a consequent significant decrease
This article is protected by copyright. All rights reserved.

10.1002/chem.201701573
Chemistry - A European Journal
FULL PAPER
of the required amount of polar co-solvent (33% v/v). An even
lower quantity of MeCN (26% v/v) was required for the complete
solubilization of 2 (entry 3). In this case, as derivative 2 bears
two equivalents of NHPI moieties, a 2% mol was used, in order
to compare the different catalysts in terms of real NHPI units’
concentration in CU. The same trend, with an analogous volume
of MeCN required, was confirmed by testing the solubility of 4%
position of the phenolic ring. For this reason, we synthesized
NHPI derivatives 8d and 8e, bearing 4-hexylphenoxy and 4-
hexyloxylphenoxy substituents respectively, which resulted to be
much more soluble in CU (entries 7 and 8). In particular, it was
possible to completely solubilize 4% mol 8d in neat CU even at
room temperature, while compound 8e required the addition of
12% of MeCN.
OH
R2
R1
R2
O
O
Cl
O
CN
CN
4a-e
O2N
CN
R3
N
OH
N
OH
R3
R1
O
O
K2CO3, DMF
25 or 100 °C,
2 days
CN
O
O
O
8b
8a
5a-e
3
Yield > 96%
NH2OH*HCl, Py
H2O / NaOH
24 - 48 h,
Reflux
N
OH
O
Mw: 120 °C, 1h
Yield > 85%
O
8c
Yield > 94%
O
O
O
R2
HO
N
Ac2O
N
OH
O
COOH Mw: 140 °C
O
O
O
O
1h
8d
R3
R1
COOH
8e
6a-e
Scheme 3. Multi-step procedure for the synthesis of the new lipophilic NHPIs.
of compound 8a, the half-unit of the corresponding catalyst 2
This trend was confirmed by determining the minimum
(entry 4), while 8b showed a behavior analogous to catalyst 1
(entry 5).
temperature required for a complete solubilization in neat
cumene (Table 2). As expected, NHPI could not be dissolved
even by heating up to 140 °C (entry 1), and the same result was
also observed both for the previously reported catalyst 2 (entry
4), and for the new derivatives 8a and 8b (entries 5 and 6,
respectively), while just a slight decrease of the minimum
temperature was measured for compound 8c (entry 7).
On the contrary, Ishii’s lipophilic catalyst 1 reached a
complete solubilization at 63-65 °C (entry 2), once again
confirming the beneficial effect of introducing a lipophilic tail on
the NHPI unit. In this case, a minimum temperature of 52-53 °C
was also measured when operating with 1% of the catalyst
(entry 3), which simulates the standard oxidation conditions.[12]
However, as already observed in Table 1, the best result
was obtained for compound 8d, which resulted the unique
derivative completely soluble in neat CU at room temperature
(entry 9). The analogous derivative 8e, whose importance will be
discussed in depth in the next sections, also showed a high
solubility, with minimum temperature ranges of 47-50 °C (entry
9) and 34-37 °C (entry 10) with 4% and 1% catalyst/CU mol ratio,
respectively. These values are in any case significantly lower
than those reported for Ishii’s catalyst 1 (entries 2 and 3),
testifying the higher lipophilic character of derivative 8e.
Unexpectedly, compound 8c, which should exhibit a higher
lipophilic character due to the presence of three methyl groups
on the phenolic ring, required a relatively high volume of MeCN
(41% v/v) (entry 6). We hypothesized that the presence of the
methyl groups could negatively affect the possible π-staking
interactions with cumene, deleting the beneficial effects of an
increased lipophilic character.
Table 2. Minimum temperatures for complete solubilization of
catalysts in neat cumene.a)
Catalyst
amount
Minimum
Temperature
Entry
Compounds
mmol
% mol
4%
4%
1%
2%
4%
4%
4%
4%
4%
1%
°C
> 140
63-65
52-53
> 140
> 140
> 140
123-126
28-29
47-50
34-37
1
2
3
4
5
6
7
8
9
NHPI
1
0.14
0.14
0.036
0.072
0.14
0.14
0.14
0.14
0.14
0.036
1
2
8a
8b
8c
8d
8e
8e
Cumene oxidation catalyzed by 8d
The solubility observed for compound 8d allowed to conduct
a solvent-free oxidation of alkylaromatics catalyzed by N-
hydroxyphthalimide derivatives (Table 3) for the very first time.
Propionaldehyde (5% mol) was used as suitable initiator,[21-23]
being able to promote PINO generation even at low
temperatures by molecule-induced homolysis,[24] via in situ
formation of the corresponding peracid (Scheme 4). Under these
reaction conditions, 18% of CU was converted to CHP, with a
selectivity of 97%, while by operating in the absence of 8d, only
2 % of conversion was observed (entry 2).
10
a)
General conditions: 0.5 mL of CU (3.6 mmol) added with
reported % mol of selected compounds at 28 °C, and slowly
heated up to the temperature of complete solution
homogeneity.
As a consequence, we opted to vacate the ortho positions of
the phenol substituents and to increase the lipophilic character
of N-hydroxy compounds by introducing alkyl chains in para
This article is protected by copyright. All rights reserved.

10.1002/chem.201701573
Chemistry - A European Journal
FULL PAPER
Table 3: Solvent-free aerobic oxidation of cumene catalyzed by new lipophilic NHPIs.a)
CHP
Catalyst
Compound
Temp.
(°C)
45
MeCN
Conversion
Selectivity
(%)
97
Entry
Initiator
% mol
2%
/
(µL)
(%)
18
2
Scheme 4. Molecule-induced homolysis mechanism for
1
CH3CH CHO (5%)
8d
-
generation of PINO with ald2ehydes.
