Thermo-analytical determination of intermediates in the copper catalysed rearrangement of o-toluic acid to meta-cresol

Article abstract of DOI:10.1039/c3cy00119a

The intermediates formed during the conversion of o-toluic acid 2 to m-cresol 1 in the presence of Cu(ii) according to the Keading process, together with the temperatures at which crucial changes occur, have been investigated by means of X-ray diffractome

Full text of DOI:10.1039/c3cy00119a

Catalysis
Science & Technology
View Article Online
View Journal | View Issue
PAPER
Thermo-analytical determination of intermediates in
the copper catalysed rearrangement of o-toluic acid to
meta-cresol†
Cite this: Catal. Sci. Technol., 2013,
3, 2227
Abraham C. Sunil, Ernst H. G. Langner, Charlene Marais and
Barend C. B. Bezuidenhoudt*
The intermediates formed during the conversion of o-toluic acid 2 to m-cresol 1 in the presence of
Cu(II) according to the Keading process, together with the temperatures at which crucial changes occur,
have been investigated by means of X-ray diffractometry [which confirmed the formation of the
paddlewheel copper complex tetrakis(m2-2-methylbenzoato)bis(2-methylbenzoic acid)copper(II) 5 with
two apical o-toluic acid ligands], differential scanning calorimetry (DSC), thermogravimetric analysis
(TGA), NMR spectroscopy, infrared spectrometry (IR) and Maldi-Tof mass spectrometry. The major
chemical change observed at 164 1C might be ascribed to the dissociation of the apical o-toluic acid
ligands from 5 to give tetrakis(m2-2-methylbenzoato)copper(II) 6 accompanied by water loss. O–C bond
formation between the carboxylate oxygens and ortho-carbons of the toluic acid moieties in adjacent
paddlewheels of the stepped polymeric 6 is proposed to explain the formation of copper(I) 2-methyl-6-
{[(2-methylphenyl)carbonyl]oxy}benzoate 8 via intermediate 7 and initiates the decomposition observed
by DSC at 236 1C. Decarboxylation of 8 at 249.5 1C gives 3-methylphenyl 2-methylbenzoate 4, which
can be hydrolysed to o-toluic acid 2 and the target compound, m-cresol 1.
Received 18th February 2013,
Accepted 8th May 2013
DOI: 10.1039/c3cy00119a
www.rsc.org/catalysis
was identified as a possible relatively cheap alternative technol-
ogy for m-cresol 1 production (Scheme 1). The DOW phenol
Introduction
Cresols are important industrial starting materials for many
phenolic compounds and are currently obtained commercially
via a three step process from toluene. In a process similar to the
one for the production of phenol, toluene is reacted with
propylene to give a mixture of o-, m- and p-isopropyltoluene.
This mixture gives the corresponding cresols and acetone via
cleavage of the hydroperoxides obtained upon oxidation by air.1
Due to their close boiling points, m- and p-cresols are
however not separable by distillation, which leaves an elabo-
rated adduct crystallisation process, derivatisation or chroma-
tography as the only options for separation. As a result of these
difficulties, synthetic m-cresol is very expensive.1
process entails the thermal decomposition of copper(II) benzoate,
Since it is known that m-cresol 1 can be produced selectively
from o-toluic acid 2 and with this acid being readily available
from the corresponding xylene, Keading’s DOW phenol process2–4
Department of Chemistry, University of the Free State, PO Box 339, Bloemfontein,
9300, South Africa. E-mail: bezuidbc@ufs.ac.za; Fax: +27 (0)51 444 6384;
Tel: +27 (0)51 401 9021
† Electronic supplementary information (ESI) available: NMR, IR, MS, and
MALDI-TOF MS. CCDC 684436. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c3cy00119a
Scheme 1 Cu(II) catalysed m-cresol
1 formation. (Only m-cresol indicated,
though o- and p-products and intermediates are not excluded.6)
c
This journal is The Royal Society of Chemistry 2013
Catal. Sci. Technol., 2013, 3, 2227--2233 2227

View Article Online
Paper
Catalysis Science & Technology
presumably in dimeric form,4–6 to phenyl benzoate via the
decarboxylation of the preferred o-benzoyl salicylic acid, followed
by hydrolysis of the phenyl benzoate in moist air yielding the
desired phenol together with benzoic acid.7–9 In an attempt to
find the temperatures where crucial changes occur and to
identify the intermediates involved in those steps, a crystal
structure of the first intermediate, tetrakis(m2-2-methylbenzoato)-
bis(2-methylbenzoic acid)copper(II)10 5 (vide infra), was obtained,
which was subjected to heating under solventless conditions
during differential scanning calorimetry (DSC) and thermogravi-
metric analysis (TGA-MS). Intermediates and products were ana-
lysed using NMR spectroscopy, infrared spectrometry (IR) and
Maldi-Tof mass spectrometry.
