Arene oxidation with malonoyl peroxides

Article abstract of DOI:10.1021/acs.orglett.5b00953

Malonoyl peroxide 7, prepared in a single step from the commercially available diacid, is an effective reagent for the oxidation of aromatics. Reaction of an arene with peroxide 7 at room temperature leads to the corresponding protected phenol which can be unmasked by aminolysis. An ionic mechanism consistent with the experimental findings and supported by isotopic labeling, Hammett analysis, EPR investigations, and reactivity profile studies is proposed.

Full text of DOI:10.1021/acs.orglett.5b00953

Letter
pubs.acs.org/OrgLett
Arene Oxidation with Malonoyl Peroxides
†
†
†
‡
Andrei Dragan, Tomasz M. Kubczyk, Julian H. Rowley, Stephen Sproules,
,†
and Nicholas C. O. Tomkinson*
†WestCHEM, Department of Pure and Applied Chemistry, Thomas Graham Building, University of Strathclyde, 295 Cathedral
Street, Glasgow, G1 1XL, U.K.
‡
WestCHEM, School of Chemistry, University of Glasgow, Joseph Black Building, University Avenue, Glasgow, G12 8QQ, U.K.
*
S Supporting Information
ABSTRACT: Malonoyl peroxide 7, prepared in a single step from
the commercially available diacid, is an effective reagent for the
oxidation of aromatics. Reaction of an arene with peroxide 7 at room
temperature leads to the corresponding protected phenol which can
be unmasked by aminolysis. An ionic mechanism consistent with the
experimental findings and supported by isotopic labeling, Hammett
analysis, EPR investigations, and reactivity profile studies is proposed.
2
he oxidation and functionalization of hydrocarbons is a
central facet of the chemical industry for the production
Scheme 1. C−H Oxidation Using Phthaloyl Peroxide
T
of high-tonnage commodities and the preparation of high-value
pharmaceuticals, agrochemicals, and fine chemicals. Therefore,
methods for selective oxidation of C−H bonds are of great
1
importance. Phenols represent a key class of oxidized
2
hydrocarbon. While there are a small number of reports in
which arenes are oxidized to the respective phenols using
3
peroxides and strong acids as additives, the oxidation of
aromatic C−H bonds still presents a synthetic challenge
specifically with respect to avoiding overoxidation.
A recent report from the laboratories of Houk and Siegel
described a metal-free oxidation of aromatic carbon−hydrogen
bonds which was proposed to proceed through an intriguing
4
reverse-rebound mechanism (Scheme 1). Reaction of
mesitylene 2 with 1.3 equiv of phthaloyl peroxide 1 in
hexafluoroisopropanol (HFIP) followed by basic solvolysis gave
the phenol 3 (97%).
The method outlined in Scheme 1 represents a significant
advance in arene oxidation. The proposed mechanistic pathway
for the transformation suggested homolytic fission of the weak
oxygenoxygen bond leading to diradical 4. Addition of this
radical to the arene gives 5 which through H atom abstraction
provides the observed product 6. Ester hydrolysis leads to the
phenol 3 (97%, two steps). The procedure has wide functional
group tolerance, and arene overoxidation did not prove
problematic. We believed two fundamental opportunities
existed for development of this procedure: First, phthaloyl
peroxide 1 is known to be very shock sensitive and explodes
9
dihydroxylation of alkenes. This reagent provides significant
advantages over phthaloyl peroxide 1 within the syn-
dihydroxylation reaction in terms of yield, selectivity, reaction
1
0,11
rate, substrate scope, and operating temperature.
experience in understanding the mechanism of reactions
Given our
12
involving the peroxide 7, together with the specific advantages
provided in alkene dihydroxylation, we elected to investigate
the reactivity of 7 in the oxidation of arenes. Within this report
we show that 7 reacts with arenes in the presence of a hydrogen
bond donor to give the corresponding functionalized aromatic
and present data consistent with the arene and peroxide
reacting through an ionic pathway.
5
,6
violently when heated, representing a significant hazard.
Second, the proposed reverse-rebound mechanism leading to 6
was based upon theoretical studies. Provision of experimental
evidence to support this pathway would be of great importance
to the understanding and development of this procedure.
In recent years we have been interested in the chemistry of
cyclic diacylperoxides and have shown that malonoyl peroxide
As a starting point to our investigations we reacted
mesitylene 2 with malonoyl peroxide 7 under the conditions
Received: April 2, 2015
7
8
7
,
and related derivatives, are effective for the syn-
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.orglett.5b00953
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
2
reported by Houk and Siegel (0.07 M, 40 °C). Reaction of
mesitylene 2 with 1.3 equiv of peroxide 7 in HFIP gave the
adduct 8 (94%, 20 h). More conveniently, this reaction could
be performed at room temperature (25 °C) rather than 40 °C
without compromise in yield (8, 98%) (Scheme 2). Cleavage of
group tolerance mirrored that obtained with phthaloyl
peroxide, which bodes well for further reaction development.