2
CH3CH2CHO (5%)
CH3CH2CHO (5%)
AIBN (2%)
/
45
-
79
3
8d
8d
/
2%
2%
/
70
-
23
43
20
27
15
18
28
30
45
32
19
28
52
54
95
4
70
-
-
96
5
AIBN (2%)
70
69
6b)
7c)
8c)
9
CH3CH2CHO (5%)
CH3CH2CHO (5%)
CH3CH2CHO (5%)
CH3CH2CHO (5%)
CH3CH2CHO (5%)
AIBN (2%)
2
1%
2%
2%
2%
2%
2%
2%
2%
2%
2%
2%
45
300
10.000
10.000
25
97
8d
NHPI
8d
8d
8d
8d
8e
8e
8e
8e
45
89
45
81
45
97
10
11
12
13
14
15
16
70
25
95
70
25
96
CH3CH2CHO (5%)
CH3CH2CHO (5%)
CH3CH2CHO (5%)
AIBN (2%)
45
100
-
94
45
97
45
25
96
70
-
95
AIBN (2%)
70
25
95
a)
General Reaction Conditions: Cumene (2 mL, 14.3 mmol), O2 (1 atm), 8d or 8e, 6 h, Conversion and selectivity were
obtained by H-NMR with internal standard. b) see ref. 18. c) 5 mmol of cumene were reacted in 10 mL of MeCN. Difference
1
among independent measurements are within 2%.
An increase of temperature up to 70 °C resulted in just a
slight increment in CU conversion (23%, entry 3). We assumed
that this low temperature effect could be associated to the
nature of the initiator, which could be quickly oxidized and
consumed at higher temperatures. For this reason, we repeated
the reaction at 70 °C in the presence of AIBN (entry 4), and in
this case we found a significantly higher conversion (43%),
preserving the selectivity of the process.
While confirming the catalytic activity of 8d, the conversions
obtained using propionaldehyde as initiator were under our
expectations, based on the comparison with our previous work
on 2 (see entry 6). In order to shade light on this behavior, we
compared the catalytic efficiency of the new lipophilic derivative
8d with that of NHPI, by conducting the oxidation of CU in MeCN
(entries 6 and 7). The results revealed a comparable reactivity
between 8d and NHPI, and thus we initially supposed that the
poor conversion observed at 45 °C in neat CU could be ascribed
to the different polarity of the reaction medium. Indeed, an
increase of the polarity of the reaction medium could lead to the
stabilization of the transition state in the HAT reaction of path ii
(Scheme 1), resulting in higher catalytic performances.
Seemingly pointing to the above hypothesis, a significantly
higher conversion (28%) was observed by performing the same
reaction with the addition of 25 μl of MeCN to the oxidation
solution (entry 9). An analogous beneficial effect was also
observed at 70 °C (comparison between entries 3 and 10), when
operating with propionaldehyde initiator. With AIBN no
differences were observed (entries 4 and 11).
Notably, however, oxidations carried out in the presence of a
higher amount of MeCN in solution (100 μl, entry 12), do not
show any further increase in CU conversion, thus a simple
solvent polarity effect should be ruled out, and another
explanation sought.
As, from a structural point of view, the NHPI moiety
possesses both hydrogen bond (HB) acceptor (C=O) and donor
(N-OH) groups, catalyst 8d might display self-complementary
and, thus, the tendency to aggregate into dimers (or higher HB
adducts) when dissolved in low polarity media (see Scheme 4).
This article is protected by copyright. All rights reserved.

10.1002/chem.201701573
Chemistry - A European Journal
FULL PAPER
mechanism as described in path ii of Scheme 1, being the NO-H
bond directly involved in the self-aggregation process. DFT
calculations performed in the gas phase for NHPI could provide
an initial tentative guess at the dimeric structure that could be
found in apolar solution. Interestingly, calculated HB
Odonor∙∙∙Oacceptor distances (ca. 2.75 Å) are similar to the HB
distance found in the only X-ray determined structure of NHPI
Scheme 4. Top: HB driven self-association into dimeric species
for a NHPI derivative in apolar medium; Bottom: Electrostatic
potential surface (ESP) for the two most stable dimers of NHPI,
calculated in gas phase at 298 K (B3LYP/6-311G(d,p)). See ESI
(Section Computational Study) for further information.
As 8d is employed in neat cumene, an eventually forming
dimeric species (or a higher aggregate), might hamper the HAT
Figure 2. Plot of the chemical shift, δ, of the aromatic H2 signal
belonging to the NHPI moiety in 8d (●) and 9 (○), versus their
molar concentration, measured in CDCl3 at 305K
(ca. 2.68 Å).[25]
In order to verify the possibility of aggregation, we first
investigated whether any concentration dependent effect on the
1H-NMR spectrum of 8d could be observed in CDCl3.
The results of this experiment are shown in Figure 1 where
the spectra of 8d at different concentrations are reported. A
significant downfield shift of all the signals in the aromatic region
can be seen. Plot of the chemical shift, δ, of the aromatic H2
signal belonging to the NHPI moiety in 8d versus its molar
concentration is shown in Figure 2 (●). Also, we synthesized
compound 9 (Figure 2), analogous to 8d but for its N-OMe group,
which lacks any HB donating capability. Being 9 incapable to
form any dimeric HB adducts, it represents an optimal model
compound for comparison with 8d.
The data clearly point to a self-association process for 8d
and they can be fitted by non-linear least square methods with a
binding isotherm based on a 1:1 association model which
affords a self-association constant K equal to 2.3 ± 0.2 M-1 (see
Fig. SI 31). Model compound 9, instead, under the same
experimental conditions, shows a markedly different behavior,
i.e., a linear decrease of the δ of the H2 signal of 9 (see Figure 2,
○ data points). This strongly confirms the validity of our initial
hypothesis: 8d can indeed self-associate into HB dimeric
adducts at high concentrations.
Small though the determined K value might appear, it can be
easily demonstrated that under the oxidation reaction conditions
(catalyst concentration ca. 0.145 M, Table 3) it results in a
significant association, with the monomeric species reduced to
ca. 70% of the total catalyst concentration (see Fig SI 36). Ishii’s
catalyst, 1, soluble in CDCl3, was also tested and it was found to
behave similarly to 8d (K equals 3.3 ± 0.2 M-1, see Fig. SI 33).