Results and discussion
As Keading,2–4 Toland5 and Buijs6 proposed the formation of
a copper salt of the (substituted) benzoic acid that decom-
poses thermally to eventually form the (substituted) phenol, our
first aim was to prepare and identify the salt formed from
o-toluic acid 2. X-ray diffractometry revealed the formation of
tetrakis(m2-2-methylbenzoato)bis(2-methylbenzoic acid)copper(II)
5 after refluxing o-toluic acid 2 in toluene (bp. 110 1C) in the
presence of basic copper(II) carbonate and magnesium oxide.10
This gives credence to the dinuclear paddlewheel Cu(II)2benzoate4,
which Buijs6 proposed to be the key reagent in the DOW phenol
process (vide infra).
With the structure of the starting copper salt unambiguously
determined, attention was subsequently turned towards deter-
mining the temperatures at which chemical or physical
changes to this salt are being effected. The thermal decomposi-
tion of tetrakis(m2-2-methylbenzoato)bis(2-methylbenzoic acid)-
copper(II) 5 was thus investigated by means of DSC and TGA
under inert solventless conditions.
Fig. 1 DSC scans of (a) tetrakis(m2-2-methylbenzoato)bis(2-methylbenzoic acid)-
copper(II) 5 and (b) o-toluic acid 2 at a heating and cooling rate of 10 1C minꢀ1
.
The IR spectrum of 5 contained bands characteristic of
syn–syn bidentate benzoate ligands9,11 with typical n(OCO)assym
(1568 cmꢀ1) and n(OCO)sym (1393 cmꢀ1) values and Dn =
175 cmꢀ1 (Fig. 2a). This spectrum also contained bands very
An endothermic transition was observed during the first
DSC heating cycle (cycle 1) of tetrakis(m2-2-methylbenzoato)-
bis(2-methylbenzoic acid)copper(II) 5 (Fig. 1a) at 164.37 1C.
Upon cooling no typical peak accompanying crystallisation
within 10 degrees below 164 1C was observed, which implies
that the transition at 164.37 1C represents a chemical change
rather than a phase transition. The newly formed product from
this reaction subsequently crystallised at ca. 78 1C upon cooling
(cycle 1). From cycles 2 and 3 it is evident that the newly formed
compound has a melting point of 81.6 1C and is thermally
stable up to 220 1C.
In an attempt to determine the structure of the second
intermediate in the reaction sequence, the product 6 from the
1
transition at 164 1C was thus analysed using H NMR and IR
spectroscopies. Despite the fact that the 1H NMR spectra of
both the starting material [tetrakis(m2-2-methylbenzoato)bis
(2-methylbenzoic acid)copper(II) 5] and this intermediate pro-
duct showed poor resolution due to the paramagnetic nature of
copper(II) [and copper(m)], the differences in the aromatic
regions of these spectra [d 9.10, 7.28 and 6.74 vs. d 9.62, 7.75,
5.96] clearly indicated a chemical change to have taken place
during the heating cycle of the DSC experiment.
Fig. 2 IR spectra of (a) tetrakis(m2-2-methylbenzoato)bis(2-methylbenzoic acid)-
copper(II) 5, (b) tetrakis(m2-2-methylbenzoato)copper(II) 6 and (c) o-toluic acid 2.
c
2228 Catal. Sci. Technol., 2013, 3, 2227--2233
This journal is The Royal Society of Chemistry 2013

View Article Online
Catalysis Science & Technology
Paper
Scheme 2 Pyrolysis products from tetrakis(m2-2-methylbenzoato)bis(2-methylbenzoic acid)copper(II) 5.
similar to the carbonyl stretch (nCQO) and C–OH vibrations
(C–O stretch and C–O–H bend) of o-toluic acid 2 (nCQO
=
1685 cmꢀ1 for 5 and 1681 cmꢀ1 for 2; nC–OH = 1271 cmꢀ1 for
both 5 and 2) as well as an OH stretching band (nOH = 2783–
3330 cmꢀ1 for 5 and 2111–3463 cmꢀ1 for 2) (Fig. 2a and c).