We therefore believe that malonoyl peroxide 7 represents an
effective alternative to the explosive phthaloyl peroxide 1 in the
oxidation of arenes.
In order to understand the reaction further a series of
experiments were undertaken to probe the mechanistic course
of the process. The reverse-rebound mechanism proposed for
the reaction of phthaloyl peroxide 1 was based upon theoretical
Scheme 2. Malonoyl Peroxide Mediated Oxidation of
Mesitylene
2
calculations. We sought to obtain experimental data using
malonoyl peroxide 7 to underpin knowledge of this trans-
formation. An analogous reverse-rebound mechanism with
peroxide 7 would involve homolytic cleavage of the peroxide
bond to give the diradical species 17 which on addition to the
arene would give the intermediate 18. Rearomatization of 18
through H atom abstraction would lead to the observed
product 8 (Scheme 3). In addition to this reverse-rebound
the ester (MeNH , EtOH, 25 °C, 1 h) gave the corresponding
2
phenol 3 (92% over two steps). Delighted by these excellent
preliminary results and given the distinct advantages of
malonoyl peroxide 7 over phthaloyl peroxide 1 outlined
above, we examined this transformation further.
Scheme 3. Potential Mechanistic Course of Reaction
Brief optimization of the reaction conditions showed that
transformations could be performed at room temperature and
considerably higher concentration (0.5 M) than those reported
previously (0.07 M) (see Supporting Information).
Houk and Siegel showed their transformation to have broad
functional group tolerance, describing the oxidation of over 50
2
substrates. Direct comparison of the reaction using malonoyl
peroxide 7 to those obtained for phthaloyl peroxide 1 on a
selection of these substrates showed the transformation to
proceed with similar yield and selectivity (Figure 1). Functional
mechanism we also considered an ionic pathway based on the
10
established reactivity of 7 with alkenes. Thus, nucleophilic
attack of the electron-rich arene 2 on the weak peroxide bond
would lead to intermediate 19, which upon rearomatization
would give 8.
Based upon reactivity patterns an ionic process appeared
plausible: the reaction required electron-rich aromatics to
proceed, with electron-donating groups directing the addition
of the peroxide to the ortho/para positions. Interestingly,
aromatic substrates containing electron-withdrawing groups
proved unreactive within this transformation. For example,
acetophenone, benzoic acid, methyl benzoate, and benzonitrile
all proved to be unreactive to peroxide 7 under the optimized
reaction conditions (7 (2 equiv), HFIP 0.5 M, 25 °C).
Conducting the reaction of mesitylene 2 and malonoyl peroxide
7
in HFIP (i) in the presence of light and air; (ii) exposed to
light under an argon atmosphere; (iii) in the dark in an aerobic
environment; and (iv) in the dark under an inert atmosphere
showed no significant differences in outcome with reactions
giving the product 8 (90−92% yield) after 6 h. In addition,
conducting the reaction at 4 °C, in the dark under an argon
atmosphere, gave the product (83% yield) after 24 h.
Combined, these observations show the reaction to proceed
under very mild conditions and did not rule out an ionic
pathway.
a
Figure 1. Arene oxidation with malonoyl peroxide 7. Isolated yield of
b
phenol using phthaloyl peroxide 1 reported in ref 2. The reaction was
perfomed at 50 °C. 1:1.6 mixture of o- and p-isomers obtained. 1:1
mixture of o- and p-isomers obtained. 1.1 equiv of malonoyl peroxide
c
d
e
18
Further investigations were carried out using O isotopically
7
used.
labeled malonoyl peroxide 20 as a mechanistic probe, which
B
DOI: 10.1021/acs.orglett.5b00953
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
was prepared from cyclopropane-1,1-dicarboxylic acid (see
Supporting Information for full details). Treatment of
mesitylene 2 with 1.35 equiv of malonoyl peroxide 20 (25
oxygen atom to achieve the results observed. It is not clear how
this selectivity would occur within a reverse-rebound process.