1
Figure 1. H-NMR spectra of 8d in CDCl3 at 305 K at different
concentrations ranging from 0.5 to 9.2 x 10-3 M, from bottom to
top
This article is protected by copyright. All rights reserved.

10.1002/chem.201701573
Chemistry - A European Journal
FULL PAPER
As catalyst 8d, in the present work, succeeds in operating in
neat cumene, we were interested in studying the same self-
association phenomenon under conditions which could be as
similar as possible to those employed in the real catalytic
of the tertiary benzyl C-H bond in CU respect to that of the
secondary one in 8d. In order to verify this hypothesis and to
identify the possible oxidation products, we promoted an
oxidation of 8d in the absence of CU under drastic conditions (in
MeCN and at 70 °C, using an equivalent amount of AIBN as
radical initiator). At the end of the reaction, 82% of the initial
catalyst was converted, as determined by 1H-NMR analysis
(entry 1, Table 4). As expected, the oxidation products derived
from the conversion of benzyl C-H to a mixture of the
corresponding alcohol, hydroperoxide and ketone (see Fig. SI
41 and 43).
Even if the oxidation of 8d under the standard reaction
conditions in neat CU occurs to a small extent, and negligibly
affects the catalytic efficiency, we set up to overcome this
potential limit by designing the analogous compound 8e, where
an oxygen atom is inserted between the phenolic ring and the
alkyl chain. While the solubility of the new catalyst in CU
appeared just slightly lower, as previously discussed (see entry
8, Table 1, and entries 9 and 10, Table 2), 8e resulted to be
much more resistant to direct oxidation, with only 12% of
conversion (see Fig. SI 47). due to the attack on the α-position
to the oxygen atom (entry 2, Table 4).
We tested 8e as catalyst in the solvent-free oxidation of
cumene at 45 °C, with 5% of propionaldehyde, and we observed
analogous results to those obtained with 8d, in terms of both CU
conversion and CHP selectivity (entry 13, Table 3). Also in this
case, the addition of tiny amounts of MeCN (25 µL) resulted in a
significant increase in CU conversion, from 18% to 28% (entry
14, Table 3), confirming one more time the activation role of the
polar solvent at lower temperatures.
Figure 3. Plot of the chemical shift, δ, of the aromatic H2 signal
belonging to the NHPI moiety in 8d (●) and 9 (○) versus their
molar concentration, measured in toluene-d8 at 305 K
process. Thus, we repeated the above dilution experiment in
[26]
toluene-d8
and the results are reported in Figure 3, where a
plot of the chemical shift of the H2 signal of the NHPI moiety in
8d versus its molar concentration is shown (● data points). It
appears evident that the monotonic behavior displayed in CDCl3
is now replaced by a more complex profile. Indeed, by dilution
an initial upfield shift in the high concentration regime is
observed, followed by a downfield shift in the low concentration
regime (minimum at ca. 0.025 M). Again, self-association is the
likely explanation for observed data.[27] We speculate that at low
concentration of 8d, an initial dimeric HB-driven species forms,
similarly to what occurs in CDCl3, while at higher concentration,
due to the low polarity of toluene solvent, larger aggregates start
to appear. The same experiment performed in a toluene:MeCN
mixture (95:5) shows only nominal changes of 8d signals  upon
its dilution (see Figure SI 37).
Again, the comparison with the model compound 9 (○ data
points) substantiates our hypothesis concerning the nature of
the adduct formed in the low concentration regime, which is
most likely HB driven. As to the linear behavior of the chemical
shift of the H2 signal in model 9, observed in both toluene-d8 and
in CDCl3, it can be explained as due to a non-specific and
inefficient aggregation of the compound probably due to -
stacking interactions.[28] Additional confirmation of the
aggregation tendency of compound 8d can be also obtained by
ATR-IR and fluorescence studies (See ESI - Section ATR-IR
and Fluorescence analysis).
By moving up to 70 °C, with 2% AIBN as initiator, we
obtained the highest conversion (52 %), with a selectivity in
hydroperoxide of 95% (entries 15 and 16, Table 3). Even under
these reaction conditions, we did not observe any oxidation
products deriving from the catalyst, and this confirms the quality
of the catalyst design.
Table 4. Stability of catalysts 8d and 8e to oxidation conditions.a)
Catalyst
Conversionb)
Entry
(%)
82
1
2
8d
8e
12
a)
MeCN (2 ml), catalyst (0.29 mmol), O2 (1 atm), 6 h. Initiator:
AIBN (0.29 mmol). b) Conversion determined by H-NMR analysis
(See ESI). Difference among independent measurements are
within 2% based on three different experiments.
1
Conclusions
NHPI homogeneous catalysis for oxidative processes has
revealed to be a promising system for several applications,
especially for the selective aerobic oxidation of alkylaromatics to
the corresponding hydroperoxides. However, as of yet this
approach suffered from the necessity to use high amounts of co-
solvent, due to the polar character of the NHPI molecule and
solubility issues. Despite significant efforts in the last decade
devoted to work out this problem, most of the proposed solutions
were not able to consider both aspects of solubility and reactivity.
In this context, this work opens the way to a new class of
Cumene oxidation catalysed by 8e
By conducting the aerobic oxidation of CU at 70 °C in the
presence of 8d, we observed the formation of tiny amounts of
side products clearly derived from degradation of the catalyst.
We assumed that they could be associated to the partial
oxidation onto the benzyl position of the catalyst, in spite of the
high molar ratio between CU and 8d and of the higher reactivity
This article is protected by copyright. All rights reserved.

10.1002/chem.201701573
Chemistry - A European Journal
FULL PAPER
for 2 days. The solution was diluted in 150 ml of water and cooled at 0 °C
overnight, obtaining a pink-white solid precipitate. Average yield: 94%.
lipophilic NHPI catalysts, bearing alkyl aromatic tails on the
NHPI moiety. The new catalysts are designed to improve their
solubility in apolar media while maintaining unaltered the NO-H
BDE value of the NHPI moiety, and thus preserving the catalytic
efficiency in terms of conversion and selectivity in
hydroperoxides. The new catalysts allowed, for the first time, to
perform solvent-free oxidations of alkylaromatics under mild
conditions (45 °C and atmospheric pressure), obtaining good
1
4-phenoxyphthalonitrile (5a): H-NMR (400 MHz, DMSO-d6) δ 8.09 (d,
1H), 7.75 (s, 1H), 7.51 (t, 2H), 7.38 (dd, 1H), 7.32 (t, 1H), 7.20 (d, 2H).