These absorption bands were thus allocated to the two apical
o-toluic acid ligands in 5. From the IR spectrum of the DSC
product 6 obtained at 164 1C, it was clear that the apical o-toluic
acid ligands have dissociated from complex 5 (Scheme 2) as the
CQO, C–OH and OH bands associated with it were absent
(Fig. 2b). The paddlewheel itself seemed to be intact judged by
the presence of the n(OCO)assym (1560 cmꢀ1) and n(OCO)sym
(1399 cmꢀ1) bands.
During the third DSC cycle (Fig. 1a), an endothermic transi-
tion at 236 1C signalled the onset of complete thermal decom-
position of the complex.
Fig. 3 TGA scans with corresponding MS traces (for H2O, CO2, CO and toluene)
of tetrakis(m2-2-methylbenzoato)bis(2-methylbenzoic acid)copper(II) 5.
In an attempt to get a better understanding of the DSC
results of tetrakis(m2-2-methylbenzoato)bis(2-methylbenzoic
acid)copper(II) 5, o-toluic acid 2 was also subjected to DSC
analysis, confirming the melting point of o-toluic acid 2 to be
at 104 1C (Fig. 1b), while decomposition of the acid started at
120 1C and reached a maximum at 234 1C. The absence of a
melting point peak corresponding to the o-toluic acid 2 formed
during the pyrolysis of 5 (Scheme 2 and Fig. 1a) can thus be
rationalized in terms of this decomposition of the acid at
temperatures of above 120 1C.
a 24% mass loss between 140 and 200 1C with the midpoint at
164.3 1C (corresponding to the temperature where the chemical
transition was observed in DSC cycle 1 in Fig. 1a) was observed
as shown in the TGA scan in Fig. 3. For this process the mass
spectrometer detected a rise in water concentration,‡ whereas
DSC and IR have proven the reaction in Scheme 2 to take place
at this temperature, releasing o-toluic acid 2 (vide supra).
Under the inert atmospheric conditions maintained during
the TGA, this water loss detected cannot be ascribed to the
It was thus decided to subject tetrakis(m2-2-methylbenzoato)
bis(2-methylbenzoic acid)copper(II) 5 to TGA analysis under
inert solventless conditions and analyse the gaseous products
to be liberated, if any, by mass spectrometry. For compound 5,
‡ It is important to note that the mass spectrometer connected to the TGA
scanned only for fragments with molecular masses of 18, 44, 28 and 92.
c
This journal is The Royal Society of Chemistry 2013
Catal. Sci. Technol., 2013, 3, 2227--2233 2229

View Article Online
Paper
Catalysis Science & Technology
Scheme 3 Proposed process steps for the formation of m-cresol 1 from o-toluic acid 2 in the presence of Cu(II).‡‡
O2-catalyzed regeneration of Cu(II) from Cu(I).4,7 In our opinion, to Scheme 2 or the loss of water of crystallisation from 5. Even
the water loss could either be ascribed to anhydride‡ formation though no water seemed to be present in the X-ray structure of
from and/or decomposition of o-toluic acid 2 formed according tetrakis(m2-2-methylbenzoato)bis(2-methylbenzoic acid)copper(II) 5,
c
2230 Catal. Sci. Technol., 2013, 3, 2227--2233
This journal is The Royal Society of Chemistry 2013

View Article Online
Catalysis Science & Technology
Paper
it has been documented that water present in crystals might be
To exclude the possibility that the DSC product 6 formed at
overlooked if no proper site placement is possible,12 rendering 164 1C might be either ester 3 or 4, 3-methylphenyl 2-methyl-
the second possibility plausible. With regard to the first sce- benzoate15 4 was prepared by synthesis from o-toluic acid 2 and
nario, the liberation of water was neatly observed between 120 m-cresol 1 (SOCl2, py, DMAP) and the boiling point was
and 220 1C when o-toluic acid 2 was subjected to TGA.
determined to be 260 1C. Methodology for the preparation of
The solid product obtained from the transition at 164 1C (6) 2-benzoyloxybenzoic acids by Cheong, Duncanson and Grif-
underwent further decomposition with an initial 2.5% mass fiths16 was used to also prepare the novel 2-methyl-6-{[(2-
loss at 216 1C, which can also be ascribed to the liberation of methylphenyl)carbonyl]oxy}benzoic acid 3 from 2-hydroxy-6-
H2O. This step still falls within the range where o-toluic acid 2 methylbenzoic acid and o-toluoylchloride in the presence of
decomposes to liberate water.
pyridine and DMAP. Both esters 3 and 4 were thus analysed
The 18.5% mass loss at 249.6 1C corresponds to the loss of using IR spectroscopy. In both cases CQO bands characteristic
both H2O and CO2, followed by further decomposition above of esters15 (n = 1734 cmꢀ1 for 4 and 1735 cmꢀ1 for 3) were
250 1C with the loss of H2O, CO2, CO and toluene, in agreement observed. As these were absent in the IR spectrum of the DSC
with the total decomposition observed in cycle 3 in Fig. 1a.