The data assembled at this stage were consistent with an
ionic interaction between the aromatic substrate and the
peroxide 7. To gain further evidence for the formation of a
carbocation intermediate, a Hammett analysis was conducted
on the arene oxidation. Monosubstituted mesitylene derivatives
were reacted with 1 equiv of malonoyl peroxide 7 (Figure 3),
°
C, 16 h) gave the ester 21 with two labels incorporated in the
structure (Scheme 4). Mass spectrometric analysis was
Scheme 4. Reaction of Mesitylene with Isotopically Labelled
Peroxide 20
consistent with incorporation of one label in the carboxylic
acid terminus of the molecule and a second label in the
carbonyl oxygen atom of the ester. No labeled oxygen was
observed in the newly formed carbonoxygen bond. To
corroborate this finding, a sample of 21 was treated with
methylamine and the crude reaction mixture was analyzed by
GCMS. The observed products were 2,4,6-trimethylphenol 3,
Figure 3. Hammett analysis of arene oxidation with peroxide 7.
and peroxide consumption was monitored against an internal
1
standard by H NMR spectroscopy. Determination of the initial
rates provided a linear Hammett plot based on literature σ
meta
14
values.
1
8
The Hammett plot (Figure 3) showed an excellent linear
relationship for both electron-withdrawing and -donating
substituents in the meta-position indicating that the same
mechanism is in operation with each substrate examined. The
gradient of the line, ρ, gave a moderate negative value of −2.7
indicating a considerable buildup of positive charge during the
transition state of the reaction, consistent with an aryl
carbocation intermediate (e.g., 19). As expected, reactions of
substrates with a meta-electron-donating substituent (e.g., 28)
proceeded at a significantly faster rate than those containing a
meta-electron-withdrawing group (e.g., 14 and 30).
amides 22 and 24 containing either one or two O labels, and
amide 23 resulting from decarboxylation. These findings show
1
8
that no scrambling of the O label from the peroxide 20 is
observed during the course of reaction and are consistent with
addition of an arene nucleophile to the weak O−O peroxide
bond.
It was possible that reversible formation of the labeled
diradical 25 could proceed without scrambling, if the rate of
peroxide bond formation is quicker than bond rotation (Figure
1
3
2
). This would mean that 26 and 27 would not be present in
solution. However, the observed selectivity in C−O bond
formation would require addition of the diradical 25 to the
aromatic ring to occur specifically through the unlabeled
To probe further the possibility of the reaction proceeding
through the diradical intermediate 17, we examined the
homolytic bond cleavage of peroxide 7 through DFT
calculations ((U)B3LYP/6-31+G(d)). Of importance was the
use of trifluoroethanol as the solvent in a CPCM, providing a
more accurate reflection of the reaction medium when
compared to the previous approaches to modeling this class
2
−1
of transformation. A transition state TS1 (29.7 kcal mol ) to
−1
form the diradical species 17 (24.8 kcal mol ) from peroxide 7
was found (Figure 4), a particularly high barrier for a reaction
that proceeds readily at 4 °C.
The potential of 17 being present within the reaction mixture
was also examined using electron paramagnetic resonance
(
EPR) spectroscopy experiments using 5,5-dimethyl-1-pyrro-
line-N-oxide (DMPO), a spin trap used in EPR spectroscopy
for the detection of oxygen based radicals. DMPO was inert
15
Figure 2. Potential scrambling of labels on formation of diradical.
C
DOI: 10.1021/acs.orglett.5b00953
Org. Lett. XXXX, XXX, XXX−XXX

Organic Letters
Letter
(b) White, M. C. Science 2012, 335, 807. (c) Company, A.; Lloret, J.;
Gomez, L.; Costas, M. In Catalysis by Metal Complexes; Springer:
Dordrecht, 2012; Vol. 38, pp 143−228. (d) Newhouse, T.; Baran, P. B.
Angew. Chem., Int. Ed. 2011, 50, 3362.
(2) (a) Quideau, S.; Deffieux, D.; Douat-Casassus, C.; Pouysegu, L.
Angew. Chem., Int. Ed. 2011, 50, 586. (b) Rappoport, Z. The Chemistry
of Phenols; Wiley-VCH: Weinheim, 2003. (c) George, T.; Mabon, R.;
Sweeney, G.; Sweeney, J. B.; Tavassoli, A. J. Chem. Soc., Perkin Trans. 1
2
000, 2529. (d) Tyman, J. H. P. Synthetic and Natural Phenols;
Elsevier: New York, 1996.
3) (a) Derbyshire, D. H.; Waters, W. A. Nature 1950, 165, 401.
(
(
(
b) Hart, H.; Buehler, C. A. J. Org. Chem. 1964, 29, 2397.
c) Hamilton, G. A.; Hanifin, J. W.; Friedman, J. P. J. Am. Chem.
Soc. 1966, 88, 5269. (d) Kovacic, P.; Kurz, M. E. J. Am. Chem. Soc.
1
966, 88, 2068. (e) Kovacic, P.; Kurz, M. E. Tetrahedron Lett. 1966, 7,
2
689. (f) Kurz, M. E.; Johnson, G. J. J. Org. Chem. 1970, 35, 2152.