13C-NMR (400 MHz, DMSO-d6) δ 161.52, 154.28, 136.78, 131.13,
126.29, 123.17, 122.46, 120.79, 117.18, 116.35, 115.84, 108.65.
4-(p-chlorophenoxy)phthalonitrile (5b): 1H-NMR (400 MHz, DMSO-d6)
δ 8.11 (d, 1H), 7.81 (s, 1H), 7.55 (d, 2H), 7.45 (dd, 1H), 7.25 (d, 2H). 13C-
NMR (400 MHz, DMSO-d6) δ 161.16, 153.24, 136.78, 130.97, 130.21,
123.32, 122.83, 122.67, 117.24, 116.30, 115.79, 109.07.
4-(2,4,6-trimethylphenoxy)phthalonitrile (5c): 1H-NMR (400 MHz,
DMSO-d6) δ 8.03 (d, 1H), 7.54 (s, 1H), 7.14 (dd, 1H), 7.01 (s, 2H), 2.27
(s, 3H), 2.00 (s, 6H). 13C-NMR (400 MHz, DMSO-d6) δ 161.22, 147.46,
136.94, 135.96, 130.45, 130.25, 120.97, 120.01, 117.35, 116.41, 115.90,
107.82, 20.82, 16.09.
yields, high selectivity in hydroperoxides and
a higher
productivity in comparison with previously reported processes.
While the use of a polar co-solvent is thus no longer required
with our novel lipophilic catalysts, this work also uncovers an
intrinsic limit to the implementation of solvent free conditions.
Indeed, a detailed study on catalyst 8d reveals its tendency
towards HB-driven aggregation in low polarity solvents, as
1
demonstrated by H-NMR investigation, also confirmed by ATR-
General procedure for the synthesis of 5d and 5e: In a round-bottom
flask of 50 ml, 4-phthalonitrile 3 (1 g, 5.78 mmol), 4-hexylphenol (4d, 1.05
g, 5.78 mmol) or 4-(p-hexyloxyl)phenol (4e, 1.12g, 5.78 mmol), K2CO3
(1.6 g, 11.56 mmol) were dissolved in DMF (20ml) and stirred at 100 °C
for 24 hours. The solution was diluted with 50 ml of water and extracted
with ethyl acetate. The organic layer was washed with water (3 x 15 ml),
dried with Na2SO4, filtered and concentrated in vacuum. 5d (1.67 g of
pure pink-white solid) and 5e (1.72 light brown solid) were obtained (93%
yield in both cases).
IR and fluorescence spectroscopy. This finding allowed us to
rationalize those data which showed, under certain experimental
conditions, a lower than expected catalytic efficiency (entry 1,
Table 1), but more generally, it shows that the addition of tiny
amounts of MeCN to the reaction medium is necessary to allow
the catalytic system to show its full potential.
Finally, catalyst stability under the reaction conditions has
been also investigated. The partial oxidation of 8d on the benzyl
position suggests the use of the analogous 4-(p-
hexyloxylphenoxy)-N-hydroxyphthalimide 8e, which maintains a
high solubility in cumene, reproduces the catalytic efficiency of
8d in terms of yield, selectivity and productivity, but importantly
is not converted during the oxidation process.
4-(p-hexylphenoxy)phthalonitrile (5d): 1H-NMR (400 MHz, DMSO-d6)
δ 8.07 (d, 1H), 7.70 (s, 1H), 7.35 (dd, 1H), 7.31 (d, 2H), 7.10 (d, 2H), 2.61
(t, 2H), 1.59 (m, 2H), 1.29 (m, 6H), 0.86 (t, 3H). 13C-NMR (400 MHz,
DMSO-d6) δ 164.14, 163.88, 163.46, 152.80, 140.17, 131.87, 130.67,
125.93, 122.43, 120.67, 111.62, 34.91, 31.52, 31.35, 28.76, 22.48, 14.37.
4-(p-hexyloxylphenoxy)phthalonitrile (5e):
1H-NMR (400 MHz,
In conclusion, compounds 8d and 8e resulted to be the
most soluble N-hydroxy catalysts ever reported, in particular
compound 8e features a higher chemical stability and HAT
efficiency with respect to compound 1.
DMSO-d6) δ 8.06 (d, 1H), 7.68 (s, 1H), 7.32 (dd, 1H), 7.13 (d, 2H), 7.04
(d, 2H), 3.97 (t, 2H), 2.07 (m, 2H), 1.42 (m, 2H), 1.31 (m, 4H), 0.88 (t,
3H). 13C-NMR (400 MHz, DMSO-d6) δ 162.41, 156.97, 147.09, 136. 67,
122.31, 122.22, 121.67, 117.06, 116.54, 116.36, 115.84, 108.05, 68.41,
31.45, 29.12, 25.65, 22.52, 14.32.
Experimental Section
General procedure for the synthesis of 6a-c: In a round-bottom flask
of 100 ml, 5a-c (5.7 mmol, 1.25 g, 1,46 g, and 1.44 g, respectively) and
sodium hydroxide (4.5g, 113.28 mmol) were dissolved in 30 ml of water.
The solution was stirred for 24 hours at reflux. The solution was diluted
with 80 ml of water and HCl conc. was added until pH 1-2. The resulting
solution was cooled at 0 °C for 12 hours. A white solid was obtained and
filtered, obtaining a minimum yield in 4a-c of 94%.