product, it can be concluded that esters 3 and 4 were formed at
Since H2O, CO2 and CO losses are not very diagnostic and temperatures of above 164 1C in later steps. Similarly, no
could occur from any part of the molecule or a fragment of it, any indication of the presence of m-cresol 1 could be found in the
conclusion as to the structures of intermediate products and the IR spectrum of the product formed at 164 1C.
transformations involved in the mass loss from 5 is highly risky.
It was therefore decided to subject the solid amorphous
product 6 from the 164 1C transition of 5 to MS analysis.
Conclusion
MALDI-TOF MS analysis (ꢀve) of the reaction mixture after
Considerable progress towards understanding the process asso-
the second DSC cycle on tetrakis(m2-2-methylbenzoato)bis
ciated with the conversion of o-toluic acid 2 to m-cresol 1 in the
(2-methylbenzoic acid)copper(II) 5 failed to show a peak corres-
presence of Cu(II) with regard to intermediates formed and the
ponding to tetrakis(m2-2-methylbenzoato)copper(II) 6, but gave
temperatures at which crucial changes occur has been made.
insight into the possible decomposition products thereof.§
Based on the evidence with regard to intermediates obtained in
MALDI-TOF MS confirmed the presence of o-toluic acid 2
previous and the current studies, the steps shown in Scheme 3
([2]ꢁꢀ, m/z 136.052) and an adduct of 3-methylphenyl 2-methyl-
can be proposed for this important conversion.
benzoate 4 with copper¶ ([9]ꢁꢀ, m/z 288.645) (Scheme 3).
The structure of the starting copper salt was unambiguously
Furthermore, a peak corresponding to the copper(I) salt of
determined by means of X-ray diffractometry to be that of
tetrakis(m2-2-methylbenzoato)bis(2-methylbenzoic acid)copper(II)
5. The major change at 164 1C observed in both the DSC and
TGA (Fig. 1a and 3) can be ascribed to dissociation of the apical o-
toluic acid ligands from tetrakis(m2-2-methylbenzoato)bis(2-
methylbenzoic acid)copper(II) 5 to give tetrakis(m2-2-methylben-
zoato)copper(II) 6 (according to IR), accompanied by water loss
(TGA-MS; probably water of crystallisation and/or water released
during anhydride formation and/or the decomposition of the
dissociated o-toluic acid 2). As it is known that paddlewheel
structures similar to 6 tend to exist as stepped polymeric struc-
tures by coordination of the second lone pair of the benzoate
oxygen to a Cu of a neighbouring paddlewheel in the apical
position,6,17 solventless decomposition of complex 6, starting at
236 1C** (DSC), might be explained by O–C bond formation
between the carboxylate oxygens and ortho-carbons of toluic acid
moieties in adjacent paddlewheels to form intermediate 7. Such a
packing pattern of 6 would in addition facilitate regioselective
ortho oxygenation as encountered before.18,19 This steric course of
the reaction previously led to Kaeding’s postulation4 of a flat six
membered ring transition state involving nucleophilic attack of
oxygen on the position ortho to the carboxyl group. From thermal
decomposition experiments of benzoic acid-1-14C as well as cupric
ꢀ
2-methyl-6-{[(2-methylphenyl)carbonyl]oxy}benzoic acid (3) ([8]ꢁ
,
m/z 332.011) was observed. These esters (3 and 4) are analogous to
the o-benzoyl salicylic acid and phenyl benzoate intermediates
isolated and identified by Kaeding et al.8,13 during the oxidation of
benzoic acid to phenol in the DOW phenol process. Furthermore,
a peak corresponding to 6-methylsalicylic acid 10 ([10]ꢁꢀ, m/z
151.878), previously also encountered by Kaeding and Shulgin14
after the pyrolysis of the basic cupric salt of o-toluic acid 2, and
which can in our case be explained by the hydrolysis of 3,8 was
observed. In addition, a peak which might be ascribed to toluene
complexed to copper (11) was observed at m/z 245.178 and is
corroborated by the observed release of toluene during the TGA of
tetrakis(m2-2-methylbenzoato)bis(2-methylbenzoic acid)copper(II) 5.