(g) Olah, G. A.; Ohnishi, R. J. Org. Chem. 1978, 43, 865. (h) Olah, G.
A.; Fung, A. P.; Keumi, T. J. Org. Chem. 1981, 46, 4305. (i) Olah, G.
A.; Fung, A. P.; Keumi, T. J. Org. Chem. 1981, 56, 6148.
(
4) Yuan, C.; Liang, Y.; Hernandez, T.; Berriochoa, A.; Houk, K. N.;
Siegel, D. Nature 2013, 499, 192.
5) Greene, F. D. J. Am. Chem. Soc. 1956, 78, 2246.
−1
Figure 4. Peroxide homolytic cleavage (kcal mol ).
(
to the peroxide 7 in HFIP, as shown by control experiments.
Mixtures of mesitylene 2 and malonoyl peroxide 7 in both the
presence and absence of DMPO were analyzed by EPR
spectroscopy. In each of these experiments no radicals were
detected providing further important evidence in support of an
ionic pathway.
(6) For a recent report which prepares phthaloyl peroxide using flow
techniques followed by a solvent switch, see: Eliasen, M.; Thedford, R.
P.; Claussen, K. R.; Yuan, C.; Siegel, D. Org. Lett. 2014, 16, 3628.
(
7) (a) Griffith, J. C.; Jones, K. M.; Picon, S.; Rawling, M. J.; Kariuki,
B. M.; Campbell, M.; Tomkinson, N. C. O. J. Am. Chem. Soc. 2010,
32, 14409. (b) Picon, S.; Rawling, M.; Campbell, M.; Tomkinson, N.
C. O. Org. Lett. 2012, 14, 6250.
1
In summary, reaction of an arene with malonoyl peroxide 7
at room temperature in the presence of a hydrogen bond donor
leads to the corresponding functionalized aromatic. Exper-
imental findings supported by isotopic labeling, Hammett
analysis, EPR studies, and reactivity profile studies support an
ionic reaction pathway. The importance of the phenol
functional group in imparting unique structural, physical, and
electronic properties within molecules suggests this simple,
effective, and high yielding method for the oxidation of
aromatics will provide a useful route for the late stage aromatic
functionalization. Given the specific advantages of malonoyl
peroxide 7 over phthaloyl peroxide 1, it is expected that this
methodology will be of great use in the introduction of the
phenol functionality.
(
(
8) Jones, K. M.; Tomkinson, N. C. O. J. Org. Chem. 2012, 77, 921.
9) Rawling, M. J.; Tomkinson, N. C. O. Org. Biomol. Chem. 2013,
11, 1434.
(10) (a) Greene, F. D. J. Am. Chem. Soc. 1956, 78, 2250. (b) Greene,
F. D.; Rees, W. W. J. Am. Chem. Soc. 1958, 80, 3432.
(
11) Yuan, C.; Axelrod, A.; Varela, M.; Danysh, L.; Siegel, D.
Tetrahedron Lett. 2011, 52, 2540.
12) Rawling, M. J.; Rowley, J. H.; Campbell, M.; Kennedy, A. R.;
Parkinson, J. A.; Tomkinson, N. C. O. Chem. Sci. 2014, 5, 1777.
13) Fujimori, K.; Oshibe, Y.; Hirose, Y.; Oae, S. J. Chem. Soc., Perkin
Trans. 2 1996, 413.
(
(
(
(
14) Brown, H. C.; Okamoto, Y. J. Am. Chem. Soc. 1958, 80, 4979.
15) Buettner, G. R.; Mason, R. P. Spin trapping methods for
detecting superoxide and hydroxyl free radicals in vitro and in vivo. In
Oxidative Stress and Aging, Critical Reviews; Cutler, R. G., Rodriguez,
H., Eds.; World Scientific Publishing: River Edge, NJ, 2003; pp 27−38.
ASSOCIATED CONTENT
■
*
S
Supporting Information
Analytical data, experimental procedures, and NMR spectra for
all compounds reported. The Supporting Information is
available free of charge on the ACS Publications website at
DOI: 10.1021/acs.orglett.5b00953.
AUTHOR INFORMATION
Corresponding Author
■
*E-mail: Nicholas.Tomkinson@strath.ac.uk.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
The authors thank the EPSRC and University of Strathclyde for
financial support and the EPSRC Mass Spectrometry Service,
Swansea for high-resolution spectra.
REFERENCES
■
(
1) For selected general reviews, see: (a) Thirunavukkasrasu, V. S.;
Kozhushkov, S. I.; Ackerman, L. Chem. Commun. 2014, 50, 29.
D
DOI: 10.1021/acs.orglett.5b00953
Org. Lett. XXXX, XXX, XXX−XXX