General Methods
All chemicals, reagents and solvents were obtained from commercial
sources and were used without further purification, if not mentioned
otherwise. Catalysts 1 and 2 were synthesised following the procedure
[12]
reported in the literature
1H-NMR spectra of the products were
4-phenoxyphthalic acid (6a): 1H-NMR (400 MHz, DMSO-d6) δ 13.01 (s,
2H), 7.76 (d, 1H), 7.46 (t, 2H), 7.24 (t, 1H), 7.11 (m, 4H). 13C-NMR (400
MHz, DMSO-d6) δ 168.88, 167.91, 159.78, 155.46, 136.92, 131.83,
130.87, 126.31, 125.33, 120.44, 119.22, 116.96. MS (ESI): m/z = 281.0
[M+Na]+.
recorded at 305 K with a Bruker Avance-400 MHz NMR spectrometer
(Bruker, Billerica, MA, USA). ESI-MS mass spectra were collected on a
Bruker Esquire 3000+ with electrospray ionization source and ion-trap
detector. The samples were analyzed by direct infusion of suitable
solutions (methanol) in the spectrometer source. The IR spectra reported
1
4-(p-chlorophenoxy)phthalic acid (6b): H-NMR (400 MHz, DMSO-d6)
in ESI were obtained by
a Varian 640 high-performance ATRIR
δ 13.01 (s, 2H), 7.78 (d, 1H), 7.50 (d, 2H), 7.16 (m, 4H). 13C-NMR (400
MHz, DMSO-d6) δ 168.77, 167.91, 159.29, 154.50, 136.92, 131.86,
130.70, 129.14, 126.84, 122.11, 119.53, 117.34. MS (ESI): m/z = 315.0
[M+Na]+.
4-(2,4,6-trimethylphenoxy)phthalic acid (6c): 1H-NMR (400 MHz,
DMSO-d6) δ 13.01 (s, 2H), 7.74 (d, 1H), 7.01 (s, 2H), 6.88 (m, 2H), 2.28
(s, 3H), 2.02 (s, 6H). 13C-NMR (400 MHz, DMSO-d6) δ 169.17, 167.83,
159.78, 148.06, 137.31, 135.31, 132.07, 130.46, 130.27, 124.85, 115.99,
113.93, 20.81, 16.16. MS (ESI): m/z = 323.1 [M+Na]+.
spectrometer. The fluorescence maps reported in ESI were obtained by a
spectrofluorometer Fluorolog® (See ESI for further information). The
NMR studies of the aggregation of NHPIs were conducted in CDCl3 and
toluene-d8; more details, are reported in the ESI. The melting points were
measured on a Büchi 535 apparatus. Flash column chromatography was
performed by using 40–63 μm silica gel packing; the eluent was chosen
in order to move the desired components to Rf 0.35 on analytical TLC.
Synthesis of the catalysts
Synthesis of 6d:In a round-bottom flask of 250 ml, 5d (3.27g, 10.74
mmol) and NaOH (8.59g, 215 mmol) were dissolved in 60 ml of water.
The solution was refluxed for 2 days under stirring, and the diluted with
additional 80 ml of water. HCl 37% (v/v) was added until a white
precipitate was formed. The resulting solution was cooled at 0 °C for 12
General procedure for the synthesis of 5a-c: In a round-bottom flask
of 50 ml, 4-phthalonitrile 3 (1 g, 5.78 mmol), phenol 4a-c (5.78 mmol, 544
mg, 743 mg, and 787 mg, respectively), K2CO3 (1.6 g, 11.56 mmol) were
dissolved in DMF (20ml). The solution was stirred at room temperature
This article is protected by copyright. All rights reserved.

10.1002/chem.201701573
Chemistry - A European Journal
FULL PAPER
hours to favor the product precipitation. The white solid was finally filtered
(3.08 g, 86 % yield) and used without further purification for the next
synthetic step.
(m, 4H), 0.89 (t, 3H). 13C-NMR (400 MHz, DMSO-d6) δ 164.15,164.09,
163.90, 156.71, 147.84, 131.86, 125.88, 122.27, 121.80, 116.48, 111.06,
68.41, 31.46, 29.13, 25.65, 22.52, 14.34. MS (ESI): m/z = 356.1 [M+H]+,
m/z = 378.1 [M+Na]+, mp = 87.4 °C.
4-(p-hexylphenoxy)phthalic acid (6d): 1H-NMR (400 MHz, DMSO-d6) δ
13.01 (s, 2H), 7.75 (d, 1H), 7.27 (d, 2H), 7.08 (dd, 1H), 7.04 (m, 3H), 2.58
(t, 2H), 1.57 (m, 2H), 1.28 (m, 6H), 0.85 (t, 3H). 13C-NMR (400 MHz,
DMSO-d6) δ 164.14, 163.88, 163.46, 152.80, 140.17, 131.87, 130.67,
125.93, 122.34, 120.67, 111.62, 34.91, 31.52, 31.35, 28.76, 22,48, 14.37.
MS (ESI): m/z = 365.1 [M+Na]+.
Procedure for the synthesis of 9: In a round-bottom flask of 50 ml,
compound 8d (300 mg, 0.884 mmol) was dissolved in 15 ml of acetone.
NaOH (36 mg, 0,884 mmol) was added and, after sonication, the mixture
was diluted with water to obtain a homogeneous red clear solution.
Finally, CH3I (1.53 g, 10.6 mmol) was added in five aliquots, one each
hour. After 5 hours, the resulting colorless solution was concentrated
under reduced pressure, leading to the precipitation of a yellow solid,
which was filtered and purified by flash chromatography column
(hexane/ethyl acetate 4/1) obtaining 304 mg of pure 9 (97 % yield).
Synthesis of 6e:In a round-bottom flask of 250 ml, 5e (1.73g, 5.4 mmol)
and NaOH (4.32g, 108 mmol) were dissolved in 100 ml of water. The
solution was refluxed for 2 days under stirring. Work up was analogous to
that reported for compound 6d. A green solid was finally filtered (1.77g,
92%) and used without further purification for the next synthetic step.