2
benzoate in the presence of H2O, Schoo et al.19,20 could prove
§ Laser induced pyrolysis of metal carboxylates has been reported.22
¶ The formation of metallic copper from Cu(I) during the DOW-phenol process
has been described by Buijs et al.9 and might be ascribed to the disproportiona-
tion of two Cu(I) moieties to Cu(II) and Cu(m).
** At this temperature, virtually all o-toluic acid 2 released according to Scheme 2
would have decomposed (vide supra: DSC decomposition starts at 120 1C and
reaches a maximum at 234 1C).
8 As crystalline 5 was pyrolysed, no basic cupric salts could be present.
c
This journal is The Royal Society of Chemistry 2013
Catal. Sci. Technol., 2013, 3, 2227--2233 2231

View Article Online
Paper
Catalysis Science & Technology
that oxygenation occurs at the position ortho to the carboxyl group acid 3 was recorded neat on a Spectrum BX FT-IR from Perkin
and that hydrogen exchange occurs at the same position. Toland5 Elmer, whereas all other infrared spectra were recorded on a
also favoured a heterolytic mechanism, though a homolytic route Bruker Tensor 27 IR spectrophotometer using KBr disks. Thermal
could not be excluded. Kaeding et al.,8 after realizing that changes were studied thermogravimetrically under argon at a
copper(m) and Cu2O are similar in colour and that the latter heating rate of 10 1C minꢀ1 using a Mettler-Toledo TGA/
Cu(I) species was mostly formed in the thermal decomposition of SDTA851e fitted with a Pfeiffer Thermostar MS, whereas DSC
Cu(II) benzoates, supported the homolytic radical cage mecha- was performed on a Mettler-Toledo DSC822 in a nitrogen atmo-
nism postulated by Schoo and co-workers.19,20 For both mechan- sphere at a similar heating rate.
isms postulated, the copper(II) benzoate was thought to adopt
orientation 12 with the carboxylate oxygen of one toluate moiety
m-Cresol 12
in close proximity to the ortho carbon of a second toluate moiety. 1H NMR (300 MHz, CDCl3) dH 7.18 (1H, t, J 7.18 Hz, OH); 6.84–
Buijs6,21 on the other hand postulated electrophilic aromatic 6.81 (1H, m, H-5); 6.74–6.71 (2H, m, H-4, H-6); 5.81 (1H, s, H-2);
substitution with ligand to metal electron transfer in stepped 2.35 (3H, s, CH3), nmax/cmꢀ1 (KBr) 1588.41, 1490.34, 1461.72,
polymeric Cu(II)2benzoate4 paddlewheel structures. In our sce- 1277.47, 1264.66, 1244.89, 1151.98, 925.64, 771.63, 732.22, 686.29.
nario, either heterolytic or homolytic C–O bond formation would
o-Toluic acid 2
give access to intermediate 7, releasing either o-toluate anions or
2-methylbenzoyloxy radicals 2a. Proton or Hꢁ abstraction from
n
max/cmꢀ1 (KBr) 3063.17, 2971.66, 2929.29, 2689.87, 2649.88,
intermediate 7 would subsequently result in the formation of 2558.81, 2521.85, 1680.92 (CO), 1601.07, 1574.67, 1489.74,
copper(I) 2-methyl-6-{[(2-methylphenyl)carbonyl]oxy} benzoate†† 1456.43, 1435.20, 1408.16, 1380.88, 1315.40, 1271.71 (C–OH),
8 (observed by MALDI-TOF MS).‡‡ Decarboxylation of the latter 1199.80, 1171.37, 1087.52, 1037.36, 915.58, 831.35, 803.21,
to give 3-methylphenyl 2-methylbenzoate 4 (observed by MALDI- 765.47, 738.32, 658.41, 566.99, 532.43, 483.20.
TOF MS and identified in the pyrolysis reaction mixture of the
copper(II) salt of o-toluic acid 2 also reported by Toland5) seem to
3-Methylphenyl 2-methylbenzoate 4
occur at 249.5 1C, whereas the mass loss above 258 1C observed in o-Toluic acid 2 (1.00 g, 7.34 mmol), thionyl chloride (1.31 g,
the TGA of tetrakis(m2-2-methylbenzoato)bis(2-methylbenzoic 0.800 ml, 11.0 mmol, 1.5 eq.) and dichloromethane (10 ml)
acid)copper(II) 5 corresponds to the decomposition observed in were refluxed at 40 1C. After the acid had completely reacted
the DSC (Fig. 1a) and also coincides with the boiling point of (TLC), the solvent and HCl gas were removed in vacuo on a
3-methylphenyl 2-methylbenzoate 4. Hydrolysis of 4 would finally rotary evaporator. Dichloromethane (10 ml) was added to the
yield o-toluic acid 2 and the industrially important m-cresol 1. concentrate, followed by the addition of pyridine (10 ml,
These transformations might also be accompanied by other side 0.12 mol, 16 eq.) and m-cresol 1 (166.54 mg, 1.5401 mmol,
reactions, leading to the release of water at 249.6 1C amongst 0.21000 eq.). The reaction mixture was stirred at room temper-
others, and might differ from reactions taking place in solvents. ature for 2 hours, and then dimethyl amino pyridine (DMAP,
one spatula) was added. After 12 hours, the reaction was
stopped following TLC-analysis (hexane–EtOAc 80 : 20, v/v).