4-(p-hexyloxylphenoxy)-N-methoxyphthalimide (9): 1H-NMR (400
MHz, CDCl3) δ 7.70 (d, 1H), 7.24 (d, 1H), 7.20 (dd, 1H), 7.17 (d, 2H),
6,92 (d, 2H), 3.98 (t, 3H), 2,56 (t, 2H), 1.57 (m, 2H), 1.27 (m, 6H), 0.83 (t,
3H). 13C-NMR (400 MHz, CDCl3) δ 164.29, 163.03, 162.98, 152.33,
140.53, 131.36, 130.24, 125.64, 122.41, 121.81, 120.34, 111.80, 65.84,
35.34, 31.69, 31.47, 28.97, 22.59, 14.07. MS (ESI): m/z = 354.1 [M+H]+,
m/z = 376.1 [M+Na]+, 391.8 [M+K]+
1
4-(p-hexyloxylphenoxy)phthalic acid (6e): H-NMR (400 MHz, DMSO-
d6) δ 13.01 (s, 2H), 7.75 (d, 1H), 7.09-6.99 (m, 6H), 3.97 (t, 2H), 1.72 (m,
2H), 1.43 (m, 2H), 1.32 (m, 4H), 0.89 (t, 3H). 13C-NMR (400 MHz,
DMSO-d6) δ 169.05, 167.81, 160.89, 156.40, 148.19, 137.02, 131.81,
125.47, 122.09, 118.16, 116.35, 115.87, 68.37, 31.46, 29.15, 25.65,
22.52, 14.23. MS (ESI): m/z = 359 [M+H]+, m/z = 381.1 [M+Na]+.
Tests of Solubility
General procedure for the synthesis of 8a-e: In a two-neck round-
bottom flask of 50 ml, 6a-e (5.3 mmol, 1.38 g, 1. 56 g, 1.58 g, 1.83 g, and
1.91 g, respectively) were dissolved in acetic anhydride (30 ml) and
heated at 140 °C for 1 hour in a microwave reactor with automatic control
of the power (Micro-SYNTH Labstation; Milestone Inc. USA). The
solution was concentrated under reduced pressure (60 °C and < 20
mbar), obtaining derivatives 7a-e. In the same two-necks round-bottom
flask containing 7a-e, pyridine (30 ml) and hydroxylamine hydrochloride
(3.7 g, 53.3 mmol) were added. The mixture was heated at 120 °C for 1
hour in the microwave reactor. The resulting orange solution was
evaporated under reduce pressure (60 °C, > 10 mbar), washed with a
solution of diluted HCl and extracted with ethyl acetate (3 x 15 ml). The
organic layer was washed with water, dried with Na2SO4 anhydrous,
filtered and evaporated under vacuum. All the catalysts resulted in a
yellow solid. 8a-c were purified by flash chromatography (eluent
hexane/ethyl acetate 1/1). 8d-e were purified by crystallization, dissolving
them in the minimum quantity of toluene and adding hexane until the
formation of a pale yellow solid. In all cases a minimum yield of 85 % was
observed.
Determination of the minimum volume of MeCN to be added for
complete solubilization in cumene: The minimum volumes of MeCN
required for the complete solubilization of each catalyst in CU were
obtained by adding the required amount of catalyst (4% mol respect to
CU) to a two-neck round-bottom flask containing CU (0.5 ml). The
mixture was stirred for 5 min at room temperature. Fixed volumes of
MeCN (25µl) were injected into the flask with a calibrated syringe every 5
minutes, until complete solubilization of the catalyst was observed.
Determination of the minimum temperature for complete
solubilization in cumene: In a two-neck round-bottomed flask, the
selected amount of each catalyst (generally 4% mol) was added to 0.5 ml
of cumene (3.6 mmol). The mixture was put in an oil bath under stirring at
room temperature. The temperature was increased of 5 °C every 20
minutes, until complete solubilization of the catalyst was observed.
Cumene aerobic oxidations
General procedure for cumene oxidations in MeCN under diluted
conditions:Aerobic oxidations were performed in a 50 mL two-necked
round-bottom flask equipped with a condenser and containing CU (0.6g,
5 mmol), MeCN (10 ml), and an appropriate amount of catalyst, as
reported in Table 3. Either AIBN (2% mol) or propionaldehyde (5% mol)
were used as initiators. The solution was stirred for 6 h at 850 rpm with a
magnetic stirrer bar at the selected temperature (45 or 70 °C) under an
oxygen atmosphere.
4-(phenoxy)-N-hydroxyphthalimide (8a): 1H-NMR (400 MHz, DMSO-
d6) δ 10.77 (s, 1H), 7.83 (d, 1H), 7.52 (t, 2H), 7.33 (m, 2H), 7.25 (s, 1H),
7.19 (d, 2H). 13C-NMR (400 MHz, DMSO-d6) δ 164.11, 163.84, 163.08,
155.07, 131.91, 131.01, 125.95, 125.82, 122.93, 122.68, 120.74, 112.01.
MS (ESI): m/z = 278.0 (M+Na)+, m/z = 532.6 (2M+Na)+.
4-(p-chlorophenoxy)-N-hydroxyphthalimide (8b): 1H-NMR (400 MHz,
DMSO-d6) δ 10.75 (s, 1H), 7.84 (d, 1H), 7.54 (d, 2H), 7.36 (dd, 1H), 7.31
(s, 1H), 7.22 (d, 2H). 13C-NMR (400 MHz, DMSO-d6) δ 164.03, 163.76,
162.56, 154.10, 131.94, 130.82, 129.66, 125.94, 123.37, 122.99, 122.40,
112.46. MS (ESI): m/z = 312.0 (M+Na)+.
4-(2,4,6-trimethylphenoxy)-N-hydroxyphthalimide (8c): 1H-NMR (400
MHz, DMSO-d6) δ 10.68 (s, 1H), 7.79 (d, 1H), 7.13 (dd, 1H), 7.03 (m,
3H). 13C-NMR (400 MHz, DMSO-d6) δ 164.15, 163.95, 162.89, 148.01,
135.65, 132.19, 130.36, 126.11, 122.02, 119.88, 109.36, 20.81, 16.14.
MS (ESI): m/z = 320.1 (M+Na)+.
4-(p-hexylphenoxy)-N-hydroxyphthalimide (8d): 1H-NMR (400 MHz,
DMSO-d6) δ 10.72 (s, 1H), 7.82 (d, 1H), 7.31 (m, 3H), 7.21 (s, 1H), 7.09
(d, 2H). 13C-NMR (400 MHz, DMSO-d6) δ 164.14, 163.88, 163.46,
152.80, 140.17, 131.87, 130.67, 125.93, 122.34, 120.67, 111.62, 34.91,
31.52, 31.35, 28.76, 22.48, 14.37. MS (ESI): m/z = 340.1 (M+H)+, m/z =
362.1 (M+Na)+, mp = 89.4 °C.
General procedure for oxidations in neat cumene: Aerobic oxidations
were performed in a 50 mL two-necked round-bottom flask equipped with
a condenser and containing CU (1.7g, 14.3 mmol), and an appropriate
amount of catalyst, as reported in Table 3. Either AIBN (2% mmol) or
propionaldehyde (5% mmol) were used as initiators. Variable amounts of
MeCN were also added, as indicated in Table 3. The solution was let to
react for 6 h under magnetic stirring (850 rpm), at the selected
temperature (45 or 70 °C), under an oxygen atmosphere.
Computational Methods
Ab initio and density functional calculations were performed by using the
Gaussian 09 program package [29] and Gaussview as the interface
program. The optimizations of all derivatives of NHPI was performed at
the B3LYP/6-311G(d,p) level of theory. The electrostatitic potential
4-(p-hexyloxylphenoxy)-N-hydroxyphthalimide (8e): 1H-NMR (400
MHz, DMSO-d6) δ 10.76 (s, 1H), 7.81 (d, 1H), 7.27 (dd, 1H), 7.18 (s, 1H),
7.14 (d, 2H), 7.05 (d, 2H), 3.99 (t, 2H), 1.73 (m, 2H), 1.43 (m, 2H), 1.33
This article is protected by copyright. All rights reserved.

10.1002/chem.201701573
Chemistry - A European Journal
FULL PAPER
surface (ESP) was calculated and generated at B3LYP/6-311G(d,p) level
of theory. Zero point energy (ZPE) was included in each result
O. Porta, F. Recupero, C. Punta, C. Gambarotti, R. Spaccini,
WO2009/115276A1. (d) E. Bencini, L. Melone, N. Pastori, S. Prosperini,
F. Recupero, C. Punta, C. Gambarotti, WO2011/001244A1. (e) F.
Recupero, C. Punta, L. Melone, S. Prosperini, N. Pastori,
WO2011/161523A1.
Acknowledgements
[15] K. Chen, J. Yao, Z. Chen, H. Li, J. Catal. 2015, 331, 76-85.
[16] Y. Kadoh, K. Oisaki, M. Kanai, Chem. Pharm. Bull. 2016, 64, 737-753.
[17] N. Sawatari, T. Yokota, S. Sakaguchi, Y. Ishii, J. Org. Chem. 2001, 66,
7889-7891.
The authors thank MIUR for continual support of their free-
radical chemistry (PRIN 2010-2011, project 2010PFLRJR_005).
[18] M. Petroselli, P. Franchi, M. Lucarini, C. Punta, L. Melone,
ChemSusChem 2014, 7, 2695-2703.
Keywords
N-hydroxyphthalimide;
aerobic
oxidation;
[19] J. Liu, Chem. Lett. 2010, 39 (11), 1194-1196.
homogeneous catalysis; solvent-free; hydrogen atom transfer.
[20] EP1449833 A1. 2004, Wako Pure Chemical Ind. LTD.
[21] L.Melone, C. Gambarotti, S. Prosperini, N. Pastori, F. Recupero, C.
Punta, Adv. Synth. Catal. 2011, 353, 147-154.
[22] L. Melone, S. Prosperini, C. Gambarotti, N. Pastori, F. Recupero, C.
Punta, J. Mol. Cat. A: Chemical, 2012, 355, 155-160.
References
[23] (a) F. Minisci, C. Gambarotti, M. Pierini, O. Porta, C. Punta, F.
Recupero, M. Lucarini, V. Mugnaini, Tetrahedron Lett. 2006, 47,1421–
1424. (b) R. Spaccini, L. Liguori, C. Punta, H.-R. Bjørsvik,
ChemSusChem 2012, 5, 261-265.
[1]
[2]
T. Newhouse, P.S. Baran, Angew. Chem. Int. Ed. 2011, 50, 3362-3374.
(a) J. M. Howell, K. Feng, J. R. Clark, L. J. Trzepkowski, M. C. White, J.
Am. Chem. Soc. 2015, 137, 14590-14593. (b) A. Badalyan, S. S. Stahl,
Nature 2016, 535, 406-410.
[24] N. Turrà, U. Neuenschwander, I. Hermans, ChemPhysChem 2013, 14,
1666-1669.
[25] M. Fang-Ming, W. Jin-Ling, M. Xiu-Shen Acta Crystallogr. Sect. C:
Cryst. Struct. Commun. 1995, 51, 712.
[3]
(a) C. C. Pattillo, I. I. Strambeanu, P. Calleja, N. A. Vermeulen, T.
Mizuno, M. C. White, J. Am. Chem. Soc. 2016, 138, 1265-1272. (b) S.
M. Paradine, J. R. Griffin, J. Zhao, A. L. Petronico, S. M. Miller, M. C.
White, Nature Chem. 2015, 7, 987-994.
[26] Toluene and cumene can be considered very similar in terms of their
capacity to solvate NHPI derivatives such as 8d and 9.
[27] The observed behavior prevents any precise determination of the
association constant pertaining to the HB adduct formation, which
however can be estimated to be of the same order of magnitude of that
observed in CDCl3.
[4]
[5]
A.-F. Voica, A. Mendoza, W. R. Gutekunst, J. O. Fraga, P. S. Baran,
Nature Chem. 2012, 4, 629-635.