Dichloromethane (50 ml) was added to the reaction mixture,
Experimental
which was washed with 3 M HCl (3 ꢂ 40 ml), H2O (2 ꢂ 40 ml)
2-Hydroxy-6-methylbenzoic acid was bought from Apollo Scien-
and saturated aq. sodium bicarbonate solution (40 ml). The
dichloromethane phase was dried over sodium sulphate and
the solvent removed in vacuo. The reaction mixture was thus
purified by PLC (hexane–EtOAc 80 : 20, v/v) to give 3-methyl-
phenyl 2-methylbenzoate15 4 (RF 0.52, 98.12 mg, 9.81%, tR 31.2
tific Ltd, whereas all other commercially available reagents
were bought from Sigma-Aldrich and used as received. TLC
was performed on precoated Merck plastic sheets (Silica Gel 60
F254). Preparative plates (PLC), 20 ꢂ 20 cm, silica Gel PF254
(1.0 mm) were air-dried and used without prior activation.
Compounds were recovered from the absorbent with acetone.
Flash column chromatography was performed using Merck
Kieselgel 60 (230–400 mesh) under N2 pressure. NMR spectra
were recorded on Bruker AM 300 or Bruker AM 600 FT spectro-
meters at 296 K in CDCl3 or acetone-d6. Maldi-Tof spectra were
collected using a Bruker Microflex LRF20 in the negative mode
with the minimum laser power required to observe signals. The
IR spectrum of 2-methyl-6-{[(2-methylphenyl)carbonyl]oxy}benzoic
1
min) as a colourless oil, H NMR (600 MHz, (CD3)2CO) dH 8.13
(1H, dd, J 1.24 Hz, J 8.52 Hz, H-6); 7.56 (1H, dt, J 1.24 Hz, H-5);
7.42–7.40 (2H, m, H-3, H-4); 7.35 (1H, t, J 7.4 Hz, H-50); 7.14–
7.07 (3H, m, H-20, H-40, H-60); 2.65 (3H, s, CH3); 2.39 (3H, s,
CH3); dC (150.9 MHz, (CD3)2CO); dC 205.9 (CO), 166.2, 152.0, 141.4,
140.3, 133.4, 132.7, 131.7, 129.7, 127.2, 126.7, 123.3, 119.7, 21.9,
21.3; nmax/cmꢀ1 (KBr) 1734.27 (CO), 1224.73, 1141.39, 1043.16,
767.94, 732.60, 685.81; m/z (EI) 226 (M), 119 (100%), 91;
C15H14NaO2+ requires 249.0886, found 249.0892 (MNa+, ES +ve).
†† The formation of Cu(I) has been reported before.9
2-Methyl-6-{[(2-methylphenyl)carbonyl]oxy}benzoic acid 3
‡‡ Though only homolytic C–O bond formation is shown, a heterolytic mecha-
nism cannot be excluded. Likewise, Hꢁ abstraction from 6 by a 2-methylbenzoy-
loxy radical 2a cannot be excluded, though aryloxy radicals are inclined to rather
add to unsaturated systems than to hydrogen abstraction.23 Decarboxylation of
2-methylbenzoyloxy radicals 2a,24 o-toluic acid 2, and 6-methylsalicylic acid 10
can also contribute towards CO2 release.