(a) V. A. Schmidt, R. K. Quinn, A. T. Brusoe, E. J. Alexanian, J. Am.
Chem. Soc., 2014, 136, 14389-14392. (b) R. K. Quinn, Z. A. Könst, S.
E. Michalak, Y. Schmidt, A. R. Szklarski, A. R. Flores, S. Nam, D. A.
Horne, C. D. Vanderwal, E. J. Alexanian, J. Am. Chem. Soc. 2016, 138,
696-702.
[28] In principle, this behaviour, that is the signal drift observed as the
concentration is varied, could be also considered as due to the change
of the solvent inherent dielectric constant due to the addition of
increasing quantities of 9 or 8d. In any case, whether this could be a
valid consideration or not, this phenomenon does not weaken the
validity of the proposed thesis relative to the formation of HB dimers
and higher aggregates for 8d.
[6]
(a) F. Recupero, C. Punta, Chem. Rev. 2007, 107, 3800-3842. (b) C.
Galli, P. Gentili, O. Lanzalunga, Angew. Chem. Int. Ed. 2008, 47, 4790-
4796. (c) S. Coseri, Catal. Rev. 2009, 51, 218-292. (d) C. Punta, C.
Gambarotti, in Ideas in Chemistry and Molecular Sciences: Advances in
Synthetic Chemistry; Pignataro, B., Ed.; Wiley-VCH Verlag GmbH & Co.
KGaA, Weinheim, 2010, pp 3-24. (d) S. Wertz, A. Studer, Green Chem.
2013, 15, 3116-3134. (e) L. Melone, C. Punta, Beilstein J. Org. Chem.
2013, 9, 1296-1310.
[29] Gaussian 03, Revision D.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel,
G. E. Scuseria, M. A. Robb, J. R. Cheeseman, V. G. Zakrzewski, J. A.
Montgomery, Jr., R. E. Stratmann, J. C. Burant, S. Dapprich, J. M.
Millan, A. D. Daniels, K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V.
Barone, M. Cossi, R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S.
Clifford, J. Ochterski, G. A. Petersson, P. Y. Ayala, Q. Cui, K.
Morokuma, D. K. Malick, A. D. Rabuck, K. Raghavachari, J. B.
Foresman, J. Cioslowski, J. V. Ortiz, A. G. Baboul, B. B. Stefanov, G.
Liu, A. Liashenko, P. Piskorz, I. Komaroi, R. Gomperts, R. L. Martin, D.
J. Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, M.
Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W. Wong, J. L.
Andres, C. Gonzales, M. Head-Gordon, E. S. Replogle, J. A. Pople,
Gaussian, Inc., Wallingford CT, 2004.
[7]
[8]
[9]
(a) E.J. Horn, B.R. Rosen, Y. Chen, J. Tang, K. Chen, M. D. Eastgate,
P.S. Baran, Nature 2016, 533, 77-81. (b) D.P. Hruszkewycz, K.C. Miles,
O.R. Thiel, S.S. Stahl, Chem. Sci. 2017, 8, 1282-1287.
R. Amorati, M. Lucarini, M. Mugnaini, G. F. Pedulli, F. Minisci, F.
Recupero, F. Fontana, P. Astolfi, L. Greci, J. Org. Chem.2003, 68,
1747-1754.
X. Yang, Y. Wang, L. Zhou, C. Chen, W. Zhang, J. Xu, J. Chem.
Technol. Biotechnol. 2010, 85, 564-568.
[10] P.P. Toribio, A. Gimeno-Gargallo, M.C. Capel-Sanchez, M.P. de Frutos,
J.M. Campos-Martin, J.L.G. Fierro, Appl. Catal. A 2009, 363, 32-39.
[11] K. Kasperczyk, B. Orlińska, J. Zawadiak, Cent. Eur. J. Chem. 2014,
12(11), 1176-1182.
[12] (a) D. Levin, J.M. Dakka, J.E. Stanat, WO/2007/073916. (b) T.J.
Chen, J.C. Cheng, T.E. Helton, J.S. Buchanan, WO/2009/038900. (c)
F. M. Benitez, J. M. Dakka, E. J. Mozeleski, S. Zushma, J. S.
Buchanan, J. E. Stanat, WO/2009/058531. (d) J. R. Lattner, M. P.
Hagemeister, J. E. Stanat, J. M. Dakka, S. Zushma, WO/2010042273.
(e) K. Wang, R. Garcia, F. M. Benitez, WO/2013/043272. (f) M. P. de
Frutos, P. P.Toribio, R. Martos Calvente, J. M. Campos-Martin, J.L.G.
Fierro, EP1,520,853B1.
[13] (a) L. Melone, S. Prosperini, G. Ercole, N. Pastori, C. Punta, J. Chem.
Technol. Biotechnol. 2014, 89, 1370-1378. (b) L. Melone, P. Franchi, M.
Lucarini, C. Punta, Adv. Synth. Catal. 2013, 355, 3210-3220.
[14] (a) F. Minisci, O. Porta, F. Recupero, C. Punta, C. Gambarotti, M.
Pierini, WO2008/037435A1. (b) F. Minisci, O. Porta, F. Recupero, C.
Punta, C. Gambarotti, R. Spaccini, WO2009/115275A1. (c) F. Minisci,
This article is protected by copyright. All rights reserved.

10.1002/chem.201701573
Chemistry - A European Journal
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
FULL PAPER
Novel lipophilic N-
Manuel Petroselli, Lucio Melone,
hydroxyphthalimides catalysts
allowed firstly to conduct catalytic
oxidation of neat cumene for the
very first time, and secondly to
discover the limits of solvent free
conditions wherein hydrogen bond
(HB) driven aggregation of the
catalyst occurs.
Massimo Cametti* and Carlo Punta*
Page No. – Page No.
Lipophilic N-hydroxyphthalimide
Catalysts for the Aerobic Oxidation
of Cumene: Towards Solvent-Free
Conditions ... and Back
((Insert TOC Graphic here: max.
width: 5.5 cm; max. height: 5.0 cm))
This article is protected by copyright. All rights reserved.