2-Hydroxy-6-methylbenzoic acid 10 (110.8 mg, 0.79 mmol, 1.0 eq.)
was dissolved in dichloromethane (5 ml) and cooled down to 0 1C
in an ice bath, and then pyridine (0.1 ml, 1.2 mmol, 1.6 eq.) was
added. o-Toluoylchloride (0.1 ml, 0.77 mmol, 1.0 eq.) was dissolved
c
2232 Catal. Sci. Technol., 2013, 3, 2227--2233
This journal is The Royal Society of Chemistry 2013

View Article Online
Catalysis Science & Technology
Paper
in dichloromethane (5 ml) in another flask and cooled down to Acknowledgements
0 1C, and it was added dropwise to the 2-hydroxy-6-methylbenzoic
Financial support from SASOL and the collection of DSC and
TGA scans by Chris Joubert are gratefully acknowledged. The
authors also thank Caryl Janse van Rensburg, University of
KwaZulu-Natal for the accurate mass analyses.
acid 10–pyridine solution over a period of 30 minutes. Dichloro-
methane (2 ml) was added to ensure a complete transfer of the
acid chloride. A spatula point of DMAP was added, after which the
temperature of the reaction mixture was allowed to rise to room
temperature. The reaction mixture was stirred at room temper-
ature for 2 days, and then more DMAP (spatula tip) was added.
After a further 3 days, the reaction mixture was poured onto ice,
acidified with 3 N HCl and extracted into dichloromethane (3 ꢂ
50 ml). The combined dichloromethane layers were dried over
Na2SO4, filtered in vacuo and the solvent was removed under
reduced pressure. The crude reaction mixture was separated by
PLC (hexane : EtOAc 9 : 1, v/v, CH3COOH) to give 2-methyl-6-{[(2-
methylphenyl)carbonyl]-oxy}benzoic acid 3 (Rf 0.09; Rf 0.81 in
hexane : EtOAc 1 : 1, v/v, CH3COOH, 111.6 mg, 54%) as a yellow
Notes and references
1 A. K. Mukhopadhyay, Industrial Chemical Cresols and Down-
stream Derivatives, Marcel Dekker, New York, 2005, pp. 33–58.
2 W. W. Keading, R. O. Lindblom and R. G. Temple, Ind. Eng.
Chem., 1961, 53, 805.
3 W. W. Keading, R. O. Lindblom, R. G. Temple and H. I.
Mahon, Ind. Eng. Chem. Proc. Des. Dev., 1965, 4(1), 97.
4 W. W. Keading, J. Org. Chem., 1961, 26, 3144.
5 W. G. Toland, J. Am. Chem. Soc., 1961, 2507.
6 W. Buijs, J. Mol. Catal. A: Chem., 1999, 146, 237.
7 W. W. Kaeding and G. R. Collins, J. Org. Chem., 1965,
30(11), 3750.
8 W. W. Kaeding, H. O. Kerlinger and G. R. Collins, J. Org.
Chem., 1965, 30(11), 3754.
1
amorphous solid, H NMR (600 MHz, CDCl3) dH 9.46 (br. s, 1H,
COOH), 8.10 (br. d, 1H, J = 7.65 Hz, H-60), 7.46–7.43 (m, 1H, H-40),
7.41 (t, 1H, J = 7.91 Hz, H-4), 7.29–7.23 (m, 2H, H-30, H-50), 7.15 (br.
d, 1H, J = 7.15 Hz, H-3 or H-5), 7.10 (br. d, 1H, J = 7.97 Hz, H-3 or
H-5), 2.61 (s, 3H, Me), 2.44 (s, 3H, Me); 13C NMR (150.9 MHz,
CDCl3) dC 171.65, 165.85, 149.11, 141.39, 139.06, 132.84, 131.91,
131.46, 131.29, 128.63, 128.48, 128.24, 125.61, 120.89, 21.76, 20.52;
9 W. Buijs, P. Comba, D. Corneli, Y. Y. Mengerink, H. Pritzkow
and M. Schickedanz, Eur. J. Inorg. Chem., 2001, 3134.
10 A. C. Sunil, B. C. B. Bezuidenhoudt and J. M. J. van Rensburg,
Acta Crystallogr., Sect. E: Struct. Rep. Online, 2008, 64, m553.
11 (a) F. T. Greenaway and L. J. Norris, Inorg. Chim. Acta, 1988, 279;
n
max/cmꢀ1 (neat) 2923.59, 1735.32 (CO), 1687.91 (CO), 1604.41,
1575.88, 1468.16, 1454.51, 1378.89, 1305.22, 1285.88, 1248.57,
1232.65, 1173.65, 1139.61, 1129.28, 1085.15, 1060.12, 926.51,
899.75, 858.93, 816.62, 770.12, 689.99, 673.81; m/z (EI) 270 (M,
trace), 253 (trace), 119 (100%), 105 (2), 91 (31), 77 (3.6); m/z
(MALDI-TOF, ꢀve) 269.126 (100%), 270.130 (35.5), 271.069 (8),
272.132 (14); C16H14O4Na+ requires 293.0790, found 293.0790
(MNa+, ES +ve).
ˇ´
(b) V. Zelenak, M. Sabo, W. Massa and P. Llewellyn, Inorg. Chim.
Acta, 2004, 357, 2049; (c) A. N. Wein, R. Cordeiro, N. Owens and
H. Olivier, J. Fluorine Chem., 2009, 130, 197; (d) B. S. Manhas and
A. K. Sardana, Synth. React. Inorg., Met.-Org., Nano-Met. Chem.,
2005, 35, 171; (e) B. S. Manhas, A. K. Sardana and S. B. Kalia,
Synth. React. Inorg. Met.-Org. Chem., 2003, 33(4), 519.
Tetrakis(l2-2-methylbenzoato)bis(2-methylbenzoic
12 Cambridge Structural Database, Version 5.30, Cambridge
Crystallographic Data Centre, Cambridge, UK, 2008.
13 W. W. Kaeding, J. Org. Chem., 1964, 29(9), 2556.
14 W. W. Kaeding and A. T. Shulgin, J. Org. Chem., 1962, 27, 3551.
15 D. Hausigk, Liebigs Ann. Chem., 1969, 726, 216.
16 Y.-K. Cheong, P. Duncanson and D. V. Griffiths, Tetra-
hedron, 2008, 64, 2329.
acid)copper(II) 5
o-Toluic acid 2 (2.52 g, 18.5 mmol), basic copper(II) carbonate
(CuCO3ꢃCu(OH)2, 0.73 g, 3.3 mmol, 0.18 eq.) and magnesium
oxide (0.19 g, 4.7 mmol, 0.25 eq.) were heated in refluxing
toluene (30 ml) for 24 hours. The product was extracted and
crystallised from diethyl ether (50 ml) to yield tetrakis(m2-2-
methylbenzoato)bis(2-methylbenzoic acid)copper(II)10 5 (1.31 g,
45.2%) as dark bluish-green cubes, 1H NMR (300 MHz, CDCl3) d
9.10 (s), 7.28 (s), 6.74 (s), 3.45 (s), 2.34 (s), 1.32 (s); nmax/cmꢀ1
(KBr) 2929.83, 2650.09, 2534.08, 1685.09 (CO), 1614.70, 1592.73,
1568.13 (OCO), 1505.55, 1455.27, 1426.99, 1392.80 (OCO),
1270.89 (C–OH), 1195.30, 1153.35, 1091.80, 1037.59, 855.91,
789.51, 742.96, 700.24, 670.79, 565.02, 476.19, 409.90.
´
17 F. P. W. Agterberg, H. A. J. Provo Kluit, W. L. Driessen,
H. Oevering, W. Buijs, M. T. Lakin, A. L. Spek and J. Reedijk,
Inorg. Chem., 1997, 36, 4321.
18 W. W. Kaeding, Pet. Refin., 1964, 43(11), 173.
19 W. Schoo, J. U. Veenland, J. A. Bigot and F. L. J. Sixma,
Recueil, 1961, 80, 134.
20 W. Schoo, J. U. Veenland, Th. T. de Boer and F. L. J. Sixma,
Recueil, 1963, 82, 172.
Tetrakis(l2-2-methylbenzoato)copper(II) 6
21 W. Buijs, Top. Catal., 2003, 24(1–4), 73.
ˆ
1H NMR (300 MHz, CDCl3) dH 9.62 (s); 7.75 (m); 7.68 (m); 5.96 22 A. Rotaru, C. Constantinescu, A. Madruleanu, P. Rotaru,
(s); 4.23 (m); 2.85 (d); 1.31 (m); 0.93 (m); nmax/cmꢀ1 (KBr)
A. Moldovan, K. Gyoryova, M. Dinescu and V. Balek, Thermo-
¨
´
3854.10, 3750.23, 3675.33, 3649.44, 3446.88, 3073.75, 3023.86,
chim. Acta, 2010, 498, 81.
2966.05, 2927.80, 1606.70, 1584.59, 1559.63 (OCO), 1503.91, 23 M. E. Kurz and M. Pellegrini, J. Org. Chem., 1969, 35(4), 990.
1488.90, 1458.11, 1438.08, 1399.39 (OCO), 1288.17, 1156.25, 24 B. Abel, J. Assmann, P. Botschwina, M. Buback, M. Kling,
1101.93, 1052.39, 857.26, 805.79, 787.47, 737.73, 696.38, 671.11,
563.26, 499.38, 473.18.
R. Oswald, S. Schmatz, J. Schroeder and T. White, J. Phys. Chem.
A, 2003, 107, 5157.
c
This journal is The Royal Society of Chemistry 2013
Catal. Sci. Technol., 2013, 3, 2227--2233 2233